Pp

XD

h

���

a

ARR2A

KOPS

1

nmoarifwaisratO

Mf

0h

Neuroscience Letters 530 (2012) 127– 132

Contents lists available at SciVerse ScienceDirect

Neuroscience Letters

jou rn al h om epage: www.elsev ier .com/ locate /neule t

lexin A3 is involved in semaphorin 3F-mediated oligodendrocyterecursor cell migration

in Xiang, Xuan Zhang, Qi-Lin Huang ∗

epartment of Neurosurgery, Xinqiao Hospital Affiliated to the Third Military Medical University, , China

i g h l i g h t s

Plexin A3 is expressed in oligodendrocyte precursor cells.The expression of plexin A3 is downregulated in mature oligodendrocytes.Plexin A3 plays an important role in oligodendrocyte precursor cell migration.

r t i c l e i n f o

rticle history:eceived 15 June 2012eceived in revised form2 September 2012

a b s t r a c t

Class 3 semaphorins are expressed in the neurodevelopmental or damage repair phase of the centralnervous system (CNS). They play an important role in guiding axon growth and directing cell migra-tion, including the migration of oligodendrocyte precursor cells (OPCs). As co-receptors for semaphorin

ccepted 27 September 2012

eywords:ligodendrocyte precursor cellslexin A3

3F(sema3F), the expression and role of neuropilin-2 (NRP2) and plexin A3 in OPC migration are unclear.Using RT-PCR, Western blot analysis, and immunofluorescence, we demonstrated that primary OPCs andimmature oligodendrocytes from neonatal rats express NRP2 and plexin A3. After transfection with NRP2siRNA and plexin A3 siRNA, the number of migrating OPCs attracted to sema3F remarkably decreased.These results suggest that plexin A3 is expressed in OPCs and immature oligodendrocytes and is involvedin OPC migration.

emaphorin signal

. Introduction

Oligodendrocyte precursor cells (OPCs) are widespread myeli-ating cells of the adult central nervous system (CNS). Theigration and recruitment of OPCs are important for the devel-

pment of the CNS, as well as the repair of demyelinated areasfter spinal cord injury, and multiple sclerosis (MS) [9]. Althoughemyelination can be extensive, it is most often insufficient and lim-ted to the periphery of demyelinated lesions [7]. The semaphorinamily is key in providing directional guidance for migrating OPCs,hich are expressed during neurodevelopment or damage repair,

nd they play an important role in guiding axon growth and direct-ng cell migration. In the semaphorin family, overexpression ofemaphorin 3F (sema3F) accelerates both OPC recruitment and theemyelination rate [7,13]. Since plexin A3 and neuropilin-2 (NRP2)

re essential receptors for the transmission of sema3F signals [15],he roles of the cooperative receptor, which may be important forPC migration induced by sema3F, need to be defined.

∗ Corresponding author at: Department of Neurosurgery, Xinqiao Hospital, Thirdilitary Medical University, Chongqing 400037, China. Tel.: +86 23 6877 4610;

ax: +86 23 6875 5602.E-mail address: hqlmdxq@yahoo.com (Q.-L. Huang).

304-3940/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.neulet.2012.09.058

© 2012 Elsevier Ireland Ltd. All rights reserved.

In the present study, the expression of NRP2 and plexin A3was detected in OPC primary cultures. In addition, OPC migrationinduced by sema3F was observed by transwell migration assaysafter siRNA-mediated inhibition of plexin A3 and NRP2 mRNAexpression.

2. Materials and methods

2.1. Reagents

Before the research was performed, the experiment plan wasreviewed and approved by the Ethics Committee of the ThirdMilitary Medical University. Sprague-Dawley (SD) rats were fromthe Laboratory Animal Center, Daping Hospital, Third MilitaryMedical University. Dulbecco’s modified Eagle’s media (DMEM;11960), B-27 supplement (12587), neurobasal medium (NBM),penicillin/streptomycin (15140), Hanks balanced salt solution(HBSS; 14025), and Dulbecco’s phosphate-buffered saline (DPBS)without Mg2+ and Ca2+ (14190-144) were all obtained from

Invitrogen/Gibco (Carlsbad, CA, USA). Fetal bovine serum with-out l-glutamine or sodium pyruvate (FBS; SH300700) was fromHyclone (Shanghai, China). Sodium pyruvate (P2256), bovineserum albumin (BSA; A9647), poly-l-lysine (P1399), DNase I

1 nce Le

(kwpfiPp

FaFp

28 X. Xiang et al. / Neuroscie

D5025), Trypsin: l-1-tosylamido-2-phenylethylchloromethyletone (TPCK)-treated trypsin (T1426), and Trypan blue (T8154)ere purchased from Sigma (St. Louis, MO, USA). B27/NBM, human

latelet-derived growth factor-AA (PDGF-AA) (100-13A), and basicbroblast growth factor (bFGF) (100-18B) were purchased fromeprotech (Rocky Hill, NJ, USA). The following reagents were alsourchased from the designated companies: Trizol reagent (Tiangen

ig. 1. Immunofluorescence analysis of NRP2 and plexin A3 expression in primary oligodentibodies, and then incubated with antibodies to NRP2 and plexin A3. The sample was

ITC (for NRP2 and plexin A3). DAPI was used to counterstain the nuclei. The pictures arelexin A3 were detected in the immature oligodendrocytes; (Q–X) both NRP2 and plexin

tters 530 (2012) 127– 132

Biotech, China), First Strand cDNA Synthesis kit (Toyobo Co. Ltd.,China), PVDF membranes (Millipore, Billeria, MA, USA), anti-NRP2(Abcam, Cambridge, UK), anti-plexin A3 (Santa Cruz Biotechnology,

Santa Cruz, CA, USA), �-actin primary antibody (Santa Cruz), NRP2siRNA, plexin A3 siRNA, and control siRNA (Santa Cruz), ECL kit(Amersham, UK), and recombinant sema3F (R&D, Minneapolis, MN,USA).

ndrocyte cultures. Purified OPCs were stained with anti-NG2, anti-O4, and anti-MBPvisualized with IgG-specific antisera labeled with TRITC (for NG2, O4, or MBP) and: (A–H) both NRP2 and plexin A3 were detected in the OPCs; (I–P) both NRP2 and

A3 were barely detected in the mature oligodendrocytes.

ce Let

2

codcahswcdtsftaoiwp5iTncbemd

Fpda

X. Xiang et al. / Neuroscien

.2. Cell culture and purification

OPC cultures were prepared as previously described [14]. Briefly,erebral cortices from postnatal day 1–2 SD rats were dissectedut after the meninges were removed. The cortical tissues wereiced into ∼1 mm3 chunks, digested for 15 min at 37 ◦C in HBSSontaining 0.25% trypsin and 60 �g/ml dNase I, and centrifugedt 100 × g for 5 min. The pellet was then dissociated until nearlyomogenous. The suspension was passed through a 70 �m celltrainer, and was then plated in a 75 cm2 tissue culture flask coatedith 100 �g/ml poly-l-lysine. The cells were maintained in DMEM

ontaining 10% FBS at 37 ◦C in humidified air with 5% CO2 for 10ays. The medium was completely changed every 3 days. Afterhe medium was changed on the tenth day, flasks were sealed andhaken at 200 rpm at 37 ◦C for 1 h. Then the medium was discardedrom the flasks to eliminate microglia cells. After 2 h of equilibra-ion in the incubator, the flasks were shaken at 200 rpm overnightt 37 ◦C. Medium with detached cells was collected and platedn untreated glass tissue culture Petri dishes for 30 min to elim-nate contaminated microglias and astrocytes. The cell suspension

as collected and plated onto coverslips coated with 100 �g/mloly-l-lysine in DMEM containing 10% FBS at a cell density of0,000 cells/cm2 in tissue culture dishes (for immunocytochem-

stry) or flasks (for RT-PCR, Western blotting, and transwell assays).he cells were incubated in a tissue culture incubator at 37 ◦C. Theext morning, the culture medium was changed to defined mediaonsisting of B27/NBM containing 10 ng/ml PDGF-AA and 10 ng/ml

FGF, and the cells were fed every other day. For differentiationxperiments, PDGF-AA and bFGF were withdrawn from the cultureedium after 3 days, and the cells were cultured for a total of 6–7

ays.

ig. 2. Expression of NRP2 and plexin A3 mRNA and protein in the oligodendrocyte linerformed as described in Section 2. Samples of mRNA and protein were obtained from procytes (MBP). No significant difference in NRP2 and plexin A3 expression was observend plexin A3 were highly expressed in NG2 and O4 cells, as determined by both RT-PCR

ters 530 (2012) 127– 132 129

2.3. Immunocytochemistry

Cells were fixed in 4% paraformaldehyde (PFA), and incubatedwith the following primary antibodies: mouse monoclonal anti-NG2 (1:200) to detect OPCs, mouse monoclonal anti-O4 (1:200)to detect immature oligodendrocytes, goat polyclonal anti-MBP(myelin basic protein) (1:200) to detect mature oligodendrocytes,rabbit polyclonal anti-NRP2 (1:200), and rabbit polyclonal anti-plexin A3 (1:200). Cells were then incubated with the followingsecondary antibodies: FITC-goat anti-rabbit IgG (1:500), TRITC-goatanti-mouse IgG (1:500), FITC-mouse anti-rabbit IgG (1:500), andTRITC-mouse anti-goat IgG (1:500). The percentage of each cell typewas obtained by counting the number of DAPI-positive cells. Thenumber of cells with positive staining was counted in 15 randomlyselected fields using a 40× objective lens.

2.4. Reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was directly extracted from primary OPCs using theTrizol reagent. According to the manufacturer’s instructions, theRNA was converted to cDNA, and then 2 �g of the cDNA was used forthe PCR reaction (20 �l). Primers and conditions of the PCR reactionare listed in Table 1. PCR primers were designed based on publishedrat sequences. As a housekeeping gene internal control, �-actin wasused. PCR reaction products were visualized on ethidium-bromide-stained 1% agarose gels.

2.5. Protein preparation and Western blot analysis

Total protein from cells in different phases of development wasextracted as previously described and quantified using a DU 640

eage. RT-PCR (A and B) and Western blotting (C and D) in primary rat OPCs wererimary cells of OPCs (NG2), immature oligodendrocytes (O4), or mature oligoden-d between NG2-positive cells and O4-positive cells. Compared to MBP cells, NRP2

and Western blot analyses (#p < 0.01) (n = 4).

130 X. Xiang et al. / Neuroscience Le

Table 1Primer sequences used for PCR.

Protein Sequence (5′–3′) Product size

NRP2 F GGGGAGAGGGCTATGAAGATGAG 281 bpR GCCGTCGTAGAGCTCAAAGTTGT

Plexin A3 F CGCTCCCTCAACCGTTCTTC 267 bp

pSPistdwwrm

2

Nmwct(biPM

2

w

Fcb

R GCCACCCACCAGCACCATAACAˇ-actin F ACCCCGTGCTGCTGACCGAG 245 bp

R TCCCGGCCAGCCAGGTCCA

rotein analyzer [1]. Proteins (30 �g) were separated on 6% or 8%DS-PAGE gels. Separated proteins were transferred to 0.45 �m/lVDF membranes that were blocked at room temperature for 2 hn Tris-buffered saline with 0.1% Tween-20 (TBS-T) containing 5%kim milk. The membranes were then probed with a 1:1000 dilu-ion (diluted in TBS-T) of rabbit polyclonal anti-NRP2, a 1:700ilution of anti-plexin A3, and a 1:500 dilution of anti-�-actin,hich was used as a loading control. The secondary antibody usedas a 1:700 dilution of horseradish peroxidase (HRP)-labeled anti-

abbit IgG. Protein bands were visualized by ECL according to theanufacturer’s instructions.

.6. siRNA preparation and transfection

siRNA oligonucleotide duplexes specific to rat plexin A3 andRP2 were prepared, and transfected into OPCs according to theanufacturer’s instructions. Briefly, before transfection the OPCsere purified for 3 days and plated at a density of about 1 × 106

ells per well. On day 2.5, according to the manufacturer’s instruc-ions, plexin A3 siRNA, NRP2 siRNA, and control siRNA solutions1:100) were prepared, separately transfected into cells, and incu-ated for 6 h. The transfected cells were then washed and cultured

n B27/NBM for 48 h. Transfection efficiency was measured by RT-CR. Cell viability after each siRNA treatment was assessed by theTT assay.

.7. Analysis of the transwell assay

After siRNA transfection and trypsinization, transwell assaysere conducted in OPCs using 24-well transwell chambers. Cells

ig. 3. PCR assays after siRNA transfection. (A–C) The expression of NRP2 and plexin A3ontrol group (#p < 0.01), while no obvious differences were observed between both theoth NRP2 siRNA and plexin A3 siRNA treatment was assessed by the MTT assay. No obvi

tters 530 (2012) 127– 132

(1 × 103) suspended in B27/NBM were seeded onto the upperchamber, whereas B27/NBM containing 10% recombinant sema3Fwere added to the lower wells. After 24 h of culture at 37 ◦Cunder 5% CO2, cells on the upper surface of the membrane wereremoved, whereas migratory cells on the lower membrane sur-face were fixed in paraformaldehyde and stained with crystalviolet. Migration was assessed by counting the number of cellsthat migrated on three independent membranes using a phasecontrast microscope (200×). Three visual fields were chosen atrandom on the central and peripheral sections of the mem-branes, and the cells that migrated through the 8 �m pores werecounted 5 separate times. The obtained value was then nor-malized to that of the control cultures to produce the relativeratio.

2.8. Statistical analysis

All data are presented as mean ± SD. Differences between themeans of separate groups were assessed by two independent sam-ple t-tests for single comparisons using SPSS software, version13.0. P values less than 0.05 were considered statistically signifi-cant.

3. Results

3.1. Cell culture and immunofluorescence

Similar to previous reports, OPCs, oligodendrocytes, astrocytes,and microglia were present in the OPCs culture flasks [14,1]. OPCs(92.12% ± 2.23%) were obtained after 3 days of incubation in thepresence of PDGF-AA and bFGF. Immature (68.56 ± 27.33) andmature oligodendrocytes (75.42 ± 14.75) were separately purifiedon days 3 and 6 after which PDGF-AA and bFGF were sepa-rately withdrawn. Double-labeled fluorescent staining showedNRP2 and plexin A3 staining on the surface of the cell bodies

and in the processes of OPCs and immature oligodendrocytes(Fig. 1). However, when OPCs differentiated into mature oligo-dendrecytes, little staining of NRP2 and plexin A3 was seen(Fig. 1).

decreased after treatment with NRP2 siRNA or plexin A3 siRNA compared to the control and normal groups. (D) The cell viability after different concentrations ofous cytotoxicity was found after treatment with either siRNA.

X. Xiang et al. / Neuroscience Letters 530 (2012) 127– 132 131

Fig. 4. Effect of NRP2 and plexin A3 siRNA on OPC migration induced by sema3F. (A) OPCs that migrated through the pore of membranes (the clear circles in the images)were stained with crystal violet. 1, normal OPCs without stimulation; 2, normal OPCs stimulated with sema3F; 3, control siRNA treatment without sema3F stimulation; 4,stimulation with sema3F after control siRNA treatment; 5, stimulation with sema3F after NRP2 siRNA treatment; 6, stimulation with sema3F after plexin A3 siRNA treatment;7, OPCs after NRP2 siRNA treatment without stimulation; 8, OPCs after plexin A3 siRNA treatment without stimulation. (B) Quantitative analysis of sema3F-induced OPCmigration and the effect of NRP2 and plexin A3 siRNA compared to the normal and control groups. While control siRNA had no effect on OPC migration, sema3F stimulationclearly promoted OPC migration (*p < 0.01); stimulation with sema3F after NRP2 siRNA or plexin A3 siRNA treatment slightly induced OPC migration (#p < 0.01); OPC migrationw mulat

3

Nbadipod

3e

dNdAg(dts

as minimized after NRP2 siRNA or plexin A3 siRNA treatment without sema3F sti

.2. RT-PCR and Western blot analyses

RT-PCR and Western blot analyses were performed to detectRP2 and plexin A3 expression. Consistent with a previous reporty Cohen et al. in which NRP2 expression was detected in A2B5nd O4-positive cells, but downregulated in O1-positive oligoden-rocytes [1], we not only confirmed NRP2 expression in OPCs and

mmature oligodendrocytes, but we also detected both NRP2 andlexin A3 mRNA and protein expression in OPCs and immatureligodendrocytes. This expression decreased in mature oligoden-rocytes (#p < 0.01) (Fig. 2B and D).

.3. siRNA transfection inhibited plexin A3 and NRP2 mRNAxpression

RT-PCR was used to examine the efficiency of siRNA knock-own of plexin A3 and NRP2 in OPCs. Compared to normal cells,RP2 or plexin A3 expression in control cells did not significantlyecrease (Fig. 3, Panels A–C). The mRNA expression of both plexin3 and NRP2 considerably decreased compared to control siRNAroups after the separate transfection of plexin A3 and NRP2 siRNA

#p < 0.05) (Fig. 3, Panels A–C). The efficiency of NRP2 siRNA knock-own was 64–79% and of plexin A3 was 64–73% (Fig. 3C), indicatinghat siRNA is an effective way to downregulate the mRNA expres-ion of NRP2 and plexin A3 in OPCs. The MTT assay indicated

ion (�p < 0.01) (n = 5).

no obvious cytotoxicity with either siRNA at a concentration of0.125–2 �g/100 �l (Fig. 3D).

3.4. Effect of sema3F on OPC migration

Migration assays were performed in transwells to assess theeffect of NRP2 and plexin A3 siRNA on OPC migration. OPCs weretreated with NRP2 and plexin A3 siRNA or control siRNA. Thenthe cells were stimulated with sema3F and compared to untreatedOPCs. We confirmed that sema3F could attract normal or controlsiRNA-treated OPC migration much more than unstimulated OPCs(*p < 0.01), whereas control siRNA did not affect OPC migration(1–4 of Fig. 4A; Fig. 4B). We found that sema3F-induced migrationwas suppressed by both NRP2 and plexin A3 siRNA respectively(#p < 0.01) (4–6 of Fig. 4A; Fig. 4B). After withdrawal of sema3F,siRNA-mediated OPC migration decreased even further (�p < 0.01)(5–8 of Fig. 4A; Fig. 4B). These findings suggest that NRP2 and plexinA3 may play key roles in the migration of OPCs induced by sema3F.

4. Discussion

Although OPCs appear during the embryonic phase of brain andspinal cord development, it has been confirmed that they are themain type of proliferating cells in the postnatal and adult brain, anddistribute almost uniformly in both gray and white matter areas

1 nce Le

[m

vs[gNcdTho

ripsdsts

wtdANstApwpN

nfaapnadARfFbtdifin

fdeesaaip

[

[

[

[

[

[

[

32 X. Xiang et al. / Neuroscie

17]. The recruitment of OPCs to the lesion is considered one of theost important steps of remyelination in the CNS [13,6].Our study confirmed that sema3F can attract OPC migration in

itro. During many developmental and repair processes in the CNS,ema3F is secreted by astrocytes, microglia, and oligodendrocytes13], and plays an important role in cell migration, growth coneuidance, and synapse formation [2,4,11]. As a receptor of sema3F,RP2 takes part in many kinds of cell migration [2,16,3]. Our studyonfirmed that NRP2 is only expressed in OPCs and immature oligo-endrocytes, and is downregulated in mature oligodendrocytes.his indicates that NRP2 is also involved in OPC migration, but mayave little effect on dendrite guidance and stretching of matureligodendrocytes.

It is well known that plexin A3 is a high-affinity cooperativeeceptor of NRP2 [7,15]. Although several reports have documentedts involvement in cell migration and axon guidance, the role oflexin A3 in OPC migration is still unclear [11,7]. In the presenttudy, we showed that the expression of plexin A3 in oligoden-rocytes is similar to that of NRP2, indicating that they may haveimilar effects during OPC migration and differentiation. Therefore,his study demonstrates the role of the sema3F–NRP2–plexin A3ignaling during OPC migration.

In the present study, plexin A3 and NRP2 expressions in OPCsere inhibited by siRNA transfection. Transwell analysis was used

o examine the effect of sema3F on OPC migration. A noticeableecrease in migration was observed after both NRP2 and plexin3 siRNA were respectively transfected into OPCs. Furthermore,RP2 siRNA interference was more effective than that by plexin A3

iRNA, indicating that NRP2 and plexin A3 both are essential forransmitting sema3F signals to OPCs and inducing them to travel.s the direct binding partner of sema3F, NRP2 binds not only tolexin A3, but also to other receptors such as VEGFR2 and VEGFR3,hich are also involved in OPC migration [5]. This array of bindingartners may lead to the difference in migratory effects betweenRP2 and plexin A3.

Type A plexins transduce class 3 semaphorin signaling via theeuropilin co-receptor, most of which transmit repellent signals

rom various migrating cells [11,8]. It is interesting that sema3Flso acts as an attractive signal for migrating cells such as OPCsnd cerebellar granule cells [2,10]. It is well known that thelexin family plays a key role in transmitting extracellular sig-als into endochylema. Upon semaphorin engagement, the GTPasectivating protein (GAP) domain of plexin is activated, which isocumented as the Rho GTPase binding domain (RBD) in plexin. The RBD can bind with GTPases including Rac, RhoD, RND1, andND2, but due to their different structures and sequences, the sub-

amilies of plexin A have diverse affinities for those GTPases [12].or instance, plexin A3 tends to bind with Rnd1, whereas stronginding with Rho and weak binding with Rac1 have been seeno plexin A1 [12]. These different affinities may activate variousownstream signals and lead to diverse biological effects, includ-

ng opposing effects on cell migration induced by the semaphorinamily. Until now, the precise mechanism of how those GTPasesnteract with plexin A3 remains unclear; thus, further studies areeeded.

Recently, the expressions of sema3F, NRP2, and plexin A3 wereound to be upregulated in adult OPCs after CNS demyelinationamage [7]. After demyelination, sema3F in local lesions hadarly and strong upregulation, which coincided with the increasedxpressions of NRP2 and plexin A3 in OPCs. In addition, the expres-ions of sema3F, NRP2, and plexin A3 consistently decreased when

n injury occurred [7]. These phenomena matched with our studynd we would presume that following acute demyelination, earlyncreases in sema3F–NRP2–plexin A3 signaling maybe attract aool of future remyelinating OPCs. After arriving at the damaged

[

tters 530 (2012) 127– 132

area, OPCs lose their dynamic ability to differentiate into matureoligodendrocytes and take part in the role of reforming neuralnetworks.

Since OPCs play an important role in CNS development, demyeli-nating diseases, and injury repair, understanding the mechanism ofmigration may be helpful for the exploration of new treatments forthese diseases.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgments

We thank Weihua Zhang for polishing the grammar. We alsothank Medjaden Bioscience Limited for assisting in the preparationof this manuscript.

References

[1] R.I. Cohen, D.M. Rottkamp, D. Maric, J.L. Barker, L.D. Hudson, A role forsemaphorins and neuropilins in oligodendrocyte guidance, Journal of Neuro-chemistry 85 (2003) 1262–1278.

[2] S. Ding, J.H. Luo, X.B. Yuan, Semaphorin-3F attracts the growth cone of cerebel-lar granule cells through cGMP signaling pathway, Biochemical and BiophysicalResearch Communications 356 (2007) 857–863.

[3] B. Favier, A. Alam, P. Barron, J. Bonnin, P. Laboudie, P. Fons, M. Mandron,J.P. Herault, G. Neufeld, P. Savi, J.M. Herbert, F. Bono, Neuropilin-2 interactswith VEGFR-2 and VEGFR-3 and promotes human endothelial cell survival andmigration, Blood 108 (2006) 1243–1250.

[4] J.C. Gant, O. Thibault, E.M. Blalock, J. Yang, A. Bachstetter, J. Kotick, P.E.Schauwecker, K.F. Hauser, G.M. Smith, R. Mervis, Y. Li, G.N. Barnes, Decreasednumber of interneurons and increased seizures in neuropilin 2 deficient mice:implications for autism and epilepsy, Epilepsia 50 (2009) 629–645.

[5] B. Le Bras, M.J. Barallobre, J. Homman-Ludiye, A. Ny, S. Wyns, T. Tammela, P.Haiko, M.J. Karkkainen, L. Yuan, M.P. Muriel, E. Chatzopoulou, C. Bréant, B. Zalc,P. Carmeliet, K. Alitalo, A. Eichmann, J.L. Thomas, VEGF-C is a trophic factor forneural progenitors in the vertebrate embryonic brain, Nature Neuroscience 9(2006) 340–348.

[6] J.M. Mangin, V. Gallo, The curious case of NG2 cells: transient trend or gamechanger? ASN Neuro 3 (2011) e00052.

[7] G. Piaton, M.S. Aigrot, A. Williams, S. Moyon, V. Tepavcevic, I. Moutkine,J. Gras, K.S. Matho, A. Schmitt, H. Soellner, A.B. Huber, P. Ravassard, C.Lubetzki, Class 3 semaphorins influence oligodendrocyte precursor recruit-ment and remyelination in adult central nervous system, Brain 134 (2011)1156–1167.

[8] A. Shimizu, A. Mammoto, J.E. Italiano Jr., E. Pravda, A.C. Dudley, D.E. Ingber, M.Klagsbrun, ABL2/ARG tyrosine kinase mediates SEMA3F-induced RhoA inacti-vation and cytoskeleton collapse in human glioma cells, Journal of BiologicalChemistry 283 (2008) 27230–27238.

[9] J.R. Siebert, D.J. Stelzner, D.J. Osterhout, Chondroitinase treatment followingspinal contusion injury increases migration of oligodendrocyte progenitor cells,Experimental Neurology 231 (2011) 19–29.

10] N. Spassky, F. de Castro, B. Le Bras, K. Heydon, F. Quéraud-LeSaux, E.Bloch-Gallego, A. Chédotal, B. Zalc, J.L. Thomas, Directional guidance ofoligodendroglial migration by class 3 semaphorins and netrin-1, Journal ofNeuroscience 22 (2002) 5992–6004.

11] K.E. Waimey, P.H. Huang, M. Chen, H.J. Cheng, Plexin-A3 and plexin-A4 restrictthe migration of sympathetic neurons but not their neural crest precursors,Developmental Biology 315 (2008) 448–458.

12] H. Wang, P.K. Hota, Y. Tong, B. Li, L. Shen, L. Nedyalkova, S. Borthakur, S. Kim,W. Tempel, M. Buck, H.W. Park, Structural basis of Rnd1 binding to plexin RhoGTPase binding domains (RBDs), Journal of Biological Chemistry 286 (2011)26093–26106.

13] A. Williams, G. Piaton, M.S. Aigrot, A. Belhadi, M. Théaudin, F. Petermann, J.L.Thomas, B. Zalc, C. Lubetzki, Semaphorin 3A and 3F: key players in myelin repairin multiple sclerosis, Brain 130 (2007) 2554–2565.

14] Z. Yang, M. Watanabe, A. Nishiyama, Optimization of oligodendrocyte progeni-tor cell culture method for enhanced survival, Journal of Neuroscience Methods149 (2005) 50–56.

15] A. Yaron, P.H. Huang, H.J. Cheng, M. Tessier-Lavigne, Differential requirementfor plexin-A3 and -A4 in mediating responses of sensory and sympathetic neu-rons to distinct class 3 semaphorins, Neuron 45 (2005) 513–523.

16] H. Yasuoka, R. Kodama, M. Tsujimoto, K. Yoshidome, H. Akamatsu, M. Naka-

hara, M. Inagaki, T. Sanke, Y. Nakamura, Neuropilin-2 expression in breastcancer: correlation with lymph node metastasis, poor prognosis, and regulationof CXCR4 expression, BMC Cancer 9 (2009) 220.

17] X. Zhu, D.E. Bergles, A. Nishiyama, NG2 cells generate both oligodendrocytesand gray matter astrocytes, Development 135 (2008) 145–157.