Noninvasive Strategies to Optimise Brain Plasticity: From ...downloads.hindawi.com/journals/specialissues/672750.pdfNeural Plasticity Guest Editors: Alessandro Sale, Anthony J. Hannan,

Neural Plasticity

Guest Editors: Alessandro Sale, Anthony J. Hannan, Lamberto Maffei, and Andrea Guzzetta

Noninvasive Strategies to Optimise Brain Plasticity: From Basic Research to Clinical Perspectives

Noninvasive Strategies to Optimise BrainPlasticity: From Basic Research to ClinicalPerspectives

Neural Plasticity

Noninvasive Strategies to Optimise BrainPlasticity: From Basic Research to ClinicalPerspectives

Guest Editors: Alessandro Sale, Anthony J. Hannan,Lamberto Maffei, and Andrea Guzzetta

Copyright © 2013 Hindawi Publishing Corporation. All rights reserved.

This is a special issue published in “Neural Plasticity.”All articles are open access articles distributed under theCreativeCommonsAttribu-tion License, which permits unrestricted use, distribution, and reproduction in anymedium, provided the original work is properly cited.

Editorial Board

Robert Adamec, CanadaShimon Amir, CanadaMichel Baudry, USAMichael S. Beattie, USAClive Raymond Bramham, NorwayAnna Katharina Braun, GermanySumantra Chattarji, IndiaRobert Chen, CanadaDavid Diamond, USAM. B. Dutia, UKRichard Dyck, CanadaZygmunt Galdzicki, USAPreston E. Garraghty, USAPaul E. Gold, USAManuel B. Graeber, AustraliaAnthony Hannan, Australia

George W. Huntley, USAYuji Ikegaya, JapanLeszek Kaczmarek, PolandJeansok J. Kim, USAEric Klann, USAMaThlgorzata Kossut, PolandFrederic Libersat, IsraelStuart C. Mangel, UKAage R. Møller, USADiane K. O’Dowd, USASarah L. Pallas, USAA. Pascual-Leone, USAMaurizio Popoli, ItalyBruno Poucet, FranceLucas Pozzo-Miller, USAVilayanur S. Ramachandran, USA

Kerry J. Ressler, USASusan J. Sara, FranceTimothy Schallert, USAMenahem Segal, IsraelPanagiotis Smirniotis, USAIvan Soltesz, USAMichael G. Stewart, UKNaweed I. Syed, CanadaDonald A. Wilson, USAJ. R. Wolpaw, USAChun-Fang Wu, USAJ. M. Wyss, USALin Xu, ChinaMin Zhuo, Canada

Contents

Noninvasive Strategies to Optimise Brain Plasticity: From Basic Research to Clinical Perspectives,Alessandro Sale, Anthony J. Hannan, Lamberto Maffei, and Andrea GuzzettaVolume 2013, Article ID 863970, 2 pages

Brain Reorganization following Intervention in Children with Congenital Hemiplegia: A SystematicReview, E. Inguaggiato, G. Sgandurra, S. Perazza, A. Guzzetta, and G. CioniVolume 2013, Article ID 356275, 8 pages

Activity-Dependent NPAS4 Expression and the Regulation of Gene Programs Underlying Plasticity inthe Central Nervous System, Jose Fernando Maya-VetencourtVolume 2013, Article ID 683909, 12 pages

Environment, Leptin Sensitivity, and Hypothalamic Plasticity, Marco Mainardi, Tommaso Pizzorusso,and Margherita MaffeiVolume 2013, Article ID 438072, 8 pages

System Consolidation of Spatial Memories in Mice: Effects of Enriched Environment, Joyce Bonaccorsi,Simona Cintoli, Rosa Mastrogiacomo, Sigrid Baldanzi, Chiara Braschi, Tommaso Pizzorusso,Maria Cristina Cenni, and Nicoletta BerardiVolume 2013, Article ID 956312, 12 pages

Gene Expression Patterns Underlying the Reinstatement of Plasticity in the Adult Visual System,Ettore Tiraboschi, Ramon Guirado, Dario Greco, Petri Auvinen, Jose Fernando Maya-Vetencourt,Lamberto Maffei, and Eero CastrenVolume 2013, Article ID 605079, 8 pages

Noninvasive Strategies to Promote Functional Recovery after Stroke, Alessio Faralli, Matteo Bigoni,Alessandro Mauro, Ferdinando Rossi, and Daniela CarulliVolume 2013, Article ID 854597, 16 pages

Case Study of Ecstatic Meditation: fMRI and EEG Evidence of Self-Stimulating a Reward System,Michael R. Hagerty, Julian Isaacs, Leigh Brasington, Larry Shupe, Eberhard E. Fetz, and Steven C. CramerVolume 2013, Article ID 653572, 12 pages

Quality and Timing of Stressors Differentially Impact on Brain Plasticity and Neuroendocrine-ImmuneFunction in Mice, Sara Capoccia, Alessandra Berry, Veronica Bellisario, Davide Vacirca, Elena Ortona,Enrico Alleva, and Francesca CirulliVolume 2013, Article ID 971817, 8 pages

Hindawi Publishing CorporationNeural PlasticityVolume 2013, Article ID 863970, 2 pageshttp://dx.doi.org/10.1155/2013/863970

EditorialNoninvasive Strategies to Optimise Brain Plasticity:From Basic Research to Clinical Perspectives

Alessandro Sale,1 Anthony J. Hannan,2 Lamberto Maffei,1 and Andrea Guzzetta3

1 Neuroscience Institute, National Research Council (CNR), 56124 Pisa, Italy2 Florey Institute of Neuroscience andMental Health, Melbourne Brain Centre, University of Melbourne, Parkville, VIC 3010, Australia3 SMILE Lab, Department of Developmental Neuroscience, IRCCS Stella Maris Scientific Institute, 56128 Calambrone (Pisa), Italy

Correspondence should be addressed to Alessandro Sale; [email protected]

Received 21 November 2013; Accepted 21 November 2013

Copyright © 2013 Alessandro Sale et al.This is an open access article distributed under the Creative CommonsAttribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Brain plasticity can be defined as the capacity of cerebral neu-rons and neural circuits to change, structurally and function-ally, in response to experience. This fundamental property isessential formaturation of sensory functions during develop-ment, for the adaptability of our behaviour to the environ-ment through learning and memory processes and for brainrepair in response to disease and trauma.

Given its relevance for primary brain processes, it is notsurprising that great effort is beingmade inmultiple laborato-ries to elaborate intervention procedures aimed at enhancingneural plasticity in the brain. In addition to their theoreticalrelevance, these studiesmay pave theway for novel paradigmsor therapeutic agents for rehabilitation and recovery fromnervous system injury. Among the possible experimentalapproaches that can be used to promote brain plasticity, ofgreat relevance are those based on noninvasive procedurescharacterised by their capability to boost the potential forplasticity retained by neural circuitries without being associ-atedwith dangerous side effects. Some paradigms appear par-ticularly worthy of interest, in light of their powerful impacton brain health, and include exposure to enriched environ-ments characterised by high levels of sensory, motor, andcognitive stimulation, behavioural interventions based on theenhancement of sensory stimuli (such as perceptual learn-ing), and dietary manipulations aimed at the optimisation ofcaloric intake and food balance.

One of the most exciting findings resulting fromthe application of these procedures is the demonstration,obtained primarily in the paradigmatic visual system butextending to other systems and functions, that the adult brain

is not “hard-wired” with immutable neuronal circuits but canbe pushed to unfold a high degree of plasticity even well pastthe end of the so-called “critical periods,” sensitive phasesduring early development when plasticity levels are particu-larly high. This special issue provides a collection of severalpapers addressing the impact of a number of noninvasiveprocedures on developmental and adult brain plasticity, witha focus on both animal models and human research.

J. Bonaccorsi et al. concentrated on system consolidation,a crucialmechanismmediated by the hippocampus and othermedial temporal lobe structures and underlying the preciserecall of already acquired memories.The authors contributeda very innovative research paper providing the first evidencethat exposure of adult mice to environmental enrichmentaffects the time-dependent process of spatial memory systemconsolidation, inducing an earlier recruitment of the medialprefrontal cortex and also the progressive activation of a dis-tributed cortical network that is not activated in mice rearedin standard housing conditions.

Antidepressant drugs such as the selective serotoninreuptake inhibitor (SSRI) fluoxetine (Prozac) have beenrecently shown to have a major impact on brain plasticity,qualifying as powerful enviromimetics, substances that canbe used tomimic the beneficial effects induced by exposure toenvironmental enrichment. In particular, a chronic treatmentwith fluoxetine has been previously shown to reopen forms ofjuvenile-like plasticity in the adult visual cortex of rodents.In this special issue, E. Tiraboschi et al. continued in thisestablished research field by using fluoxetine-induced plas-ticity in the adult rat visual cortex as an experimental model

2 Neural Plasticity

to investigate possible modulatory effects on gene expres-sion by means of microarrays and RT-PCR. They provideevidence that the combination of fluoxetine and monoculardeprivation (i.e., the closure of one eye) induces significantchanges in the expression of genes belonging to differentbiological classes, such as chromatin structure remodelling,transcription factors, extracellular matrix, and excitatory andinhibitory neurotransmission.

Closely related to the paper by E. Tiraboschi et al., J. F.Maya-Vetencourt provides a review article on the emergingrole in brain plasticity played by the recently discoveredneuronal-specific and activity-dependent transcription fac-tor NPAS4, which has been proved to be involved in asvarious processes as neural circuits’ reorganization after cere-bral ischemia and brain injury, amygdala and hippocampal-dependent memory, homeostatic plasticity, and neurogen-esis. The author provides a useful discussion of the linkbetween NPAS4 expression and the regulation of GABA-mediated inhibitory transmission and discusses how futurelines of research might concentrate on NPAS4 as a possiblemediator for the established effects of environmental enrich-ment and fluoxetine administration on adult visual cortexplasticity.

When looking at the effects of environmental and phar-macological manipulations, one critical factor to be consid-ered is the possibility that the selected intervention proce-dures might also induce unpredictable amounts of undesiredstress, which can neutralise their impact on brain andbehaviour. While the literature on the impact of acute andchronic stress is huge, one major challenge is to understandthe conditions under which the harmful effects of a stress-ful situation can be converted into a potential benefit forbrain plasticity. S. Capoccia et al. provide a research paperinvestigating the effects of stressors of different nature andlength on hippocampal plasticity, with the associated changesin the immune and neuroendocrine activation. The authorsdemonstrate that while prolonged stress in mice is associatedwith immunosuppression and lowering of brain-derivedneurotrophic factor (BDNF) levels, opposite changes areelicited by brief exposure to stressful stimuli. The results arediscussed in terms of possible hormetic effects set in motionbymild stress, resulting in the activation of a greater flexibilityfor resourcemanagement inmoderately challenging environ-mental conditions.

One fundamental source of modification of the extracel-lular and intracellular milieu is food intake and the ensu-ing modulation of energy metabolism. Dietary factors areincreasingly recognised as powerful regulators of neural plas-ticity, exerting their effects on the brain by affecting molec-ular events related to synaptic plasticity, neuronal signalling,and, ultimately, mental health. M. Mainardi et al. contributewith a timely review on this exciting matter, focusing on theliterature dealing with neural plasticity induced by environ-mental stimulation (e.g., environmental enrichment, physicalexercise, dietary restriction, and high-fat diet) on the arcuatenucleus of the hypothalamus, the primary sensor of plasmaticleptin levels.

The special issue also contains significant contributionscentred on the effects of environment on brain plasticity in

humans. E. Inguaggiato et al. surveyed the literature on theimpact of noninvasive rehabilitation strategies in childrenwith unilateral cerebral palsy, providing the first review onthis subject. The selected literature discussed by the authorsemployed totally nonpharmacological procedures such asconstraint-inducedmovement therapy, occupational therapy,motor training, and virtual reality exposure.

One of the most common and severely disabling neuraldiseases is stroke, a leading cause of permanent adult disabil-ity. Enhancing neural plasticity in patients with stroke mightconsiderably affect their functional output, boosting therecovery process by eliciting and facilitating the spontaneousreparative potential of the brain. In their review, F. Faralli et al.discuss how, and to what extent, noninvasive interventionstrategies such as mirror therapy, action observation, andmental practice affect poststroke recovery. The review nicelyintegrates preclinical studies with clinical evidence, bridgingthe translational gap and providing a list of possible molec-ular factors underlying the beneficial effects elicited byenvironmental stimulation.

An emerging and very attractive area of research inexperience-dependent neuroplasticity is meditation, whichappears to elicit plasticity processes affecting higher cognitivefunctions. The research paper by M. R. Hagerty et al. reportsan fMRI and EEG study of the brain of a trained meditatorin the course of ecstatic meditation during a Buddhistconcentration technique called jhana.The authors documentthe areas activated during this practice and relate them tothe subjective reports of emotions and psychophysical states.A striking finding is the activation of the dopamine/opioidreward system during meditation stages corresponding tosubjective reports of joy, particularly relevant if one considersthat it is achieved through a totally self-stimulating procedurebased on internal mental processes.

Our hope is that this special issue will serve to emphasizethe relevance of environment-based intervention strategiesin eliciting brain plasticity under both physiological andpathological conditions. This area of investigation will likelyemerge as one of the most successful in the fields of brainrepair, neurology, psychiatry, and mental health.

Alessandro SaleAnthony J. Hannan

Lamberto MaffeiAndrea Guzzetta


Review ArticleBrain Reorganization following Intervention in Children withCongenital Hemiplegia: A Systematic Review

E. Inguaggiato,1,2 G. Sgandurra,2 S. Perazza,2,3 A. Guzzetta,2 and G. Cioni2,4

1 Scuola Superiore Sant’Anna, Piazza Martiri della Liberta, I-56127 Pisa, Italy2 Department of Developmental Neuroscience, IRCCS Stella Maris Scientific Institute, Via dei Giacinti 2,Calambrone, I-56128 Pisa, Italy

3 Physical and Rehabilitation Medicine, University of Rome Tor Vergata, I-00173 Rome, Italy4Department of Clinical and Experimental Medicine, University of Pisa, I-56126 Pisa, Italy

Correspondence should be addressed to A. Guzzetta; [email protected]

Received 26 July 2013; Revised 29 October 2013; Accepted 30 October 2013

Academic Editor: Alessandro Sale

Copyright © 2013 E. Inguaggiato et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Noninvasive rehabilitation strategies for children with unilateral cerebral palsy are routinely used to improve handmotor function,activity, and participation. Nevertheless, the studies exploring their effects on brain structure and function are very scarce. Recently,structural neuroplasticity was demonstrated in adult poststroke patients, in response to neurorehabilitation. Our purpose is toreview current evidence on the effects of noninvasive intervention strategies on brain structure or function, in children withunilateral cerebral palsy. The main literature databases were searched up to October 2013. We included studies where the effectsof upper limb training were evaluated at neurofunctional and/or neurostructural levels. Only seven studies met our selectioncriteria; selected studies were case series, six using the intervention of the constraint-induced movement therapy (CIMT) andone used virtual reality therapy (VR). CIMT and VR seem to produce measurable neuroplastic changes in sensorimotor cortexassociated with enhancement of motor skills in the affected limb. However, the level of evidence is limited, due to methodologicalweaknesses and small sample sizes of available studies. Well-designed and larger experimental studies, in particular RCTs, areneeded to strengthen the generalizability of the findings and to better understand the mechanism of intervention-related brainplasticity in children with brain injury.

1. Introduction

Unilateral cerebral palsy (U-CP) is the most common typeof cerebral palsy (CP), with an incidence of 1 in 1000 live-births [1]. Typically, the upper limb (UL) is more involvedthan the lower, with impairments of spasticity, sensation, andreduced strength. Effective use of the arm and hand to reach,grasp, release, and manipulate objects is often compromised.Children with hemiplegia usually have the intellectual capac-ity to attend regular school; however, impaired arm functionrestricts their participation in educational, leisure, and latervocational roles [2].

U-CP can result from a wide variety of brain lesions,with respect to the timing of insults (acquired during thepre-, peri- or postnatal period), and the type of structuralpathology (brain malformations, periventricular lesions, and

corticosubcortical lesions) [3]. U-CP often leads to delays inmotor development or deconditioning of the affected limb,as individuals are inclined to functional compensation withthe intact limb rather than attempting to use the involvedlimb [4]; this may result in suppression of developmentof cortical representation of the affected limb, and it mayfurther inhibit its functional use [5, 6]. When the lesionoccurs at an early stage of development, either during theintrauterine life or soon after birth, the mechanisms ofplastic (re-)organization of the sensory motor system can bedifferent from those observed at later stages of development[7]. Primary motor control of the hemiplegic upper limbcan be eventually maintained within the spared tissue of theaffected hemisphere (ipsilesional reorganization), or it can bereorganized within the unaffected hemisphere, as a result ofthe complete withdraw of the crossing fibers from the affected

2 Neural Plasticity

hemisphere and the survival of the fast-conducting ipsilateralmotor projections from the unaffected one (contralesionalreorganization) [8]. The type of reorganization can be influ-enced by the size and site of damage, but it appears stronglyinfluenced also by the experience following damage, that is,by the complex interaction between residual motor outputfrom the affected hemisphere and somatosensory feedbackfrom the affected limb [9].

In general terms, adaptive plasticity of the central nervoussystem (CNS) refers to functional and structural changesin the brain, which are advantageous to offset or improvefunctions; the term denotes several capacities including theability to adapt to changes in the environment and to storeinformation in memory associated with learning [10]. Thereis abundant evidence that the structure of certain braincircuits can change in response to environmental stimuli [11].Recently, structural neuroplasticity has been demonstratedin response to neurorehabilitation intervention in adultpoststroke patients. Gauthier et al. [12] have shown in strokepatients treated with CIMT a significant increase in graymatter volume in several regions, including bilateral primarysensory and motor areas, both hippocampi, and anteriorsupplementary motor area contralateral to the motor deficit[12].

In children with U-CP, several types of interventionhave been used to improve abilities of the affected limb(e.g., neurodevelopmental treatment, neuromuscular electri-cal stimulation, constraint-induced movement therapy, etc.).Compared to adult poststroke research, a relatively smallnumber of studies investigated the effects of rehabilitation onbrain reorganization. The purpose of this study has been toevaluate current evidence on brain reorganization in childrenwith U-CP following noninvasive intervention strategies.

2. Methods

Articles were identified through comprehensive searches ofcomputerized bibliographic databases: PubMed, MedLine(1973 to October 2013), CINAHL (Cumulative Index to Nurs-ing and Allied Health Literature) (up from 1994 to October2013), Web of Science (1992 to October 2013), and ERIC (pre-1966 to October 2013). We also searched for reviews on thistopic on the Cochrane Central Register of Controlled Trials,with no result.

The search explored Medical Subject Headings (MeSH)terms and text words:

(1) “cerebral palsy” or “hemiplegia”,(2) “child” or “adolescent” or “infant”,(3) “therapy” or “training” or “intervention”,(4) “MRI” or “fMRI” or “EEG” or “TMS” or “PET” or

“MEG” or “reorganization”.

Selection Criteria. To be included in this systematic review,studies had to meet the following criteria.

(1) Participants were diagnosed with U-CP.(2) Interventions to improve outcome were noninvasive

and did not include drugs.

(3) Outcomes included functional activities and evidenceof brain reorganization through neurophysiologicalexperiments, carried out before and after the inter-vention.

Studies were excluded if they

(1) reported only clinical measures as outcomes;(2) were case reports;(3) were not published in English.

The initial search yield was reviewed by only one revieweron the basis of title and abstract. All the studies emergedfrom the search focused on upper limb (UL) intervention.The search strategy allowed to identify 12 articles thatmet ourinclusion/exclusion criteria. The full-text articles were exam-ined by 3 reviewers, and the eligibility for study inclusionwas assessed independently; in case of mismatched opinionbetween the 3 reviewers, the eligibility of the study wasdiscussed together and consensuswas reached. Following oursearch in the different databases, only 5 eligible studies wereidentified while two additional ones were selected withintheir reference lists. The final analysis included 7 studies. Thegeneral purpose of the studies was to evaluate the effects ofnoninvasive rehabilitation strategies on brain reorganizationand on functional improvement of affected upper limb (UL)in unilateral cerebral palsy. In Figure 1, flow chart describesstudy selection and reasons for exclusion.

3. Results

3.1. StudyDesigns andParticipants. Selected studieswere caseseries; no controlled studies were found.

We found seven trials specifically targeted on childrenwith unilateral cerebral palsy; only in one case a participanthad bilateral impairment with right arm sparing [13].The agerange of participants was between 2.1 and 7.6 years in onestudy [14], between 7 and 14 years in two studies [13, 15],between 13 and 15 years in another study [16], and between10 and 30 years in the others [17–19]. Some studies wereperformed by the same research group and some subjects par-ticipated in more than one study [17–19]. Table 1 summarizesthe characteristics of the population for each study.

3.2. Type of Interventions. The most frequently proposedintervention, in six of the seven studies, was the constraint-induced movement therapy (CIMT); CIMT was used inassociation with neurodevelopmental treatment (NDT) [15],in association with occupational therapy (OT) [13], in associ-ation with intensive motor training [14] or during a trainingcamp in associationwith individual and peer groups activities[17–19]. Standard CIMT for children involves a restraintworn on the non-affected upper limb for 90% of wakinghours and 6hours/day of intensive intervention using shapingtechniques and massed practice typically over a 2-weekperiod [20]. The standard CIMT model has been adapted tobe less intensive (<6 hours/day) due to concerns that youngchildren are unable to participate in such intensive therapyregimen [21]. The remaining study employed an innovative

Neural Plasticity 3

7 studies were excluded:

Articles screened by title or abstract

n = 356

Articles identified according to our inclusion/exclusion criteria

n = 12

Articles excluded due to inappropriate population, outcome measures,

purpose, etc.n = 345

Full-text articles were examined by 3 reviewers independently

7 studies were included in final analysis

Two articles were selected within reference lists of eligible studies

[Junger et al.; 2007, Walther et al.; 2009]

Articles identified on Medline n = 147

Articles identified on CINAHL n = 29

Articles identified on web of science n = 30

Articles identifiedon ERIC n = 0

Articles identifiedon PubMed n = 150

2 = absent intervention; 1 = review;1 = absent neurophysiological

outcomes; 1 = used botulinum toxin;2 = case report

Figure 1: Flow chart of search strategy and selection process.

treatment strategy: virtual reality (VR) [16], a virtual envi-ronment system that uses new technologies to make thepatient perceptions similar to those coming from real-lifeactivities. In none of the studies children received botulinuminjections or upper limb surgery for the affected upper limbin the 6 months prior to intervention. Table 2 describes thecharacteristics of each UL intervention.

3.3. Outcome Measures. Selected studies were case series;no controlled studies were found. Studies aimed to evaluatethe effects of noninvasive intervention on (i) functionalityof UL, through scales and/or questionnaires and (ii) brainreorganization, through neuroimaging and neurophysiologi-cal techniques (i.e., MRI, fMRI, TMS, MEG). In three of thestudies [17–19], the different patterns of corticospinal reorga-nization (ipsilesional versus contralesional) were determinedby using TMS. Outcome measures were applied both beforeand after the intervention; in a few studies, assessments werealso recorded during followup [15, 18].

3.3.1. Upper Limb Function. The effects of noninvasive inter-vention on the functionality of the hemiplegic upper limbwere monitored using different functional measures, scales,or questionnaires. The type of clinical assessments varied

among studies; we therefore grouped the functional motoroutcomes in 3 categories, according to the dimensions ofthe International Classification of Functioning. Disabilityand Health (ICF-CY): (a) body functions and structures,(b) activities, and (c) participation [22]. (Table 1s sum-marizes clinical assessment and corresponding results; seeTable 1s in the Supplementary Material available online athttp://dx.doi.org/10.1155/2013/356275). Most of the studiesinvestigated functional motor outcomes according to at leastone of the dimensions of the International Classificationof Functioning, Disability and Health (ICF). The most fre-quently used outcomes measures were WMFT [17–19] in 3/7papers and P-MAL [13, 14, 18] in 3/7 papers.

3.3.2. Brain Reorganization. To evaluate the effects of theintervention on brain reorganization, 6 studies used func-tional magnetic resonance imaging (fMRI) [13, 15–19]. AllMRI experiments were performed on 1.5T scanner, butthe fMRI procedure and the tasks performed during theexamination were different among studies. Half of the sixstudies used, as fMRI task, active and passive movementsof the paretic and the nonparetic hands, while the otherhalf only performed an active task on the paretic hand. Onestudy combined fMRI with Transcranial Magnetic Stimu-lation (TMS) [18] and another combined fMRi, TMS, and

4 Neural Plasticity

Table 1: Population and study design.

Study Design Patients M : F Age Diagnosis: congenitalU-CP Lesion/etiopathogenesis Type of

reorganization§

[13] Case series 5∗ n/a 7–13 ys 4R-CP, 1bilateralCP∗ n/a n/a

[14] Case series 10 6 : 4 2.1–7.6(3.3 ± 1.6) 8R-CP; 2L-CP 3L-FP; 4L-PV; 1R-F;

1L-FT; 1R-FP n/a

[15] Case series 10 4 : 6 7–14 ys(11 ± 2.5) 4L-CP; 6R-CP

3 malformative,3 prenatal, 1 connatal,2 early acquired, 1 n/a

n/a

[16] Case series 3 2 : 1 13–15 ys R-CP2 patients: perinatal

stroke,1 patient: IVH

n/a

[17] Case series 10 5 : 5 10–30 ys(median 14 ys) 6R-CP; 4L-CP

unilateralcortical-subcortical

infarction in the MCAterritory

7/10 ipsilesional3/10 mixed

[18] Case series 7 3 : 4 10–30 ys(median 16 ys) 5R-CP; 2L-CP



ipsilesional

[19]Case series dividedinto: contralesional,

ipsilesional

169/167/16

8 : 85 : 43 : 4

10–31 ys11–31 ys10–30 ys

6L-CP, 10R-CP4L-CP, 5R-CP2L-CP, 5R-CP



9/16 contralesional7/16 ipsilesional

∗A participant had bilateral involvement with right arm sparing; §assessed by Transcranial Magnetic Stimulation (TMS). Abbreviations: M: male; F: female; ys:years; L: left; R: right; U-CP: unilateral cerebral palsy, IVH: intraventricular hemorrhage; MCA: middle cerebral artery; FP: frontoparietal; PV: periventricular;F: frontal; TP: temporal-parietal; CIMT: constraint-induced movement therapy; VR: virtual reality; NDT: neurodevelopmental treatment; OT: occupationaltherapy; n/a: not available.

Table 2: Characteristics of the UL intervention programs.

Study Treatment Duration Frequency Intensityper day Environment Activities Restraining device or therapy

system

[13] Modify CIMT +OT 3weeks Weekly n/a In home

Bloorview Kidsrehabilitationtherapy manual

3 weeks continuous casting of theaffected arm and hand

[14] CIMT + intensivemotor training 15 days∗ Weekdays 5 hrs N/a Shaping technique Less-affected arm is continuously

restrained in a long arm cast

[15] modify CIMT +NDT 2weeks Weekdays 4 hrs

Outpatientclinic, home,playgroup

Chosencollaboratively

between child andtherapists

Removable cast on nonaffectedarm for 90% of the waking hrsincluded weekend

[16] VR 2months Weekdays 30min In home2 games: “sliders”,“chase away abutterfly”

5DT5 Ultra Glove + Play Station 3game console

[17]CIMT +

individual/peergroup activities

12 days Daily n/a Training campIndividual (2 hrs)and peer group

activities

Tailored glove fortified on palmerside and fingers (wearing time:10 hrs/day)

[18]CIMT +


12 days Daily n/a Training campIndividual (2 hrs)and peer group

activities


[19]CIMT +


12 days Daily 10 hrs Training campIndividual (2 hrs)and peer groupactivities (8 hrs)


∗On the last 2 days of treatment, the cast is removed and training is focused on bilateral activities. Abbreviations: CIMT: constraint-inducedmovement therapy;VR: virtual reality; NDT: neurodevelopmental treatment; OT: occupational therapy; n/a: not available; min: minutes; hrs: hours.

Neural Plasticity 5

Magnetoencephalography (MEG) [19]. TMS procedure wasthe same in the two studies; MEPs were recorded from theflexor pollicis brevis muscle from paretic and non-paretichands by surface electromyography for amplitude and globaltransmission time [18, 19]. For MEG, somatosensory evokedmagnetic fields (SEFs) were elicited by tactile stimulationof the paretic and non-paretic hands. Authors analyzed thetactile evoked magnetic field of the early SEF (amplitudeand latency) [19]. Only one study evaluated the effectsof reorganization trough voxel-based morphometry (VBM)analysis to determine gray matter change [14].

3.4. Findings. Details of the findings are reported in Table 3.To explore the effects of intervention on brain reorganization,neurofunctional techniques were used in 6/7 studies (fMRIin 6 studies, TMS in 2, MEG in 1), while a neurostructuraltechnique (VBM) was used in 1/7 studies.

3.4.1. Effects on the Hemisphere Contralateral to the PH(Affected orMost AffectedHemisphere). Themain neurofunc-tional effect upon the (most) affected hemisphere, observedafter intervention, consisted of an enlargement of M1 orM1/S1 activation during active motor tasks of the paretichand, demonstrated either at the single subject level [17] orat the group level [13, 15, 18, 19]. Less consistent findingswere observed on fMRI during passive motor tasks of theparetic hand with enlargement of M1/S1 activation at thesingle subject level in one study [17], not confirmed at agroup level in another study. Tasks performed by the non-paretic hand, both active and passive, did not seem to affectbrain reorganization within the affected hemisphere. Thisgeneral fMRI pattern was confirmed also when stratifyingsubjects according to motor reorganization (i.e., ipsilesionalversus contralesional) [19]. In studies usingTMS, a significantincrease of M1-MEPs amplitude was observed, following theTMS stimulation of the affected hemisphere [18].This findingwas clearly not observed in subjects with contralesional reor-ganization of motor function, as no MEPs could be elicitedin these subjects by stimulation of the affected hemisphere[19]. In the study usingMEG, increased amplitude of the SEFswas observed in the affected hemisphere following tactilefinger stimulation of the paretic hand, irrespective of the typeof motor reorganization, while a reduction in SEFs latencywas only observed in subjects with ipsilesional reorganization[19].

The only study exploring neurostructural changes,through VBM analysis [14], found an increased volume ofM1/S1 in the affected hemisphere, together with an increasedvolume of the hippocampus.

3.4.2. Effects on the Hemisphere Ipsilateral to the PH (Non-Affected or Least Affected Hemisphere). No clear neurofunc-tional effects were observed upon the unaffected (or leastaffected) hemisphere. In a minority of cases, changes at thesingle subject level were observed on fMRI, at the level ofM1/S1 or the Cerebellum [17]. OnTMS, a small but significantdecrease of M1-MEPs amplitude was observed, followingTMS stimulation of the unaffected hemisphere, limited to

those subjects with a contralesional reorganization [19]. Noeffects were observed using MEG, with the exception ofa reduction of SEFs latency in the unaffected hemispherefollowing tactile finger stimulation of the non-paretic hand,limited to the subgroup of subjects with ipsilesional reorga-nization [19].

The only study exploring neurostructural changes,through VBM analysis, found an increased volume of M1in the unaffected hemisphere, together with an increasedvolume of the hippocampus [14].

3.4.3. Correlation of Brain Reorganization with FunctionalImprovement. The correlation between functional motorimprovement in the upper limb and degree of neuroplasticchanges was explored in 4/7 studies and significant cor-relations were found. In 3 studies CIMT was used andthe training-related improvements were positively correlatedwith the extent of the area of activation [15], the lateralityindex [13] and volume increase at VBM [14]. In the study onVR, a correlation between motor function and fMRI signalduring active motor tasks was found [16].

4. Discussion

Despite the high number of studies exploring the functionaleffects of neurorehabilitation in children with unilateralcerebral palsy, relatively little is known on the neurobiologicalunderpinnings of such effects. The main common findingreported in the reviewed studies is the enlargement ofthe primary hand motor area contralateral to the paretichand, following intervention. This was valid across differentstudies, both for CIMT [13–15, 17–19] and VR trainings [16],using various hand motor tasks such as finger tapping [15],hand opening/closing [16] and rubber ball pressing [17–19]. Contralateral primary motor and sensory cortex werethe most frequently involved but increased activation couldbe also found in the supplementary motor area [18], thepremotor cortex [17] and the cerebellum [16, 17]. More ingeneral, the effect results into a shift in the laterality index dueto the increased activity in the (most) affected hemisphereafter therapy, not counterbalanced by a similar effect in theunaffected (or least affected) one.

The effect of training on hand passive motor task activa-tion was less clear. Of the three papers exploring this question[17–19], one showed significant enlargements in about half ofthe tested subjects [17], the second one found no significantchanges [18], while the third one found changes only in thesubgroup of subjects with contralesional reorganization. Thethree studies however used different statistical approaches(single subject versus group analysis), making the threestudies poorly comparable and potentially less conflicting.

In two of the six studies [14, 18], effects of training wereexplored with different means other than fMRI.Walther et al.[18] used TMS to determine the changes in corticospinalexcitability following CIMT training, and recorded increasedamplitude MEPs in the paretic hand from the contralateralprimary motor cortex. No effect was observed for the non-paretic hand. Sterling et al. [14] explored the effects of training

6 Neural Plasticity

Table 3: Neuroimaging and neurophysiological outcome measures and results.

CIMTFunctional magnetic resonanceimaging PH N-PH Notes

Active movements fMRI task

Four-finger/wrist extension/flexion[13]

2/4 LI shift to contralateralhemisphere, 2/4 reduced LI(group stat; 𝑛 = 4)

—

Finger tapping [15] 6/7 (M1c) ↑ area of activation2/7 (M1c) ↑ signal (1–3%) —

3/10 were excluded due toartifacts (2/10) orclaustrophobia.fMRI task was tested on 5controls who showed M1cactivation.

Rubber ball press [17] 1/3 (M1S1c + M1S1i + CBMi +PMC) ↑ area of activation

4/10 M1S1c ↑ area of activation1/10 M1S1i ↑ area of activation

3/10 were excluded for PH taskdue to movement artifacts.

Rubber ball press [18] (M1S1c + SMA) ↑ area ofactivation (group stat; 𝑛 = 5) No changes (group stat; 𝑛 = 5) 2/7 were excluded for inability

to perform the task.

Rubber ball press [19]Ipsilesional group:(M1S1c + SMA) ↑ area ofactivation (group stat; 𝑛 = 5)

No changes (group stat; 𝑛 = 5) 2/7 were excluded for inabilityto perform the task.

Contralesional group(M1S1c + CBMi/c) ↑ area ofactivation(M1i) ↓ activation(group stat; 𝑛 = 6)

No changes (group stat; 𝑛 = 6) 3/9 were excluded formovement artefacts.

Passive movements

Flexion/extension at themetacarpophalangeal of fingersII–V of the patient’s hand [17]

4/8 (M1S1c) ↑ area of activation1/8 (M1S1i) ↑ area of activation2/8 (IHF) ↑ area of activation1/8 (CMBi) ↑ area of activation4/8 no changes observed

2/10 M1S1c ↑ area of activation1/10 IHF ↑ area of activation

2/10 were excluded for PH taskdue to movement artifacts.


No changes (group stat; 𝑛 = 7) No changes (group stat; 𝑛 = 7)


Ipsilesional group: no changes(group stat; 𝑛 = 7)

Ipsilesional group: no changes(group stat; 𝑛 = 7)

Contralesional groupParietal operculumc + M2S2c ↓activation(group stat; 𝑛 = 9)

Contralesional groupM1S1c ↓ activation(group stat; 𝑛 = 9)

Voxel-based morphometry Posttreatment—pretreatment Pretreatment—baseline Notes

VBM [14] (M1S1c + M1i + Hippocampi) ↑volume (group stat; 𝑛 = 10)

No changes (group stat;𝑛 = 10)

Transcranial magnetic stimulation PH N-PH Notes

TMS [18] (M1-MEPs) ↑ amplitude(group stat; 𝑛 = 7)

No changes(group stat; 𝑛 = 7)

TMS [19] amplitude Ipsilesional group:(M1-MEPs) ↑ amplitude(group stat; 𝑛 = 7)

Ipsilesional group:No changes(group stat; 𝑛 = 7)

Contralesional group:(M1-MEPs) ↓ amplitude(group stat; 𝑛 = 9)

Contralesional group:(M1-MEPs) ↓ amplitude(group stat; 𝑛 = 9)

TMS [19] conduction time

Ipsilesional group:No changes (group stat; 𝑛 = 7)Contralesional group:No changes (group stat; 𝑛 = 9)

Ipsilesional group:No changes (group stat; 𝑛 = 7)Contralesional group:No changes (group stat; 𝑛 = 9)

Neural Plasticity 7

Table 3: Continued.

CIMTMagnetoencephalography PH N-PH Notes

MEG [19] latency

Ipsilesional group:↓ early-SEF latency (group stat;𝑛 = 7)Contralesional group:No changes in early-SEF latency(group stat; 𝑛 = 8)

Ipsilesional group:↓ early-SEF latency (groupstat; 𝑛 = 7)Contralesional group:No changes in early-SEFlatency.(group stat; 𝑛 = 8)

1/9 was excluded due to strongmagnetic artefacts.

MEG [19] amplitude

Ipsilesional group:↑ early-SEF amplitude (groupstat; 𝑛 = 7)Contralesional group:↑ early SEF amplitude(group stat; 𝑛 = 8)

Ipsilesional group:≈early-SEF amplitude (groupstat; 𝑛 = 7)Contralesional group:No changes SEF amplitude(group stat; 𝑛 = 8)

1/9 was excluded due to strongmagnetic artefacts.

VRFunctional magnetic resonanceimaging PH N-PH Notes

Active movements fMRI task

Hand open/close [16] 2/3 (M1c) ↑ area of activation2/3 (CBM) ↑ area of activation — Training dose was variable in

the 3 cases.Abbreviations: PH: paretic hand; N-PH: nonparetic hand; fMRI: functional magnetic resonance imaging; VBM: voxel-based morphometry; TMS: transcranialmagnetic stimulation, MEG: magnetoencephalography; M1: primary motor cortex; FP: frontoparietal; M1S1: primary sensory motor cortex, M2S2c secondarysensory motor cortex c/i: indicate contralateral/ipsilateral, CBM: cerebellum, IHF interhemispheric fissure (including cingulate motor area supplementarymotor area), PMC: premotor cortex; LI: lateral index; LI is calculated [(contralateral − ipsilateral)/(contralateral + ipsilateral)]; SMA: supplementary motorarea, MEPs: motor evoked potentials; SEF: somatosensory evoked potentials.

on a structural level by using VBM analysis. This is alsothe only paper with an actual control condition consistingof a same-length interval pretraining used to explore brainchanges unrelated to intervention. It is of interest that whileno changes were observed from baseline to pretreatment, asignificant volume increase was observed at a group levelposttreatment in the primary motor cortex bilaterally, in thecontralateral primary sensory area and in both hippocampi.

Not surprisingly, a key factor influencing treatment-related brain neuroplasticity appeared to be the type ofreorganization of the corticospinal tract (i.e., ipsilesionalor contralesional). Type of reorganization was taken intoaccount in 3/7 studies; these studies came from the sameresearch group, with the most recent one [19] confirming andexpanding the results of the previous two [17, 18]. Althoughthe overall small figures do not allow for definite conclusions,there appears to be enough evidence supporting the existenceof two types of treatment-related neuroplasticity with themain hallmark of an increase in M1 excitability in subjectswith ipsilesional reorganization and of a decrease in M1excitability in subjects with contralesional reorganization.

Positive effects of training on handmotor function, in theselected studies, were almost invariably reported, althoughthe outcome measures used were very heterogeneous. Whencorrelating functional improvements with the amount ofplastic brain reorganization, significant results were generallyobserved after intervention, including enlarged area of M1S1activation in fMRI, increased M1-MEPs amplitude fromstimulation of the affected hemisphere, and increased M1S1

brain volumes on VBM. However, data are too scatteredand heterogeneous to allow for definite conclusions on thepossible correlations between neurobiological changes andfunctional improvements.

Themain limitation of the findings of this review is relatedto the number and type of papers found in our systematicsearch. Studies included in this review consist of quasi-experimental or descriptive pre-post designs. Their level ofevidence, based upon a modified Sackett score [23] adaptedto include PEDro ratings, is between 2b for CIMT studiesand 5 for VR. It is of great interest that none of the studiesselectedwas a randomized controlled study. Although severalRCTs have been performed comparing different trainings inchildren with unilateral cerebral palsy, some ethical prob-lems might have hindered the possibility of testing controlsubjects with relatively invasive techniques such as TMS.Nevertheless, since MRI, MEG, and EEG techniques, whenused without sedation, can be considered noninvasive, thereis no obvious reason why RCTs have not yet been performedusing these methods. Lack of RCTs might be more simplyjustified by this field of research being relatively new and thistype of study design being more complex.

In summary, noninvasive rehabilitation strategies seem toproduce measurable neuroplastic changes in sensory motorcortex associated with enhancement of motor skills in theaffected limb. This conclusion is however largely restricteddue to the strong limitations of the reviewed studies, themostrelevant of which concerns their methodological characteris-tics. It is also important to underline that the selected studies

8 Neural Plasticity

only investigated the effects of two types of intervention,namely, CIMT and VR, making therefore our conclusionsnot applicable to other approaches. For the same reason,this review cannot provide any contribution to the definitionof the type of intervention that should be recommendedin children with U-CP. Well-designed experimental studieswith larger sample sizes should be carried out to strengthenthe generalizability of these preliminary findings. Moreover,for further studies it would be important to investigatethe clinical outcomes according to the dimensions of ICFwith the best measures created for children with hemiplegiaconsidering psychometric properties. More researches, andin particular RCT studies, are needed to better understand themechanism of brain plasticity in children with brain injuryand to inform and fine-tune current or novel rehabilitationstrategies in children with cerebral palsy.

Conflict of Interests

The authors declare that there is no conflict of interests.

References

[1] http://www.cerebralpalsysource.com/.[2] L. Sakzewski, J. Ziviani, and R. Boyd, “Systematic review and

meta-analysis of therapeutic management of upper-limb dys-function in childrenwith congenital hemiplegia,”Pediatrics, vol.123, no. 6, pp. e1111–e1122, 2009.

[3] M. Staudt,W.Grodd, C.Gerloff,M. Erb, J. Stitz, and I. Krageloh-Mann, “Two types of ipsilateral reorganization in congenitalhemiparesis: a TMS and fMRI study,” Brain, vol. 125, no. 10, pp.2222–2237, 2002.

[4] J. Held, “Recovery of function after brain damage: theoreticalimplications for therapeutic intervention,” inMovement Science:Foundations for Physical Therapy in Rehabilitation, J. H. Carrand R. B. Shepherd, Eds., pp. 189–211, Aspen, Oxford, UK, 2ndedition, 2000.

[5] P. Cicinelli, R. Traversa, and P. M. Rossini, “Post-stroke reor-ganization of brain motor output to the hand: a 2–4 monthfollow-up with focal magnetic transcranial stimulation,” Elec-troencephalography and Clinical Neurophysiology, vol. 105, no.6, pp. 438–450, 1997.

[6] J. Liepert, H. Bauder, W. H. R. Miltner, E. Taub, and C. Weiller,“Treatment-induced cortical reorganization after stroke inhumans,” Stroke, vol. 31, no. 6, pp. 1210–1216, 2000.

[7] J. A. Eyre, “Corticospinal tract development and its plasticityafter perinatal injury,” Neuroscience and Biobehavioral Reviews,vol. 31, no. 8, pp. 1136–1149, 2007.

[8] M. Staudt, C. Gerloff, W. Grodd, H. Holthausen, G. Nie-mann, and I. Krageloh-Mann, “Reorganization in congenitalhemiparesis acquired at different gestational ages,” Annals ofNeurology, vol. 56, no. 6, pp. 854–863, 2004.

[9] A. Guzzetta, P. Bonanni, L. Biagi et al., “Reorganisation ofthe somatosensory system after early brain damage,” ClinicalNeurophysiology, vol. 118, no. 5, pp. 1110–1121, 2007.

[10] M.V. Johnston, “Clinical disorders of brain plasticity,”Brain andDevelopment, vol. 26, no. 2, pp. 73–80, 2004.

[11] R. Chen, L. G. Cohen, and M. Hallett, “Nervous systemreorganization following injury,”Neuroscience, vol. 111, no. 4, pp.761–773, 2002.

[12] L. V. Gauthier, E. Taub, C. Perkins, M. Ortmann, V. W. Mark,and G. Uswatte, “Remodeling the brain: plastic structural brainchanges produced by different motor therapies after stroke,”Stroke, vol. 39, no. 5, pp. 1520–1525, 2008.

[13] T. L. Sutcliffe, W. J. Logan, and D. L. Fehlings, “Pedi-atric constraint-induced movement therapy is associated withincreased contralateral cortical activity on functional magneticresonance imaging,” Journal of Child Neurology, vol. 24, no. 10,pp. 1230–1235, 2009.

[14] C. Sterling, E. Taub, D. Davis et al., “Structural neuroplasticchange after constraint-induced movement therapy in childrenwith cerebral palsy,” Pediatrics, vol. 131, no. 5, pp. e1664–e1669,2013.

[15] S. M. Cope, X. C. Liu, M. D. Verber, C. Cayo, S. Rao, andJ. C. Tassone, “Upper limb function and brain reorganizationafter constraint-induced movement therapy in children withhemiplegia,” Developmental Neurorehabilitation, vol. 13, no. 1,pp. 19–30, 2010.

[16] M. R. Golomb, B. C. McDonald, S. J. Warden et al., “In-homevirtual reality videogame telerehabilitation in adolescents withhemiplegic cerebral palsy,” Archives of Physical Medicine andRehabilitation, vol. 91, no. 1, pp. 1.e1–8.e1, 2010.

[17] H. Juenger, M. Linder-Lucht, M. Walther, S. Berweck, V.Mall, and M. Staudt, “Cortical neuromodulation by constraint-induced movement therapy in congenital hemiparesis: an fMRIstudy,” Neuropediatrics, vol. 38, no. 3, pp. 130–136, 2007.

[18] M. Walther, H. Juenger, N. Kuhnke et al., “Motor cortexplasticity in ischemic perinatal stroke: a transcranial magneticstimulation and functionalMRI study,” Pediatric Neurology, vol.41, no. 3, pp. 171–178, 2009.

[19] H. Juenger, N. Kuhnke, C. Braun et al., “Two types of exercise-induced neuroplasticity in congenital hemiparesis: a tran-scranial magnetic stimulation, functional MRI, and magne-toencephalography study,” Developmental Medicine and ChildNeurology, vol. 55, no. 10, pp. 941–951, 2013.

[20] E. Taub, A. Griffin, J. Nick, K. Gammons, G. Uswatte, and C.R. Law, “Pediatric CI therapy for stroke-induced hemiparesis inyoung children,”Developmental Neurorehabilitation, vol. 10, no.1, pp. 3–18, 2007.

[21] A. Eliasson, L. Krumlinde-Sundholm, K. Shaw, and C. Wang,“Effects of oncstraint-induced movement therapy in youngchildren with hemiplegic cerebral palsy: an adapted model,”Developmental Medicine and Child Neurology, vol. 47, no. 4, pp.266–275, 2005.

[22] M. Leonardi and A. Martinuzzi, “ICF and ICF-CY for an inno-vative holistic approach to persons with chronic conditions,”Disability and Rehabilitation, vol. 31, supplement 1, pp. S83–S87,2009.

[23] D. L. Sackett,W. S. Richardson,W.Rosenberg, andR. B.Haynes,Eds., Evidence-BasedMedicine: How to Practice and Teach EBM,Churchill Livingstone, New York, NY, USA, 2nd edition, 2000.


Review ArticleActivity-Dependent NPAS4 Expression andthe Regulation of Gene Programs Underlying Plasticity inthe Central Nervous System

José Fernando Maya-Vetencourt1,2

1 Centre for Nanotechnology Innovation, Italian Institute of Technology, Piazza San Silvestro 12, 56127 Pisa, Italy2 Centre for Neuroscience and Cognitive Systems, Italian Institute of Technology, Corso Bettini 31, 38068 Rovereto, Italy

Correspondence should be addressed to Jose Fernando Maya-Vetencourt; [email protected]

Received 5 May 2013; Accepted 9 July 2013


Copyright © 2013 Jose Fernando Maya-Vetencourt. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

The capability of the brain to change functionally in response to sensory experience ismost active during early stages of developmentbut it decreases later in life when major alterations of neuronal network structures no longer take place in response to experience.This view has been recently challenged by experimental strategies based on the enhancement of environmental stimulation levels,genetic manipulations, and pharmacological treatments, which all have demonstrated that the adult brain retains a degree ofplasticity that allows for a rewiring of neuronal circuitries over the entire life course. A hot spot in the field of neuronal plasticitycentres on gene programs that underlie plastic phenomena in adulthood.Here, I discuss the role of the recently discovered neuronal-specific and activity-dependent transcription factor NPAS4 as a critical mediator of plasticity in the nervous system. A betterunderstanding of howmodifications in the connectivity of neuronal networks occurmay shed light on the treatment of pathologicalconditions such as brain damage or disease in adult life, some of which were once considered untreatable.

1. Introduction

The interaction between genetic and environmental factorslies behind the neuronal representation of sensory stimuli inthe nervous system. The environment largely modifies brainstructure and function through mechanisms of neuronalplasticity. Sensory experience actually drives the refinementof immature neural circuitries into organized patterns ofsynaptic connectivity that subserve adult brain functions [1].

Environmental influences play a key role in sculptingthe central nervous system architecture during early life,when neural circuitries are highly sensitive to experience(reviewed in [2, 3]). This seems to be a period of time (socalled critical period) in which an individual acquires anindelible memory of relevant stimuli in the environment,which ensures proper development of sensory functionsand/or behaviours. An emerging view in the field of plasticityis that the effects caused by early developmental experience

in the remodeling of neural networks seem to be activelypreserved by the late appearance of structural and functionalfactors that restrict plasticity over the time course. Thisfeature seems to be of relevance in terms of adaptive functionsbut determines diminished plasticity in the adult brain, whichin turn severely restricts the functional reorganization ofthe nervous system thus posing a limit for feasible clinicalinterventions after brain injury or disease in humans (forreview see [4, 5]).

The capacity of neural circuitries to change in response tosensory experience is of high relevance in fields of neuronalrehabilitation and brain repair. This is clear, for instance,in the case of stroke, which is a major cause of long-termdisability for which there is currently no clinical treatment.Reactivating juvenile-like plasticity in the adult brain wouldbe beneficial in poststroke patients, whose recovery dependson a reorganization of neuronal networks in adult life(reviewed in [6]).

2 Neural Plasticity

How does experience modify synaptic circuitries inthe brain? Experience-dependent modifications of brainfunctions depend, at least partially, on gene expressionpatterns that have evolved to meet specific environmentaldemands. The structure and function of the BDNF geneare a compelling example of physiological mechanisms bywhich experience-dependent plasticity is achieved. Sincepromoter areas of the BDNF gene are differentially regulatedby distinct neurotransmitter systems, the levels of whichvary in response to environmental influences (for reviewsee [7]), BDNF protein synthesis in the brain is regulatedby experience in a spatiotemporal-dependent manner [8,9]. This neurotrophin drives different forms of synapticplasticity and therefore epitomizes how the nervous systemmediates fast adaptive responses to changing environmentalconditions.

A hot spot in the neuroscience field is the identificationof physiological mechanisms associated with experiencethat trigger alterations in the pattern of DNA methylationand/or posttranslational modifications of histones that inturn control the expression of genes underlying phenomenaof plasticity in the brain (reviewed by [10, 11]). Indeed,epigenetic mechanisms that exert a long-lasting control ofgene expression by modifying chromatin structure ratherthan changing the DNA sequence itself have been recog-nized as experience-dependent mechanisms that regulate theoccurrence of brain plasticity ([12–15], reviewed in [16–18]).

Transcriptional mechanisms that are mediated by imme-diate early genes (IEGs) and lie behind the occurrence ofplasticity in the nervous system have also been subject ofrecent studies ([19], for review see [20, 21]). It is becomingincreasingly clear that experience-dependent plasticity isachieved when neuronal activity triggers intracellular signalpathways that promote the induction of IEGs (e.g., c-Fos, c-Jun, CREB, and Zif268) that in turn control the expressionof downstream targets, the products of which then work viathe activation of structural and functional mechanisms thateventually modify the strength of synaptic connections so asto change the computational properties of neural networks inthe brain (reviewed in [20–22]).

In this review, I shall focus on the role of the recentlydiscovered neuronal-specific transcription factor NPAS4 asa key regulator of brain plasticity and cognition. It has beensuggested thatNPAS4may be involved in phenomena of plas-ticity after local [23, 24] and global [24, 25] cerebral ischemia,seizures [26, 27], and brain injury [25, 27, 28]. More recently,it has been reported that theNPAS4 transcription factor playsa key role in mediating a transcriptional program underlyingamygdala-dependent [29] and hippocampal-dependent [30]processes ofmemory, social, and cognitive functions [31].Theupregulation of NPAS4 in the striate nucleus after chronicamphetamine administration [32], which is a pharmacolog-ical model of plasticity with high relevance for mechanismsof addiction [33, 34], has also been described. Moreover,impairments of neurogenesis and deficits in fear [35] andspatial memories [36] by social isolation and chronic stressseem to be associated with the transcriptional suppressionof the NPAS4 gene [37], suggesting a central role for thistranscription factor as a mediator of plasticity. Here, I will

highlight recent advances that have brought to light someof the structural and functional mechanisms underlying theaction of NPAS4 in experience-dependent plasticity. I shallalso cover novel findings onNPAS4-mediated gene programsthat lie behind phenomena of cortical plasticity caused byeither pharmacological treatments or experimental strategiesbased on the enhancement of environmental stimulationlevels in adult life.

2. Neuronal Activity and NPAS4-MediatedGene Expression Patterns

Studies aimed at the identification of genes that mediate theactivity-dependent regulation of inhibitory synapses forma-tion during development, revealed that NPAS4 is an IEGinduced by neuronal activity that seems to lie behind home-ostatic mechanisms that keep neuronal firing in response tosensory experience within normal levels [38].

The exposure of primary neuronal cultures to high levelsof potassium chloride leads to membrane depolarizationand calcium influx through L-type voltage-sensitive calciumchannels [39]. The resulting increase in intracellular calciumlevels then triggers calcium-dependent signaling pathwaysthat eventually mediate changes in gene transcription. Theanalysis of DNA microarrays upon this experimental design,in cortical neurons of young mice when development ofinhibitory synapses is underway, revealed that NPAS4 is atranscription factor regulated by neuronal activity, whoseexpression parallels the development of inhibitory synapticcontacts [38].

NPAS4 is selectively induced by calcium influx onlyin neurons but not in other cell types. The expression ofNPAS4, unlike other activity-dependent transcription fac-tors such as CREB and c-Fos, is triggered selectively byexcitatory synaptic transmission but not by neurotrophicfactors [38]. As observed in the cortex, NPAS4 expressionin primary hippocampal neurons increases with the forma-tion and maturation of synaptic contacts that occur duringdevelopment, presumably, because of enhanced endogenousspontaneous levels of activity. Of note, NPAS4 expressionin response to stimulation of primary sensory areas hasbeen reported; visual experience in mice after one week ofdark exposure actually increases mRNA and protein levels ofNPAS4 in visual cortex pyramidal cells [38]. Interestingly, theexpression of NPAS4 seems to take place predominantly inexcitatory neurons.

The induction of NPAS4 promotes GABA-mediatedinhibitory transmission during development. Studies in hip-pocampal cell cultures, using shRNA interference (shRNAi)against NPAS4 and immunohistochemistry for both theGABA synthetizing enzyme GAD65 and the GABAA-receptor 𝛾2 subunit as pre- and postsynaptic markers,respectively, revealed that the downregulation of NPAS4expression markedly reduces inhibitory synaptic contactsformation on perisomatic and dendritic regions of excitatoryneurons, suggesting that this transcription factor positivelyregulates the number of inhibitory synapses that form duringearly life. These findings were confirmed by recordings of

Neural Plasticity 3

miniature inhibitory postsynaptic currents (mIPSCs) in CA1pyramidal cells, which decrease in amplitude after NPAS4downregulation by NPAS4-shRNAi infection [38]. Further-more, experiments performed in conditional knockout mice(𝑁𝑃𝐴𝑆4flx/flx) in which the NPAS4 gene is selectively deletedby CRE-mediated recombination revealed that CRE expres-sion leads to a significant increase of interevent intervals ofmIPSCs, thus showing that CA1 pyramidal neurons lackingNPAS4 receive fewer inhibitory synaptic inputs. In contrast,increasing NPAS4 levels in cultured hippocampal neuronsenhances the formation of inhibitory synapses, as suggestedby a marked increase in the number of the GABAA-receptor𝛾2 subunits. In line with this, expression of NPAS4 in CA1pyramidal neurons increases the amplitude of mIPSCs whiledecreasing mIPSCs interevent intervals, consistent with anenhanced inhibitory synaptic signaling [38].

Notably, modifications of excitatory synaptic transmis-sion also seem to occur after alterations ofNPAS4 expression.In addition to the induction of genes that control thedevelopment of inhibition, NPAS4 also seems to regulate agene program that includes a wide variety of transcriptionfactors, genes encoding channel proteins, G-protein signalingmolecules, protein kinases and phosphatases, and genesinvolved in membrane receptors trafficking and synaptictransmission [38].Moreover, it has been reported thatNPAS4mediates BDNF expression in primary cortical neurons [30,38, 40, 41]. BDNF is reduced in neurons with decreasedlevels of NPAS4 after lentiviral NPAS4-shRNAi infectionand primary cell cultures from NPAS4 knockout miceconsistently show a similar reduction of depolarization-induced BDNF expression [38]. Chromatin immunoprecip-itation (ChIP) studies have shown that NPAS4 binds to theBDNF promoters I and IV in membrane-depolarized neu-rons, indicating that NPAS4 directly mediates the activity-dependent BDNF transcription. This phenomenon seems tounderlie, at least partially, the effect of NPAS4 in increasingthe formation of inhibitory synapses, as the number ofinhibitory synaptic contacts induced byNPAS4 is moderatelyattenuated in cells in which BDNF has been knocked downby BDNF-shRNAi infection. Accordingly, the enhancementof inhibition caused by NPAS4 in CA1 neurons is partiallybut not totally attenuated by knocking down BDNF levels[38].

In summary, NPAS4 induction in response to excitatorytransmission appears to mediate a reduction of neuronalactivity levels and therefore may function as a homeostaticmechanism during phases of enhanced excitability [38].To what extent NPAS4 mediates, directly or indirectly, thedevelopment of inhibitory synaptic contacts formed by dif-ferent types of GABAergic interneurons on excitatory cellsis an open question that remains to be explored. Furtherstudies of NPAS4 physiological functions may shed lighton mechanisms by which experience-dependent neuronalactivity regulates the balance between inhibition and exci-tation in the brain and how alterations in such a balancemay contribute to pathological conditions such as Downsyndrome, Autism, and Rett syndrome in which inhibitorytransmission seems to be altered [42–44].

3. NPAS4 Upregulates a Gene Program ThatUnderlies Memory Formation

The formation and storage of memories are a classicalexample of experience-dependent plasticity mechanisms thatallow an individual to modify behaviour by learning. Whatstructural and functional changes occur in the brain as welearn? It is well established that there are stages in memorythat are encoded as modifications in the strength of synapsesthat correlate with behavioural phases of short- and long-term memory.

Pioneering studies from molluscs to flies, and mam-mals revealed highly conserved signal transduction pathwaysthat are critical for the occurrence of synaptic plasticityunderlying the establishment of long-term memories. Theseconserved pathways involve calcium-mediated activation ofintracellular protein kinases, translocation of these proteinsto the nucleus, and subsequent activation of transcriptionfactors that mediate gene transcription (for review see [45]).Activation of Glutamate N-methyl-D-aspartate (NMDA)-receptors [46], for instance, seems to induce phosphorylationof CREB, which causes alterations of chromatin structurethat allow for the induction of gene programs and denovo synthesis of proteins that eventually mediate long-term changes of synaptic transmission during learning [47](Figure 1).

In rodents, the hippocampus is involved in the formationof memory for new environments or contexts (reviewed by[48]), this phenomenon being dependent on the activationof the CA3 hippocampal area [49–51]. Contextual memoryformation can be examined using the contextual fear con-ditioning (CFC) task (for review see [52]), which consists ofexposure of an animal to a given context in which an electricshock, thatmay ormay not be accompanied by a tone, occurs.After training in this protocol, wild-type animals normallyremember and associate the context with the aversive shockexperience, which can be later evaluated in terms of freezingbehaviour; 1 hour or 24 hours after training, the animalsare exposed to the same aversive context to explore eithershort- or long-term contextual memory, respectively. Usingthis experimental paradigm, a novel role for NPAS4 in theregulation of contextualmemory formation has been recentlyuncovered [30].

These studies initially evaluated the expression of theIEGs c-Fos, Arc, and NPAS4 in the dorsal hippocampus ofmice that were exposed to the CFC task and sacrificed atdifferent time points. Notably, NPAS4 expression was foundto peak much before that of c-Fos and Arc; NPAS4 mRNAreached its peak after 5min of training, returning to basallevels of expression after 4.5 hours. Instead, c-Fos and Arcreached their peak levels of expression after 30min of training[30]. These findings highlight a hierarchical genetic programin which NPAS4 is upstream of several other IEGs in thedorsal hippocampal area. This notion was later confirmedby the observation that conditional deletion of the NPAS4gene by CRE recombination in hippocampal neurons of𝑁𝑃𝐴𝑆4

flx/flx transgenic mice results in a marked loss of c-Fos,Arc, and Zif268 expression [30].

4 Neural Plasticity

Presynaptic neuron

Glutamate

AMPA receptorsNa+

Na+Na+

Ca2+

Ca2+AMPA receptorsinsertion

Dendritic spine ofpostsynaptic neuron Substrate

phosphorylationATP

cAMP

TranslationPostsynaptic neuron Protein

kinase ACREB Nucleus

TranscriptionDNA

CREB phosphorylation

Ca2+/calmodulinkinase proteins

NMDAreceptor

Proteinkinases

Figure 1: Molecular mechanisms underlying long-lasting modifica-tions of synaptic transmission. After presynaptic glutamate release,the NMDA channel opens only when the postsynaptic neuronis sufficiently depolarized. As a result, the permeability of Ca2+increases and Ca2+ ions activate postsynaptic protein kinases. Thesekinases may then act to insert new AMPA receptors into thepostsynaptic spine, thereby increasing the sensitivity to glutamate.The activation of second-messenger pathways (e.g., ↑ cAMP) thatsubsequently set in motion the catalytic subunit of the proteinkinase A results in the phosphorylation of the transcriptionalregulator CREB. This turns on the expression of a number of genes(those containing the CRE promoter area) that produce long-lastingstructural and functional changes on the synapses.

Learning and memory deficits were also evaluated inNPAS4 knockout (NPAS4−/−) mice. After 5min of training inthe CFC test, robust freezing behaviour was observed in bothwild-type and NPAS4−/− littermates, indicating that learningcapabilities were normal in NPAS4−/− animals. In contrast,freezing behaviour was significantly reduced 1 hour and 24hours after CFC training, showing that both short-term andlong-term memory formation is impaired in NPAS4−/− mice[30].

After CFC training, NPAS4 expression was localizedmainly in the CA3 area of the hippocampus. The selectivedeletion of NPAS4 in CA3 but not in CA1 impaired long-term contextual memory formation; 24 hours after CFCtraining, 𝑁𝑃𝐴𝑆4flx/flx mice injected in CA3 with a virusexpressing the CRE recombinase showed attenuated freezingresponses as comparedwithwild-type or𝑁𝑃𝐴𝑆4flx/flx animalsinjected in CA1 [30], thus demonstrating that deletingNPAS4specifically in CA3 replicates the memory deficits seen in theNPAS4 knockout.

The issue of whether NPAS4 expression in the CA3area of the NPAS4−/− background leads to the expressionof the NPAS4-mediated gene program and rescues memoryformation was also investigated. Remarkably, the expressionof NPAS4 in CA3 completely reversed the short-term andlong-term contextual memory deficits previously observed inthe NPAS4−/− background; NPAS4-expressing mice in CA3but not in CA1 showed similar freezing behaviour as wild-type control animals after either 1 hour or 24 hours of trainingin the CFC behavioural task [30]. Consistently, the sameexperimental design also induced c-Fos expression in CA3.

In summary, this elegant set of experiments demonstratesthat the activity-dependent transcription factor NPAS4is a key mediator of plastic phenomena that underliehippocampal-dependent contextual memory formation. Onthe one hand, acute deletion of theNPAS4 gene inCA3 resultsin a dramatic diminishment of IEGs expression and impairedcontextualmemory formation.On the other hand, expressionof NPAS4 mRNA in NPAS4 knockout animals effectivelyrestores both IEGs expression and memory formation.

4. Role of NPAS4 in the Regulation ofHomeostatic Plasticity

Thefirstmodel to provide a specificmechanism formodifica-tions of synaptic transmission involved in associative learningwas advanced by Donald Hebb in 1949; it was proposedthat modifications in the strength of synapses might occuronly if the use of those synapses was associated with andcontributes to the generation of action potentials in thepostsynaptic neuron (reviewed by [53]). Hebb’s principle hasbeen summarized as follows: “neurons that fire together wiretogether” whereas “neurons that fire out of synchrony losetheir connection.” Thus, an essential feature of this postulateis that modifications of synaptic transmission depend oncoincidence activity of the presynaptic and the postsynapticneuron. NMDA-receptors actually function as coincidencedetectors in synaptic plasticity, as they open and mediateexcitatory synaptic transmission only when the presynapticrelease of glutamate is coupled to the postsynaptic depolar-ization ([54, 55], for review see [56]), thus fulfilling Hebb’srule at molecular level.

Although Hebbian mechanisms provided an initial andimportant framework for the interpretation of neuronalnetwork alterations, it has become clear that there aremechanisms ofmetaplasticity controlling changes of synapticplasticity (reviewed by [57]). Indeed, due to positive feedback,Hebbian plasticity could lead to a saturation of the synapticstrength in the absence of proper constraints. There isnow a general consensus that homeostatic mechanisms areregulatory adjustments that work to maintain the stabilityand functionality of neuronal networks when modificationsof synaptic transmission are underway (reviewed in [58]).

A classical form of homeostatic plasticity is epitomized bythe Bienenstock-Cooper-Munro (BCM) model [59], whichstates that synaptic inputs driving postsynaptic firing to highlevels result in an increase in synaptic strength, whereasinputs that trigger low levels of postsynaptic firing result

Neural Plasticity 5

in a decrement of synaptic transmission. The threshold forneuronal activation in the BCM model is not fixed butchanges itself as a function of postsynaptic activity, thethreshold slides as tomake potentiationmore likelywheneveraverage activity is low, and less likely when average activityis high (reviewed by [60]). This is thought to maintain thestability of synapses in neuronal circuitries upon changes ofsynaptic transmission.

Mechanisms of homeostatic plasticity described so far(for review see [57, 58]) include (i) synaptic scaling (i.e.,scaling of the strength of excitatory synapses dependingon the average activity of the postsynaptic neuron) and(ii) the regulation of intrinsic excitability (i.e., changing theway in which postsynaptic neurons integrate synaptic inputsand fire action potentials). The identification of molecularsubstrates underlying these forms of homeostatic plasticity,however, still needs further research. Hence, the discoverythat the activity-dependent expression ofNPAS4 is implicatedin a transcriptional program that regulates neuronal firingresponses to excitatory transmission by enhancing inhibition[38] is of high relevance for homeostatic plasticity research. Itwill be interesting to evaluate mechanisms of metaplasticityinNPAS4−/− knockout animals or in conditional𝑁𝑃𝐴𝑆4flx/flxmice after deletion of the NPAS4 gene by selective CRErecombination.

5. NPAS4 and Structural Plasticity inthe Nervous System

Experience-dependent functional modifications of neuronalcircuitries in the brain are accompanied by structural rear-rangements of neuronal connectivity. Excitatory synapticstructures such as dendritic spines, for instance, are particu-larly sensitive to experience during development. A total lackof visual experience in early life (dark rearing) actually modi-fies spinesmorphology and density in the visual system, thesetwo phenomena being partially reversible by subsequent lightexposure [61]. In agreement with this notion, monoculardeprivation during the critical period influences motility,turnover, number, and morphology of dendritic spines in thevisual cortex [62–66]. These findings highlight a correlationbetween the structural remodeling of single synapses andfunctional modifications of neural circuitries in response tochanging environmental conditions.

Does structural plasticity contribute to experience-dependent changes of neuronal connectivity? This questionhas been recently addressed by signal optical imaging offunctional responses to visual stimulation and by longitudi-nal two-photon imaging experiments, showing that dendriticspine dynamics of pyramidal neurons in themouse neocortexis maximal during early stages of development but decreasesthereafter, in parallel to the decline of functional plasticitythat occurs over development ([65], for review see [67]).Although most studies on structural plasticity have focusedonmodifications in excitatory cells, there is also evidence thatstructural plasticity occurs in inhibitory neurons. It has beendemonstrated that GABAergic interneurons in superficiallayers of the visual cortex exhibit dendritic arbor growth and

remodeling in adult life [68]. Moreover, structural modifica-tions of inhibitory synapses onto pyramidal excitatory cellsseem to be amajor component of plasticity in the adultmouseneocortex [69–71]. The dynamic turnover of dendritic spineson pyramidal neurons and the remodeling of interneuronsdendritic arbors actually appear to be a common featureamong primary sensory areas [71]. In summary, cortical plas-ticity seems to be associated with a structural rearrangementof excitatory connections during early development whereasstructural modifications of dendritic arbors in GABAergicinterneurons seem to correlate with adult cortical plasticity.

An unresolved and interesting question in the fieldis whether NPAS4 activates downstream targets associatedwith structural plasticity in the nervous system. Very recentstudies suggest that NPAS4 may be involved, at least in part,in some forms of structural plasticity. There is evidence thatdifferentiation-induced neurite outgrowth in cell cultures isinhibited if NPAS4 expression is knocked down, whereasoverexpression of NPAS4 appears to accelerate neurite out-growth [72]. Moreover, depolarization-induced neurite out-growth is impaired in the hippocampus of NPAS4 knockoutanimals. This phenomenon appears to depend on phospho-rylation of the protein synapsin-I by the cyclin-dependentprotein kinase CDK5 and NPAS4 seems to mediate CDK5expression by binding to the CDK5 gene promoter area [72].Whether these findings bear any physiological significancein naturally occurring processes of neuronal plasticity is anopenquestion to be explored. Itmay be interesting to examinewhether dendritic spines in excitatory neurons and dendriticarbors in GABAergic cells are fewer and/or lessened in thevisual cortex ofNPAS4−/− mice or in conditional𝑁𝑃𝐴𝑆4flx/flxmice after deletion of the NPAS4 gene by selective CRErecombination.

Since the expression of NPAS4 seems to take placepredominantly in excitatory neurons [38], another interest-ing consideration would be to investigate whether NPAS4influences dendritic spines turnover and density in pyramidalcells in the visual cortex. This issue is of particular relevancefor the process of plasticity reactivation late in life as thereis evidence that the maturation of the extracellular matrixcomposition during development stabilizes neuronal con-nectivity patterns while inhibiting structural and functionalplasticity of dendritic spines [73]. The degradation of extra-cellular matrix components known as chondroitin sulphateproteoglycans (CSPGs) by exogenous administration of thebacterial enzyme chondroitinase actually reinstates oculardominance plasticity in adulthood [74], probably by mod-ifying dendritic spines dynamics and associated neuronalconnectivity changes in the visual cortex.

6. NPAS4 and the Regulation of CriticalPeriod Plasticity in the Visual System

The extent to which environmental influences modify brainstructure and function has been extensively studied in thedeveloping visual system. An experience-dependent reorga-nization of eye-specific inputs during early life is actually

6 Neural Plasticity

the major mechanism by which neuronal connectivity isestablished in the primary visual cortex.

Themonocular deprivation paradigm has been a classicalmodel to assess neuronal plasticity in the visual system. Earlyelectrophysiological and anatomical studies in cats and mon-keys revealed that short periods of sensory deprivation byunilateral eye closure during early life cause major structuraland functional modifications of visual cortical circuitries.Visual cortex responsiveness markedly shifts in favour ofthe non deprived eye after monocular deprivation duringthe critical period ([75–79]). In addition, the deprived eyebecomes amblyopic; its visual acuity (spatial resolution) andcontrast sensitivity are severely impaired [77, 79–81].

Is the activity-dependent NPAS4 expression involved inthe occurrence of critical period plasticity? Before addressingthis question, a brief overview of the developmental func-tional organization of the visual system should be considered.Sensory experience during early life signals the time courseof the critical period by promoting the transfer of the proteinOtx2 from the retina to the visual cortex, where Otx2 appearsto drive the maturation of parvalbumin-positive GABAergicinterneurons [82] (for review see [83, 84]). The experience-dependent developmental maturation of GABA-mediatedinhibition then establishes the threshold for both the startand the end of the critical period for visual cortical plasticity[85–87]. Indeed, transgenic mice with reduced levels of intra-cortical inhibition due to the lack of the GABA-synthetizingenzyme GAD65 exhibit no modifications of visual cortexresponsiveness after monocular deprivation in early life,whereas enhancing inhibition by exogenous administrationof GABAA-receptor agonists in the knockout backgroundrescues the impairment of plasticity [86, 87]. On the otherhand, transgenic animals that showan acceleratedmaturationof intracortical inhibition due to BDNF overexpression inforebrain regions display a precocious development of thevisual system and an accelerated end of the critical period forocular dominance plasticity [85].

In summary, an initial threshold of inhibition triggers asensitive period in which neuronal networks in the visualsystem are highly susceptible to sensory experience, whereasa second inhibitory threshold signals the end of this phase ofenhanced plasticity (for review see [88]). Since the transcrip-tional program activated by NPAS4 enhances inhibition bypromoting the expression of genes that direct the formationof inhibitory synaptic contacts [38], it emerges clearly thatNPAS4 expression is likely involved in the regulation of thecritical period for visual cortex plasticity. It will be interestingto evaluate whether NPAS4−/− animals with reduced levelsof inhibition [38] show impairments of ocular dominanceplasticity in response to monocular deprivation during earlydevelopment. This could be complemented by studies ofplasticity in wild-type animals after NPAS4 downregulationby selective NPAS4-shRNAi infection or in 𝑁𝑃𝐴𝑆4flx/flxmice with a selective deletion of the NPAS4 gene by CRErecombination in the developing visual cortex. Moreover, theinduction of theNPAS4-mediated gene program by infectionwith an NPAS4 expressing virus in the visual cortex of eitherwild-type or NPAS4−/− animals may be a feasible strategy

to evaluate whether NPAS4 overexpression accelerates visualsystem development and the time course of the criticalperiod for plasticity. Another interesting issue is whether theactivation of the retinogeniculocortical transfer of the proteinOtx2 in the visual pathway correlates with the activity-dependent expression of NPAS4 in pyramidal neurons of theprimary visual cortex.

An alternative approach to assess the impact of NPAS4expression in visual cortical plasticity may be rearing animalsin total darkness from birth. Experiments that combinedark rearing and electrophysiology as a functional readouthave demonstrated that the absence of visual inputs duringdevelopment leads to a delayed maturation of the visualcortex [89]. It will be exciting to evaluate whether darkrearing decreases the expression of NPAS4 in the visualcortex and whether the effects of dark rearing in the visualsystem of wild-type animals are prevented by selectiveNPAS4expression.

7. NPAS4 and Visual CortexPlasticity in Adult Life

Converging lines of evidence attribute the decline of plasticitythat occurs with age to the maturation of intracorticalinhibitory circuitries [85–87]. Consistently, it is possible torestore a high degree of plasticity in adult life by reducinglevels of inhibition [90]. This is in line with the observationthat experimental paradigms, such as dark exposure [91–93],environmental enrichment [94–96], food restriction [97],long-term fluoxetine treatment [15, 98, 99], exogenous IGF-1 administration [100], and genetic manipulations [101, 102],all promote plasticity in adult life by shifting the intracorticalinhibitory/excitatory ratio in favour of excitation.

Recent studies in rodents [5] and cats [103] have revealedthat the process of plasticity reactivation appears to be amultifactorial event that comprises the action of differentstructural and functional mechanisms, working in parallelor in series, the sum of which results in the activation ofintracellular signal pathways regulating the expression ofplasticity genes (reviewed by [18, 104]). Indeed, experience-dependent modifications of chromatin structure that controlgene transcription are recruited as targets of plasticity-associated processes in adulthood [14, 15, 97, 101].

Does NPAS4 drive mechanisms of visual cortex plasticityin adult life? There is evidence that NPAS4 mediates theactivity-dependent expression of BDNF [30, 38, 40, 41],a neurotrophin that has been clearly linked to multipleforms of synaptic plasticity in diverse brain areas ([105–110],for review see [111]). Of note, chronic infusion of BDNFinto the visual cortex restores susceptibility to monoculardeprivation in adulthood [98], whereas the impairment ofBDNF-trkB signaling effectively prevents the process ofplasticity reactivation caused by fluoxetine in the adult visualsystem [15]. These findings portray the activity-dependentNPAS4 transcription factor as an appealing candidate forthe regulation of visual cortical plasticity in adulthood. Thefollowing is an overview of the potential role of NPAS4 as

Neural Plasticity 7

a mediator of plasticity induced by different noninvasiveexperimental approaches in the adult visual system.

7.1. Pharmacological and Environmental-Like StimulationApproaches. Compelling experimental evidence for NPAS4-mediated transcriptional mechanisms that lie behind phe-nomena of visual cortex plasticity in adulthood has beenrecently obtained using themonocular deprivation paradigmand chronic treatment with fluoxetine as a pharmacologicalstrategy for the induction of plasticity.

There is evidence that the plastic outcome caused byfluoxetine in adulthood is accompanied by increased lev-els of serotonin, reduced levels of GABAergic inhibition,and increased BDNF expression in the visual cortex [98].More recently, it was demonstrated that the reinstatementof plasticity caused by fluoxetine is paralleled by epigeneticmodifications of chromatin structure that promote gene tran-scription. On the one hand, an increased histone acetylationstatus at BDNF promoter regions occurs in concomitancewith BDNF expression [15]. On the other, a reduction inthe methylation status at the NPAS4 promoter area paral-lels an enhanced NPAS4 expression after pharmacologicaltreatment [101]. Of note, the impairment of serotonergicsignaling prevents the remodeling of chromatin structurecaused by the pharmacological treatment in gene promoterareas [15]. This points toward a hierarchical model in whichserotonin seems to be the primummovens in a series of signaltransduction pathways leading to epigenetic modifications ofchromatin structure and subsequent expression of plasticitygenes in the adult visual cortex (for review see [18]). Inthis context, NPAS4 is upstream BDNF expression andseems to direct the gene program mediating this plasticphenomenon. Electrophysiological experiments in combina-tion with gene delivery by lentiviral infection have actuallyshown that NPAS4 expression in the visual cortex of naıveanimals restores susceptibility to monocular deprivation inadulthood. Consistently, NPAS4 downregulation by NPAS4-shRNAi in the adult visual system effectively prevents plasticevents caused by fluoxetine treatment [101].

How does NPAS4 fit into a model of enhanced plasticity,which correlates with a decrease of inhibition in adulthood?Given that NPAS4 expression increases the formation ofinhibitory synaptic contacts [38], one might expect NPAS4to be inversely correlated with the occurrence of plasticity;that is, NPAS4 expression should occlude plasticity whereasNPAS4 knockdown should enhance it. This scenario, how-ever, seems to be unlikely as there is evidence that knockingdown NPAS4 expression by NPAS4-shRNAi infection in thevisual cortex of naıve animals does not restore visual cortexsusceptibility to monocular deprivation in adult life [101].Instead, based on extensive data in the hippocampus, it seemsreasonable to hypothesize that the activity-dependentNPAS4expression caused by serotonin (fluoxetine) in the adult visualcortex may turn on a transcriptional program that upregu-lates the expression of plasticity genes while facilitating, inparallel or in series, a functional reorganization of inhibitorycircuitries thatmight contribute to the homeostasis of corticalexcitability during this phase of enhanced plasticity [101].

This is in line with two recent observations: (i) NPAS4interacts with a wide variety of neuronal activity-regulatedgene expression enhancers and promoters in the nervoussystem [112] and (ii) different homeostatic mechanisms assistto keep neuronal activity within normal levels as synapticmodifications of neuronal circuitries are underway in thevisual cortex [113].

The available experimental evidence for a combinedaction of fluoxetine-induced serotonergic signaling, reducedlevels of inhibition, and NPAS4 expression in driving adultvisual cortex plasticity is consistent with a model (Figure 2)in which the serotonin-mediated shift of the intracorticalinhibitory/excitatory balance that occurs in favour of excita-tion [15, 98] may induce the activity-dependent expressionof NPAS4 [101]. This in turn could mediate the expressionof plasticity genes and subsequently promote the formationof inhibitory synaptic contacts on excitatory neurons as acompensatory mechanism for the reduction of the inhibitorytone after fluoxetine treatment. In line with this notion,there is evidence that the fluoxetine-induced mechanism ofdisinhibition in the visual cortex of adult animals after abrief period of monocular deprivation is accompanied by anincrease in elongations of GABAergic interneuron dendriticbranch tips in superficial cortical layers [99]. Hence, thispoints toward a compensatory mechanism for the reductionof inhibition that involves the formation and/or strength-ening of inhibitory contacts on neighbouring excitatoryneurons, that is, a mechanism in which NPAS4 is likely to beinvolved.

Understanding how NPAS4 expression regulatesinhibitory synapse density and function in the adult visualsystem is important to interpret these findings. It maybe interesting to evaluate the time course of expressionof GABAergic markers in the adult visual cortex (e.g.,GAD65, GAD67, VGAT, GABAA-receptor 𝛾2 subunits, andparvalbumin) by means of immunohistochemistry in a naıvebackground versus a background of NPAS4 knockdown or𝑁𝑃𝐴𝑆4

flx/flx mice with a selective deletion of the NPAS4 geneby CRE recombination.This could also be assessed in a naıvebackground before and after long-term fluoxetine treatmentand may be complemented by electrophysiological analysisof spontaneous and evoked IPSCs to examine functionalconnectivity. The role of monocular deprivation in thisexperimental design should also be examined. The use ofDNA microarrays in this context might be of relevance forthe identification of NPAS4 downstream targets associatedwith structural and functional modifications in the adultvisual system after fluoxetine treatment [114].

The enhancement of environmental stimulation levelshas recently proved to be a powerful and noninvasive strat-egy to promote juvenile-like plasticity in the adult visualsystem. Environmental enrichment (EE) is an experimentalparadigm characterized by enhanced sensory-motor activityand social stimulation that has a profound impact on brainstructure and function (reviewed by [115]). In rodents, ithas been demonstrated that short period of EE in adult lifereactivates ocular dominance plasticity [94, 116] and thereis evidence that resetting adult visual cortex circuitries to a

8 Neural Plasticity

Long-term FLX treatment

↑ 5-HT transmission

↑ Gene transcription activation

↑ NPAS4 expression: ↑ Downstream NPAS4 target genes

↑ BDNF-trkB signaling ↑ Dendritic spines remodeling ↑ Inhibitory synapses

↑ Epigenetic remodeling of chromatin structure

↑ Inhibition↑ Plasticity

↑ Plasticity and homeostatic response

↓ Inhibition/excitation ratio

Figure 2: The process of plasticity reactivation in the adult visual system. The reinstatement of plasticity caused by FLX in adult life isassociated with signal transduction pathways that involve the activation of long-distance serotonergic transmission, a downregulation oflocal intracortical inhibitory circuitries and enhanced NPAS4 expression.The experimental evidence is consistent with a model in which theincreased serotonergic signaling shifts the inhibitory/excitatory balance, thus activating intracellular mechanisms that eventually promoteepigenetic modifications of chromatin structure that, in turn, allow for the expression of plasticity genes in adult life, among which NPAS4plays a key role. NPAS4 seems to turn on a transcriptional program that underlies structural and functional plasticity while facilitating,in parallel or in series, a reorganization of inhibitory circuitries that might contribute to the homeostasis of cortical excitability by drivinginhibition during this phase of enhanced plasticity.The transitory expression of NPAS4 target genesmay ultimately set inmotion downstreamphysiological mechanisms that provide a permissive environment for changes in adult visual cortical circuitries (e.g., enhanced Bdnf-trkB signaling, removal of extracellular matrix components that are inhibitory for plasticity, and enhanced dendritic spines density andremodeling). Continuous arrows represent established interactions between molecular and cellular processes mentioned (boxes). Dashedlines represent interactions that remain to be ascertained. Reproduced from [101] with permission.

more plastic stage by EE favours the rescue of sensory func-tions after long-term deprivation [95, 96, 117, 118]. In humans,enriching the environment in terms of body massage triggersplastic phenomena that accelerate the maturation of visualfunctions during development [119].

Does NPAS4 play a role in the effects caused by EEin visual cortical plasticity? This is a likely scenario asthe reinstatement of plasticity caused by EE in adulthoodis accompanied by an increment in serotonin signaling,reduced levels of inhibition, and enhanced BDNF expression[94], much as in the case of long-term fluoxetine treatment.

It is worth noting that decreasing BDNF signaling byexogenous administration of antisense oligonucleotides inthe visual cortex of adult animals exposed to enriched envi-ronmental conditions prevents partially but not totally theshift of ocular dominance in response to monocular depri-vation. Considering that NPAS4 directly promotes BDNFexpression, this suggests that the effects caused by EE in

visual cortical plasticity could be only partially dependenton NPAS4 expression. Another possibility is that NPAS4drives phenomena of plasticity even in the absence of BDNFsignaling and therefore could lie behind the plasticizingeffects of EE in adult life. This is, however, an open questionthat remains to be explored. It will be interesting to assessthe effects caused by EE in adult visual cortex plasticity inNPAS4−/− knockout animals or in 𝑁𝑃𝐴𝑆4flx/flx mice with aselective deletion of the NPAS4 gene by CRE recombination.

Brief periods of visual deprivation by dark exposure andfood restriction have also proved to be effective approachesto reactivate plasticity in the adult visual system. Juvenile-like ocular dominance plasticity can actually be restored inadult animals if monocular deprivation is preceded by visualdeprivation [93] or by a reduction of the caloric intake [97].It may be interesting to investigate the effects caused bybrief periods of dark exposure and food restriction in theNPAS4−/− knockout versus a naıve background.

Neural Plasticity 9

8. Conclusion and Implications

Long regarded as a rather static and unchanging structure,the adult brain has increasingly been recognized as a systemthat retains a degree of plasticity that allows for structuraland functionalmodifications of neuronal networks if exposedto certain experimental conditions. In animal models, ithas been demonstrated that both pharmacological treat-ments and experimental paradigms based upon manipula-tion of environmental stimulation levels effectively promotea rewiring of visual cortical circuitries in adulthood. It seemsreasonable to speculate that these noninvasive approaches,when combined with appropriate instructive environmen-tal stimuli, could be exploited for clinical applications inhumans. Since the decline of plasticity that occurs over thelife course severely restricts the functional reorganization ofneuronal circuitries, these studies are beginning to elucidatephysiological processes that lie behind modifications ofneuronal networks’ connectivity in adulthood in which theactivity-dependent transcription factor NPAS4 seems to playa critical role. In light of this, structural and functional mech-anisms leading to the activity-dependent NPAS4 expressionarise as potential therapeutic targets for future developmentof drugs that could be used in a variety of pathologicalconditions in which a reorganization of neuronal circuitriesis needed late in life. Long-term fluoxetine administration,indeed, not only inducesNPAS4 expression [101], but also hasbeen proved successful in promoting the functional recoveryfrom stroke in humans [120], which is a major cause oflong-term disability for which there is currently no clinicaltreatment.

Acknowledgment

Jose Fernando Maya-Vetencourt is supported by a researchfellowship provided by the Italian Institute of Technology,Italy.

References

[1] L. C.Katz andC. J. Shatz, “Synaptic activity and the constructionof cortical circuits,” Science, vol. 274, no. 5290, pp. 1133–1138,1996.

[2] N. Berardi, T. Pizzorusso, and L.Maffei, “Critical periods duringsensory development,”Current Opinion in Neurobiology, vol. 10,no. 1, pp. 138–145, 2000.

[3] C. N. Levelt and M. Hubener, “Critical-period plasticity in thevisual cortex,” Annual Review of Neuroscience, vol. 35, pp. 309–330, 2012.

[4] T. K. Hensch and P. M. Bilimoria, “Re-opening windows:manipulating critical periods for brain development,” Cere-brum, vol. 2012, p. 11, 2012.

[5] D. Bavelier, D. M. Levi, R. W. Li, Y. Dan, and T. K. Hensch,“Removing brakes on adult brain plasticity: from molecular tobehavioral interventions,” The Journal of Neuroscience, vol. 30,no. 45, pp. 14964–14971, 2010.

[6] R. Zorowitz and M. Brainin, “Advances in brain recovery andrehabilitation,” Stroke, vol. 42, no. 2, pp. 294–297, 2011.

[7] T. Aid, A. Kazantseva, M. Piirsoo, K. Palm, and T. Timmusk,“Mouse and rat BDNF gene structure and expression revisited,”

Journal of Neuroscience Research, vol. 85, no. 3, pp. 525–535,2007.

[8] T. Timmusk, N. Belluardo, H. Persson, and M. Metsis, “Devel-opmental regulation of brain-derived neurotrophic factor mes-senger RNAs transcribed from different promoters in the ratbrain,” Neuroscience, vol. 60, no. 2, pp. 287–291, 1994.

[9] T. Timmusk, U. Lendahl, H. Funakoshi, E. Arenas, H. Persson,and M. Metsis, “Identification of brain-derived neurotrophicfactor promoter regions mediating tissue-specific, axotomy-,and neuronal activity-induced expression in transgenic mice,”Journal of Cell Biology, vol. 128, no. 1-2, pp. 185–199, 1995.

[10] E. Borrelli, E. J. Nestler, C. D. Allis, and P. Sassone-Corsi,“Decoding the epigenetic language of neuronal plasticity,”Neuron, vol. 60, no. 6, pp. 961–974, 2008.

[11] L. Zocchi and P. Sassone-Corsi, “Joining the dots: from chro-matin remodeling to neuronal plasticity,” Current Opinion inNeurobiology, vol. 20, no. 4, pp. 432–440, 2010.

[12] J. M. Levenson, K. J. O’Riordan, K. D. Brown, M. A. Trinh, D.L. Molfese, and J. D. Sweatt, “Regulation of histone acetylationduring memory formation in the hippocampus,”The Journal ofBiological Chemistry, vol. 279, no. 39, pp. 40545–40559, 2004.

[13] A. Fischer, F. Sananbenesi, X.Wang, M. Dobbin, and L.-H. Tsai,“Recovery of learning andmemory is associatedwith chromatinremodelling,” Nature, vol. 447, no. 7141, pp. 178–182, 2007.

[14] E. Putignano, G. Lonetti, L. Cancedda et al., “Developmentaldownregulation of histone posttranslational modifications reg-ulates visual cortical plasticity,” Neuron, vol. 53, no. 5, pp. 747–759, 2007.

[15] J. F. Maya-Vetencourt, E. Tiraboschi, M. Spolidoro, E. Castren,and L. Maffei, “Serotonin triggers a transient epigenetic mech-anism that reinstates adult visual cortex plasticity in rats,” TheEuropean Journal of Neuroscience, vol. 33, no. 1, pp. 49–57, 2011.

[16] J. Graff, D. Kim, M. M. Dobbin, and T. Li-Huei, “Epigeneticregulation of gene expression in physiological and pathologicalbrain processes,” Physiological Reviews, vol. 91, no. 2, pp. 603–649, 2011.

[17] E. D. Nelson and L. M. Monteggia, “Epigenetics in the maturemammalian brain: effects on behavior and synaptic transmis-sion,” Neurobiology of Learning and Memory, vol. 96, no. 1, pp.53–60, 2011.

[18] J. F. Maya-Vetencourt and N. Origlia, “Visual cortex plasticity:a complex interplay of genetic and environmental influences,”Neural Plasticity, vol. 2012, Article ID 631965, 14 pages, 2012.

[19] E. Nedivi, D. Hevroni, D. Naot, D. Israell, and Y. Citri, “Numer-ous candidate plasticity-related genes revealed by differentialcDNA cloning,” Nature, vol. 363, no. 6431, pp. 718–722, 1993.

[20] S. Loebrich and E. Nedivi, “The function of activity-regulatedgenes in the nervous system,” Physiological Reviews, vol. 89, no.4, pp. 1079–1103, 2009.

[21] J. H. Leslie and E. Nedivi, “Activity-regulated genes asmediatorsof neural circuit plasticity,” Progress in Neurobiology, vol. 94, no.3, pp. 223–237, 2011.

[22] C. M. Alberini, “Transcription factors in long-term memoryand synaptic plasticity,” Physiological Reviews, vol. 89, no. 1, pp.121–145, 2009.

[23] M. Rickhag, T. Wieloch, G. Gido et al., “Comprehensiveregional and temporal gene expression profiling of the ratbrain during the first 24 h after experimental stroke identi-fies dynamic ischemia-induced gene expression patterns, andreveals a biphasic activation of genes in surviving tissue,”Journal of Neurochemistry, vol. 96, no. 1, pp. 14–29, 2006.

10 Neural Plasticity

[24] M. Shamloo, L. Soriano, D. Von Schack et al., “Npas4, a novelhelix-loop-helix PAS domain protein, is regulated in responseto cerebral ischemia,”TheEuropean Journal of Neuroscience, vol.24, no. 10, pp. 2705–2720, 2006.

[25] N. Ooe, K. Motonaga, K. Kobayashi, K. Saito, and H. Kaneko,“Functional characterization of basic helix-loop-helix-PAS typetranscription factor NXF in vivo putative involvement in an“on demand” neuroprotection system,”The Journal of BiologicalChemistry, vol. 284, no. 2, pp. 1057–1063, 2009.

[26] W. D. Flood, R. W. Moyer, A. Tsykin, G. R. Sutherland, and S.A. Koblar, “Nxf and Fbxo33: novel seizure-responsive genes inmice,” The European Journal of Neuroscience, vol. 20, no. 7, pp.1819–1826, 2004.

[27] S.-J. Zhang, M. Zou, L. Lu et al., “Nuclear calcium signalingcontrols expression of a large gene pool: identification of a geneprogram for acquired neuroprotection induced by synapticactivity,” PLoS Genetics, vol. 5, no. 8, Article ID e1000604, 2009.

[28] I. Hester, S. McKee, P. Pelletier et al., “Transient expressionof Nxf, a bHLH-PAS transactivator induced by neuronal pre-conditioning, confers neuroprotection in cultured cells,” BrainResearch, vol. 1135, no. 1, pp. 1–11, 2007.

[29] J. E. Ploski, M. S. Monsey, T. Nguyen, R. J. DiLeone, and G.E. Schafe, “The neuronal PAS domain protein 4 (Npas4) isrequired for new and reactivated fear memories,” PLoS ONE,vol. 6, no. 8, Article ID e23760, 2011.

[30] K. Ramamoorthi, R. Fropf, G.M. Belfort et al., “Npas4 regulatesa transcriptional program in CA3 required for contextualmemory formation,” Science, vol. 334, no. 6063, pp. 1669–1675,2011.

[31] L. Coutellier, S. Beraki, P. M. Ardestani, N. L. Saw, and M.Shamloo, “Npas4: a neuronal transcription factor with a keyrole in social and cognitive functions relevant to developmentaldisorders,” PloS ONE, vol. 7, no. 9, Article ID e46604, 2012.

[32] M. L. Guo, B. Xue, D. Z. Jin et al., “Upregulation of Npas4protein expression by chronic administration of amphetaminein rat nucleus accumbens in vivo,”Neuroscience Letters, vol. 528,no. 2, pp. 210–214, 2012.

[33] T. Z. Shaw-Lutchman, S. Impey, D. Storm, and E. J. Nestler,“Regulation of CRE-mediated transcription in mouse brain byamphetamine,” Synapse, vol. 48, no. 1, pp. 10–17, 2003.

[34] W. Renthal, T. L. Carle, I. Maze et al., “ΔFosB mediatesepigenetic desensitization of the c-fos gene after chronicamphetamine exposure,” The Journal of Neuroscience, vol. 28,no. 29, pp. 7344–7349, 2008.

[35] J. Yun, H. Koike, D. Ibi et al., “Chronic restraint stress impairsneurogenesis and hippocampus-dependent fear memory inmice: possible involvement of a brain-specific transcriptionfactor Npas4,” Journal of Neurochemistry, vol. 114, no. 6, pp.1840–1851, 2010.

[36] D. Ibi, K. Takuma, H. Koike et al., “Social isolation rearing-induced impairment of the hippocampal neurogenesis is asso-ciated with deficits in spatial memory and emotion-relatedbehaviors in juvenile mice,” Journal of Neurochemistry, vol. 105,no. 3, pp. 921–932, 2008.

[37] Y. Furukawa-Hibi, J. Yun, T. Nagai, and K. Yamada, “Tran-scriptional suppression of the neuronal PAS domain 4 (Npas4)gene by stress via the binding of agonist-bound glucocorticoidreceptor to its promoter,” Journal of Neurochemistry, vol. 123, no.3, pp. 866–875, 2012.

[38] Y. Lin, B. L. Bloodgood, J. L. Hauser et al., “Activity-dependentregulation of inhibitory synapse development by Npas4,”Nature, vol. 455, no. 7217, pp. 1198–1204, 2008.

[39] P. L. Greer and M. E. Greenberg, “From synapse to nucleus:calcium-dependent gene transcription in the control of synapsedevelopment and function,”Neuron, vol. 59, no. 6, pp. 846–860,2008.

[40] N. Ooe, K. Saito, N. Mikami, I. Nakatuka, and H. Kaneko,“Identification of a novel basic helix-loop-helix-PAS factor,NXF, reveals a Sim2 competitive, positive regulatory role indendritic-cytoskeleton modulator drebrin gene expression,”Molecular and Cellular Biology, vol. 24, no. 2, pp. 608–616, 2004.

[41] P. Pruunsild, M. Sepp, E. Orav, I. Koppel, and T. Timmusk,“Identification of cis-elements and transcription factors regulat-ing neuronal activity-dependent transcription of human BDNFgene,”The Journal of Neuroscience, vol. 31, no. 9, pp. 3295–3308,2011.

[42] J. L. R. Rubenstein and M. M. Merzenich, “Model of autism:increased ratio of excitation/inhibition in key neural systems,”Genes, Brain and Behavior, vol. 2, no. 5, pp. 255–267, 2003.

[43] V. S. Dani, Q. Chang, A. Maffei, G. G. Turrigiano, R. Jaenisch,and S. B. Nelson, “Reduced cortical activity due to a shift in thebalance between excitation and inhibition in a mouse model ofrett syndrome,” Proceedings of the National Academy of Sciencesof the United States of America, vol. 102, no. 35, pp. 12560–12565,2005.

[44] F. Fernandez, W. Morishita, E. Zuniga et al., “Pharmacotherapyfor cognitive impairment in amousemodel of down syndrome,”Nature Neuroscience, vol. 10, no. 4, pp. 411–413, 2007.

[45] E. R. Kandel, “The molecular biology of memory storage: adialogue between genes and synapses,” Science, vol. 294, no.5544, pp. 1030–1038, 2001.

[46] R. G. M. Morris, E. Anderson, G. S. Lynch, and M. Baudry,“Selective impairment of learning and blockade of long-termpotentiation by an N-methyl-D-aspartate receptor antagonist,AP5,” Nature, vol. 319, no. 6056, pp. 774–776, 1986.

[47] S. Impey, D. M. Smith, K. Obrietan, R. Donahue, C. Wade, andD. R. Storm, “Stimulation of cAMP response element (CRE)-mediated transcription during contextual learning,” NatureNeuroscience, vol. 1, no. 7, pp. 595–601, 1998.

[48] G. Neves, S. F. Cooke, and T. V. P. Bliss, “Synaptic plasticity,memory and the hippocampus: a neural network approach tocausality,” Nature Reviews Neuroscience, vol. 9, no. 1, pp. 65–75,2008.

[49] T. Nakashiba, J. Z. Young, T. J. McHugh, D. L. Buhl, andS. Tonegawa, “Transgenic inhibition of synaptic transmissionreveals role of CA3 output in hippocampal learning,” Science,vol. 319, no. 5867, pp. 1260–1264, 2008.

[50] K. Nakazawa, L. D. Sun, M. C. Quirk, L. Rondi-Reig, M. A.Wilson, and S. Tonegawa, “Hippocampal CA3NMDAreceptorsare crucial for memory acquisition of one-time experience,”Neuron, vol. 38, no. 2, pp. 305–315, 2003.

[51] I. Lee and R. P. Kesner, “Different contributions of dorsalhippocampal subregios to emory acquisation and retrieval incontextual fear-conditioning,” Hippocampus, vol. 14, no. 3, pp.301–310, 2004.

[52] R. G. Phillips and J. E. LeDoux, “Differential contributionof amygdala and hippocampus to cued and contextual fearconditioning,” Behavioral Neuroscience, vol. 106, no. 2, pp. 274–285, 1992.

[53] N. Le Roux, M. Amar, and P. Fossier, “Acquiring new informa-tion in a neuronal network: fromHebb’s concept to homeostaticplasticity,” Journal de la Societe de Biologie, vol. 202, no. 2, pp.143–160, 2008.

Neural Plasticity 11

[54] B. M. Kampa, J. Clements, P. Jonas, and G. J. Stuart, “Kineticsof Mg2+ unblock of NMDA receptors: implications for spike-timing dependent synaptic plasticity,” Journal of Physiology, vol.556, no. 2, pp. 337–345, 2004.

[55] J. A. Kauer, R. C.Malenka, andR.A.Nicoll, “NMDAapplicationpotentiates synaptic transmission in the hippocampus,”Nature,vol. 334, no. 6179, pp. 250–252, 1988.

[56] P. Paoletti, “Molecular basis of NMDA receptor functionaldiversity,” The European Journal of Neuroscience, vol. 33, no. 8,pp. 1351–1365, 2011.

[57] W. C. Abraham, “Metaplasticity: tuning synapses and networksfor plasticity,”Nature ReviewsNeuroscience, vol. 9, no. 5, pp. 387–399, 2008.

[58] G. G. Turrigiano, “The self-tuning neuron: synaptic scaling ofexcitatory synapses,” Cell, vol. 135, no. 3, pp. 422–435, 2008.

[59] E. L. Bienenstock, L. N. Cooper, and P. W. Munro, “Theory forthe development of neuron selectivity: orientation specificityand binocular interaction in visual cortex,” The Journal ofNeuroscience, vol. 2, no. 1, pp. 32–48, 1982.

[60] N. S. Desai, “Homeostatic plasticity in the CNS: synaptic andintrinsic forms,” Journal of Physiology Paris, vol. 97, no. 4–6, pp.391–402, 2003.

[61] W. Wallace and M. F. Bear, “A morphological correlate ofsynaptic scaling in visual cortex,” The Journal of Neuroscience,vol. 24, no. 31, pp. 6928–6938, 2004.

[62] A. Majewska and M. Sur, “Motility of dendritic spines in visualcortex in vivo: changes during the critical period and effectsof visual deprivation,” Proceedings of the National Academy ofSciences of the United States of America, vol. 100, no. 26, pp.16024–16029, 2003.

[63] N. Mataga, Y. Mizuguchi, and T. K. Hensch, “Experience-dependent pruning of dendritic spines in visual cortex by tissueplasminogen activator,” Neuron, vol. 44, no. 6, pp. 1031–1041,2004.

[64] S. B. Hofer, T. D. Mrsic-Flogel, T. Bonhoeffer, and M. Hubener,“Experience leaves a lasting structural trace in cortical circuits,”Nature, vol. 457, no. 7227, pp. 313–317, 2009.

[65] D. Tropea, A. K. Majewska, R. Garcia, and M. Sur, “Structuraldynamics of synapses in vivo correlate with functional changesduring experience-dependent plasticity in visual cortex,” TheJournal of Neuroscience, vol. 30, no. 33, pp. 11086–11095, 2010.

[66] S. Oray, A. Majewska, and M. Sur, “Dendritic spine dynam-ics are regulated by monocular deprivation and extracellularmatrix degradation,”Neuron, vol. 44, no. 6, pp. 1021–1030, 2004.

[67] A. Holtmaat and K. Svoboda, “Experience-dependent struc-tural synaptic plasticity in the mammalian brain,” NatureReviews Neuroscience, vol. 10, no. 9, pp. 647–658, 2009.

[68] W.-C. A. Lee, H. Huang, G. Feng et al., “Dynamic remodelingof dendritic arbors in GABAergic interneurons of adult visualcortex,” PLoS Biology, vol. 4, no. 2, p. e29, 2006.

[69] T. Keck, V. Scheuss, R. I. Jacobsen et al., “Loss of sensory inputcauses rapid structural changes of inhibitory neurons in adultmouse visual cortex,” Neuron, vol. 71, no. 5, pp. 869–882, 2011.

[70] D. van Versendaal, R. Rajendran, H. M. Saiepour et al., “Elim-ination of inhibitory synapses is a major component of adultocular dominance plasticity,”Neuron, vol. 74, no. 2, pp. 374–383,2012.

[71] J. L. Chen,G.H. Flanders,W.-C.A. Lee,W.C. Lin, andE.Nedivi,“Inhibitory dendrite dynamics as a general feature of the adultcortical microcircuit,” The Journal of Neuroscience, vol. 31, no.35, pp. 12437–12443, 2011.

[72] J. Yun, T.Nagai, Y. Furukawa-Hibi et al., “Neuronal PerArnt Sim(PAS) domain protein 4 (NPAS4) regulates neurite outgrowthand phosphorylation of synapsin I,” The Journal of BiologicalChemistry, vol. 288, pp. 2655–2664, 2013.

[73] L. de Vivo, S. Landi, M. Panniello et al., “Extracellular matrixinhibits structural and functional plasticity of dendritic spinesin the adult visual cortex,” Nature Communications, vol. 4, p.1484, 2013.

[74] T. Pizzorusso, P. Medini, N. Berardi, S. Chierzi, J. W. Fawcett,and L. Maffei, “Reactivation of ocular dominance plasticity inthe adult visual cortex,” Science, vol. 298, no. 5596, pp. 1248–1251, 2002.

[75] T. N. Wiesel and D. H. Hubel, “Single-cell responses in striatecortex of kittens deprived of vision in one eye,” Journal ofneurophysiology, vol. 26, pp. 1003–1017, 1963.

[76] D. H. Hubel and T. N. Wiesel, “The period of susceptibility tothe physiological effects of unilateral eye closure in kittens,”TheJournal of Physiology, vol. 206, no. 2, pp. 419–436, 1970.

[77] A. Antonini and M. P. Stryker, “Rapid remodeling of axonalarbors in the visual cortex,” Science, vol. 260, no. 5115, pp. 1819–1821, 1993.

[78] S. LeVay, T. N. Wiesel, and D. H. Hubel, “The development ofocular dominance columns in normal and visually deprivedmonkeys,” The Journal of Comparative Neurology, vol. 191, no.1, pp. 1–51, 1980.

[79] C. Blakemore andR.C.Van Sluyters, “Reversal of the physiolog-ical effects ofmonocular deprivation in kittens: further evidencefor a sensitive period,”The Journal of Physiology, vol. 237, no. 1,pp. 195–216, 1974.

[80] C. Blakemore, F. Vital-Durand, and L. J. Garey, “Recoveryfrom monocular deprivation in the monkey. I. Reversal ofphysiological effects in the visual cortex,” Proceedings of theRoyal Society of London - Biological Sciences, vol. 213, no. 1193,pp. 399–423, 1981.

[81] G. T. Prusky and R. M. Douglas, “Developmental plasticity ofmouse visual acuity,”The European Journal of Neuroscience, vol.17, no. 1, pp. 167–173, 2003.

[82] S. Sugiyama, A. A. Di Nardo, S. Aizawa et al., “Experience-dependent transfer of Otx2 homeoprotein into the visual cortexactivates postnatal plasticity,” Cell, vol. 134, no. 3, pp. 508–520,2008.

[83] A. Rebsam and C. A. Mason, “Otx2’s Incredible Journey,” Cell,vol. 134, no. 3, pp. 386–387, 2008.

[84] S. Sugiyama, A. Prochiantz, and T. K. Hensch, “From brainformation to plasticity: insights on Otx2 homeoprotein,” Devel-opment Growth and Differentiation, vol. 51, no. 3, pp. 369–377,2009.

[85] Z. J. Huang, A. Kirkwood, T. Pizzorusso et al., “BDNF regulatesthe maturation of inhibition and the critical period of plasticityin mouse visual cortex,” Cell, vol. 98, no. 6, pp. 739–755, 1999.

[86] T. K. Hensch, M. Fagiolini, N. Mataga, M. P. Stryker, S.Baekkeskov, and S. F. Kash, “Local GABA circuit control ofexperience-dependent plasticity in developing visual cortex,”Science, vol. 282, no. 5393, pp. 1504–1508, 1998.

[87] M. Fagiolini andT.K.Hensch, “Inhibitory threshold for critical-period activation in primary visual cortex,”Nature, vol. 404, no.6774, pp. 183–186, 2000.

[88] T. K. Hensch, “Critical period plasticity in local cortical cir-cuits,” Nature Reviews Neuroscience, vol. 6, no. 11, pp. 877–888,2005.


[89] M. Fagiolini, T. Pizzorusso, N. Berardi, L. Domenici, and L.Maffei, “Functional postnatal development of the rat primaryvisual cortex and the role of visual experience: dark rearing andmonocular deprivation,”Vision Research, vol. 34, no. 6, pp. 709–720, 1994.

[90] A. Harauzov, M. Spolidoro, G. DiCristo et al., “Reducingintracortical inhibition in the adult visual cortex promotesocular dominance plasticity,” The Journal of Neuroscience, vol.30, no. 1, pp. 361–371, 2010.

[91] H.-Y. He, B. Ray, K. Dennis, and E. M. Quinlan, “Experience-dependent recovery of vision following chronic deprivationamblyopia,” Nature Neuroscience, vol. 10, no. 9, pp. 1134–1136,2007.

[92] K. L. Montey and E. M. Quinlan, “Recovery from chronicmonocular deprivation following reactivation of thalamocorti-cal plasticity by dark exposure,”Nature Communications, vol. 2,no. 1, article 317, 2011.

[93] H. Y. He and E. M. Quinlan, “Visual deprivation reactivatesrapid ocular dominance plasticity in adult visual cortex,” TheJournal of Neuroscience, vol. 226, pp. 2951–2955, 2006.

[94] L. Baroncelli, A. Sale, A. Viegi et al., “Experience-dependentreactivation of ocular dominance plasticity in the adult visualcortex,” Experimental Neurology, vol. 226, pp. 100–109, 2010.

[95] A. Sale, J. F. Maya-Vetencourt, P. Medini et al., “Environmen-tal enrichment in adulthood promotes amblyopia recoverythrough a reduction of intracortical inhibition,” Nature Neuro-science, vol. 10, pp. 679–681, 2007.

[96] L. Baroncelli, J. Bonaccorsi, M. Milanese et al., “Enrichedexperience and recovery from amblyopia in adult rats: Impactof motor, social and sensory components,”Neuropharmacology,vol. 62, no. 7, pp. 2387–2396, 2012.

[97] M. Spolidoro, L. Baroncelli, E. Putignano, J. F.Maya-Vetencourt,A. Viegi, and L.Maffei, “Food restriction enhances visual cortexplasticity in adulthood,” Nature Communications, vol. 2, no. 1,article 320, 2011.

[98] J. F. Maya-Vetencourt, A. Sale, A. Viegi et al., “The antidepres-sant fluoxetine restores plasticity in the adult visual cortex,”Science, vol. 320, no. 5874, pp. 385–388, 2008.

[99] J. L. Chen,W.C. Lin, J.W. Cha, P. T. So, Y. Kubota, and E.Nedivi,“Structural basis for the role of inhibition in facilitating adultbrain plasticity,”Nature Neuroscience, vol. 14, no. 5, pp. 587–596,2011.

[100] J. F. Maya-Vetencourt, L. Baroncelli, A. Viegi et al., “IGF-1restores visual cortex plasticity in adult life by reducing localGABA levels,” Neural Plasticity, vol. 2012, Article ID 250421, 10pages, 2012.

[101] J. F. Maya-Vetencourt, E. Tiraboschi, D. Greco et al., “Experie-nce-dependent expression ofNPAS4 regulates plasticity in adultvisual cortex,”The Journal of Physiology, vol. 590, pp. 4777–4787,2012.

[102] H. Morishita, J. M. Miwa, N. Heintz, and T. K. Hensch, “Lynx1,a cholinergic brake, limits plasticity in adult visual cortex,”Science, vol. 330, no. 6008, pp. 1238–1240, 2010.

[103] L. Bachatene, V. Bharmauria, S. Cattan, and S. Molotchnikoff,“Fluoxetine and serotonin facilitate attractive-adaptationin-duced orientation plasticity in adult cat visual cortex,” TheEuropean Journal of Neuroscience, vol. 38, no. 1, pp. 2065–2077,2013.

[104] D. Tropea, A. VanWart, andM. Sur, “Molecular mechanisms ofexperience-dependent plasticity in visual cortex,” PhilosophicalTransactions of the Royal Society B, vol. 364, no. 1515, pp. 341–355, 2009.

[105] A. K. McAllister, D. C. Lo, and L. C. Katz, “Neurotrophinsregulate dendritic growth in developing visual cortex,” Neuron,vol. 15, no. 4, pp. 791–803, 1995.

[106] A. K. McAllister, L. C. Katz, and D. C. Lo, “Neurotrophin reg-ulation of cortical dendritic growth requires activity,” Neuron,vol. 17, no. 6, pp. 1057–1064, 1996.

[107] E. S. Levine, R. A. Crozier, I. B. Black, and M. R. Plummer,“Brain-derived neurotrophic factor modulates hippocampalsynaptic transmission by increasing N-methyl-D-aspartic acidreceptor activity,” Proceedings of the National Academy of Sci-ences of the United States of America, vol. 95, no. 17, pp. 10235–10239, 1998.

[108] G. Carmignoto, T. Pizzorusso, S. Tia, and S. Vicini, “Brain-derived neurotrophic factor and nerve growth factor potentiateexcitatory synaptic transmission in the rat visual cortex,” TheJournal of Physiology, vol. 498, no. 1, pp. 153–164, 1997.

[109] H. W. Horch and L. C. Katz, “BDNF release from singlecells elicits local dendritic growth in nearby neurons,” NatureNeuroscience, vol. 5, no. 11, pp. 1177–1184, 2002.

[110] S. Meis, T. Endres, and V. Lessmann, “Postsynaptic BDNF sig-nalling regulates long-term potentiation at thalamo-amygdalaafferents,” Journal of Physiology, vol. 590, no. 1, pp. 193–208, 2012.

[111] A. K. McAllister, L. C. Katz, and D. C. Lo, “Neurotrophins andsynaptic plasticity,” Annual Review of Neuroscience, vol. 22, pp.295–318, 1999.

[112] T.-K. Kim, M. Hemberg, J. M. Gray et al., “Widespread tran-scription at neuronal activity-regulated enhancers,”Nature, vol.465, no. 7295, pp. 182–187, 2010.

[113] T. D. Mrsic-Flogel, S. B. Hofer, K. Ohki, R. C. Reid, T.Bonhoeffer, and M. Hubener, “Homeostatic regulation of eye-specific responses in visual cortex during ocular dominanceplasticity,” Neuron, vol. 54, no. 6, pp. 961–972, 2007.

[114] E. Tiraboschi, R. Guirado, D. Greco et al., “Gene expressionpatterns underlying the reinstatement of plasticity in the adultvisual system,” Neural Plasticity, vol. 2013, Article ID 605079, 8pages, 2013.

[115] H. Van Praag, G. Kempermann, and F. H. Gage, “NeuralConsequences of environmental enrichment,” Nature ReviewsNeuroscience, vol. 1, no. 3, pp. 191–198, 2000.

[116] M. Mainardi, S. Landi, L. Gianfranceschi et al., “Environmen-tal enrichment potentiates thalamocortical transmission andplasticity in the adult rat visual cortex,” Journal of NeuroscienceResearch, vol. 88, no. 14, pp. 3048–3059, 2010.

[117] P. Tognini, I. Manno, J. Bonaccorsi, M. C. Cenni, A. Sale, andL. Maffei, “Environmental enrichment promotes plasticity andvisual acuity recovery in adult monocular amblyopic rats,” PLoSONE, vol. 7, no. 4, Article ID e34815, 2012.

[118] L. Baroncelli, C. Braschi, and L. Maffei, “Visual depth per-ception in normal and deprived rats: effects of environmentalenrichment,” Neuroscience, vol. 236, pp. 313–319, 2013.

[119] A. Guzzetta, S. Baldini, A. Bancale et al., “Massage acceleratesbrain development and the maturation of visual function,” TheJournal of Neuroscience, vol. 29, no. 18, pp. 6042–6051, 2009.

[120] F. Chollet, J. Tardy, J. F. Albucher et al., “Fluoxetine for motorrecovery after acute ischaemic stroke (FLAME): a randomisedplacebo-controlled trial,” The Lancet Neurology, vol. 10, no. 2,pp. 123–130, 2011.


Review ArticleEnvironment, Leptin Sensitivity, and Hypothalamic Plasticity

Marco Mainardi,1 Tommaso Pizzorusso,1,2 and Margherita Maffei3,4

1 CNR Neuroscience Institute, Via Moruzzi 1, 56124 Pisa, Italy2 Department of Neuroscience, Psychology, Drug Research and Child Health NEUROFARBA, University of Florence,Via San Salvi 12, 50135 Florence, Italy

3 Dulbecco Telethon Institute at Endocrinology Unit, University Hospital of Pisa, Via Paradisa 2, 56124 Pisa, Italy4CNR Institute of Food Sciences, Via Roma 64, 83100 Avellino, Italy

Correspondence should be addressed to Marco Mainardi; [email protected]

Received 5 April 2013; Accepted 25 June 2013


Copyright © 2013 Marco Mainardi et al.This is an open access article distributed under theCreativeCommonsAttribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Regulation of feeding behavior has been a crucial step in the interplay between leptin and the arcuate nucleus of the hypothalamus(ARC). On one hand, the basic mechanisms regulating central and peripheral action of leptin are becoming increasingly clear. Onthe other hand, knowledge on how brain sensitivity to leptin can be modulated is only beginning to accumulate. This point is ofparamount importance if one considers that pathologically obese subjects have high levels of plasmatic leptin. A possible strategyfor exploring neural plasticity in the ARC is to act on environmental stimuli. This can be achieved with various protocols, namely,physical exercise, high-fat diet, caloric restriction, and environmental enrichment. Use of these protocols can, in turn, be exploitedto isolate keymolecules with translational potential. In the present review, we summarize present knowledge about themechanismsof plasticity induced by the environment in the ARC. In addition, we also address the role of leptin in extrahypothalamic plasticity,in order to propose an integrated view of how a single diffusible factor can regulate diverse brain functions.

1. Introduction

Food intake is one of the most relevant aspects of metabolichomeostasis. Feeding behavior is the result of the integratedaction of peripheral organs and brain. In particular, thearcuate nucleus of the hypothalamus (ARC) has a prominentrole in sensing long-term energy stores and in consequentlyregulating food intake. Indeed, the ARC is positioned aroundthe third ventricle, where the blood-brain barrier is partic-ularly leaky, and is the primary target organ of leptin. Inparticular, leptin exerts its action by modulating the activityof the twomain cell populations of theARC,NPY, andPOMCneurons.

2. Overview of Neural Circuits ControllingResponse to Leptin

In the ARC, two main neuronal populations of leptin-respo-nsive neurons exist. Their activity has opposite behavioraloutcomes and can be identified according to the expression

of the neuropeptides proopiomelanocortin (POMC) and 𝛼-melanocyte-stimulating hormone (𝛼-MSH), or Neuropep-tide y (NPY) and agouti-related protein (AgRP), respectively[1]. POMC neurons are considered to be the principal cellsof the ARC [2]. However, recent work has shown thatat least a fraction of POMC cells displays a GABAergicphenotype [3]; the precise role of this population still has tobe addressed in detail. On the other hand, NPY cells can beconsidered as ARC interneurons, since they are GABAergicand synapse onto POMC neurons to modulate their activity[2]. In addition to this local role, however, NPY cells also sendprojections to nearby hypothalamic nuclei [4].

Both POMC and NPY neurons express the active form ofthe leptin receptor (ObRb). Leptin has easy access to theARC,since permeability of the blood-brain barrier around thiscircumventricular organ is higher than in the rest of the brain.ObRb binding to its ligand activates in both POMC and NPYneurons two main signal transduction pathways: STAT3 andPI3 Kinase/AKT [5]. However, leptin has opposite actions onthe two neuronal populations of the ARC. Indeed, activation

2 Neural Plasticity

of leptin signaling results in repression of transcription ofthe NPY gene, whereas it has an opposite outcome onPOMC mRNA synthesis [5]. Moreover, leptin is able tomodulate the electrical activity of bothNPY and POMC cells,by acting on L-type calcium channels [6], whereas it alsooperates on TRPC nonselective cation channels on the lattercell type alone [7]. This results in decrease of Ca2+ influxin NPY neurons, which are thus hyperpolarized, whereasan opposite action is exerted on POMC neurons, whichbecome depolarized. The net effect of leptin on the ARC is tofavor POMC neuron activity, in order to promote activity ofanorexigenic pathways at the expense of orexigenic pathways,with the final action of repressing food intake and increasingenergy expenditure.

ARCneurons in turn send projections to nearby hypotha-lamic nuclei, in addition to subcortical areas (preoptic area)and to the nucleus of tractus solitarius of the vagus nerve[4]. Thus, leptin sensing by the ARC is able to interact withandmodulate the activity of several brain regions involved inmetabolism and feeding behavior.

The first experiments aimed at elucidating the role of themediobasal hypothalamus were based on chemical ablation;this coarse approach leads to the development of hyperphagia[8] and contributes to highlight the importance of this brainarea in feeding behavior regulation. More recently, geneticapproaches have permitted amore precise ablation of selectedneural populations of the ARC. Gene ablation of POMCneurons causes hyperphagia [9]. On the other hand, the out-come of AgRP/NPY neurons manipulation is more variable,ranging from the expected hypophagia if performed in adult-hood [10] to paradoxical hyperphagia if performed duringimmediate postnatal development [11]. This age-dependenteffect may reflect behavioral adaptation to precocious loss ofAgRP neurons, in addition to suggesting that other pathways,possibly outside of the ARC, can compensate for loss ofthese neuronal types. However, the preeminence of NPY andPOMC neurons in regulating feeding behavior should not bequestioned. Indeed, elegant experiments using optogeneticactivation of AgRP neurons have clearly shown that activa-tion of this neuronal population leads to immediate voraciousfeeding [12]. Further experiments using a similar approach onPOMC neurons still have to be performed. Recent evidenceshows that targeted pharmacological activation of POMCneurons has an effect on food intake and body weight onlyafter 3 days of treatment, wher as no effect is detectedacutely within 24 hours from treatment [13]. Basing on theseresults, we predict that POMC neuron stimulation wouldresult in the repression of food intake only if performedchronically for several days, highlighting a subtler role ofthese cells in repressing food intake, with a reduction infeeding characterized by a slower kinetic compared to theimmediate effect produced by AgRP neuron stimulation [12].

On the other hand, optogenetics can be exploited alsofor inhibiting the activation of NPY and POMC neurons,by means of halorhodopsin expression [14]. Compared togene ablation, this approach has the advantage of preservingthe neuronal population under investigation, thus, allowingto test the functionality of the circuit without substantiallyaltering it.

3. Leptin and Its Peripheral Effects

Leptin derives from the Greek term lepthos (thin) and iscoded by the ob gene. The ob/ob mice, harboring a pointmutation in this gene, are diabetic, hyperphagic, infertile, andshow other abnormalities including a less efficient immunesystem, high bone mass [15], and impaired thermogenesis[16, 17]. When treated with the recombinant hormone, theirfood intake and body weight are normalized to that of thelean controls and most of the previously mentioned alter-ations are completely rescued [18–21]. Similar observationswere reported for the few human obese subjects carryinga homozygous mutation in the leptin gene that exhibited adramatic phenotype and greatly benefited from recombinantleptin therapy [22].

Leptin is mainly expressed by the white adipose tissue(WAT) [23], and long before its molecular identification itspresence and nature (a hormone) had been postulated byDougColeman in the 1960s, based on parabiosis experimentsthat utilized ob/ob and wild type mice [24]. During theseexperiments ob/obmice lost weight implying that a regulatorof body weight, which they naturallymissed, was provided bythe blood flow of WT mice.

In those years, body weight regulation was modeled byHervey as a typical homeostatic mechanism in which a con-troller (already hypothesized to reside in the hypothalamus)was informed by the status of the energy stores through anafferent signal able to sense them. In turn the controller,through an efferent signal, was able to control the energystores and the afferent signal levels in a feedback loop, typicalof the endocrine mechanisms [25]. Leptin cloning, followedone year later by the identification of its receptor (OB-R)[26], gave a hard support to this hypothesis. Leptin filledperfectly the requirements of the afferent signal: (i) it wasproduced by theWAT, the largest energy store in the body, (ii)its expression was directly related to fat mass [27, 28], (iii) itwas a bioactive molecule [19] able to reduce body weight, andfinally (iv) the active formof its receptor (OB-Rb)was presentin the ARC [29]. Concerning point (ii), the upstream signalslinking increase in adiposity to induction in leptin expressionin WAT are still a matter of investigation: it is known thatinsulin, glucocorticoids, and the tumor necrosis factor alpha(TNF-𝛼) [30, 31] participate to leptin regulation, but evidenceso far accumulated does not account for the fine modulationof this hormone.

Leptin discovery was key to disclose a huge amountof knowledge in the field of body weight homeostasis andclarified which molecular pathways concur to determine thefine tuning of satiety and appetite in the brain.However, someaspects of theHerveymodel regarding the efferent signals stillneed to be tested. Until recently leptin-brain relationship hasin fact been considered a one-way dialogue with no feedbackon the periphery, if we exclude the indirect effects on WATcaused by energy unbalance.

Given the clear physiological properties and behav-ioral outcomes of ARC neurons, the process of metabolichomeostasis regulation appears quite straightforward, withappetite repression and energy expenditure being linearly

Neural Plasticity 3

regulated by plasma leptin concentration. However, obesepatients typically display high leptin levels, which happensin the absence of any leptin receptor mutation [32]. Thisraises the possibility that the pathological phenotype is theresult of cerebral leptin insensitivity, possibly as a result ofwrong set-point hypothalamic circuits to the levels of thishormone. Thus, a crucial issue is to understand if and howleptin sensitivity can be regulated by acting on hypothalamiccircuits.

4. Neural Plasticity Induced byEnvironmental Stimuli in the ArcuateNucleus of the Hypothalamus

Neural plasticity is the process by which neural circuitsadapt their response to variations in stimuli coming fromthe environment. Concerning the hypothalamus, plasticitywas first studied as a result of changes in the internalenvironment, namely, blood concentration of estrogen. It wasfirst shown that GABAergic synapses on ARC neurons arenegatively regulated by estradiol [33]. Subsequent work hasalso demonstrated that synapses formed on dendritic spines,which are mainly excitatory, are sensitive to the estrous cycle[34].

A growing body of the literature addresses the impact onneural plasticity of genetic manipulation [35]. On the otherhand and despite the importance of this topic, surprisinglyfew works address the issue of understanding how theenvironment affects neural plasticity. A better knowledge ofthis process is instrumental in finding the key molecules thatmediate the beneficial effects of environment onmetabolism,with the ultimate goal of designing effective therapies. Oneof the purposes of the present review is to highlight howrecent findings by our and other groups have shed new lighton this challenging topic suggesting a fascinating scenario inwhich the environment, acting as a plasticizing agent on thebrain, is able to modulate leptin production and action in theperiphery.

During immediate postnatal development, leptin hasbeen shown to regulate correct development of ARC pro-jections to the paraventricular nucleus of the hypothalamus(PVH). Indeed, leptin-deficient ob/ob mice display sparserprojections to the paraventricular nucleus of the hypothala-mus compared to wild-type mice [36]. On the other hand,leptin treatment of ob/obmice is able to normalize the densityof these projections, but only if treatment is performedduringimmediate postnatal life, whereas it is ineffective in the adult[36]. This has led to hypothesizing the existence of a “criticalperiod” for proper development of hypothalamic circuits, inclose analogy to what has been shown for cortical sensorysystems [37].

In addition, leptin is able to modulate the firing pat-tern of ARC neurons. Indeed, leptin-deficient ob/ob micedisplay increased inhibitory currents and decreased excita-tory synapses on POMC neurons, whereas NPY neuronsreceive more excitatory synapses. Replacement of leptin isable to normalize these alterations, demonstrating that adulthypothalamic circuits display a plastic potential [38].

On the other hand, one of themost interesting problems isto understand how hypothalamic plasticity can bemodulatedby environmental stimuli, such as the life style or food intake.To this regard, the main experimental models are (i) physicalexercise by running wheel activity; (ii) high-fat diet (HFD)exposure; (iii) caloric restriction. Physical exercise protocolsaim at increasing movement activity of the animal; this isusually achieved by placing a running wheel in the cage,with constant or intermittent access [39]. The number ofanimals that are housed in the same cage is variable, rangingfrom a few units to a single individual. It should be noted,however, that this may represent a confounding factor forresult interpretation, since a low amount of social interaction(i.e., low number of animals in a cage) has been shown toaffect food intake [40]. HFD is achieved by changing the stan-dard chow diet for laboratory rodents with special formulascontaining a higher amount of fat and carbohydrates [41, 42].HFD ultimately leads to pathological obesity and type IIdiabetes and is a valid model for simulating the interplaybetween excessive caloric intake, metabolic syndrome, andbrain [43]. On the other hand, various protocols have beenset up for studying the effect of a reduction in caloric intakeon the brain. The simplest protocol is caloric restrictionthat is based on reducing the total daily intake of food tovalues that span from 70% to 90% of normal intake [44]. Afurther development is represented by intermittent fastingthat consists of a one day on-one day off food availabilityschedule. Since the animals usually experience a reboundfeeding the day after fasting, the total caloric balance, alsoin this case, is usually reduced to 70–90% of normal intake[45, 46].

More recently, environmental enrichment (EE) has beenadded to this list [47]. Compared to the simple physicalexercise protocol, EE adds cognitive, sensory, and socialstimulations bymeans of rearing larger groups of animals (upto ten, compared to the usual three to five) inwider cages withseveral objects to explore that are frequently changed [48, 49].A summary of these protocols is presented in Table 1.

5. Molecular Mediators ofHypothalamic Plasticity

Transition to HFD induces an adaptation in ARC circuitswhich is mediated by a transient increase in the expressionof the extracellular matrix component polysialic acid (PSA),whose role in promoting neural plasticity is well known in thecortex [50].The increase in polysialic acid synthesis correlateswith a change in the firing pattern of POMC neurons, whichdisplay a higher excitatory activity, possibly to compensatethe excess nutrient intake with an inhibition of feeding[51]. Of note, the role of extracellular matrix in regulatingadaptation to dietary change was also demonstrated usingmice deficient for the tissue inhibitor of metalloproteinase-2 (TIMP-2). TIMP-2 knockouts display higher weight gainand hyperleptinaemia after transition to HFD [52]. However,these compensatory reactions will eventually be ineffective,since HFD will result invariably in obesity and loss of leptinsensitivity. On the other hand, enhancing motor activity with

4 Neural Plasticity

Table 1: Summary of the main protocols for studying the interactions between environment, leptin, and neural plasticity, with their mainfeatures and effects on the ARC. The effects of the various rearing protocols on leptin levels, STAT3 phosphorylation (pSTAT3) in the ARC,and neural plasticity are described as compared to the standard rearing condition (first row).

Protocol Animals percage

Runningwheel Objects Food Leptin pSTAT3 in

the ARCNeural plasticity in the

ARCStandardcondition 3 to 5 No No Standard chow diet,

ad libitumControlvalue

Controlvalue Control value

Physical exercise 1 to 3 Yes No Standard chow diet,ad libitum ↓ ↑

Not observed onsynaptic puncta.

No data available forother markers

Diet-induced obesity(DIO) 3 to 5 No No

High-fat,high-carbohydrate diet,

ad libitum↑ ↓

Extracellular MatrixRearrangements

Caloric restriction 3 to 5 No No

Standard chow diet,70% to 90% of normal

caloric intake,sometimes coupled tointermittent fasting

↓No dataavailable No data available

Environmentalenrichment 6 to 10 Yes Yes Standard chow diet,

ad libitum ↓ ↑

Modulation ofexcitation to inhibition

ratio

forced physical exercise has been demonstrated to producelasting improvements in the status of rats with diet-inducedobesity (DIO) fed a high-energy diet. Indeed, a brief periodof physical exercise reduces body weight and enhances theeffects of leptin without requiring a switch to standard diet;moreover, these benefits are maintained even after the end ofthe exercising period. At the hypothalamic level, this corre-lates with increased STAT3 phosphorylation, thus pointingto increased leptin sensitivity [42]. Analogous benefits wasobserved when wild-type mice were subjected to EE sincebirth. Indeed, they showed a reduction in plasma leptin levels,despite no significant differences in food intake or bodyweight compared to controls. Strikingly, this pointed to anenvironmentally induced reprogramming of the set-point forhypothalamic response to the anorexigenic signal transmittedby leptin. This was accompanied by an enhancement inglucose tolerance and food intake repression in response toleptin injection and by an increase in STAT3 phosphorylationin the ARC. In addition, excitatory and inhibitory synapseson the soma of POMC and NPY neurons were quantified tolook for explicit signs of neural plasticity. It was found thatEE results in an increase in the excitatory tone on POMCneurons and in opposite changes on NPY neurons. Notably,in comparison to physical exercise alone, rearrangementof hypothalamic synapses were observed only in EE mice,whereas increased leptin sensitivity and STAT3 phosphory-lation were common to both conditions [47]. It is worthnoting that, if EE is terminated, not all these changes aremaintained. Indeed, the increased STAT3 phosphorylation inresponse to leptin is no more observed, whereas an imprinton ARC synapses can still be observed [47]. It is temptingto speculate that this imprint will help in reinstating higherSTAT3 phosphorylation in the ARC, after transitioning againto the EE protocol.Moreover, further workwould be required

to investigate whether the precocious EE experience can alsoinduce resistance to diet-induced obesity.

An important aspect of studies on experience-inducedplasticity is the search for the keymolecular mediators. In thecase of EE, we have shown that enriched animals displayedan increased synthesis of brain-derived neurotrophic factor(BDNF) [47]. Interestingly, this molecule is the ideal bridgebetween control of feeding behavior and plasticity. Firstly,BDNF-deficientmice are obese [53]. Projections fromPOMCneurons of the ARC act on the ventromedial hypothalamicnucleus (VMH) to induce synthesis of BDNF via activationof the melanocortin-4 receptor (MC4R) [54]. BDNF in turnis anterogradely transported to synaptic sites, where it isreleased in an activity-dependent manner [55]. One of theeffects of synaptically released BDNF is a facilitation inconsolidation of neural circuits, since this neurotrophin isrequired for the establishment of long-term potentiation(LTP) [56]. However, the precise role of BDNF in repressingfood intake still has to be completely elucidated [57].

On the other hand, BDNF had initially been shown to bea master controller of plasticity. BDNF-deficient mice displaycognitive impairments and reduced long-term potentiationin the hippocampus [58, 59]. In addition, manipulation ofBDNF levels also modulates cortical plasticity [60]. Strik-ingly, increased BDNF levels are observed as a result ofEE, both during the immediate postnatal development andadulthood [61, 62].

In addition, recent work has uncovered a previouslyunknown and beneficial role of the increase of BDNF causedby exposure to EE. In mice inoculated with B16 melanomacells or in the MC38 colon cancer model, EE amelioratedthe pathological phenotype by reducing tumor size [63].Of note, a decrease in serum leptin level was observed inthose experiments, as also reported by us [47]. Strikingly,

Neural Plasticity 5

hypothalamic overexpression of BDNF per se was able toreplicate the effect of EE on cancer outcome [63]. Moreover,the same results could not be obtained by exposing mice torunning wheel activity alone. In the same line of findings isour observation that EE, but not physical exercise, was ableto induce synaptic plasticity [47]. Another interesting findingby Cao and colleagues [63] is that the beneficial effects ofhypothalamic BDNF are due to activation of adrenergic 𝛽receptors, since administration of the 𝛽-blocker propranololabolishes the decrease in serum leptin associated with EE.Taken together, these results add functional value to thedirect polysynaptic projection from the ARC to adiposetissue that has been described anatomically [64] and lead tohypothesizing that the feedback from the brain to WAT canconsist of sympathetic fibers.

Another interesting aspect of the effects of EE on theleptin-hypothalamus-adipose tissue axis is the induction ofWAT transdifferentiation. Indeed, the appearance of multi-loculated adipose cells, characterized by a gene expressionprofile resembling that of brown adipose tissue (BAT) cells,has been described in EE mice [65] and put in correlationwith resistance to diet-induced obesity. Even if the resultsshould be interpreted cautiously, since the perirenal adiposetissue contains per se a fraction of brown adipose tissue (Maf-fei M and Barone I, unpublished observation), these “brite”cells can be responsible for the peripheral changes leading toimproved metabolic homeostasis in animals undergoing EE.Again, these effects appear downstream of sympathetic out-put to the adipose tissue driven by hypothalamic activation[65].

But what translates the beneficial effects of a period ofenhanced physical, cognitive, and social stimulations intolasting changes of neural circuits? A first hint comes fromthe experiments showing rapid modulation of ARC synapticcurrents by leptin [66]. However, these seminal experimentsonly showed the acute effect of leptin on neural activity. Ourgroup has shown that EE induces robust synaptic plasticityin the ARC, namely, through a global reduction in inhibitionand an opposite change of excitation [47]. We have analyzedmore in detail ARC synaptology in mice raised in EE sincebirth and found that POMC neurons receive more excitatoryinputs, whereas more inhibitory synapses impinge on NPYneurons [47]. These findings indicate that (i) modulation ofexcitation/inhibition ratio is related to the metabolic changesobserved in EE and (ii) the increase in leptin sensitivity thatoccurs in mice raised in EE since birth could be explainedby an increase in the excitatory input on POMC neurons. Ofnote, plasticity in the ARC appears to be regulated accordingto the same mechanisms observed in the cerebral cortex[62, 67]. Thus, modulation of the excitation/inhibition ratioseems to be a common motif in brain plasticity.

A further factor to be taken into account is that theARC has been shown to be one of those few sites whereneurogenesis takes place also in adulthood [68]. Moreover,suppression of neurogenesis has been observed in the ARCof mice with diet-induced obesity [41]. Of note, EE inducesan increase of adult neurogenesis in the hippocampus [69],whereas the ARC has not been yet investigated to this regard.Although still to be fully demonstrated, we could then build

a hypothetical scenario in analogy to what is observed inthe hippocampus: neurogenesis in the ARC may act as afurther mechanism to sense environmental stimulation andtranslates it into a sort of “metabolic” memory.

In this context, ARC neurogenesis can also be seenas an important source of neurons to be integrated withnewly formed networks and can collaborate with synapticrearrangements to the reprogramming of leptin sensitivityobserved in EE.

6. Leptin and Nonhypothalamic Plasticity

Obesity has also been linked to the induction of cognitivedeficits [70], but the underlying mediators are still poorlyunderstood. A growing body of evidence is now pointing toleptin as a regulator of plasticity also outside of hypothalamicand metabolism-related circuits. In fact, expression of leptinreceptor isoforms is observed in various parts of the brain,including the hippocampus and cortex [71].

Induction of hyperleptinaemia in rats results in impair-ment in hippocampal synaptic plasticity, as assessed withinduction of long-term potentiation by stimulation of theSchaffer collateral pathway. This deficit is reversed by low-ering blood leptin levels by mild caloric restriction [72]. Inaddition, fasting induces activation of the ERKpathway in thehypothalamus, and ob/ob mice are deficient in this response[73]. ERK is one of the key transducers of plasticity-inducingstimuli in the cortex [74, 75]. Further work will be required toassess whether ERK is also sensitive to leptin action outsidethe hypothalamus. Of note, caloric restriction per se has beenshown to be a powerful inducer of neural plasticity, up to thepoint of restoring visual cortical plasticity in adult rats [76].In this paper, leptin levels were not measured, but it is highlylikely that caloric restriction resulted in their decrease, asalready shown [77]. Moreover, EE has been shown to restorecortical plasticity in adult rats [62] in addition to decreasingplasmatic leptin levels and increasing leptin sensitivity [47].This suggests that interfering with leptin action on corticalcircuits can be a valuable strategy to restore plasticity. Animalfindings can be put in correlation with analogous results inhumans, showing that three months of caloric restrictioncause an improved cognitive performance [78].

On the other hand, leptin supplementation also can exerta positive effect on plasticity and cognitive performance.Indeed, leptin supplementation has been shown to restoreproper cognitive development in a pediatric patient with amutated leptin gene [79]. Accordingly, hippocampal long-termpotentiation and performance in theMorrisWaterMazetest are impaired in db/db mice and in the hyperleptinaemicZucker rat strain [80]. Moreover, leptin increases hippocam-pal neurogenesis [81].

Thus, it is tempting to speculate that also at the extrahy-pothalamic level, lower circulating leptin associated to higherleptin sensitivity can be relevant for adult plasticity, whereasin immediate postnatal life adequate levels of this hormoneare required to support correct development of neural cir-cuits.

6 Neural Plasticity

7. Conclusions and Future Perspectives

Increasing experimental evidence is showing that leptin isnot only a hormone but also a molecular link betweenmetabolism, neural plasticity, and cognition. Consideringthat the ARC is the primary sensor of leptin, it also appears tobe a possible orchestrator of the polyvalent effects of leptin,also if one takes into account its widespread projections. Onthe other hand, thewide expression ofObRb in extrahypotha-lamic sites calls for a search of direct effects of leptin.

Experiments performed using modulation of environ-mental inputs, either by EE, physical exercise, or variation ofcaloric intake, point to a crucial role of BDNF, excitatory tone,extracellular matrix, and neurogenesis in neural plasticityassociated with variations in leptin levels. Notably, theseparameters have long been known asmediators of hippocam-pal and cortical plasticity [82, 83], pointing to the existence ofcommon principles regulating adaptation of neural circuitsto variations in experience throughout the brain. Moreover,all these variables have been shown to be regulated by EE, inaddition to the reduction in leptin levels. On the other hand,efferent signals originating in the brain to regulate activityof WAT (the main leptin-producing site) are beginning toemerge, with remarkable transdifferentiation effect on theadipocyte that can be switched to a “brite” phenotype incertain conditions.

We believe that future experimental efforts should beaimed at establishing a clear hierarchy between all factorsmentioned, in order to build a coherent picture of the leptin-brain interaction. This will be an instrumental step for iso-lating the master controller(s) of hypothalamic and extrahy-pothalamic plasticity, with the ultimate goal of designingeffective therapies for metabolic syndromes.

Acknowledgments

This work was supported by grant “SEED project EXTRA-PLAST IIT” to Tommaso Pizzorusso.

References

[1] M.W. Schwartz, S. C. Woods, D. Porte Jr., R. J. Seeley, and D. G.Baskin, “Central nervous system control of food intake,”Nature,vol. 404, no. 6778, pp. 661–671, 2000.

[2] E. E. Jobst, P. J. Enriori, and M. A. Cowley, “The electro-physiology of feeding circuits,” Trends in Endocrinology andMetabolism, vol. 15, no. 10, pp. 488–499, 2004.

[3] M. S. Dicken, R. E. Tooker, and S. T. Hentges, “Regulation ofgaba and glutamate release from proopiomelanocortin neuronterminals in intact hypothalamic networks,” The Journal ofNeuroscience, vol. 32, no. 12, pp. 4042–4048, 2012.

[4] M. G. Myers Jr., H. Munzberg, G. M. Leinninger, and R. L.Leshan, “The geometry of leptin action in the brain: morecomplicated than a simple ARC,” Cell Metabolism, vol. 9, no. 2,pp. 117–123, 2009.

[5] L. Plum, B. F. Belgardt, and J. C. Bruning, “Central insulin actionin energy and glucose homeostasis,” The Journal of ClinicalInvestigation, vol. 116, no. 7, pp. 1761–1766, 2006.

[6] J.-H.Wang, F.Wang,M.-J. Yang et al., “Leptin regulated calciumchannels of neuropeptide Y and proopiomelanocortin neurons

by activation of different signal pathways,” Neuroscience, vol.156, no. 1, pp. 89–98, 2008.

[7] J. Qiu, Y. Fang, O. K. Rønnekleiv, and M. J. Kelly, “Leptinexcites proopiomelanocortin neurons via activation of TRPCchannels,”The Journal of Neuroscience, vol. 30, no. 4, pp. 1560–1565, 2010.

[8] H. T. Bergen, T. M. Mizuno, J. Taylor, and C. V. Mobbs,“Hyperphagia and weight gain after gold-thioglucose: relationto hypothalamic neuropeptide Y and proopiomelanocortin,”Endocrinology, vol. 139, no. 11, pp. 4483–4488, 1998.

[9] L. Yaswen, N. Diehl, M. B. Brennan, and U. Hochgeschwender,“Obesity in the mouse model of pro-opiomelanocortin defi-ciency responds to peripheral melanocortin,” Nature Medicine,vol. 5, no. 9, pp. 1066–1070, 1999.

[10] E. Gropp, M. Shanabrough, E. Borok et al., “Agouti-relatedpeptide-expressing neurons are mandatory for feeding,”NatureNeuroscience, vol. 8, no. 10, pp. 1289–1291, 2005.

[11] A. Mesaros, S. B. Koralov, E. Rother et al., “Activation of Stat3signaling in AgRP neurons promotes locomotor activity,” CellMetabolism, vol. 7, no. 3, pp. 236–248, 2008.

[12] Y. Aponte, D. Atasoy, and S. M. Sternson, “AGRP neurons aresufficient to orchestrate feeding behavior rapidly and withouttraining,” Nature Neuroscience, vol. 14, no. 3, pp. 351–355, 2011.

[13] C. Zhan, J. Zhou, Q. Feng et al., “Acute and long-term suppres-sion of feeding behavior by POMC neurons in the brainstemand hypothalamus, respectively,” The Journal of Neuroscience,vol. 33, no. 8, pp. 3624–3632, 2013.

[14] F. Zhang, V. Gradinaru, A. R. Adamantidis et al., “Optogeneticinterrogation of neural circuits: technology for probing mam-malian brain structures,”Nature Protocols, vol. 5, no. 3, pp. 439–456, 2010.

[15] V. K. Yadav, F. Oury, N. Suda et al., “A serotonin-dependentmechanism explains the leptin regulation of bone mass,appetite, and energy expenditure,” Cell, vol. 138, no. 5, pp. 976–989, 2009.

[16] J. M. Friedman and J. L. Halaas, “Leptin and the regulation ofbody weight in mammals,” Nature, vol. 395, no. 6704, pp. 763–770, 1998.

[17] P. Lindstrom, “The physiology of obese-hyperglycemic mice[ob/ob mice],”The Scientific World Journal, vol. 7, pp. 666–685,2007.

[18] F. F. Chehab, “A broader role for leptin,”Nature Medicine, vol. 2,no. 7, pp. 723–724, 1996.

[19] J. L. Halaas, K. S. Gajiwala, M. Maffei et al., “Weight-reducingeffects of the plasma protein encoded by the obese gene,”Science, vol. 269, no. 5223, pp. 543–546, 1995.

[20] M. A. Pelleymounter, M. J. Cullen, M. B. Baker et al., “Effectsof the obese gene product on body weight regulation in ob/obmice,” Science, vol. 269, no. 5223, pp. 540–543, 1995.

[21] J. Salvador, J. Gomez-Ambrosi, andG. Fuhbeck, “Perspectives inthe therapeutic use of leptin,” Expert Opinion on Pharmacother-apy, vol. 2, no. 10, pp. 1615–1622, 2001.

[22] I. Sadaf Farooqi and S. O’Rahilly, “Leptin: a pivotal regulatorof human energy homeostasis,” American Journal of ClinicalNutrition, vol. 89, no. 3, pp. 980S–984S, 2009.

[23] Y. Zhang, R. Proenca, M. Maffei, M. Barone, L. Leopold, and J.M. Friedman, “Positional cloning of the mouse obese gene andits human homologue,” Nature, vol. 372, no. 6505, pp. 425–432,1994.

[24] D. L. Coleman, “A historical perspective on leptin,” NatureMedicine, vol. 16, no. 10, pp. 1097–1099, 2010.

Neural Plasticity 7

[25] M.-D. Li, “Leptin and beyond: an odyssey to the central controlof body weight,” Yale Journal of Biology and Medicine, vol. 84,no. 1, pp. 1–7, 2011.

[26] G.-H. Lee, R. Proenca, J. M.Montez et al., “Abnormal splicing ofthe leptin receptor in diabetic mice,” Nature, vol. 379, no. 6566,pp. 632–635, 1996.

[27] E. Ravussin, R. E. Pratley,M.Maffei et al., “Relatively lowplasmaleptin concentrations precede weight gain in Pima Indians,”Nature Medicine, vol. 3, no. 2, pp. 238–240, 1997.

[28] M. Maffei, H. Fei, G.-H. Lee et al., “Increased expression inadipocytes of ob RNA inmice with lesions of the hypothalamusand with mutations at the db locus,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 92, no.15, pp. 6957–6960, 1995.

[29] L. A. Tartaglia, M. Dembski, X. Weng et al., “Identification andexpression cloning of a leptin receptor, OB-R,” Cell, vol. 83, no.7, pp. 1263–1271, 1995.

[30] M. E. Trujillo,M.-J. Lee, S. Sullivan et al., “Tumor necrosis factor𝛼 and glucocorticoid synergistically increase leptin productionin human adipose tissue: role for p38mitogen-activated proteinkinase,” Journal of Clinical Endocrinology and Metabolism, vol.91, no. 4, pp. 1484–1490, 2006.

[31] S. K. Fried, M. R. Ricci, C. D. Russell, and B. Laferrere,“Regulation of leptin production in humans,” The Journal ofNutrition, vol. 130, no. 12, pp. 3127S–3131S, 2000.

[32] M. Mapfei, J. Halaas, E. Ravussin et al., “Leptin levels in humanand rodent:measurement of plasma leptin and obRNA in obeseand weight-reduced subjects,”NatureMedicine, vol. 1, no. 11, pp.1155–1161, 1995.

[33] A. Parduca, J. Perez, and L. M. Garcia-Segura, “Estradiolinduces plasticity of GABAergic synapses in the hypothalamus,”Neuroscience, vol. 53, no. 2, pp. 395–401, 1993.

[34] E. Csakvari, Z. Hoyk, A. Gyenes, D. Garcia-Ovejero, L. M.Garcia-Segura, and A. Parducz, “Fluctuation of synapse densityin the arcuate nucleus during the estrous cycle,” Neuroscience,vol. 144, no. 4, pp. 1288–1292, 2007.

[35] L. M. Zeltser, R. J. Seeley, and M. H. Tschop, “Synapticplasticity in neuronal circuits regulating energy balance,”NatureNeuroscience, vol. 15, no. 10, pp. 1336–1342, 2012.

[36] S. G. Bouret, S. J. Draper, and R. B. Simerly, “Trophic action ofleptin on hypothalamic neurons that regulate feeding,” Science,vol. 304, no. 5667, pp. 108–110, 2004.


[38] S. Pinto, A. G. Roseberry, H. Liu et al., “Rapid rewiring ofarcuate nucleus feeding circuits by leptin,” Science, vol. 304, no.5667, pp. 110–115, 2004.

[39] C. M. Novak, P. R. Burghardt, and J. A. Levine, “The use of arunning wheel to measure activity in rodents: relationship toenergy balance, general activity, and reward,” Neuroscience andBiobehavioral Reviews, vol. 36, no. 3, pp. 1001–1014, 2012.

[40] R. A. Hammond, “Social influence and obesity,” Current Opin-ion in Endocrinology,Diabetes andObesity, vol. 17, no. 5, pp. 467–471, 2010.

[41] D. E. G. McNay, N. Briancon, M. V. Kokoeva, E. Maratos-Flier, and J. S. Flier, “Remodeling of the arcuate nucleus energy-balance circuit is inhibited in obesemice,”The Journal of ClinicalInvestigation, vol. 122, no. 1, pp. 142–152, 2012.

[42] C. M. Patterson, S. G. Bouret, A. A. Dunn-Meynell, and B.E. Levin, “Three weeks of postweaning exercise in DIO rats

produces prolonged increases in central leptin sensitivity andsignaling,” American Journal of Physiology, vol. 296, no. 3, pp.R537–R548, 2009.

[43] S. K. Panchal and L. Brown, “Rodent models for metabolicsyndrome research,” Journal of Biomedicine and Biotechnology,vol. 2011, Article ID 351982, 14 pages, 2011.

[44] E. M. Mercken, B. A. Carboneau, S. M. Krzysik-Walker, and R.de Cabo, “Of mice and men: the benefits of caloric restriction,exercise, and mimetics,” Ageing Research Reviews, vol. 11, no. 3,pp. 390–398, 2012.

[45] B. Martin, M. P. Mattson, and S. Maudsley, “Caloric restrictionand intermittent fasting: two potential diets for successful brainaging,”Ageing Research Reviews, vol. 5, no. 3, pp. 332–353, 2006.

[46] A. Coppola, Z.-W. Liu, Z. B. Andrews et al., “A centralthermogenic-like mechanism in feeding regulation: an inter-play between arcuate nucleus T3 and UCP2,” Cell Metabolism,vol. 5, no. 1, pp. 21–33, 2007.

[47] M. Mainardi, G. Scabia, T. Vottari et al., “A sensitive periodfor environmental regulation of eating behavior and leptinsensitivity,” Proceedings of the National Academy of Sciences ofthe United States of America, vol. 107, no. 38, pp. 16673–16678,2010.

[48] A. Sale, N. Berardi, and L. Maffei, “Enrich the environment toempower the brain,” Trends in Neurosciences, vol. 32, no. 4, pp.233–239, 2009.

[49] H. van Praag, G. Kempermann, and F. H. Gage, “Neuralconsequences of environmental enrichment,” Nature ReviewsNeuroscience, vol. 1, no. 3, pp. 191–198, 2000.

[50] G. Di Cristo, B. Chattopadhyaya, S. J. Kuhlman et al., “Activity-dependent PSA expression regulates inhibitory maturation andonset of critical period plasticity,” Nature Neuroscience, vol. 10,no. 12, pp. 1569–1577, 2007.

[51] A. Benani, C. Hryhorczuk, A. Gouaze et al., “Food intakeadaptation to dietary fat involves PSA-dependent rewiringof the arcuate melanocortin system in mice,” The Journal ofNeuroscience, vol. 32, no. 35, pp. 11970–11979, 2012.

[52] G. Lluri, G. D. Langlois, P. D. Soloway, and D. M. Jaworski,“Tissue inhibitor of metalloproteinase-2 (TIMP-2) regulatesmyogenesis and 𝛽1 integrin expression in vitro,” ExperimentalCell Research, vol. 314, no. 1, pp. 11–24, 2008.

[53] J. Gray, G. S. H. Yeo, J. J. Cox et al., “Hyperphagia, severe obesity,impaired cognitive function, and hyperactivity associated withfunctional loss of one copy of the brain-derived neurotrophicfactor (BDNF) gene,” Diabetes, vol. 55, no. 12, pp. 3366–3371,2006.

[54] B. Xu, E. H. Goulding, K. Zang et al., “Brain-derived neu-rotrophic factor regulates energy balance downstream ofmelanocortin-4 receptor,” Nature Neuroscience, vol. 6, no. 7, pp.736–742, 2003.

[55] C. A. Altar and P. S. Distefano, “Neurotrophin trafficking byanterograde transport,” Trends in Neurosciences, vol. 21, no. 10,pp. 433–437, 1998.

[56] A. Barco, S. Patterson, J. M. Alarcon et al., “Gene expressionprofiling of facilitated L-LTP in VP16-CREB mice reveals thatBDNF is critical for the maintenance of LTP and its synapticcapture,” Neuron, vol. 48, no. 1, pp. 123–137, 2005.

[57] J. Cordeira and M. Rios, “Weighing in the role of BDNF in thecentral control of eating behavior,”Molecular Neurobiology, vol.44, no. 3, pp. 441–448, 2011.

8 Neural Plasticity

[58] L. Minichiello, A. M. Calella, D. L. Medina, T. Bonhoeffer, R.Klein, andM. Korte, “Mechanism of TrkB-mediated hippocam-pal long-term potentiation,” Neuron, vol. 36, no. 1, pp. 121–137,2002.

[59] J.-S. Mu, W.-P. Li, Z.-B. Yao, and X.-F. Zhou, “Deprivationof endogenous brain-derived neurotrophic factor results inimpairment of spatial learning andmemory in adult rats,” BrainResearch, vol. 835, no. 2, pp. 259–265, 1999.

[60] A. Bartoletti, L. Cancedda, S. W. Reid et al., “Heterozygousknock-out mice for brain-derived neurotrophic factor showa pathway-specific impairment of long-term potentiation butnormal critical period for monocular deprivation,”The Journalof Neuroscience, vol. 22, no. 23, pp. 10072–10077, 2002.

[61] L. Cancedda, E. Putignano, A. Sale, A. Viegi, N. Berardi,and L. Maffei, “Acceleration of visual system development byenvironmental enrichment,” The Journal of Neuroscience, vol.24, no. 20, pp. 4840–4848, 2004.

[62] A. Sale, J. F. Maya Vetencourt, P. Medini et al., “Environ-mental enrichment in adulthood promotes amblyopia recoverythrough a reduction of intracortical inhibition,” Nature Neuro-science, vol. 10, no. 6, pp. 679–681, 2007.

[63] L. Cao, X. Liu, E.-J. D. Lin et al., “Environmental and geneticactivation of a brain-adipocyte BDNF/leptin axis causes cancerremission and inhibition,” Cell, vol. 142, no. 1, pp. 52–64, 2010.

[64] S. Stanley, S. Pinto, J. Segal et al., “Identification of neuronalsubpopulations that project from hypothalamus to both liverand adipose tissue polysynaptically,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 107, no.15, pp. 7024–7029, 2010.

[65] L. Cao, E. Y. Choi, X. Liu et al., “White to brown fat phenotypicswitch induced by genetic and environmental activation of ahypothalamic-adipocyte axis,”CellMetabolism, vol. 14, no. 3, pp.324–338, 2011.

[66] M. A. Cowley, J. L. Smart, M. Rubinstein et al., “Leptin activatesanorexigenic POMC neurons through a neural network in thearcuate nucleus,” Nature, vol. 411, no. 6836, pp. 480–484, 2001.

[67] M. Mainardi, S. Landi, L. Gianfranceschi et al., “Environmen-tal enrichment potentiates thalamocortical transmission andplasticity in the adult rat visual cortex,” Journal of NeuroscienceResearch, vol. 88, no. 14, pp. 3048–3059, 2010.

[68] M. V. Kokoeva, H. Yin, and J. S. Flier, “Neurogenesis in thehypothalamus of adult mice: potential role in energy balance,”Science, vol. 310, no. 5748, pp. 679–683, 2005.

[69] J. Brown, C. M. Cooper-Kuhn, G. Kempermann et al., “En-riched environment and physical activity stimulate hippocam-pal but not olfactory bulb neurogenesis,” European Journal ofNeuroscience, vol. 17, no. 10, pp. 2042–2046, 2003.

[70] Z. B. Yu, S. P. Han, X. G. Cao, and X. R. Guo, “Intelligencein relation to obesity: a systematic review and meta-analysis,”Obesity Reviews, vol. 11, no. 9, pp. 656–670, 2010.

[71] J. G. Mercer, N. Hoggard, L. M. Williams, C. B. Lawrence, L.T. Hannah, and P. Trayhurn, “Localization of leptin receptormRNA and the long form splice variant (Ob-Rb) in mousehypothalamus and adjacent brain regions by in situ hybridiza-tion,” FEBS Letters, vol. 387, no. 2-3, pp. 113–116, 1996.

[72] C. A. Grillo, G. G. Piroli, A. N. Evans et al., “Obesity/hyper-leptinemic phenotype adversely affects hippocampal plasticity:effects of dietary restriction,” Physiology and Behavior, vol. 104,no. 2, pp. 235–241, 2011.

[73] Y. Morikawa, E. Ueyama, and E. Senba, “Fasting-inducedactivation of mitogen-activated protein kinases (ERK/p38) in

the mouse hypothalamus,” Journal of Neuroendocrinology, vol.16, no. 2, pp. 105–112, 2004.

[74] S. Davis and S. Laroche, “Mitogen-activated protein kinase/extracellular regulated kinase signalling and memory stabiliza-tion: a review,” Genes, Brain and Behavior, vol. 5, supplement 2,pp. 61–72, 2006.

[75] G. Di Cristo, N. Berardi, L. Cancedda et al., “Requirement ofERK activation for visual cortical plasticity,” Science, vol. 292,no. 5525, pp. 2337–2340, 2001.


[77] K. P. Kinzig, S. L. Hargrave, and E. E. Tao, “Central and periph-eral effects of chronic food restriction and weight restorationin the rat,” American Journal of Physiology, vol. 296, no. 2, pp.E282–E290, 2009.

[78] A. V. Witte, M. Fobker, R. Gellner, S. Knecht, and A. Floel,“Caloric restriction improves memory in elderly humans,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 106, no. 4, pp. 1255–1260, 2009.

[79] G. J. Paz-Filho, T. Babikian, R. Asarnow et al., “Leptin replace-ment improves cognitive development,” PLoS ONE, vol. 3, no. 8,Article ID e3098, 2008.

[80] X.-L. Li, S. Aou, Y. Oomura, N. Hori, K. Fukunaga, and T. Hori,“Impairment of long-term potentiation and spatial memory inleptin receptor-deficient rodents,” Neuroscience, vol. 113, no. 3,pp. 607–615, 2002.

[81] J. C. Garza, M. Guo, W. Zhang, and X.-Y. Lu, “Leptin increasesadult hippocampal neurogenesis in vivo and in vitro,” TheJournal of Biological Chemistry, vol. 283, no. 26, pp. 18238–18247,2008.

[82] M. Caleo and L. Maffei, “Neurotrophins and plasticity in thevisual cortex,” Neuroscientist, vol. 8, no. 1, pp. 52–61, 2002.

[83] K. K. Cowansage, J. E. Ledoux, and M.-H. Monfils, “Brain-derived neurotrophic factor: a dynamic gatekeeper of neuralplasticity,”CurrentMolecular Pharmacology, vol. 3, no. 1, pp. 12–29, 2010.


Research ArticleSystem Consolidation of Spatial Memories in Mice:Effects of Enriched Environment

Joyce Bonaccorsi,1,2 Simona Cintoli,2,3 Rosa Mastrogiacomo,3 Sigrid Baldanzi,3

Chiara Braschi,2 Tommaso Pizzorusso,2,4 Maria Cristina Cenni,2 and Nicoletta Berardi2,4

1 Scuola Normale Superiore, Piazza dei Cavalieri 7, 56100 Pisa, Italy2 Institute of Neuroscience of the CNR, 56124 Pisa, Italy3 Pisa University, 56100 Pisa, Italy4Department of Neuroscience, Psychology, Drug Research and Child Health, NEUROFARBA, Universita di Firenze,Viale Pieraccini, 50100 Firenze, Italy

Correspondence should be addressed to Nicoletta Berardi; [email protected]


Academic Editor: Andrea Guzzetta

Copyright © 2013 Joyce Bonaccorsi et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Environmental enrichment (EE) is known to enhance learning and memory. Declarative memories are thought to undergo a firstrapid and local consolidation process, followed by a prolonged process of system consolidation, which consist in a time-dependentgradual reorganization of brain regions supporting remote memory storage and crucial for the formation of enduring memories.At present, it is not known whether EE can affect the process of declarative memory system consolidation. We characterized thetime course of hippocampal and cortical activation following recall of progressively more remote spatial memories.Wild-typemiceeither exposed to EE for 40 days or left in standard environment were subjected to spatial learning in the Morris water maze andto the probe test 1, 10, 20, 30, and 50 days after learning. Following the probe test, regional expression of the inducible immediateearly gene c-Fos wasmapped by immunohistochemistry, as an indicator of neuronal activity.We found that activation of themedialprefrontal cortex (mPFC), suggested to have a privileged role in processing remote spatial memories, was evident at shorter timeintervals after learning in EEmice; in addition, EE induced the progressive activation of a distributed cortical network not activatedin non-EE mice. This suggests that EE not only accelerates the process of mPFC recruitment but also recruits additional corticalareas into the network supporting remote spatial memories.

1. Introduction

Environmental enrichment (EE) is an experimental protocolclassically defined as “a combination of complex inanimateand social stimulation” [1] and which provides animals withthe opportunity to attain high levels of voluntary physicalactivity on running wheels and to enhance exploration,cognitive activity, and social interaction. Several studies pointout that animals reared in EE show marked brain changesat functional, anatomical, and molecular levels [2–13] andin particular changes in plasticity factors and mechanisms[14, 15]. EE can indeed be used as a noninvasive strategy tomodulate brain plasticity throughout life; EE can acceleratethe development of the central nervous system [16–18], can

reopen plasticity windows in the adult cortex [19, 20] andcauses a significant improvement in learning and memory[14, 21–23], especially evident in aged animals [22, 24–32], orin animal models of neurodegenerative diseases [33–35].

Declarative memories depend initially on the medialtemporal lobe system, including the hippocampus, but, overdays to weeks, as these memories mature, they becomeincreasingly dependent on other brain regions such as theneocortex [36–39]. This process of time-dependent gradualreorganization of the brain regions that supports remotememory storage and underlies the formation of enduringmemories, is known as system-level memory consolidationor system consolidation [40, 41]. It has been demonstratedthat the progressive stabilization of long-lasting memories is

2 Neural Plasticity

due to the reactivation of hippocampal-cortical connections[42, 43] and the strengthening of corticocortical connections,involving cortical plasticity mechanisms [41, 44–48]. Despiteall the evidences showing that EE enhances cortical plasticityand learning and memory, there is no evidence on whetherand how EE could affect the process of time-dependentsystem consolidation.

This study aimed at testing whether EE could affect thesystem-level memory consolidation process using a spatialmemory paradigm and characterizing the time course ofhippocampal and of cortical activation following recall ofprogressively more remote memories.

Spatial memory is a declarative type of memory.The hip-pocampus plays an essential role in the formation of spatialmemories [49–53]; subsequently, spatial memories become,in a gradualmanner, additionally dependent on other corticalregions. Many studies point out that the privileged final stor-age site for remote spatial memories is the prefrontal cortex(PFC) and in particular the medial PFC (mPFC), includingthe anterior cingulate, prelimbic, and infralimbic cortices[47, 48, 54, 55]. Recently, however, it has been suggested thatremote memories recall involves the coordinated activationof a broader network of cortical brain regions [56–58]. Wehave therefore characterized the time course of activation ofa number of cortical areas in addition to themPFC.We foundthat EE not only induces an earlier recruitment of mPFC,but also induces the progressive activation of a distributedcortical network that is not activated in standard housedmice.

2. Materials and Methods

2.1. Animal Treatment. Male and female C57BL/6 mice of 2months of age were used in this study. All the procedureswere approved by the Italian Ministry of Health.The animalswere housed in an animal room with a 12 h/12 h light/darkcycle, with food and water available ad libitum. At 2 monthsof age, the animals were assigned to one of the followingrearing conditions for 40 days: environmental enrichment(EE, 𝑛 = 24) or standard condition (SC, 𝑛 = 24). The SCrearing condition consisted of a 26×18×18 cm cage housing 3animals.The EE rearing condition was achieved using a largecage (44 × 62 × 28 cm) containing several food hoppers, onerunning wheel for voluntary physical exercise, and differentlyshaped objects (tunnels, toys, shelters, stairs) that wererepositioned twice a week and completely substituted withothers once a week [33]. Two additional groups of controlanimals, age and gender matched to SC and EE groups, werehoused in home cage standard condition (HC-SC, 𝑛 = 8) orin home cage enriched condition (HC-EE, 𝑛 = 7), and theydid not perform any behavioural task.

2.2. Morris Water Maze (MWM). The hidden platform ver-sion of theMWM test was performed [59]. A large water tankof 120 cm of diameter was filled with white opaque water at22∘C. An escape platform of 11 cm of side was submerged 1 cmbelow the water surface and placed in the center of the SWquadrant.The platformwasmaintained in this position for all

the swim trials through the test. Mice were trained to swimto the platform in 4 daily trials, starting in pseudorandomlyvaried locations, with a 30min interval, during 7 consecutivedays. The trial was complete once the mouse found theplatform or 60 sec had elapsed. If the mouse failed to find theplatform on a given trial, the experimenter guided the mouseonto the platform. Once reaching the platform, each mousewas allowed to rest for 20 s on it. After each trial each mousewas returned to its home cage where it rested until the nexttrial. After the completion of training, spatial memory wasassessed in a probe test; a recall probe trial was performedafter 1, 10, 20, 30, and 50 days after the end of learning. Weused an automated tracking system (Noldus Ethovision XT)for recording behavioural data from training and probe tests.For each trial we measured the latency (in sec) to reach theplatform, the total distance (in cm) swam in order to reachthe platform, and the average swim speed (in cm/s). For eachprobe trial we measured the amount of time spent in thetarget zone (23 cm in radius, centered on the location of theplatform during training) and the average time spent in threeother equivalent zones in each quadrant [55, 60].

2.3. Immunohistochemistry. Mice were anaesthetized andperfused via intracardiac infusion with 0.1M PBS and then4% paraformaldehyde (PFA, dissolved in 0.1M phosphatebuffer, pH 7.4) 90min after the completion of behavioraltesting. Brains were removed, fixed overnight in PFA, andthen transferred to 30% sucrose solution and stored at 4∘C.Coronal sections were cut at 40 𝜇m thickness on a freez-ing microtome (Sliding Leica microtome SM2010R, LeicaMicrosystems), and free-floating sections were prepared forimmunohistochemistry. After a blocking step in 10% NGSand 0.5% Triton X-100 in PBS, sections were incubated in asolution containing 1% NGS, 0.3% Triton X-100, and anti-c-Fos primary rabbit polyclonal antibody (1 : 3000 rabbit antic-Fos polyclonal antibody, Calbiochem, USA) for 36 h at4∘C. Subsequently, sections were transferred in a solutioncontaining 1%NGS, 0.1% Triton X-100, and 1 : 200 anti-rabbitbiotinylated secondary antibody (Vector Labs) in PBS. Thiswas followed by incubation inABCkit (Vector Labs) and finaldetection with DAB reaction kit (Vector Labs). Sections werefinally mounted on gelatinized slides, dehydrated, and sealedwith DPX mounting medium (VWR International, UK).

2.4. Analysis of c-Fos Positive Cells. Counting of c-Fos pos-itive cells in different brain areas was performed using aCCD camera (MBF Bioscience, Germany) mounted on aZeiss Axioskop (Zeiss, Germany)microscope and the Stereo-Investigator software (MBF Bioscience). Brain structureswere anatomically defined according to a mouse brain atlas(Paxinos and Franklin [61]), and the regions of interestselected for measurement of c-Fos-positive nuclei were(numbers indicate the distance in millimeters of the sectionsfrom bregma) infralimbic cortex (IL, +1.94mm); secondarymotor cortex (M2, +0.98mm); anterior cingulate cortex, area1 and area 2 (aCC, +0.98mm); dentate gyrus (DG,−1.94mm);CA1 field of dorsal hippocampus (dCA1, −1.94mm); CA3field of dorsal hippocampus (dCA3, −1.94mm); posterior

Neural Plasticity 3

parietal association cortex (pPtA, −1.94mm); primary audi-tory cortex (Au1, −3.16mm); primary visual cortex (V1,−4.16mm); medial entorhinal cortex (MEnt, −4.16mm).Thenumber of c-Fos-positive cells was counted at 20x magnifi-cation, following a “blind procedure”, in 10–40 fields (50 ×50𝜇m or 100 × 100 𝜇m) per section according to the sizeof brain structure and their density calculated (cells/mm2),using at least 5 sections for each structure.

2.5. Statistics. All results were expressed as mean ± s.e.m.,and all statistical analysis were performed using statisticalsoftware package SigmaStat. For MWM performance in thelearning phase, a two-way analysis of variance (ANOVA) forrepeated measures (RM) was performed, considering bothfactor housing condition (EE or SC) and factor learning day,with post hoc analysis Holm-Sidak method. Performance ineach probe test was compared with one-way ANOVA acrosscircular zones (target zone versus the average of other zones)for each housing condition. The c-Fos protein expression ineach area was analyzed with a two-way ANOVA for housingcondition factor and probe test day factor, with post hocanalysis Holm-Sidak method.

3. Results

3.1. Hippocampus Is Activated following SpatialMemory Recallat All Temporal Points Tested Both in EE and SC Mice. Totest whether EE can affect the system consolidation process,we trained C57BL/6 mice, housed in different conditions(standard condition, SC 𝑛 = 24 or environmental enrichmentEE 𝑛 = 24), in a spatial learning task, using the Morris watermaze, and we analyzed the following parameters referredto the average of 4 daily trials, during 7 consecutive days:latency to find the platform (s), total distance swam (cm),and mean swim speed (cm/s). For the distance swam and thelatency to reach the platform during acquisition, a significantlearning effect for both housing conditionswas found, but nota significant difference between the two groups (two-way RMANOVA, for latency 𝑃 = 0.016; for distance swan 𝑃 < 0.001).For the latency parameter only, we found a housing condition× day interaction: multiple comparisons showed that themain differences resided on days 4 and 5 (Figure 1(a)) (Two-way RMANOVA, post hoc analysis Holm-Sidak method, forday 4 𝑃 = 0.005; for day 5 𝑃 = 0.001). We also measuredthe mean swim velocity throughout the test, in order toexclude differences in navigation speed (data not shown): weobserved a significant decrease in the mean swim velocitythrough the test (two-way RM ANOVA, 𝑃 < 0.001), butneither a difference between housing condition (𝑃 = 0.276)nor a housing condition × day interaction (𝑃 = 0.163).

Spatial memory was evaluated in a probe test in whichthe hidden platformwas removed.Weperformed recall probetests at 1, 10, 20, 30, and 50 days, andwe quantified explorationin the target zone, a circular zone (radius: 23 cm) in quadrantwhere the platform was placed during training, and theaverage time spent in three other equivalent zones in eachquadrant, for SC andEEmice (Figure 1(b)) [55, 60].We founda significant difference between target zone versus the others

in probe tests at 1, 10, 20, and 30 days, for both groups (seeFigure 1(b) for details).

After the probe test, mice were sacrificed and the proteinc-Foswas immunolabeled as an indicator of neuronal activity.c-Fos expression was calculated as the density of numberof c-Fos-positive cells in mm2. First we investigated c-Fos expression in the hippocampus, the structure that isresponsible for the formation of long term spatial memory[49–53].

Levels of c-Fos expression in the hippocampus of controlmice (Home cage mice, HC-EE and HC-SC mice) did notdiffer between housing conditions (HC-SC 𝑛 = 8; HC-EE𝑛 = 7; one-way ANOVA, 𝑃 = 0.736); levels of c-Fos proteinfor EE and SC mice were significantly greater than those intheir home-cage controls at all retention intervals, (two-wayANOVA, post hoc analysis Holm-Sidak method, all 𝑃 values<0.05), suggesting that the hippocampus is involved both inthe formation and delayed recall of the spatial memory. Wefound a similar c-Fos expression in the EE and SC mice in allprobe tests (two-way ANOVA, post hoc analysis Holm-Sidakmethod for housing condition, 𝑃 = 0.731).

We then focused on the dorsal hippocampus (dHCP),known to be specifically involved in spatial memory [60];again, we observed the same c-Fos expression pattern in EEand SC group; activation increased with increasing retentioninterval up to 30 days (see Figure 2 for details).

3.2. EE Induces an Early Recruitment of theMPFC. Theresultsfor c-Fos expression in the mPFC, the final memory storagesite in the cortex, show that both the aCC and the IL havethe same time course of c-Fos protein expression pattern; c-fos expression at 1 day did not differ from that in home cageanimals, both for EE and SCmice, and therewas no differencebetween EE and SC or HC-EE and HC-SC mice (two-way ANOVA, post hoc analysis Holm-Sidak method, all 𝑃values>0.05); however, starting from the probe test at 10 days,c-Fos expression was greater in EE group than HC-EE andSC groups, with a further increase at 20 days (see Figure 3 fordetails); only for EE animals did the c-Fos protein expressiondiffer from that of HC control animals. For the probe testsat 30 and 50 days we found that the c-Fos expression in SCanimals differed from that of HC-SC animals; values of SCand EE groups did not differ (two-way ANOVA, post hocanalysis Holm-Sidak method; see Figure 3 for details).

3.3. EE Induces the Involvement of Distributed Cortical Net-work in Supporting Remote Spatial Memory. To examine thetime-dependent reorganization of neuronal activation in abrain-wide manner, we observed c-Fos protein expression inother cortical areas that are important for the constructionof spatial maps. Using several techniques, Wang et al. [62]determined that distinct areas of extrastriate visual cortex aregateways for ventral and dorsal streams in the mouse. Thedorsal stream includes the network hippocampus—medialentorhinal cortex [63]—posterior parietal cortex [64] forspatial navigation; in addition, the dorsal stream is connectedto auditory cortex and to frontal areas, such as cingulatecortex, infralimbic cortex, and motor areas [62]. First we

4 Neural Plasticity

Esca

pe la

tenc

y (s

)

Days

Latency to find the platform

EESC

1 2 3 4 5 6 70

10

20

30

40

50

60

∗∗

(a)

1 day 10 days

Target zone Other zones

Tim

e in

zone

(s)

20 days

0

5

10

15

20

25


Tim

e in

zone

(s)

0

5

10

15

20

25


Tim

e in

zone

(s)

0

5

10

15

20

25


Tim

e in

zone

(s)

0

5

10

15

20

25


Tim

e in

zone

(s)

0

5

10

15

20

2530 days 50 days

∗

∗∗∗

∗

∗

∗∗

SCEE

SCEE

(b)

Figure 1: (a) Performances of SC (𝑛 = 24) and EE (𝑛 = 24) mice in the MWM.There is a significant learning effect (two-way RM ANOVA,𝑃 = 0.016), and a housing condition × day interaction (two-way RMANOVA post hoc analysis Holm-Sidak method, for day 4, 𝑃 = 0.005; forday 5, 𝑃 = 0.001). ∗= statistical significance; error bars = s.e.m. (b) Evaluation of spatial memory for SC (1 day 𝑛 = 5; 10 days 𝑛 = 5; 20 days𝑛 = 5; 30 𝑛 = 5, 50 𝑛 = 4) and EE (1 day 𝑛 = 5; 10 days 𝑛 = 5; 20 days 𝑛 = 5; 30 𝑛 = 5; 50 𝑛 = 4) mice. Time spent in the target zone (𝑇), wherethe platform was placed, versus other equivalent zones (𝑂), for recall probe tests. One-way ANOVA, post hoc analysis Holm-Sidak method, 1day probe test: in SC group, 𝑇 versus𝑂, 𝑃 = 0.044; in EE group, 𝑇 versus𝑂, 𝑃 = 0.036; 10 days probe test: in SC group, 𝑇 versus𝑂, 𝑃 = 0.037;in EE group, 𝑇 versus 𝑂, 𝑃 = 0.043; 20 days probe test: in SC group, 𝑇 versus 𝑂, 𝑃 < 0.01; in EE group, 𝑇 versus 𝑂, 𝑃 = 0.048; 30 days probetest: in SC group, 𝑇 versus 𝑂, 𝑃 = 0.025; in EE group, 𝑇 versus 𝑂, 𝑃 = 0.048; 50 days probe test: in SC group, 𝑇 versus 𝑂, 𝑃 = 0.070; in EEgroup, 𝑇 versus 𝑂, 𝑃 = 0.226. ∗= statistical significance; error bars = s.e.m.

Neural Plasticity 5

0

100

200

300

400

500

600

DG

dHCP

0

100

200

300

400

500

600

700

dCA1

Home cage Retention delay

1d 10d 20d 30d 50dHome cage Retention delay

1d 10d 20d 30d 50dc-

Fos c

ount

s (ce

ll/m

m2)

c-Fo

s cou

nts (

cell/

mm

2)

∘

∘

∘

∘

(a)

HC-SC SC 1d Sc 20d SC 30d SC 50d

HC-EE EE 1d EE 10d EE 20d EE 30d EE 50d

SC 10d

(b)

Figure 2: (a) c-Fos expression in subregions of dHPC for EE (open columns) and SC (filled columns) mice subjected to recall probe testsat 1, 10, 20, 30, and 50 days. DG: two-way ANOVA post hoc analysis Holm-Sidak method, SC versus EE, 𝑃 = 0.968; HC-SC versus HC-EE,𝑃 = 0.503; SC versus HC-SC and EE versus HC-EE 𝑃 < 0.05 for all retention intervals. Statistical differences for factor day were foundbetween 1 and 20 days, 1 and 30 days, and 50 days versus 20 and 30 days, all 𝑃 < 0.05. dCA1: two-way ANOVA post hoc analysis Holm-Sidakmethod, SC versus EE, 𝑃 = 0.242; HC-SC versus HC-EE 𝑃 = 0.979. Statistical difference factor day were found between 1 and 10, 20 and 30days, all 𝑃 < 0.05. ◼Statistical significance between EE and HC-EE; ◻statistical significance between SC and HC-SC; ∘statistical significancefor factor day; error bars = s.e.m. (b) Representative panel of c-Fos protein expression in DG for EE and SC animals, for all recall probe tests;scale bar 100 𝜇m.

investigated c-Fos expression in MEnt and in pPta, and wefound that activation at 1 day in both areas was similar in EEmice, SCmice, anddid not differ from that in their home-cagecontrols (two-way ANOVA, post hoc analysis Holm-Sidakmethod, all𝑃 values > 0.05); however, in the other probe testsperformed, significant differences between EE and SC miceand between EE and HC-EE mice were found (see Figure 4for details). Then, we observed c-Fos expression in V1 andM2, for they are connected to the dorsal stream and thereis a direct monosynaptic connection between motor and

visual cortices [65]. We found that EE group was statisticallydifferent from SC and HC-EE groups in M2, only in probetest performed at 20, 30, and 50 days; instead, in V1, we didnot find any difference between EE and SC group, only anincrease in c-Fos expression for the late retention delays inboth groups (see Figure 5 for details). Finally, we investigatedc-Fos expression in Au1, a sensory cortex not supposed tobe involved in spatial learning, and we demonstrated thatactivation in this area was similar in all groups (see Figure 6for details).

6 Neural Plasticity

ILaCC

0

200

400

600

800

1000


SCEE

1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

0

200

400

600

800

1000


SCEE

1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

∗

∗

∗

∗

§

§§

§

##

#

(a)

HC-SC SC 1d Sc 20d SC 30d SC 50d

HC-EE EE 1d EE 10d EE 20d EE 30d EE 50d

SC 10d

(b)

Figure 3: (a) c-Fos protein expression in aCC and IL. aCC: two-way ANOVA post hoc analysis Holm-Sidak method, HC-SC versus HC-EE𝑃 = 0.384; statistical significance for EE versus HC-EE at 10, 20, 30, and 50 days, and for SC versus HC-SC at 30 and 50 days, all 𝑃 values <0.05. Statistical difference between EE and SC at 1 day, 𝑃 = 0.376; at 10 and 20 days, 𝑃 < 0.01; at 30 days, 𝑃 = 0.899; at 50 days, 𝑃 = 0.383.Statistical difference within EE group were found between 10 and 20 days, and 1 day versus 10, 20, 30, and 50 days, all 𝑃 < 0.05. Statisticaldifference within SC group were found between 1 and 30 days, and 10 day versus 30 and 50 days, all 𝑃 < 0.05. IL: two-way ANOVA post hocanalysis Holm-Sidak method, HC-SC versus HC-EE, 𝑃 = 0.451; statistical significance for EE versus HC-EE at 10, 20, 30, and 50 days, andfor SC versus HC-SC at 30 and 50 days, all 𝑃 values < 0.05. Statistical differences between EE and SC at 1 day, 𝑃 = 0.300; at 10 and 20 days,𝑃 < 0.01; at 30 days, 𝑃 = 0.084; at 50 days, 𝑃 = 0.055. Statistical differences within EE group were found between 10 and 20 days, and 1 dayversus 10, 20, 30, and 50 days, all 𝑃 < 0.05. Statistical difference within SC group were found between 1 and 30 days, and 10 day and 30 days, all𝑃 < 0.05. ◼Statistical significance between EE and HC-EE; ◻statistical significance between SC and HC-SC; ∗statistical significance betweenEE and SC; §

= statistical significance for factor day within EE group; #= statistical significance for factor day within SC group; error bars =

s.e.m. (b) Representative panel of c-Fos protein expression in aCC for EE and SC animals, for all recall probe tests; scale bar 100 𝜇m.

4. Discussion

In this study, we provide the first evidence that EE can affectthe time-dependent spatial memory system consolidation.C57BL/6 mice, housed in standard or in enriched condition,

were subjected to spatial learning and then tested up to 50days after learning to evaluate consolidation process. Usingthe expression of c-Fos protein as an indicator of neuronalactivity in a brain-wide manner we have found indicationsfor a difference both in the time course and in the network of

Neural Plasticity 7

MEnt pPta

0

200

400

600

800

1000

1200

SCEE


1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

0

200

400

600

800

1000

1200

SCEE


1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

∗

∗

∗

∗

∗∗

∗∗

§

§§§

§

§

Figure 4: c-Fos protein expression inMEnt and pPta.MEnt: two-way ANOVA post hoc analysis Holm-Sidak method, HC-SC versus HC-EE𝑃 = 0.914; statistical significance for EE versus HC-EE at 10, 20, 30 and 50 days. Statistical difference between EE and SC at 1 day, 𝑃 = 0.782;at 10, 20, 30 and 50 days, 𝑃 < 0.01. Statistical difference within EE group were found between 1 day versus 10, 20, 30 and 50 days, and 20 daysversus 10, 30 and 50 days, all 𝑃 values < 0.05. pPta: two-way ANOVA post hoc analysis Holm-Sidakmethod, HC-SC versus HC-EE 𝑃 = 0.497;statistical significance for EE versus HC-EE at 10, 20, 30 and 50 days, all 𝑃 values < 0.05. Statistical difference between EE and SC at 1 day,𝑃 = 0.300; at 10 and 20 days, 𝑃 < 0.01; at 30 days, 𝑃 = 0.084; at 50 days, 𝑃 = 0.055. Statistical difference within EE group were found between 1day versus 20, 30 and 50 days, and 20 day versus 10 and 30 days, all 𝑃 values < 0.05. Statistical difference within SC group were found between1 and 30 days, and 10 day and 30 days, all 𝑃 values < 0.05. ◼Statistical significance between EE and HC-EE; ◻statistical significance betweenSC and HC-SC; ∗statistical significance between EE and SC; §

= statistical significance for factor day within EE group; error bars = s.e.m.

cortical areas recruited for recent and remote recall betweenEE and non-EE animals.

In agreement with previous studies [47, 55] there wasa progressive increase in c-Fos protein expression in bothaCC and IL, as consolidation process proceeded. We showedthat EE induces an earlier recruitment of aCC and IL withrespect to SC mice; these areas were recruited followingrecall of spatial memory in the EE group as early as 10 daysafter learning, while they were recruited only 30 days afterlearning in SC mice. The final storage site in the cortex couldbe the aCC, while the IL could correlate with motivationalaspects of performance, encoding other significant aspectsof the environment, such as salient landmarks or preferredlocations [66]. The aCC was found to be activated afterremote memory recall in a number of tasks [47, 48, 54, 55],and, conversely, inactivation of the aCC disrupted recall ofremote five-arm discrimination [47], contextual fear [48],andMWM [55] memories.The aCC is highly interconnectedto other prefrontal regions and is reciprocally connectedto sensory, motor, and limbic cortices [67, 68]; therefore,this connectivity places the aCC in favorable position, rais-ing the possibility that this region coordinates retrieval ofremote memories stored in distributed cortical networks.The earlier recruitment of aCC and IL in EE animals could

imply an earlier independence of spatial memory recall fromhippocampal activation in EE animals. Indeed, in animalsprovided with running wheels, a component of EE, block ofhippocampal activation ceased to block recall of contextualfear memory at shorter time distance from learning withrespect to sedentary animals [69].

In addition,we demonstrated that EE induces the involve-ment of a distributed cortical network in supporting remotespatial memory which is not activated in non EE animals.Weobserved that MEnt and pPta were activated following mem-ory recall at 10 days in EE group. Both areas were included inthe dorsal network for spatial navigation [60]; the entorhinalcortex contains a spatial representation of environment andplays an interface role between the hippocampus and neo-cortex [70]; instead, the parietal cortex, specifically the mul-tisensory posterior region, translates coordinate informationfrom spatial maps in the entorhinal cortex and hippocampusinto egocentric representations [59, 71]. We also investigatedc-Fos protein expression in V1 andM2, and we found that EEgroup showed a greater activation in M2 than SC group, forprobe test performed at 20, 30, and 50 days. In V1, instead, wedid not find any difference between EE and SC group, onlyan increase in c-Fos expression for the late retention delaysin both groups; a recent study [58] showed that V1 could

8 Neural Plasticity

M2 V1

0

200

400

600

800

1000

1200

1400

0

200

400

600

800

1000

1200

1400


SCEE

1d 10d 20d 30d 50d Home cage Retention delay

SCEE

1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

c-Fo

s cou

nts (

cell/

mm

2)

∗

∗

∗

∘

∘

§

§

Figure 5: c-Fos protein expression in V1 and M2. M2: two-way ANOVA post hoc analysis Holm-Sidak method, HC-SC versus HC-EE,𝑃 = 0.416; statistical significance for EE versus HC-EE at 20, 30, and 50 days. Statistical difference between EE and SC at 1 day, 𝑃 = 0.288;at 10 days, 𝑃 = 0.154; at 20, 30, and 50 days, 𝑃 < 0.05. Statistical differences within EE group were found between 10 and 20 days, and 1 dayversus 20, 30, and 50 days, all 𝑃 values < 0.05. V1: two-way ANOVA post hoc analysis Holm-Sidak method, HC-SC versus HC-EE 𝑃 = 0.907;statistical significance for EE versus HC-EE at 20, 30, and 50 days, and for SC versus HC-SC at 20, 30, and 50 days, all 𝑃 values < 0.05.Statistical significance for factor day was found between 1 day versus 20, 30, and 50 days, and 10 days versus 20, 30, and 50 days, all 𝑃 values< 0.05. ◼Statistical significance between EE and HC-EE; ◻statistical significance between SC and HC-SC; ∗= statistical significance betweenEE and SC; ∘statistical significance for factor day §

= statistical significance within EE group; error bars = s.e.m.

belong to the network of fear contextual memory, althoughits activation did not change between recent and remotememories. In our study, the V1 pattern activation could beinduced by its engagement in attentional process on accountof the spatial complex task. The involvement of M2 and V1 isnot surprising since they are connected to the dorsal stream[62], and a direct monosynaptic connection between motorand visual cortices was identified [65].

For the hippocampus we found no difference betweenEE and SC animals. Our results are consistent with theidea that hippocampus is responsible for encoding spatialmemory [49–53]; its activation in remote spatial memoryrecall is not in agreement with studies that showed a pro-gressive reduction in hippocampus activationwith increasingretention interval [47, 48, 54], though it is in line with thehypothesis that remote memory never becomes totally inde-pendent from the hippocampus [72]. In a more recent study,indeed, Lopez et al. [73] demonstrated that hippocampusrecruitment in the recall of remote memories was influencedby the environmental conditions, such as cue saliency andcomplexity of the task in whichmemories are initially formedand subsequently recalled; thus the rich spatial details andthe complexity of the training in MWM could account forthe hippocampal activation found also for remote memory

0

200

400

600

800

1000

1200Au1


SCEE

1d 10d 20d 30d 50d

c-Fo

s cou

nts (

cell/

mm

2)

Figure 6: c-Fos protein expression in Au1. The data showed nodifference between groups (two-way ANOVA, housing conditionfactor, 𝑃 = 0.996, day of probe test factor, 𝑃 = 0.943, housingcondition × day of probe test interaction, 𝑃 = 0.938). In all probetests, EE and SC groups did not differ from their home cage controls.

Neural Plasticity 9

recall. Moreover, it has been found that that precise real-time inhibition of the dorsal CA1 region, using optogeneticmethod, was sufficient to impair remote recall [74].

We found that EE mice were faster, considering latencyparameter, in learning the position of the target platformin comparison to SC mice, but there was no significantdifference between groups in the probe tests. These resultsare consistent with other studies in the literature which used,in rodents living in EE or provided with running wheels,intensive learning protocol for the MWM such as that usedby us (4 daily trials for 7 consecutive days) [21]. We decidedto use a behavioural protocol that maximizes the results ofspatial learning task because our purpose was to examinepossible differences between EE and SC groups in the timecourse of system consolidation without confounding effectsdue to differences in the results of learning, that is in theprobe test, and also to be confident in the formation of aremote memory. The results of c-Fos data are an indicationthat the same result (a successful probe test) can be obtainedthrough a different balance of hippocampal and corticalactivation during system consolidation. Also in the Mavieland Bontempi paper [47] response accuracy in animals testedon either day 1 or 30 was similar while cortical activationstrongly differed. The faster recruitment of cortical areasfound in EE animals and the activation of a distributedcortical network, involving prefrontal and other associativecortices, activated in EE but not in SC animals, could suggestthat the quality of the recalled memory is different in the twogroups of animals; indeed, activation of prefrontal cortex hasbeen correlated with development of the ability to create amemory that is vivid and rich in contextual details in humans[75] and activation of associative sensory cortices supportsmemory storage and retrieval of sensory stimuli that haveacquired a behavioral salience with the experience [56].

How could EE act on the recruitment of cortical net-works during system consolidation? One possibility is viaits well-known action on hippocampal neurogenesis. It hasbeen proposed that new neurons generated in the DGbecome functionally integrated into existing neural circuits[76]; in fact the spatial training when new neurons aremore receptive to surrounding neuronal activity favoredtheir subsequent recruitment upon remote memory retrieval[77, 78]. Thus, these tagged adult-generated neurons, oncemature, are recruited into hippocampal networks underlyingremote spatial memory representation. Therefore tamperingwith the level of hippocampal neurogenesis could interferein the hippocampus-only dependent period of memory.Indeed, it has been demonstrated that new hippocampalneurons were recruited into neuronal networks supportingretrieval of remote spatial memory and that the enhancedneurogenesis by voluntary running-wheel exercise sped upthe disengaging from hippocampus [69]. Since EE wasfound to increase hippocampal neurogenesis and promotethe survival of newly generated neurons [14], it is plausiblethat EE may accelerate the recruitment of extrahippocampalareas. In its turn, EE action on hippocampal neurogenesisis likely mediated by neurotrophins, such as brain-derivedneurotrophic factor (BDNF), or by insulin growth factor-l(IGF-1) [9, 10], which affect hippocampal neurogenesis and

hippocampal and cortical plasticity [14, 79, 80]. IGF-1 playsan important role in cell growth and development, and italso upregulates neurogenesis in the adult hippocampus [81–83]. In the adult brain, IGF-1 has been shown to mediateboth the neuroprotective effects of physical exercise and theenhancement caused by exercise in hippocampal plasticityand in learning and memory [79]. Moreover IGF-1 mediatesthe increased expression of BDNF subsequent to EE andphysical exercise [84–86]. BDNF has been shown to regulateadult hippocampal neurogenesis, to mediate EE effects on it,to modulate plasticity during learning by activating signalingpathways that modify local synaptic targets and have long-term effects on transcription, and to mediate the expressionof hippocampal LTP, in both the early and late phases [80, 83,87, 88].

Another nonexclusive possibility is that molecular mech-anisms could “tag” the activated synapses in hippocampal andcortical networks at the time of memory encoding; this earlytagging could guide the reactivation of proper hippocampal-cortical connections associated with the specific memory. Arecent study showed indeed that cortical tagging seems to behighly specific for precise memory trace and the impairmentof early cortical tagging tampers with the postacquisitionhippocampal-cortical dialogue, preventing the formation ofremote memory [41]. Moreover, they demonstrated thatearly tagging triggers specific signaling cascades, leading tohistone-tail acetylation in the cortex and that the histonedeacetylase inhibitor improves remote memory retrieval[41]. Thus, early tagging acts on epigenetic modificationthat could mediate remote memory formation and retrieval,so EE could modulate the long-term memory formationand consolidation through chromatin remodeling, such ashistone-tail acetylation, known to be increased in EE.


The authors declare no conflict of interests.

Authors’ Contribution

Joyce Bonaccorsi and Simona Cintoli equally contributed tothis work.

References

[1] M. R. Rosenzweig, E. L. Bennett, M. Hebert, and H. Morimoto,“Social grouping cannot account for cerebral effects of enrichedenvironments,” Brain Research, vol. 153, no. 3, pp. 563–576, 1978.

[2] M. R. Rosenzweig, E. L. Bennett, andD. Krech, “Cerebral effectsof environmental complexity and training among adult rats,”Journal of Comparative and Physiological Psychology, vol. 57, no.3, pp. 438–439, 1964.

[3] C. Beaulieu and M. Colonnier, “Effect of the richness of theenvironment on the cat visual cortex,” Journal of ComparativeNeurology, vol. 266, no. 4, pp. 478–494, 1987.

[4] R. L. Holloway Jr., “Dendritic branching: some preliminaryresults of training and complexity in rat visual cortex,” BrainResearch, vol. 2, no. 4, pp. 393–396, 1966.


[5] Y. Kozorovitskiy, C. G. Gross, C. Kopil et al., “Experienceinduces structural and biochemical changes in the adult primatebrain,” Proceedings of the National Academy of Sciences of theUnited States of America, vol. 102, no. 48, pp. 17478–17482, 2005.

[6] A. H. Mohammed, S. W. Zhu, S. Darmopil et al., “Environmen-tal enrichment and the brain,” Progress in Brain Research, vol.138, pp. 109–133, 2002.

[7] D. Young, P. A. Lawlor, P. Leone, M. Dragunow, and M.J. During, “Environmental enrichment inhibits spontaneousapoptosis, prevents seizures and is neuroprotective,” NatureMedicine, vol. 5, no. 4, pp. 448–453, 1999.

[8] G.Kempermann,H.G.Kuhn, and F.H.Gage, “More hippocam-pal neurons in adult mice living in an enriched environment,”Nature, vol. 386, no. 6624, pp. 493–495, 1997.

[9] B. R. Ickes, T. M. Pham, L. A. Sanders, D. S. Albeck, A. H.Mohammed, and A.-C. Granholm, “Long-term environmentalenrichment leads to regional increases in neurotrophin levelsin rat brain,” Experimental Neurology, vol. 164, no. 1, pp. 45–52,2000.

[10] T. M. Pham, S. Soderstrom, B.Winblad, and A. H.Mohammed,“Effects of environmental enrichment on cognitive functionand hippocampal NGF in the non-handled rats,” BehaviouralBrain Research, vol. 103, no. 1, pp. 63–70, 1999.

[11] M. R. Rosenzweig, E. L. Bennett, and M. C. Diamond, “Effectsof differential environments on brain anatomy and brainchemistry,” Proceedings of the annual meeting of the AmericanPsychopathological Association, vol. 56, pp. 45–56, 1967.

[12] S. Rasmuson, T. Olsson, B. G. Henriksson et al., “Environmen-tal enrichment selectively increases 5-HT1A receptor mRNAexpression and binding in the rat hippocampus,” MolecularBrain Research, vol. 53, no. 1-2, pp. 285–290, 1998.

[13] F. Naka, T. Shiga,M. Yaguchi, andN. Okado, “An enriched envi-ronment increases noradrenaline concentration in the mousebrain,” Brain Research, vol. 924, no. 1, pp. 124–126, 2002.

[14] H. Van Praag, G. Kempermann, and F. H. Gage, “Neuralconsequences of environmental enrichment,” Nature ReviewsNeuroscience, vol. 1, no. 3, pp. 191–198, 2000.

[15] C. Rampon, C. H. Jiang, H. Dong et al., “Effects of environmen-tal enrichment on gene expression in the brain,” Proceedings ofthe National Academy of Sciences of the United States of America,vol. 97, no. 23, pp. 12880–12884, 2000.

[16] A. Sale, E. Putignano, L. Cancedda et al., “Enriched environ-ment and acceleration of visual system development,” Neu-ropharmacology, vol. 47, no. 5, pp. 649–660, 2004.

[17] S. Landi, M. C. Cenni, L. Maffei, and N. Berardi, “Environmen-tal enrichment effects on development of retinal ganglion celldendritic stratification require retinal BDNF,” PLoS One, vol. 2,no. 4, Article ID e346, 2007.

[18] A. Sale, M. C. Cenni, F. Clussi, E. Putignano, S. Chierzi, andL. Maffei, “Maternal enrichment during pregnancy acceleratesretinal development of the fetus,” PLoSOne, vol. 2, no. 11, ArticleID e1160, 2007.

[19] A. Sale, J. F. M. Vetencourt, P. Medini et al., “Environmen-tal enrichment in adulthood promotes amblyopia recoverythrough a reduction of intracortical inhibition,” Nature Neuro-science, vol. 10, no. 6, pp. 679–681, 2007.

[20] L. Baroncelli, J. Bonaccorsi, M. Milanese et al., “Enrichedexperience and recovery from amblyopia in adult rats: impactof motor, social and sensory components,”Neuropharmacology,vol. 62, no. 7, pp. 2387–2396, 2012.

[21] H. Van Praag, G. Kempermann, and F. H. Gage, “Runningincreases cell proliferation and neurogenesis in the adult mousedentate gyrus,” Nature Neuroscience, vol. 2, no. 3, pp. 266–270,1999.

[22] J. C. Bennett, P. A. McRae, L. J. Levy, and K. M. Frick, “Long-term continuous, but not daily, environmental enrichmentreduces spatial memory decline in aged male mice,” Neurobi-ology of Learning and Memory, vol. 85, no. 2, pp. 139–152, 2006.

[23] S. N. Duffy, K. J. Craddock, T. Abel, and P. V. Nguyen,“Environmental enrichment modifies the PKA-dependenceof hippocampal LTP and improves hippocampus-dependentmemory,” Learning and Memory, vol. 8, no. 1, pp. 26–34, 2001.

[24] B. A. Doty, “The effects of cage environment upon avoidanceresponding of aged rats,” Journals of Gerontology, vol. 27, no. 3,pp. 358–360, 1972.

[25] R. A. Cummins, R. N.Walsh, and O. E. B. Olsen, “Environmen-tally induced changes in the brains of elderly rats,” Nature, vol.243, no. 5409, pp. 516–518, 1973.

[26] J. M. Warren, C. Zerweck, and A. Anthony, “Effects of environ-mental enrichment on old mice,” Developmental Psychobiology,vol. 15, no. 1, pp. 13–18, 1982.

[27] G. Winocur, “Environmental influences on cognitive decline inaged rats,”Neurobiology of Aging, vol. 19, no. 6, pp. 589–597, 1998.

[28] M. Soffie, K. Hahn, E. Terao, and F. Eclancher, “Behavioural andglial changes in old rats following environmental enrichment,”Behavioural Brain Research, vol. 101, no. 1, pp. 37–49, 1999.

[29] G. Kempermann, D. Gast, and F. H. Gage, “Neuroplasticityin old age: sustained fivefold induction of hippocampal neu-rogenesis by long-term environmental enrichment,” Annals ofNeurology, vol. 52, no. 2, pp. 135–143, 2002.

[30] S. Kobayashi, Y. Ohashi, and S. Ando, “Effects of enrichedenvironments with different durations and starting times onlearning capacity during aging in rats assessed by a refined pro-cedure of theHebb-Williamsmaze task,” Journal of NeuroscienceResearch, vol. 70, no. 3, pp. 340–346, 2002.

[31] K. M. Frick and S. M. Fernandez, “Enrichment enhances spatialmemory and increases synaptophysin levels in aged femalemice,” Neurobiology of Aging, vol. 24, no. 4, pp. 615–626, 2003.

[32] P. Leal-Galicia, M. Castaneda-Bueno, R. Quiroz-Baez, andC. Arias, “Long-term exposure to environmental enrichmentsince youth prevents recognitionmemory decline and increasessynaptic plasticity markers in aging,” Neurobiology of Learningand Memory, vol. 90, no. 3, pp. 511–518, 2008.

[33] N. Berardi, C. Braschi, S. Capsoni, A. Cattaneo, and L. Maffei,“Environmental enrichment delays the onset ofmemory deficitsand reduces neuropathological hallmarks in a mouse modelof Alzheimer-like neurodegeneration,” Journal of Alzheimer’sDisease, vol. 11, no. 3, pp. 359–370, 2007.

[34] L. Baroncelli, C. Braschi,M. Spolidoro, T. Begenisic, A. Sale, andL. Maffei, “Nurturing brain plasticity: impact of environmentalenrichment,” Cell Death and Differentiation, vol. 17, no. 7, pp.1092–1103, 2010.

[35] J. Nithianantharajah and A. J. Hannan, “Enriched environ-ments, experience-dependent plasticity and disorders of thenervous system,” Nature Reviews Neuroscience, vol. 7, no. 9, pp.697–709, 2006.

[36] L. R. Squire and P. Alvarez, “Retrograde amnesia and memoryconsolidation: a neurobiological perspective,” Current Opinionin Neurobiology, vol. 5, no. 2, pp. 169–177, 1995.

[37] J. L. McClelland, B. L. McNaughton, and R. C. O’Reilly, “Whythere are complementary learning systems in the hippocampus


and neocortex: insights from the successes and failures ofconnectionist models of learning and memory,” PsychologicalReview, vol. 102, no. 3, pp. 419–457, 1995.

[38] P. J. Bayley and L. R. Squire, “The neuroscience of remotememory,” Current Opinion in Neurobiology, vol. 17, no. 2, pp.185–196, 2007.

[39] S.-H. Wang and R. G. M. Morris, “Hippocampal-neocorticalinteractions in memory formation, consolidation, and recon-solidation,” Annual Review of Psychology, vol. 61, pp. 49–79,2010.

[40] B. Bontempi and P. W. Frankland, “The organization of recentand remote memories,”Nature Reviews Neuroscience, vol. 6, no.2, pp. 119–130, 2005.

[41] E. Lesburgueres, O. L. Gobbo, S. Alaux-Cantin, A. Hambucken,P. Trifilieff, and B. Bontempi, “Early tagging of cortical networksis required for the formation of enduring associative memory,”Science, vol. 331, no. 6019, pp. 924–928, 2011.

[42] E. Shimizu, Y.-P. Tang, C. Rampon, and J. Z. Tsien, “NMDAreceptor-dependent synaptic reinforcement as a crucial processformemory consolidation,” Science, vol. 290, no. 5494, pp. 1170–1174, 2000.

[43] P. W. Frankland, C. O’Brien, M. Ohno, A. Kirkwood, and A. J.Silva, “𝛼-CaMKII-dependent plasticity in the cortex is requiredfor permanent memory,”Nature, vol. 411, no. 6835, pp. 309–313,2001.

[44] M. L. Hayashi, S.-Y. Choi, B. S. Rao et al., “Altered corticalsynaptic morphology and impaired memory consolidation inforebrain- specific dominant-negative PAK transgenic mice,”Neuron, vol. 42, no. 5, pp. 773–787, 2004.

[45] L. Restivo, G. Vetere, B. Bontempi, and M. Ammassari-Teule,“The formation of recent and remotememory is associated withtime-dependent formation of dendritic spines in the hippocam-pus and anterior cingulate cortex,” Journal of Neuroscience, vol.29, no. 25, pp. 8206–8214, 2009.

[46] G. Vetere, L. Restivo, C. J. Cole et al., “Spine growth in theanterior cingulate cortex is necessary for the consolidation ofcontextual fear memory,” Proceedings of the National Academyof Sciences of the United States of America, vol. 108, no. 20, pp.8456–8460, 2011.

[47] T. Maviel, T. P. Durkin, F. Menzaghi, and B. Bontempi, “Sites ofneocortical reorganization critical for remote spatial memory,”Science, vol. 305, no. 5680, pp. 96–99, 2004.

[48] P. W. Frankland, B. Bontempi, L. E. Talton, L. Kaczmarek, andA. J. Silva, “The involvement of the anterior cingulate cortex inremote contextual fear memory,” Science, vol. 304, no. 5672, pp.881–883, 2004.

[49] J. M. O’Keefe and L. Nadel, The Hippocampus as a CognitiveMap, Oxford Unversity Press, Oxford, UK, 1978.

[50] N. Burgess, E. A. Maguire, and J. O’Keefe, “The human hip-pocampus and spatial and episodic memory,” Neuron, vol. 35,no. 4, pp. 625–641, 2002.

[51] R. G. M. Morris, E. I. Moser, G. Riedel et al., “Elementsof a neurobiological theory of the hippocampus: the role ofactivity-dependent synaptic plasticity inmemory,”PhilosophicalTransactions of the Royal Society B, vol. 358, no. 1432, pp. 773–786, 2003.

[52] K. Nakazawa, T. J. McHugh, M. A. Wilson, and S. Tonegawa,“NMDA receptors, place cells and hippocampal spatial mem-ory,” Nature Reviews Neuroscience, vol. 5, no. 5, pp. 361–372,2004.

[53] S. Leutgeb, J. K. Leutgeb, M.-B. Moser, and E. I. Moser, “Placecells, spatial maps and the population code for memory,”Current Opinion in Neurobiology, vol. 15, no. 6, pp. 738–746,2005.

[54] B. Bontempi, C. Laurent-Demir, C. Destrade, and R. Jaffard,“Time-dependent reorganization of brain circuitry underlyinglong-termmemory storage,”Nature, vol. 400, no. 6745, pp. 671–675, 1999.

[55] C. M. Teixeira, S. R. Pomedli, H. R. Maei, N. Kee, and P. W.Frankland, “Involvement of the anterior cingulate cortex in theexpression of remote spatial memory,” Journal of Neuroscience,vol. 26, no. 29, pp. 7555–7564, 2006.

[56] T. Sacco and B. Sacchetti, “Role of secondary sensory corticesin emotional memory storage and retrieval in rats,” Science, vol.329, no. 5992, pp. 649–656, 2010.

[57] R. Romero-Granados, A. Fontan-Lozano, J.M.Delgado-Garcıa,and A.M.Carrion, “From learning to forgetting: behavioral, cir-cuitry, and molecular properties define the different functionalstates of the recognition memory trace,” Hippocampus, vol. 20,no. 5, pp. 584–595, 2010.

[58] A. L. Wheeler, C. M. Teixeira, A. H. Wang et al., “Identificationof a functional connectome for long-term fearmemory inmice,”PLoS Computational Biology, vol. 9, no. 1, Article ID e1002853,2013.

[59] R. G. M. Morris, P. Garrud, J. N. P. Rawlins, and J. O’Keefe,“Place navigation impaired in rats with hippocampal lesions,”Nature, vol. 297, no. 5868, pp. 681–683, 1982.

[60] E. Moser, M.-B. Moser, and P. Andersen, “Spatial learningimpairment parallels the magnitude of dorsal hippocampallesions, but is hardly present following ventral lesions,” Journalof Neuroscience, vol. 13, no. 9, pp. 3916–3925, 1993.

[61] G. Paxinos and K. Franklin, The Mouse Brain in StereotaxicCoordinates, Academic Press, San Diego, Calif, USA, 1997.

[62] Q. Wang, E. Gao, and A. Burkhalter, “Gateways of ventral anddorsal streams in mouse visual cortex,” Journal of Neuroscience,vol. 31, no. 5, pp. 1905–1918, 2011.

[63] T. Hafting, M. Fyhn, S. Molden, M.-B. Moser, and E. I. Moser,“Microstructure of a spatial map in the entorhinal cortex,”Nature, vol. 436, no. 7052, pp. 801–806, 2005.

[64] J. R. Whitlock, R. J. Sutherland, M. P. Witter, M.-B. Moser, andE. I. Moser, “Navigating from hippocampus to parietal cortex,”Proceedings of the National Academy of Sciences of the UnitedStates of America, vol. 105, no. 39, pp. 14755–14762, 2008.

[65] A. Di Garbo, M. Mainardi, S. Chillemi, L. Maffei, and M.Caleo, “Environmental enrichment modulates cortico-corticalinteractions in the mouse,” PLoS One, vol. 6, no. 9, Article IDe25285, 2011.

[66] V. Hok, E. Save, P. P. Lenck-Santini, and B. Poucet, “Coding forspatial goals in the prelimbic/infralimbic area of the rat frontalcortex,” Proceedings of the National Academy of Sciences of theUnited States of America, vol. 102, no. 12, pp. 4602–4607, 2005.

[67] H. B.M. Uylings, H. J. Groenewegen, and B. Kolb, “Do rats havea prefrontal cortex?” Behavioural Brain Research, vol. 146, no. 1-2, pp. 3–17, 2003.

[68] B. F. Jones, H. J. Groenewegen, and M. P. Witter, “Intrinsic con-nections of the cingulate cortex in the rat suggest the existenceof multiple functionally segregated networks,” Neuroscience,vol. 133, no. 1, pp. 193–207, 2005.

[69] T. Kitamura, Y. Saitoh, N. Takashima et al., “Adult neurogenesismodulates the hippocampus-dependent period of associativefear memory,” Cell, vol. 139, no. 4, pp. 814–827, 2009.


[70] E. I.Moser andM.-B.Moser, “Ametric for space,”Hippocampus,vol. 18, no. 12, pp. 1142–1156, 2008.

[71] M. Moscovitch, R. S. Rosenbaum, A. Gilboa et al., “Functionalneuroanatomy of remote episodic, semantic and spatial mem-ory: a unified account based on multiple trace theory,” Journalof Anatomy, vol. 207, no. 1, pp. 35–66, 2005.

[72] L. Nadel and M. Moscovitch, “Memory consolidation, retro-grade amnesia and the hippocampal complex,”Current Opinionin Neurobiology, vol. 7, no. 2, pp. 217–227, 1997.

[73] J. Lopez, K. Herbeaux, B. Cosquer et al., “Context-dependentmodulation of hippocampal and cortical recruitment duringremote spatial memory retrieval,” Hippocampus, vol. 22, no. 4,pp. 827–841, 2012.

[74] I. Goshen, M. Brodsky, R. Prakash et al., “Dynamics of retrievalstrategies for remote memories,” Cell, vol. 147, no. 3, pp. 678–689, 2011.

[75] N. Ofen, Y.-C. Kao, P. Sokol-Hessner, H. Kim, S. Whitfield-Gabrieli, and J. D. E. Gabrieli, “Development of the declarativememory system in the human brain,” Nature Neuroscience, vol.10, no. 9, pp. 1198–1205, 2007.

[76] W. Deng, J. B. Aimone, and F. H. Gage, “New neurons and newmemories: how does adult hippocampal neurogenesis affectlearning and memory?” Nature Reviews Neuroscience, vol. 11,no. 5, pp. 339–350, 2010.

[77] S. Trouche, B. Bontempi, P. Roullet, and C. Rampon, “Recruit-ment of adult-generated neurons into functional hippocampalnetworks contributes to updating and strengthening of spatialmemory,” Proceedings of the National Academy of Sciences of theUnited States of America, vol. 106, no. 14, pp. 5919–5924, 2009.

[78] M. Arruda-Carvalho, M. Sakaguchi, K. G. Akers, S. A. Josselyn,and P. W. Frankland, “Posttraining ablation of adult-generatedneurons degrades previously acquired memories,” The Journalof Neuroscience, vol. 31, no. 42, pp. 15113–15127, 2011.

[79] N. D. Aberg, K. G. Brywe, and J. Isgaard, “Aspects of growthhormone and insulin-like growth factor-I related to neuro-protection, regeneration, and functional plasticity in the adultbrain,”TheScientificWorldJournal, vol. 6, pp. 53–80, 2006.

[80] E.G.Waterhouse andB.Xu, “New insights into the role of brain-derived neurotrophic factor in synaptic plasticity,” Molecularand Cellular Neuroscience, vol. 42, no. 2, pp. 81–89, 2009.

[81] A. J. D’Ercole, P. Ye, A. S. Calikoglu, and G. Gutierrez-Ospina,“The role of the insulin-like growth factors in the centralnervous system,”Molecular Neurobiology, vol. 13, no. 3, pp. 227–255, 1996.

[82] J. A. D’ercole, P. Ye, and J. R. O’Kusky, “Mutant mouse modelsof insulin-like growth factor actions in the central nervoussystem,” Neuropeptides, vol. 36, no. 2-3, pp. 209–220, 2002.

[83] E. Lee andH. Son, “Adult hippocampal neurogenesis and relatedneurotrophic factors,” BMB Reports, vol. 42, no. 5, pp. 239–244,2009.

[84] E. Carro, A. Nunez, S. Busiguina, and I. Torres-Aleman, “Circu-lating insulin-like growth factor Imediates effects of exercise onthe brain,” Journal of Neuroscience, vol. 20, no. 8, pp. 2926–2933,2000.

[85] Q. Ding, S. Vaynman, M. Akhavan, Z. Ying, and F. Gomez-Pinilla, “Insulin-like growth factor I interfaces with brain-derived neurotrophic factor-mediated synaptic plasticity tomodulate aspects of exercise-induced cognitive function,” Neu-roscience, vol. 140, no. 3, pp. 823–833, 2006.

[86] S. Landi, F. Ciucci, L. Maffei, N. Berardi, and M. C. Cenni,“Setting the pace for retinal development: environmental

enrichment acts through insulin-like growth factor 1 and brain-derived neurotrophic factor,” Journal of Neuroscience, vol. 29,no. 35, pp. 10809–10819, 2009.

[87] C. Rossi, A. Angelucci, L. Costantin et al., “Brain-derivedneurotrophic factor (BDNF) is required for the enhancementof hippocampal neurogenesis following environmental enrich-ment,” The European Journal of Neuroscience, vol. 24, no. 7, pp.1850–1856, 2006.

[88] K. K. Cowansage, J. E. Ledoux, and M.-H. Monfils, “Brain-derived neurotrophic factor: a dynamic gatekeeper of neuralplasticity,”CurrentMolecular Pharmacology, vol. 3, no. 1, pp. 12–29, 2010.


Research ArticleGene Expression Patterns Underlying the Reinstatement ofPlasticity in the Adult Visual System

Ettore Tiraboschi,1,2 Ramon Guirado,1 Dario Greco,3 Petri Auvinen,4

Jose Fernando Maya-Vetencourt,5,6 Lamberto Maffei,7 and Eero Castrén1

1 Neuroscience Centre, University of Helsinki, 00790 Helsinki, Finland2 SARS Institute, University of Bergen, 5020 Bergen, Norway3 Finnish Institute of Occupational Health, 00250 Helsinki, Finland4 Institute of Biotechnology, University of Helsinki, 00790 Helsinki, Finland5 Centre for Nanotechnology Innovation, Italian Institute of Technology, 56127 Pisa, Italy6Centre for Neuroscience and Cognitive Systems, Italian Institute of Technology, 38068 Rovereto, Italy7 Neuroscience Institute, CNR, 56100 Pisa, Italy

Correspondence should be addressed to Eero Castren; [email protected]

Received 1 May 2013; Accepted 10 June 2013


Copyright © 2013 Ettore Tiraboschi et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The nervous system is highly sensitive to experience during early postnatal life, but this phase of heightened plasticity decreaseswith age. Recent studies have demonstrated that developmental-like plasticity can be reactivated in the visual cortex of adultanimals through environmental or pharmacological manipulations. These findings provide a unique opportunity to study thecellular and molecular mechanisms of adult plasticity. Here we used the monocular deprivation paradigm to investigate large-scale gene expression patterns underlying the reinstatement of plasticity produced by fluoxetine in the adult rat visual cortex.We found changes, confirmed with RT-PCRs, in gene expression in different biological themes, such as chromatin structureremodelling, transcription factors, molecules involved in synaptic plasticity, extracellular matrix, and excitatory and inhibitoryneurotransmission. Our findings reveal a key role for several molecules such as the metalloproteases Mmp2 and Mmp9 or theglycoprotein Reelin and open up new insights into the mechanisms underlying the reopening of the critical periods in the adultbrain.

1. Introduction

Use-dependent plasticity shapes neuronal networks withinsensory systems during early life to optimally representsensory stimuli [1]. Experience-dependent organization ofeye-specific inputs is amajormechanismwhereby refinementof synaptic connectivity is achieved in the developing visualsystem [2–4]. Monocular deprivation during developmentleads to a loss of cortical connectivity of the deprived eyeresulting in a shift of the ocular dominance in the visualcortex, which will become permanent if the MD persiststo adulthood [5, 6]. Although neuronal plasticity of thedeveloping brain gradually decreases with age [7], recentfindings suggest that it can be reactivated in the adult visual

cortex [8] and other regions, such as the amygdala [9]. Avariety of experimental manipulations, including enzymatictreatments [10, 11], environmental enrichment [12–15], foodrestriction [16], genetic manipulations [17, 18], and othermanipulations, promote this kind of plasticity [19–21].

Although themechanisms behind the adult induced plas-ticity are still unclear, we are beginning to understand the keyfactors involved. For example, the developmental maturationof neuronal inhibition, mainly through the parvalbumincontaining interneurons [22, 23], is known to be involvedin both the opening and the closure of the critical period[20]. Several extracellular matrix components, such as PSA-NCAM or the perineuronal nets, have been shown to playa role in the maturation of the inhibitory circuitries and

2 Neural Plasticity

Saline (Sal)BV

MD

BV

MD

D23

D21

Fluoxetine (Flx)

D0

(a)

BV-Sal versus BV-Flx BV-Sal versus MD-Sal

MD-Sal versus MD-Flx

1507

5837

1

15921138

(b)

Figure 1: Experimental design and Venn diagram. (a) Schematic graph showing the different animal groups and experimental conditionsconducted. (b) Venn diagram showing the number of genes differentially expressed when comparing the effects of monocular deprivation(BV-Sal versus MD-Sal), the effects of fluoxetine in animals with binocular vision (BV-Sal versus BV-Flx), and those of fluoxetine in animalswith monocular deprivation (MD-Sal versus MD-Flx).

experimental manipulations removing these extracellularmatrix components, can trigger an early closure [24] or areopening of the critical period, respectively [11]. Similarly, avariety of other molecules, such as transcription factors [25]or proteins involved in chromatin structure remodeling [26],are also key factors in regulating the closure and reopening ofthe critical period.

The main pharmacological approaches to experimentallyregulate critical period plasticity in the adulthood are thoseaffecting the action of ascending projection systems, suchas the serotoninergic or cholinergic systems [21, 27–29]. Inthis line, we have investigated the plastic effects of antide-pressants, such as fluoxetine, that modulate serotoninergictransmission, and we have shown that these drugs, in a long-term treatment, are able to trigger critical period plasticity inthe adult brain [8], through an early epigenetic modificationthat regulates gene expression [29].

Here, we have used fluoxetine in combination with anexperience-dependent paradigm of visual deprivation, toanalyze the large-scale gene expression patterns, to under-stand the temporal-dependent changes that allow the reopen-ing of the critical periods in the adult brain.

2. Experimental Procedures

2.1. Animal Treatment, Fluoxetine Administration, and Surgi-cal Procedures. A total amount of 32 Long-Evans hooded ratswere used in this study, equally distributed in 4 experimentalgroups (𝑛 = 8 animals per group), as explained later(Figure 1(a)). Animals were group-housed under standardconditions with food and water ad libitum in plexiglas cages(40 × 30 × 20 cm) and kept in a 12 : 12 light/dark cycle.Adult rats at the postnatal day 70 (P70) were systemicallytreatedwith fluoxetine (fluoxetine-hydrochloride, 0.2mg/mL

drinking water) for 23 days. Control animals were housedunder the same standard conditions drinking tap water.

Three weeks after the beginning of the fluoxetine treat-ment, rats were anaesthetized with avertin (1mL/100 g) andmounted on a stereotaxic apparatus to perform the eye-lid suture for monocular deprivation (MD). Eyelids wereinspected daily until complete cicatrisation; subjects witheven minimal spontaneous reopening were excluded. Greatcare was taken during the first days after MD to preventinflammation or infection of the deprived eye through topicalapplication of antibiotic and cortisone.

2.2. DNA Microarrays and Data Analysis. Two days afterMD, the binocular region of the primary visual cortex wasdissected. For all microarray experiments, total RNA waspurified using RNA extraction kit (Macherey Nagel), andAmino Allyl cRNA labeling Kit (Ambion) was used tolabel cRNA according to manufacturer’s standard protocols.Agilent Whole Rat Genome Microarray Kits (4 × 44K) werehybridized following provided protocols.

Images fromhybridizedmicroarrays were segmented andthe median intensity of each spot was estimated by thesoftware GenePix v.5.0 (axon). Data was imported into thesoftware (http://cran.r-project.org/) and preprocessed by thebioconductor package limma. The statistical analysis usedwas a linear model followed by 𝑡-test for finding the differ-entially expressed genes. In order to increase the reliability ofthe statistical analysis, we only considered significant thosegenes with a 𝑃 value less than 0.01. In addition, we alsoincrease the reliability of the analysis through validation ofthe results using multiple RT-PCRs. Lists of significant geneswere screened by the DAVID 6.7 annotation tools in orderto find overrepresented biological themes. Default DAVIDparameters were used.

Neural Plasticity 3

2.3. Real-Time PCR. RNA purification was performed accor-ding to the standard Trizol procedure (Invitrogen). PurifiedRNA was treated with DNAse (Fermentas) and cDNA wassynthesised from 1 𝜇g of RNA (Invitrogen). Real-time PCRwas carried out to determine relative enrichment in thesamples using the Sybr Green method according to themanufacturer instructions (SYBRGreen Imaster, Light cycler480, Roche Diagnostics). The comparative Ct method [30]was used to determine the normalized changes of the targetgene relative to a calibrator reference; in particular, valueswere normalized toGAPDH levels. As calibrator referencewereferred to Ct from water-treated animal samples.

3. Results

3.1. DNA Microarrays. Previous studies have shown that 7days period of monocular deprivation in fluoxetine-treatedadult rats is sufficient to bring about a change in the oculardominance. To reveal early transcriptional changes thatprecede and underlie the functional change, we analysedgene expression, using DNA microarrays, at two days afterMD. Microarray analysis revealed only relatively few genesthat were significantly regulated by either FLX (𝑛 = 197,see Supplementary Table S1 in the Supplementary Materialavailable online at http://dx.doi.org/10.1155/2013/605079) orMD alone (𝑛 = 239, Table S2), treatments that themselves donot produce any changes in the ocular dominance plasticity.However, the combination of FLX and MD, the treatmentthat promotes changes in ocular dominance, altered theexpression of a significantly larger number of genes (𝑛 =1603, Table S3, Figure 1(b)). Notably 1237 out of 1603 (77%)of the genes in the group receiving both MD and FLXwere downregulated, whereas in the groups receiving eitherMD or FLX, 111 out of 239 and 88 out of 197 genes weredownregulated, respectively, comprising of roughly 50%of allthe regulated genes.Hence, the combination of the treatmentsapparently has two major effects on gene expression; first, itincreases the number of regulated genes when compared tothe single treatments, and second, it has a striking effect ondownregulating most of the genes, indicating that silencingof genes normally expressed during basal conditions is likelyinvolved in the triggering of plasticity of the adult brain.

The representation of biological themes was screenedusing Fisher’s Exact test on the lists of differentially expressedgenes in each comparison. Chronic fluoxetine treatmentinduced a regulation of genes related to chromatin remod-elling, nervous system development, and plasticity, as wellas regulation of gene expression and transcription in thebinocular visual cortex (Table S4).

MD altered the expression of a significant number ofgenes related to cell differentiation, cell plasticity, and neuro-genesis. Several genes of the ion homeostasis and regulationof transcription were also found overexpressed (Table S5).

The combination of MD and fluoxetine treatment down-regulated the majority of the differentially expressed genes,altering the expression of genes represented in a variety offunctional processes, including genes related to neuronaldevelopment, plasticity, and apoptosis. In addition, genes

involved in the synaptic transmission, ion and intracellularcalcium homeostasis, and vesicular secretion were found dif-ferentially expressed. Blood circulation and lipid metabolismwere among the most significantly overrepresented families(Table S6).

3.2. RT-PCR. To provide validation of the microarray data,we next examined single patterns of gene expression bymeans of real-time PCR, in the same experimental groupsand using the same experimental paradigm (Figure 1(a)). Inparticular, we focused our attention on genes whose expres-sion may alter molecular and cellular processes involved inthe closure of the critical period for visual cortex plasticity,such as the balance of inhibitory and excitatory transmission[22, 31, 32], transcription factors regulating gene expression[25], extracellular matrix remodeling [11], myelination [33],and chromatin structure remodeling [26, 29], as well asgenes involved in processes of synaptic plasticity, neuronaldifferentiation, and outgrowth (see Table 1).

3.2.1. Inhibitory Neurotransmission. We observed that fluox-etine produced a significant increase in the expression ofgenes involved in inhibitory neurotransmission when com-paring both animals with binocular vision and animals withmonocular deprivation with their respective controls (BV-Sal versus BV-Flx and MD-Sal versus MD-Flx; Figure 2(a)).Specifically, in animals with binocular vision, we foundan increased expression of the vesicular GABA transporter(VGAT; 60% increased expression; 𝑃 = 0.001), while inrats with monocular deprivation together with fluoxetinetreatment, we observed an increase in the expression ofGABRA4 (30% increased expression; 𝑃 = 0.02).

3.2.2. Excitatory Neurotransmission. We did not observemany changes in the composition of NMDA receptor sub-units in either of the experimental groups (see Figure 2(b)and Table 1). The only significant change we found was adecrease in the expression of the NR2A subunit (NMDA-2A;20% decreased expression; 𝑃 = 0.04) in the animals withmonocular deprivation treated with fluoxetine.

3.2.3. Transcription Factors. We detected increases in thegene expression of transcription factors in the animals withbinocular vision treated with fluoxetine (Figure 2(c)). Inparticular, NFKB1 and DLX1 increased their expression (50%and 30% increased expression, resp.;𝑃 = 0.04 and 0.03, resp.).However, in those animals with monocular deprivation,fluoxetine treatment produced a decrease in the expressionof transcription factors, such as EGR-2 (𝑃 = 0.04).

3.2.4. Synaptic Plasticity. The expression of Reelin, transcriptthat encodes a glycoprotein that mediates synaptic plasticityat hippocampal level [34], was significantly increased in MDanimals treated with fluoxetine (Figure 2(d); 40% increasedexpression; 𝑃 = 0.001). The expression of additionaltranscripts that encode proteins involved in neuronal dif-ferentiation and outgrowth processes as well as synapticplasticity increased in both groups. In animals with binocular

4 Neural Plasticity

Table 1: RT-PCR analysis.

Gene symbol Entrez gene ID Treatment comparison 𝑃 value Fold changeKCNQ3 29682 BV-Sal versus BV-Flx 0.01 1.3KCNQ3 29682 MD-Sal versus MD-Flx 0.02 1.3MMP9 81687 BV-Sal versus BV-Flx 0.13 1.1MMP9 81687 MD-Sal versus MD-Flx 0.01 1.5VGAT 83612 BV-Sal versus BV-Flx 0.001 1.6VGAT 83612 MD-Sal versus MD-Flx 0.72 1.1DLX1 296500 BV-Sal versus BV-Flx 0.03 1.3DLX1 296500 MD-Sal versus MD-Flx 0.25 1.0EGR2 114090 BV-Sal versus BV-Flx 0.86 1.1EGR2 114090 MD-Sal versus MD-Flx 0.04 0.7EGR4 25129 BV-Sal versus BV-Flx 0.29 1.2EGR4 25129 MD-Sal versus MD-Flx 0.38 0.9mGluR1 24414 BV-Sal versus BV-Flx 0.47 1.0mGluR1 24414 MD-Sal versus MD-Flx 0.98 1.1HDAC3 15183 BV-Sal versus BV-Flx 0.02 1.3HDAC3 15183 MD-Sal versus MD-Flx 0.92 1.1KCNV1 60326 BV-Sal versus BV-Flx 0.02 1.2KCNV1 60326 MD-Sal versus MD-Flx 0.10 1.2NFKB1 81736 BV-Sal versus BV-Flx 0.04 1.5NFKB1 81736 MD-Sal versus MD-Flx 0.52 1.0CLCN3 84360 BV-Sal versus BV-Flx 0.04 1.2CLCN3 84360 MD-Sal versus MD-Flx 0.01 1.5NR1 24408 BV-Sal versus BV-Flx 0.45 0.9NR1 24408 MD-Sal versus MD-Flx 0.43 1.1NR2A 14811 BV-Sal versus BV-Flx 0.63 1.1NR2A 14811 MD-Sal versus MD-Flx 0.04 0.8NR2B 24410 BV-Sal versus BV-Flx 0.51 1.1NR2B 24410 MD-Sal versus MD-Flx 0.07 0.8GABRA1 29705 BV-Sal versus BV-Flx 0.34 1.1GABRA1 29705 MD-Sal versus MD-Flx 0.27 0.9GABRA4 140675 BV-Sal versus BV-Flx 0.29 1.1GABRA4 140675 MD-Sal versus MD-Flx 0.02 1.3Reelin 24718 BV-Sal versus BV-Flx 0.15 1.2Reelin 24718 MD-Sal versus MD-Flx 0.001 1.4MMP2 17390 BV-Sal versus BV-Flx 0.02 0.5MMP2 17390 MD-Sal versus MD-Flx 0.01 1.6MBP 24547 BV-Sal versus BV-Flx 0.01 0.6MBP 24547 MD-Sal versus MD-Flx 0.01 0.6

vision, fluoxetine promoted an increase in the expressionof CLCN3 (20% increased), KCNV1 (20% increased), andKCNQ3 (30% increased), which encode ion channels thatmediate chloride and potassium conductance (𝑃 < 0.05), andin animals with monocular deprivation fluoxetine producedalso an increase in the expression of CLCN3 (50% increasedexpression; 𝑃 = 0.01).

3.2.5. Extracellular Matrix. The expression of MMP2 andMMP9 was markedly changed between animals treated

with fluoxetine and with binocular vision and those withmonocular deprivation (Figure 2(e)). MMP2 and MMP9encode for proteolytic enzymes that degrade extracellularmatrix components [35–37] and play a key role in mediatingsynaptic plasticity at the level of the hippocampus [38, 39]. Inparticular, MMP2 gene expression was decreased in animalstreated with fluoxetine alone (50% decrease; 𝑃 = 0.02),while animals with combined monocular deprivation andchronic fluoxetine treatment had an increased expression ofboth MMP2 (60% increased; 𝑃 = 0.01) and MMP9 (50%increased; 𝑃 = 0.01).

Neural Plasticity 5

2

1.5

1

0.5

0

Chan

ge in

gen

e exp

ress

ion

Binocular vision MD

VGATGABRA1

GABRA4Control

(a) Inhibitory neurotransmission

Chan

ge in

gen

e exp

ress

ion

2

1.5

1

0.5

0Binocular vision MD

ControlmGluR1NR1NR2A

NR2B

(b) Excitatory neurotransmission

Chan

ge in

gen

e exp

ress

ion

2

1.5

1

0.5


DLX1EGR-2EGR-4

NF𝜅B1Control

(c) Transcription factors

Chan

ge in

gen

e exp

ress

ion

2

1.5

1

0.5


KCNQ3KCNV1CLCN3

ReelinControl

(d) Synaptic plasticity

Chan

ge in

gen

e exp

ress

ion

2

1.5

1

0.5


MMP9MMP2Control

(e) Extracellular matrix

Chan

ge in

gen

e exp

ress

ion

2

1.5

1

0.5


MBPHDAC3Control

(f) Chromatin remodeling and myelination

Figure 2: Effects of fluoxetine in the expression of genes involved in critical period plasticity. qRT-PCR mRNA fold change comparisonbetween the effects of fluoxetine in rats with binocular vision and monocular deprivation. Statistical data is grouped by binocular vision ormonocular deprivation, and in each part the position of the line represents the fold change of the fluoxetine treated group (coloured line)with respect to the saline treated group (grey dashed line) (BV-Sal versus BV-Flx and MD-Sal versus MD-Flx, resp.). All gene expression wasnormalized using GAPDH as a control gene.

6 Neural Plasticity

3.2.6. Chromatin Remodeling and Myelination. Changes inthe expression of transcripts that encode an enzyme that reg-ulate chromatin susceptibility to transcription were detectedin animals with binocular vision after chronic fluoxetinetreatment. In particular, we found that Hdac3 expression wasenhanced (Figure 2(f); 30% increased𝑃 = 0.02). On the otherhand, the expression of MBP, which encodes a basic proteinof myelin, a repressive factor for visual cortex plasticity [33],was significantly reduced by fluoxetine treatment in animalswith both binocular vision (40% decreased; 𝑃 = 0.01) andmonocular deprivation (40% decreased; 𝑃 = 0.001).

4. Discussion

This study provides a large-scale analysis of changes inpatterns of gene expression associated with the reopeningof the critical period of plasticity in the adult visual systeminduced by the combination of fluoxetine treatment andmonocular deprivation. Our findings suggest a scenariowhere an enhanced serotoninergic transmission inducedby long-term fluoxetine treatment induces a shift of theinhibitory-excitatory balance [8, 29], which in turn promotesan alteration in the expression of genes involved in differentbiological themes that may underlie the functional modifica-tions in the adult visual cortex related with the reopening ofthe critical period plasticity [40].

Our results reveal that the process of plasticity reactiva-tion in adulthood involves both (i) a transient activation ofneural mechanisms normally present during early stages ofbrain development and (ii) a removal of molecular factorsthat inhibit plasticity in adulthood [19]. Gene expressionpatterns involved in processes of synaptic plasticity, neuronaldifferentiation, and outgrowth were, indeed, differentiallyregulated by chronic fluoxetine treatment.

The increased expression of Reelin may represent anexample of the first mechanism. Reelin is an extracellularglycoprotein involved in the migration and correct devel-opment of the cerebral cortex [41, 42]. Reelin is highlyexpressed by Cajal-Retzius neurons during development, butits expression is limited to a subpopulation of interneuronsduring the adulthood [43, 44]. Although the function ofReelin in adult neurons remains unclear, its overexpressionhas been shown to enhance plasticity and learning, affectingpresynaptic transmission [34, 45]. Our results demonstratean upregulation of Reelin after chronic fluoxetine treatment,suggesting that the overexpression of molecules involvedin the juvenile plasticity plays an important role in thereopening of the critical periods during the adulthood.

The proteolytic enzyme Mmp2, on the other hand, maydrive mechanisms of synaptic plasticity by degrading extra-cellular matrix components that are inhibitory for plasticity,as observed in the adult hippocampus [39]. Increase ofMmp2expression, indeed, was paralleled by a decrease of Mbp: abasic component of myelin, which is a repressive factor forvisual cortex plasticity [33]. Our analysis of gene expressionpoints towards a downregulation ofMbp following long-termantidepressant treatment, supporting the hypothesis that theremoval of factors that are inhibitory for plasticity may pro-vide a permissive environment for structural and functional

changes of neuronal circuitries in the adult nervous system[19].

Chronic fluoxetine administration has been shown topromote structural changes in both excitatory [46, 47] andinhibitory circuits [48–50]. Although there is evidence thatlong-term fluoxetine administration promotes a reductionof GABA-mediated inhibition in adult visual cortical cir-cuitries [8, 29], a compensatory mechanism might explainthe increase in the expression of VGAT or GABRA4 thatwe observe in our experiment. These results are also inagreement with previous studies, in which fluoxetine treat-ment in combination with monocular deprivation producesan increase in the elongation of the tips of interneuronaldendrites [50], supporting the idea that inhibitory neuro-transmission plays a key role in the reopening of the criticalperiods [20, 22, 23]. Similarly, the change of NMDA receptorsubunit composition, evidenced by the decrease in NMDA-2A gene expression following antidepressant treatment, isparticularly interesting in this respect. The expression ofthe NR2A subunit has been correlated with a progressivedecrease of NMDA receptor currents during development[51, 52]. This raises the possibility that a decrement of theNR2A/B ratio may increase NMDA receptors sensitivitythus causing the strengthening of synapses required for thepotentiation of the nondeprived input [53].

Another highly significant notion that emerges from ourdata is that the changes promoted by the combination offluoxetine withmonocular deprivation, regarding the expres-sion of transcription factors and proteins of the extracellularmatrix, are opposed to those promoted by fluoxetine alone.This indicates that these molecules might be underlying thestructural plasticity changes driven by monocular depriva-tion to produce the shift in the ocular dominance and itsconsolidation in the visual system [54].

Our findings support the hypothesis that the therapeuticeffect of antidepressant drugs is dependent on changes inneuronal plasticity [55, 56]. Importantly, these results openup new insights into the understanding of the mechanismsunderlying the reopening of the critical period in the adultbrain, by providing the basis of gene expression patterns fora visual deprivation paradigm that demonstrates the abilityof the nervous system to translate environmental stimuli intostructural and functional changes of neural circuitries.


Ettore Tiraboschi and RamonGuirado contributed equally tothis work.

References

[1] L. C.Katz andC. J. Shatz, “Synaptic activity and the constructionof cortical circuits,” Science, vol. 274, no. 5290, pp. 1133–1138,1996.


[3] V. B. Mountcastle, “The columnar organization of the neocor-tex,” Brain, vol. 120, part 4, pp. 701–722, 1997.

Neural Plasticity 7

[4] D. Tropea, A. vanWart, and M. Sur, “Molecular mechanisms ofexperience-dependent plasticity in visual cortex,” PhilosophicalTransactions of the Royal Society B, vol. 364, no. 1515, pp. 341–355, 2009.

[5] D.H.Hubel andT.N.Wiesel, “Theperiod of susceptibility to thephysiological effects of unilateral eye closure in kittens,” Journalof Physiology, vol. 206, no. 2, pp. 419–436, 1970.

[6] T. N. Wiesel and D. H. Hubel, “Single-cell responses in striatecortex of kittens deprived of vision in one eye,” Journal ofNeurophysiology, vol. 26, pp. 1003–1017, 1963.

[7] T. K. Hensch, “Critical period plasticity in local cortical cir-cuits,” Nature Reviews Neuroscience, vol. 6, no. 11, pp. 877–888,2005.

[8] J. F. Maya-Vetencourt, A. Sale, A. Viegi et al., “The antidepres-sant fluoxetine restores plasticity in the adult visual cortex,”Science, vol. 320, no. 5874, pp. 385–388, 2008.

[9] N. N. Karpova, A. Pickenhagen, J. Lindholm et al., “Fear erasurein mice requires synergy between antidepressant drugs andextinction training,” Science, vol. 334, no. 6063, pp. 1731–1734,2011.

[10] C. Orlando, J. Ster, U. Gerber, J. W. Fawcett, and O. Raineteau,“Perisynaptic chondroitin sulfate proteoglycans restrict struc-tural plasticity in an integrin-dependent manner,” The Journalof Neuroscience, vol. 32, Article ID 18017a, pp. 18009–18017, 2012.

[11] T. Pizzorusso, P. Medini, N. Berardi, S. Chierzi, J. W. Fawcett,and L. Maffei, “Reactivation of ocular dominance plasticity inthe adult visual cortex,” Science, vol. 298, no. 5596, pp. 1248–1251, 2002.

[12] L. Baroncelli, A. Sale, A. Viegi et al., “Experience-dependentreactivation of ocular dominance plasticity in the adult visualcortex,” Experimental Neurology, vol. 226, no. 1, pp. 100–109,2010.

[13] A. Sale, J. F. Maya-Vetencourt, P. Medini et al., “Environmen-tal enrichment in adulthood promotes amblyopia recoverythrough a reduction of intracortical inhibition,” Nature Neuro-science, vol. 10, no. 6, pp. 679–681, 2007.

[14] A. Sale, N. Berardi, and L. Maffei, “Enrich the environment toempower the brain,” Trends in Neurosciences, vol. 32, no. 4, pp.233–239, 2009.

[15] M. Scali, L. Baroncelli, M. C. Cenni, A. Sale, and L. Maffei,“A rich environmental experience reactivates visual cortexplasticity in aged rats,” Experimental Gerontology, vol. 47, no. 4,pp. 337–341, 2012.


[17] J. F. Maya-Vetencourt, E. Tiraboschi, D. Greco et al., “Expe-rience-dependent expression of NPAS4 regulates plasticity inadult visual cortex,”TheJournal of Physiology, vol. 590, pp. 4777–4787, 2012.

[18] S. Miyata, Y. Komatsu, Y. Yoshimura, C. Taya, and H. Kitagawa,“Persistent cortical plasticity by upregulation of chondroitin 6-sulfation,”Nature Neuroscience, vol. 15, no. 3, pp. 414–422, 2012.

[19] D. Bavelier, D. M. Levi, R. W. Li, Y. Dan, and T. K. Hensch,“Removing brakes on adult brain plasticity: from molecular tobehavioral interventions,” The Journal of Neuroscience, vol. 30,no. 45, pp. 14964–14971, 2010.

[20] M. Beurdeley, J. Spatazza, H. H. C. Lee et al., “Otx2 binding toperineuronal nets persistently regulates plasticity in the maturevisual cortex,” The Journal of Neuroscience, vol. 32, pp. 9429–9437, 2012.

[21] H. Morishita, J. M. Miwa, N. Heintz, and T. K. Hensch, “Lynx1,a cholinergic brake, limits plasticity in adult visual cortex,”Science, vol. 330, no. 6008, pp. 1238–1240, 2010.

[22] Z. J. Huang, A. Kirkwood, T. Pizzorusso et al., “BDNF regulatesthe maturation of inhibition and the critical period of plasticityin mouse visual cortex,” Cell, vol. 98, no. 6, pp. 739–755, 1999.

[23] D.G. Southwell, R. C. Froemke,A.Alvarez-Buylla,M. P. Stryker,and S. P. Gandhi, “Cortical plasticity induced by inhibitoryneuron transplantation,” Science, vol. 327, no. 5969, pp. 1145–1148, 2010.

[24] G. Di Cristo, B. Chattopadhyaya, S. J. Kuhlman et al., “Activity-dependent PSA expression regulates inhibitory maturation andonset of critical period plasticity,” Nature Neuroscience, vol. 10,no. 12, pp. 1569–1577, 2007.

[25] T. A. Pham, S. Impey, D. R. Storm, and M. P. Stryker, “Cre-mediated gene transcription in neocortical neuronal plasticityduring the developmental critical period,” Neuron, vol. 22, no.1, pp. 63–72, 1999.

[26] E. Putignano, G. Lonetti, L. Cancedda et al., “Developmentaldownregulation of histone posttranslational modifications reg-ulates visual cortical plasticity,” Neuron, vol. 53, no. 5, pp. 747–759, 2007.

[27] Y. Wang, Q. Gu, and M. S. Cynader, “Blockade of serotonin-2Creceptors by mesulergine reduces ocular dominance plasticityin kitten visual cortex,” Experimental Brain Research, vol. 114,no. 2, pp. 321–328, 1997.

[28] Q. Gu andW. Singer, “Involvement of serotonin in developmen-tal plasticity of kitten visual cortex,” The European Journal ofNeuroscience, vol. 7, no. 6, pp. 1146–1153, 1995.

[29] J. F. Maya-Vetencourt, E. Tiraboschi, M. Spolidoro, E. Castren,and L. Maffei, “Serotonin triggers a transient epigenetic mech-anism that reinstates adult visual cortex plasticity in rats,” TheEuropean Journal of Neuroscience, vol. 33, no. 1, pp. 49–57, 2011.

[30] K. J. Livak and T. D. Schmittgen, “Analysis of relative geneexpression data using real-time quantitative PCR and the 2-ΔΔCT method,”Methods, vol. 25, no. 4, pp. 402–408, 2001.

[31] M. Fagiolini andT.K.Hensch, “Inhibitory threshold for critical-period activation in primary visual cortex,”Nature, vol. 404, no.6774, pp. 183–186, 2000.

[32] T. K. Hensch, M. Fagiolini, N. Mataga, M. P. Stryker, S.Baekkeskov, and S. F. Kash, “Local GABA circuit control ofexperience-dependent plasticity in developing visual cortex,”Science, vol. 282, no. 5393, pp. 1504–1508, 1998.

[33] A. W. McGee, Y. Yang, Q. S. Fischer, N. W. Daw, and S.H. Strittmatter, “Neuroscience: experience-driven plasticity ofvisual cortex limited by myelin and nogo receptor,” Science, vol.309, no. 5744, pp. 2222–2226, 2005.

[34] E. J. Weeber, U. Beffert, C. Jones et al., “Reelin and ApoEreceptors cooperate to enhance hippocampal synaptic plasticityand learning,” The Journal of Biological Chemistry, vol. 277, no.42, pp. 39944–39952, 2002.

[35] T. M. Reeves, M. L. Prins, J. Zhu, J. T. Povlishock, and L. L.Phillips, “Matrix metalloproteinase inhibition alters functionaland structural correlates of deafferentation-induced sproutingin the dentate gyrus,” The Journal of Neuroscience, vol. 23, no.32, pp. 10182–10189, 2003.

[36] J. Zuo, T. A. Ferguson, Y. J. Hernandez, W. G. Stetler-Stevenson,and D. Muir, “Neuronal matrix metalloproteinase-2 degradesand inactivates a neurite- inhibiting chondroitin sulfate proteo-glycan,” The Journal of Neuroscience, vol. 18, no. 14, pp. 5203–5211, 1998.

8 Neural Plasticity

[37] V. W. Yong, C. Power, P. Forsyth, and D. R. Edwards, “Metal-loproteinases in biology and pathology of the nervous system,”Nature Reviews Neuroscience, vol. 2, no. 7, pp. 502–511, 2001.

[38] V. Nagy, O. Bozdagi, A. Matynia et al., “Matrix metalloprot-einase-9 is required for hippocampal late-phase long-termpotentiation and memory,”The Journal of Neuroscience, vol. 26,no. 7, pp. 1923–1934, 2006.

[39] S. E. Meighan, P. C. Meighan, P. Choudhury et al., “Effectsof extracellular matrix-degrading proteases matrix metallopro-teinases 3 and 9 on spatial learning and synaptic plasticity,”Journal of Neurochemistry, vol. 96, no. 5, pp. 1227–1241, 2006.

[40] J. F. Maya-Vetencourt and N. Origlia, “Visual cortex plasticity:a complex interplay of genetic and environmental influences,”Neural Plasticity, vol. 2012, Article ID 631965, 14 pages, 2012.

[41] J. Herz and Y. Chen, “Reelin, lipoprotein receptors and synapticplasticity,” Nature Reviews Neuroscience, vol. 7, no. 11, pp. 850–859, 2006.

[42] G. D’Arcangelo, G. G. Miao, S.-C. Chen, H. D. Soares, J. I.Morgan, and T. Curran, “A protein related to extracellularmatrix proteins deleted in themousemutant reeler,”Nature, vol.374, no. 6524, pp. 719–723, 1995.

[43] S. H. Fatemi, “Reelin glycoprotein: structure, biology and rolesin health and disease,” Molecular Psychiatry, vol. 10, no. 3, pp.251–257, 2005.

[44] E. Soriano and J. A. del Rıo, “The cells of cajal-retzius: still amystery one century after,” Neuron, vol. 46, no. 3, pp. 389–394,2005.

[45] S. Hellwig, I. Hack, J. Kowalski et al., “Role for reelin in neuro-transmitter release,” The Journal of Neuroscience, vol. 31, no. 7,pp. 2352–2360, 2011.

[46] T. Hajszan, N. J. MacLusky, and C. Leranth, “Short-termtreatment with the antidepressant fluoxetine triggers pyramidaldendritic spine synapse formation in rat hippocampus,” TheEuropean Journal of Neuroscience, vol. 21, no. 5, pp. 1299–1303,2005.

[47] R. Guirado, E. Varea, E. Castillo-Gomez et al., “Effects ofchronic fluoxetine treatment on the rat somatosensory cortex:activation and induction of neuronal structural plasticity,”Neuroscience Letters, vol. 457, no. 1, pp. 12–15, 2009.

[48] S. H. Fatemi, T. J. Reutiman, and T. D. Folsom, “Chronicpsychotropic drug treatment causes differential expression ofReelin signaling system in frontal cortex of rats,” SchizophreniaResearch, vol. 111, no. 1–3, pp. 138–152, 2009.

[49] R. Guirado, D. Sanchez-Matarredona, E. Varea, C. Crespo, J.M. Blasco-Ibanez, and J. Nacher, “Chronic fluoxetine treatmentin middle-aged rats induces changes in the expression ofplasticity-related molecules and in neurogenesis,” BMC Neuro-science, vol. 13, no. 1, article 5, 2012.

[50] J. L. Chen,W.C. Lin, J.W. Cha, P. T. So, Y. Kubota, and E.Nedivi,“Structural basis for the role of inhibition in facilitating adultbrain plasticity,”Nature Neuroscience, vol. 14, no. 5, pp. 587–596,2011.

[51] A. C. Flint, U. S. Maisch, J. H. Weishaupt, A. R. Kriegstein, andH. Monyer, “NR2A subunit expression shortens NMDA recep-tor synaptic currents in developing neocortex,” The Journal ofNeuroscience, vol. 17, no. 7, pp. 2469–2476, 1997.

[52] G. Carmignoto and S. Vicini, “Activity-dependent decrease inNMDA receptor responses during development of the visualcortex,” Science, vol. 258, no. 5084, pp. 1007–1011, 1992.

[53] K.K.A.Cho, L.Khibnik, B.D. Philpot, andM. F. Bear, “The ratioof NR2A/B NMDA receptor subunits determines the qualities

of ocular dominance plasticity in visual cortex,” Proceedings ofthe National Academy of Sciences of the United States of America,vol. 106, no. 13, pp. 5377–5382, 2009.

[54] A.W. Lyckman, S. Horng, C. A. Leamey et al., “Gene expressionpatterns in visual cortex during the critical period: synapticstabilization and reversal by visual deprivation,” Proceedings ofthe National Academy of Sciences of the United States of America,vol. 105, no. 27, pp. 9409–9414, 2008.

[55] E. Castren and T. Rantamaki, “The role of BDNF and its recep-tors in depression and antidepressant drug action: reactivationof developmental plasticity,” Developmental Neurobiology, vol.70, no. 5, pp. 289–297, 2010.

[56] N. Tsankova, W. Renthal, A. Kumar, and E. J. Nestler, “Epi-genetic regulation in psychiatric disorders,” Nature ReviewsNeuroscience, vol. 8, no. 5, pp. 355–367, 2007.


Review ArticleNoninvasive Strategies to Promote FunctionalRecovery after Stroke

Alessio Faralli,1,2 Matteo Bigoni,3 Alessandro Mauro,3,4

Ferdinando Rossi,1,2 and Daniela Carulli1,2

1 Department of Neuroscience, Neuroscience Institute of Turin, University of Turin, Regione Gonzole 10, 10043 Orbassano (Turin), Italy2 Neuroscience Institute Cavalieri-Ottolenghi (NICO), University of Turin, Regione Gonzole 10, 10043 Orbassano (Turin), Italy3 IRCCS Istituto Auxologico Italiano, Corso Goffredo Mameli 197, 28921 Verbania, Italy4Department of Neuroscience, University of Turin, Via Cherasco 15, 10126 Turin, Italy

Correspondence should be addressed to Daniela Carulli; [email protected]



Copyright © 2013 Alessio Faralli et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Stroke is a common and disabling global health-care problem, which is the third most common cause of death and one of the maincauses of acquired adult disability in many countries. Rehabilitation interventions are a major component of patient care. In the lastfew years, brain stimulation, mirror therapy, action observation, or mental practice with motor imagery has emerged as interestingoptions as add-on interventions to standard physical therapies. The neural bases for poststroke recovery rely on the concept ofplasticity, namely, the ability of central nervous system cells to modify their structure and function in response to external stimuli.In this review, we will discuss recent noninvasive strategies employed to enhance functional recovery in stroke patients and we willprovide an overview of neural plastic events associated with rehabilitation in preclinical models of stroke.

1. Introduction

Stroke is an acute neurological syndrome caused by disrup-tion of the cerebral blood supply. About 80% of strokes areischaemic, resulting from an obstruction of blood flow, whileabout 15% are due to a primary intracerebral hemorrhage.Stroke is one of the leading causes of chronic adult disabilityand death in western industrialized countries [1]. Neurolog-ical deficits reflect the location of the tissue damage and, inparticular, the extent of the neuronal loss. Neurons deprivedof their normal metabolic substrates cease to function inseconds and show signs of structural damage after only 2min-utes. As energy-dependent processes fail, neurons are unableto maintain their normal transmembrane ionic gradients,resulting in ion and water imbalance that triggers apoptoticand necrotic cell death cascades and, ultimately, leads to focalneurological signs and symptoms. According to the WHO’sinternational classification of function, disability, and health(ICF, WHO 2001), the impairment of brain functions mayoriginate different activity limitations (disability) and partic-ipation restriction (handicap).

Motor impairments, including hemiparesis, incoordina-tion, and spasticity, are themost commondeficits after stroke.However, functional recovery frequently occurs followingstroke, although its extent is highly variable. Some patientswith initial severe hemiparesis may eventually achieve fullrecovery, while others have little or no improvement andremain permanently disabled.There aremany reasons for thevariable degrees of recovery, including the age of the patient,the location and extent of the lesion, and individual variationsin anatomical and functional connections [2].

The neural bases for poststroke recovery rely on the con-cept of plasticity [3], namely, the ability of central nervoussystem (CNS) cells to modify their structure and function inresponse to a variety of external stimuli (experience). Theplastic/reparative properties of the brain are determined bythe balance between cell-intrinsic mechanisms and extrin-sic regulatory molecules, which is regulated by activity-dependent processes and different kinds of interaction withthe external world [4, 5]. Molecules in the adult CNS milieu,such as myelin-associated proteins (e.g., Nogo, MAG, andOmgp), factors secreted by astrocytes near the stroke site

2 Neural Plasticity

(e.g., chondroitin sulfate proteoglycans), and repulsive axonalguidance cues (e.g., semaphorins, netrins, and members ofthe ephrin family), constrain axonal sprouting and hamperthe formation of new connections [6]. In preclinical strokemodels it has been shown that pharmacological blockadeof Nogo, Nogo receptor antagonism, or digestion of chon-droitin sulfate proteoglycans by chondroitinase induce axonalsprouting and promote functional recovery [7–9]. Blockingthe semaphorin pathway reduces cortical damage after stroke[10]. Other growth inhibitors, such as EphA4 and ephrin-A5, have also recently been identified, which limit functionalrecovery and are promising targets for repair after stroke [11,12]. Interestingly, inhibition of ROCK, a downstream targetof several growth inhibitors, greatly improves outcome afterischemic stroke [12]. Several studies have also uncoveredpharmacological targets that promote a neuronal growth statein the adult CNS. For example, inosine triggers a serine/thre-onine kinase (Mst3b), enhancing axonal sprouting [13, 14].

The therapeutic potential of replacement strategies inlaboratorymodels of stroke is also under investigation. Trans-plantation of neural progenitor cells, bone marrow-derivedmesenchymal stem cells or human-induced pluripotent stemcells into the ischemically lesioned brain have been proved tobe a safe and efficient approach to promote significant func-tional recovery in experimental animals [15–17]. Nonetheless,the mechanisms underlying the beneficial effects of celltransplantation in the ischemic CNS remain uncertain, and,most importantly, to date there is no clear evidence thatdonor cells may directly contribute to the structural repair ofneuronal circuits.

In addition to pharmacological or replacement therapies,clinical and preclinical studies are currently focusing on non-invasive strategies for post-stroke rehabilitation. Clinical datashow that neurologic deficits following stroke can be treatedby physical therapy [18]. Motor rehabilitation after hemi-paretic stroke typically involves combinatory approaches,including neurofacilitation techniques, task-specific training,and task-oriented training [19, 20]. Furthermore, stroke units,in which patients have access to daily skill training therapiesin highly stimulating environments, such as during physi-cal, occupational, or language therapy, result in decreaseddeficits, increased performance on self-care tasks, lower 1-year mortality, and lower probability to be in a nursing homeat followup [21]. Finally, in recent years brain stimulation,mirror therapy, action observation, or mental practice withmotor imagery is emerging as interesting options as add-oninterventions to standard physical therapies [22].

Here, we will provide an overview of recent noninva-sive strategies employed to enhance functional recovery inpatients after stroke and discuss the current knowledge ofrehabilitative strategies and the associated neural plasticevents in preclinical models of stroke.

2. Novel Noninvasive Strategies forPatients Rehabilitation

Stroke rehabilitation aims to guarantee that stroke survivorsreach the maximum physical, functional, and psychosocialrecovery possible within the limits of their impairment. In

order to help stroke patients to fully participate in life, thefinal goal of rehabilitation should be to maximize perfor-mance of activities of daily living and independence.Throughlearning-dependent processes, rehabilitation facilitates andshapes the recovery that would occur spontaneously. Recov-ery of stroke patients is extremely heterogeneous and deter-mined by a combination of processes including functionalrestoring of damaged nervous tissue, relearning of lost skillsthrough reorganization of spared pathways (plasticity), adap-tation, and compensation for deficits. Compensation reflectsthe use of alternate behavioral strategies in order to solvea specific task. Most recovery of specific neurological focaldeficits occurs during the first 3 to 6months after stroke, but itis largely accepted that improvements can continue for yearsafter stroke [23].

General principles of stroke rehabilitation include thestart of intensive rehabilitation programs carried out in astroke unit within the first few days after stroke [24, 25]. Evi-dence demonstrates that comprehensive intensive rehabilita-tion, as well as the presence of a structured multidisciplinaryteam, may be more effective than less intense programs [26].In agreement with the learning nature of the rehabilita-tive process, involvement, engagement, and motivation ofpatients, families and caregivers are crucial to obtain goodoutcome.

Most recent neurorehabilitative approaches are based ona task-oriented model of motor learning, whose main featureis an intensive training with specific tasks in an environmen-tal context (task-specific and context-specific trainings; [27–30]). In this context, a number of new rehabilitative tech-niques potentially capable of stimulating cerebral plasticityhave been proposed and tested in the last years. Among thesetechniques, large interest is devoted to treatment approachesaimed to improve motor functions, including constraint-induced movement therapy, mental practice, mirror therapy,virtual reality, robotics, and brain stimulation techniques.

Constraint-InducedMovementTherapy (CIMT) involvesthe restriction of usage of the unaffected limb, forcing the useof the paretic one, and aiming to contrast the maladaptive“learned nonuse” of the paretic limb (the subject learns toignore the damaged limb because of its lack of functionalityand learns to use exclusively the healthy limb). A numberof studies including randomized controlled trials and aCochrane review have shown that CIMT is effective inimprovingmotor performance in human patients after stroke[31, 32] with a large effect size and robust effects especiallyon arm function [33]. In particular, the ECXITE trial [31]demonstrated that daily intensive CIMT training for upperlimb paresis was superior to the control treatment 3 to 9months after stroke, and that a modest improvement inmotor function persisted in the CIMT group after 2 years.Important limitations to the routine use of CIMT trainingderive from the fact that it is labor intensive and suitable onlyfor patients with some conservation of motor functions (inparticular wrist and finger), thus its use is recommended onlyfor selected patients.

Mental practice with motor imagery is considered apromising additional treatment to improve motor functions

Neural Plasticity 3

of severely affected upper limb [33], although its clinical effec-tiveness is not yet clearly proven. This approach grounds onthe statement that imaging a movement requires activationof brain circuits involved in the preparation and executionof the same movement and consists in a repetitive cognitivetraining during which the patient imagines performing a taskor body movement without actually physically performing it.It has been demonstrated that mental practice may modu-late cerebral perfusion and neural activity in brain regionssimilar to those activated during actual movements [34, 35].Following few systematic reviews [36, 37] suggesting thatmental practice may be beneficial for post-stroke disabilitiesin addition to conventional treatments, a recent Cochranereview [38] concluded that there is only limited evidencethat mental practice may increase the effectiveness of usualphysiotherapy and occupational therapy.

Another approach based on multisensory stimulationis represented by the mirror therapy. In this technique amirror is placed at 90∘ in the patient midsagittal plane, sothat the paretic limb is hidden behind the mirror and thepatient watches the image on the mirror of the unaffectedarm as if it was the affected arm. In a certain sense, thepatient receives the impression that the affected limb isfunctioning. It has been demonstrated [39] that viewingthe image of one’s moving hand reflected by the mirrorincreases the excitability of neurons in the ipsilateral primarymotor cortex more than directly viewing the inactive hand.Mirror therapy effects (as well as those related to mentalpractice) may be related to the activity of the so-calledmirrorneurons, which discharge both following performance ofmotor acts and simply observing the same action done byanother individual [40, 41]. In fact, by means of fMRI ithas been demonstrated [42] that prolonged and repetitiveobservation of an action may enhance the activity in theventral premotor cortex, the supplementary motor area, andthe superior temporal gyrus. A recent systematic review [43]including 14 studies and a total of 567 patients treated withmirror therapy concluded that, when compared to otherrehabilitative approaches, this treatment has a significanteffect on motor function even though this result is stronglyinfluenced by the type of intervention used as control. Thus,it remains unclear if mirror therapy should replace othertreatments for motor rehabilitation after stroke, while its roleas additional intervention is confirmed. Moreover, mirrortherapy improves activities of daily living, but this statementis limited by the small number of studies (four) examiningthis effect.

Virtual reality technologies represent a relatively newapproach for rehabilitation.Thevirtual reality idea is based onthe possibility that a computer can generate a three-dimen-sional graphical environment from numerical data [44], sothat, by using visual, aural, or haptic devices, the operator canexperience the environment as if it were a part of the world. Akey feature of all virtual reality applications is interaction: vir-tual environments are created to allow the user to interact alsowith virtual objects within the environment. In some systems,the interaction may be achieved via a mouse or a joystickbutton, while in others, a representation of the user’s handmay be generated within the environment with movement

of the virtual hand reflecting the user’s hand, thus allowinga more natural interaction with objects. Therefore, virtualreality represents a unique instrument to achieve severalrequirements for effective rehabilitation, such as repetitivepractice, feedback about performance, and motivation toendure practice [45, 46]. Specifically, by using virtual reality itis possible to drive and control exercises for patient rehabilita-tion within a functional, purposeful, and motivating context[45]. Moreover virtual reality technologies play a pivotal rolein the construction of telerehabilitation systems.

Different virtual reality approaches have been used, inparticular, for upper limb motor rehabilitation. A Cochranereview published two years ago [47], analyzing 19 randomisedand quasi-randomised trials of virtual reality that involved565 participants, concluded that there is a limited evidencethat virtual reality and interactive video games may bebeneficial in improving arm function and activity of dailyliving function when compared with the conventional treat-ments. Another, contemporary meta-analysis [48], including12 studies (5 randomized controlled trials and 7 observationalstudies) for a total of 195 patients, showed that in thelarge majority (11 over 12) of these studies virtual realityadded a significant benefit on arm motor recovery afterstroke.However, to gain convincing evidence of virtual realityeffectiveness in poststroke rehabilitation, further research isneeded based on good randomized controlled trials.

In the last years a growing interest has been addressedto robot-assisted rehabilitative treatments after stroke. Intheory, robotic devices may help administer an intenserepetitive training to facilitate recovery. Several studies havedemonstrated a significant result in motor recovery of theupper limb of patients who trained with robotic devices butno significant effect on functional ability [49]. However, theconclusion of a randomized controlled trial (UL-Robot [50])and a Cochrane meta-analysis [51] limited the significance ofthese results. In the UL-Robot trial two groups of patientsreceiving 36 therapy sessions over 12 weeks of robot-assistedtherapy or intensive conventional physical therapy, respec-tively, were compared with patients receiving usual (notintensive) care. The study failed to demonstrate a superiorityof the intensive robot therapy when compared to intensiveconventional physical therapy, but both techniques weresuperior to usual care, suggesting that intensity of trainingmay be a crucial factor for motor recovery. The Cochranereview [51], including 19 trials and 666 patients, concludedthat electromechanical and robot-assisted arm training afterstroke may improve generic activities of daily living as well asparetic arm function, but not arm muscle strength.

A phase III randomized and controlled trial (LEAPS-[52]), designed to test the efficacy of a popular techniquethat utilizes partial body-weight support with treadmilltraining, was concluded in 2011.TheLEAPS trial included 408patients randomly assigned to three groups: two groups weresubjected to a locomotor training with treadmill and body-weight support (one group initiating treatment 2months afterstroke and the second 6 months after stroke), the third groupreceived a home exercise program.The results were somewaysurprising: no significant difference was found between thethree groups concerning the improvement in walk speed,

4 Neural Plasticity

motor recovery, balance, functional status and quality oflife. Thus, locomotor training with body-weight support andtreadmill cannot be considered superior to a structured,progressive, and intensive at home treatment. Also, in thistrial all intensive interventions were more effective whencompared to non-intensive and structured care.

A promising robotic interface has been recently devel-oped by Courtine’s group to evaluate, enable, and train pat-tern generation and balance during walking in rats. Thedevise continuously and independently assists or perturbspropulsion and balance along four degrees of freedom, whilerats are progressing overground within a large workspace. Inamodel of stroke, this robotic interface improves equilibriummaintenance, thereby contributing to skilled locomotion[53].

The use of noninvasive techniques of brain stimulation tostimulate adaptive plasticity is very appealing, and the resultsobtained are exciting. Two main techniques are available toobtain both cortical enhancement and inhibition: repetitivetranscranial magnetic stimulation (rTMS) and transcranialdirect current stimulation (tDCS). rTMS, using a coil placedon the scalp, generates a focal magnetic field, which induces(transiently, focally, and reversibly) an electric current in theunderlying cortex. Low frequency stimulation (in the rangeof 1Hz) reduces cortical excitability, while higher stimulationfrequencies increase the cortical excitability. In tDCS, weakdirect currents are delivered to the cortex through twoelectrodes that polarize the underlying tissue. Electrodeposition is crucial to modulate the distribution and directionof the current flow: anodal stimulation has an excitatoryeffect by cortical neuron depolarization, while cathodal tDCShyperpolarizes neurons by decreasing cortical excitability.In general, two different approaches can be described usingnoninvasive brain stimulation: one addressed to increaseexcitability of ipsilateral damaged hemisphere (e.g., by stim-ulating primary motor cortex), and the other one directed toreduce the activity of intact surrounding or contralateral areathat can produce intra- or interhemispheric inhibition.

The purpose of these applications is to restore theunbalance between intact and lesioned hemisphere accordingto the interhemispheric competition model [54]. Moreover,Bestmann and coworkers suggested an unexpected role ofthe contralesional dorsal premotor cortex, with an elegantdemonstration by means of rTMS which showed the sup-porting activity of contralesional dorsal premotor cortex toipsilesional sensorimotor regions in particular for greaterclinical and neurophysiological impaired patients [55]. Theapplication of these approaches have produced very promis-ing results, in both acute and chronic stroke patients, recentlyreviewed by Corti et al. [56]. That review suggests that rTMSapplied to the affected hemisphere is safe and could beconsidered effective for modulating brain function and con-tributing tomotor recovery after stroke.However, the authorsstressed the need of double-blinded, sham-controlled PhaseII and Phase III clinical trials involving larger sample sizes tovalidate this treatment. In a meta-analysis of 18 randomizedcontrolled trials dedicated to the effects of rTMS on upperlimb motor impairment, Hsu et al. [57] found a significant

effect size (0.55–95% CI, 0.37–0.72) for motor outcome func-tion, with more clear effects for subcortical stroke and low-frequency rTMS applied to the unaffected hemisphere. Talelliet al. questioned about the real duration and anticipated sizeof the treatment effects in chronic stroke patients. In suchpatients they showed with a small semirandomized clinicaltrial that rTMS application does not augment the gains froma late rehabilitation program [58]. The need for randomizedcontrolled trial is even more evident to validate efficacy oftDCS, considering that its use in stroke patients is quite new[59, 60]. Recently, Khedr et al. provide an interesting evidencethat both anodal and cathodal tDCS are superior to shamstimulation in enhancing the effect of rehabilitation trainingto improve motor recovery after subacute stroke in a pilotrandomized controlled trial [61]. However, it must be stressedthat our knowledge about mechanisms underlying brainstimulation are largely incomplete.Thus, different paradigmsof brain stimulation will likely appear in the next future.

3. Noninvasive Therapies inAnimal Models of Stroke

3.1. Enriched Environment. Rehabilitative conditions instroke units, such as physical therapy and various kinds ofstimulating activities, can be partially mimicked in animalstudies by housing the animals in an enriched environment(EE). EE is awidely employed paradigm to study the influenceof external stimuli on brain plasticity in animal models bothin physiological conditions and after damage [62]. Environ-mental enrichment refers to housing conditions that facilitateenhanced sensory, social, cognitive stimulation, and motoractivity. Home cages used for enrichment are larger thanstandard cages to allow room for several objects, whichgenerally vary in composition, shape, size, texture, smell, andcolour. Enrichment may also involve access to runningwheels for enhanced voluntary exercise (Figure 1). Keyaspects appear to be the provision of environmental complex-ity, with objects that offer a range of opportunities for visual,somatosensory and olfactory stimulation, and environmentalnovelty, obtained by changing the objects and their positionin the cage, which might provide additional cognitivestimulation. Increased complexity and novelty also leadto greater levels of physical activity. Social interactions arealso favored by housing rather large groups of animals ofboth sexes together (see for review [63]). Several studiesshow that in experimental models of stroke, EE stronglypromotes recovery of motor functions, such as skilled limbfunction [64–68] and gait [69]. Compensatory mechanismshave been shown to substantially contribute to functionalimprovement after stroke [70, 71]. Compensation reflectsthe use of alternate behavioral strategies in order to solve aspecific task [70, 72]. To what extent EE enhances functionaloutcome after stroke due to compensation for lost functionsrather than their restoration is not entirely clear. Witte andcoworkers addressed this question by focusing on the timecourse of functional recovery versus motor compensationin skilled forelimb movements after cerebral ischemia inrats. The skilled reaching task allows the distinction betweenrecovery and compensation by quantitative (reaching

Neural Plasticity 5

Enriched environment

Motor training

Runningwheel

Skilledreaching

Treadmilltraining

Social interaction

Tactile stimulation

Paintbrush Whiskerstimulation

Brain stimulation

tDCS rTMS

Plasticity andfunctional recovery

Figure 1: Figure 1 summarizes some of the most used noninvasive strategies to promote neural plasticity and functional recovery inexperimental models of stroke.

success) and qualitative (movement pattern) analysis. It hasbeen shown that EE facilitates effective compensation inskilled reaching, while it does not promote restitution offunction. Namely, rotatingmovements of the forelimb duringreaching are permanently impaired and require functionalcompensation through intensified use of the upper body [68].

Interestingly, in one of the first studies on the effects ofEE on stroke animals, Ohlsson and Johansson [64] addressedwhether preoperative and postoperative environments candifferently influence functional outcome after focal brainischemia. Rats were subjected to ligation of the right middlecerebral artery (MCA) then transferred from a non enrichedto an enriched environment or reared in an EE already beforethe operation. Rats kept in an EE before and after the MCAligation improved sooner and to a slightly higher degreethan those placed in the EE only after the ischemia. Thebeneficial effects of EE in the animals enriched before MCAligation suggest that complex experiences during healthyconditions may provide a “brain reserve” against late braindamage, according to previous findings [73–78]. Among theEE-induced changes in physiological conditions, the devel-opment of new synapses [79, 80] and dendritic spines [81–83] has been demonstrated. In addition, there is evidence thatexposure to EE reduces the expression of growth-inhibitorymolecules in the intact CNS tissue [84, 85]. Therefore, itis conceivable that reduced inhibitory mechanisms togetherwith a “reserve” of synapses in enriched animals may provideneuroprotection and facilitate functional compensation afterstroke.

3.2. Motor Training. A bulk of evidence highlights the func-tional benefits induced bymotor training after focal ischemicinjury in humans. A useful method of training for chronicand acute individuals after a stroke is treadmill training [86](Figure 1). When applied to ischemic rats starting 24 h afterischemia, it leads to a significant reduction of infarct volumeand improves neurological function [87]. Interestingly, func-tional recovery after stroke (such as forelimb foot placing,parallel bar crossing, and rope or ladder climbing) can befurther improved by complex motor training (which canbe obtained by using rotarod) rather than simple repetitiveexercise, such as treadmill training [88]. This suggests thatrepeated complex movements involving motor balance andcoordination are more effective for functional recovery afterstroke than either simple activity or inactivity.

In line with this view, specific behavioral experience,such as skilled-reaching training (Figure 1), after focal exper-imental infarct, provides substantial behavioral recovery ofskilled hand function in monkeys [89]. In experimentalanimals, skilled reaching training consists of daily practiceof the impaired forelimb to retrieve food pellets. This kindof rehabilitation provides positive reinforcement (i.e., foodreward) associated with use of the impaired limb, therebyencouraging animals to practice “spared” motor function orpromoting development of compensatory motor strategies,resulting in lessened functional deficiency.

Interestingly, by combining both enriched living con-ditions and daily skilled-reaching training, Biernaskie andCorbett [66] obtained dramatic long-term improvement both

6 Neural Plasticity

in skilled use of the impaired forelimb and digits and inlimb placement in stroke rats. These findings reinforced theidea that skilled learning therapy coupled with enrichedsurroundings may facilitate neurologic recovery in humans.It should be noted, however, that the effect of forced exerciseon functional recovery after stroke is controversial. Forcedexercise, such as treadmill running or constraint-inducedmovement therapy, has been shown to enhance the functionalrecovery of motor skills after experimental ischemic stroke[90, 91]. Other studies, however, demonstrate that tread-mill running produced negative physiological adaptationsinduced by stress [92], and a constraint-induced movementstudy did not show improved functional outcome after brainischemia [93].

3.3. Social Stimuli. Patients with high levels of social supportor large social networks exhibit more rapid and extensivefunctional recovery after stroke than socially isolated individ-uals [94, 95]. The importance of social influences on strokeoutcome have been also highlighted in experimental animalsby Johansson and Ohlsson [96] (Figure 1). These authorsassessed the relative importance of postoperation physicalactivity and social interaction for functional outcome. Ratswere housed together in a large cage with no equipment orhoused individually in cages with free access to a runningwheel and compared to rats kept in an EE. Interestingly, ratshoused together in a large cage with no activity-stimulatingfacilities improve more than rats housed in individual cageswith access to a running wheel. However, rats housed in anEE improve significantly more than the other two groups,suggesting that, although increased physical or social activityalone might result in some of the beneficial effects observedwith enrichment, they do not fully account for the broaderbehavioural improvements observed following exposure tocomplex stimuli.

To study social influences on experimental stroke out-come, DeVries’ group addressed the effects of social isola-tion versus pair housing on stroke-induced infarct size andfunctional recovery in mice.They observed that pair housingdecreased infarct size and improved functional outcomeof stroke mice when compared to socially isolated mice[97]. Social interaction influences locomotor activity [98]and introduces auditory, olfactory, and visual stimuli, whichin turn may influence pathophysiological mechanisms andrecovery. Further, the same authors asked whether one aspectof social interaction, namely, physical contact, may mediatethe effects of social interaction [99]. To control for the ele-ment of physical contact during pair housing, the experimentincluded the use of standard cages fitted with a grid partitionthat allowed the experimentalmouse to see, hear, and smell itspartner but not engage in physical contact. Interestingly, onlypaired animals that were in unobstructed physical contactshowed smaller infarct volumes and exhibited recovery oflocomotor activity following MCA occlusion, indicating thatphysical contact during social interactions influences strokeoutcome. Further clinical research is, therefore, needed todetermine the influence of physical contact on patient recov-ery.

3.4. Tactile Stimulation. Another potential noninvasive treat-ment that might have a significant impact upon recoveryof skilled motor behaviors after stroke is tactile stimulation(Figure 1). When stroke rats are given tactile stimulation,which involves petting animals individually with a babyhairbrush or a paintbrush, they show dramatic improvementin the single pellet reaching task relative to untreated lesionedanimals [100].These data suggest that massage therapymightbe beneficial in resolving motor deficits in human strokepatients.

Interestingly, intermittent single whisker stimulation, ifinitiated within 2 h of permanent MCA occlusion in the rat,induces complete protection from ischemic stroke by 24 hafter injury, preventing the expected damage and deficits.Namely, animals that receive early stimulation treatmentshowno sign of infarct. An initial absent or severely disruptedwhisker functional representation is followed by gradualrecovery to baseline responses over the treatment period.Evoked subthreshold activity and spiking and blood flowlevels, which are severely decreased immediately after occlu-sion, return gradually to preocclusion levels. Blood flow datasuggest that the protection induced by early stimulation isdue to reorganized blood flow via collateral vessels (interar-terial connections). In contrast, animals that do not receivetreatment until 3 h post-MCA occlusion show compromisedfunction and large infarcts [101, 102]. These studies raisehope for the development of stimulation-based strategies tomitigate stroke pathology in humans.

3.5. Noninvasive Brain Stimulation Techniques

3.5.1. tDCS. Recent studies employed animal models toinvestigate the positive effects of tDCS and define the optimaltime window of its application after stroke (Figure 1). Bothearly (1 day after ischemia) and late (1 week after ischemia)anodal tDCS treatments exert beneficial effects on cognition,behavioral function (i.e., improvedBarnesmaze performanceand motor behavioral index scores), and neural plastic-ity, without exacerbating ischemic volume and metabolicalteration [103]. However, only the rats receiving late tDCStreatment showed improvement in the beam balance test[103]. Accordingly, in the study by Jiang et al. [104] anodal andcathodal tDCS applications from day 1 to day 3 after cerebralinfarction do not improve the beam walking test scores ofrats on day 3, but significant amelioration of motor functionis observed if the animals receive continuous application oftDCS till day 7 or 14. These findings suggest that late applica-tion of tDCS may result in stronger motor function improve-ment than earlier intervention after stroke. Accordingly, onestudy, in which anodal tDCS was applied during five dailysessions to the ipsilesional primary motor cortex in acutestroke patients starting on the 2nd day, did not reveal anysignificant difference in motor function between the tDCSand sham groups, indicating that tDCS application from day2 to day 5 after stroke does not promote functional recovery[105]. LTP and LTD may be candidates processes to explainthe cellular correlates for tDCS-induced effects [106, 107].

Neural Plasticity 7

3.5.2. rTMS. Despite the observed beneficial effects inhumans (see for review [108]), the cellular/molecular mech-anisms underlying rTMS action are far from clear. It is likelythat rTMS induces LTP or LTD, which, in turn, produceenduring changes on neocortical excitability and synapticconnections [109–111]. In humans, an increase in motor-evoked potential amplitude [110, 112], regional cerebral bloodflow, glucose metabolism [113], and EEG response amplitude[109] has been reported. Studies in animal models (Figure 1)have shown that rTMS effects depend on changes in NMDAreceptor activity [114]. Interestingly,Wang et al. [115] providedthe first evidence that rTMS induces changes in BDNF-TrkBsignaling in the rat brain, which are reflected in lymphocytes.Transcription of glial fibrillary acidic protein (GFAP) isincreased in astrocytes of the mouse dentate gyrus (the mag-nitude of this response depends on the number of stimulustrains), suggesting that rTMS induces the first stage of areactive response that is similar to what occurs followingnervous tissue injury [116]. However, the consequences ofrTMS on experimental animals after stroke have been poorlyinvestigated. Zhang et al. [117] report a significant recovery ofneurological severity score in stroke rats treated with TMS,which is accompanied by increased expression of c-Fos andBDNF in the cerebral cortex surrounding the infarction areas.

4. Is There a Critical Period forSuccessful Rehabilitation?

After clinical stroke, the initiation of physical rehabilitationprograms varies from days to several weeks after the insult.Determining whether there is a period during which thepoststroke brain is most sensitive to physical rehabilitation isessential to maximize the functional gains from such therapy.Biernaskie et al. [118] hypothesized that implementing reha-bilitative treatment early after the strokewould enhance func-tional outcome. To characterize a potential “critical period”for successful rehabilitation after stroke, animals receivedenriched rehabilitative training at 5 d, 14 d, or 30 d after MCAocclusion. Early initiation of enriched rehabilitation (5 d afterstroke) provides enhanced functional outcome relative toischemic animals receiving delayed rehabilitation, suggestingthat the poststroke brain is in a state of heightened sensitivityto behavioral experience. In line with those findings, Barbayet al. [119] demonstrate a time-dependent, rehabilitation-induced map reorganization after ischemic injury in pri-mates. Similarly, early treadmill training (started 24 h post-MCA occlusion) was found to have significant effects inreducing brain infarct volume and in improving neurologicfunction, when compared with late training (started 1 weekpost-MCA occlusion, [87]). Nevertheless, some evidencessuggest that early training after focal brain ischemia in ratsexacerbates brain damage and worsens the general outcomeafter excessive use of the impaired limb. Namely, when theintact forelimb is constrained immediately after the surgicalprocedure, thus forcing the animal to overuse the impairedforelimb for postural support and movements, functionalimprovement is reduced [120, 121]. The intensity of trainingmay contribute to early exclusive use-dependent exaggerationof injury. For example, in the study by Yang et al. [87],

the intensity of treadmill training for 30min/day seems tobe mild compared to forced use by casting procedures.Excessive sensorimotor activation too early after the insultmay exacerbate injury through a use dependent, NMDA-mediated process, possibly stimulating an excitotoxic cascade[122]. This process may dissipate over days, explaining whyrehabilitative experience beginning 3–5 d after insult doesnot worsen injury size or behavioral outcome [89, 123].In addition, during the first week after injury, the tissuesurrounding the infarct is reported to show decreased phasicinhibition and thus become hyperexcitable [124]. However,Carmicheal’s group show that while phasic GABA signaling isreduced in the first weeks after stroke, tonic GABA signalingis potentiated in peri-infarct motor neurons. Behavioral andelectrophysiological studies in mice suggest that the overalleffect in terms of motor cortex circuitry is a diminishedneuronal excitability, which when reversed leads to recovery.Therefore, the precise signaling systems in brain excitabilitythat are deleterious in the early phases, become beneficialin later phases of recovery (see for a comprehensive reviewon brain excitability in stroke [125]). Rehabilitation mayact by affecting this delicate balance between hypo- andhyperexcitability of neuronal circuits in peri-infarct cortex.

Interestingly, immediate exposure to EE improves func-tional outcome, despite exacerbation of ischemic injury [67,126], perhaps as a consequence of removal of functionallyabnormal neurons. Nonetheless, early EE combined withtraining enhances recovery when compared with conditionsinwhich rehabilitation is started later and is not accompaniedby any exacerbation of injury [118]. In addition, a “window ofopportunity” extends also to neurovascular changes, whichcan facilitate full protection [101].

In summary, the efficacy of rehabilitative therapy afterstroke is influenced by the time of its initiation, with mildintensity physical training provided early after brain injurybeing beneficial for functional improvement. Delaying thebeginning of rehabilitation may instead reduce the efficacyof treatment and, as a consequence, more intense or longerduration therapies are required to achieve the same func-tional gains.

5. Cellular and Molecular Correlates ofRehabilitation-Induced Plasticity

5.1. Neuritic Plasticity, Reorganization of Connectivity, andCircuit Rewiring. Much of the recovery after stroke is likelydue to brain plasticity, with some areas of the brain takingover the functions previously performed by the damagedregions. Proposed mechanisms include: (i) redundancy ofbrain circuitry with alternative pathways taking over whenanother one has been damaged; (ii) unmasking of previouslyexisting but functionally inactive networks; (iii) sproutingof fibers from surviving neurons with formation of newsynapses [127, 128]. The mechanisms involved likely dependon the extent of injury. When damage to a functional systemis partial, within-system recovery is possible, whereas aftercomplete destruction, substitution by a functionally relatedsystem may be the only alternative [129].

8 Neural Plasticity

In stroke patients, improved arm and hand movementand clinical scores have been found in correlation with anenlargement of the hand region in the ipsilesional cortex[130–135]. However, the exact mechanisms behind thesechanges remain elusive. Activity changes in specific corti-cal areas may result from a reduction in inhibition fromhorizontal or callosal connections [136]. Alternatively, newconnections may form due to lesion-induced sprouting atthe cortical or subcortical level [137, 138]. Reorganizationof neuronal connectivity around the lesion site and also inthe undamaged contralateral cortex has been detected [139–141]. Interestingly, following an ischemic subtotal lesion of therat forelimb motor cortex, spontaneous recovery of forelimbfunction is correlated with hindlimb corticospinal neuronsforming new connections with cervical, forelimb-related,spinal cord neurons [142].

Rewiring of connections after stroke is further enhancedby rehabilitation. For example, while an ischemic lesionconfined to a small portion of the representation of one handresults in a further loss of hand territory in the adjacent,undamaged cortex, early rehabilitative training prevents theloss of hand territory adjacent to the infarct. In some instan-ces, the hand representation expands into regions formerlyoccupied by representations of the elbow and shoulder.Functional reorganization in the undamaged motor cortex isaccompanied by behavioral recovery of skilled hand function[89]. Moreover, stroke rats housed in an EE or receivingtactile stimulation [100] have significantly increased dendriticbranching and spine density on pyramidal cortical neuronsthan control stroke rats, suggestive of increased sprouting ofintracortical connections in the enriched/stimulated group[143]. Indeed, EE can influence a number of factors, suchas functional enforcement of existing neuronal circuits,sprouting, formation of new connections, and angiogenesis[63, 144]. EE may also modulate ischemia-induced glutamateexcitotoxicity, thus leading to attenuated oxidative damageand neurodegeneration [145]. One candidate mechanismunderlying the beneficial effects of EE on functional recoveryafter stroke involves upregulation of neurotrophic factors[146, 147], which may stimulate neuritic remodeling andsynaptogenesis.

Cortical neurons that sprout a new connection afterstroke activate a neuronal growth program that consists oftranscription factors, cell adhesion, axonal guidance, andcytoskeletal modifying molecules [148]. It is known that EEmodulates the expression of several genes in the infarctedcortex [149]. Namely, postischemic EE or social interactionmodulate the expression of substances associated with neu-ronal plasticity, such as nerve growth factor-induced geneA (NGFI-A) and NGFI-B. NGFI-A (also known as Egr1,krox24, zif/268, and TIS8), a transcription factor belongingto the early growth response family [150], is associated withstabilisation of LTP and learning [151, 152]. NGFI-A targetgenes are synapsin-I and -II, which are involved in synapticvesicle trafficking and release as well as synaptogenesis [153–155]. Synapsin-I and –II are increased in the ipsilateral cortexof stroke rats following skilled training [156]. In addition,NGFI-A is a master switch for the initiation of inflammatory

gene expression under ischemic stress [157]. NGFI-B (alsoknown as Nur77, N10, TIS1, or TR3), a member of thesteroid/thyroid receptor family without any known ligand[158], has also been associated with LTP [159]. At one monthfollowing MCA occlusion the mRNA expression of NGFI-A and NGFI-B is increased after EE in the cerebral cortexand the hippocampus [160]. However, other reports show adecreased expression of NGFI-A in both cortices of EE rats[161–163], likely reflecting the suppression of postischemicinflammation in the brain. Differences in the intensity andthe duration of exercise administered to the rats may accountfor the different results obtained.

5.2. Compensatory Neurogenesis. Postlesional plasticity inthe adult brain is not restricted to structural modifications atthe level of axons, dendrites, and synapses but also comprisesthe generation, differentiation, and maturation of newneurons in circumscribed brain regions (reviewed by [164]).Numerous studies utilizing different experimental modelshave shown that an ischemic CNS lesion leads to a substantialincrease in proliferation of neural stem cells and subsequentlyincreased generation of new neurons in the subgranular zoneof the dentate gyrus and in the subventricular zone (SVZ)(see for review [143]). Dentate neurogenesis is stimulatedby focal ischemic infarcts even when the site of the injury islocated in remote cortical brain areas [165, 166]. Newbornneurons in the SVZ are recruited to infarcted areas and maystart to express region-specific mature neuronal markers[167–172]. However, newborn cells expressing matureand region-appropriate neuronal markers have only beenobserved in the ischemic striatum but not in the cerebralcortex, with low fractions of newly generated cells survivinginto maturity [167, 168, 173]. Possible reasons for the reducedincidence of neuronal replacement in the ischemic striatumand cortex could be low cell survival or hampered neuronalphenotypic maturation due to detrimental factors in theperilesional environment, lack of neurotrophic supportand of necessary developmental cues. Notably, ablationof doublecortin-positive neuronal precursors from therostral SVZ and dentate gyrus abolishes neurogenesis andassociated neuronal migration induced by focal cerebralischemia. This results in increased infarct size and worsenedneurologic deficits, indicating that neurogenesis contributesto neuroprotection and short-term functional outcome afterexperimental stroke in mice [174]. Those beneficial effectsmay depend on the release of chemical mediators (e.g.,growth factors) by immature neurons [175].

Studies on the effects of EE and exercise on the adultgerminal niches in intact animals have shown that both theseparadigms lead to increased neurogenesis in the hippocam-pus and the SVZ [63]. However, environmental and physicalactivities affect the lesioned brain differently. For example,postischemic EE enhances cell proliferation in the SVZ, withstronger effects in the chronic poststroke phase [171, 176],while wheel-running exercise after neocortical infarctionattenuates the early poststroke activation of the SVZ germinalniche [176]. Interestingly, no effect of EE or exercise onhippocampal progenitor cell proliferation is reported after

Neural Plasticity 9

transient global ischemia in rats [177], suggesting that com-mon pathways of regulation by lesion and environmentalinterventions may exist [178–180]. In contrast, specific reha-bilitative training of the impaired forelimb (skilled reachingtraining) is able to increase dentate neurogenesis relativeto nontrained stroke rats, although at lower levels whencompared with sham-operated animals. Moreover, increasedlevels of newborn granule cells generated in the dentate gyruscorrelate with better functional outcomes [181, 182].

Interestingly, postischemic EE combined with spatiallearning (which simulates occupational therapy in humanrehabilitation and activate hippocampus and prefrontal cor-tex) restores the perturbed dentate gyrus neuroblast pro-duction resulting from focal ischemic insult and increasesneuroprotection in the ischemic penumbra [183].

5.3. The Contribution of Glial Cells to Postlesion Plasticityand Repair. The lack or inadequacy of endogenous neuronalreplacement after brain lesions encouraged investigations onthe role of glial cells in poststroke recovery process. Increas-ing evidence indicate that glial cells crucially contribute to thedegenerative and regenerative processes following ischemicbrain lesions [184, 185]. Also, some of the beneficial effectsof EE on the postischemic brain might be mediated by adynamic modulation of different glial populations.

It is well known that astrocytes are essential for optimalneuronal function and take an active part in synaptic genera-tion and plasticity as well as in maintenance of neuronal andsynaptic homeostasis [186–188]. Recently, it has been revealedthat astroglia may represent neural stem cells in the adultbrain and may also direct neuronal differentiation of adultneural stem cells [189–191].

After brain insults like stroke, astrocytes play a multi-faceted role [184].They immediately proliferate in response tothe lesion, increase their expression of GFAP, and contributeto the formation of the glial scar [192, 193]. Reactive astrogliamight provide a protective environment in the perilesionalzone by shielding neurons from oxidative stress [194, 195]or producing antiapoptotic and trophic factors. Accordingly,they might promote neuronal survival, synaptic remodelling,and neurite outgrowth [184, 196–199]. Postischemic EE ordaily training of the impaired forelimb enhances astrogliosisin the perilesional area [171, 176, 193]. Reactive astroglia,although representing an impediment for axon growth, mayfulfill important protective and reparative functions afterischemic injuries and rehabilitation [199–203].

Immediately after the ischemic insult, resting microgliachange their morphology from a ramified to an activatedhyperramified phenotype and express the CD68 antigen[204]. The activated microglia migrate towards the lesion,remove the necrotic tissue by phagocytosis, and therebybecome macrophages [205, 206]. Some macrophages derivefrom monocytes that cross the blood-brain barrier afterthe ischemic lesion [207, 208]. Besides the degradation ofnecrotic cells, activated microglia and macrophages releasegrowth factors and scavenge-free radicals [209, 210]. How-ever, activated microglia could also harm the injured brainwith the synthesis of potentially toxic substances like nitric

oxide and reactive oxygen radicals or the release of gluta-mate and proinflammatory cytokines [209, 211–216]. Indeed,recent studies show that suppression of activated microgliaand macrophages significantly improve functional recoveryafter focal ischemic infarcts [217, 218]. In stroke animalsexposed to EE or training a reduction of proliferatingmicroglia andmacrophages is observed, which may favor thebetter functional outcome observed [193].

Finally, proliferation and survival of immature andmature oligodendrocytes are only slightly influenced by EE.It has been shown that EE increases the number of NG2-positive glia, in intact ipsi-and contralateral cortical regionsremote from the infarct [219]. NG2-positive cells possesssome characteristics of multipotent progenitor cells, maysupport neuronal function, and can turn intomyelin-formingoligodendrocytes [220–223]. However, the role of this cellpopulation in the injured brain is still obscure.

6. Conclusions

Novel noninvasive interventions for stroke patients, such asmental practice, mirror therapy, virtual reality, robotics, andbrain stimulation techniques, are emerging as potentially effi-cient strategies to promote functional recovery, but in mostcases only when provided in combination with physical reha-bilitation [43, 224]. The expansion of rehabilitative programswith a wide range of possible interventions is more likely thekey to obtain optimal results, by stimulating different repara-tive and adaptive brain processes. Particularly, the use of non-invasive techniques of brain stimulation to promote adaptiveplasticity, such as tDCS and rTMS, is very appealing, and theresults obtained in preclinical and clinical models of strokeare exciting. In this context, however, randomized controlledtrials are needed to validate the efficacy of these techniques.Moreover, a deeper understanding of the underlying mech-anisms is necessary. This knowledge may allow the identifi-cation of biological markers suitable to monitor plastic pro-cesses in human patients undergoing specific rehabilitativeprograms, predict the outcome of the treatments, and opti-mise existing procedures. In conclusion, in the last few yearsthere has been an enormous progress in the field of rehabil-itative trials after stroke, for example, in terms of standard-ized interventions and tools for assessment of function andpatient selection (e.g., recruitment of homogeneous groupsof patients). Crucial issues, however, remain to be addressedin future studies, including the sample wideness, repeatabilityof the results, and effective outcome measurements.

Acknowledgment

This work is partially supported by the Italian Ministry ofHealth (RF-2009-1472190).

References

[1] P. L. Kolominsky-Rabas, M.Weber, O. Gefeller, B. Neundoerfer,and P. U. Heuschmann, “Epidemiology of ischemic stroke sub-types according to TOAST criteria: incidence, recurrence, and


long-term survival in ischemic stroke subtypes: a population-based study,” Stroke, vol. 32, no. 12, pp. 2735–2740, 2001.

[2] P. Langhorne, J. Bernhardt, and G. Kwakkel, “Stroke rehabilita-tion,”The Lancet, vol. 377, no. 9778, pp. 1693–1702, 2011.

[3] S. T. Carmichael, “Targets for neural repair therapies afterstroke,” Stroke, vol. 41, no. 10, pp. S124–S126, 2010.

[4] F. Rossi, S. Gianola, and L. Corvetti, “Regulation of intrinsicneuronal properties for axon growth and regeneration,”Progressin Neurobiology, vol. 81, no. 1, pp. 1–28, 2007.

[5] S. Foscarin, F. Rossi, and D. Carulli, “Influence of the envi-ronment on adult CNS plasticity and repair,” Cell and TissueResearch, vol. 349, no. 1, pp. 161–167, 2012.

[6] G. Yiu and Z. He, “Glial inhibition of CNS axon regeneration,”Nature Reviews Neuroscience, vol. 7, no. 8, pp. 617–627, 2006.

[7] J. Lee, J. Kim, M. Sivula, and S. M. Strittmatter, “Nogo receptorantagonism promotes stroke recovery by enhancing axonalplasticity,” Journal of Neuroscience, vol. 24, no. 27, pp. 6209–6217,2004.

[8] J. J. Hill, K. Jin, X. O. Mao, L. Xie, and D. A. Greenberg, “Intrac-erebral chondroitinase ABC and heparan sulfate proteoglycanglypican improve outcome fromchronic stroke in rats,”Proceed-ings of the National Academy of Sciences of the United States ofAmerica, vol. 5, no. 109, pp. 9155–9160, 2012.

[9] S. Soleman, P. K. Yip, D. A. Duricki, and L. D. F.Moon, “Delayedtreatment with chondroitinase ABC promotes sensorimotorrecovery and plasticity after stroke in aged rats,” Brain, vol. 135,no. 4, pp. 1210–1223, 2012.

[10] A. Pekcec, K. Yigitkanli, J. E. Jung et al., “Following experimen-tal stroke, the recovering brain is vulnerable to lipoxygenase-dependent semaphorin signaling,” FASEB Journal, vol. 27, no. 2,pp. 437–445, 2013.

[11] J. J. Overman, A. N. Clarkson, I. B. Wanner et al., “A role forephrin-A5 in axonal sprouting, recovery, and activity-depend-ent plasticity after stroke,” Proceedings of the National Academyof Sciences of the United States of America, vol. 109, no. 33, pp.2230–2239, 2012.

[12] R. Lemmens, T. Jaspers, W. Robberecht, and V. N. Thijs,“Modifying expression of EphA4 and its downstream targetsimproves functional recovery after stroke,” Human MolecularGenetics, vol. 22, no. 11, pp. 2214–2220, 2013.

[13] B. Lorber, M. L. Howe, L. I. Benowitz, and N. Irwin, “Mst3b, anSte20-like kinase, regulates axon regeneration in mature CNSand PNS pathways,” Nature Neuroscience, vol. 12, no. 11, pp.1407–1414, 2009.

[14] L. I. Benowitz and S. T. Carmichael, “Promoting axonal rewiringto improve outcome after stroke,” Neurobiology of Disease, vol.37, no. 2, pp. 259–266, 2010.

[15] C. Cossetti, C. Alfaro-Cervello, M. Donega, G. Tyzack, and S.Pluchino, “New perspectives of tissue remodelling with neuralstem and progenitor cell-based therapies,” Cell and TissueResearch, vol. 349, no. 1, pp. 321–329, 2012.

[16] O. Honmou, R. Onodera, M. Sasaki, S. G. Waxman, and J.D. Kocsis, “Mesenchymal stem cells: therapeutic outlook forstroke,”Trends inMolecularMedicine, vol. 18, no. 5, pp. 292–297,2012.

[17] K. Oki, J. Tatarishvili, J. Wood et al., “Human-induced pluripo-tent stem cells form functional neurons and improve recoveryafter grafting in stroke-damaged brain,” Stem Cells, vol. 30, no.6, pp. 1120–1133, 2012.

[18] T.H.Murphy andD.Corbett, “Plasticity during stroke recovery:from synapse to behaviour,” Nature Reviews Neuroscience, vol.10, no. 12, pp. 861–872, 2009.

[19] J. D. Schaechter, “Motor rehabilitation and brain plasticity afterhemiparetic stroke,” Progress in Neurobiology, vol. 73, no. 1, pp.61–72, 2004.

[20] M. F. Levin, J. A. Kleim, and S. L.Wolf, “What domotor “recov-ery” and “compensationg” mean in patients following stroke?”Neurorehabilitation and Neural Repair, vol. 23, no. 4, pp. 313–319, 2009.

[21] R. Teasell, J. Bitensky, N. Foley, andN. A. Bayona, “Training andstimulation in post stroke recovery brain reorganization,”Topicsin Stroke Rehabilitation, vol. 12, no. 3, pp. 37–45, 2005.

[22] S. L. Small, G. Buccino, and A. Solodkin, “The mirror neuronsystem and treatment of stroke,” Developmental Psychobiology,vol. 54, no. 3, pp. 293–310, 2012.

[23] S. J. Page, D. R. Gater, and P. Bach-Y-Rita, “Reconsidering themotor recovery plateau in stroke rehabilitation,” Archives ofPhysical Medicine and Rehabilitation, vol. 85, no. 8, pp. 1377–1381, 2004.

[24] J. Bernhardt, M. N. Thuy, J. M. Collier, and L. A. Legg, “Veryearly versus delayed mobilisation after stroke,” Cochrane Data-base of Systematic Reviews, no. 1, Article ID CD006187, 2009.

[25] Stroke Unit Trialists’ Collaboration, “Organised inpatient(stroke unit) care for stroke,” Cochrane Database of SystematicReviews, no. 1, Article ID CD000197, 2002.

[26] TheEuropean StrokeOrganisation (ESO) Executive Committeeand ESO Writing Committee, “Guidelines for management ofischaemic stroke and transient ischaemic attack,” Cerebrovascu-lar Diseases, vol. 25, pp. 457–507, 2008.

[27] G. Kwakkel, R. C.Wagenaar, J. W. R. Twisk, G. J. Lankhorst, andJ. C. Koetsier, “Intensity of leg and arm training after primarymiddle-cerebral-artery stroke: a randomised trial,” The Lancet,vol. 354, no. 9174, pp. 191–196, 1999.

[28] R. F. Macko, F. M. Ivey, and L. W. Forrester, “Task-oriented aer-obic exercise in chronic hemiparetic stroke: training protocolsand treatment effects,”Topics in Stroke Rehabilitation, vol. 12, no.1, pp. 45–57, 2005.

[29] I. G. L. van de Port, S. Wood-Dauphinee, E. Lindeman, andG. Kwakkel, “Effects of exercise training programs on walkingcompetency after stroke: a systematic review,” The AmericanJournal of Physical Medicine and Rehabilitation, vol. 86, no. 11,pp. 935–951, 2007.

[30] P. Govender and L. Kalra, “Benefits of occupational therapy instroke rehabilitation,” Expert Review of Neurotherapeutics, vol.7, no. 8, pp. 1013–1019, 2007.

[31] S. L. Wolf, C. J. Winstein, J. P. Miller et al., “Effect of constraint-induced movement therapy on upper extremity function 3 to 9months after stroke: the EXCITE randomized clinical trial,”TheJournal of the AmericanMedical Association, vol. 296, no. 17, pp.2095–2104, 2006.

[32] V. Sirtori, D. Corbetta, L. Moja, and R. Gatti, “Constraint-induced movement therapy for upper extremities in strokepatients,” Cochrane Database of Systematic Reviews, no. 4,Article ID CD004433, 2009.

[33] P. Langhorne, F. Coupar, and A. Pollock, “Motor recovery afterstroke: a systematic review,”The Lancet Neurology, vol. 8, no. 8,pp. 741–754, 2009.

[34] P. L. Jackson, M. F. Lafleur, F. Malouin, C. L. Richards, andJ. Doyon, “Functional cerebral reorganization following motorsequence learning through mental practice with motor image-ry,” NeuroImage, vol. 20, no. 2, pp. 1171–1180, 2003.

[35] S. J. Page, J. P. Szaflarski, J. C. Eliassen, H. Pan, and S. C. Cramer,“Cortical plasticity followingmotor skill learning duringmental


practice in stroke,” Neurorehabilitation and Neural Repair, vol.23, no. 4, pp. 382–388, 2009.

[36] S. M. Braun, A. J. Beurskens, P. J. Borm, T. Schack, and D. T.Wade, “The effects of mental practice in stroke rehabilitation: asystematic review,” Archives of Physical Medicine and Rehabili-tation, vol. 87, no. 6, pp. 842–852, 2006.

[37] A. Zimmermann-Schlatter, C. Schuster, M. A. Puhan, E. Sieki-erka, and J. Steurer, “Efficacy of motor imagery in post-strokerehabilitation: a systematic review,” Journal of NeuroEngineeringand Rehabilitation, vol. 5, article 8, 2008.

[38] R. E. Barclay-Goddard, T. J. Stevenson, W. Poluha, and L. Thal-man, “Mental practice for treating upper extremity deficits inindividuals with hemiparesis after stroke,” Cochrane Databaseof Systematic Reviews, no. 5, Article ID CD005950, 2011.

[39] M. I. Garry, A. Loftus, and J. J. Summers, “Mirror, mirror onthe wall: viewing a mirror reflection of unilateral hand move-ments facilitates ipsilateral M1 excitability,” Experimental BrainResearch, vol. 163, no. 1, pp. 118–122, 2005.

[40] G. Rizzolatti and L. Craighero, “The mirror-neuron system,”Annual Review of Neuroscience, vol. 27, pp. 169–192, 2004.

[41] G. Rizzolatti and C. Sinigaglia, “The functional role of the pari-eto-frontal mirror circuit: interpretations and misinterpreta-tions,” Nature Reviews Neuroscience, vol. 11, no. 4, pp. 264–274,2010.

[42] D. Ertelt, S. Small, A. Solodkin et al., “Action observation has apositive impact on rehabilitation of motor deficits after stroke,”NeuroImage, vol. 36, no. 2, pp. T164–T173, 2007.

[43] H. Thieme, J. Mehrholz, M. Pohl, J. Behrens, and C. Dohle,“Mirror therapy for improving motor function after stroke,”Cochrane Database of Systematic Reviews, no. 3, Article IDCD008449, 2012.

[44] G. Riva, “Applications of virtual environments in medicine,”Methods of Information in Medicine, vol. 42, no. 5, pp. 524–534,2003.

[45] H. Sveistrup, “Motor rehabilitation using virtual reality,” Journalof NeuroEngineering and Rehabilitation, vol. 1, article 10, 2004.

[46] M. K. Holden, “Virtual environments for motor rehabilitation:review,” Cyberpsychology and Behavior, vol. 8, no. 3, pp. 187–211,2005.

[47] K. E. Laver, S. George, S. Thomas, J. E. Deutsch, and M. Crotty,“Virtual reality for stroke rehabilitation,” Cochrane Database ofSystematic Reviews, no. 9, Article ID CD008349, 2011.

[48] G. Saposnik and M. Levin, “Virtual reality in stroke rehabilita-tion: ameta-analysis and implications for clinicians,” Stroke, vol.42, no. 5, pp. 1380–1386, 2011.

[49] G. Kwakkel, B. J. Kollen, and H. I. Krebs, “Effects of robot-assisted therapy on upper limb recovery after stroke: a system-atic review,” Neurorehabilitation and Neural Repair, vol. 22, no.2, pp. 111–121, 2008.

[50] A. C. Lo, P. D. Guarino, L. G. Richards et al., “Robot-assistedtherapy for long-term upper-limb impairment after stroke,”TheNew England Journal of Medicine, vol. 362, no. 19, pp. 1772–1783,2010.

[51] J. Mehrholz, T. Platz, J. Kugler, andM. Pohl, “Electromechanicaland robot-assisted arm training for improving arm functionand activities of daily living after stroke,” Cochrane Database ofSystematic Reviews, no. 4, Article ID CD006876, 2008.

[52] P. W. Duncan, K. J. Sullivan, A. L. Behrman et al., “Body-weight—supported treadmill rehabilitation after stroke,” TheNew England Journal of Medicine, vol. 364, no. 21, pp. 2026–2036, 2011.

[53] N. Dominici, U. Keller, H. Vallery et al., “Versatile roboticinterface to evaluate, enable and train locomotion and balanceafter neuromotor disorders,” Nature Medicine, vol. 18, no. 7, pp.1142–1147, 2012.

[54] F. C. Hummel and L. G. Cohen, “Non-invasive brain stimu-lation: a new strategy to improve neurorehabilitation afterstroke?”The Lancet Neurology, vol. 5, no. 8, pp. 708–712, 2006.

[55] S. Bestmann, O. Swayne, F. Blankenburg et al., “The role of con-tralesional dorsal premotor cortex after stroke as studied withconcurrent TMS-fMRI,” Journal of Neuroscience, vol. 30, no. 36,pp. 11926–11937, 2010.

[56] M. Corti, C. Patten, and W. Triggs, “Repetitive transcranialmagnetic stimulation of motor cortex after stroke: a focusedreview,” The American Journal of Physical Medicine and Reha-bilitation, vol. 91, no. 3, pp. 254–270, 2012.

[57] W. Y. Hsu, C. H. Cheng, K. K. Liao, I. H. Lee, and Y. Y.Lin, “Effects of repetitive transcranial magnetic stimulation onmotor functions in patientswith stroke: ameta-analysis,” Stroke,vol. 43, no. 7, pp. 1849–1857, 2012.

[58] P. Talelli, A. Wallace, M. Dileone et al., “Theta burst stimulationin the rehabilitation of the upper limb: a semirandomized,placebo-controlled trial in chronic stroke patients,” Neuroreha-bilitation and Neural Repair, vol. 26, no. 8, pp. 976–987, 2012.

[59] G. P. Schjetnan, J. Faraji, G. A.Metz,M. Tatsuno, and A. Luczak,“Transcranial direct current stimulation in stroke rehabilita-tion: a review of recent advancements,” Stroke Research andTreatment, vol. 2013, Article ID 170256, 14 pages, 2013.

[60] W. W. Feng, M. G. Bowden, and S. Kautz, “Review of transcra-nial direct current stimulation in poststroke recovery,” Topics inStroke Rehabilitation, vol. 20, no. 1, pp. 68–77, 2013.

[61] E. M. Khedr, O. A. Shawky, D. H. El-Hammady et al., “Effect ofanodal versus cathodal transcranial direct current stimulationon stroke rehabilitation: a pilot randomized controlled trial,”Neurorehabilitation and Neural Repair, 2013.

[62] B. Will, R. Galani, C. Kelche, and M. R. Rosenzweig, “Recoveryfrom brain injury in animals: relative efficacy of environmentalenrichment, physical exercise or formal training (1990–2002),”Progress in Neurobiology, vol. 72, no. 3, pp. 167–182, 2004.

[63] J. Nithianantharajah and A. J. Hannan, “Enriched environ-ments, experience-dependent plasticity and disorders of thenervous system,” Nature Reviews Neuroscience, vol. 7, no. 9, pp.697–709, 2006.

[64] A. L. Ohlsson and B. B. Johansson, “Environment influencesfunctional outcome of cerebral infarction in rats,” Stroke, vol.26, no. 4, pp. 644–649, 1995.

[65] B. B. Johansson, “Functional outcome in rats transferred to anenriched environment 15 days after focal brain ischemia,” Stroke,vol. 27, no. 2, pp. 324–326, 1996.

[66] J. Biernaskie and D. Corbett, “Enriched rehabilitative trainingpromotes improved forelimb motor function and enhanceddendritic growth after focal ischemic injury,” Journal of Neuro-science, vol. 21, no. 14, pp. 5272–5280, 2001.

[67] A. Risedal, B. Mattsson, P. Dahlqvist, C. Nordborg, T. Olsson,and B. B. Johansson, “Environmental influences on functionaloutcome after a cortical infarct in the rat,” Brain Research Bulle-tin, vol. 58, no. 3, pp. 315–321, 2002.

[68] M. Knieling, G. A.Metz, I. Antonow-Schlorke, andO.W.Witte,“Enriched environment promotes efficiency of compensatorymovements after cerebral ischemia in rats,” Neuroscience, vol.163, no. 3, pp. 759–769, 2009.


[69] Y. Wang, B. Bontempi, S. M. Hong et al., “A comprehensiveanalysis of gait impairment after experimental stroke and thetherapeutic effect of environmental enrichment in rats,” Journalof Cerebral Blood Flow andMetabolism, vol. 28, no. 12, pp. 1936–1950, 2008.

[70] I. Q. Whishaw, “Loss of the innate cortical engram for actionpatterns used in skilled reaching and the development ofbehavioral compensation following motor cortex lesions in therat,” Neuropharmacology, vol. 39, no. 5, pp. 788–805, 2000.

[71] J. W. Krakauer, “Motor learning: its relevance to stroke recoveryand neurorehabilitation,” Current Opinion in Neurology, vol. 19,no. 1, pp. 84–90, 2006.

[72] G. A. Metz, I. Antonow-Schlorke, and O. W. Witte, “Motorimprovements after focal cortical ischemia in adult rats aremediated by compensatory mechanisms,” Behavioural BrainResearch, vol. 162, no. 1, pp. 71–82, 2005.

[73] A. vanDellen, C. Blakemore, R. Deacon, D. York, and A. J. Han-nan, “Delaying the onset of Huntington’s in mice,” Nature, vol.404, no. 6779, pp. 721–722, 2000.

[74] G. W. Arendash, M. F. Garcia, D. A. Costa, J. R. Cracchiolo, I.M. Wefes, and H. Potter, “Environmental enrichment improvescognition in aged Alzheimer’s transgenic mice despite stable 𝛽-amyloid deposition,” NeuroReport, vol. 15, no. 11, pp. 1751–1754,2004.

[75] J. L. Jankowsky, T.Melnikova, D. J. Fadale et al., “Environmentalenrichment mitigates cognitive deficits in a mouse model ofAlzheimer’s disease,” Journal of Neuroscience, vol. 25, no. 21, pp.5217–5224, 2005.

[76] O. Lazarov, J. Robinson, Y. P. Tang et al., “Environmental enrich-ment reduces A𝛽 levels and amyloid deposition in transgenicmice,” Cell, vol. 120, no. 5, pp. 701–713, 2005.

[77] L. Restivo, F. Ferrari, E. Passino et al., “Enriched environmentpromotes behavioral and morphological recovery in a mousemodel for the fragile X syndrome,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 102, no.32, pp. 11557–11562, 2005.

[78] L.Mandolesi, P. de Bartolo, F. Foti et al., “Environmental enrich-ment provides a cognitive reserve to be spent in the case of brainlesion,” Journal of Alzheimer’s Disease, vol. 15, no. 1, pp. 11–28,2008.

[79] A. M. Turner andW. T. Greenough, “Differential rearing effectson rat visual cortex synapses. I. Synaptic and neuronal densityand synapses per neuron,” Brain Research, vol. 329, no. 1-2, pp.195–203, 1985.

[80] J. Nithianantharajah, H. Levis, andM.Murphy, “Environmentalenrichment results in cortical and subcortical changes inlevels of synaptophysin and PSD-95 proteins,” Neurobiology ofLearning and Memory, vol. 81, no. 3, pp. 200–210, 2004.

[81] W. T. Greenough, F. R. Volkmar, and J. M. Juraska, “Effects ofrearing complexity on dendritic branching in frontolateral andtemporal cortex of the rat,” Experimental Neurology, vol. 41, no.2, pp. 371–378, 1973.

[82] C. J. Faherty,D.Kerley, andR. J. Smeyne, “AGolgi-Coxmorpho-logical analysis of neuronal changes induced by environmentalenrichment,”Developmental Brain Research, vol. 141, no. 1-2, pp.55–61, 2003.

[83] B. Kolb, R.Gibb, andG.Gorny, “Experience-dependent changesin dendritic arbor and spine density in neocortex vary qualita-tively with age and sex,” Neurobiology of Learning and Memory,vol. 79, no. 1, pp. 1–10, 2003.

[84] A. Sale, J. F. Maya Vetencourt, P. Medini et al., “Environmen-tal enrichment in adulthood promotes amblyopia recovery

through a reduction of intracortical inhibition,” Nature Neuro-science, vol. 10, no. 6, pp. 679–681, 2007.

[85] S. Foscarin,D. Ponchione, E. Pajaj et al., “Experience-dependentplasticity and modulation of growth regulatory molecules atcentral synapses,” PLoS ONE, vol. 6, no. 1, Article ID e16666,2011.

[86] B. H. Dobkin, “Motor rehabilitation after stroke, traumaticbrain, and spinal cord injury: common denominators withinrecent clinical trials,” Current Opinion in Neurology, vol. 22, no.6, pp. 563–569, 2009.

[87] Y. R. Yang, R. Y.Wang, and P. S. Wang, “Early and late treadmilltraining after focal brain ischemia in rats,”Neuroscience Letters,vol. 339, no. 2, pp. 91–94, 2003.

[88] Y. Ding, J. Li, Q. Lai et al., “Motor balance and coordinationtraining enhances functional outcome in rat with transientmiddle cerebral artery occlusion,” Neuroscience, vol. 123, no. 3,pp. 667–674, 2004.

[89] R. J. Nudo, B.M.Wise, F. SiFuentes, andG.W.Milliken, “Neuralsubstrates for the effects of rehabilitative training on motorrecovery after ischemic infarct,” Science, vol. 272, no. 5269, pp.1791–1794, 1996.

[90] E. Taub, G. Uswatte, and T. Elbert, “New treatments in neu-rorehabilitation founded on basic research,” Nature ReviewsNeuroscience, vol. 3, no. 3, pp. 228–236, 2002.

[91] M. W. Kim, M. S. Bang, T. R. Han et al., “Exercise increasedBDNF and trkB in the contralateral hemisphere of the ischemicrat brain,” Brain Research, vol. 1052, no. 1, pp. 16–21, 2005.

[92] A. Moraska, T. Deak, R. L. Spencer, D. Roth, and M. Fleshner,“Treadmill running produces both positive and negative phys-iological adaptations in Sprague-Dawley rats,” The AmericanJournal of Physiology, vol. 279, no. 4, pp. R1321–R1329, 2000.

[93] H. D.Muller, K.M.Hanumanthiah, K. Diederich, S. Schwab,W.Schabitz, and C. Sommer, “Brain-derived neurotrophic factorbut not forced arm use improves long-term outcome afterphotothrombotic stroke and transiently upregulates bindingdensities of excitatory glutamate receptors in the rat brain,”Stroke, vol. 39, no. 3, pp. 1012–1021, 2008.

[94] A.Colantonio, S. V.Kasl, A.M.Ostfeld, andL. F. Berkman, “Psy-chosocial predictors of stroke outcomes in an elderly popula-tion,” Journals of Gerontology, vol. 48, no. 5, pp. S261–S268, 1993.

[95] T. A. Glass, D. B.Matchar,M. Belyea, and J. R. Feussner, “Impactof social support on outcome in first stroke,” Stroke, vol. 24, no.1, pp. 64–70, 1993.

[96] B. B. Johansson and A. L. Ohlsson, “Environment, social inter-action, and physical activity as determinants of functional out-come after cerebral infarction in the rat,” Experimental Neurol-ogy, vol. 139, no. 2, pp. 322–327, 1996.

[97] T. K. S. Craft, E. R.Glasper, L.McCullough et al., “Social interac-tion improves experimental stroke outcome,” Stroke, vol. 36, no.9, pp. 2006–2011, 2005.

[98] S. Pietropaolo, J. Feldon, and B. K. Yee, “Nonphysical contactbetween cagemates alleviates the social isolation syndrome inC57BL/6 male mice,” Behavioral Neuroscience, vol. 122, no. 3,pp. 505–515, 2008.

[99] K. Karelina, G. J. Norman, N. Zhang, and A. C. DeVries,“Social contact influences histological and behavioral outcomesfollowing cerebral ischemia,” Experimental Neurology, vol. 220,no. 2, pp. 276–282, 2009.

[100] R. L. Gibb, C. L. R. Gonzalez, W. Wegenast, and B. E. Kolb,“Tactile stimulation promotes motor recovery following corti-cal injury in adult rats,” Behavioural Brain Research, vol. 214, no.1, pp. 102–107, 2010.


[101] C. C. Lay, M. F. Davis, C. H. Chen-Bee, and R. D. Frostig, “Mildsensory stimulation completely protects the adult rodent cortexfrom ischemic stroke,”PLoSONE, vol. 5, no. 6, Article ID e11270,2010.

[102] C. C. Lay, M. F. Davis, C. H. Chen-Bee, and R. D. Frostig, “Mildsensory stimulation reestablishes cortical function during theacute phase of ischemia,” Journal of Neuroscience, vol. 31, no. 32,pp. 11495–11504, 2011.

[103] K. J. Yoon, B. M. Oh, and D. Y. Kim, “Functional improvementand neuroplastic effects of anodal transcranial direct currentstimulation (tDCS) delivered 1 day versus 1 week after cerebralischemia in rats,” Brain Research, vol. 1452, pp. 61–72, 2012.

[104] T. Jiang, R. X. Xu, A. W. Zhang et al., “Effects of transcranialdirect current stimulation on hemichannel pannexin-1 and neu-ral plasticity in rat model of cerebral infarction,” Neuroscience,vol. 226, pp. 421–426, 2012.

[105] C. Rossi, F. Sallustio, S. Di Legge, P. Stanzione, and G. Koch,“Transcranial direct current stimulation of the affected hemi-sphere does not accelerate recovery of acute stroke patients,”European Journal of Neurology, vol. 20, no. 1, pp. 202–204, 2013.

[106] M. A. Nitsche, A. Schauenburg, N. Lang et al., “Facilitationof implicit motor learning by weak transcranial direct currentstimulation of the primary motor cortex in the human,” Journalof Cognitive Neuroscience, vol. 15, no. 4, pp. 619–626, 2003.

[107] M. A. Nitsche, S. Doemkes, T. Karakose et al., “Shaping theeffects of transcranial direct current stimulation of the humanmotor cortex,” Journal of Neurophysiology, vol. 97, no. 4, pp.3109–3117, 2007.

[108] N. Bolognini, A. Pascual-Leone, and F. Fregni, “Using non-invasive brain stimulation to augment motor training-inducedplasticity,” Journal of NeuroEngineering and Rehabilitation, vol.6, no. 1, article 8, 2009.

[109] S. K. Esser, R. Huber, M.Massimini, M. J. Peterson, F. Ferrarelli,and G. Tononi, “A direct demonstration of cortical LTP inhumans: a combined TMS/EEG study,” Brain Research Bulletin,vol. 69, no. 1, pp. 86–94, 2006.

[110] A. Quartarone, V. Rizzo, S. Bagnato et al., “Rapid-rate pairedassociative stimulation of the median nerve and motor cortexcan produce long-lasting changes in motor cortical excitabilityin humans,” Journal of Physiology, vol. 575, no. 2, pp. 657–670,2006.

[111] V. Di Lazzaro, P. Profice, F. Pilato, M. Dileone, A. Oliviero, andU. Ziemann, “The effects ofmotor cortex rTMSon corticospinaldescending activity,”Clinical Neurophysiology, vol. 121, no. 4, pp.464–473, 2010.

[112] A. Conte, D. Belvisi, E. Iezzi, F. Mari, M. Inghilleri, and A. Be-rardelli, “Effects of attention on inhibitory and facilitatory phe-nomena elicited by paired-pulse transcranial magnetic stimula-tion in healthy subjects,” Experimental Brain Research, vol. 186,no. 3, pp. 393–399, 2008.

[113] H. R. Siebner, M. Peller, F. Willoch et al., “Lasting corticalactivation after repetitive TMS of the motor cortex: a glucosemetabolic study,” Neurology, vol. 54, no. 4, pp. 956–963, 2000.

[114] F. Wang, X. Geng, H. Y. Tao, and Y. Cheng, “The restorationafter repetitive transcranial magnetic stimulation treatment oncognitive ability of vascular dementia rats and its impactson synaptic plasticity in hippocampal ca1 area,” Journal ofMolecular Neuroscience, vol. 41, no. 1, pp. 145–155, 2010.

[115] H.Wang, D. Crupi, J. Liu et al., “Repetitive transcranialmagnet-ic stimulation enhances BDNF-TrkB signaling in both brain andlymphocyte,” Journal of Neuroscience, vol. 31, no. 30, pp. 11044–11054, 2011.

[116] M. Fujiki and O. Steward, “High frequency transcranial mag-netic stimulation mimics the effects of ECS in upregulatingastroglial gene expression in the murine CNS,”Molecular BrainResearch, vol. 44, no. 2, pp. 301–308, 1997.

[117] X. Zhang, Y. Mei, C. Liu, and S. Yu, “Effect of transcranialmagnetic stimulation on the expression of c-Fos and brain-derived neurotrophic factor of the cerebral cortex in rats withcerebral infarct,” Journal of Huazhong University of Science andTechnology—Medical Science, vol. 27, no. 4, pp. 415–418, 2007.

[118] J. Biernaskie, G. Chernenko, and D. Corbett, “Efficacy of reha-bilitative experience declines with time after focal ischemicbrain injury,” Journal of Neuroscience, vol. 24, no. 5, pp. 1245–1254, 2004.

[119] S. Barbay, E. J. Plautz, K. M. Friel et al., “Behavioral and neuro-physiological effects of delayed training following a smallischemic infarct in primary motor cortex of squirrel monkeys,”Experimental Brain Research, vol. 169, no. 1, pp. 106–116, 2006.

[120] A. Risedal, J. Zeng, and B. B. Johansson, “Early training mayexacerbate brain damage after focal brain ischemia in the rat,”Journal of Cerebral Blood Flow andMetabolism, vol. 19, no. 9, pp.997–1003, 1999.

[121] S. T. Bland, T. Schallert, R. Strong, J. Aronowski, and J. C.Grotta,“Early exclusive use of the affected forelimb aftermoderate tran-sient focal ischemia in rats: functional and anatomic outcome,”Stroke, vol. 31, no. 5, pp. 1144–1152, 2000.

[122] J. L. Humm, D. A. Kozlowski, S. T. Bland, D. C. James, and T.Schallert, “Use-dependent exaggeration of brain injury: is gluta-mate involved?”ExperimentalNeurology, vol. 157, no. 2, pp. 349–358, 1999.

[123] T. A. Jones, C. J. Chu, L. A. Grande, and A. D. Gregory, “Motorskills training enhances lesion-induced structural plasticity inthe motor cortex of adult rats,” Journal of Neuroscience, vol. 19,no. 22, pp. 10153–10163, 1999.

[124] R. Domann, G.Hagemann,M. Kraemer, H. J. Freund, andO.W.Witte, “Electrophysiological changes in the surrounding braintissue of photochemically induced cortical infarcts in the rat,”Neuroscience Letters, vol. 155, no. 1, pp. 69–72, 1993.

[125] S. T. Carmichael, “Brain excitability in stroke: the yin and yangof stroke progression,” Archives of Neurology, vol. 69, no. 2, pp.161–167, 2012.

[126] R. Farrell, S. Evans, andD.Corbett, “Environmental enrichmentenhances recovery of function but exacerbates ischemic celldeath,” Neuroscience, vol. 107, no. 4, pp. 585–592, 2001.

[127] R. G. Lee and P. van Donkelaar, “Mechanisms underlying func-tional recovery following stroke,”Canadian Journal of Neurolog-ical Sciences, vol. 22, no. 4, pp. 257–263, 1995.

[128] R. Chen, L. G. Cohen, andM.Hallett, “Nervous system reorgan-ization following injury,” Neuroscience, vol. 111, no. 4, pp. 761–773, 2002.

[129] R. J. Seitz and H. J. Freund, “Plasticity of the human motorcortex,” Advances in Neurology, vol. 73, pp. 321–333, 1997.

[130] P. Cicinelli, R. Traversa, and P. M. Rossini, “Post-stroke reorga-nization of brainmotor output to the hand: a 2–4month follow-up with focal magnetic transcranial stimulation,” Electroen-cephalography andClinical Neurophysiology—Electromyographyand Motor Control, vol. 105, no. 6, pp. 438–450, 1997.

[131] R. Traversa, P. Cicinelli, A. Bassi, P.M. Rossini, andG. Bernardi,“Mapping of motor cortical reorganization after stroke: a brainsimulation study with focal magnetic pulses,” Stroke, vol. 28, no.1, pp. 110–117, 1997.


[132] J. Liepert,W. H. R.Miltner, H. Bauder et al., “Motor cortex plas-ticity during constraint,induced movement therapy in strokepatients,” Neuroscience Letters, vol. 250, no. 1, pp. 5–8, 1998.

[133] R. S. Marshall, G. M. Perera, R. M. Lazar, J. W. Krakauer, R. C.Constantine, and R. L. DeLaPaz, “Evolution of cortical activa-tion during recovery fromcorticospinal tract infarction,” Stroke,vol. 31, no. 3, pp. 656–661, 2000.

[134] N. S. Ward, J. M. Newton, O. B. C. Swayne et al., “Motor systemactivation after subcortical stroke depends on corticospinalsystem integrity,” Brain, vol. 129, part 3, pp. 809–819, 2006.

[135] N. S. Ward, J. M. Newton, O. B. C. Swayne et al., “The relation-ship between brain activity and peak grip force is modulatedby corticospinal system integrity after subcortical stroke,” Euro-pean Journal of Neuroscience, vol. 25, no. 6, pp. 1865–1873, 2007.

[136] H. Adesnik and M. Scanziani, “Lateral competition for corticalspace by layer-specific horizontal circuits,” Nature, vol. 464, no.7292, pp. 1155–1160, 2010.

[137] S. T. Carmichael, L. Wei, C. M. Rovainen, and T. A. Woolsey,“New patterns of intracortical projections after focal corticalstroke,” Neurobiology of Disease, vol. 8, no. 5, pp. 910–922, 2001.

[138] J. H. Kaas, H. X. Qi, M. J. Burish, O. A. Gharbawie, S. M. Onifer,and J. M. Massey, “Cortical and subcortical plasticity in thebrains of humans, primates, and rats after damage to sensoryafferents in the dorsal columns of the spinal cord,” ExperimentalNeurology, vol. 209, no. 2, pp. 407–416, 2008.

[139] T. A. Jones and T. Schallert, “Overgrowth and pruning of den-drites in adult rats recovering from neocortical damage,” BrainResearch, vol. 581, no. 1, pp. 156–160, 1992.

[140] R. P. Stroemer, T. A. Kent, and C. E. Hulsebosch, “Neocorticalneural sprouting, synaptogenesis, and behavioral recovery afterneocortical infarction in rats,” Stroke, vol. 26, no. 11, pp. 2135–2144, 1995.

[141] N. Dancause, “Vicarious function of remote cortex followingstroke: recent evidence from human and animal studies,”Neuroscientist, vol. 12, no. 6, pp. 489–499, 2006.

[142] M. L. Starkey, C. Bleul, B. Zorner et al., “Back seat driving:hindlimb corticospinal neurons assume forelimb control fol-lowing ischaemic stroke,” Brain, vol. 135, no. 11, pp. 3265–3281,2012.

[143] M. Komitova, B. B. Johansson, and P. S. Eriksson, “On neu-ral plasticity, new neurons and the postischemic milieu: anintegrated view on experimental rehabilitation,” ExperimentalNeurology, vol. 199, no. 1, pp. 42–55, 2006.

[144] B. B. Johansson and P. V. Belichenko, “Neuronal plasticity anddendritic spines: effect of environmental enrichment on intactand postischemic rat brain,” Journal of Cerebral Blood Flow andMetabolism, vol. 22, no. 1, pp. 89–96, 2002.

[145] T. L. Briones, M. Rogozinska, and J. Woods, “Modulationof ischemia-induced NMDAR1 activation by environmentalenrichment decreases oxidative damage,” Journal of Neuro-trauma, vol. 28, no. 12, pp. 2485–2492, 2011.

[146] O. L. Gobbo and S. M. O’Mara, “Impact of enriched-envi-ronment housing on brain-derived neurotrophic factor andon cognitive performance after a transient global ischemia,”Behavioural Brain Research, vol. 152, no. 2, pp. 231–241, 2004.

[147] M. Ploughman, S. Granter-Button, G. Chernenko et al., “Exer-cise intensity influences the temporal profile of growth factorsinvolved in neuronal plasticity following focal ischemia,” BrainResearch, vol. 1150, no. 1, pp. 207–216, 2007.

[148] S. Li, J. J. Overman, D. Katsman et al., “An age-related sproutingtranscriptome provides molecular control of axonal sprouting

after stroke,”Nature Neuroscience, vol. 13, no. 12, pp. 1496–1506,2010.

[149] K. Keyvani, N. Sachser, O. W. Witte, and W. Paulus, “Geneexpression profiling in the intact and injured brain followingenvironmental enrichment,” Journal of Neuropathology andExperimental Neurology, vol. 63, no. 6, pp. 598–609, 2004.

[150] J. Milbrandt, “A nerve growth factor-induced gene encodes apossible transcriptional regulatory factor,” Science, vol. 238, no.4828, pp. 797–799, 1987.

[151] J. L. W. Yau, T. Olsson, R. G. M. Morris, J. Noble, and J. R.Seckl, “DecreasedNGFI-A gene expression in the hippocampusof cognitively impaired aged rats,”Molecular BrainResearch, vol.42, no. 2, pp. 354–357, 1996.

[152] M. Dragunow, “A role for immediate-early transcription factorsin learning and memory,” Behavior Genetics, vol. 26, no. 3, pp.293–299, 1996.

[153] G. Thiel, S. Schoch, and D. Petersohn, “Regulation of synapsinI gene expression by the zinc finger transcription factorzif268/egr-1,” The Journal of Biological Chemistry, vol. 269, no.21, pp. 15294–15301, 1994.

[154] D. Petersohn, S. Schoch, D. R. Brinkmann, and G. Thiel, “Thehuman synapsin II gene promoter. Possible role for the tran-scription factors zif268/egr-1, polyoma enhancer activator 3,and AP2,” The Journal of Biological Chemistry, vol. 270, no. 41,pp. 24361–24369, 1995.

[155] A. Ferreira and M. Rapoport, “The synapsins: beyond theregulation of neurotransmitter release,” Cellular and MolecularLife Sciences, vol. 59, no. 4, pp. 589–595, 2002.

[156] A. S. Pagnussat, F. Simao, J. R. Anastacio et al., “Effects ofskilled and unskilled training on functional recovery and brainplasticity after focal ischemia in adult rats,” Brain Research, vol.1486, pp. 53–61, 2012.

[157] S. F. Yan, T. Fujita, J. Lu et al., “Egr-1, a master switch coordinat-ing upregulation of divergent gene families underlying ischemicstress,” Nature Medicine, vol. 6, no. 12, pp. 1355–1361, 2000.

[158] J. Milbrandt, “Nerve growth factor induces a gene homologousto the glucocorticoid receptor gene,” Neuron, vol. 1, no. 3, pp.183–188, 1988.

[159] M. Dragunow, W. Abraham, and P. Hughes, “Activation ofNMDA and muscarinic receptors induces nur-77 mRNA inhippocampal neurons,”Molecular Brain Research, vol. 36, no. 2,pp. 349–356, 1996.

[160] P. Dahlqvist, A. Ronnback, A. Risedal et al., “Effects of postis-chemic environment on transcription factor and serotoninreceptor expression after permanent focal cortical ischemia inrats,” Neuroscience, vol. 119, no. 3, pp. 643–652, 2003.

[161] P. Dahlqvist, L. Zhao, I. M. Johansson et al., “Environmentalenrichment alters nerve growth factor-induced gene A and glu-cocorticoid receptor messenger RNA expression after middlecerebral artery occlusion in rats,” Neuroscience, vol. 93, no. 2,pp. 527–535, 1999.

[162] J. Nygren and T. Wieloch, “Enriched environment enhancesrecovery of motor function after focal ischemia in mice, anddownregulates the transcription factor NGFI-A,” Journal ofCerebral Blood Flow and Metabolism, vol. 25, no. 12, pp. 1625–1633, 2005.

[163] Y. Shono, C. Yokota, Y. Kuge et al., “Gene expression associatedwith an enriched environment after transient focal ischemia,”Brain Research, vol. 1376, pp. 60–65, 2011.

[164] M. Cayre, P. Canoll, and J. E. Goldman, “Cell migration in thenormal and pathological postnatal mammalian brain,” Progressin Neurobiology, vol. 88, no. 1, pp. 41–63, 2009.


[165] S. Keiner, J.Walter, J. Oberland, and C. Redecker, “Contributionof constitutively proliferating precursor cell subtypes to dentateneurogenesis after cortical infarcts,” BMC Neuroscience, vol. 11,article 146, 2010.

[166] J. Walter, S. Keiner, O. W. Witte, and C. Redecker, “Differentialstroke-induced proliferative response of distinct precursor cellsubpopulations in the young and aged dentate gyrus,” Neuro-science, vol. 169, no. 3, pp. 1279–1286, 2010.

[167] A. Arvidsson, T. Collin, D. Kirik, Z. Kokaia, and O. Lindvall,“Neuronal replacement from endogenous precursors in theadult brain after stroke,” Nature Medicine, vol. 8, no. 9, pp. 963–970, 2002.

[168] J. M. Parent, Z. S. Vexler, C. Gong, N. Derugin, and D. M. Fer-riero, “Rat forebrain neurogenesis and striatal neuron replace-ment after focal stroke,” Annals of Neurology, vol. 52, no. 6, pp.802–813, 2002.

[169] R. Zhang, Z. Zhang, C. Zhang et al., “Stroke transientlyincreases subventricular zone cell division from asymmetric tosymmetric and increases neuronal differentiation in the adultrat,” Journal of Neuroscience, vol. 24, no. 25, pp. 5810–5815, 2004.

[170] J. E. Gotts and M. Chesselet, “Mechanisms of subventricularzone expansion after focal cortical ischemic injury,” Journal ofComparative Neurology, vol. 488, no. 2, pp. 201–214, 2005.

[171] M. Komitova, B. Mattsson, B. B. Johansson, and P. S. Eriksson,“Enriched environment increases neural stem/progenitor cellproliferation and neurogenesis in the subventricular zone ofstroke-lesioned adult rats,” Stroke, vol. 36, no. 6, pp. 1278–1282,2005.

[172] D. Koketsu, Y. Furuichi, M. Maeda, N. Matsuoka, Y. Miyamoto,and T. Hisatsune, “Increased number of new neurons in theolfactory bulb and hippocampus of adult non-human primatesafter focal ischemia,” Experimental Neurology, vol. 199, no. 1, pp.92–102, 2006.

[173] T. Collin, A. Arvidsson, Z. Kokaia, and O. Lindvall, “Quanti-tative analysis of the generation of different striatal neuronalsubtypes in the adult brain following excitotoxic injury,” Exper-imental Neurology, vol. 195, no. 1, pp. 71–80, 2005.

[174] K. Jin, X.Wang, L. Xie, X. O.Mao, andD. A. Greenberg, “Trans-genic ablation of doublecortin-expressing cells suppresses adultneurogenesis and worsens stroke outcome inmice,” Proceedingsof the National Academy of Sciences of the United States ofAmerica, vol. 107, no. 17, pp. 7993–7998, 2010.

[175] Y. Sun, K. Jin, L. Xie et al., “VEGF-induced neuroprotection,neurogenesis, and angiogenesis after focal cerebral ischemia,”Journal of Clinical Investigation, vol. 111, no. 12, pp. 1843–1851,2003.

[176] M. Komitova, L. R. Zhao, G. Gido, B. B. Johansson, and P.Eriksson, “Postischemic exercise attenuates whereas enrichedenvironment has certain enhancing effects on lesion-inducedsubventricular zone activation in the adult rat,” EuropeanJournal of Neuroscience, vol. 21, no. 9, pp. 2397–2405, 2005.

[177] T. L. Briones, E. Suh, H. Hattar, and M. Wadowska, “Dentategyrus neurogenesis after cerebral ischemia and behavioraltraining,” Biological Research for Nursing, vol. 6, no. 3, pp. 167–179, 2005.

[178] J. L. Trejo, E. Carro, and I. Torres-Aleman, “Circulating insulin-like growth factor I mediates exercise-induced increases in thenumber of new neurons in the adult hippocampus,” Journal ofNeuroscience, vol. 21, no. 5, pp. 1628–1634, 2001.

[179] K. Fabel, K. Fabel, B. Tam et al., “VEGF is necessary for exercise-induced adult hippocampal neurogenesis,” European Journal ofNeuroscience, vol. 18, no. 10, pp. 2803–2812, 2003.

[180] L. Cao, X. Jiao, D. S. Zuzga et al., “VEGF links hippocampalactivity with neurogenesis, learning and memory,” NatureGenetics, vol. 36, no. 8, pp. 827–835, 2004.

[181] F. Wurm, S. Keiner, A. Kunze, O. W. Witte, and C. Redecker,“Effects of skilled forelimb training on hippocampal neurogen-esis and spatial learning after focal cortical infarcts in the adultrat brain,” Stroke, vol. 38, no. 10, pp. 2833–2840, 2007.

[182] C. Zhao, J. Wang, S. Zhao, and Y. Nie, “Constraint-inducedmovement therapy enhanced neurogenesis and behavioralrecovery after stroke in adult rats,” Tohoku Journal of Experi-mental Medicine, vol. 218, no. 4, pp. 301–308, 2009.

[183] Y. Matsumori, S. M. Hong, Y. Fan et al., “Enriched environmentand spatial learning enhance hippocampal neurogenesis andsalvages ischemic penumbra after focal cerebral ischemia,”Neurobiology of Disease, vol. 22, no. 1, pp. 187–198, 2006.

[184] M. Nedergaard and U. Dirnagl, “Role of glial cells in cerebralischemia,” Glia, vol. 50, no. 4, pp. 281–286, 2005.

[185] M. Nilsson and M. Pekny, “Enriched environment and astro-cytes in central nervous system regeneration,” Journal of Reha-bilitation Medicine, vol. 39, no. 5, pp. 345–352, 2007.

[186] J. L. Ridet, S. K. Malhotra, A. Privat, and F. H. Gage, “Reactiveastrocytes: cellular and molecular cues to biological function,”Trends in Neurosciences, vol. 20, no. 12, pp. 570–577, 1997.

[187] E. M. Ullian, K. S. Christopherson, and B. A. Barres, “Role forglia in synaptogenesis,” Glia, vol. 47, no. 3, pp. 209–216, 2004.

[188] M. Pekny, U. Wilhelmsson, Y. R. Bogestal, and M. Pekna, “Therole of astrocytes and complement system in neural plasticity,”International Review of Neurobiology, vol. 82, pp. 95–111, 2007.

[189] H. Song, C. F. Stevens, and F. H. Gage, “Astroglia induce neuro-genesis from adult neural stem cells,” Nature, vol. 417, no. 6884,pp. 39–44, 2002.

[190] A.Alvarez-Buylla andD.A. Lim, “For the long run:maintaininggerminal niches in the adult brain,” Neuron, vol. 41, no. 5, pp.683–686, 2004.

[191] S. Robel, B. Berninger, and M. Gotz, “The stem cell potential ofglia: lessons from reactive gliosis,”Nature Reviews Neuroscience,vol. 12, no. 2, pp. 88–104, 2011.

[192] H. K. Kimelberg, “Astrocytic swelling in cerebral ischemia as apossible cause of injury and target for therapy,”Glia, vol. 50, no.4, pp. 389–397, 2005.

[193] S. Keiner, F. Wurm, A. Kunze, O. W. Witte, and C. Redecker,“Rehabilitative therapies differentially alter proliferation andsurvival of glial cell populations in the perilesional zone ofcortical infarcts,” Glia, vol. 56, no. 5, pp. 516–527, 2008.

[194] R. P. Kraig and C. B. Jaeger, “Ionic concomitants of astroglialtransformation to reactive species,” Stroke, vol. 21, no. 11,supplement, pp. III184–III187, 1990.

[195] J. X. Wilson, “Antioxidant defense of the brain: a role forastrocytes,” Canadian Journal of Physiology and Pharmacology,vol. 75, no. 10-11, pp. 1149–1163, 1997.

[196] K. Ruscher, D. Freyer, M. Karsch et al., “Erythropoietin is aparacrine mediator of ischemic tolerance in the brain: evidencefrom an in vitro model,” Journal of Neuroscience, vol. 22, no. 23,pp. 10291–10301, 2002.

[197] G. Trendelenburg, K. Prass, J. Priller et al., “Serial analysis ofgene expression identifies metallothionein-II as major neuro-protective gene in mouse focal cerebral ischemia,” Journal ofNeuroscience, vol. 22, no. 14, pp. 5879–5888, 2002.

[198] M. F. Anderson, F. Blomstrand, C. Blomstrand, P. S. Eriksson,and M. Nilsson, “Astrocytes and stroke: networking for sur-vival?”Neurochemical Research, vol. 28, no. 2, pp. 293–305, 2003.


[199] G. Trendelenburg and U. Dirnagl, “Neuroprotective role ofastrocytes in cerebral ischemia: focus on ischemic precondi-tioning,” Glia, vol. 50, no. 4, pp. 307–320, 2005.

[200] T. G. Bush, N. Puvanachandra, C. H. Horner et al., “Leukocyteinfiltration, neuronal degeneration, and neurite outgrowth afterablation of scar-forming, reactive astrocytes in adult transgenicmice,” Neuron, vol. 23, no. 2, pp. 297–308, 1999.

[201] M. Pekny, C. B. Johansson, C. Eliasson et al., “Abnormal reactionto central nervous system injury in mice lacking glial fibrillaryacidic protein and vimentin,” Journal of Cell Biology, vol. 145, no.3, pp. 503–514, 1999.

[202] J. R. Faulkner, J. E. Herrmann, M. J. Woo, K. E. Tansey, N.B. Doan, and M. V. Sofroniew, “Reactive astrocytes protecttissue and preserve function after spinal cord injury,” Journal ofNeuroscience, vol. 24, no. 9, pp. 2143–2155, 2004.

[203] C. M. Liberto, P. J. Albrecht, L. M. Herx, V. W. Yong, and S.W. Levison, “Pro-regenerative properties of cytokine-activatedastrocytes,” Journal of Neurochemistry, vol. 89, no. 5, pp. 1092–1100, 2004.

[204] T. Morioka, A. N. Kalehua, and W. J. Streit, “Characterizationof microglial reaction after middle cerebral artery occlusion inrat brain,” Journal of Comparative Neurology, vol. 327, no. 1, pp.123–132, 1993.

[205] W. J. Streit, S. A.Walter, and N. A. Pennell, “Reactive microglio-sis,” Progress in Neurobiology, vol. 57, no. 6, pp. 563–581, 1999.

[206] W. J. Streit, “Microglial response to brain injury: a briefsynopsis,” Toxicologic Pathology, vol. 28, no. 1, pp. 28–30, 2000.

[207] I. Bechmann, J. Priller, A. Kovac et al., “Immune surveillance ofmouse brain perivascular spaces by blood-bornemacrophages,”European Journal of Neuroscience, vol. 14, no. 10, pp. 1651–1658,2001.

[208] S. Jander, M. Schroeter, and A. Saleh, “Imaging inflammationin acute brain ischemia,” Stroke, vol. 38, no. 2, supplement, pp.642–645, 2007.

[209] S. Elkabes, E. M. DiCicco-Bloom, and I. B. Black, “Brainmicroglia/macrophages express neurotrophins that selectivelyregulate microglial proliferation and function,” Journal of Neu-roscience, vol. 16, no. 8, pp. 2508–2521, 1996.

[210] F. Imai, H. Suzuki, J. Oda et al., “Neuroprotective effect of exoge-nous microglia in global brain ischemia,” Journal of CerebralBlood Flow and Metabolism, vol. 27, no. 3, pp. 488–500, 2007.

[211] K. M. Boje and P. K. Arora, “Microglial-produced nitric oxideand reactive nitrogen oxidesmediate neuronal cell death,”BrainResearch, vol. 587, no. 2, pp. 250–256, 1992.

[212] J. von Zahn, T. Moller, H. Kettenmann, and C. Nolte, “Microgl-ial phagocytosis is modulated by pro-and anti-inflammatorycytokines,” NeuroReport, vol. 8, no. 18, pp. 3851–3856, 1997.

[213] M. Aschner, J. W. Allen, H. K. Kimelberg, R. M. LoPachin, andW. J. Streit, “Glial cells in neurotoxicity development,” AnnualReview of Pharmacology and Toxicology, vol. 39, pp. 151–173,1999.

[214] C. L. Gibson, T. C. Coughlan, and S. P. Murphy, “Glial nitricoxide and ischemia,” Glia, vol. 50, no. 4, pp. 417–426, 2005.

[215] S. W. Barger, M. E. Goodwin, M. M. Porter, and M. L. Beggs,“Glutamate release from activatedmicroglia requires the oxida-tive burst and lipid peroxidation,” Journal of Neurochemistry,vol. 101, no. 5, pp. 1205–1213, 2007.

[216] S. T. Dheen, C. Kaur, and E. Ling, “Microglial activation and itsimplications in the brain diseases,” Current Medicinal Chemis-try, vol. 14, no. 11, pp. 1189–1197, 2007.

[217] K. A. Hewlett and D. Corbett, “Delayed minocycline treatmentreduces long-term functional deficits and histological injury ina rodent model of focal ischemia,” Neuroscience, vol. 141, no. 1,pp. 27–33, 2006.

[218] Z. Liu, Y. Fan, S. J. Won et al., “Chronic treatment with min-ocycline preserves adult new neurons and reduces functionalimpairment after focal cerebral ischemia,” Stroke, vol. 38, no. 1,pp. 146–152, 2007.

[219] M. Komitova, E. Perfilieva, B. Mattsson, P. S. Eriksson, and B. B.Johansson, “Enriched environment after focal cortical ischemiaenhances the generation of astroglia andNG2 positive polyden-drocytes in adult rat neocortex,” Experimental Neurology, vol.199, no. 1, pp. 113–121, 2006.

[220] A. Nishiyama,M.Watanabe, Z. Yang, and J. Bu, “Identity, distri-bution, and development of polydendrocytes: NG2-expressingglial cells,” Brain Cell Biology, vol. 31, no. 6-7, pp. 437–455, 2002.

[221] S. Belachew, R. Chittajallu, A. A. Aguirre et al., “PostnatalNG2 proteoglycan-expressing progenitor cells are intrinsicallymultipotent and generate functional neurons,” Journal of CellBiology, vol. 161, no. 1, pp. 169–186, 2003.

[222] A. Aguirre and V. Gallo, “Postnatal neurogenesis and gliogene-sis in the olfactory bulb fromNG2-expressing progenitors of thesubventricular zone,” Journal of Neuroscience, vol. 24, no. 46, pp.10530–10541, 2004.

[223] E. Boda and A. Buffo, “Glial cells in non-germinal territories:insights into their stem/progenitor properties in the intact andinjured nervous tissue,” Archives Italiennes de Biologie, vol. 148,no. 2, pp. 119–136, 2010.

[224] A. J. Butler and S. J. Page, “Mental practice with motor imagery:evidence for motor recovery and cortical reorganization afterstroke,”Archives of Physical Medicine and Rehabilitation, vol. 87,no. 12, pp. 2–11, 2006.


Research ArticleCase Study of Ecstatic Meditation: fMRI and EEG Evidence ofSelf-Stimulating a Reward System

Michael R. Hagerty,1 Julian Isaacs,2 Leigh Brasington,3 Larry Shupe,4

Eberhard E. Fetz,5 and Steven C. Cramer6

1 University of California, Davis and Wellspring Institute, Davis, CA 95616, USA2Wellspring Institute, San Rafael, CA 94903, USA3 Barre Center for Buddhist Studies, Barre, MA 01005, USA4University of Washington, Seattle, WA 98195, USA5 Physiology & Biophysics, University of Washington, Seattle, WA 98195, USA6Department of Neurology and Anatomy & Neurobiology, University of California, Irvine, CA 92697, USA

Correspondence should be addressed to Michael R. Hagerty; [email protected]

Received 26 February 2013; Accepted 3 April 2013


Copyright © 2013 Michael R. Hagerty et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

We report the first neural recording during ecstatic meditations called jhanas and test whether a brain reward system plays a rolein the joy reported. Jhanas are Altered States of Consciousness (ASC) that imply major brain changes based on subjective reports:(1) external awareness dims, (2) internal verbalizations fade, (3) the sense of personal boundaries is altered, (4) attention is highlyfocused on the object of meditation, and (5) joy increases to high levels. The fMRI and EEG results from an experienced meditatorshow changes in brain activity in 11 regions shown to be associated with the subjective reports, and these changes occur promptlyafter jhana is entered. In particular, the extreme joy is associated not onlywith activation of cortical processes but alsowith activationof the nucleus accumbens (NAc) in the dopamine/opioid reward system. We test three mechanisms by which the subject mightstimulate his own reward system by external means and reject all three. Taken together, these results demonstrate an apparentlynovel method of self-stimulating a brain reward system using only internal mental processes in a highly trained subject.

1. Introduction

Ecstatic experiences have been reported in every majorreligion, and psychologists have long advocated research inthese areas [1, 2]. Neuroscience can contribute to these issuesby documenting the brain activity of expert meditators, someof whom have trained to enter these states with volitionalcontrol. The type of meditation studied here is a Buddhistconcentration technique called jhana that induces an AlteredState of Consciousness (ASC) in the framework of Vaitl et al.[3] and whose short-term goal is joy or happiness. Becausehappiness is a fundamental goal of many people and is theobject of the new discipline of positive psychology [4, 5],imaging the brain of an individual who claims to generatejoy without any external rewards or cues could point the waytoward improved training in joy and greater resilience in the

face of external difficulties. Of particular interest is the neuralmechanisms by which happiness is generated.

Jhanameditations consist of a set of 8 sequential practicesthat were first codified by Buddhists over 2000 years ago [6].All are reported to be ecstatic, in that they generate greatjoy while in an ASC that is dissociated from external cuesor stimuli. The first three practices are, to our knowledge,the only meditations to specifically target short-term joyor happiness (see [7, 8] for other meditations that generateASCs). Figure 1 shows a schematic of the reported jhanaexperiences on 2 dimensions of interest. Joy or happinessis shown on the 𝑥-axis, and vigilance for external stimuliis plotted on the 𝑦-axis. Meditators progress in sequencefrom normal resting consciousness (rest) to AC, a prepara-tory meditation concentrating on the breath. When internalconcentration is strong enough, J1 is entered, accompanied

2 Neural Plasticity

Rest

Access concentration(AC)

J3: contentment and happiness

J2: joy and bliss

J1: physical pleasure

J5: “infinite space”

J4: equanimity and peace

J8: neither perception nor nonperception

J7: “nothingness”

J6: “infinite consciousness”

Vigilance for

external stimuli

Joy or happiness

Figure 1: Schematic of the reported experiences in 8 jhanas relative to resting consciousness and access concentration (AC) on 2 dimensionsof interest. Joy or happiness is shown on the 𝑥-axis, and vigilance for external stimuli is plotted on the 𝑦-axis.The typical meditation sequenceis rest to AC to J1, J2, and J3 (the three jhanas highest in joy or happiness), then to J4–J8, all of which are said to be higher in happiness thanrest or AC. Each jhana is reported to be deeper and more remote from external stimuli than the last.

by strong physical pleasure—“better than sexual orgasm”([9] p.151)—and greatly reduced vigilance with smaller startleresponses. In J2 joy “permeates every part of the body,”but with less physical pleasure. In J3, the character of joychanges to “deep contentment and serenity.” J4 is describedby “equanimity—a profound peace and stillness.”The higher-numbered jhanas J5–J8 are characterized by more subtleand profound perceptions. J5 is called “infinite space,” J6is “infinite consciousness,” J7 is “nothingness,” and J8 isnamed “neither perception nor non-perception.” Each jhanais reported to be deeper and more remote from externalstimuli than the last, yielding the ranking shown on the𝑦-axisin Figure 1. J1–J3 are the highest on joy or happiness, with J4–J8 intermediate, yielding the ranking on the 𝑥-axis. All areclassified by Lutz et al. [10, 11] as concentration rather thanopen awareness meditations.

Previous studies have shown that long-term meditatorshave higher volume of grey matter compared to matchedcontrols [12, 13], and randomized experiments show thatsubjects benefit from as little as 4 weeks of training in theareas of attention regulation [14, 15] and emotion regulation

[11, 16–18]. Heretofore, all of the emotion studies have testedsubjects’ ability to learn to downregulate negative emotions,particularly their response to stress. In contrast, the presentstudy examines the ability to up-regulate positive emotion,which involves different neural pathways [19, 20]).

Perhaps the most thoroughly studied system related topositive emotion is the dopamine system, which gives rise topleasure and mediates positive reinforcement [21, 22]. Bothanimal and human studies show that when a delivered rewardis greater than expected, dopaminergic neurons in theVentralTegmental Area (VTA) in the brain stem are activated. TheVTA in turn innervates the nucleus accumbens (NAc) in theventral striatum, which leads to higher centers in the orbitalfrontal cortex (OFC).Human studies have shown that activityin the medial OFC at the time of a reward correlates withsubjective reports of pleasure for olfactory [23], gustatory[24], and musical stimuli [25]. Studies have shown that thissystem is activated for a diverse array of stimuli, includingfood [26], sex [27],music [25], humor [28],monetary rewards[29], and maternal love [30]. But it has never been shownthat this dopamine system can be activated without external

Neural Plasticity 3

cues or rewards by volitional mental activity.Themechanismby which such a mental activity can self-stimulate positiveemotions would be of great interest. One hypothesis is thatthe full dopamine pathway is stimulated beginning withthe VTA and progressing upward. An alternate hypothesisis that the subjective report of pleasure is caused only byexpectancy effects (such as a belief that a high-priced winemust taste better; see [31] or [32]) and that the lower partsof the dopamine system do not participate. Yet a thirdalternativemechanism is that the subjective pleasure is due tosubtle rhythmic bodymovements which are known to inducepleasurable altered states [3].

The dopamine reward system has also been shown to bestimulated bymost drugs of abuse andplays an important rolein addiction [33]. An important question is whether jhanameditators are subject to addiction and tolerance effects thatcan result from stimulation of the dopamine reward system.

Besides the dopamine system, Pecina et al. [34] documentthat the opioid system mediates pleasure in animal studies.Unfortunately, it shares a pathway very close to that of thedopamine system in the NAc. Discrimination between thetwo systems would require microinjection studies and isbeyond the spatial discrimination of typical fMRI studies.Hence, the current paper limits itself to detecting activationin the region shared by these two reward pathways.

Experientially, all jhanas in Figure 1 are reported to sharethe following 5 characteristics that may have specific braincorrelates: (1) external awareness dims and startle responsesdiminish, (2) internal verbalizations fade completely orbecome “wispy”, (3) one’s sense of body boundaries andorientation in space are altered, (4) attention is highly focusedon the object of meditation, and (5) happiness increases tovery high levels and can be maintained for long periods oftime. Jhana is distinguished from some other ASCs becauseit does not include visual or auditory hallucinations (as insome organic disorders and drug experiences) nor does itinclude cross-sense synesthesia (such as “seeing” the bell ringor “feeling” a bird sing). The correspondences expected fromknown functions of brain regions can be articulated in theform of the following a priori hypotheses.

H1: Jhanas should show decreased activation compared tothe rest state in the visual (BA 17–19) and auditory (BA 41-42) processing areas. Since all jhanas share the experientialcharacteristic that external awareness dims, then the brainregions associated with vision and hearing should becomeless active.

H2: Jhanas should show decreased activation compared tothe rest state in Broca’s area (BA 44,45) and in Wernicke’s area(BA 39,40). Because internal verbalization fades in jhana, thebrain regions associated with speech should become idle orless active.

H3: Jhanas should show decreased activation compared tothe rest state in the orientation area (BA5). Since the normalsense of personal boundaries is altered, the orientation areaof the brain should show changes from normal rest. Newberg

and Iversen [8] showed that monks and nuns experiencing“union with God” exhibit decreased activation in this area.

H4: Jhanas should show increased activation compared tothe rest state in the Anterior Cingulate Cortex (ACC) (BA32,33). Because attention is highly focused on the object ofmeditation in the jhanas, we would expect high activity in theACC, which regulates and monitors attention.

H5: Jhanas should show increased activation compared tothe rest state in the dopamine reward system of the brain (NAcin the ventral striatum and medial OFC). A broad range ofexternal rewards stimulate this system (food, sex, beautifulmusic, and monetary awards), so extreme joy in jhana maybe triggered by the same system (the VTA is also part ofthis system, but is too small to image with standard fMRImethods, but see [35] for successful imaging methods).

H6: Jhanas should show no increased activation comparedto the rest state in the areas responsible for rhythmicmovement,including motor cortex (BA4), primary somatosensory cortex(BA 1,2,3), and cerebellum. Increased activity in these areaswould support an alternative hypothesis that the rewardsystem is being stimulated not by internalmeans but by subtlerhythmic movements that are known to induce ecstatic states[3].

The activation of brain regions during these six subjectivejhana experiences can now be examined via fMRI and EEG.

2. Methods

The subject is a long-term Buddhist practitioner (53-year-oldmale, left-handed). At the time of recording, he had 17 yearsof training consisting of about 6,000 hours of practice andwas trained in the Sri Lankan tradition of jhanas by Khema[6] (the length of training was estimated based on his dailypractice and the time spent on meditative retreats, countingone day of retreat as 8 hours of sitting meditation). At thetime of testing, this subject was to our knowledge the onlyperson in the US who had the requisite training in jhana whowas willing to submit to the experimental protocol.The fMRIscanning was done several months after the EEG recording.

The subject signed informed consent, and a neurologicalexamwas performed, confirming the absence of neurologicaldisease. He had no medical conditions and was on nomedications.The subject meditated in his standard sequence,starting with access concentration (AC), progressing throughJ1, J2,. . .J8, then returning through J7, J6, and so forth, backdown to J1. For each jhana state, the subject signaled with adouble finger tap using an MR-compatible force transducer[36] when he was beginning the transition to the next higher-number jhana state, then clicked the mouse once when hehad reached the state. He clicked three times to indicate hewas transitioning downward to the next lower-number jhanastate. Resting periods were recorded before or after jhanas.

The protocol did not use a random assignment of statesbecause each jhana builds on the previous one, and the time

4 Neural Plasticity

required to transition from one state to another was variable.Hence, the standard sequence was used. This sequence hadbeen very well practiced, making state identification easy forour subject. The duration of each jhana state averaged about120 sec, with about 30 sec transition between states.

2.1. fMRI Recording and Analysis. We acquired gradientecho T2∗-weighted echo-planar images (EPIs) with blood-oxygen-level-dependent (BOLD) contrasts on a GE 1.5-Teslascanner (repetition time TR of 2.5 s and TE of 50ms). Atotal of 421 volumes were collected, with 20 axial slices pervolume and slice thickness of 7mm, going from vertex toinferior cerebellum with no skip between slices. Two T1-weighted structural images were also acquired, the first ahigh-resolution volumetric series and the second a lowerresolution scan in-plane with the functional data. Threeperiods of rest were interspersedwith 2 periods of tapping theforce transducer for control purposes, then subject enteredAC followed by J2, J3, J4, and J5. The fMRI recordingthen ended due to scanner memory limitations (421 volumemaximum). J1 was not practiced because the associated headmovements would induce excessive artifact.

Statistical parametric mapping [37] served to preprocessand analyze the data. The first four volumes were discardeddue to tissue nonsaturation, and each remaining volumewas motion corrected to the 5th volume. All images werenormalized to a standard MNI template and smoothedusing an isometric Gaussian kernel with a full width at halfmaximumof 8mm.High-pass filteringwas increased to 4096seconds because the experimental design consisted of a verylow frequency of 625 s (from rest to J5). The time signatureof the epochs was modeled as a series of boxcar functionsconvolved with a canonical hemodynamic response function(HRF).The general linear model estimated the percent signalchange of each event (jhana versus rest versus AC) as afunction of the convolved time signature. The two contrastsof interest in testing the planned hypotheses were jhana-restand jhana-AC. In addition, J2 was contrasted with each ofthe other states in order to investigate specific differencesbetween jhana levels of meditation. For each a priori ROIspecified in the hypotheses, an anatomicalmaskwas preparedfrom theWFU PickAtlas software [38] and the mean percentsignal change was calculated for each contrast usingMarsBar[39].Themasks used in this study were Brodmann’s area (BA)17 OR 19, BA 41 OR 42, BA 44 OR 45, BA 39 OR 40, BA 5OR 7, BA 32 OR 33, BA 1 OR 2 OR 3, and BA 4 (where “OR”refers to the logical addition of two masks), cerebellum, andMed OFC. Finally, the NAc was approximated with sphericalmasks of radius 5mm centered at (±10, 9, −4) using thelocation identified by Kirk et al. [40] and Knutson et al. [41].

2.2. EEGRecording andAnalysis. TheEEGsystemused a 256-channel Geodesic Sensor Net (System v.2.0 from ElectricalGeodesics, OR), sampled at 500Hz and referenced to thevertex (Cz). Sections of the recording that showed eyemovements or muscular artifacts were manually excludedfrom the study. The data was bandpassed with a digital high-pass filter at .4Hz and a hardware low-pass filter at 200Hz.

A 60Hz notch filter was employed to remove 60Hz lineartifacts. Six epochs of 4 seconds each were extracted fromeach of the 21 states (2 resting states and 19 jhana states).

For each electrode and for each 4 s epoch, the powerspectral distribution was computed by usingWelch’s method,which averages power values across sliding and overlapping500ms time windows. Spectral bands were defined to beconsistent with previous research: theta band was from 4 to6Hz, alpha1 band from 6 to 8Hz, alpha2 band from 8 to10Hz, alpha3 from 10 to 12.5Hz, beta from 12.5 to 25Hz, andgamma from 25 to 42Hz. The last is consistent with Lutzet al. [42] who analyzed only the gamma range. The first 3bands are congruent with Aftanas et al. [43] who analyzedonly those bands. However, we did not perform the analysisof alpha dominant frequency to establish frequency bandboundaries individually for our subject, as Aftanas et. al.[43]did, although our band frequencies are close to theirs.All power estimates are reported as a ratio of the power ina selected band to total power from 4 to 42Hz.

Electrode positions were matched with underlyinganatomical ROIs using the probabilistic maps developed byOkamoto et al. [44] who correlated the anatomical MRI’s of17 healthy adults with the overlying electrodes placed in thestandard 10–20 position.

3. Results

3.1. fMRI. Table 1 reports a formal assessment of the 6 apriori hypotheses. The first row of Table 1 tests H1, where thefirst column shows the subjective experience during jhana(that external awareness dims), the second column showsthe ROI associated with that experience (the primary andassociative visual cortex BA 17,19), and the third columnshows predicted change in activity during jhana comparedto rest (activity will be less during jhana). Column 4 showsthat the actual contrast is −.81, a difference that is significant(𝑡 = −4.3, 𝑃 < .001) and in the predicted direction.The last column of Table 1 uses an alternative comparisonstandard, calculating the BOLD signal contrast for Jhanarelative to access concentration (Jhana-AC). That columnconfirms that the contrast is also negative, supporting thereports in Figure 1. The next row shows that the contrastin primary auditory and association cortex (BA 41, 42) wasalso negative and significant, again supporting H1. Similarly,H2 (that internal verbalization fades) is strongly supportedby significant negative contrasts in Broca’s area (BA 44, 45)and in Wernicke’s area (BA 39, 40). H3 (an altered sense ofpersonal boundaries) is strongly supported with large andsignificant negative signal contrasts in the orientation area(BA 5, 7). H4 (that attention is highly focused) ismore weaklyconfirmed, with both BOLD signal contrasts in the ACCpositive compared to rest, though column 5 shows that thecontrast Jhana-AC failed to reach significance. H5 is stronglyconfirmed, with both the NAc and Med OFC recordingsignificantly higher BOLD signal during jhana than duringboth rest and AC meditation. The last rows of Table 1 showthe test of an alternative hypothesis (H6) that the ecstaticjoy in jhanas may be caused by subtle rhythmic movements,resulting in higher BOLD signal during jhana in the primary

Neural Plasticity 5

Table 1: Mean percent BOLD signal change in a priori defined ROIs related to 6 hypotheses on jhana activity contrasted with rest and ACmeditation, followed by its two-sided 𝑡 test (corrected for multiple comparisons). Contrasts labeled simply “jhana” refer to the pooled activityover all recorded jhanas 2–5. All 22 planned contrasts are in the predicted direction.

Subjective report duringjhanas

A priori ROI (MNI coordinatesof centroid of ROI)

Predicted sign ofcontrast (jhana-rest)

BOLD contrast(jhana-rest)

BOLD contrast(jhana-AC)

(1) “External awareness dims”Visual: BA 17, 19(±30 −80 6) (−) −.81 𝑡 = −4.3∗∗ −.73 𝑡 = −4.0∗∗

Auditory: BA 41, 42(±55 −26 12) (−) −.63 𝑡 = −2.5∗ −.25 𝑡 = −1.0

(2) “Internal verbalizationfades”

Broca: BA 44, 45(±54 18 12) (−) −.84 𝑡 = −4.6∗∗ −.85 𝑡 = −4.8∗∗

Wernicke: BA 39, 40(±51 −51 34) (−) −.76 𝑡 = −3.7∗∗ −.70 𝑡 = −3.5∗∗

(3) “Altered sense of personalboundaries”

Orientation: BA 5, 7(±17 −59 52) (−) −1.8 𝑡 = −6.9∗∗ −1.4 𝑡 = −5.6∗∗

(4) “Attention is highlyfocused”

ACC: BA 32, 33(±8 36 14) (+) .62 𝑡 = 2.86∗ .10 𝑡 = .44

(5) “Ecstatic joy experienced”N Ac

(±10 9 −4) (+) .88 𝑡 = 3.5∗∗ .94 𝑡 = 3.8∗∗

Med OFC(±8 50 −9) (+) 1.44 𝑡 = 7.2∗∗ .49 𝑡 = 2.6∗

(6) Less rhythmic movement

Somatosens: BA 1, 2, 3(±39 −28 53) (−) −1.50 𝑡 = −7.3∗∗ −1.38 𝑡 = −6.9∗∗

Prim Motor: BA 4 (±35 −23 53) (−) −1.47 𝑡 = −5.8∗∗ −1.38 𝑡 = −5.6∗∗

Cerebellum (±0 −61 −34) (−) −.77 𝑡 = 4.3∗∗ −.62 𝑡 = −3.6∗∗∗∗

𝑃 < .001.∗

𝑃 < .05.BA: Brodmann’s area, NAc: nucleus accumbens, Med OFC: medial orbitofrontal cortex, and ACC: anterior cingulate cortex.

somatosensory cortex, the primary motor cortex, and thecerebellum. This alternative hypothesis was strongly rejectedin all 3 areas.

In addition to testing the six a priori hypotheses, standardSPM5 statistical tests using post hoc analysis were computedfor all brain tissue. Figure 2 displays all cortical surfaces withpost hoc 𝑡 values greater than +3 (in red and yellow) or−3 (in blue and green) in the contrast (jhana-rest). It showsvery extensive but “patchy” areas of activation, with 63 clus-ters significantly positive, and 27 clusters were significantlynegative, suggesting an overall pattern of diffuse activationduring jhana. Perhaps the most evident results in Figure 2are that transition to jhana is associated with selectivedecreases in BOLD signal in the parietal and posterior frontallobes (confirmed by a priori tests above) and with selectiveincreases in the right temporal region.

Given that the data support the six hypotheses, we thendisaggregated the results to explore whether the differentjhana meditation states produced different brain activationpatterns. Figure 3(a) plots the BOLD signal of each statecontrasted with J2, with a separate line for each of the ROIsfrom H1 to H3. For example, the line labeled “orientation”plots the BOLD signal (relative to J2) on the 𝑦-axis as afunction of meditation state on the 𝑥-axis, progressing fromrest to AC to J2 and on through J5. It shows a steep declinefrom rest and AC to J2, and another steep decline to J3, thenreaches a global maximum at J4, followed by a return tothe low levels of J3. Interestingly, the remaining four lines

Posterior AnteriorL R

PosteriorAnterior

Figure 2: Cortical surfaces with post hoc 𝑡 values greater than +3(in red and yellow) or −3 (in blue and green) as calculated by SPM5using theBOLDcontrast (jhana-rest).Note that transition to jhana isassociated with selective increases in BOLD signal in right temporalregion and with decreases in parietal lobe and posterior frontal lobe.

in Figure 3(a) are highly correlated with the “orientation”line, showing similar patterns of decline, steep increases atJ4, and return to low values at J5. The correlation suggests

6 Neural Plasticity

Rest AC J2 J3 J4 J5

BOLD

sign

al co

mpa

red

to J2

OrientVisualAud

BrocaWernicke

2

1.5

1

0.5

0−0.5

−1

−1.5

−2

−2.5

−3

(a)

Rest AC J2 J3 J4 J5

BOLD

sign

al co

mpa

red

to J2

N AcMed OFC

ACC

0.5

0

−0.5

−1

−1.5

−2

−2.5

−3

(b)

Figure 3: Average BOLD signal of each meditation state contrasted with J2 is shown on the 𝑦-axis, with a separate line for each ROI. The𝑥-axis denotes each state from rest to AC to J2–J5. The mean SE of the signal contrasts averaged over the ROIs and states was ± .3. Note thehigh correlation between the lines in (a) and the steep increase in “NAc” at J2 in (b) (J1 was not recorded due to head movement artifacts).

an association between the ROIs, the most likely being thatreduced activation of vision and audition will “deafferent” theorientation area from its normal inputs, causing an alteredsense of orientation.

Figure 3(a) also gives a more nuanced view of individualjhanas than the pooled results in Table 1. While the averagejhana shows lower activation than rest and AC (as predictedby H1–H3), the individual jhanas show great variability, withlower activation in J2, J3, and J5, (as predicted by traditionalreports in Figure 1), but J4 shows activation equal to or higherthan rest. We caution that this figure plots single meditationstates of an individual, so that a single distractor event couldgreatly alter the activation pattern during ameditation. In thiscase, a distractor event may have occurred during J4, causingincreased activity in visual, auditory, and orientation area(however, the subject did not report any distractions duringdebriefing). A final deviation from predictions is that nodecline in activation occurs after J2, whereas Figure 1 wouldpredict that activity will decline with each successive jhana inareas associated with sensing external stimuli.

Figure 3(b) plots the BOLD contrast (relative to J2) of theremaining ROIs as a function of meditation state on the 𝑥-axis, progressing from rest through J5. The line denoted as“NAc,” shows a very steep increase in activation from rest andAC to J2, consistent with Figure 1. But activity in the NAcdeclines during J3 to near that of rest and AC and declineseven further in J5, consistent with a dopamine depletionhypothesis in later jhanas. The line for Med OFC showsmoderate decline during J3 and reaches its maximum at J4.This pattern contrasts with the predictions of Figure 1 whereJ4 is reported as less joyful than J2 and J3. Finally, the linelabeled “ACC” shows increased monitoring from rest to J2,declining to lower monitoring at J3 and J5, but spiking at J4.Since Figure 3(b) shows that J2 was the only jhana to activate

the complete dopamine pathway, tests of the alternativehypothesis were conducted on J2 alone. Consistent with thepooled results in Table 1, the alternative hypothesis (H6) thatsubtle rhythmic movements triggered joy in J2 was rejected,with significantly lower activity in areas associated withmovement during J2 compared to rest in BA 1,2,3 (𝑡 = −4.7,𝑃 < .001), BA 4 (𝑡 = −4.5, 𝑃 < .001), and in the cerebellum(𝑡 = −1.75 n.s.). All signs were in the opposite direction fromthat predicted by the alternative hypothesis.

Figure 4 shows more detailed dynamics of the statetransitions, with the time course of the BOLD signal averagedover all voxels in three a priori specified ROIs during the 417fMRI scans. Figure 4(a) shows average BOLD signal for theorientation area BA 5 and 7, with the blue line representingthe right side and the red line representing the left. The blackspikes extending from the 𝑥-axis represent events where themeditator signaled a transition to a higher state with a mouseclick. Note the steep drop during the transitions from AC toJ2 and J2 to J3. These drops are not caused by the clickingaction because they do not appear during transition from resttoAC.Thedrops occurred promptly after the subject signaledthat he was starting to transition, beginning within 2 scans(5 sec) and reaching minimumwithin 8 scans (20 sec) duringthe AC to J2 transition, with similarly prompt transitionsfrom J2 to J3. Figure 4(b) shows the BOLD signal in theright and left ACC regions, with similar steep and promptdrops during the transitions from AC to J2, J2 to J3, andJ4 to J5. Finally, Figure 4(c) shows the BOLD signal in theright and left medial OFC, with even steeper drops duringthe transitions from AC to J2, J2 to J3, J3 to J4, and J4 to J5.

3.2. EEG Results. The EEG data were first examined foroutliers and missing data. There were no bad channels, sospatial interpolation was not required. Though no missing

Neural Plasticity 7

128

133

138

143

148

153

1 51 101 151 201 251 301 351 401Time (scans)

BOLD

sign

al in

BA

5, 7

Rest A.C J2 J3 J4 J5

RightLeft

Event

(a)

RightLeft

Event

147

149

151

153

155

157

159

161

1 51 101 151 201 251 301 351 401Time (scans)

BOLD

sign

al in

ACC


(b)

RightLeft

Event

107108109110111112113114115116

511 101 151 201 251 301 351 401Time (scans)

BOLD

sign

al in

med

ial O

FC


(c)

Figure 4: Time course (in fMRI scans) of BOLD signal for three a priori defined ROIs (blue line shows right side of ROI, and red line showsright side) graphing transitions between rest, access concentration, and jhanas. Figure 4(a) shows BOLD signal averaged for all voxels in BA5 and 7 (orientation area), Figure 4(b) shows average BOLD in ACC, and Figure 4(c) shows average BOLD in medial OFC. Note the promptdrop in signal during transition events, including the decline in BA 5,7 activity during jhanas and the increase in OFC signal during jhanas(J1 was not recorded due to head movement artifacts).

data was found, all of the data for J1 are outliers, with putativegamma power at least 10 times the gamma power of otherjhanas and rest. It is likely that much of the gamma was dueto muscle tension because of head movements. Hence J1 isexcluded from analysis because it was more than 4 standarddeviations away from any other state. All data for remainingstates were approximately normally distributed.

Statistical tests for the planned comparisons that test H1–H6 are presented in Table 2. Similar to Table 1, column 1shows the subjective experience, column 2 shows the ROIsand the scalp electrode locations (from [44]) associated withthat experience, and column 3 shows the predicted directionof contrasts between jhana and rest. Column 4 shows theactual gamma power (25–42Hz) measured at that scalplocation. In the case of the first row, the gamma power at

O1 (overlying the primary and associative visual cortex BA17,19) showed no significant difference between jhana and rest.Examining all rows of column 4 shows that gamma powerincreased significantly only in the electrode locations overly-ing the ACC and the Med OFC, consistent with H4 and H5.However, in locations overlying regions expected to decreaseactivation (H1, H2, H3, and H6), all showed nonsignificantcontrasts in the gamma range (with the exception of C4,which was in the predicted direction). We also examinedcontrasts in the alpha1 range (not shown), which Laufs etal. [45] demonstrated are negatively correlated with fMRIactivation. Twelve of the 14 contrasts testing H1, H2, H3,and H6 showed significant increases in the alpha1 range,consistent with the hypotheses. We integrated the powerinformation from many bands in column 5, which calculates

8 Neural Plasticity

Table 2: Contrasts in the spectral power of the EEG signal in selected bands at a priori defined scalp locations related to 6 hypotheses on jhanaactivity compared with rest, followed by its 𝐹-test on the null hypothesis that jhana activity is equal to rest activity. All 𝐹 statistics are withdegrees of freedom of (1,502). Contrasts labeled simply “jhana” refer to the pooled activity over all recorded jhanas 2–5. In the last (summary)column, all significant differences are in the direction predicted by the 6 hypotheses.

Subjective report duringjhanas

A priori ROI (scalpelectrode locations)

Predicted sign ofcontrast (jhana-rest)

Contrast of power ingamma range(jhana-rest)

Contrast in power of(gamma + beta) −(alpha1 + theta)

(1) “External awarenessdims”

Visual: BA 17, 19(O1) (−) Ns −.08 𝐹 = 5.6∗

(O2) Ns −.11 𝐹 = 8.3∗

Auditory: BA 41, 42(T3) (−) Ns Ns(T4) Ns −.10 𝐹 = 15∗∗

(2) “Internal verbalizationfades”

Broca: BA 44, 45(FC5) (−) Ns −.07 𝐹 = 9.3∗

Wernicke: BA 39, 40(Tp7) (−) Ns Ns

(3) “Altered sense ofpersonal boundaries”

Orientation: BA 5, 7(P1) (−) Ns −.08 𝐹 = 6.6∗

(P2) Ns −.12 𝐹 = 13∗∗

(P3) Ns −.06 𝐹 = 5.0∗

(P4) Ns −.11 𝐹 = 16∗∗

(4) “Attention is highlyfocused”

ACC: BA 32, 33(AFz) (+) +.030 𝐹 = 21∗∗ +.12 𝐹 = 27∗∗

(Fz) +.014 𝐹 = 8.0∗ +.07 𝐹 = 7.8∗

(FCz) −.015 𝐹 = 5.1∗ Ns

(5) “Ecstatic joyexperienced”

N Ac (+) (unobservable) (unobservable)Med OFC(Fp1) (+) +.104 𝐹 = 57∗∗ +.42 𝐹 = 149∗∗

(Fp2) +.092 𝐹 = 38∗∗ +.35 𝐹 = 75∗∗

(6) Less rhythmicmovement

Somatosens: BA 1, 2, 3(C3) (−) Ns −.07 𝐹 = 8.7∗

(C4) −.013 𝐹 = 4.6∗ −.11 𝐹 = 15∗∗

Prim motor: BA 4(FC3) (−) Ns Ns(FC4) Ns Ns

Cerebellum (−) (unobservable) (unobservable)∗

𝑃 < .05.∗∗

𝑃 < .001.BA: Brodmann’s area, NAc: nucleus accumbens, Med OFC: medial orbitofrontal cortex, and ACC: Anterior cingulate cortex.

the difference in power between the high frequencies (gamma+ beta) minus the power in the lower frequencies (alpha1 +theta) for jhana compared to rest. Consistent with column 4,the largest increases in activation during jhana are observednear the Med OFC (H5), accompanied by smaller but verysignificant increases in ACC (H4). Significant declines inactivity during jhana are observed near BA 17,19, BA 41,42, BA44,45, BA 5,7, and BA 1,2,3, consistent with those hypotheses(H1, H2, H3, and H6).

4. Discussion

The fMRI and EEG recordings provide mutually consistentevidence on the neural correlates of ecstatic meditationscalled jhanas. In the cortical regions associated with externalawareness, verbalization, and orientation (H1, H2, and H3),Table 1 shows a lower fMRI BOLD signal during jhanacontrasted with rest. In addition, Table 2 shows that theEEG signal shifted to the lower-power bands of theta and

Neural Plasticity 9

alpha1, although it is acknowledged that spatial localization ofcortical function with scalp EEG has some limitations. In theregion associated with executive control (H4) and the regionassociatedwith subjective happiness (H5), the fMRI in Table 1showed higher BOLD signal during jhana contrasted withrest, while the EEG in Table 2 showed a shift to higher powerin the beta and gamma bands. In addition, the subcorticalimaging from the fMRI was able to distinguish whether thesubjective happiness (H5) was associated with activation ofthe dopamine/opioid reward system or due to purely corticalexpectation effects. Table 1 (in the rowH5) shows very strongactivation of the NAc in the ventral striatum indicatingthat the full pathway was activated in at least one of thejhanas.

Examining individual jhanas revealed several deviationsfrom the predictions derived from subjective reports inFigure 1. First, activity in orientation and visual areas doesdecline below rest and AC but does not decline furtherafter J3, contrary to reports that each succeeding jhana goesdeeper. Second, activity in the NAc peaks during J2 and thendrops quickly, contrary to reports that J3 is equally joyful. Weconclude that full activation of the dopamine reward systemoccurred only in J2, while J3 activated only the Med OFCportion of the reward system.

Previous imaging of the dopamine/opioid reward systemhas always used external stimuli to activate it (e.g., actual foodor drink was consumed or photos of loved ones cued a shortperiod of attachment). In contrast, jhana meditators claimthat they can voluntarily generate increased happiness purelyby volitional mental processes and for extended periods. Wetested this claim in several ways. First, we examined ROIsassociated with somatosensory and motor coordination,which would be active if the subject wasmaking subtle rhyth-mic movements known to trigger ecstatic ASCs [3]. Theseareas were not found to show increases but instead showedsignificant decreases in activity during J2, consistent with theclaim that the reward system is triggered without physicalcues or imaginedmovements. Another alternative hypothesisis that the subject was using indirect mental processes tostimulate the reward system such as evoking a visual orauditory memory of a happy time, which in turn wouldtrigger the reward system. However, our evidence in Tables1 and 2 (row H1) showed that the cortical ROIs associatedwith vision and hearing declined significantly in activityduring jhana (and Figure 3(a) confirms this specifically forJ2), contrary to this alternate hypothesis. Finally, evidence onlateralized brain activation such as those related to H2 andWernicke’s areamust be interpreted with some caution, as thesubject examined in the current study was left handed. Lefthandedness can be associated with structural and functionalchanges in brain symmetry, as compared to the majority ofhuman subjects, who are strongly right handed [46, 47], andthis fact might have influenced some results in Figures 3 and4.

4.1. Mechanisms of Action. Our data would reject four possi-ble cortical mechanisms (expectations, rhythmic movement,visual memories, and auditory memories) by which thesubject might have self-stimulated his own reward system

during J2. Several other pathways are possible that ourexperiment did not test. First, it is known that reciprocalconnections exist between the NAc and the medial OFC, sothat it might be possible to activate a feedback loop betweenthe two. Under normal conditions, the feedback loop wouldbe quickly interrupted by shifting attention to everchanginginput from visual, auditory, or somatic senses, but thesecortical areas have been downregulated, and attention maybe tightly focused on reinforcing the feedback loop.The loopmight be realized by creating a series of very short tasks thatcan each be completed successfully, allowing a new goal tobe achieved and reward attained with each newmoment.Theclassicmeditation instructions for breathingwould constitutesuch a task, wherein the student is instructed: “When thatin-breath finishes, you know that moment. You see in yourmind that last moment of the in-breath. You then see the nextmoment as a pause between breaths, and then many moremoments of pause until the out-breath begins. . .Weare awareonly of the beautiful breath, without effort and for a very longtime.” ([9] p.16).

Other possiblemechanisms of action could comprise sub-cortical activations that might have reward characteristics.For example, shifting control of breathing from the voluntarymotor cortex to the involuntary medullary rhythmicity areain the brain stem might be perceived as relaxing, as wellas giving rise to a common altered experience of “feelinglike I am being breathed, not in control.” Also, rhythmicmovements might be maintained below the level of corticalcontrol, since spinal reflexes are now known to mediaterhythmic movements as complex as coordinating leg move-ments related to walking.

Our results also shed light on the magnitude of theactivation of the dopamine reward system. Subjective reportsfrom the subject indicated extremely high magnitude ofreward, comparing J1 (which was not recorded due to headmovement) to continuous multiple orgasms, J2 to “openinga birthday gift and getting exactly what you most wishedfor,” and J3 to postcoital bliss. Yet the objective activationof the reward system in J2 was not extreme. The apparentmismatch between extreme subjective reports and moderateobjective activation can be explained by the signal-to-noiseratio of the circuits. When most other cortical activity isreduced, as in this subject, a much smaller reward signalcan be detected and will be perceived as more intense thanwhen cortical “noise” from other sources is high, as innormal awareness. Indeed, during normal awareness it takesdrug-induced hyperstimulation of the dopamine pathways togenerate such extreme subjective reports. If this signal-to-noise view is correct, then jhana’s reduced sense awarenessis not incidental to achieving extreme pleasure but is acontributing condition.

Despite the moderate level of activation, caution is advis-able with any voluntary stimulation of the reward systems.Drugs of abuse can generate short-term bliss but can quicklyincrease tolerance, requiring ever greater doses of the drugto create the same level of pleasure. They can also createwithdrawal symptoms during abstinence [33]. In contrastto the drugs, jhana meditators report negative tolerancebecause they can achieve the same state more quickly with


less effort over time, and no withdrawal symptoms havebeen reported when meditation is stopped. Nevertheless,Figure 3 shows that NAc activity dropped below normalresting consciousness in J5, whichmay be a sign of short-termtolerance and neurotransmitter depletion.

4.2. Implications. Our experiment is to our knowledge thefirst that compares brain states in five different meditations(AC and J2–J5), finding strong differences between ACmed-itation and jhana, and smaller but still significant differencesbetween jhana states. These in turn differ from the TibetanBuddhist compassion meditation reported by Lutz et al. [42]where EEG gamma frequencies were dominant and from thealpha dominance of Transcendental Meditation [48]. Takentogether, the multiplicity of brain states suggests that theremay be a vast array of ASCs available through meditation,depending on which brain regions are given awareness andwhich are inhibited from awareness [49]. If there are a largenumber of possible ASCs, it is likely that only some wouldhave survival value. For example, the state of mystical unionwith all beings might be helpful in encouraging cooperationwith all people in the tribe, so that evolution may haveselected certain of these ASCs to be more easily learned andretained.

However, the same reasoning would suggest that theability to self-stimulate the brain’s reward system would bedysfunctional in the struggle for survival and procreationbecause it could short-circuit the system that motivatessurvival actions. Organisms that are adept at self-stimulationwould quickly die out if they fail to respond to environmentaldemands or to pass on their genes. This reasoning sug-gests caution in making autonomous self-stimulation moreavailable, but we point out that the modern environmentalready allows unprecedented stimulation of the dopaminereward system with plentiful food and drugs of abuse. Ameditation that stimulates the reward system without theharmful effects of obesity and environmental damage couldbe beneficial in themodern environment. On the other hand,a meditation that short-circuits the desire to get an educationand work for long-term goals could become dysfunctional.Rather than simply stimulating the reward system in responseto traditional goals of food and sex, it would be beneficial toregulate the system and focus it on long-term goals that aremore adaptive.

This case study provides guidelines for larger studies onjhana meditation in several areas. First, it demonstrates thatjhana is not so fragile that it can be destroyed by the presenceof curious experimenters or by intrusive sounds of MRIscanners. Hence, it can be scientifically investigated. Second,the transition time to move from one jhana to another in apracticed subject is much shorter (between 5 and 20 seconds)than we expected, in line with other meditations that do notproduce such extremeASCs [42].With short transition times,it might be feasible to use better randomized designs thatalternate control states with meditations (however, the shorttransition times here may be due to the subject’s internalknowledge of readiness to transition, and he may not beable to transit “on demand”). Third, the experiment could beshortened if interest is focused only on the reward system

because only J2 shows strong self-activation of the NAc.Fourth, the simple “resting” condition used here could bereplaced with better controls that have been demonstrated toincrease happiness, such as “remembering a happy event inyour life” or visualizing a loved one.

More potential subjects will become available as moreEnglish-speaking students are being trained in jhana medita-tion [9, 50]. How thesemeditators achieve periods of extremejoy without common negative side effects could contributeto the scientific “pursuit of happiness” and could pave theway for novel paradigms for rehabilitation and recovery fromnervous system injury.


The authors declared that they have no conflict of interests.

Acknowledgments

This research is dedicated to the memory of Dr. DouglasFinlayson of Seattle, WA, who initiated this study.Thanks arealso due to Rick Mendius and Rick Hanson for comments onearlier drafts. Earlier versions were presented at theMind andLife Summer Research Institute, June 6-12, 2008, in Garrison,NY, and at Cognitive Neuroscience Society, March 18, 2009,in San Francisco, CA.

References

[1] W. James, The Varieties of Religious Experience: A Study inHuman Nature, Harvard University, Cambridge, Mass, USA,1902.

[2] A. Maslow, Religions, Values and Peak-Experiences, Ohio StateUniversity Press, Columbus, Ohio, USA, 1964.

[3] D. Vaitl, J. Gruzelier, G. A. Jamieson et al., “Psychobiology ofaltered states of consciousness,” Psychological Bulletin, vol. 131,no. 1, pp. 98–127, 2005.

[4] S. Lyubomirsky,The How of Happiness: A Scientific Approach toGetting the Life you Want, Penguin Press, New York, NY, USA,2008.

[5] M. A. Cohn, B. L. Fredrickson, S. L. Brown, J. A. Mikels, and A.M. Conway, “Happiness unpacked: positive emotions increaselife satisfaction by building resilience,” Emotion, vol. 9, no. 3, pp.361–368, 2009.

[6] A. Khema, Visible Here and Now, Shambhala Publications,Boston, Mass, USA, 2001.

[7] M. Beauregard and V. Paquette, “Neural correlates of a mysticalexperience in Carmelite nuns,” Neuroscience Letters, vol. 405,no. 3, pp. 186–190, 2006.

[8] A. B. Newberg and J. Iversen, “The neural basis of the complexmental task ofmeditation: neurotransmitter andneurochemicalconsiderations,”Medical Hypotheses, vol. 61, no. 2, pp. 282–291,2003.

[9] A. Brahm,Mindfulness, Bliss, and Beyond: A Meditator’s Hand-book, Wisdom Publications, Boston, Mass, USA, 2006.

[10] A. Lutz,H.A. Slagter, J. D.Dunne, andR. J.Davidson, “Attentionregulation and monitoring in meditation,” Trends in CognitiveSciences, vol. 12, no. 4, pp. 163–169, 2008.

[11] A. Lutz, J. Brefczynski-Lewis, T. Johnstone, and R. J. Davidson,“Regulation of the neural circuitry of emotion by compassion


meditation: effects of meditative expertise,” PLoS ONE, vol. 3,no. 3, Article ID e1897, 2008.

[12] E. Luders, A. W. Toga, N. Lepore, and C. Gaser, “The under-lying anatomical correlates of long-term meditation: largerhippocampal and frontal volumes of gray matter,” NeuroImage,vol. 45, no. 3, pp. 672–678, 2009.

[13] S. W. Lazar, C. E. Kerr, R. H. Wasserman et al., “Meditationexperience is associated with increased cortical thickness,”NeuroReport, vol. 16, no. 17, pp. 1893–1897, 2005.

[14] J. A. Brefczynski-Lewis, A. Lutz, H. S. Schaefer, D. B. Levinson,and R. J. Davidson, “Neural correlates of attentional expertise inlong-termmeditation practitioners,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 104, no.27, pp. 11483–11488, 2007.

[15] K. A. MacLean, E. Ferrer, S. R. Aichele et al., “Intensivemeditation training leads to improvements in perceptual dis-crimination and sustained attention,” Psychological Science, vol.21, pp. 820–830, 2010.

[16] R. J. Davidson, J. Kabat-Zinn, J. Schumacher et al., “Alterationsin brain and immune function produced by mindfulnessmeditation,”PsychosomaticMedicine, vol. 65, no. 4, pp. 564–570,2003.

[17] Y. Y. Tang, Y. Ma, J. Wang et al., “Short-term meditationtraining improves attention and self-regulation,” Proceedings ofthe National Academy of Sciences of the United States of America,vol. 104, no. 43, pp. 17152–17156, 2007.

[18] R. Walsh and S. L. Shapiro, “The meeting of meditative disci-plines and western psychology: a mutually enriching dialogue,”American Psychologist, vol. 61, no. 3, pp. 227–239, 2006.

[19] N. D. Daw, S. Kakade, and P. Dayan, “Opponent interactionsbetween serotonin and dopamine,”Neural Networks, vol. 15, no.4-6, pp. 603–616, 2002.

[20] J. B. Nitschke, I. Sarinopoulos, and K. L. Mackiewicz, “Func-tional neuroanatomy of aversion and its anticipation,”NeuroIm-age, vol. 29, no. 1, pp. 106–116, 2006.

[21] W. Schultz, “Multiple reward signals in the brain,” NatureReviews Neuroscience, vol. 1, no. 3, pp. 199–207, 2000.

[22] A. Rangel, C. Camerer, and P. R. Montague, “A framework forstudying the neurobiology of value-based decision making,”Nature Reviews Neuroscience, vol. 9, no. 7, pp. 545–556, 2008.

[23] I. E. de Araujo, E. T. Rolls, M. I. Velazco, C. Margot, andI. Cayeux, “Cognitive modulation of olfactory processing,”Neuron, vol. 46, no. 4, pp. 671–679, 2005.

[24] M. L. Kringelbach, “The human orbitofrontal cortex: linkingreward to hedonic experience,” Nature Reviews Neuroscience,vol. 6, no. 9, pp. 691–702, 2005.

[25] A. J. Blood and R. J. Zatorre, “Intensely pleasurable responsesto music correlate with activity in brain regions implicated inreward and emotion,” Proceedings of the National Academy ofSciences of theUnited States of America, vol. 98, no. 20, pp. 11818–11823, 2001.

[26] M. L. Kringelbach, J. O’Doherty, E. T. Rolls, and C. Andrews,“Activation of the human orbitofrontal cortex to a liquid foodstimulus is correlated with its subjective pleasantness,” CerebralCortex, vol. 13, no. 10, pp. 1064–1071, 2003.

[27] B. A. Arnow, J. E. Desmond, L. L. Banner et al., “Brain activationand sexual arousal in healthy, heterosexual males,” Brain, vol.125, no. 5, pp. 1014–1023, 2002.

[28] D. Mobbs, M. D. Greicius, E. Abdel-Azim, V. Menon, and A.L. Reiss, “Humor modulates the mesolimbic reward centers,”Neuron, vol. 40, no. 5, pp. 1041–1048, 2003.

[29] K. Preuschoff, P. Bossaerts, and S. R. Quartz, “Neural differ-entiation of expected reward and risk in human subcorticalstructures,” Neuron, vol. 51, no. 3, pp. 381–390, 2006.

[30] A. Bartels and S. Zeki, “The neural correlates of maternal andromantic love,” NeuroImage, vol. 21, no. 3, pp. 1155–1166, 2004.

[31] H. Plassmann, J. O’Doherty, B. Shiv, and A. Rangel, “Marketingactions can modulate neural representations of experiencedpleasantness,” Proceedings of the National Academy of Sciencesof the United States of America, vol. 105, no. 3, pp. 1050–1054,2008.

[32] I. E. de Araujo, E. T. Rolls, M. I. Velazco, C. Margot, andI. Cayeux, “Cognitive modulation of olfactory processing,”Neuron, vol. 46, no. 4, pp. 671–679, 2005.

[33] P. W. Kalivas and N. D. Volkow, “The neural basis of addiction:a pathology of motivation and choice,” American Journal ofPsychiatry, vol. 162, no. 8, pp. 1403–1413, 2005.

[34] S. Pecina, K. S. Smith, and K. C. Berridge, “Hedonic hot spots inthe brain,”The Neuroscientist, vol. 12, no. 6, pp. 500–511, 2006.

[35] K. D’Ardenne, S. M. McClure, L. E. Nystrom, and J. D. Cohen,“BOLD responses reflecting dopaminergic signals in the humanventral tegmental area,” Science, vol. 319, no. 5867, pp. 1264–1267,2008.

[36] S. C. Cramer, R. M. Weisskoff, J. D. Schaechter et al., “Motorcortex activation is related to force of squeezing,”Human BrainMapping, vol. 16, no. 4, pp. 197–205, 2002.

[37] Welcome Trust Centre for Neuroimaging, “SPM5: StatisticalParametric Mapping Software,” 2005, http://www.fil.ion.ucl.ac.uk/spm/software/spm5/.

[38] J. A. Maldjian, P. J. Laurienti, R. A. Kraft, and J. H. Burdette, “Anautomated method for neuroanatomic and cytoarchitectonicatlas-based interrogation of fMRI data sets,” NeuroImage, vol.19, no. 3, pp. 1233–1239, 2003.

[39] M. Brett, J.-L. Anton, R. Valabregue, and J.-B. Poline, “Regionof interest analysis using an SPM toolbox,” NeuroImage, vol. 16,no. 2, 2002.

[40] U. Kirk, M. Skov, M. S. Christensen, and N. Nygaard, “Braincorrelates of aesthetic expertise: a parametric fMRI study,”Brainand Cognition, vol. 69, no. 2, pp. 306–315, 2009.

[41] B. Knutson, S. Rick, G. E. Wimmer, D. Prelec, and G. Loewen-stein, “Neural predictors of purchases,”Neuron, vol. 53, no. 1, pp.147–156, 2007.

[42] A. Lutz, L. L. Greischar, N. B. Rawlings, M. Ricard, and R. J.Davidson, “Long-term meditators self-induce high-amplitudegamma synchrony during mental practice,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 101, no. 46, pp. 16369–16373, 2004.

[43] L. I. Aftanas and S. A. Golocheikine, “Human anterior andfrontal midline theta and lower alpha reflect emotionallypositive state and internalized attention,” Neuroscience Letters,vol. 310, no. 1, pp. 57–60, 2001.

[44] M. Okamoto, H. Dan, K. Sakamoto et al., “Three-dimensionalprobabilistic anatomical cranio-cerebral correlation via theinternational 10-20 system oriented for transcranial functionalbrain mapping,” NeuroImage, vol. 21, no. 1, pp. 99–111, 2004.

[45] H. Laufs, A. Kleinschmidt, A. Beyerle et al., “EEG-correlatedfMRI of human alpha activity,” NeuroImage, vol. 19, no. 4, pp.1463–1476, 2003.

[46] A. M. Galaburda, M. LeMay, T. L. Kemper, and N. Geschwind,“Right-left asymmetries in the brain. Structural differencesbetween the hemispheres may underlie cerebral dominance,”Science, vol. 199, no. 4331, pp. 852–856, 1978.


[47] S. Knecht, A. Floel, B. Drager et al., “Degree of language later-alization determines susceptibility to unilateral brain lesions,”Nature Neuroscience, vol. 5, no. 7, pp. 695–699, 2002.

[48] F. Travis and R. K. Wallace, “Autonomic and EEG Patterns dur-ing eyes-closed rest and transcendental meditation practice,”Consciousness and Cognition, vol. 8, no. 3, pp. 302–318, 1999.

[49] J. H. Austin, Selfless Insight, MIT Press, Cambridge, Mass, USA,2009.

[50] R. Shankman, The Experience of Samadhi: An In-Depth Explo-ration of Buddhist Meditation, Shambala Press, Boston, Mass,USA, 2008.


Research ArticleQuality and Timing of Stressors Differentially Impact on BrainPlasticity and Neuroendocrine-Immune Function in Mice

Sara Capoccia,1 Alessandra Berry,1 Veronica Bellisario,1 Davide Vacirca,2 Elena Ortona,2

Enrico Alleva,1 and Francesca Cirulli1

1 Section of Behavioural Neuroscience, Department of Cell Biology and Neurosciences, Istituto Superiore di Sanita,Viale Regina Elena 299, 00161 Rome, Italy

2 Section of Biomarkers in Degenerative Diseases, Department of Cell Biology and Neurosciences, Istituto Superiore di Sanita,Viale Regina Elena 299, 00161 Rome, Italy

Correspondence should be addressed to Francesca Cirulli; [email protected]

Received 23 January 2013; Revised 8 March 2013; Accepted 12 March 2013


Copyright © 2013 Sara Capoccia et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

A growing body of evidence suggests that psychological stress is a major risk factor for psychiatric disorders.The basic mechanismsare still under investigation but involve changes in neuroendocrine-immune interactions, ultimately affecting brain plasticity. In thisstudy we characterized central and peripheral effects of different stressors, applied for different time lengths, in adult male C57BL/6Jmice. We compared the effects of repeated (7 versus 21 days) restraint stress (RS) and chronic disruption of social hierarchy (SS)on neuroendocrine (corticosterone) and immune function (cytokines and splenic apoptosis) and on a marker of brain plasticity(brain-derived neurotrophic factor, BDNF ). Neuroendocrine activation did not differ between SS and control subjects; by contrast,the RS group showed a strong neuroendocrine response characterized by a specific time-dependent profile. Immune functionand hippocampal BDNF levels were inversely related to hypothalamic-pituitary-adrenal axis activation. These data show a finemodulation of the crosstalk between central and peripheral pathways of adaptation and plasticity and suggest that the length ofstress exposure is crucial to determine its final outcome on health or disease.

1. Introduction

Stressful events are well-known risk factors that can promoteneurochemical changes ultimately involved in the pathophys-iology of psychiatric disorders such asmajor depression [1–3].Any change of the internal or external milieu may representa source of stress triggering a complex and coordinated setof physiological responses involving (among others) the acti-vation of the hypothalamus-pituitary-adrenal (HPA) axis [4–6]. Although adaptive on the short run, prolonged exposureto glucocorticoids hormones (GC), secreted following stress,may exhaust the capacity of an organism to cope with furtherstressors and, given the catabolic nature of these adrenalglucocorticoids, lead to an impairment in brain plasticity[5, 7, 8].

Stress begins in the brain with the perception and inter-pretation of the stressful event and affects the brain itself as

well as the rest of the body through plastic changes, leading toadaptation. The connection between central stress responsepathways and peripheral targets involves the alteration of anumber of neurochemical and/or inflammatory factors thatultimately affect neuronal functioning and/or survival [8, 9].One of the most representative players implicated in theseevents is the neurotrophin brain-derived neurotrophic factor(BDNF), which is involved in synaptic and morphologicalplasticity of the brain both during development (with max-imal levels during times of neuronal growth, differentiationand synaptogenesis) as well as at adulthood [10–13]. Highlevels of this neurotrophin are found in the hippocampus,a brain region expressing also high levels of receptors forGC (GR) and playing a main role in the negative feedbackregulation of the HPA axis, a pathway often disinhibitedin depressed subjects [14]. A growing body of evidenceshows that chronic stress decreases the expression of BDNF

2 Neural Plasticity

contributing to neuronal atrophy in the hippocampus andthat antidepressant treatment reverses or blocks these effects,restoring brain plasticity [8, 9, 15, 16].

By being able to directly affect HPA axis activity [9,17] and being produced by cells outside the nervous sys-tem (including immune cells, adipocytes, endocrine, andendothelial cells), BDNF has a key position in integratingneural, immune, and endocrine responses to stress [8, 18, 19].Indeed, the central nervous system and the immune systemare known to be engaged in an intense bidirectional crosstalkwhich can be affected by stress and which involves multiplemediators, including cytokines and growth factors [20]. Asan example, the immune signaling cytokines, particularly theproinflammatory ones such as interleukin-6 (IL-6) or tumornecrosis factor-alpha (TNF-𝛼), are elevated following stressexposure and can thwart brain plasticity eliciting depressivesymptoms, which are amenable to antidepressant treatment[20]. However, the directionality of the effects of stress is stilla matter of intense investigation: for instance, GC releasedin response to stress can act both enhancing and inhibitingimmune responses and by decreasing or increasing levels ofneurotrophins [21–24]. Such opposite effects might coexistin light of the fact that during stress, multiple interactingmediators are activated in a nonlinear network influencingdifferent systems and functions [25]. Factors such as theduration (acute versus chronic) of stress as well as the timeof exposure to GC, relative to the activation and time courseof the immune response, might differently impact healthoutcome [21]. Progress in understanding the pathophysiologyof stress would greatly benefit from further preclinical studiesincorporating both the permissive as well as the inhibitoryrole of GC in immune-endocrine interactions and mimicconditions experienced in everyday life [26].

Restraint stress (RS) and the chronic disruption of thesocial hierarchy (SS) are two of the most widely usedexperimental paradigms that can induce stress in mice. Thefirst relies on a combination of psychological and physicalstimuli and is considered a reliable model of severe stress inhumans [27, 28]; the latter represents a comprehensive andethologically relevant paradigm inducing chronic stress andleading to anxiety and/or depressive-like symptoms as oftenreported in stress-precipitated major depression [29–31].Thus, the main aim of the present study was to characterizecentral and peripheral effects of different stressors, appliedfor different time lengths on neuroendocrine and immuneresponses in adult male C57BL/6J mice. Specifically, wecompared the effects of repeated (7 versus 21 days) RS and SSon neuroendocrine (circulating corticosterone) and immune(circulating cytokines and splenic apoptosis) function and ona marker of brain plasticity (hippocampal BDNF) in orderto identify a specific neuroendocrine profile in response toa selective type of stress.

2. Materials and Methods

2.1. Animals. Experimental subjects were adult maleC57BL/6J mice purchased from a commercial breeder(Charles River, Calco, Italy). Upon arrival, all animalswere group-housed in the same room provided by air

conditioning (temperature 21 ± 1∘C, relative humidity60 ± 10%), in transparent Plexiglas cages (29 cm × 12 cm ×14 cm), under a reversed 12/12 h light/dark cycle with lightsoff from 0800 to 2000 h. Pellet food (standard diet Altromin-R, Rieper, Italy) and tap water were continuously available.All stressors were administered randomly throughoutthe active phase of the day. A Social Interaction Test wasused as a challenge to assess HPA axis response followingthe social stress procedure and took place between 1700–2000 h, that is, during the corticosterone (CORT) circadiantrough. All subjects were sacrificed at the end of the stressprocedure. Animal handling and experimental procedureswere performed in accordance with the EC guidelines(EC Council Directive 86/609 1987) and with the Italianlegislation on animal experimentation (Decreto L.vo 116/92).

2.2. Experimental Procedures

2.2.1. Experiment I: Effects of Restraint Stress on Neuroen-docrine and Immune Responses. Experimental subjects were15 mice divided into three groups: 7 days restraint stress (RS7,𝑛 = 5), 21 days restraint stress (RS21, 𝑛 = 5), and unhandledcontrols (CTRL, 5 subjects left undisturbed in their homecage). All subjects undergoing the same treatment conditionwere group-housed. The restraint procedure consisted inremoving subjects from their home cage and putting eachof them in a conical 50mL falcon tube, provided with holesfor breathing, on a laboratory bench under dim light for3 consecutive hrs/day. The stress was administered eachday at random times in order to prevent habituation tothe procedure. Animals from the RS21 were used to assessstress-related changes in CORT levels so to have repeatedmeasures for each subject during days 1, 7, and 21. Onthese days the procedure was administered at a fixed timesin order to take into account circadian rhythm, that is,from 1700 to 2000. Blood samples were collected by tailnick at 0 (basal) and 180min from the onset of stress (i.e.,at 2000). At the end of stress all mice (CTRL, RS7, andRS21) were sacrificed, trunk blood was collected to assesslevels of the proinflammatory cytokines interleukin 6 (IL-6)and tumor necrosis factor-alpha (TNF-𝛼), and of the anti-inflammatory cytokine Interleukin 10 (IL-10), [32, 33]. Brainand spleen were dissected out in order to assess, respectively,hippocampal BDNF levels and lymphocyte apoptosis.

2.2.2. Experiment II: Effects of Social Stress on Neuroendocrineand Immune Responses. Experimental subjects were 48 adultmale mice divided into three groups: 7 days social stress (SS7,𝑛 = 16), 21 days social stress (SS21, 𝑛 = 16), and controls(CTRL, 16 subjects group-housed). All mice undergoing theSS procedure were ear-marked and housed into 4 cages (4mice/cage) and social structure was disrupted twice a weekfor one or three weeks by replacing one mouse with a novelunfamiliar selected randomly from another cage [34, 35].Sawdust was replaced at the same time in all cages. Controlmice were also ear-marked and housed in stable groups of4 mice/cage. Cages were cleaned and sawdust replaced twicea week mimicking the handling procedure of the SS groups[31].

Neural Plasticity 3

The activity of the HPA axis was assessed in responseto a 20-minute acute stress (Social Interaction Test) andblood samples for CORT evaluation were collected from 8subjects per group (CTRL, SS7, SS21) right before (basal) and30min following the end of stress. Briefly, the night before theSocial Interaction Test, all subjects were individually housedto stimulate social interactions [36–38]. On the day of test,micewere placed in a novel cage, identical to the holding cage,ideally subdivided in three equal parts, with an unfamiliarconspecific of the same strain, weight, and sex that had beenpreviously isolated (standard opponent). Standard opponentswere marked with a yellow, scentless, and nontoxic paint [31].

At the end of the test mice that did not undergo theSocial Interaction test (5 mice for each group, randomlychosen) were sacrificed, and trunk blood was collected toassess also levels of IL-6, TNF-𝛼, and IL-10 [32, 33]. Brainsand spleen were dissected out, in order to assess, respectively,hippocampal BDNF levels and lymphocyte apoptosis.

2.3. Radioimmunoassay for Corticosterone Determination-RIA. Blood samples (100 𝜇L, approximate volume) were col-lected individually in potassium EDTA coated tubes (1.6mgEDTA/mL blood, Sarstedt, Germany). All samples werekept on ice and later centrifuged at 3000 rpm for 15minat +4∘C. Blood plasma was transferred to Eppendorf tubesfor CORT determination and stored at −20∘C until furtheranalysis. CORTwasmeasured using a commercially availableradioimmunoassay (RIA) kit containing 125Iodine-labeledCORT; 5 𝜇L of plasma was sufficient to carry out CORTmeasurement. Sensitivity of the assay was 0.125mg/dL, inter-and intra-assay variationwas less than 10 and 5%, respectively(MP Biomedicals Inc., CA, USA). Vials were counted for2min in a gamma-scintillation counter (Packard MinaxiGamma counter, Series 5000).

2.4. BDNF Measurement. BDNF evaluation was carried outwith an enzyme-linked immunosorbent assay kit (BDNFEmax ImmunoAssay System numberG7610, Promega,Madi-son, Wisconsin, USA) following the instructions providedby the manufacturer. Following sacrifice brains were quicklyremoved and the hippocampus was dissected out andimmediately stored at −80∘C until used. Brain tissues werehomogenized in a lysis buffer and centrifuged at 14000 rpm,and the supernatant was used for BDNF analyses. Briefly,BDNF standard and brain samples were distributed in 96-well immunoplates precoated with monoclonal anti-mouseBDNF antibody (100mL/well) and incubated for 2 h at roomtemperature. After washing, plates were incubated with ananti-human BDNF antibody for 2 h at room temperature.Theplates werewashed again and then incubatedwith an anti-IgYhorseradish peroxidase (HRP) for 1 h at room temperature.Tetramethylbenzidine (TMB)/peroxidase substrate solutionwas added to the wells to produce colorimetric reactionmeasured at 450 nm with a microplate reader (Dynatech MR5000, Dynatech Laboratories, Chantilly, VA, USA). BDNFconcentrations were determined from the regression line forthe BDNF standard incubated under similar conditions ineach assay. The sensitivity of the assay was about 15 pg/mg

of BDNF, and the cross-reactivity with other related neu-rotrophic factors (NGF, NT-3, and NT-4) is considered nil[39].

2.5. Cytokines Determination. Quantitative evaluation ofTNF-𝛼, IL-6, and IL-10 in sera from trunk blood of stressedand control mice was determined by ELISA kits (R&DSystems, Inc., Minneapolis, USA) according to the manu-facturer’s instructions. Briefly, standards, controls, and serawere placed into the wells and incubated 2 h at room temper-ature. After washing 5 times, the enzyme-linked polyclonalantibody specific for mouse cytokines was added to the wellsand then, afterwashing, the substrate solutionwas added.Theenzyme reaction was read at 450 nm (correction wavelengthset at 570 nm). The samples values were read off the standardvalue.

2.6. Splenocytes Apoptosis. Spleens were gently removed andsuspended in ice-cold culture RPMI-1640 medium (GIBCOBRL, Grand Island, NY). Splenocytes were isolated frommicespleen by flushing 5mL of RPMI-1640medium into spleen byneedle and syringe. Cells were then centrifugated at 1200 rpmin order to remove cellular debris. Cells were resuspended insupplemented RPMI 1640 and counted on a hemocytometerin trypan blue to ensure viability. Average viability was >90%.Splenocytes were then cultured in RPMI-1640 medium with10% FBS (Euroclone, Pero, Italy), 2mM glutamine (Sigma, StLouis, MO), and 50𝜇g/mL gentamycin (Sigma). Apoptosiswas measured after 1 h of culture. Apoptosis was quantifiedusing FITC-conjugated annexin V (AV) and propidiumiodide (PI) apoptosis detection kit (Marine Biological Lab-oratory, Woods Hole, MA) according to the manufacturer’sprotocol. Reported data are referred to AV-positive apoptoticcells. AV binds to phosphatidylserine which is exposed atthe outer surface of the cell membrane already at earlystages of apoptosis and remains so during the subsequentprocess of apoptosis. By defining apoptotic cells as those cellsstaining with AV, irrespective of PI staining, we were able todetect early (AV+/PI− cells) as well as late (AV+/PI+ cells)apoptotic cells. In this study, we analyzed specifically “earlyapoptosis” in which the nuclear changes are observed first, incontrast to the changes seen in the later stages of apoptosisand then in the necrosis, which usually begin with cellmembrane damage [40, 41]. Acquisition was performed ona FACSCalibur cytometer (BD Immunocytometry Systems)and 50.000 events per sample were run. Data were analyzedusing the Cell Quest Pro (BD Immunocytometry Systems)software.

2.7. Statistical Analysis. Data were analyzed using parametricanalysis of variance (ANOVA) with “condition” (controland stress) as between-subjects factor (BDNF, cytokines,apoptosis, CORT) and “day” (1, 7, 21) and “time” (0 and 180)as within-subject repeated measures (CORT assessment onlyfor the restraint stress).Post hoc comparisonswere performedusing theTukey’s test. Statistical analysiswas performedusingStatview II (Abacus Concepts, CA, USA). Data are expressedas mean + SEM. A significance level of 0.05 was chosen.

4 Neural Plasticity

0

100

200

300

400

RSCTRL

Day 1 Day 7 Day 210 0 0 180180180

CORT

(ng/

mL)

Restraint stress

∗

(a)

0

100

200

300

400

7 SSCTRL

21 SS

∗

Basal Social challenge

Social stress

CORT

(ng/

mL)

(b)

Figure 1: Effect of restraint stress on CORT secretion in mice. All subjects undergoing RS showed a reduced response of the HPA axis onday 7 (a). Effect of social stress on CORT secretion in mice. The response to an acute challenge (represented by the Social Interaction Test)was effective in inducing an increase in CORT secretion in all groups, with no differences in relation to social stress exposure (b). Results arepresented as mean + S.E.M. ∗𝑃 < 0.05.

3. Results

3.1. Experiment I: Restraint Stress

3.1.1. Corticosterone. Restraint stress was effective in chal-lenging the HPA axis. In fact, RS subjects showed overallhigher CORT levels (main effect of condition: F(1,8) = 17.917,𝑃 = 0.0029) compared to controls, particularly 180 minutesfrom the onset of stress (interaction between condition andtime: F(1,8) = 12.022, 𝑃 = 0.0085). Moreover, a blunted HPAaxis response characterized RS subjects on day 7 (main effectof days F(1,8) = 3.618 𝑃 = 0.0505; see Figure 1(a)).

3.1.2. BDNF. BDNF evaluation was performed on 4 mice ineach group since values from some subjects (1 subject for eachgroup)were found to be outliers andwere therefore discardedfrom the analysis (Grubbs’ test performed by GraphPadSoftware).

A time-dependent effect of RS was found for hippocam-pal BDNF levels (see Figure 2). In particular, post hoccomparisons show a decrease in BDNF levels following 21days of RS compared to RS7 (main effect of condition F(2,9) =4.164, 𝑃 = 0.0500). The latter group did not differ from theCTRL subjects. However, it is worth noticing that the lackof difference between these two groups might be related toa reduced power of the statistical test (0.574) suggesting thatthis result suffers from a low number of experimental subjects(only 4 animals per experimental group) possibly maskingother significant trends (RS7 versus CTRL).

3.1.3. Cytokine Production. Following 7 days of restraintstress a tendency to increase was observed for levels of IL-6 (F(2,12) = 3.480; 𝑃 = 0.0643, Figure 3(a)). This trend

4

3

2

1

0CTRL RS7 RS21

∗

BDN

F (p

g/m

g tis

sue)

Figure 2: Effects of restraint stress on hippocampal BDNF levels.BDNF levels were decreased following a chronic 21 days restraintprocedure compared to 7 days of repeated restraint. Data shown aremean + S.E.M. ∗𝑃 < 0.05.

reached statistical significance when assessing TNF-𝛼 (maineffect of condition: F(2,12) = 5.558; 𝑃 = 0.0196, Figure 3(b))that returned to basal levels after 21 days. By contrast, IL-10 increased only after 21 days of restraint (main effect oftreatment: F(2,12) = 5.345; 𝑃 = 0.0219, Figure 3(c)).

3.1.4. Splenocytes Apoptosis. Splenocytes apoptosis slightlydecreased after 21 days of restraint (main effect of condition:F(2,12) = 8.597; 𝑃 = 0.0048, post hoc RS7 days versus RS21

Neural Plasticity 5

0

10

20

30IL-6

(pg/

mL)

Restraint stress

(a)

IL-10

39

41

45

43

0

∗

(pg/

mL)

Restraint stress

(b)

0

100

200

CTRL RS7 RS21

∗

(pg/

mL)

Restraint stressTNF-𝛼

(c)

Annexin V

0

4

8

12

CTRL RS7 RS21

Posit

ive c

ells

(%)

∗

Restraint stress

(d)

IL-6

0

30

3

6

9

CTRL SS7 SS21

(pg/

mL)

Social stress

(e)

IL-10

0

3

6

9

12

45

CTRL SS7 SS21

(pg/

mL)

Social stress

∗∗

(f)

Annexin V

0

4

8

12

CTRL SS7 SS21

Social stress

∗

Posit

ive c

ells

(%)

(g)

Figure 3: Effect of RS and SS on the immune system response. Restraint stress procedure. Following 7 days of restraint stress theproinflammatory cytokine TNF-𝛼 increases, (b) while the increase in IL-6 during days 7 and 21 just missed statistical significance (a); bycontrast the anti-inflammatory cytokine IL-10 increased following 21 days (c). The percentage of apoptotic splenocytes was found to bedecreased following 21 days of stress (d). Social stress procedure. Levels of IL-10 were increased already after 7 days of the SS procedure (f).Splenocytes apoptosis was increased after 7 days of SS and decreased following 21 days (g). No difference is evident as for levels of IL-6 (e).Results are presented as mean + S.E.M. ∗∗𝑃 < 0.01, ∗𝑃 < 0.05.

days 𝑃 < 0.01; see Figure 3(d)). No difference was foundbetween CTRL and RS7.

3.2. Experiment II—Social Stress

3.2.1. Corticosterone. TheSocial Interaction Test was effectivein inducing the activation of the HPA axis in all groups,

regardless of their stress history (effect of social challenge:F(1,29) = 153,515; 𝑃 < 0.0001, Figure 1(b)).

3.2.2. BDNF. Social stress, per se, did not affect hippocampalBDNF levels (no main effect of condition: F(2,12) = 2.015 𝑃 =0.1759, data not shown).

6 Neural Plasticity

3.2.3. Cytokine Production. Social condition did not affectsignificantly the production of IL-6, even if a slight increaseafter 7 days of social stress was observed, (effect of condition:F(2,12) = 3.075 𝑃 = 0.0835, see Figure 3(e)). By contrast,serum levels of IL-10 increased after 7 days of the socialstress procedure (main effect of condition: F(2,12) = 13.217𝑃 = 0.0009, see Figure 3(f)). Differences in serum TNFlevels between control and social stressed mice appearedundetectable (data not shown).

3.2.4. Splenocytes Apoptosis. Splenocytes apoptosis increasedafter 7 days of social stress and decreased following 21 days(main effect of treatment: F(2,12) = 47.932; 𝑃 < 0.0001, seeFigure 3(g)).

4. Discussion

Data from this study show that chronic RS is a powerful stres-sor eliciting strong neuroendocrine, and immune responsesand that brief versus prolonged exposure to this stress resultsin a differential activation of these systems in mice. Inaddition, we were able to identify a specific neuroendocrine-immune profile associated to specific changes in hippocam-pal BDNF levels. Results suggest a fine modulation of thecrosstalk between central and peripheral pathways of adap-tation and plasticity and that the length of stress exposureis crucial to determine its final outcome on health ordisease.

Allostasis—or “stability through change”—is defined asany neural, neuroendocrine and immune activation leadingto adaptation in the face of stressful challenges [4]. Whilein the short run, activation of these systems is essential tothe maintenance of homeostasis and survival yet, over longertime intervals, it imposes a cost-allostatic load—that canaccelerate disease processes or participate to pathologicalchanges associated, among others, to immunosuppression[4]. When we studied the characteristics of the diversestressors applied for different lengths of times, we found thatupon prolonged exposure to RS (21 days), an increase in theimmunogenic/allostatic load was observed, mirrored by apeak in CORT levels comparable to that observed on day 1.This was associated to a suppression of the immune systemwith decreased levels of the proinflammatory cytokine TNF-𝛼 and increased levels of the anti-inflammatory cytokine IL-10. In addition, a decrease in hippocampal BDNF levels wasfound, suggesting a reduction in the ability to cope withprolonged stress (brain plasticity).

Analyzing more in detail the effects of RS, we foundthat, following 7 days of this procedure, CORT elevationwas significantly lower than on the first stimulation (day1), suggesting an habituation of the system to the chronicprocedure, as previously shown [42]. This effect, whichappears to be mediated by limbic regions [8], is likely tohave consequences for the functioning of a number of GC-sensitive systems, including the immune system. Indeed,reduced CORT levels could disinhibit immune function,leading to a proinflammatory response, as suggested by theincrease in the levels of TNF-𝛼.

A number of evidence support the hypothesis that amoderate increase in the levels of proinflammatory cytokines,such as TNF-𝛼, might result in an overall “priming effect”on the immune system, leading to better abilities to copewith further physiologic/stressful stimuli [21, 43–46]. Worthnoticing, the response to stress observed does not onlyinvolve peripheral targets, but also extends to central medi-ators. In fact, after 7 days of RS, hippocampal BDNF proteinlevels showed a trend towards an increase, possibly reflectinga neuroprotective mechanism. By contrast, after 21 days ofRS, BDNF levels were found to be decreased and this wasassociated to an augmented anti-inflammatory response bythe immune systemwith increased IL-10 levels and a return ofTNF-𝛼 to basal levels. It must be emphasized that while acutechanges in BDNF levels might represent a coping responseto stressful events, and thus being beneficial, prolongedexposure to stressors and increased allostatic load would leadto detrimental effects as reduced BDNF signaling in the adultbrain may be involved in the pathophysiology of psychiatricdisorders [47–49].

A fine regulation of apoptosis might positively affectoptimal immune function. This is achieved by maintaininglymphocyte homeostasis by a continuous removal of cells thathave been activated once they have served their function.Therefore, inappropriate induction of such a mechanismcould result in a variety of pathological effects such asautoimmune diseases, while the maintenance of physio-logically regulated levels of apoptosis might exert a ben-eficial/protective effect [50]. In this context, the observeddecrease in apoptosis levels following 21 days of RS, suggestsa long-term impairment of the immune system response.

Compared to RS, SS resulted in an overall lower responseof the HPA axis as well as of the immune system. Differ-ences were both quantitative and qualitative. In particular,no change in TNF-𝛼 could be detected, while an earlierincrease in IL-10 was observed compared to RS, suggestingan anticipated anti-inflammatory reaction in response to thisspecific stressor. Differently from RS data, splenic apoptosisincreased after 7 days of SS, suggesting that it might representa reliable early stress-sensitive physiological marker.

Taken together, data from this study clearly indicate adifferential role of psychophysical versus social and of briefversus prolonged stress on neuroendocrine and immunefunction suggesting that the quality and the extent ofthe stress period are crucial in determining individualneuroendocrine-immune responses to external challenges.In addition, and more intriguingly, all these peripheralresponses were associated to specific changes in hippocampalBDNF levels. We hypothesize that this neurotrophin mightrepresent a key modulator of neuro-immunoendocrine path-ways, playing a pivotal role in the orchestration/maintenanceof the brain and peripheral plasticity leading to optimalcoping strategies to stressful events [8]. Future studies shouldemploy pharmacological challenges aimed at investigatingsuch interactions.

While this is a first attempt to mimic some of thequalitative and temporal features of “stress,” studies areongoing to extend the range of mediators analyzed andthe peripheral targets, to evaluate more extensively the role

Neural Plasticity 7

of acute versus chronic stressors on neuroendocrine andimmune function. In addition a thorough characterizationof the specific changes occurring in brain plasticity in otherregions involved in neuroendocrine-immune integration willhelp elucidating the mechanisms underlying the benefi-cial/pathological effects of stress increasing the translationalvalue of these studies.


All authors declare that no conflict of interests, financial orotherwise, exists with regard to this research.


S. Capoccia and A. Berry contributed equally to this paper.

Acknowledgments

The authors thank L. T. Bonsignore for technical support.Funding for this study was provided by the ItalianMinistry ofHealth (Ricerca Finalizzata 2008, Fasc10BF to EA), (RicercaFinalizzata 2009, Fasc.12AF to FC), and Fondazione Veronesi2012 to FC.

References

[1] F. Cirulli, N. Francia, A. Berry, L. Aloe, E. Alleva, and S. J.Suomi, “Early life stress as a risk factor for mental health:role of neurotrophins from rodents to non-human primates,”Neuroscience and Biobehavioral Reviews, vol. 33, no. 4, pp. 573–585, 2009.

[2] E. R. De Kloet, R. H. DeRijk, and O. C. Meijer, “Therapyinsight: is there an imbalanced response of mineralocorticoidand glucocorticoid receptors in depression?” Nature ClinicalPractice Endocrinology and Metabolism, vol. 3, no. 2, pp. 168–179, 2007.

[3] K. S. Kendler, L. M. Karkowski, and C. A. Prescott, “Causalrelationship between stressful life events and the onset of majordepression,” American Journal of Psychiatry, vol. 156, no. 6, pp.837–841, 1999.

[4] B. S. McEwen, “The neurobiology of stress: from serendipity toclinical relevance,” Brain Research, vol. 886, no. 1-2, pp. 172–189,2000.

[5] B. S. McEwen, “Protection and damage from acute and chronicstress: allostasis and allostatic overload and relevance to thepathophysiology of psychiatric disorders,” Annals of the NewYork Academy of Sciences, vol. 1032, pp. 1–7, 2004.

[6] F. Eskandari and E.M. Sternberg, “Neural-immune interactionsin health and disease,” Annals of the New York Academy ofSciences, vol. 966, pp. 20–27, 2002.

[7] B. S. McEwen, “Central effects of stress hormones in health anddisease: understanding the protective and damaging effects ofstress and stress mediators,” European Journal of Pharmacology,vol. 583, no. 2-3, pp. 174–185, 2008.

[8] F. Cirulli and E. Alleva, “The NGF saga: from animal models ofpsychosocial stress to stress-related psychopathology,” Frontiersin Neuroendocrinology, vol. 30, no. 3, pp. 379–395, 2009.

[9] L. Tapia-Arancibia, F. Rage, L. Givalois, and S. Arancibia, “Phys-iology of BDNF: focus on hypothalamic function,” Frontiers inNeuroendocrinology, vol. 25, no. 2, pp. 77–107, 2004.

[10] G. S. Griesbach, J. Vincelli, D. L. Tio, and D. A. Hovda, “Effectsof acute restraint-induced stress on glucocorticoid receptorsand brain-derived neurotrophic factor after mild traumaticbrain injury,” Neuroscience, vol. 210, pp. 393–402, 2012.

[11] G. Naert, G. Ixart, T. Maurice, L. Tapia-Arancibia, and L. Gival-ois, “Brain-derived neurotrophic factor and hypothalamic-pituitary-adrenal axis adaptation processes in a depressive-likestate induced by chronic restraint stress,”Molecular and CellularNeuroscience, vol. 46, no. 1, pp. 55–66, 2011.

[12] F. Cirulli, A. Berry, L. T. Bonsignore et al., “Early life influ-ences on emotional reactivity: evidence that social enrichmenthas greater effects than handling on anxiety-like behaviors,neuroendocrine responses to stress and central BDNF levels,”Neuroscience and Biobehavioral Reviews, vol. 34, no. 6, pp. 808–820, 2010.

[13] H. Thoenen, “Neurotrophins and neuronal plasticity,” Science,vol. 270, no. 5236, pp. 593–598, 1995.

[14] R.M. Sapolsky, “Depression, antidepressants, and the shrinkinghippocampus,” Proceedings of the National Academy of Sciencesof the United States of America, vol. 98, no. 22, pp. 12320–12322,2001.

[15] R. S. Duman and L. M. Monteggia, “A neurotrophic model forstress-relatedmood disorders,” Biological Psychiatry, vol. 59, no.12, pp. 1116–1127, 2006.

[16] E. Castren, V. Voikar, and T. Rantamaki, “Role of neurotrophicfactors in depression,” Current Opinion in Pharmacology, vol. 7,no. 1, pp. 18–21, 2007.

[17] F. D. Jeanneteau, W. M. Lambert, N. Ismaili et al., “BDNFand glucocorticoids regulate corticotrophin-releasing hormone(CRH) homeostasis in the hypothalamus,” Proceedings of theNational Academy of Sciences of the United States of America,vol. 109, no. 4, pp. 1305–1310, 2012.

[18] L. Aloe, E. Alleva, A. Bohm, and R. Levi-Montalcini, “Aggres-sive behavior induces release of nerve growth factor frommousesalivary gland into the bloodstream,” Proceedings of the NationalAcademy of Sciences of the United States of America, vol. 83, no.16, pp. 6184–6187, 1986.

[19] W. A. Nockher and H. Renz, “Neurotrophins in clinical diag-nostics: pathophysiology and laboratory investigation,” ClinicaChimica Acta, vol. 352, no. 1-2, pp. 49–74, 2005.

[20] S. Hayley, M. O. Poulter, Z. Merali, and H. Anisman, “Thepathogenesis of clinical depression: stressor- and cytokine-induced alterations of neuroplasticity,” Neuroscience, vol. 135,no. 3, pp. 659–678, 2005.

[21] F. S. Dhabhar, “Enhancing versus suppressive effects of stresson immune function: implications for immunoprotection andimmunopathology,” NeuroImmunoModulation, vol. 16, no. 5,pp. 300–317, 2009.

[22] M. A. Smith, S. Makino, R. Kvetnansky, and R. M. Post, “Effectsof stress on neurotrophic factor expression in the rat brain,”Annals of the New York Academy of Sciences, vol. 771, pp. 234–239, 1995.

[23] G. Barbany and H. Persson, “Regulation of neurotrophinmRNAexpression in the rat brain by glucocorticoids,”EuropeanJournal of Neuroscience, vol. 4, no. 5, pp. 396–403, 1992.

[24] F. Marmigere, L. Givalois, F. Rage, S. Arancibia, and L. Tapia-Arancibia, “Rapid induction of BDNF expression in the hip-pocampus during immobilization stress challenge in adult rats,”Hippocampus, vol. 13, no. 5, pp. 646–655, 2003.

8 Neural Plasticity

[25] B. S. Mcewen and P. J. Gianaros, “Central role of the brain instress and adaptation: links to socioeconomic status, health, anddisease,” Annals of the New York Academy of Sciences, vol. 1186,pp. 190–222, 2010.

[26] B. S. McEwen, C. A. Biron, K. W. Brunson et al., “The role ofadrenocorticoids as modulators of immune function in healthand disease: neural, endocrine and immune interactions,” BrainResearch Reviews, vol. 23, no. 1-2, pp. 79–133, 1997.

[27] S. Chiba, T. Numakawa, M. Ninomiya, M. C. Richards, C.Wakabayashi, and H. Kunugi, “Chronic restraint stress causesanxiety- and depression-like behaviors, downregulates gluco-corticoid receptor expression, and attenuates glutamate releaseinduced by brain-derived neurotrophic factor in the prefrontalcortex,” Progress in Neuro-Psychopharmacology and BiologicalPsychiatry, vol. 39, no. 1, pp. 112–119, 2012.

[28] L. K. Singh, X. Pang, N. Alexacos, R. Letourneau, and T. C.Theoharides, “Acute immobilization stress triggers skin mastcell degranulation via corticotropin releasing hormone, neu-rotensin, and substance P: a link to neurogenic skin disorders,”Brain, Behavior, and Immunity, vol. 13, no. 3, pp. 225–239, 1999.

[29] D. J. Nutt and D. J. Stein, “Understanding the neurobiology ofcomorbidity in anxiety disorders,” CNS Spectrums, vol. 11, no.10, supplement 12, pp. 13–20, 2006.

[30] J. F. Cryan and A. Holmes, “Model organisms: the ascent ofmouse: advances in modelling human depression and anxiety,”Nature Reviews Drug Discovery, vol. 4, no. 9, pp. 775–790, 2005.

[31] A. Berry, V. Bellisario, S. Capoccia, A. Calza, E. Alleva, andF. Cirulli, “Social deprivation stress is a triggering factor forthe emergence of anxiety- and depression-like behavioursand leads to reduced brain BDNF levels in C57BL/6J mice,”Psychoneuroendocrinology, vol. 37, no. 6, pp. 762–772, 2012.

[32] L. R. Frick, M. L. Palumbo, M. P. Zappia, M. A. Brocco, G. A.Cremaschi, and A. M. Genaro, “Inhibitory effect of fluoxetineon lymphoma growth through the modulation of antitumor T-cell response by serotonin-dependent and independent mecha-nisms,” Biochemical Pharmacology, vol. 75, no. 9, pp. 1817–1826,2008.

[33] L. R. Frick, M. L. Barreiro Arcos, M. Rapanelli et al., “Chronicrestraint stress impairs T-cell immunity and promotes tumorprogression in mice,” Stress, vol. 12, no. 2, pp. 134–143, 2009.

[34] R. Avitsur, D. A. Padgett, and J. F. Sheridan, “Social interactions,stress, and immunity,”Neurologic Clinics, vol. 24, no. 3, pp. 483–491, 2006.

[35] R. Avitsur, J. L. Stark, F. S. Dhabhar, K. A. Kramer, and J. F.Sheridan, “Social experience alters the response to social stressin mice,” Brain, Behavior, and Immunity, vol. 17, no. 6, pp. 426–437, 2003.

[36] F. Cirulli, M. L. Terranova, and G. Laviola, “Affiliation in peri-adolescent rats: behavioral and corticosterone response to socialreunion with familiar or unfamiliar partners,” PharmacologyBiochemistry and Behavior, vol. 54, no. 1, pp. 99–105, 1996.

[37] J. B. Panksepp, K. A. Jochman, J. U. Kim et al., “Affiliativebehavior, ultrasonic communication and social reward areinfluenced by genetic variation in adolescent mice,” PLoS ONE,vol. 2, no. 4, article e351, 2007.

[38] M. L. Terranova, G. Laviola, and E. Alleva, “Ontogeny ofamicable social behavior in the mouse: gender differences andongoing isolation outcomes,”Developmental Psychobiology, vol.26, no. 8, pp. 467–481, 1993.

[39] L. Aloe, F. Properzi, L. Probert et al., “Learning abilities, NGFand BDNF brain levels in two lines of TNF-𝛼 transgenic

mice, one characterized by neurological disorders, the otherphenotypically normal,” Brain Research, vol. 840, no. 1-2, pp.125–137, 1999.

[40] C. E. Perandones, V. A. Illera, D. Peckham, L. L. Stunz, and R.F. Ashman, “Regulation of apoptosis in vitro in mature murinespleen T cells,” Journal of Immunology, vol. 151, no. 7, pp. 3521–3529, 1993.

[41] A. H. Wyllie, J. F. R. Kerr, and A. R. Currie, “Cell death: thesignificance of apoptosis,” International Review of Cytology, vol.68, pp. 251–306, 1980.

[42] S. Bhatnagar, R. Huber, N. Nowak, and P. Trotter, “Lesionsof the posterior paraventricular thalamus block habitua-tion of hypothalamic-pituitary-adrenal responses to repeatedrestraint,” Journal of Neuroendocrinology, vol. 14, no. 5, pp. 403–410, 2002.

[43] O.Watters and J. J. O’Connor, “A role for tumor necrosis factor-alpha in ischemia and ischemic preconditioning,” Journal ofNeuroinflammation, vol. 8, p. 87, 2011.

[44] M. J. Ehrke, S. Verstovsek, D. L. Maccubbin et al., “Protectivespecific immunity induced by doxorubicin plus TNF-𝛼 com-bination treatment of EL4 lymphoma-bearing C57BL/6 mice,”International Journal of Cancer, vol. 87, no. 1, pp. 101–109, 2000.

[45] R. van Horssen, T. L. M. Ten Hagen, and A. M. M. Eggermont,“TNF-𝛼 in cancer treatment: molecular insights, antitumoreffects, and clinical utility,”Oncologist, vol. 11, no. 4, pp. 397–408,2006.

[46] M. J. Ehrke, S. Verstovsek, S. K. Pocchiari et al., “Thymic anti-tumor effectors in mice cured of lymphoma by cyclophos-phamide plus TNF-𝛼 therapy: phenotypic and functional char-acterization up to 20 months after initial tumor inoculation,”International Journal of Cancer, vol. 76, no. 4, pp. 579–586, 1998.

[47] C. A. Altar, “Neurotrophins and depression,”Trends in Pharma-cological Sciences, vol. 20, no. 2, pp. 59–61, 1999.

[48] E. Castren, “Is mood chemistry?”Nature Reviews Neuroscience,vol. 6, no. 3, pp. 241–246, 2005.

[49] S. Sen, R. Duman, and G. Sanacora, “Serum brain-derived neu-rotrophic factor, depression, and antidepressant medications:meta-analyses and implications,” Biological Psychiatry, vol. 64,no. 6, pp. 527–532, 2008.

[50] C. P. Reddy Avula, A. R. Muthukumar, K. Zaman, R. McCarter,and G. Fernandes, “Inhibitory effects of voluntary wheel exer-cise on apoptosis in splenic lymphocyte subsets of C57BL/6mice,” Journal of Applied Physiology, vol. 91, no. 6, pp. 2546–2552, 2001.

Noninvasive Strategies to Optimise Brain Plasticity: From ...downloads.hindawi.com/journals/specialissues/672750.pdfNeural Plasticity Guest Editors: Alessandro Sale, Anthony J. Hannan,

Documents