Nicotine exposure during adolescence: cognitive performance and brain gene expression in adult heterozygous reeler mice

ORIGINAL INVESTIGATION

Nicotine exposure during adolescence: cognitive performanceand brain gene expression in adult heterozygous reeler mice

Emilia Romano & Federica De Angelis & Lisa Ulbrich &

Antonella De Jaco & Andrea Fuso & Giovanni Laviola

Received: 28 June 2013 /Accepted: 30 November 2013# Springer-Verlag Berlin Heidelberg 2013

AbstractRationale We have recently reported nicotine-induced stimu-lation of reelin and glutamic acid decarboxylase 67 (GAD67)mRNA expression levels in the brain of heterozygous reelermice (HRM), a putative animal model for the study of symp-toms relevant to major behavioral disorders.Objectives We aimed to evaluate long-term behavioral effectsand brain molecular changes as a result of adaptations tonicotine exposure in the developing HRM males.Methods Adolescent mice (pnd 37–42) were exposed to oralnicotine (10 mg/l) in a 6-day free-choice drinking schedule.As expected, no differences in total nicotine intake betweenWT (wild-type) mice and HRM were found.Results Long-term behavioral effects and brain molecularchanges, as a consequence of nicotine exposure during ado-lescence, were only evidenced in HRM. Indeed, HRM per-severative exploratory behavior and poor cognitive perfor-mance were modulated to WT levels by subchronic exposureto nicotine during development. Furthermore, the expectedreduction in the expression of mRNA of reelin and GAD67 inbehaviorally relevant brain areas of HRM appeared persistent-ly restored by nicotine. For brain-derived neurotrophic factor

(BDNF) mRNA expression, no genotype-dependent changesappeared. However, expression levels were increased by pre-vious nicotine in brains from both genotypes. The mRNAencoding for nicotine receptor subunits (α7, β2 and α4) didnot differ between genotypes and as a result of previousnicotine exposure.Conclusion These findings support the hypothesis of pre-existing vulnerability (based on haploinsufficiency of reelin)to brain and behavioral disorders and regulative short- andlong-term effects associated with nicotine modulation.

Keywords Adolescence . Nicotine . Tobacco . Behavioraldisorders . Cognition . Animal models . Reelin

AbbreviationsGAD67 Glutamic acid decarboxylaseHRM Heterozygous reeler miceBDNF Brain-derived neurotrophic factornAChRs Acetylcholine nicotinic receptorsDNMT1 DNA methyltransferase 1

Introduction

Nicotine is the main addictive and neuroactive compound intobacco smoke and the most commonly abused drug. Tobaccosmoking is highly prevalent in individuals with psychiatricsymptoms (Batel 2000; Salin-Pascual et al. 2003) and, amongthem, the comorbidity with schizophrenia is particularly high.Several hypotheses have been proposed to explain these phe-nomena, including a self-medication theory (Kumari andPostma 2005; Leonard et al. 2007). It is possible that nicotineconsumption is associated with amelioration in a number ofcognitive deficits related to the disorders (D'Souza and

E. Romano :G. Laviola (*)Section of Behavioural Neuroscience, Department of Cell Biologyand Neuroscience, Istituto Superiore di Sanità, Viale Regina Elena299, 00161 Rome, Italye-mail: [email protected]

E. RomanoBambino Gesù Children’s Hospital IRCCS, Rome, Italy

F. De Angelis : L. Ulbrich :A. De JacoDepartment of Biology and Biotechnology Charles Darwin,Sapienza University of Rome, P. le AldoMoro, 5, 00185 Rome, Italy

A. FusoDepartment of Psychology, Section of Neuroscience, SapienzaUniversity of Rome, Via dei Marsi, 78, 00183 Rome, Italy

PsychopharmacologyDOI 10.1007/s00213-013-3388-y

Markou 2012; Kumari and Postma 2005; Levin and Rezvani2007; Ochoa and Lasalde-Dominicci 2007).

The classical pharmacological action of nicotine takesplace directly through the stimulation of the acetylcholinenicotinic receptors (nAChRs), with subsequent modulationof several other neurotransmitters (Wonnacott et al. 1989). Itis important to note that in brain areas of major psychiatricpatients, both the expression and the function of nAChR, inparticular α4, β2 and α7 subunits are reported to be down-regulated (Breese et al. 2000; Freedman et al. 2000; Freedmanet al. 1995; Leonard et al. 2000; Olincy et al. 1997) andgenerally neuropathology of GABAergic systems also occurs(Akbarian et al. 1995; Benes et al. 2007; Guidotti et al. 2000;Guidotti et al. 2005; Lewis et al. 2005). Such pathologicalphenotype is also characterized by a decreased expression ofglutamic acid decarboxylase 67 (GAD67) and increased ex-pression of DNA methyltransferases (DNMTs) analysed bymeasuring messenger levels (Guidotti et al. 2010; Malokuet al. 2011; Ruzicka et al. 2007; Veldic et al. 2007; Zhubiet al. 2009). Current research suggests that DNMT1 overex-pression in the upper cortical layers of GABAergic neurons inthe brains of psychiatric patients is responsible for hyperme-thylation of specific GABAergic gene promoters (Guidottiet al. 2000), including GAD67 and reelin (Veldic et al.2007).

With respect to animal models, heterozygous reeler mice(HRM) are characterized by down-regulation of GAD67 geneexpression in the prefrontal cortex, hippocampus and cerebel-lum (Costa et al. 2001; Liu et al. 2001; Tueting et al. 1999) inassociation with reduced reelin gene expression (D'Arcangeloet al. 1995; Fatemi et al. 2000; Romano et al. 2013; Tuetinget al. 2006). HRM have been proposed as a suitable animalmodel for studying behavioral and cognitive symptoms asso-ciated with major psychiatric disorders (Laviola et al. 2009;Tueting et al. 2006). Indeed, HRM are characterized by be-havioral alterations such as a decreased contextual fear condi-tioning (Qiu et al. 2006) and prepulse inhibition (Tremolizzoet al. 2002; Tueting et al. 1999), increased motor impulsivity(Ognibene et al. 2007) and perseverative behavior (Macriet al. 2010).

A number of reports also indicate that both GAD67 andreelin genes are up-regulated in cortical and hippocampalbrain areas in mice, whereas DNMT1 is down-regulated,following repeated treatment with nicotine (Satta et al.2008). Therefore, changes in mRNA expression levels ofthese genes are established as a model for evaluatingnicotine effects at a molecular level in the brain. In thisline, we recently reported that the reduced gene expres-sion of reelin and GAD67 in HRM was consistentlyrestored to WT levels following acute nicotine injections(Romano et al. 2013).

Exposure to nicotine during the still plastic period of ado-lescence (Adriani et al. 2002a; Laviola et al. 1999) has been

shown to exert long-term consequences on brain, behaviorand function (Adriani et al. 2003, 2004; Slotkin et al. 2008).Most behavioral symptoms typically manifest during earlypubertal period (Woods 1998), therefore we have chosento expose 5-week-old mice to nicotine (Adriani et al.2002a; Laviola et al. 2003). Indeed, the adolescence win-dow is considered a vulnerable and "critical" ontogeneticperiod for fine-tuning of neural structures (Adriani andLaviola 2004; Brenhouse and Andersen 2011; Laviola andMarco 2011; Spear 2000b). Furthermore, human adoles-cents usually undergo their first encounter with mostpsychoactive compounds around this specific develop-mental phase (Kandel and Chen 2000; Laviola et al.1999; Spear 2000a). In this report, we extend our previousinvestigations on adolescent mice and nicotine (Adrianiet al. 2004; Adriani et al. 2002a) to the HRM model ofgenetic vulnerability.

Adult HRM and their WT littermates were assessed forlong-term adaptations, derived from developmental nico-tine exposure, based on the levels of environmental explo-ration and behavioral disinhibition and for cognitive per-formance. At the same time we were also interested inverifying brain gene expression changes at the messengerlevel derived from nicotine exposure during adolescence.It is known that aberrant regulation of brain-derived neu-rotrophic factor (BDNF) gene has been implicated in theaetiology and pathogenesis of several cognitive disorders,including schizophrenia (Ikeda et al. 2008; Toyooka et al.2002; Weickert et al. 2003, 2005). Since it is also knownthat epigenetic mechanisms dynamically regulate BDNF(Lu and Martinowich 2008), we have analyzed the effectsof nicotine exposure on BDNF expression (Romano et al.2013). Moreover, reelin is a direct effector of BDNFduring brain development (Ringstedt et al. 1998) andBDNF regulates GABAergic function and induces theexpression of GAD67 in neurons (Arenas et al. 1996;Yamada et al. 2002). Within this context, it has beenrecently reported that reelin and GAD67 gene expressionundergo a short-term modulation in HRM specific brainareas after acute nicotine treatment (Romano et al. 2013).We have observed a marked reduction in reelin andGAD67 mRNA expression in the prefrontal cortex, hippo-campus, cerebellum and striatum of control HRM andmeasured a specific increment of two mRNA in responseto nicotinic treatment. Here we report investigations onlong-term adaptations on the mRNA expression levels ofBDNF, reelin and GAD67 in adult mice with a history ofexposure to nicotine during adolescence. Furthermore,since developmental exposure to nicotine might affectdensity of neuronal nicotinic receptors (Flores et al.1992; Marks et al. 1983; Schwartz and Kellar 1983), wehave also compared the mRNA levels of the single recep-tor subunits in HRM and WT mice.

Psychopharmacology

Materials and methods

Subjects

B6C3Fe heterozygous female and wild-type male mice, orig-inally purchased from Jackson Laboratories (USA) were bredin our laboratory. Two females (12 females in total) werehoused with one male in Plexiglas boxes (33×13×14 cm),with sawdust bedding and a metal top. Animals were providedwith tap water and food pellets (Mucedola, Settimo Milanese,Italy) ad libitum. Mice were maintained on a reversed 12:12 hlight/dark cycle (lights on at 18:30 h). Room temperature wasmaintained at 21±1 °C with a relative humidity 60±10 %.After ca. 2 weeks, the male was removed and dams werehoused individually and checked daily for delivery. The dayof birth was the postnatal day (pnd 0). At weaning (pnd 25),offspring were housed in pairs according to sex and genotype.Only male offspring were used in this study. The same micewere assessed in all behavioral tests (see Fig. 1); however, forsome experimental paradigms we excluded some animalsfrom the analysis as they were outliers (outside of the rangemean±3 SD). Adult tests were performed through a schedulethat minimized test battery carryover effects and maximizedthe information obtained by a single individual. Additionally,the test characterized by the lowest invasiveness was per-formed first and the other was performed after a long periodof time. All tests were performed under red light. All proce-dures were performed according to European Communitiesguidelines (EC Council Directives 86/609/EEC and 2010/63/EU), Italian national laws on animal experimentation (DecretoL.vo 116/92) and formally approved by Italian Ministry ofHealth. All efforts were made to take care of animal welfareand to minimize animal suffering, to reduce the number ofanimals used, strictly following the 3R' principles.

Genotyping

Mice genotype was determined from tail samples at weaningusing the PCR genotyping protocol previously described(Laviola et al. 2006).

Experimental procedure for nicotine exposure

Mice were allowed to self-administer nicotine solution duringadolescence (pnd 37–42). The temporal window was selectedaccording to Adriani et al. (2002a), where mice of the sameage were characterized by the absence of a clear profile ofeither preference or aversion for nicotine oral consumption, asassessed in a free-choice drinking schedule.

Mice were adapted to a 22-h water-restriction schedule,during which the water bottle was removed from the homecage, while food pellets were provided ad libitum. Micelearned to have their regular daily fluid intake within the

time-constraints allowed by the schedule. For the daily 2-hdrinking session (The access to water/nicotine solution wasallowed from 9:30 to 11:30A.M. only), mice were gentlyplaced individually in drinking cages of the same size andshape as their home cage. In each cage, there were two bottles(see below). Food pellets were available on the floor (Adrianiet al. 2002a). The weights of the animals were monitored eachday for the possible effects of water restriction and consump-tion of nicotine. In order to measure each animal's fluid intake,bottles were weighed before and after the drinking session.Bottles were filled with fresh solution each day. The weightloss due to evaporation was calculated from two identicalbottles placed in an empty cage ("blank"), and subtracted fromthe bottle weight.

For both genotypes, mice were grouped in "Water" and"Nicotine treated". Animals from the Nicotine group weregiven (2 h/day for 6 days) a choice for drinking from twobottles containing either tap water or a nicotine solution(10 mg/l). Mice in the Water group were exposed to the same2-h daily drinking schedule, with the difference that bothbottles contained only tap water. To avoid a bias representedby potential preferences for either the right or the left side ofthe cage, the position of the nicotine bottle was regularlyreversed as previously reported (Adriani et al. 2002a, b).

Solutions

Nicotine drinking solution was prepared by dissolving nico-tine hydrogen tartrate (Sigma, USA) in tap water and adjustedto the pH of drinking water (pH 7). The drug concentration(10 mg/l expressed as nicotine base) was selected on the basisof literature data (Adriani et al. 2002a,b).

Design and data analysis

Data were analyzed by two way ANOVA, using Stat Viewsoftware (version 4.0). The general model was a 2 genotypes(WT vs. HRM) × 2 solutions (Tap water vs. Nicotine solution)× day. An "inversion" factor (before vs. after the bottle posi-tion was reversed) was also included. For the light/dark, thedependent variables in the ANOVA were time spent in eachcompartment, number of crossing, latency to enter the darkcompartment. For the response to an object in the open-fieldtest, the dependent variables in the ANOVA were crossing,rearing and percent of time spent exploring the object duringthe 5 min test. For the Hole-board test, latency and number ofholes explored were used as the dependent variable. For thespontaneous alternation in the T-maze task, the sequence ofentries and the latency before the first entry were used as thedependent variable. The percent alternation was calculated asratio (actual alternation/possible alternation)×100. For the ac-quisition task in the T-maze, the level of acquisition (allcorrect answers as dependent variables) was expressed as

Psychopharmacology

percentage of correct responses considering first and last daysas repeatedmeasure. For the molecular evaluation, the relativeamount of specific cDNA, for the different mRNA evaluated,were considered as dependent variables. Post hoc compari-sons were performed using Tukey's test, which can also beused in the absence of significant ANOVA results (for details,see Wilcox 1987).

Behavioral evaluation

Light/dark test

Two weeks after the end of nicotine self-consumption, adultWTand HRMmice were assessed for anxiety-like behavior inthe light/dark paradigm (De Filippis et al. 2010). Tests wereconducted between 9A.M. and 5P.M. Mice were individuallyplaced inside the experimental apparatus consisting of anopaque Plexiglas rectangular box with smooth walls,subdivided into two compartments (20×14×27 cm each).One compartment was white and brightly illuminated(100 lx), whereas the other one was black and topped toprevent the light from entering. The floor of the apparatuswas cleaned with 50% diluted ethanol after each testing. Micewere placed into the center of the white, brightly lit area. Thefollowing measures were recorded: (1) latency to enter thedark compartment, (2) number of crossings in the light com-partment, (3) time spent in each compartment. The behavioralprofile expressed by each animal was subsequently scored bya trained observer blind to the genotype of the mice, using acomputer and specific software (THE OBSERVER v2.0 forDOS; Noldus Information Technology, Wageningen, Nether-lands). The whole session lasted 9 min for each mouse.

Response to an object in the open-field test

Locomotor behavior of mice was evaluated in the open-fieldtest, placing each individual animal in the centre of a squaredarena (40×40×40 cm), around 20 days following nicotineexposure. After 10 min of free exploration, an object

(rectangular box, 2×3×7 cm) was inserted in the centre ofthe arena and the animals' exploratory/emotional responsewas measured for further 5 min. All sessions were recordedand videotapes were scored using ethological software (THEOBSERVER v2.0 for DOS; Noldus Information Technology).Behavioral parameters included number of crossings, whichrepresents a measure of horizontal locomotor activity, obtain-ed from the total number of line crossings with both forepaws,and the frequency of rearing of mice on the hind paws. Duringthe second part of the session, time spent in the object squarewas measured (Ognibene et al. 2007). After each test, the floorwas cleaned with 50 % diluted ethanol.

Hole-board test

Mice were subjected to a hole-board test to assess levels ofexploratory behavior. Tests were carried out under red lightfrom 10A.M. to 1P.M. , around 30 days following nicotineexposure. The hole-board consisted of a square platform(40×40 cm), containing 4×4 equally spaced (7 cm) holes, each1.5 cm in diameter. The mice were placed individually on theplatform and the latency and the number of holes explored(head dips), in 10 min was recorded. A head dip was scored ifthe animal's head entered the hole at least up to eye level.Furthermore the number of different holes visited was alsorecorded (Calamandrei et al. 1996; Moy et al. 2008).

Spontaneous alternation in the T-maze task

Working memory was assessed by means of the variant of theT-maze spontaneous alternation procedure (Hughes 2004;Lalonde 2002), 50 days following nicotine exposure. Briefly,the T-maze apparatus consisted of two short arms (21×8 cm)and one long arm (25×8 cm). Without previous habituation,each mouse was subjected to two trials per day for 4 days.Each mouse was individually placed at the centre of long armof the T maze and given the choice to turn right (R) or left (L)for a single 2-min period (cut off). The animals were removedimmediately after their first choice. The sequence of entries

Fig. 1 Experimental design

Psychopharmacology

(all four paws into a given arm) and the latency before the firstentry were recorded. Spontaneous alternation was defined as achoice to enter two different arms on the same day. Thepercentage alternation is the ratio (actual alternation/possiblealternation)×100.

Acquisition task in the T-maze

All mice were gradually introduced to a food restriction reg-imen for 7–10 days, until they had reached 85 % of free-feeding weight. After 4 days of habituation, animals weresubjected (at around 120 days following nicotine exposure)to five daily trials (one session per day for 10 consecutivedays), during which they were allowed to explore the T-mazeand obtain rewards (a small piece of cereal flake). Mice werethen given time to consume the reinforcing pellet. The rein-forced arm was on the left side for half of the mice, and on theright side for the other half. Mice were returned to theirhome cage after the session for a short interval of 1 min.A repeated association between the arm and the properreward was thus established over the days. The level ofacquisition was evaluated considering all correct answers.For each mouse, one arm was designated as the correctarm.

Molecular evaluation from brain samples

Real-time PCR analysis

RNA was extracted from homogenized dissected brain re-gions with the RNeasy Lipid Tissue mini kit (Qiagen,Milano Italy) following instructions of the manufacturer;1 μg of total RNA was used for cDNA synthesis, and1 μg of total cDNA was used for each real-time reactionin triplicate for each sample as previously described (Ricceriet al. 2011) at an annealing temperature of 60 °C. Amplifi-cation efficiency for each primer pair was determined byamplification of a linear standard curve (from 0.16 to100 ng) of total cDNA as assessed by A260/A280 spectro-photometry. Standard curves displayed a good linearity andamplification efficiency (90–99 %) for all primer pairs.Primers sequences were previously published: BDNF(Zaheer et al. 2006), reelin and GAD67 (Kundakovic et al.2007) and β-actin (Fuso et al. 2008). Total cDNA amountswere standardized by normalization to the β-actin controland presented as the fold-increase over control samples (WTTap water). Expression levels of the genes of interest werealso normalized to the reference genes GAPDH and 18Swith similar results (data not shown). Messenger levels forα7, β2 and α4 nicotinic subtypes between WT mice andHRM were also analysed using the primer sets reported inTable S1.

Results

Body weight

Body weight measurement was used to monitor changesrelated to genotype and nicotine exposure. No significantchanges were found (F (1,21)=0.025, NS; WT tap water15.39±0.44, HRM tap water 14.23±0.80, WT nicotine solu-tion 16.35±0.31, HRM nicotine solution 15.58±0.33) (datanot shown).

Total fluid intake

ANOVA statistical analysis indicated that the mice of bothgenotypes either fromWater or Nicotine groups (seeMaterialsand methods), drank as a whole a comparable amount of fluid(F (1,21)=1.490, NS) (96.012±3.98; 91.162±2.17, value ofmean of total fluid intake (ml/kg); nicotine contained in thepercentage of total fluid intake over the course of the study 51±2 % and 56±2 % for WT and HRM, respectively).

As expected, on the basis of previous work mice assessedin the middle adolescence (age range 37–48 years in Adrianiet al. 2002a) did not exhibit a preference for the nicotinesolution over tap water (drug: F (1,15)=2.801, NS). Further,no significant differences emerged (F (1,15)=1.032, NS) be-tween WTand HRM mice for the mean of nicotine consump-tion over the course of 6 days expressed as nicotine dose (mg/kg) (0.908±0.083 and 1.031±0.065, respectively).

Long-term consequences of developmental nicotine exposure:behavioral profile

Light/dark test

No significant differences in the behavioral response to thelight/dark test were found between adult mice from twogenotypes or as a function of previous drug exposure. Inparticular, HRM spent a similar amount of time as the corre-sponding WT control mice in the light, intimidating compart-ment of the apparatus (F (1,21)=1.289; NS) (WT 105.89±7.39, HRM 121.25±6.47). Not even previous drug exposurehas changed this profile (F (1,21)=0.148; NS) (Tap water110.25±9.31, Nicotine solution 116.49±5.79). Mice of thetwo genotypes and treatment groups did not differ for activityrate measured within the light compartment (F (1,21)=1.288and F (1,21)=0.299 [NS], respectively) (WT 9.82±0.96,HRM 7.64±1.06 and Tap water 9.417±1.63, Nicotine solu-tion 8.21±0.77). No significant genotype (WT 100.9±26.14,HRM 103.0±24.74) or treatment effects (Tap water 100.7±24.74, Nicotine solution 100.9±23.66) were found for latencyto enter the dark compartment (F (1,21)=0.003 and F(1,21)=0.010 [NS], respectively).

Psychopharmacology

Response to an object in the open field test

The number of crossing in the arena and rearing episodeswere not expressed with different frequencies between the twogenotypes and treatment groups (NS). When considering theapproach to an object placed in the centre of the arena, in theabsence of simple main effects of genotype (F (1,21) = 2.308;p =NS) (WT 60.11±5.41, HRM 64.49±4.90) or drug(F (1,21) = 0.259; p =NS) (Tap water 65.45±7.65, Nicotinesolution 61.21±3.99), a significant genotype by drug interac-tion appeared (F(1,21)=6.116; p <0.05). Post hoc analysisrevealed that adult HRM control mice showed persistentexploration of the object compared to corresponding WTlittermates (see Fig. 2a). Furthermore, in the absence of effectsin theWTmice, developmental nicotine exposure was respon-sible for a significant reduction of time devoted to objectexploration (p <0.05) by HRM. As a consequence, HRMreached the performance typical of WT mice.

Hole-board test

An additional measure of exploratory behavior was achievedby testing adult mice in a Hole-board apparatus. ANOVAcarried out on total number of visits (head dips) yielded amain effect of genotype (F (1, 20) = 3.987; p <0.05). As awhole, a reduction of exploratory behavior characterizedHRM mice, which visited a lower number of holes than WTlittermates (WT 6.20±1.29, HRM 2.857±0.90). No signifi-cant effects were found for drug exposure (F(1, 20) = 1.074;NS) (Tap water 5.62±1.58, Nicotine solution 3.56±0.92). Theanalysis of latency to first head dip yielded a main effect ofdrug (F(1, 20) = 4.206; p <0.05), with developmental nicotineexposure being associated as a long term effect, with a con-sistent increment of this interval bymice from both genotypes.Only a trend for genotype by drug interaction appeared in theANOVA (F(1, 20)=3.396; p =0.08). In particular, post hocanalysis (see Materials and methods) revealed that in theabsence of consequences in WT group, HRM exposed tonicotine during adolescence appeared associated with an in-creased latency compared to corresponding HRM controls(see Fig. 2b).

Spontaneous alternation in a T-maze task

No significant consequences of either genotype (WT 3.27±0.24, HRM 3.28±0.16) or drug exposure (Tap water 3.12±0.23, Nicotine solution 3.35±0.17) were found for spontane-ous alternation in the T-maze (F (1,21)=0.049, p =NS andF(1,21)=0.507, p =NS, respectively). However, when con-sidering the latency to fist entry, in the absence of genotype bydrug interaction (F (1,15)=0.10, NS) (see Fig. 2c), a signifi-cant main effect of genotype and of previous drug exposure(F (1,15) = 5.557, p <0.05 and F (1,15) = 4.879, p <0.05,

respectively) was evidenced. Specifically, HRM were signif-icantly faster than WT mice (WT 22.21±2.79, HRM 12.75±1.92). With respect to consequence of drug exposure, micefrom both genotypes exposed to nicotine were also muchfaster than the control group (Tap water 24.04±3.89, Nicotinesolution 14.09±1.66).

Acquisition task in a T-maze

Adult mice were tested for strength of cognitive association ina T-maze task. The results of this analysis based on thepercentage of correct choices of association between the firstand the last days of the schedule, are illustrated in Fig. 2d. Onthe first day of testing, the HRM control subject (tap waterduring adolescence) was the only group to exhibit a poorperformance compared to that of all the other groups. On thelast day of testing, all groups improved in performance. TheANOVA revealed a significant genotype × drug interaction(F (1,19) = 6.192; p <0.05). Specifically, as shown in the insetin Fig. 2d, control HRM exhibited a significantly reduceddegree of reward association than corresponding WT mice.Notably, nicotine exposure during adolescence rendered allperformances (percentage of correct choices) of HRM sub-jects indistinguishable from that of WTcontrols (for post hoc,see Fig. 2d). To provide further support to these conclusions,statistical analysis of the discrimination ratio [(correctresponses/total responses)] was also performed to assesswhether the experimental groups showed a preference forone of the two arms that was above chance (chance = discrim-ination ratio of 50 in both the first and the last days of theschedule). Such analysis confirmed that none of the groupsshowed a significant preference for one of the two arms on thefirst day of testing (WT Tap water: t(3) = 1.225, NS; HRMTap water t (3) = −1.225, NS; WT Nicotine solution: t (5) =1.085, NS; HRM t (8) = 1.344, NS). By contrast, on the lastday of testing, only WT controls and HRM that receivednicotine in adolescence showed a preference for the baitedarm which was significantly different from the chance level(WT controls: t (3) = 6.928, p <0.005; HRM Nicotine solu-tion: t (8) = 5.330, p <0.001).

Molecular analysis

Semiquantitative real-time PCR was used to analyse reelin,GAD67 and BDNF transcript levels from the cortices andhippocampi of fully adult WT and HRM subjects exposed tonicotine during adolescence. Notably, these parameters werepreviously evaluated by our and other groups as a sufficientmarker of regulatory response to nicotine exposure (Romanoet al. 2013; Satta et al. 2008).

As shown in Fig. 3a and b, ANOVAyielded the significantmain effect as genotype for reelin and GAD67 gene expres-sion both in cortex (F(1,15) = 4.75, p <0.001 [WT 1±0.082,

Psychopharmacology

HRM 0.64±0.054] and F(13,15) = 6.62, p <0.001 [WT 1±0.049, HRM 0.59±0.073], respectively), and hippocampus(F (1,15) = 5.36, p <0.001 [WT 1±0.071, HRM 0.58±0.047] and F(1,15) = 5.08, p <0.001 [WT 1±0.051, HRM0.62±0.069], respectively). As expected, reelin and GAD67mRNA levels were reduced in HRM subjects compared toWT. As previously observed, no genotype effects were foundfor BDNF messenger expression (Fig. 3c).

Significant genotype × treatment interaction was evidentsince nicotine exposure during adolescence resulted in persis-tent up-regulation of reelin and GAD67 gene expression inHRM brains. Post hoc analysis revealed that in the absence ofchanges inWT group, HRM appeared specifically sensitive tonicotine effects reaching WT-like reelin and GAD67 mRNAlevels. Specifically, a significant effect was found in the cortexand hippocampus for reelin (cortex: F (1,15) = 12.44,p <0.001; hippocampus: (F (1,15) = 10.03, p <0.001) andGAD67 (cortex: F (1,15) = 11.71, p <0.001; hippocampus:(F (1,15) = 10.34, p <0.001). A main effect of treatment wasfound for BDNF expression, showing nicotine dependent up-regulation both in WT mice and HRM (cortex: F (1,15) = 4,p <0.001 [Tap water 1±0.053, nicotine solution 1.30±0.058];hippocampus: F(1,15) = 3.27, p <0.001 [Tap water 1±0.065,nicotine solution 1.33±0.06]).

Study of nicotinic receptor subunits

Brain mRNA expression of the nicotinic receptor subunitsα7,β2 and α4 was also analysed by real-time PCR in the samebrain regions from both genotypes and treatment groups. The

results obtained showed that the expression of the subunitsα7, β2 and α4 of the nicotine receptor did not differ betweenWTand HRM in the prefrontal cortex, hippocampus, striatumand cerebellum (data not shown). Moreover, no changes in themRNA expression for the subunits were found as a function ofdevelopmental nicotine exposure. Only a positive trend(higher expression) of variation in the β2 receptor mRNAexpression emerged in the cortex of HRMwith respect to WTcontrols. That, however, was not significant at the ANOVAstatistical analysis (F (1,12) = 3.922, p =0.0711) (WT 0.002±3.461E−4, HRM 0.002±1.729E−4).

Discussion

We provide here first evidence for differential long-term mod-ulation following nicotine exposure during adolescence inHRM, an animal model validated for studying symptomsrelevant to major neuropsychiatric disorders (Laviola et al.2009). We report evidence for a persistent rescue of bothemotional and cognitive profiles and of brain genes expres-sion in the HRM as a function of developmental nicotineexposure. Specifically, a modulation to WT levels wasachieved, in the following tasks: (1) perseverative exploratorybehavior, (2) disinhibition profiles in the hole-board test, (3)incorrect responses in a cognitive task. Nicotine exposureduring adolescence also persistently increased BDNF mRNAexpression in both genotypes and restored genotype-dependent deficits in reelin and GAD67messenger expressionlevels in the HRM cortex and hippocampus.

Fig. 2 a Response to the object(drug-free state). Percentage oftime spent exploring the object byadult WT and HRM. b Hole-board test (drug-free state). La-tency to first head-dipping bymice as in panel a. c Spontaneousalternation in a T-maze task (drug-free state). Latency to the firstentry in one of two arms of a T-maze apparatus by mice as inpanel a. d Acquisition task (drug-free state), by mice as in panel a .The graph shows the percentageof correct responses in the firstand last days of the test. The insetshows the same data pooled asinteraction between genotype ×drug exposure. Data are mean±SEM. Significant difference*within groups and $betweengroups.

Psychopharmacology

We did not observe significant changes in the basal expres-sion of receptor nicotine subtypes (nAChR) by relative quan-tification of HRM and WT messenger levels, or as a functionof previous drug exposure. Previous studies have alreadyexamined long-term effects of nicotine treatment administeredduring adolescence, with reported evidences of molecularchanges at receptor level (transcripts and/or proteins, respec-tively) in brain areas (Adriani et al. 2003; Slotkin et al. 2007b),possibly mediated by nicotine binding to its receptors.

We recently reported selective modulation of behavior andbrain genes expression in the HRM model, following acutetreatment with nicotine (Romano et al. 2013). This modula-tion is suggested to be mediated by epigenetic mechanismssince nicotine or other synthetic nAChR agonists (nicotineanalogues) are reported to transiently inhibit DNMT1, thusactivating genes expression in the mice brain with possibleinfluences on behavior (Satta et al. 2008; Veldic et al. 2007).

As previously outlined, we also investigated possible dif-ferences in motivation to nicotine consumption in mice fromboth genotypes. Nicotine has been reported to act on meso-limbic and nigro-striatal neurons projecting to striatum,

amygdala, and prefrontal cortex (Picciotto et al. 1998). Acutetreatment with this drug is reported to modulate anxiety,behavioral disinhibition, reward and habit-forming processes.Indeed, there is evidence that HRM present changes inmesolimbic DA transmission (Ballmaier et al. 2002) whichmight be expected to support a differential response to drugeffects. To this aim, we adopted a consolidated free-choicedrinking paradigm (Schneider et al. 2012; Zimmerberg andBrett 1992). However, also confirming previous work onmiceof the same age (adolescence) (Adriani et al. 2002a), nodifferences were found on the total intake of nicotine solution.

Despite a similar amount of nicotine ingested during ado-lescence, long-term behavioral and molecular effects of thisregimen of drug exposure were only found in HRM. Micefrom both genotypes were investigated at adulthood for theirexploratory and cognitive behavior in a drug-free state as afunction of previous nicotine exposure. HRM subjects werecharacterized by a much faster approach to the object in theopen field and by a longer time spent in contact with it, incomparison to WT littermates, suggesting a perseverativeprofile and behavioral disinhibition. Present data is in

Fig. 3 Real-time PCR analysis ofreelin. Reelin mRNA expressionin cortex (a) and hippocampus(b) of WT and heterozygousreeler mice exposed or not duringadolescence to nicotine (Nicotinesolution). Real-time PCR analysisof GAD67. GAD67 mRNA ex-pression in cortex (c), hippocam-pus (d). Real-Time PCR analysisof BDNF. BDNF mRNA expres-sion in cortex (e), hippocampus(f). Data are mean±SEM. *Sig-nificant difference within groupsand & between groups

Psychopharmacology

agreement with previous literature on this mouse model(Krueger et al. 2006; Podhorna and Didriksen 2004;Salinger et al. 2003), even if contrasting results are alsoavailable (Tueting et al. 1999). A trend in support of a generaldisinhibition profile was also observed in the hole-board test.Indeed, HRM were characterized by shorter latency to firsthead-dipping than WT littermates. Pharmacological assayshave indicated that head-dips can be increased by treatmentwith diazepam or other anxiolytic drugs, and decreased byanxiogenic compounds (Kliethermes and Crabbe 2006;Takeda et al. 1998).

HRM appeared much faster than WT littermates in thelatency to first turn in one arm of a T-maze. This occurredhowever in absence of basal changes for levels of spontaneousalternation, which has been associated with functions at thelevel of the prefrontal cortex and hippocampus (Lalonde2002). As a whole, developmental nicotine affected the adultperformance in a drug-free state. Indeed, subjects of bothgenotypes with a history of nicotine exposure exhibited afurther reduction in their latency. With respect to cognitiveperformance, HRM took much longer to learn in the acquisi-tion task in comparison to WT mice, with HRM reaching aconsistently low number of correct responses. This paradigmhas been designed to recapitulate in rodents important ele-ments of the working memory task (Aultman andMoghaddam 2001) which is thought to be centred on thehippocampus and prefrontal cortex (Chudasama andRobbins 2006). Moreover, impairment in this task has beenlinked to altered dopamine transmission in rodents prefrontalcortex (Aultman and Moghaddam 2001; Kellendonk et al.2006). In a long time range, nicotine exposure during adoles-cence persistently improved the performance of HRM to reachpercentage levels of WT control mice.

These behavioral results were partially corroborated bygene expression data in behaviorally relevant brain areas,according to the analysis reported in previous literature andin our recent work (Liu et al. 2001; Romano et al. 2013;Tueting et al. 1999): a genotype-related marked reduction inbasal reelin and GAD67 genes expression was evidenced inthe prefrontal cortex and hippocampus of untreated HRMwithrespect to WT littermates. Here we report the first evidence ofpersistent elevation of reelin and GAD67 mRNA levels inbrains of adult HRM as adaptation to nicotine exposure duringadolescence. This effect was specifically dependent on themice genotype, since no significant changes emerged in WTsubjects. It can be hypothesized that exposure to nicotine canregulate the activation of reelin and GAD67 expression in theHRM that are normally expressed at low levels. According toexisting evidence, we previously reported that acute nicotineadministration resulted in DNMT1 inhibition and subsequentspecific up-regulation of gene expression in cortical (Veldicet al. 2007) and hippocampal regions in mice (Romano et al.2013; Satta et al. 2008). In our case, nicotine exposure,

occurring during adolescence, a still plastic developmentaltime window (see Introduction), resulted in persistent changesdue to drug-induced organizational events on specific genepathways. These findings raise the attention of an emergingliterature that describes the lasting effects of adolescent nico-tine exposure on cognitive and motivational systems in pre-clinical and clinical reports (Dao et al. 2011; Goriounova andMansvelder 2012).

A different mechanism of regulation should be hypothe-sized for the expression of the BDNF, since this gene showssimilar basal expression in WT and HRM in prefrontal cortexand hippocampus, in agreement with published results(Ognibene et al. 2008, Romano et al. 2013). Indeed, BDNFgene expression in selected brain areas was shown to besimilarly modulated by acute nicotine in both genotypes. Thisprofile appears in line with the modulatory effect exerted bynicotine treatment, since it is well known that BDNF expres-sion is dynamically modulated by epigenetic modifications(Cowansage et al. 2010; Roth and Sweatt 2011). Nicotineexposure during adolescence was also responsible for a per-sistent increment of BDNFmRNA in brain areas frommice ofboth genotypes. Zhang and colleagues (Zhang et al. 2002)suggested that the increase of BDNF levels might be associ-ated with enhanced neural plasticity and the reported effects ofnicotine (Picciotto and Zoli 2008; Singh et al. 2004) onattention and cognitive capacities (Singh et al. 2004). In ourexperiment, it is possible that BDNF up-regulation results inbehavioral effects only in HRM (only spontaneous alternationlatency was modified in both genotypes) due to their impairedbehavioral phenotype; our hypothesis is that in WT mice, theeffects of BDNF up-regulation do not appear to largely affectthe basal behavioral profile.

Keeping into account previous literature on the conse-quences of developmental nicotine administration (Adrianiet al. 2003; Counotte et al. 2012; Levin and Rezvani 2007;Miao et al. 1998; Slotkin et al. 2007b), this study investigatedlong-term adaptations at the mRNA level for the cholinergicnicotinic receptor subunits. It is reported that neuronal nico-tinic receptors containing α4 and β2 subunits increase fol-lowing exposure to nicotine (Flores et al. 1992) in culturedneurons expressing these receptors (Pacheco et al. 2001), bothfrom rat (Schwartz and Kellar 1983) and mouse brains (Markset al. 1983). Adriani and colleagues (2003) reported that dailynicotine administration for 10 days, during pre-puberty,caused an increase in α5, α6, β2 gene expression in brainareas, which was evident even after the adolescents had growninto fully adulthood. Furthermore, prenatal or adolescent nic-otine exposure has been reported to promote alterations inreceptors and cholinergic function that persist well into adult-hood (Slotkin et al. 2007a,b).

Since low expression of brain nicotinic receptors has beenreported in psychiatric patients (Albuquerque et al. 2009) andcontinued nicotine consumption is reported to increase the

Psychopharmacology

brain expression of the high-affinity nAChRs (Breese et al.2000), we expected to replicate some of these findings in theHRM, as a model for major neuropsychiatric symptoms. Onthe contrary to what was expected, no significant genotype-dependent effects emerged in the expression analysis of nico-tinic receptor subunits in any of the selected brain regions.Moreover, no significant outcomes from nicotine exposureemerged in the regulation of receptor subunits in any of thetwo genotypes. It is accepted that nicotine-induced up-regulation of native nAChRs results from multiple processessuch as conformational changes and subunit endoplasmicreticulum (ER) degradation (Govin et al. 2012). This up-regulation is dose dependent (Marks et al. 1983; Rowell andLi 1997) and is generally not associated with an increment inthe expression of mRNAs encoding for various nAChR sub-units (Marks et al. 1992). However, increased levels of sub-units' messenger have also been reported in adult ratspretreated with nicotine during adolescence (Adriani et al.,2003), as mentioned above. Furthermore, there is evidencethat nicotine treatment from the early gestation caused asignificant increment of α2, α4, α7, and β2 subunits mRNAsin the foetal forebrain and hindbrain (Lv et al. 2008). Based onthese observations, we measured mRNA levels for thosereceptor subunits also to correlate the results to the dataobtained on reelin, GAD67 and BDNF mRNA analysis. Ourresults did not evidence a transcriptional regulation for any ofthe receptor subunits. This can possibly be ascribed to the timeinterval chosen for the analysis (around 180 days after nico-tine exposure), which may have influenced the biochemicalmeasurements in some way. Actually, present data appear tobe in agreement with a report showing that chronic nicotinetreatment does not affect mRNA expression of nAChR sub-units (Marks et al. 1992; Saito et al. 2005). However, wecannot exclude a short-term modulation in the expression ofthe receptor subunits by the repeated exposure to nicotine. Ourdata also do not exclude the possibility of different regulationmechanisms of the receptors number in the HRM and in theWT mice.

Conclusion

We reported a selective sensitivity to nicotine modulatoryeffects on behavior and brain gene expression in HRM. Thelatter represent a validated mouse model for symptoms rele-vant to major neuropsychiatric disorders (Laviola et al. 2009).HRM and WT mice were exposed to similar nicotine dosageduring the still "plastic window" of adolescence. Notably,quite different outcomes between the two genotypes wereevidenced at brain and behavioral levels in the long term.Indeed, HRM administered with nicotine showed persistentWT-like levels of adult cognitive and exploratory performance

and reelin and GAD67 gene expression in behaviorally rele-vant brain areas.

We are not aware of any studies on adolescent smokers atrisk for major neuropsychiatric illness and possibly of follow-up in adulthood. We can add however the evidence recentlyprovided by Ross and colleagues (2013). These authors in-vestigated the effects of perinatal choline supplementation onthe development of cerebral inhibition in human infants. It isworth noting that besides other general effects, choline isreported to act as a selective agonist at nicotinic α7 subunits(Albuquerque et al. 1998; Ricceri et al. 2011). In this frame-work, Rapoport (2013) commented that so far “Improvinginfant brain inhibition is not prevention of schizophrenia”but “it represents a landmark proof of concept showing thatsuch an approach might be possible”.

As a whole, our results also support the suitability of theheterozygous reeler mouse model for the study of the interac-tion between genetic vulnerability (haploinsufficienty ofreelin and GAD67) and environmental modulation thereon(as it was studied here by means of nicotine).

Acknowledgements We are grateful to Giovanni Dominici for animalcare. E.R. is supported by a fellowship from the ERAnet "PrioMedChild",ItalianMinistry of Health (P.I.Walter Adriani). The authors are grateful toDr. Saira Shamsi for critical reading of the manuscript. This research wassupported by IRE-IFO (RF2008) "MECP2 phosphorilation and relatedkinase in Rett syndrome"to GL, E.R. is recipient of a Postdoctoralfellowship under the "NeuroGenMRI" project in the framework ofERAnet "PrioMedChild" Program.

Conflict of interest We also declare no conflict of interest.

References

Adriani W, Laviola G (2004) Windows of vulnerability to psychopathol-ogy and therapeutic strategy in the adolescent rodent model. BehavPharmacol 15:341–352

Adriani W, Macri S, Pacifici R, Laviola G (2002a) Peculiar vulnerabilityto nicotine oral self-administration in mice during early adolescence.Neuropsychopharmacology 27:212–224

AdrianiW,Macri S, Pacifici R, Laviola G (2002b) Restricted daily accessto water and voluntary nicotine oral consumption in mice: method-ological issues and individual differences. Behav Brain Res 134:21–30

Adriani W, Spijker S, Deroche-Gamonet V, Laviola G, Le Moal M, SmitAB, Piazza PV (2003) Evidence for enhanced neurobehavioralvulnerability to nicotine during periadolescence in rats. J Neurosci23:4712–4716

Adriani W, Granstrem O, Macri S, Izykenova G, Dambinova S, LaviolaG (2004) Behavioral and neurochemical vulnerability during ado-lescence in mice: studies with nicotine. Neuropsychopharmacology29:869–878

Akbarian S, Huntsman MM, Kim JJ, Tafazzoli A, Potkin SG, BunneyWE Jr, Jones EG (1995) GABAA receptor subunit gene expressionin human prefrontal cortex: comparison of schizophrenics and con-trols. Cereb Cortex 5:550–560

Albuquerque EX, Pereira EF, Braga MF, Alkondon M (1998)Contribution of nicotinic receptors to the function of synapses in

Psychopharmacology

the central nervous system: the action of choline as a selectiveagonist of alpha 7 receptors. J Physiol Paris 92:309–316

Albuquerque EX, Pereira EF, Alkondon M, Rogers SW (2009)Mammalian nicotinic acetylcholine receptors: from structure tofunction. Physiol Rev 89:73–120

Arenas E, Akerud P, Wong V, Boylan C, Persson H, Lindsay RM, AltarCA (1996) Effects of BDNF and NT-4/5 on striatonigral neuropep-tides or nigral GABA neurons in vivo. Eur J Neurosci 8:1707–1717

Aultman JM, Moghaddam B (2001) Distinct contributions of glutamateand dopamine receptors to temporal aspects of rodent workingmemory using a clinically relevant task. Psychopharmacology(Berl) 153:353–364

Ballmaier M, Zoli M, Leo G, Agnati LF, Spano P (2002) Preferentialalterations in the mesolimbic dopamine pathway of heterozygousreeler mice: an emerging animal-based model of schizophrenia. EurJ Neurosci 15:1197–1205

Batel P (2000) Addiction and schizophrenia. Eur Psychiatry 15:115–122Benes FM, Lim B, Matzilevich D, Walsh JP, Subburaju S, Minns M

(2007) Regulation of the GABA cell phenotype in hippocampus ofschizophrenics and bipolars. Proc Natl Acad Sci U S A 104:10164–10169

Breese CR, Lee MJ, Adams CE, Sullivan B, Logel J, Gillen KM, MarksMJ, Collins AC, Leonard S (2000) Abnormal regulation of highaffinity nicotinic receptors in subjects with schizophrenia.Neuropsychopharmacology 23:351–364

Brenhouse HC, Andersen SL (2011) Developmental trajectories duringadolescence in males and females: a cross-species understanding ofunderlying brain changes. Neurosci Biobehav Rev 35:1687–1703

Calamandrei G, Pennazza S, Ricceri L, Valanzano A (1996) Neonatalexposure to anti-nerve growth factor antibodies affects exploratorybehavior of developing mice in the hole board. Neurotoxicol Teratol18:141–146

Chudasama Y, Robbins TW (2006) Functions of frontostriatal systems incognition: comparative neuropsychopharmacological studies in rats,monkeys and humans. Biol Psychol 73:19–38

Costa E, Davis J, Grayson DR, Guidotti A, Pappas GD, Pesold C (2001)Dendritic spine hypoplasticity and downregulation of reelin andGABAergic tone in schizophrenia vulnerability. Neurobiol Dis 8:723–742

Counotte DS, Goriounova NA, Moretti M, Smoluch MT, Irth H,Clementi F, Schoffelmeer AN, Mansvelder HD, Smit AB, Gotti C,Spijker S (2012) Adolescent nicotine exposure transiently increaseshigh-affinity nicotinic receptors and modulates inhibitory synaptictransmission in rat medial prefrontal cortex. Faseb J 26:1810–1820

Cowansage KK, LeDoux JE, Monfils MH (2010) Brain-derived neuro-trophic factor: a dynamic gatekeeper of neural plasticity. Curr MolPharmacol 3:12–29

Dao JM, McQuown SC, Loughlin SE, Belluzzi JD, Leslie FM (2011)Nicotine alters limbic function in adolescent rat by a 5-HT1Areceptor mechanism. Neuropsychopharmacology 36:1319–1331

D'Arcangelo G, Miao GG, Chen SC, Soares HD, Morgan JI, Curran T(1995) A protein related to extracellular matrix proteins deleted inthe mouse mutant reeler. Nature 374:719–723

De Filippis B, Ricceri L, Laviola G (2010) Early postnatal behavioralchanges in the Mecp2-308 truncation mouse model of Rett syn-drome. Genes Brain Behav 9:213–223

D'Souza MS, Markou A (2012) Schizophrenia and tobacco smokingcomorbidity: nAChR agonists in the treatment of schizophrenia-associated cognitive deficits. Neuropharmacology 62:1564–1573

Fatemi SH, Earle JA, McMenomy T (2000) Reduction in Reelin immu-noreactivity in hippocampus of subjects with schizophrenia, bipolardisorder and major depression. Mol Psychiatry 5(654–63):571

Flores CM, Rogers SW, Pabreza LA, Wolfe BB, Kellar KJ (1992) Asubtype of nicotinic cholinergic receptor in rat brain is composed ofalpha 4 and beta 2 subunits and is up-regulated by chronic nicotinetreatment. Mol Pharmacol 41:31–37

Freedman R, Hall M, Adler LE, Leonard S (1995) Evidence in postmor-tem brain tissue for decreased numbers of hippocampal nicotinicreceptors in schizophrenia. Biol Psychiatry 38:22–33

Freedman R, Adams CE, Leonard S (2000) The alpha7-nicotinic acetyl-choline receptor and the pathology of hippocampal interneurons inschizophrenia. J Chem Neuroanat 20:299–306

Fuso A, Nicolia V, Cavallaro RA, Ricceri L, D'Anselmi F, Coluccia P,Calamandrei G, Scarpa S (2008) B-vitamin deprivation induceshyperhomocysteinemia and brain S -adenosylhomocysteine, de-pletes brain S-adenosylmethionine, and enhances PS1 and BACEexpression and amyloid-beta deposition in mice. Mol Cell Neurosci37:731–746

Goriounova NA, Mansvelder HD (2012) Nicotine exposure during ado-lescence leads to short- and long-term changes in spike timing-dependent plasticity in rat prefrontal cortex. J Neurosci 32:10484–10493

Govind AP, Walsh H, Green WN (2012) Nicotine-induced upregulationof native neuronal nicotinic receptors is caused by multiple mecha-nisms. J Neurosci 32:2227–2238

Guidotti A, Auta J, Davis JM, Di-Giorgi-Gerevini V, Dwivedi Y, GraysonDR, Impagnatiello F, Pandey G, Pesold C, Sharma R, Uzunov D,Costa E (2000) Decrease in reelin and glutamic acid decarboxyl-ase67 (GAD67) expression in schizophrenia and bipolar disorder: apostmortem brain study. Arch Gen Psychiatry 57:1061–1069

Guidotti A, Auta J, Davis JM, Dong E, Grayson DR, Veldic M, Zhang X,Costa E (2005) GABAergic dysfunction in schizophrenia: newtreatment strategies on the horizon. Psychopharmacology (Berl)180:191–205

Guidotti A, Auta J, Chen Y, Davis JM, Dong E, Gavin DP, Grayson DR,Matrisciano F, Pinna G, Satta R, Sharma RP, Tremolizzo L, TuetingP (2010) Epigenetic GABAergic targets in schizophrenia and bipo-lar disorder. Neuropharmacology 60:1007–1016

Hughes RN (2004) The value of spontaneous alternation behavior (SAB)as a test of retention in pharmacological investigations of memory.Neurosci Biobehav Rev 28:497–505

Ikeda Y, Yahata N, Ito I, Nagano M, Toyota T, Yoshikawa T, Okubo Y,Suzuki H (2008) Low serum levels of brain-derived neurotrophicfactor and epidermal growth factor in patients with chronic schizo-phrenia. Schizophr Res 101:58–66

Kandel DB, Chen K (2000) Extent of smoking and nicotine dependencein the United States: 1991–1993. Nicotine Tob Res 2:263–274

Kellendonk C, Simpson EH, Polan HJ,Malleret G, Vronskaya S,WinigerV, Moore H, Kandel ER (2006) Transient and selective overexpres-sion of dopamine D2 receptors in the striatum causes persistentabnormalities in prefrontal cortex functioning. Neuron 49:603–615

Kliethermes CL, Crabbe JC (2006) Pharmacological and genetic influ-ences on hole-board behaviors in mice. Pharmacol Biochem Behav85:57–65

Krueger DD, Howell JL, Hebert BF, Olausson P, Taylor JR, Nairn AC(2006) Assessment of cognitive function in the heterozygous reelermouse. Psychopharmacology (Berl) 189:95–104

Kumari V, Postma P (2005) Nicotine use in schizophrenia: the selfmedication hypotheses. Neurosci Biobehav Rev 29:1021–1034

KundakovicM, Chen Y, Costa E, Grayson DR (2007) DNAmethyltrans-ferase inhibitors coordinately induce expression of the human reelinand glutamic acid decarboxylase 67 genes. Mol Pharmacol 71:644–653

Lalonde R (2002) The neurobiological basis of spontaneous alternation.Neurosci Biobehav Rev 26:91–104

Laviola G,Marco EM (2011) Passing the knife edge in adolescence: brainpruning and specification of individual lines of development.Neurosci Biobehav Rev 35:1631–1633

Laviola G, Adriani W, Terranova ML, Gerra G (1999) Psychobiologicalrisk factors for vulnerability to psychostimulants in human adoles-cents and animal models. Neurosci Biobehav Rev 23:993–1010

Psychopharmacology

Laviola G, Macri S, Morley-Fletcher S, Adriani W (2003) Risk-takingbehavior in adolescent mice: psychobiological determinants andearly epigenetic influence. Neurosci Biobehav Rev 27:19–31

Laviola G, AdrianiW, Gaudino C,Marino R, Keller F (2006) Paradoxicaleffects of prenatal acetylcholinesterase blockade on neuro-behavioral development and drug-induced stereotypies in reelermutant mice. Psychopharmacology (Berl) 187:331–344

Laviola G, Ognibene E, Romano E, Adriani W, Keller F (2009) Gene–environment interaction during early development in the heterozy-gous reeler mouse: clues for modelling of major neurobehavioralsyndromes. Neurosci Biobehav Rev 33:560–572

Leonard S, Breese C, Adams C, Benhammou K, Gault J, Stevens K, LeeM, Adler L, Olincy A, Ross R, Freedman R (2000) Smoking andschizophrenia: abnormal nicotinic receptor expression. Eur JPharmacol 393:237–242

Leonard S, Mexal S, Freedman R (2007) Smoking, genetics and schizo-phrenia: evidence for self medication. J Dual Diagn 3:43–59

Levin ED, Rezvani AH (2007) Nicotinic interactions with antipsychoticdrugs, models of schizophrenia and impacts on cognitive function.Biochem Pharmacol 74:1182–1191

Lewis DA, Hashimoto T, Volk DW (2005) Cortical inhibitory neuronsand schizophrenia. Nat Rev Neurosci 6:312–324

Liu WS, Pesold C, Rodriguez MA, Carboni G, Auta J, Lacor P, Larson J,Condie BG, Guidotti A, Costa E (2001) Down-regulation of den-dritic spine and glutamic acid decarboxylase 67 expressions in thereelin haploinsufficient heterozygous reeler mouse. Proc Natl AcadSci U S A 98:3477–3482

Lu B, Martinowich K (2008) Cell biology of BDNF and its relevance toschizophrenia. Novartis Found Symp 289: 119–29; discussion 129–35, 193–5

Lv J, Mao C, Zhu L, Zhang H, Pengpeng H, Xu F, Liu Y, Zhang L, Xu Z(2008) The effect of prenatal nicotine on expression of nicotinereceptor subunits in the fetal brain. Neurotoxicology 29:722–726

Macri S, Biamonte F, Romano E, Marino R, Keller F, Laviola G (2010)Perseverative responding and neuroanatomical alterations in adultheterozygous reeler mice are mitigated by neonatal estrogen admin-istration. Psychoneuroendocrinology 35:1374–1387

Maloku E, Kadriu B, Zhubi A, Dong E, Pibiri F, Satta R, Guidotti A(2011) Selective alpha4beta2 nicotinic acetylcholine receptor ago-nists target epigenetic mechanisms in cortical GABAergic neurons.Neuropsychopharmacology 36:1366–1374

Marks MJ, Burch JB, Collins AC (1983) Effects of chronic nicotineinfusion on tolerance development and nicotinic receptors. JPharmacol Exp Ther 226:817–825

Marks MJ, Pauly JR, Gross SD, Deneris ES, Hermans-Borgmeyer I,Heinemann SF, Collins AC (1992) Nicotine binding and nicotinicreceptor subunit RNA after chronic nicotine treatment. J Neurosci12:2765–2784

Miao H, Liu C, Bishop K, Gong ZH, Nordberg A, Zhang X (1998)Nicotine exposure during a critical period of development leads topersistent changes in nicotinic acetylcholine receptors of adult ratbrain. J Neurochem 70:752–762

Moy SS, Nadler JJ, Poe MD, Nonneman RJ, Young NB, Koller BH,Crawley JN, Duncan GE, Bodfish JW (2008) Development of amouse test for repetitive, restricted behaviors: relevance to autism.Behav Brain Res 188:178–194

Ochoa EL, Lasalde-Dominicci J (2007) Cognitive deficits in schizophre-nia: focus on neuronal nicotinic acetylcholine receptors andsmoking. Cell Mol Neurobiol 27:609–639

Ognibene E, Adriani W, Granstrem O, Pieretti S, Laviola G (2007)Impulsivity-anxiety-related behavior and profiles of morphine-induced analgesia in heterozygous reeler mice. Brain Res 1131:173–180

Ognibene E, Adriani W, Caprioli A, Ghirardi O, Ali SF, Aloe L, LaviolaG (2008) The effect of early maternal separation on brain derivedneurotrophic factor and monoamine levels in adult heterozygous

reeler mice. Prog Neuropsychopharmacol Biol Psychiatry 32:1269–1276

Olincy A, Young DA, Freedman R (1997) Increased levels of the nicotinemetabolite cotinine in schizophrenic smokers compared to othersmokers. Biol Psychiatry 42:1–5

Pacheco MA, Pastoor TE, Lukas RJ, Wecker L (2001) Characterizationof human alpha4beta2 neuronal nicotinic receptors stably expressedin SH-EP1 cells. Neurochem Res 26:683–693

Picciotto MR, Zoli M (2008) Neuroprotection via nAChRs: the role ofnAChRs in neurodegenerative disorders such as Alzheimer's andParkinson's disease. Front Biosci 13:492–504

Picciotto MR, Zoli M, Rimondini R, Lena C, Marubio LM, Pich EM,Fuxe K, Changeux JP (1998) Acetylcholine receptors containing thebeta2 subunit are involved in the reinforcing properties of nicotine.Nature 391:173–177

Podhorna J, Didriksen M (2004) The heterozygous reeler mouse: behav-ioural phenotype. Behav Brain Res 153:43–54

Qiu S, Korwek KM, Pratt-Davis AR, Peters M, Bergman MY,Weeber EJ(2006) Cognitive disruption and altered hippocampus synaptic func-tion in Reelin haploinsufficient mice. Neurobiol Learn Mem 85:228–242

Rapoport JL (2013) Prevention of schizophrenia: an impossible dream?Am J Psychiatry 170:245–247

Ricceri L, De Filippis B, Fuso A, Laviola G (2011) Cholinergichypofunction in MeCP2-308 mice: beneficial neurobehavioural ef-fects of neonatal choline supplementation. Behav Brain Res 221:623–629

Ringstedt T, Linnarsson S, Wagner J, Lendahl U, Kokaia Z, Arenas E,Ernfors P, Ibanez CF (1998) BDNF regulates reelin expression andCajal–Retzius cell development in the cerebral cortex. Neuron 21:305–315

Romano E, Fuso A, Laviola G (2013) Nicotine restores Wt-like levels ofreelin and GAD67 gene expression in brain of heterozygous reelermice. Neurotox Res 24:205–215

Ross RG, Hunter SK, McCarthy L, Beuler J, Hutchison AK,Wagner BD,Leonard S, StevensKE, FreedmanR (2013) Perinatal choline effectson neonatal pathophysiology related to later schizophrenia risk. AmJ Psychiatry 170:290–298

Roth TL, Sweatt JD (2011) Epigenetic marking of the BDNF gene byearly-life adverse experiences. Horm Behav 59:315–320

Rowell PP, Li M (1997) Dose–response relationship for nicotine-inducedup-regulation of rat brain nicotinic receptors. J Neurochem 68:1982–1989

Ruzicka WB, Zhubi A, Veldic M, Grayson DR, Costa E, Guidotti A(2007) Selective epigenetic alteration of layer I GABAergic neuronsisolated from prefrontal cortex of schizophrenia patients using laser-assisted microdissection. Mol Psychiatry 12:385–397

Saito M, O'Brien D, Kovacs KM, Wang R, Zavadil J, Vadasz C (2005)Nicotine-induced sensitization in mice: changes in locomotor activ-ity and mesencephalic gene expression. Neurochem Res 30:1027–1035

Salinger WL, Ladrow P, Wheeler C (2003) Behavioral phenotype of thereeler mutant mouse: effects of RELN gene dosage and socialisolation. Behav Neurosci 117:1257–1275

Salin-Pascual RJ, Alcocer-Castillejos NV, Alejo-Galarza G (2003)Nicotine dependence and psychiatric disorders. Rev Invest Clin55:677–693

Satta R, Maloku E, Zhubi A, Pibiri F, Hajos M, Costa E, Guidotti A(2008) Nicotine decreases DNAmethyltransferase 1 expression andglutamic acid decarboxylase 67 promoter methylation inGABAergic interneurons. Proc Natl Acad Sci U S A 105:16356–16361

Schneider T, Bizarro L, Asherson PJ, Stolerman IP (2012) Hyperactivity,increased nicotine consumption and impaired performance in thefive-choice serial reaction time task in adolescent rats prenatallyexposed to nicotine. Psychopharmacology (Berl) 223:401–415

Psychopharmacology

Schwartz RD, Kellar KJ (1983) Nicotinic cholinergic receptor bindingsites in the brain: regulation in vivo. Science 220:214–216

Singh A, Potter A, Newhouse P (2004) Nicotinic acetylcholine receptorsystem and neuropsychiatric disorders. IDrugs 7:1096–1103

Slotkin TA, MacKillop EA, Rudder CL, Ryde IT, Tate CA, Seidler FJ(2007a) Permanent, sex-selective effects of prenatal or adolescentnicotine exposure, separately or sequentially, in rat brain regions:indices of cholinergic and serotonergic synaptic function, cell sig-naling, and neural cell number and size at 6 months of age.Neuropsychopharmacology 32:1082–1097

Slotkin TA, Ryde IT, Seidler FJ (2007b) Separate or sequential exposureto nicotine prenatally and in adulthood: persistent effects on acetyl-choline systems in rat brain regions. Brain Res Bull 74:91–103

Slotkin TA, Bodwell BE, Ryde IT, Seidler FJ (2008) Adolescent nicotinetreatment changes the response of acetylcholine systems to subse-quent nicotine administration in adulthood. Brain Res Bull 76:152–165

Spear L (2000a) Modeling adolescent development and alcohol use inanimals. Alcohol Res Health 24:115–123

Spear LP (2000b) The adolescent brain and age-related behavioral man-ifestations. Neurosci Biobehav Rev 24:417–463

Takeda H, Tsuji M, Matsumiya T (1998) Changes in head-dippingbehavior in the hole-board test reflect the anxiogenic and/or anxio-lytic state in mice. Eur J Pharmacol 350:21–29

ToyookaK, AsamaK,WatanabeY,Muratake T, TakahashiM, Someya T,Nawa H (2002) Decreased levels of brain-derived neurotrophicfactor in serum of chronic schizophrenic patients. Psychiatry Res110:249–257

Tremolizzo L, Carboni G, RuzickaWB,Mitchell CP, Sugaya I, Tueting P,Sharma R, Grayson DR, Costa E, Guidotti A (2002) An epigeneticmouse model for molecular and behavioral neuropathologies relatedto schizophrenia vulnerability. Proc Natl Acad Sci U S A 99:17095–17100

Tueting P, Costa E, Dwivedi Y, Guidotti A, Impagnatiello F, Manev R,Pesold C (1999) The phenotypic characteristics of heterozygousreeler mouse. Neuroreport 10:1329–1334

Tueting P, Doueiri MS, Guidotti A, Davis JM, Costa E (2006) Reelindown-regulation in mice and psychosis endophenotypes. NeurosciBiobehav Rev 30:1065–1077

Veldic M, Kadriu B, Maloku E, Agis-Balboa RC, Guidotti A, Davis JM,Costa E (2007) Epigenetic mechanisms expressed in basal gangliaGABAergic neurons differentiate schizophrenia from bipolar disor-der. Schizophr Res 91:51–61

Weickert CS, Hyde TM, Lipska BK, Herman MM, Weinberger DR,Kleinman JE (2003) Reduced brain-derived neurotrophic factor inprefrontal cortex of patients with schizophrenia. Mol Psychiatry 8:592–610

Weickert CS, Ligons DL, Romanczyk T, Ungaro G, Hyde TM, HermanMM, Weinberger DR, Kleinman JE (2005) Reductions inneurotrophin receptor mRNAs in the prefrontal cortex of patientswith schizophrenia. Mol Psychiatry 10:637–650

Wilcox RR (1987) New Statistical Procedures for the Social Sciences.Modern Solution To Basic Problems. Hillsdale, New Jersey

Wonnacott S, Irons J, Rapier C, Thorne B, Lunt GG (1989) Presynapticmodulation of transmitter release by nicotinic receptors. Prog BrainRes 79:157–163

Woods BT (1998) Is schizophrenia a progressive neurodevelopmentaldisorder? Toward a unitary pathogenetic mechanism. Am JPsychiatry 155:1661–1670

Yamada MK, Nakanishi K, Ohba S, Nakamura T, Ikegaya Y, NishiyamaN, Matsuki N (2002) Brain-derived neurotrophic factor promotesthe maturation of GABAergic mechanisms in cultured hippocampalneurons. J Neurosci 22:7580–7585

Zaheer A, Haas JT, Reyes C, Mathur SN, Yang B, Lim R (2006) GMF-knockout mice are unable to induce brain-derived neurotrophicfactor after exercise. Neurochem Res 31:579–584

Zhang H, Liu X, Zhang C, Mundo E, Macciardi F, Grayson DR, GuidottiAR, Holden JJ (2002) Reelin gene alleles and susceptibility toautism spectrum disorders. Mol Psychiatry 7:1012–1017

Zhubi A, Veldic M, Puri NV, Kadriu B, Caruncho H, Loza I, Sershen H,Lajtha A, Smith RC, Guidotti A, Davis JM, Costa E (2009) Anupregulation of DNA-methyltransferase 1 and 3a expressed in tel-encephalic GABAergic neurons of schizophrenia patients is alsodetected in peripheral blood lymphocytes. Schizophr Res 111:115–122

Zimmerberg B, Brett MB (1992) Effects of early environmental experi-ence on self-administration of amphetamine and barbital.Psychopharmacology (Berl) 106:474–478

Psychopharmacology