Creutzfeldt-Jakob Disease Subtype-Specific Regional and Temporal Regulation of ADP Ribosylation Factor-1-Dependent Rho/MLC Pathway at Pre-Clinical Stage

Creutzfeldt-Jakob Disease Subtype-Specific Regionaland Temporal Regulation of ADP RibosylationFactor-1-Dependent Rho/MLC Pathway at Pre-Clinical Stage

Saima Zafar1 & Matthias Schmitz1 & Neelam Younus1 & Waqas Tahir1 &

Mohsin Shafiq1& Franc Llorens1 & Isidre Ferrer2,3 & Olivier Andéoletti4 &

Inga Zerr1

Received: 25 November 2014 /Accepted: 9 March 2015 /Published online: 21 April 2015# European Union 2015

Abstract Small GTPases of the Arf family mainly activatethe formation of coated carrier vesicles. We showed that class-I Arf1 interacts specifically with full length GPI-anchoredcellular prion protein (PrPC). Several recent reports have alsodemonstrated a missing link between the endoplasmic reticu-lum and the Golgi-complex role for proper folding, but theexact molecular mechanism is not yet fully understood. In thepresent study, we identified and characterized the interactiverole of Arf1 during PrPC intracellular distribution under path-ophysiological conditions. PrPC interaction with Arf1 was in-vestigated in cortical primary neuronal cultures of PrPC wildtype and knockout mice (PrP−/−). Arf1 and PrPC co-binding

affinity was confirmed using reverse co-immunoprecipitation,co-localization affinity using confocal laser-scanning micros-copy. Treatment with brefeldin-A modulated Arf1 expressionand resulted in down-regulation and redistribution of PrPC

into cytosolic region. In the pre-symptomatic stage of the dis-ease, Arf1 expression was significantly downregulated in thefrontal cortex in tg340 mice expressing about fourfold of hu-man PrP-M129 with PrP null background that had been inoc-ulated with human sCJD MM1 brain tissue homogenates(sCJD MM1 mice). In addition, the frontal cortex of CJDhuman brain demonstrated significant binding capacity ofArf1 protein using co-immunoprecipitation analysis. We also

Author Summary Prion diseases are fatal neurodegenerative diseasesthat affect a number of different species, including humans. A long pre-clinical phase of the disease lasting months to years is a basic character-istic of the prion diseases. During the disease replication stage, the affect-ed individual does not show any signs of the disease. The clinical orsymptomatic stage of the disease is much shorter than the pre-clinicalstage. Variant CJD appears to not only be transmitted by eating contam-inated beef but also by blood transfusions. In the latter cases, the blooddonors had prion diseases in a pre-clinical stage and the prion disease wasdiagnosed at a later date. Given the potential for large numbers of peopleto have a clinically inapparent prion infection, it is critical for publichealth concerns that diagnostic tools be developed to detect the diseaseas early as possible. To address this need, we have developed a mousemodel of prion disease to study the molecular pathways involved duringdisease progression. Targeting Arf/Rho/MLC signaling pathways mightbe a promising strategy to study the most likely cause of variant rate ofdisease progression and internal homeostasis of misfolded proteins.

* Saima [email protected]; [email protected]

1 Department of Neurology, Clinical Dementia Center and DZNE,Georg-August University, University Medical Center Goettingen(UMG), Robert-Koch-Str. 40, 37075 Goettingen, Germany

2 Institute of Neuropathology, IDIBELL-University HospitalBellvitge, University of Barcelona, Hospitalet de Llobregat, Spain

3 CIBERNED (Network center for biomedical research ofneurodegenerative diseases), Ministry of Health, Institute Carlos III,Madrid, Spain

4 Institut National de la Recherche Agronomique/Ecole NationaleVétérinaire, Toulouse, France

J Mol Neurosci (2015) 56:329–348DOI 10.1007/s12031-015-0544-3

examined Arf1 expression in the brain of CJD patients withthe subtypesMM1 and VV2 and found that it was regulated ina region-specific manner. In the frontal cortex, Arf1 expres-sion was not significantly changed in either MM1 or VV2subtype. Interestingly, Arf1 expression was significantly re-duced in the cerebellum in both subtypes as compared to con-trols. Furthermore, we observed altered RhoA activity, whichin turn affects myosin light-chain (MLC) phosphorylation andArf1-dependent PI3K pathway. Together, our findings under-score a key early symptomatic role of Arf1 in neurodegener-ation. Targeting the Arf/Rho/MLC signaling axis might be apromising strategy to uncover the missing link which proba-bly influences disease progression and internal homeostasis ofmisfolded proteins.

Keywords Prion protein . Interacting protein . Arf1 . BFA .

Intracellular PrPC distribution . Creutzfeldt-Jacob disease(CJD) .MM1 . VV2 . RhoA .MLC . Frontal cortex .

Cerebellum

Introduction

In eukaryotes, a family of GTP-binding proteins (Arf, Rab,Rho, and dynamin families) regulates all aspects of vesicletrafficking, i.e., formation of vesicles on donor membranesthat facilitates vesicle docking on target membranes (Bucciet al. 2000). A regulatory factor for cell division and migrationhas been demonstrated in specific proteins of the Arf family(Sabe et al. 2006; Boulay et al. 2008; Boulay and Claing 2009;Lewis-Saravalli et al. 2013). The budding of coated vesicles isalso initiated when molecules of ubiquitously expressed Arfprotein exchange their bound GDP for GTP. This reaction ismostly catalyzed by an enzyme in the Golgi membrane (Weisand Scheller 1998). Arf protein is expressed in six differentconserved isoforms in mammalian cells and exists in solubleas well as membrane-associated forms (Schurmann et al.1994; Campa and Randazzo 2008; Popoff et al. 2011). Underphysiological conditions, Arf proteins regulate vesicular traf-ficking along with microtubules and by remodeling actin. Arfproteins also facilitate phospholipase D activation (Cockcroftet al. 1994) and are morphological/functional regulators andtransporters in the ER and Golgi complex, involved inendosomal and nuclear membrane fusion (Tsai et al. 2013).Likewise, RhoA GTPase plays a major role during cell divi-sion and cell migratory action and dysregulatory transportmachinery (Clark et al. 2000; Fritz and VanBerkum 2002;Pille et al. 2005). Arf1 is a critical modulating factor of Rho-dependent signaling pathway regulating myosin light chain(MLC) and microvesicles transportation (Schlienger et al.2014).

Most familial cases of neurodegenerative disorders are as-sociated with reported dysfunction of vesicular biogenesis,

vesicle/organelle trafficking, and synaptic signaling (Fletcherand Mullins 2010). Recent nanoparticle cell imaging studiesindicated classical clathrin-dependent/receptor-mediated path-way for PrPC−aptamer−QD complex internalization into cy-toplasm and also the involvement of different movements(i.e., membrane/confined diffusion and vesicle transportation)during distinct phases of PrPC endocytosis (Chen et al. 2012).PrPC is implicated in the conversion to the pathogenic isoformof PrPSc and is thought to be involved in redox equilibrium,differentiation, cell adhesion, transmembrane signaling, neu-ronal survival, plasticity, and has recently been linked to re-cently interdependent accumulation of β-amyloid deposits(Schneider et al. 2011; Linden et al. 2008; Weise et al. 2008;Ordonez-Gutierrez et al. 2013).

The human form of PrPC consists of a highly conservedsequence of 253 amino acids (Goldmann 1993). The biosyn-thesis of PrPC is similar to that of other proteins, bothmembrane-bound and secreted. The majority of the matureprotein is located on the plasma membrane, anchored throughthe C-terminal glycosyl-phosphatidylinositol (GPI). PrPC con-tains a specific N-terminal signal peptide (SP) which translo-cates this protein into the endoplasmic reticulum (ER) fromwhere it transits the Golgi apparatus on its way to the cellsurface (Harris 2003). PrPC resides largely on the cell surfacewithin the detergent-resistant raft domains (Gorodinsky andHarris 1995; Naslavsky et al. 1997) and constitutively cyclesbetween the plasma membrane and the endocytic compart-ment (Shyng et al. 1993).

Kinetic analysis has suggested that PrPC molecules cyclethrough the cell with a transit time of approximately 60 min(Magalhaes et al. 2002). Shyng et al. reported in 1993 thatmost of the protein is recycled without degradation and thatinternalization of PrPC may occur via clathrin-coated pits(Shyng et al. 1995a), caveolae-like membranous domains(Vey et al. 1996), or sphingolipid/cholesterol rafts (Shynget al. 1995b). The pathogenesis of prion diseases is attributedto the major conformational change of the cellular form ofprion protein (PrPC) to the diseased form (PrPSc) (Aguzzi2000; Knight and Will 2004; Aguzzi and O’Connor 2010;Prusiner 1998). The mechanism behind this misfolding tothe PrPSc form in prion diseases is poorly understood. Theaccumulation of this misfolded form in the cytosol is hypoth-esized to cause the malfunction of the transport machinery thatultimately leads to disease (Ma et al. 2002). This vesicularbiogenesis dysfunction may also be specifically involved inthe disruption of PrPC transport machinery, most probably as aresult of the activity of various interacting proteins.

Creutzfeldt-Jakob disease (CJD) is the most common priondisease in humans. It shows a broad heterogeneity of patho-physiological features and has a methionine/valine (M/V)polymorphism at codon 129 in the human prion protein gene(PRNP gene). The presence of two major types of protease-resistant, diseased forms of prion protein (PrPSc) lead to two

330 J Mol Neurosci (2015) 56:329–348

different Western blot profiles (type 1 20/21 kDa and type 219 kDa). These two PrPSc profiles yield six possible dimercombinations: MM1, MM2, MV1, MV2, VV1, and VV2(Parchi et al. 1999). The prevalent subtypes are MM1 andVV2 which represent about 67 and 15 % of all sCJD cases,respectively (Parchi et al. 1999; Gambetti et al. 2003;Dagdanova et al. 2010). Patients with theMM1 subtype sufferfrom cognitive impairment and visual deficits at an early stageof the disease, while dementia develops later in patients withthe VV2 subtype (Gambetti et al. 2003; Llorens et al. 2013;Katsube et al. 2013).

Importantly, many characteristics of the interacting earlyresponse and downstream effectors in prion diseases remainelusive, mainly because previous analyses of early mediatorsand responses in human samples and sCJD mouse models didnot take prion subtypes, cell types, and regional responses intoaccount.

In the current study, we demonstrated correlation betweenearly progressive responses of Arf1 with PrPC and subtyperegional differences in the frontal cortex and cerebellum inMM1 and VV2 sCJD patients, in an attempt to gain under-standing of the factors in vesicle/organelle trafficking and ve-sicular biogenesis in prion diseases. In addition, tg340 miceexpressing about fourfold of human PrP-M129 with PrP nullbackground inoculated with sCJD MM1 brain tissue homog-enates were analyzed during early pre-symptomatic andsymptomatic clinical stages.

Materials and Methods

Primary Culture of Mouse Cortex

Primary cultures of mouse cortex were prepared as describedpreviously (Carimalo et al. 2005). In brief, embryos were re-moved from pregnant mice at embryonic day 14 under halo-thane (Sigma) anesthesia; the mice were prnp (the PrP-encoding gene) and knockout (PrP−/−) (Bueler et al. 1992).Animals were killed by cervical dislocation and the dissectedembryonic cortex was mechanically dissociated and plated onpolyethylenimine (1mg/ml)-coated glass cover slips in culturewells. For the brefeldin-A (BFA) treatment, cultures at a den-sity of 7×105 cells/cm2 were first grown in DMEM, supple-mented with 5 % fetal bovine serum (FBS), 2.5 % fetal calfserum (FCS; Eurobio), 2 mM glutamine, and 0.1 % penicillinand streptomycin (Gibco). After 3 days in vitro (DIV), themedium was replaced with N5 medium (Kawamoto andBarrett 1986) with 180 mg/L glucose and supplemented with5 % FBS and 1 % FCS. Then 3 μM cytosine arabinoside wasadded to prevent astrocyte proliferation (resulting in at least97 % pure neuronal cultures), and 1 μM and MK-801 toprevent excitotoxicity (Knusel et al. 1990). The medium waschanged daily. On DIV 5, FCS was removed and the FBS

content was reduced to 1 %. All cell cultures were kept at37 °C in a humidified 5 % CO2 atmosphere.

Brefeldin ATreatment

For BFA treatment, the cells were treated with 1μg/ml of BFAafter 24–48 h of transient transfection for various incubationtimes in primary cortex cultures. After BFA treatment, thecells were lysed for the expression analysis and/orimmunomounted for the co-localization analysis.

Cell Viability Assays

Cell viability assay was performed as described previously(Zafar et al. 2014). Briefly, the adherent cells were grown upto 60–70 % confluency and then detached from flasks using1× Trypsin-EDTA. The cells were spun down at 4 °C for5 min at 400×g and resuspended in culture media. Cells werethen dispensed into 24-well plates (Nunc, Roskilde, Denmark)at a final concentration of 1×105 cells/well and incubatedfor 12 h. The culture media was then removed and replacedprior to MTS [3-(4, 5-dimthylthiazol-2-yl)-5-(3-carbo-ymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium, innersalt] treatment. The effect of the presence of PrPC on cellviability was measured using the MTS cell proliferation assay,which measures the reduction of MTS tetrazolium salt toformazan in metabolically active cells (Cory et al. 1991).The cells were then treated with 1:20 ratio of MTS reagent(Promega Co. Madison, WI, USA) 2 mg/ml with 0.5 % glu-cose and PMS 0.92 mg/ml with 0.5 % glucose. Cells wereincubated for 1 h at 37 °C for color development, and theabsorbance values were read at 490 nm using a Multiscanplate reader (Labsystems, VA, USA) and Accent software2.6. Background absorbance from controls was incubated inmedia with the MTS reagent and was subtracted from samplewells after the final absorbance was obtained.

Trypan blue exclusion was also used to check cell viability.In subsequent experiments, viability was determined bycounting the number of cells in ten fields (×20 objective)selected at random on cover slip containing either transfectedor un-transfected (control) cells.

The nuclear area factor (NAF) for transfected and non-transfected cells was determined by using fluorescent stainingof the nucleus using DAPI, followed by digital microscopy.The measurement of the nuclear area and nuclear circularitywas carried out using Image J analysis software (Daniel andDeCoster 2004).

Caspase-3 Activity Assay

The caspase-3-activity assay allows quantitative measurementof caspase-3 (DEVDase) protease activity, which is an earlyregulatory event in the apoptotic cell death process. The assay

J Mol Neurosci (2015) 56:329–348 331

was performed using caspase-3 activity assay kit according tothe manufacturer’s recommendations. Briefly, untreated con-trol and primary cortical cultures were lysed in the cell lysisbuffer for 15 min at 4 °C, and this was followed by centrifu-gation at 10,000×g. Protein concentration was estimated fromthe supernatants and the total cell lysate (50 μg) was thenincubated with 50 μM caspase-3-specific substrate DEVD-pNA for 4–5 h at 37 °C. The caspase-3 inhibitor Z-vad-FMK (20 mM) was used as control. Caspase-3-mediated re-lease of pNAwas measured by absorbance at 405 nm. Back-ground absorbance from the control (untreated cells) wassubtracted from the samples after the final absorbance wasobtained.

Antibodies, Reagents, and Immunoblot Analysis

Mouse anti-PrP mAbs: 6H4 (Prionics AG, Zurich, Switzer-land), SAF70 (SpiBio, Paris, France), 3F4 (Covance, MunsterGermany), mouse anti-Arf1 (Affinity BioReagents, CO,USA), mouse anti-anti-RhoA monoclonal antibody (NewEastBiosciences, King of Prussia, PA), myosin light chain 2 andphospho-myosin light chain 2 -Thr18/Ser19 (Cell signaling,Frankfurt, Germany), PI3 kinase p85, p55, and pKB (Abcam,Cambridge, UK) were used as primary antibodies. HRP-conjugated rabbit anti-mouse pAb (IBA, Gottingen, Germa-ny), anti-rabbit pAb (Santa Cruz Biotechnology, Santa Cruz,CA, USA), goat anti-mouse cy3-conjugated (Dianova, Ham-burg, Germany), goat anti-rabbit (Alexa 488-conjugated), andanti-mouse (Alexa 488-conjugated) were used as secondaryantibodies. BFA (Cell Signaling, Frankfurt, Germany), prote-ase and phosphatase inhibitor cocktail (Roche, Mannheim,Germany), Hoechst 33342 (Sigma-Aldrich, Steinheim,Germany) and halothane anesthesia, cytosine arabinoside(Sigma, St Louis, MO, USA), MK-801 (Research Biochemi-cals International), penicillin, and streptomycin (Gibco) werealso used.

Cell lysis and immunoblotting were performed as de-scribed previously (Zafar et al. 2011). Briefly, cells were lysed(50 mM Tris-HCl, pH 8, 1 % Triton X-100, 0.5 % CHAPS,1 mM DTT), and lysates were cleared of cell debris (1 min,1000×g, 4 °C). Cell lysates were supplemented with proteaseand phosphatase inhibitors (Roche) and were separated on12.5% 1-DE SDS-PAGE. Expression of recombinant proteinswas analyzed by immunoblot using overnight exposure at4 °C to anti-PrP 6H4 monoclonal antibody (1:1000), anti-PrP SAF70 monoclonal antibody (1:5000), and anti-Arf1mAb (1:1000). Membranes were then rinsed in 1× TBS-Tand incubated with the corresponding horseradishperoxidase-conjugated secondary antibody (diluted 1:2000/1:5000) for 1 h at RT. Immunoreactivity was detected afterimmersion of the membranes into enhanced chemilumines-cence (ECL) solution and exposure to ECL-Hyperfilm(Amersham Biosciences, Buckinghamshire, UK). Images

were documented using the ScanMaker4 (Microtek, Interna-tional), after correction for the background, and band intensi-ties were determined by densitometry using Labimage (ver-sion 2.7.1, Kapelan GmbH, Germany) data analyzer software.

sCJD MM1 Mice

Double transgenic mice overexpressing about fourfold levelof human PrPC with methionine at codon 129 (Met129) on amurine PrP knockout background were used, as describedpreviously (Padilla et al. 2011). Inocula were prepared fromsCJDMM1 brain tissues as 10% (w/v) homogenates. Individ-ually identified 6 to 10-week-old mice were anesthetized andinoculated with 2 mg of brain homogenate in the right parietallobe using a 25-gauge disposable hypodermic needle (six an-imals per group and time point). Mice were observed dailyand the neurological status was assessed weekly. When dis-ease progression was evident, or at the end of lifespan, animalswere euthanized, necropsy was performed, and the brain wasremoved. A part of the brain was fixed by immersion in 10 %buffered formalin to quantify spongiform degeneration andperform immunohistological procedures. The other part wasfrozen at −80 °C to extract protein. Survival time was calcu-lated for each isolate and expressed as the mean of the survivalday post-inoculation (dpi) of all mice scoring positive forPrPSc. Infection rate was determined as the proportion of micescoring positive for PrPSc from all inoculated mice.

Cases and General Processing and CJD SubtypeCharacterization

Frontal cortex and cerebellum samples from 30 pathologicallyconfirmed sCJD patients (15 of each MM1 and VV2 sub-types) and 15 age-matched control cases (CON) were usedin this study. All 45 samples were obtained from the Instituteof Neuropathology Brain Bank (HUB-ICO-IDIBELLBiobank) and Biobank of Hospital Clinic-IDIBAPS. Themean age and gender of study cases were as described previ-ously (Llorens et al. 2013). In brief, for sCJD Western blotanalysis in frontal cortex: 60 years of age in control (10 M/5F), 68 in sCJD MM1 (10 M/5 F), and 63 in sCJD VV2 (5M/10 F). In cerebellum, the mean age was 62 in control (11M/4 F), 66 sCJD MM1 (10 M/F), and 63 in sCJD VV2 (4M/10 F). After post-mortem interval ranging 1 h and 45min to24 h and 30 min, coronal sections 1-cm thick were cut fromone of the hemispheres. Along with selectively dissected areasof encephalon, coronal sections were rapidly frozen on metalplates over dry ice, sorted in separate bags, labeled with water-resistant ink, and stored at −80 °C until further use for bio-chemical investigations. The other hemisphere was immersedfixed in 4 % buffered formalin for 3 weeks for morphologicalstudies and neuropathological examination and characteriza-tion. The analysis of the codon 129 genotype of PrP gene

332 J Mol Neurosci (2015) 56:329–348

(Met: M or Val: V) was performed after isolation of genomicDNA from blood samples according to standard methods.Western blot profile of PrPSc was classified as type 1 or type2 based on electrophoretic mobility after proteinase K (PK)digestion.

Co-Immunoprecipitation

Cell lysis was performed as described above, and the insolublecell debris was removed by centrifugation at 543,000×g for15 min at 4 °C. Immunoprecipitation was performed usingmagnetic Dynabeads protein G, according to the manufac-turer’s instructions. Total cytoplasmic cell extracts orimmunoprecipitated proteins (corresponding to 2×106 cells/lane) were subjected to 12.5 % 1-DE SDS-PAGE and, aftertransferring to polyvinylidene difluoride membranes(Millipore), immunoblotting was performed as describedabove.

Immunofluorescence and Quantification Analysis

Cells were platted on chambered slides [Lab-Tek™ II; Ther-mo Fisher Scientific (Nunc GmbH&Co. KG), LangenselboldSite] and transfected with the C-terminus One STrEP-tag PrPC

for 24, 36, and 48 h. Subsequently, they were washed in 1×PBS and then fixed for 15 min with 100 % ethanol. Afterfixation, cells were permeabilized with 0.2 % Triton X-100in 1× PBS, followed by a 20 min blocking step using 0.2 %casein-solution containing Tween 20. Co-localization of PrPC

with Arf1 was detected by exposure to the primary antibodies[anti-PrP 3 F4 (1:200), mouse anti-Arf1 (1:50)] overnight at4 °C (Table 1). The monoclonal antibodies were detected byincubating slides for 60 min with Alexa 488 conjugated anti-rabbit (1:200), Alexa 488 conjugated anti- mouse (1:200), orCy3-labeled anti-mouse secondary antibody (1:200). Incuba-tion with Hoechst 33342 or with TO-PRO-3 iodide for 10 minwas performed to visualize nuclei. Finally, cover slips wereplaced on the glass slides which were then mounted withFluoromount (DAKO, Hamburg, Germany). All the stepswere carried out in a dark humid chamber and reactions werestopped by washing the cover slips three times with 1× PBS,and then they were kept dry at 4 °C and in the dark. Control

cells (not-transfected with C-terminus One STrEP-tag PrPC ortransfected with empty vector) as well as BFA-treated cellswere harvested and processed under the same conditions asSTrEP-tag PrPC-transfected cells. Confocal laser-scanningmi-croscopy was carried out using LSM510 laser-scanning mi-croscope (Zeiss, Göttingen, Germany; 543 and 633 nmHelium-Neon and 488 nm Argon excitation wavelengths) ac-cording to the manufacturer’s instructions for the localizationof PrPC and other interacting proteins, using a 63×/1.25 oilimmersion lens. Individual images were analyzed separatelyfor co-localization using LSM 5 (Zeiss) or ImageJ (WCIFplugin) software. For two-color analysis, image stacks with atotal thickness of approximately 30μmwere acquired, using adynamic range of 12 bits per pixel. Pearson’s linear correlationcoefficient (rP) was used in this study to calculate fluorescencechannel correlations to illustrate the strength and the directionof the linear relationship between two fluorescence channels.

Statistical Analysis

All results in this study were obtained from at least four inde-pendent sets of experiments and were expressed as mean±SDusing descriptive statistics. Densitometric analysis of 1-DEgels was performed using Labimage (version 2.7.1 Kapelan,Leipzig, Germany) software.

Ethics Statement

Human samples from the Institute of Neuropathology BrainBank (HUB-ICO-IDIBELL Biobank) and Biobank of Hospi-tal Clinic-IDIBAPS were obtained following the Spanish leg-islation (Ley de la Investigación Biomédica 2013 and RealDecretoBiobancos, 2014) and the approval of the local ethicscommittees.

All animal experiments were performed in accordance withthe ethical standard set by Regierungspräsidium Tübingen(Regional Council) Experimental No. FLI 231/07 file refer-ence number 35/9185.81-2. All animal experiments have beenperformed in compliance with the institutional and Frenchnational guidelines, in accordancewith the European Commu-nity Council Directive 86/609/EEC. The experimental

Table 1 Arf1 partially co-localizes with PrPC BFA (1 μg/ml) PrPc rP Coloc. coefficient PrPc (M1) Coloc. coefficient Afr1 (M2)

Primary culture of mouse cortex

− + 0.151 0.46 0.81

+1.5 h +24 h 0.039 0.59 0.92

Pearson’s correlation coefficient rp (−1≤rp≤1) demonstrated significant co-localization (0.81) between Afr1 andPrPC in primary culture of mouse cortex without BFA treatment while 1.5 h of BFA treatment showed increase inco-localization as compared to untreated control cells. Co-localization coefficients,M1 andM2, ranged between 0and 1, showing partial co-localized pixels of interest within each channel

J Mol Neurosci (2015) 56:329–348 333

protocol was approved by the INRA Toulouse/ENVT ethicscommittee.

Results

In this study, we identified and characterized the interaction ofArf1 with PrPC in primary cortical cultures in the presence andabsence of PrPC. In order to further verify the interaction ofArf1 with PrPC, we used co-immunoprecipitation and co-localization analysis. We analyzed the data using Pearson’scorrelation coefficient. We demonstrated that:

& Brefeldin A impaired Arf1 expression and activity leadingto down-regulation and redistribution of PrPC into the cy-tosolic region.

& sCJD MM1 humanized infected tg340 mouse overex-pressing human PrP showed significant, early symptom-atic expressional regulation of Arf1 in the cortical regionof the brain.

& Arf1 expression in the brain (frontal cortex and cerebel-lum) was significantly altered in both MM1 and VV2subtypes.

& Interestingly, Arf1 expression in the cerebellum was sig-nificantly lower in both subtypes than in controls.

& Meanwhile, RhoA also showed destabilized activity lead-ing to myosin light-chain (MLC) phosphorylation dereg-ulation in both mouse and human samples.

Arf1 Interacts with PrPC and Depletion of Arf1 AltersPrPC Expression and Significantly Impairs IntracellularTransport

We previously demonstrated that Arf1 could be co-affinitypurified with PrPC byOne STrEP-tag affinity chromatography(Zafar et al. 2011). However, the nature of the interactionbetween Arf1 and PrPC and the possible interactive role ofArf1 was not clear. In order to determine whether the interac-tion between Arf1 and PrPC was functional, and whether Arf1is regulated and interacts in the presence of PrPSC in priondisease subtype patients, we affinity purified Arf1 with PrPC

from primary cortical cultures in the presence and absence ofPrPC and co-immunoblotted the purified eluate separatelywith Arf1 antibody (Fig. 1). No signal was detected usingPrPC antibody combined with control, purified eluates fromtotal cell lysates (TCL) of PrP−/− primary cortical cultures. Inorder to further confirm these observations, PrPC expressingprimary cortical cultures were co-immunoprecipitated withArf1 antibody using Dynabeads G-protein coupled magneticbeads (Fig. 1a, b). Eluates from this reverse co-immunoprecipitation were immunoblotted with SAF70 PrPC

antibody to verify the PrPC interaction (Fig. 1a). Figure 1a, b

showed the co-immunoprecipitation results with PrPC boundto Arf1 antibodies and with subsequent confirmation by im-munoblotting. Positive PrPC signals detected at 27 to 37 kDaconfirmed that Arf1 was co-precipitating with the PrPC.

BFA, an inhibitor of intracellular protein transport, is rou-tinely used to study the role of Arf1 in the morphology of theGolgi apparatus, and the recruitment of coat proteins to theGolgi (Volpicelli-Daley et al. 2005). To investigate the signif-icance of the Arf1 interaction with PrPC, we separately treatedprimary cultures of mouse cortical neuronal cells (PCC) with1 μg/ml BFA. Immunoblotting showed a significant decreasein PrPC concentrations in BFA-treated cells (**P<0.01,***P<0.001) compared to control cells (Fig. 2d–f).

Furthermore, confocal laser co-immunofluorescence scan-ning (Fig. 2a) showed increased co-localization of PrPC andArf1 compared to control, untreated PCC after 1.5 h of BFAtreatment. In contrast, co-localization of PrPC and Arf1 incontrol and untreated cells resulted in decreased expression(Fig. 2a, b and Table 2).

Arf1 Alters PrPC Expression and IntracellularTransportation Impairs Cell Viability and Caspase-3Activity in PCC Cells

HpL3-4 cells expressing PrP showed decreased levels of ap-optosis, which suggests a neuroprotective function of PrPC

(Zafar et al. 2014; Wu et al. 2008). Here, we assessed theneuroprotective function of human PrPC under altered Arf1expression. Interestingly, we found that PCC expressing PrPC

with altered Arf1 expression were more viable than cells with-out PrPC (Fig. 2c).

To test the cytotoxic nature of PrPC, the activity of theapoptotic marker enzyme caspase-3 was measured in transientand PCC PrPC expressing cells. Caspase-3 activity was ana-lyzed in cell lysates incubated with the caspase-3 substrateDEVD-pNA and free pNAwas detected by fluorescence. Rel-ative caspase-3 activity was analyzed in PCC cells expressingPrPC. It was significantly increased in cells without PrPC afterbrefeldin A treatment (Fig. 2c). This significantly decreasedcaspase-3 activity of cells expressing PrPC under altered Arf1expression showed the anti-apoptotic nature of interactive as-sociation of PrPC and Arf1 in neuronal cells.

Arf1 Expression in sCJD MM1 and VV2 Subtypes

Arf1 expression was assessed byWestern blot in frontal cortexand cerebellum samples from sCJD cases with the subtypesMM1 and VV2, as well as age-matched controls. Total ho-mogenates of the frontal cortex and cerebellum were analyzedusing an Arf1 monoclonal antibody (Fig. 3). Densitometricanalysis revealed no significantly altered Arf1 expression atthe protein level in the frontal cortex in the sCJD MM1 andVV2 subtypes (Fig. 3c). Interestingly, the Arf1 expression

334 J Mol Neurosci (2015) 56:329–348

showed was significantly decreased in the cerebellum of bothMM1 and VV2 subtypes compared to age-matched controlsamples (Fig. 3b, d) showing that the response was region-specific.

Arf1 Interaction with PrPC and PrPSc in sCJD MM1and VV2 Subtypes

We previously showed C-terminus One STrEP-tag PrPC spe-cific interaction with Arf1 by using Q-TOF MS/MS analysis(Zafar et al. 2011). Furthermore, we verified this interactionby using reverse co-immunoprecipitation and co-localizationanalysis (Fig. 1). We also observed evidence of interactionsbetween endogenous Arf1 and PrPC in both human embryonickidney (HEK) 293 and human neuroblastoma cells (SH-SY5Y) (data not shown). However, the nature of Arf1 inter-action within CJD subtype-specific manner with PrPC/PrPSc

complex is not yet understood.We quantitatively co-purified Arf1 with PrPC/PrPSc from

frontal cortex and cerebellum samples from sCJD subtypesMM1 and VV2, and age-matched controls using protein G

coupled to super magnetic Dynabeads® (Fig. 4). Eluates fromthis co-immunoprecipitation with SAF32 and SAF70 PrPC

were immunoblotted with Arf1 antibody to verify the interac-tive association between Arf1 and PrPC/PrPSc complex(Fig. 4a). Figure 4b showed the results of reverse co-immunoprecipitation with Arf1 antibody immunoblotted withSAF32 and SAF70 PrPC antibody.

A positive Arf1 signal was detected in frontal cortex andcerebellum samples from sCJD subtypes MM1 and VV2, andage-matched controls. Interestingly, the intensity of the Arf1signal showed significantly different region- and subtype-specific binding capacities for PrPC and the PrPC/PrPSc com-plex (Fig. 4c, d).

In the frontal cortex, subtype MM1 showed a higher bind-ing capacity of Arf1 for the PrPC/PrPSc complex while subtypeVV2 showed a significantly lower binding capacity of Arf1for the PrPC/PrPSc complex (Fig. 4c). However, in the cere-bellum, both subtypes showed a significant reduction of bind-ing capacity of Arf1 for the PrPC/PrPSc complex (Fig. 4a).Furthermore, reverse co-immunoprecipitation analysis withArf1 antibody followed by and immunoblotting with SAF32

Fig. 1 PrPC interaction with Arf1: primary cultures of mouse cortex(PCC) were prepared from pregnant mice at embryonic day 14; themice were prnp (the PrP-encoding gene) and knockout (PrP−/−). a Totalcellular lysate co-immunoprecipitated (IP) with Arf1 antibody (lane 1PrP−/− PCC and lane 2 PrPC expressing PCC) and 3F4 PrPC antibody(lane 3 PrP−/− PCC and lane 2 PrPC expressing PCC), and thenimmunoblotted with SAF70 PrPC antibody. c Reverse co-IP with Arf1

antibody (lane 1 PrP−/− PCC and lane 2 PrPC expressing PCC), and 3F4PrPC antibody (lane 3 PrP−/− PCC and lane 2 PrPC expressing PCC), andthen immunoblotted with Arf1 antibody. Eluate from this reverse co-IPverified PrPC interaction with Arf1 (b, d). Densitometry analysis fromfour independent (±SD) CO-IP-immunoblotting experiments. Thesignificance was calculated by one-way ANOVA Friedman test(**P<0.01, ***P<0.001)

J Mol Neurosci (2015) 56:329–348 335

PrPC and SAF70 showed the same pattern of binding capacityof Arf1 for the PrPC/PrPSc complex with slight modification ofthe extent of the detected expression of PrPC/PrPSc complex at27 to 37 kDa (Fig. 4d). This confirm that Arf1 was co-precipitating with the PrPC/PrPSc complex in a regional dis-ease subtype MM1 and VV2 manner and demonstrate thestrain-specific binding capacity of Arf1 for PrPC/PrPSc com-plex during disease progression.

Arf1 Interactive Regulatory Response at Pre-symptomaticand Symptomatic Stage of sCJD MM1 in Mice

As we previously mentioned, material from MM1 patientsshowed significant down-regulation of the PrPC/PrPSc com-plex in the frontal cortex of brain region compared to age-matched control samples (Llorens et al. 2013). Here, weshowed that the association of Arf1 with the PrPC/PrPSc

336 J Mol Neurosci (2015) 56:329–348

complex was significantly higher than in control and age-matched controls (Fig. 4). Interestingly, cerebellum samplesfrom the same patients showed decreased levels of bindingcapacity. We followed the changes of Arf1 from the pre-symptomatic to the symptomatic disease stage in mice to as-sess the regulatory role of Arf1 during the progression fromearly to late disease (Fig. 5). Data showed that in the cortex ofTg340 mice, the expression of Arf1 was about four times thelevel in human PrPM129 and in infected sCJDMM1 (MM1)10 % brain homogenates. We previously showed the follow-ing progression rate: pre-symptomatic (pre-clinical) 120 dayspost-infection (dpi), early clinical symptomatic (160 dpi), andlate clinical symptomatic (183 dpi) stages in the cortex andcerebellum (Llorens et al. 2014). Interestingly, Arf1 expres-sion was significantly decreased only in the cortex of MM1mice at the pre-symptomatic (120 dpi) and early clinical stage(160 dpi) (Fig. 5a, b). The late clinical stage at about 183 dpiand terminal stage at about 210 dpi showed no significantchange compared to age-matched control mice (Fig. 5a, b).However, in the cerebellum, the pre-symptomatic and symp-tomatic stages, there were no significant changes in the ex-pression of Arf1 protein (Fig. 5a, c).We also followed the total

PrP expression in the cortex and cerebellum of sCJD MM1-inoculated Tg340 mice throughout the pre-symptomatic andsymptomatic stages of disease progression and found a signif-icantly decreased PrP expression in the clinical stages (from160 dpi to the terminal stage at 210 dpi) only in the cortex ofsCJDMM1mice (Fig. 5a, d) and no significant regulatory PrPresponse in the cerebellum (Fig. 5a, e).

Arf1-Dependent Regulatory Response of MicrovesicleShedding in sCJD MM1 and VV2 Subtypesand in Pre-symptomatic and Symptomatic Stages of sCJDMM1 Mice

Microvesicle shedding requires the phosphorylation of myo-sin light chain (MLC) (Alexander et al. 2008; Muralidharan-Chari et al. 2009). It has been reported that inhibition ofMLCK or myosin II plays a role in invadopodia-associatedvesicular release (Alexander et al. 2008) and that Thr-18/Ser-19 residue phosphorylation of MLC regulates actin cytoskel-eton assembly which is required for the microvesicle fission(Muralidharan-Chari et al. 2009). Arf1 also was shown tocontrol the activity of both GTPase families of small GTP-binding proteins, the Rho family (Sahai and Marshall 2002;Wheeler and Ridley 2004), and leads to decreased shedding ofmicrovesicles (Schlienger et al. 2014). Although phosphory-lation of the Thr-18/Ser-19 residues of MLC is required formicrovesicle fission (Muralidharan-Chari et al. 2009), our aimwas to check this Arf1-dependent microvesicular responseand Arf1 expression in the frontal cortex and cerebellum inMM1 and VV2 patients. As illustrated in Fig. 6, MLC expres-sion in the frontal cortex of MM1 and VV2 patients was notsignificantly changed (Fig. 6a). However, MLC expression inthe cerebellum was significantly down-regulated in MM1sCJD patients (Fig. 6a). Nonetheless, the phosphorylated(activated) form was significantly down-regulated in the fron-tal cortex in MM1 and VV2 patients (Fig. 8a (a1, b1, c1)). Inthe cerebellum, phospho-MLC levels in subtype MM1 weredecreased to a greater degree than in age-matched control andVV2 patients (Fig. 8a (a2, b2, c2)). MLC expression in thecortex of sCJD MM1 inoculated Tg340 mice was also signif-icantly down-regulated in pre-symptomatic and early symp-tomatic stages (from 120 and 160 dpi). In contrast, cerebellarMLC expression in pre-symptomatic and early symptomaticstages (from 120 and 160 dpi) was significantly up-regulatedcompared to age-matched controls (Fig. 7a–c). Furthermore,sCJD MM1-inoculated Tg340 mice demonstrated a signifi-cantly decreased expression of phospho-MLC levels in pre-symptomatic and early symptomatic stages (from 120 and160–210 dpi) in the cortex (Fig. 8a (a4, b4)) and the cerebel-lum (Fig. 8a (a6, b6)). In the terminal stages of the disease at183 and 210 dpi, the sCJD MM1 mice showed no regulatoryresponse of phospho-MLC in either the cortex or the cerebel-lum (Fig. 8a (c4–d4, c6–d6)).

�Fig. 2 Effect of BFA on Arf1 and PrPC expression, co-localization, cellviability, and caspase-3 activity: primary cortical neuronal cultures ofmouse (PCC) were treated with BFA (1 μg/ml) for different timeintervals and immunoblotted (a) PCC: (i) untreated primary cultures ofmouse cortical neurons and (ii) 1.5 h of BFA treatment of primary culturesof mouse cortical neurons. PrPC and Arf1 distribution was analyzed using3F4 anti-PrPC (red), anti-Arf1 (green) antibodies and DAPI (blue-nuclearstaining). At least 25 cells were observed per condition per experiment foran equal exposure time (scale bar: 10 μm). The co-localization frequencyplots of the individual pixels from paired images and co-localizationfrequency graphs were generated by ImageJ (WCIF plugin) software. bPearson’s co-localization correlation coefficient rp (−1≤rp≤1) showedincrease in co-localization as compared to control untreated cells after1.5 h of BFA treatment and graph was generated by ImageJ (WCIFplugin) software. c PCC were prepared from pregnant mice atembryonic day 14; the mice were prnp (the PrP-encoding gene) andknockout (PrP−/−). Cultures were platted on polyethylenimine(1 mg/ml)-coated glass cover slips in culture wells and treated withBFA (1 μg/ml) for 1.5 h. Cell viability was measured by MTS assayand viability values are shown as absorbance at 490 nm. Data pointsare the means±SEM of values from four different experiments. ThePCC cultures expressing PrPC displayed elevated levels of cell viabilityunder BFA treatment as compared with PrP−/− cells. The significance wasperformed by one-way ANOVA Friedman test (*P<0.05). Caspase-3activity was detected by fluorescence measurement of the cleaved pNAfrom the substrate peptide DEVD-pNA.Data points are themeans±SD ofvalues from four different experiments. The PCC cultures expressingPrPC showed decreased caspase-3 cleavage activity as compared withthe cells with PrP−/− after being treated with BFA (1 μg/ml) for 1.5 h.The significance was performed by one-way ANOVA Friedman test(*P<0.05). d PrPC, Arf1, and beta actin (control) expression wasanalyzed after 1.5 h of BFA treatment by immunoblotting using specificSAF70 PrPC, Arf1, and beta actin antibodies. e, f Densitometry analysisfrom four independent (±SD) immunoblotting experiments. Thesignificance was calculated with Student’s t test (**P<0.01,***P<0.001)

J Mol Neurosci (2015) 56:329–348 337

It has already been reported that the phosphorylation ofMLC is regulated by the Rho family of GTPases and Rho-associated kinase signaling axis (Amano et al. 1996; Kimuraet al. 1996). We observed that RhoA regulation was not sig-nificantly altered in subtypes MM1 and VV2 in the frontalcortex and cerebellum of CJD patients (Fig. 6a, c). Interest-ingly, sCJDMM1-inoculated Tg340mice demonstrated a sig-nificant increase in the expression of RhoA at the pre-clinicalstage (120 dpi) in the cerebellum of sCJD MM1 mice only(Fig. 7a, d, e).

In addition, depletion of Arf1 causes inactivation of thePI3K pathway in MDA-MB-231 cells (Boulay et al. 2008)and Rac1 activation can depend on PI3K activation (for areview, see (Welch et al. 2003). Similarly, other groupshave reported that RhoA activation can be regulated bythe PI3K signaling axis (Qiang et al. 2004; Kakinumaet al. 2008). Together, these observations suggest that

Arf1 might control Rac1 and RhoA/C activation by amechanism involving the PI3K pathway. PI3K kinase 85forms showed significant down-regulation in the cerebel-lum of MM1 patients (Fig. 9a, b). However, the PI3K ki-nase 55 form showed down-regulation in the frontal cortexof MM1 patients and in the cerebellum of both MM1 andVV2 patients (Fig. 9a, c). Interestingly, the expression lev-el of PkB was only significantly up-regulated in the frontalcortex of both subtypes (MM1 and VV2) (Fig. 9a–c). Inthe cerebellum, no PkB was detectable by Western blotanalysis (Fig. 9a, d). Furthermore, sCJD MM1-inoculatedTg340 mice demonstrated a significant increase in the ex-pression of PI3K kinase 85 form in the pre-symptomaticstage (120 dpi) in the cortex of sCJD MM1 mice only(Fig. 10a, b). There was a significant increase in the ex-pression of PI3K kinase 85 form in the cerebellum in thepre-symptomatic and symptomatic stages (from 120 to 210

Table 2 Summary of the findings in sCJD cases and at the pre-symptomatic and symptomatic stages of sCJD MM1 in mice

Summary of the findings related to level of Arf1, Rho-A, MLC2, P-MLC2, PI3-85, PI3-55, and PkB in sCJD cases and at the pre-symptomatic andsymptomatic stages of sCJD MM1 in mice

338 J Mol Neurosci (2015) 56:329–348

dpi terminal stage). In contrast, the PI3K kinase 55 formshowed a significantly decreased expression in the pre-symptomatic and early symptomatic stages (120 and 160dpi) in the cortex of sCJD MM1 mice only (Fig. 10a, d).However, the cerebellum showed a homogeneous regula-tory response in the pre-symptomatic and symptomaticstages (from 120 to the 210 dpi terminal stage) in sCJD

MM1-inoculated Tg340 mice (Fig. 10a, d, e). Furthermore,the PkB form showed a significant down-regulation in thecortex in the pre-symptomatic and symptomatic stages(120 to 183 dpi) (Fig. 10a, f). In contrast, the cerebellumof MM1 patients showed up-regulation in the pre-symptomatic and symptomatic stages (120 to 183 dpi)(Fig. 10a, g).

Fig. 3 Arf1 expression in sCJDMM1 and VV2 subtypes:Western blotting analysis of Arf1in four representative cases eachfor control, sCJDMM1 and sCJDVV2 cases; GAPDHimmunostaining was used tonormalize total protein loading (a)in the frontal cortex (b) incerebellum (c, d). Densitometryanalysis from four independent(±SD) immunoblottingexperiments of 15 control (Con),15 MM1 and 15 VV2 cases. Thesignificance was calculated withone-way ANOVA Friedman test(*P<0.05). Region-dependentsignificantly decreasedexpression is found in sCJD incerebellum as compared tocontrols. AU arbitrary units

Fig. 4 Arf1 interaction with PrPC and PrPSc in sCJD MM1 and VV2subtypes: quantitative co-purification of Arf1 with PrPC/PrPSc usingDynabeads G-protein coupled magnetic beads in frontal cortex andcerebellum samples from sCJD cases subtypes MM1 and VV2, andage-matched controls. a Eluates from co-immunoprecipitate withSAF32 PrPC were immunoblotted with Arf1 antibody. b Reverse co-

immunoprecipitation with Arf1 antibody immunoblotted with SAF32PrPC antibody. c, d Densitometry analysis from four independent (±SD)immunoblotting experiments from control, MM1, and VV2 cases. Thesignificance was calculated with Student’s t test (*P<0.05, **P<0.01,***P<0.001, ****P<0.0001)

J Mol Neurosci (2015) 56:329–348 339

Discussion

The emerging field of interactomics has led to the identifica-tion of several new proteins that interact with PrPC.Interactomic studies could possibly be useful to uncover theyet unknown function of PrPC. In a recent paper, we identified43 proteins that interacted with PrPC (Zafar et al. 2011) and 23proteins that interacted with the truncated form of PrPC (Zafaret al. 2014). Our attempts to further characterize the interactiveassociation of Arf1 in terms of PrPC expression led to a num-ber of interesting observations. Arfs are known regulators ofconstitutive membrane trafficking and localize to early/cis-Golgi and ER–Golgi intermediate compartments, an impor-tant factor for the secretory pathway and receptor-mediatedregulation (Boulay et al. 2008; Zafar et al. 2011; Cohenet al. 2007). Vesicle formation requires Arfs and GEP (gua-nine nucleotide-exchange protein, an activator of Arf1). TheseGEP catalyze cycling between inactive, Arf-GDP (largely

cytosolic), and GTP-bound active (membrane-associated)forms (Puxeddu et al. 2009), and Arf1 interacts with PrPC

(Zafar et al. 2011). Experiments were designed to investigatewhether Arf1-dependent vesicle formation has any role in thetransportation or localization of PrPC. Furthermore, the ex-pressional regulation of the Arf1 pathway in the brain (frontalcortex and cerebellum) of CJD subtype and during diseaseprogression from the early pre-clinical to the terminal clinicalstage in MM1 CJD-infected M129 human PrP overexpressingmice showed an altered response under pathophysiologicalconditions.

Most importantly, we provide new evidence that Arf1 reg-ulates in an early pre-clinical stage and in a brain region-specific manner during the disease progression supporting astrain-specific regional involvement of brain regions duringMM1 and VV2 subtype CJD disease progression. In themeantime, RhoA also shows destabilized activity leading tomyosin light-chain (MLC) phosphorylation and deregulation

Fig. 5 Expression of Arf1 andPrPC/PrPSc at pre-symptomaticand symptomatic stage of sCJDMM1 in mice: Tg340 miceexpressing about fourfold level ofhuman PrPM129 were infectedwith sCJD MM1 (MM1) 10 %brain homogenates. Samples werecollected at indicated days post-infection (dpi): pre-symptomaticstage (120 dpi) and symptomaticstages (160, 180, 210, and 240dpi). aArf1 and PrP expression inthe cortex and cerebellum wereobserved byWestern blot analysisusing Arf1 and Saf32 antibodies.b–e Densitometry analysis fromfour independent (±SD)immunoblotting experiments ofArf1 and PrPC/PrPSc Saf32 fromcontrol and sCJD MM1-infectedPrPM129 mice in cortex andcerebellum. The significance wascalculated with one-way ANOVAFriedman test (*P<0.05)

340 J Mol Neurosci (2015) 56:329–348

of the PI3K pathway and demonstrates early disease progres-sion and internal homeostasis of misfolded proteins.

PrPC Expression Significantly Impairs IntracellularTransport Under Altered Arf1 Expression

The potential involvement of Arf1 interactions in PrPC local-ization and expression was examined by inhibiting GEP withBFA. BFA is known to rapidly and fully reversibly block theintracellular transport of newly synthesized cellular proteinsimmediately distal to the distinct compartment of the ER(Ulmer and Palade 1991). Increasing concentrations of BFAtreatment were studied and showed down-regulation ofphorbol 12-myristate 13-acetate (PMA)-induced secretion ofmetalloproteinase (MMP)-9 from fibrosarcoma cells (Ho et al.2003). In PrPC expressing PCC, it showed a marked decline ofPrPC expression after BFA treatment. Using confocal laser co-immunofluorescence scanning techniques, we found thatBFA-induced Arf1 deactivation altered the sub-cellular local-ization and expression of PrPC. Based on the model of Arf1function in intra-Golgi trafficking, it is proposed that Arf1-GTP stimulates retrograde transport of PrPC molecules withinthe trans-Golgi compartment toward the ER and also to theplasma membrane. The disruption of Arf1 function by BFAduring intra-Golgi trafficking may stimulate retrograde trans-port toward the ER and decrease the expression of Arf1,which ultimately disrupts the transport of PrPC toward theplasma membrane. The decreased co-localization of Arf1and PrPC after BFA treatment compared to control, untreatedcells supports this proposition.

Intracellular proteins are processed by the Golgi apparatus,and in several neurodegenerative disorders such asAlzheimer’s disease, these Golgi stacks are hypothesized toundergo disassembly and fragmentation (Nakagomi et al.2008). This disassembly of the Golgi apparatus may disruptthe normal trans-Golgi network (TGN). Microtubules andmicrovesicles are required to maintain the organization ofthe Golgi apparatus (Rogalski and Singer 1984) and theTGN. The Golgi apparatus and the TGN are sites for sortingand secretory vesicle biogenesis, and are critically involved inthe transport of proteins to the plasmamembrane (Burgess andKelly 1984; Burgess et al. 1991). Taken together, these find-ings indicate that the Golgi apparatus plays an essential role inPrPC transport and that Arf1-dependent microvesicle transpor-tation (Schlienger et al. 2014) may complement this Arf1/PrPC pathway.

Cell Viability and Caspase-3 Activity in PCC Cells

Our analysis of cell viability and the caspase-3 activity path-way by which Arf1 may control PrPC expressional regulationand disrupt intracellular transport in PCC showed a down-regulation of apoptotic markers and demonstrated the neuro-protective function of PrPC (Zafar et al. 2014; Wu et al. 2008).Here, we studied the neuroprotective function of human PrPC

under an altered Arf1 expressional regulatory pathway. Inter-estingly, we found that PCC expressing PrPC with altered Arf1expression showed more viability capacity than cells withoutPrPC. During treatment with BFA, we found a relationshipbetween cells expressing PrPC and PrPC knockout cells and

Fig. 6 MLC2 and Rho-Aregional specific expressionalregulation in sCJD MM1 andVV2 subtypes: aWestern blottinganalysis of MLC and Rho-A infour representative cases each forcontrol, sCJD MM1, and sCJDVV2 cases, GAPDHimmunostaining was used tonormalize total protein loading.b–g Densitometry analysis fromfour independent (±SD)immunoblotting experiments of15 control (Con), 15MM1 and 15VV2 cases. The significance wascalculated with one-way ANOVAFriedman test (*P<0.05). AUarbitrary units

J Mol Neurosci (2015) 56:329–348 341

the viability/caspase-3 activity response under altered Arf1expression. Recent reports also showed that under PrPSc stressconditions, PrP induced neurotoxicity response by the markedalterations of PI3K/Akt/GSK-3, JNK, and caspase 12 activa-tion pathways (Simon et al. 2014).

A different study suggested that PrPC plays a protective roleduring oxidative stress and a proapoptotic role during ERstress-induced apoptotic cell death in response to thecaspase-3-dependent proteolytic activation of PKCδ. In addi-tion, exposure to BFA did not induce an increase in ROSgeneration for up to 60 min in cells expressing PrPC, suggest-ing that ROS are a key mediator of the H2O2-induced but notthe BFA-induced cytotoxic response (Anantharam et al. 2008).

Our study demonstrates that the viability of cells andcaspase-3 activity is PrPC- and Arf1-interaction dependent.MTS assay and caspase 3 analysis showed comparable results.

BFA-induced caspase-3 changes were suppressed in PrPC ex-pressing but not in PrP knockout cells. In a similar manner, thePrPC expressing cells displayed a markedly greater viabilitythan PrP knockout cells under BFA-induced altered Arf1 ex-pression. Together, these results suggest that cellular PrP de-creases the vulnerability to oxidative insults, and down-regulates downstream mediators of cellular stress-inducedneuronal apoptosis.

Regulation of Arf1/RhoA/MLC/PI3K PathwayExpression in sCJD MM1 and VV2 Subtypes

Many recent reports have suggested that synaptic impairmentcould be the mechanism by which cognitive dysfunction de-velops during brain aging and neurodegeneration, and that itcould involve the small Rho GTPases (Afshordel et al. 2014).

Fig. 7 Expression of MLC andRho-A at pre-symptomatic andsymptomatic stages of sCJDMM1 mice: a Western blottinganalysis of MLC and Rho-A inTg340 sCJDMM1mice with fourrepresentative cases each forcontrol, sCJD MM1 and sCJDVV2 cases; GAPDHimmunostaining was used tonormalize total protein loading.b–e Densitometry analysis fromfour independent (±SD)immunoblotting experimentsfrom control and sCJD MM1-infected PrPM129 mice in cortexand cerebellum. Tg340 sCJDMM1 mice showed significantaltered regulation of MLC andRho-A expression in the cortexand cerebellum by Western blotanalysis. The significance wascalculated with one-way ANOVAFriedman test (*P<0.05). AUarbitrary units

342 J Mol Neurosci (2015) 56:329–348

Rho GTPases are molecular switches of intracellular signalingprocesses (DeGeer and Lamarche-Vane 2013) and their activ-ity is reported to be regulated by Arf1 (Schlienger et al. 2014).Though the roles of Arf1 and Rho GTPases during neuraldevelopment have been well documented, their participationduring neurodegeneration has been far less well characterized.Herein, we discuss our current knowledge of the role andfunction of Arf1/Rho GTPases and regulators during neuro-degeneration and highlight their potential as targets for thera-peutic intervention in common neurodegenerative disorders(Fig. 11).

Arf1 expression was significantly decreased in the cerebel-lum of both subtypes of sCJD MM1 and VV2 cases andshowed regional specific responses. Also, co-purification ofArf1 with PrPC in frontal cortex and cerebellum samples fromsCJD patients with MM1 and VV2 subtypes, and age-matched controls, showed quantitative regulation in subtype-specific manner. This specific interaction of Arf1 or structuralmodification of PrPC/ PrPSc complex during the developmentof pathological symptoms controls the PI3K pathway (Boulayet al. 2008; Schlienger et al. 2014; Cohen et al. 2007). Asalready reported, Arf1 controls the PI3K pathway and

Fig. 8 Phospho-MLC (Thr18/Ser19) dot blot immunoscreening of thesCJD MM1 and VV2 subtype patients and sCJD MM1-infected mice atpre-symptomatic and symptomatic stages: phospho-MLC (Thr18/Ser19)dot blot immunoscreening of the sCJD MM1 and VV2 subtypes patientfrontal cortex (lane a1=control, b1=MM1 sCJD, c1=VV2 sCJD),cerebellum (lane a2=control, b2=MM1 sCJD, c2=VV2 sCJD) region.sCJD MM1-infected mouse cortex at pre-symptomatic and symptomaticstages (lane a3=control 120 dpi, b3=control 160 dpi, c3=control 183dpi, d3=control 210 dpi, a4=sCJD MM1-infected 120 dpi, b4=sCJDMM1-infected 160 dpi, c4=sCJD MM1-infected 183 dpi, d4=sCJD

MM1-infected 210 dpi). Cerebellum regulatory response phospho-MLC(Thr18/Ser19) at pre-symptomatic and symptomatic stages (lane a5=control 120 dpi, b5=control 160 dpi, c5=control 183 dpi, d5=control210 dpi; a5=sCJD MM1-infected 120 dpi, b5=sCJD MM1-infected160 dpi, c5=sCJD MM1-infected 183 dpi, d5=sCJD MM1-infected210 dpi). Total lysate (50 μg) was applied onto PVDF membrane andspleen from c57BL6 mice used as positive control (lane d1 and d2). b–dDensitometry analysis from four independent (±SD) immunoblottingexperiments. The significance was calculated with one-way ANOVAFriedman test (*P<0.05). AU arbitrary units

J Mol Neurosci (2015) 56:329–348 343

proliferation by regulating the function of pRB (Boulay et al.2008) and further regulating metalloproteinase (MMP) activ-ity by using BFA to inhibit phorbol 12-myristate 13-acetate(PMA)-induced secretion of MMP-9 from HT 1080 fibrosar-coma cells (Ho et al. 2003). It selectively blocks MMP-9 andexhibits specificity toward a particular MMP (Schlienger et al.2014).

Our findings show a region-specific response with asignificant decrease of the BFA-sensitive Arf isoform onlyin the cerebellum of sCJD MM1 and VV2 patients. Re-gional and temporal studies in sCJD MM1 mice onlyshowed a significant decrease in the cortex of MM1 micein the pre-symptomatic and early clinical stages of the

sCJD MM1-induced disease. There was no significant al-teration in the cortex or the cerebellum in the final clinicalstage, and this is comparable with the human disease formin the cortical region only. However, the expression of PrPis altered at a later stage of the disease. Arf1 also appears tobe the early target in the sCJD MM1-induced strain-specif-ic disease form.

Arf1 was also shown to be the key regulator forGTPases, particularly the small Rho family (Sahai andMarshall 2002; Wheeler and Ridley 2004) by reducingthe shedding of microvesicles (Schlienger et al. 2014)and, in addition, that microvesicle fission required thephosphorylation of MLC on Thr-18/Ser-19 (Alexander

Fig. 9 p85 and p55 form of PI3 kinase and PkB expressional regulationin sCJDMM1 andVV2 subtypes: aWestern blotting analysis of p85, p55form of PI3 kinase, and PkB in four representative cases each for control,sCJD MM1 and sCJD VV2 cases. b–d Densitometry analysis from fourindependent (±SD) immunoblotting experiments of 15 control (Con), 15MM1, and 15 VV2 cases. Significant down-regulation of PI3 kinase p85was observed only in MM1 sCJD in cerebellum as compared to controls

and VV2 cases. PI3 kinase p55 form showed significant down-regulationinMM1 sCJD in frontal cortex and also in both subtypes in cerebellum ascompared to controls. Significant up-regulation of PkB was observedonly in frontal cortex in both subtypes (MM1 and VV2) of sCJD ascompared to controls. AU arbitrary units. The significance wascalculated with one-way ANOVA Friedman test (*P<0.05, **P<0.01)

344 J Mol Neurosci (2015) 56:329–348

et al. 2008; Alexander et al. 2008; Muralidharan-Chariet al. 2009). It is already reported that Thr-18/Ser-19 phos-phorylation of MLC regulates actin cytoskeleton assemblywhich is required for microvesicle fission (Muralidharan-Chari et al. 2009). Regulation of stress fiber assembly isa lso rock2-dependent and phosphoryla tes MLC(Totsukawa et al. 2000). In our regional and temporal stud-ies, MLC expression in the frontal cortex did not show any

significant regulation in contrast to the cerebellum in MM1sCJD patients, in whom MLC expression was significantlydown-regulated. The phosphorylated (activated) form wassignificantly decreased in the frontal cortex in MM1 andVV2 patients and decreased only in the cerebellum inMM1 patients. Time course study in sCJD MM1-inoculated Tg340 mice showed a significant decrease inthe cortical expression of phospho-MLC levels in at pre-

Fig. 10 Expressional regulationof p85 and p55 form of PI3 kinaseand pkB at pre-symptomatic andsymptomatic stages of sCJDMM1 mice: a Western blottinganalysis of Tg340 sCJD MM1mice showed significant up-regulation of PI3 kinase p85expression only in cortex at pre-symptomatic stage (120 dpi) andalso early clinical stage (160 dpi).In cerebellum, PI3 kinase p55form showed significant down-regulation at symptomatic stages(160 and 180 dpi) as compared tocontrols. Tg340 sCJD MM1 miceshowed significant down-regulation of pkB expression onlyin cortex at pre-symptomaticstage (120 dpi) and also clinicalstage (160 and 183 dpi). Incerebellum, pkB showedsignificant up-regulation at pre-symptomatic stage (120 dpi) andalso clinical stage (160 and 183dpi) as compared to controls. b–fDensitometry analysis from fourindependent (±SD)immunoblotting experiments ofp85 and p55 form of PI3 kinaseand PkB from control and sCJDMM1-infected PrPM129 mice incortex and cerebellum. Thesignificance was calculated withone-way ANOVA Friedman test(*P<0.05)

J Mol Neurosci (2015) 56:329–348 345

clinical stages (from 120 and 183 dpi) (Fig. 8), and signif-icant cerebellar regulatory response of MLC in the pre-symptomatic stage. Phosphorylation of MLC is regulatedby the Rho family and while RhoA did not significantlychange in the frontal cortex or the cerebellum of CJD pa-tients with either the MM1 or the VV2 subtype, it is inter-esting to note that CJD MM1-inoculated Tg340 miceshowed a significant increase in the expression of RhoAin the cerebellum in the pre-clinical stage. In addition, de-pletion of Arf1 leads to inactivation of the PI3K pathway inMDA-MB-231 cells (Boulay et al. 2008), and Rac1 acti-vation can depend on PI3K activation (Welch et al. 2003).Similarly, other different research groups have reportedthat RhoA activation can be regulated by the PI3K signal-ing axis (Qiang et al. 2004; Kakinuma et al. 2008). Togeth-er, these observations suggest that Arf1 might control Rac1and RhoA/C activation by a mechanism involving thePI3K pathway. PI3K showed significant down-regulationin the cerebellum of MM1 patients, but there was no sig-nificant change in the frontal cortex. Furthermore, therewas a significant deregulation of PI3K and PkB expressionin the cortex and cerebellum of sCJD MM1-inoculatedTg340 mice in pre-clinical stages. In addition, alterationsof PI3K/Akt/GSK-3, JNK, and caspase 12 activation path-ways (Simon et al. 2014), the lower reported levels ofglycogen synthase kinase 3β (GSK3 β) with up-regulation of CDK5 in the scrapie-infected hamster braintissues (Wang et al. 2010). The drastic reduction of totaland phosphorylated GSK3 in the frontal cortex of bothsCJD subtypes (Llorens et al. 2014) demonstrated a possi-ble up-stream kinase regulatory response which is detect-able in the early pre-symptomatic stages of the disease, andwhich could be a potential early diagnostic approach usefulto reveal therapeutic strategy for misfolded proteinderegulatory response.

Conclusion

In conclusion, this study demonstrates for the first time tem-poral regional hallmarks from pre-symptomatic stages of thedisease through the clinical and terminal stages in a strain-specific subtype manner. The results showed a significant ear-ly pre-clinical response of transporter proteins in most preva-lent sCJD MM1 disease subtypes. This could be an earlyindicator of disease progression. In addition, the molecularpathways which showed deregulation in association with al-tered Arf1 pathway with destabilizes RhoA activity are myo-sin light-chain (MLC) phosphorylation and Arf1-dependentPI3K/PkB pathways. Taken together, our findings underscorea key early symptomatic role of Arf1 in neurodegenerationand specifically as an early regulatory response, and they sug-gest that targeting the Arf/Rho/MLC signaling axis might be apromising strategy to uncover the missing link which quitepossibly influences disease progression and the internal ho-meostasis of misfolded proteins.

Acknowledgments Special thanks to Dr Torres at CISA INIA whoproduced the tg340 mice.

References

Afshordel S, Wood WG, Igbavboa U, Muller WE, Eckert GP (2014)Impaired geranylgeranyltransferase-I regulation reducesmembrane-associated Rho protein levels in aged mouse brain. JNeurochem 129:732–742. doi:10.1111/jnc.12654

Aguzzi A (2000) Molecular pathology of prion diseases. Vox Sang78(Suppl 2):25

Aguzzi A, O’Connor T (2010) Protein aggregation diseases: pathogenic-ity and therapeutic perspectives. Nat Rev Drug Discov 9:237–248.doi:10.1038/nrd3050

Alexander NR, Branch KM, Parekh A, Clark ES, Iwueke IC, GuelcherSA, Weaver AM (2008) Extracellular matrix rigidity promotes

Fig. 11 Schematic representation of molecular events of ADPribosylation factor-1 dependent Rho/MLC pathway: schematicrepresentation of molecular events of ADP ribosylation factor-1-

dependent Rho/MLC pathway during disease progression in sCJDMM1 and VV2 subtypes and at pre-symptomatic and symptomaticstages of sCJD MM1 mice

346 J Mol Neurosci (2015) 56:329–348

invadopodia activity. Curr Biol 18:1295–1299. doi:10.1016/j.cub.2008.07.090

Amano M, Ito M, Kimura K, Fukata Y, Chihara K, Nakano T, MatsuuraY, Kaibuchi K (1996) Phosphorylation and activation of myosin byRho-associated kinase (Rho-kinase). J Biol Chem 271:20246–20249

Anantharam V, Kanthasamy A, Choi CJ, Martin DP, LatchoumycandaneC, Richt JA, Kanthasamy AG (2008) Opposing roles of prion pro-tein in oxidative stress- and ER stress-induced apoptotic signaling.Free Radic Biol Med 45:1530–1541. doi:10.1016/j.freeradbiomed.2008.08.028

Boulay PL, Claing A (2009) [ARF proteins: molecular switches control-ling tumour proliferation and metastasis]. Med Sci (Paris) 25:783–785. doi:10.1051/medsci/20092510783

Boulay PL, Cotton M, Melancon P, Claing A (2008) ADP-ribosylationfactor 1 controls the activation of the phosphatidylinositol 3-kinasepathway to regulate epidermal growth factor-dependent growth andmigration of breast cancer cells. J Biol Chem 283:36425–36434.doi:10.1074/jbc.M803603200

Bucci C, Thomsen P, Nicoziani P, McCarthy J, van Deurs B (2000) Rab7:a key to lysosome biogenesis. Mol Biol Cell 11:467–480

Bueler H, Fischer M, Lang Y, Bluethmann H, Lipp HP, DeArmond SJ,Prusiner SB, Aguet M, Weissmann C (1992) Normal developmentand behaviour of mice lacking the neuronal cell-surface PrP protein.Nature 356:577–582. doi:10.1038/356577a0

Burgess TL, Kelly RB (1984) Sorting and secretion of adrenocorticotro-pin in a pituitary tumor cell line after perturbation of the level of asecretory granule-specific proteoglycan. J Cell Biol 99:2223–2230

Burgess TL, Skoufias DA, Wilson L (1991) Disruption of the Golgiapparatus with brefeldin A does not destabilize the associateddetyrosinated microtubule network. Cell Motil Cytoskeleton 20:289–300. doi:10.1002/cm.970200405

Campa F, Randazzo PA (2008) Arf GTPase-activating proteins and theirpotential role in cell migration and invasion. Cell Adh Migr 2:258–262

Carimalo J, Cronier S, Petit G, Peyrin JM, Boukhtouche F, Arbez N,Lemaigre-Dubreuil Y, Brugg B, Miquel MC (2005) Activation ofthe JNK-c-Jun pathway during the early phase of neuronal apoptosisinduced by PrP106-126 and prion infection. Eur J Neurosci 21:2311–2319. doi:10.1111/j.1460-9568.2005.04080.x

Chen LQ, Xiao SJ, Hu PP, Peng L, Ma J, Luo LF, Li YF, Huang CZ(2012) Aptamer-mediated nanoparticle-based protein labeling plat-form for intracellular imaging and tracking endocytosis dynamics.Anal Chem 84:3099–3110. doi:10.1021/ac202810b

Clark EA, Golub TR, Lander ES, Hynes RO (2000) Genomic analysis ofmetastasis reveals an essential role for RhoC. Nature 406:532–535.doi:10.1038/35020106

Cockcroft S, Thomas GM, Fensome A, Geny B, Cunningham E, Gout I,Hiles I, Totty NF, Truong O, Hsuan JJ (1994) Phospholipase D: adownstream effector of ARF in granulocytes. Science 263:523–526

Cohen LA, Honda A, Varnai P, Brown FD, Balla T, Donaldson JG (2007)Active Arf6 recruits ARNO/cytohesin GEFs to the PM by bindingtheir PH domains. Mol Biol Cell 18:2244–2253. doi:10.1091/mbc.E06-11-0998

Cory AH, Owen TC, Barltrop JA, Cory JG (1991) Use of an aqueoussoluble tetrazolium/formazan assay for cell growth assays in culture.Cancer Commun 3:207–212

Dagdanova A, Ilchenko S, Notari S, YangQ, ObrenovichME, Hatcher K,McAnulty P, Huang L, Zou W, Kong Q, Gambetti P, Chen SG(2010) Characterization of the prion protein in human urine. J BiolChem 285:30489–30495. doi:10.1074/jbc.M110.161794

Daniel B, DeCoster MA (2004) Quantification of sPLA2-induced earlyand late apoptosis changes in neuronal cell cultures using combinedTUNEL and DAPI staining. Brain Res Brain Res Protol 13:144–150. doi:10.1016/j.brainresprot.2004.04.001

DeGeer J, Lamarche-Vane N (2013) Rho GTPases in neurodegenerationdiseases. Exp Cell Res 319:2384–2394. doi:10.1016/j.yexcr.2013.06.016

Fletcher DA, Mullins RD (2010) Cell mechanics and the cytoskeleton.Nature 463:485–492. doi:10.1038/nature08908

Fritz JL, VanBerkum MF (2002) Regulation of rho family GTPases isrequired to prevent axons from crossing the midline. Dev Biol 252:46–58

Gambetti P, Kong Q, Zou W, Parchi P, Chen SG (2003) Sporadic andfamilial CJD: classification and characterisation. Br Med Bull 66:213–239

Goldmann W (1993) PrP gene and its association with spongiform en-cephalopathies. Br Med Bull 49:839–859

Gorodinsky A, Harris DA (1995) Glycolipid-anchored proteins in neuro-blastoma cells form detergent-resistant complexes without caveolin.J Cell Biol 129:619–627

Harris DA (2003) Trafficking, turnover and membrane topology of PrP.Br Med Bull 66:71–85

Ho WT, Exton JH, Williger BT (2003) Arfaptin 1 inhibits ADP-ribosylation factor-dependent matrix metalloproteinase-9 secretioninduced by phorbol ester in HT 1080 fibrosarcoma cells. FEBS Lett537:91–95

Kakinuma N, Roy BC, Zhu Y, Wang Y, Kiyama R (2008) Kank regulatesRhoA-dependent formation of actin stress fibers and cell migrationvia 14-3-3 in PI3K-Akt signaling. J Cell Biol 181:537–549. doi:10.1083/jcb.200707022

Katsube M, Shiota Y, Harada T, Shibata H, Nagai A (2013) A case ofMM1+2 Creutzfeldt-Jakob disease with a longitudinal study ofEEG and MRI. Rinsho Byori 61:995–1000

Kawamoto JC, Barrett JN (1986) Cryopreservation of primary neuronsfor tissue culture. Brain Res 384:84–93

Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M,Yamamori B, Feng J, Nakano T, Okawa K, Iwamatsu A, KaibuchiK (1996) Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 273:245–248

Knight RS, Will RG (2004) Prion diseases. J Neurol NeurosurgPsychiatry 75(Suppl 1):i36–i42

Knusel B, Michel PP, Schwaber JS, Hefti F (1990) Selective and nonse-lective stimulation of central cholinergic and dopaminergic develop-ment in vitro by nerve growth factor, basic fibroblast growth factor,epidermal growth factor, insulin and the insulin-like growth factors Iand II. J Neurosci 10:558–570

Lewis-Saravalli S, Campbell S, Claing A (2013) ARF1 controls Rac1 sig-naling to regulate migration of MDA-MB-231 invasive breast cancercells. Cell Signal 25:1813–1819. doi:10.1016/j.cellsig.2013.05.011

Linden R, Martins VR, Prado MA, Cammarota M, Izquierdo I, BrentaniRR (2008) Physiology of the prion protein. Physiol Rev 88:673–728. doi:10.1152/physrev.00007.2007

Llorens F, Ansoleaga B, Garcia-Esparcia P, Zafar S, Grau-Rivera O,Lopez-Gonzalez I, Blanco R, Carmona M, Yague J, Nos C, DelRio JA, Gelpi E, Zerr I, Ferrer I (2013) PrP mRNA and proteinexpression in brain and PrP(c) in CSF in Creutzfeldt-Jakob diseaseMM1 and VV2. Prion 7:383–393. doi:10.4161/pri.26416

Llorens F, Zafar S, Ansoleaga B, ShafiqM, Blanco R, CarmonaM, Grau-Rivera O, Nos C, Gelpi E, Del Rio JA, Zerr I, Ferrer I (2014)Subtype and regional regulation of prion biomarkers in sporadicCreutzfeldt-Jakob disease. Neuropathol Appl Neurobiol. doi:10.1111/nan.12175

Ma J, Wollmann R, Lindquist S (2002) Neurotoxicity and neurodegener-ation when PrP accumulates in the cytosol. Science 298:1781–1785.doi:10.1126/science.1073725

Magalhaes AC, Silva JA, Lee KS, Martins VR, Prado VF, Ferguson SS,Gomez MV, Brentani RR, Prado MA (2002) Endocytic intermedi-ates involved with the intracellular trafficking of a fluorescent cel-lular prion protein. J Biol Chem 277:33311–33318. doi:10.1074/jbc.M203661200

J Mol Neurosci (2015) 56:329–348 347

Muralidharan-Chari V, Clancy J, Plou C, Romao M, Chavrier P, RaposoG, D’Souza-Schorey C (2009) ARF6-regulated shedding of tumorcell-derived plasma membrane microvesicles. Curr Biol 19:1875–1885. doi:10.1016/j.cub.2009.09.059

Nakagomi S, Barsoum MJ, Bossy-Wetzel E, Sutterlin C, Malhotra V,Lipton SA (2008) A Golgi fragmentation pathway in neurodegen-eration. Neurobiol Dis 29:221–231. doi:10.1016/j.nbd.2007.08.015

Naslavsky N, Stein R, Yanai A, Friedlander G, Taraboulos A (1997)Characterization of detergent-insoluble complexes containing thecellular prion protein and its scrapie isoform. J Biol Chem 272:6324–6331

Ordonez-Gutierrez L, Torres JM, Gavin R, Anton M, Arroba-EspinosaAI, Espinosa JC, Vergara C, Del Rio JA, Wandosell F (2013)Cellular prion protein modulates beta-amyloid deposition in agedAPP/PS1 transgenic mice. Neurobiol Aging. doi:10.1016/j.neurobiolaging.2013.05.019

Padilla D, Beringue V, Espinosa JC, Andreoletti O, Jaumain E, Reine F,Herzog L, Gutierrez-Adan A, Pintado B, LaudeH, Torres JM (2011)Sheep and goat BSE propagate more efficiently than cattle BSE inhuman PrP transgenic mice. PLoS Pathog 7:e1001319. doi:10.1371/journal.ppat.1001319

Parchi P, Giese A, Capellari S, Brown P, Schulz-Schaeffer W, Windl O,Zerr I, Budka H, Kopp N, Piccardo P, Poser S, Rojiani A,Streichemberger N, Julien J, Vital C, Ghetti B, Gambetti P,Kretzschmar H (1999) Classification of sporadic Creutzfeldt-Jakobdisease based on molecular and phenotypic analysis of 300 subjects.Ann Neurol 46:224–233

Pille JY, Denoyelle C, Varet J, Bertrand JR, Soria J, Opolon P, Lu H,Pritchard LL, Vannier JP, Malvy C, Soria C, Li H (2005) Anti-RhoAand anti-RhoC siRNAs inhibit the proliferation and invasiveness ofMDA-MB-231 breast cancer cells in vitro and in vivo. Mol Ther 11:267–274. doi:10.1016/j.ymthe.2004.08.029

Popoff V, Langer JD, Reckmann I, Hellwig A, Kahn RA, Brugger B,Wieland FT (2011) Several ADP-ribosylation factor (Arf) isoformssupport COPI vesicle formation. J Biol Chem 286:35634–35642.doi:10.1074/jbc.M111.261800

Prusiner SB (1998) Prions. Proc Natl Acad Sci U S A 95:13363–13383Puxeddu E, Uhart M, Li CC, Ahmad F, Pacheco-Rodriguez G,

Manganiello VC, Moss J, Vaughan M (2009) Interaction of phos-phodiesterase 3A with brefeldin A-inhibited guanine nucleotide-exchange proteins BIG1 and BIG2 and effect on ARF1 activity.Proc Natl Acad Sci U S A 106:6158–6163. doi:10.1073/pnas.0901558106

QiangYW,Yao L, Tosato G, Rudikoff S (2004) Insulin-like growth factorI induces migration and invasion of human multiple myeloma cells.Blood 103:301–308. doi:10.1182/blood-2003-06-2066

Rogalski AA, Singer SJ (1984) Associations of elements of the Golgiapparatus with microtubules. J Cell Biol 99:1092–1100

Sabe H, Onodera Y, Mazaki Y, Hashimoto S (2006) ArfGAP familyproteins in cell adhesion, migration and tumor invasion. Curr OpinCell Biol 18:558–564. doi:10.1016/j.ceb.2006.08.002

Sahai E, Marshall CJ (2002) RHO-GTPases and cancer. Nat Rev Cancer2:133–142. doi:10.1038/nrc725

Schlienger S, Campbell S, Claing A (2014) ARF1 regulates the Rho/MLC pathway to control EGF-dependent breast cancer cell inva-sion. Mol Biol Cell 25:17–29. doi:10.1091/mbc.E13-06-0335

Schneider B, Pietri M, Pradines E, Loubet D, Launay JM, Kellermann O,Mouillet-Richard S (2011) Understanding the neurospecificity ofPrion protein signaling. Front Biosci 16:169–186

Schurmann A, Breiner M, Becker W, Huppertz C, Kainulainen H,Kentrup H, Joost HG (1994) Cloning of two novel ADP-ribosylation factor-like proteins and characterization of their differ-ential expression in 3T3-L1 cells. J Biol Chem 269:15683–15688

Shyng SL, Huber MT, Harris DA (1993) A prion protein cycles betweenthe cell surface and an endocytic compartment in cultured neuro-blastoma cells. J Biol Chem 268:15922–15928

Shyng SL, Moulder KL, Lesko A, Harris DA (1995a) The N-terminaldomain of a glycolipid-anchored prion protein is essential for itsendocytosis via clathrin-coated pits. J Biol Chem 270:14793–14800

Shyng SL, Lehmann S,Moulder KL, Harris DA (1995b) Sulfated glycansstimulate endocytosis of the cellular isoform of the prion protein,PrPC, in cultured cells. J Biol Chem 270:30221–30229

Simon D, Herva ME, Benitez MJ, Garrido JJ, Rojo AI, Cuadrado A,Torres JM, Wandosell F (2014) Dysfunction of the PI3K-Akt-GSK-3 pathway is a common feature in cell culture and in vivomodels of prion disease. Neuropathol Appl Neurobiol 40:311–326.doi:10.1111/nan.12066

Totsukawa G, Yamakita Y, Yamashiro S, Hartshorne DJ, Sasaki Y,Matsumura F (2000) Distinct roles of ROCK (Rho-kinase) andMLCK in spatial regulation of MLC phosphorylation for assemblyof stress fibers and focal adhesions in 3T3 fibroblasts. J Cell Biol150:797–806

Tsai YR, Huang LJ, Lin HY, Hung YJ, Lee MR, Kuo SC, HsuMF,WangJP (2013) Inhibition of formyl peptide-stimulated phospholipase Dactivation by Fal-002-2 via blockade of the Arf6, RhoA and proteinkinase C signaling pathways in rat neutrophils. NaunynSchmiedebergs Arch Pharmacol 386:507–519. doi:10.1007/s00210-013-0851-6

Ulmer JB, Palade GE (1991) Effects of Brefeldin A on the Golgi com-plex, endoplasmic reticulum and viral envelope glycoproteins inmurine erythroleukemia cells. Eur J Cell Biol 54:38–54

VeyM, Pilkuhn S, Wille H, Nixon R, DeArmond SJ, Smart EJ, AndersonRG, Taraboulos A, Prusiner SB (1996) Subcellular colocalization ofthe cellular and scrapie prion proteins in caveolae-like membranousdomains. Proc Natl Acad Sci U S A 93:14945–14949

Volpicelli-Daley LA, Li Y, Zhang CJ, Kahn RA (2005) Isoform-selectiveeffects of the depletion of ADP-ribosylation factors 1-5 on mem-brane traffic. Mol Biol Cell 16:4495–4508. doi:10.1091/mbc.E04-12-1042

Wang GR, Shi S, Gao C, Zhang BY, Tian C, Dong CF, Zhou RM, Li XL,Chen C, Han J, Dong XP (2010) Changes of tau profiles in brains ofthe hamsters infected with scrapie strains 263 K or 139 A possiblyassociated with the alteration of phosphate kinases. BMC Infect Dis10:86. doi:10.1186/1471-2334-10-86

WeisWI, Scheller RH (1998)Membrane fusion. SNARE the rod, coil thecomplex. Nature 395:328–329. doi:10.1038/26354

Weise J, Doeppner TR, Muller T, Wrede A, Schulz-Schaeffer W, Zerr I,Witte OW, Bahr M (2008) Overexpression of cellular prion proteinalters postischemic Erk1/2 phosphorylation but not Akt phosphory-lation and protects against focal cerebral ischemia. Restor NeurolNeurosci 26:57–64

Welch HC, Coadwell WJ, Stephens LR, Hawkins PT (2003)Phosphoinositide 3-kinase-dependent activation of Rac. FEBS Lett546:93–97

Wheeler AP, Ridley AJ (2004) Why three Rho proteins? RhoA, RhoB,RhoC, and cell motility. Exp Cell Res 301:43–49. doi:10.1016/j.yexcr.2004.08.012

Wu G, Nakajima K, Takeyama N, Yukawa M, Taniuchi Y, Sakudo A,Onodera T (2008) Species-specific anti-apoptotic activity of cellularprion protein in a mouse PrP-deficient neuronal cell line transfectedwith mouse, hamster, and bovine Prnp. Neurosci Lett 446:11–15.doi:10.1016/j.neulet.2008.09.020

Zafar S, von Ahsen N, Oellerich M, Zerr I, Schulz-Schaeffer WJ,Armstrong VW, Asif AR (2011) Proteomics approach to identifythe interacting partners of cellular prion protein and characterizationof Rab7a interaction in neuronal cells. J Proteome Res. doi:10.1021/pr2001989

Zafar S, Asif AR, Ramljak S, Tahir W, Schmitz M, Zerr I (2014)Anchorless 23-230 PrP(C) interactomics for elucidation of PrP(C)protective role. Mol Neurobiol 49:1385–1399. doi:10.1007/s12035-013-8616-2

348 J Mol Neurosci (2015) 56:329–348