Neural progenitor cell implants modulate vascular endothelial growth factor and brain-derived neurotrophic factor expression in rat axotomized neurons

Neu al P ogeni o Cell Implan s Modula e Vascula Endo helial G ow h Fac o and B ain-De i ed Neu o ophic Fac o Exp ession in Ra Axo omized Neu ons Rocı ´o Tala e o ´n, Espe anza R. Ma a edona, Rosa R. de la C uz, Angel M. Pas o * Labo a o io de Fisiologı ´a y Plas icidad Neu onal, Depa amen o de Fisiologı ´a, Facul ad de Biologı ´a, Uni e sidad de Se illa, Se illa, Spain Abs ac Axo omy o cen al neu ons leads o unc ional and s uc u al al e a ions which la gely e e when neu al p ogeni o cells (NPCs) a e implan ed in he lesion si e. The new mic oen i onmen c ea ed by NPCs in he hos issue migh modula e in he damaged neu ons he exp ession o a high a ie y o molecules wi h ele an oles in he epai mechanisms, including neu o ophic ac o s. In he p esen wo k, we aimed o analyze changes in neu o ophic ac o exp ession in axo omized neu ons induced by NPC implan s. Fo his pu pose, we pe o med immuno luo escence ollowed by con ocal mic oscopy analysis o he de ec ion o ascula endo helial g ow h ac o (VEGF), b ain-de i ed neu o ophic ac o (BDNF), neu o ophin-3 (NT-3) and ne e g ow h ac o (NGF) on b ains em sec ions om a s wi h axo omy o abducens in e nuclea neu ons ha ecei ed NPC implan s (implan ed g oup) o ehicle injec ions (axo omized g oup) in he lesion si e. Con ol abducens in e nuclea neu ons we e s ongly immuno eac i e o VEGF and BDNF bu showed a weak s aining o NT-3 and NGF. Compa isons be ween g oups e ealed ha lesioned neu ons om animals ha ecei ed NPC implan s showed a signi ican inc ease in VEGF con en wi h espec o animals ecei ing ehicle injec ions. Howe e , he immuno eac i i y o BDNF, which was inc eased in he axo omized g oup as compa ed o con ol, was no modi ied in he implan ed g oup. The modi ica ions induced by NPC implan s on VEGF and BDNF con en we e speci ic o he popula ion o axo omized abducens in e nuclea neu ons since he neighbo ing abducens mo oneu ons we e no a ec ed. Simila le els o NT-3 and NGF immunolabeling we e ob ained in inju ed neu ons om axo omized and implan ed animals. Among all he analyzed neu o ophic ac o s, only VEGF was exp essed by he implan ed cells in he lesion si e. Ou esul s poin o a ole o NPC implan s in he modula ion o neu o ophic ac o exp ession by lesioned cen al neu ons, which migh con ibu e o he es o a i e e ec s o hese implan s. Ci a ion: Tala e o ´n R, Ma a edona ER, de la C uz RR, Pas o AM (2013) Neu al P ogeni o Cell Implan s Modula e Vascula Endo helial G ow h Fac o and B ain- De i ed Neu o ophic Fac o Exp ession in Ra Axo omized Neu ons. PLoS ONE 8(1): e54519. doi:10.1371/jou nal.pone.0054519 Edi o : Tim Douglas Aumann, Flo ey Ins i u e o Neu oscience & Men al Heal h, Aus alia Recei ed Oc obe 2, 2012; Accep ed Decembe 12, 2012; Published Janua y 18, 2013 Copy igh : ß2013 Tala e o ´n e al. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Funding: The s udy was unded by g an s in Spain BFU2009-07121 and BFU2012-33975 om MEC-FEDER and CVI-6053 om Jun a de Andalucı ´a-FEDER. The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing In e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . * E-mail: [email p o ec ed]s In oduc ion Axo omy o adul CNS neu ons leads o se e e mo pho unc- ional al e a ions which include synap ic s ipping and educed i ing a e [1,2]. When neu o ophic ac o s a e deli e ed o he dis al s ump o lesioned ne es, hese al e a ions a e la gely p e en ed and neu onal unc ion is es o ed [3,4,5,6], which sugges s ha physiological changes induced by axo omy a e, a leas in pa , due o he lack o a ge -de i ed ophic suppo . Neu o ophic dependence o adul neu ons is also e idenced by he axo omy-induced egula ion in he exp ession o g ow h ac o s and/o hei ecep o s [7,8,9], which migh compensa e o he loss o e og ade ophic suppo h ough al e na i e pa h- ways, such as an e og ade, pa ac ine o au oc ine [10]. By using he oculomo o sys em as he expe imen al model, we ha e p e iously desc ibed ha he ansec ion o he medial longi udinal ascicle (MLF) induces p o ound al e a ions in he axo omized in e nuclea neu ons o he abducens nucleus, which a e le disconnec ed om hei a ge [2,11]. The abducens nucleus is composed o mo oneu ons ha inne a e he ipsila e al la e al ec us muscle, and a g oup o p emo o cells, he abducens in e nuclea neu ons, whose axons cou se h ough he MLF o es ablish synap ic con ac s wi h medial ec us mo oneu ons in he con ala e al oculomo o nucleus [12]. A e hei axo omy, abducens in e nuclea neu ons show a ma ked educ ion in i ing a e, a signi ican dec ease in eye posi ion and eloci y sensi i i ies and a no iceable s ipping o synap ic a e ences [2,11]. Mo eo e , when a new a ge (emb yonic issue) is p o ided, lesioned abducens in e nuclea neu ons a e able o es o e hei i ing p ope ies and synap ic a e ences [13,14]. I was sugges ed ha neu o ophic ac o s eleased by he implan ed issue p obably media e he e-es ablishmen o hese axo omy-induced al e a- ions, a hypo hesis suppo ed by he ac ha abducens in- e nuclea neu ons exp ess he di e en ypes o k ecep o s o neu o ophins [15]. As a u he s ep, we ha e explo ed in ou cen al axo omy model he e ec s o neu al p ogeni o cell (NPC) implan s based PLOS ONE | www.plosone.o g 1 Janua y 2013 | Volume 8 | Issue 1 | e54519 on he epo ed e icacy o his cell popula ion in p omo ing unc ional imp o emen o damaged CNS neu ons [16,17,18]. Thus, axo omized abducens in e nuclea neu ons la gely eco e hei mo phophysiological p ope ies ollowing he implan o NPCs a he lesion si e in he MLF, as we ha e shown ecen ly in he ca [19]. Se e al mechanisms ha e been p oposed o be likely in ol ed in he bene i ial e ec s o he NPC implan s a e CNS inju y: sec e ion o neu o ophic and/o angiogenic ac o s, es o a ion o synap ic ansmi e elease by p o iding local einne a ion, e-es ablishmen o unc ional a e en and e e en connec ions, o change in he b ain mic oen i onmen owa ds mo e a ou able condi ions o bo h he lesioned neu ons o su i e and he implan ed p ogeni o s o in eg a e [20,21]. In line wi h his, ecen e idence sugges s ha implan ed NPCs es ablish a ‘‘c oss- alk’’ o bila e al communica ion wi h hos issue cells o ele an impo ance o he neu op o ec i e ac ion o he implan s [22,23]. NPCs used in ou s udy we e ob ained om he sub en icula zone (SVZ) o pos na al animals. In hei neu ogenic niche, SVZ cells ha e been desc ibed o exp ess he angiogenic ac o ascula endo helial g ow h ac o (VEGF) and he neu o ophic ac o s b ain-de i ed neu o ophic ac o (BDNF), ne e g ow h ac o (NGF) and neu o ophin-3 (NT-3) [24]. Mo e p ecisely, SVZ as ocy es exp ess all he men ioned angionenic and neu o ophic ac o s whe eas SVZ neu al p ogeni o s selec i ely exp ess VEGF [24]. The neu obiological in e es o VEGF is sha ply inc easing nowadays since in addi ion o i s well known angiogenic ac i i y, VEGF also exhibi s neu o ophic and neu op o ec i e p ope ies [25,26,27,28]. Mo eo e , a ole in synap ic ansmission has also been a ibu ed o his ac o . In pa icula , VEGF is able o induce a dep ession in exci a o y synap ic ansmission on o py amidal and g anule neu ons o he hippocampus and also on hypoglossal mo oneu ons [29,30]. On he o he hand, he neu o ophins BDNF, NT-3 and NGF, also exp essed in he SVZ niche, ha e been demons a ed o induce bene i ial e ec s in lesioned neu ons a e di e en ypes o b ain damage [3,5,6]. In he p esen wo k, we aimed o in es iga e possible modi ica ions o neu o ophic ac o exp ession in cen al inju ed neu ons ollowing NPC implan s which could be in ol ed in he physiological eco e y induced by hese implan s. Fo his pu pose, we ansec ed he MLF in adul a s and implan ed NPCs a he lesion si e. Changes induced by he lesion and/o he implan in he exp ession o VEGF, BDNF, NT-3 and NGF we e analyzed in he axo omized popula ion o abducens in e nuclea neu ons. In addi ion, we also e alua ed he exp ession o hese neu o ophic molecules by he implan ed NPCs. The indings o his s udy indica e ha axo omy induces changes in he exp ession o neu o ophic ac o s by inju ed cen al neu ons ha can be di e en ially modula ed by NPC implan s. We p opose ha he neu o ophic egula ion p omo ed by NPCs migh con ibu e o he es o a i e e ec s o NPC implan s. Ma e ials and Me hods Expe imen s we e pe o med in adul and 7 day pos na al (P7) Wis a a s in acco dance wi h he guidelines o he Eu opean Union (86/609/EU) and Spanish law (R.D. 120/2005 BOE 252/ 34367-91, 2005) o he use and ca e o labo a o y animals. Su gical p ocedu es used in his s udy we e app o ed by he e hics commi ee o Uni e sidad de Se illa. Neu al P ogeni o Cell Cul u e and In ec ion Neu al p ogeni o s we e isola ed om he SVZ o pos na al a s (P7) and we e expanded in he o m o neu osphe es essen ially as epo ed [31]. B ie ly, he la e al walls o he la e al en icles we e emo ed and enzyma ically dissocia ed wi h 1 mg/ml ypsin a 37uC o 15 min. The issue was hen cen i uged a 150 g o 5 min, insed in Dulbecco’s modi ied Eagle’s medium/F12 medium 1:1 (DF-12) and cen i uged again in he same condi ions. Then, he cells we e esuspended in DF-12 con aining 0.7 mg/ml o omucoid, and mechanically disagg ega ed wi h a i e polished Pas eu pipe e. The dissocia ed cells we e cen i uged, esus- pended in de ined medium (DM: DF-12 con aining B-27 supplemen , 2 mM glu amine, 100 uni s/ml penicillin, 100 mg/ ml s ep omycin and 0.25 mg/ml ampho e icin B) supplemen ed wi h 20 ng/ml EGF and 10 ng/ml FGF-2, and main ained in an a mosphe e o 5% CO 2 ,a 37uC. A e 1–2 days, cell agg ega es known as neu osphe es we e o med (Figu e 1A). Cells we e subcul u ed 48 hou s a e isola ion and hen e e y 3–4 days. Neu osphe e-de i ed cells we e immunoposi i e o he neu al p ecu so ma ke nes in and we e able o gene a e glial and neu onal cells when hey we e induced o di e en ia e a e pla ing on adhesi e subs a e (suppo ing in o ma ion S1 and Figu e S1). Neu osphe es we e in ec ed wi h a len i i al ec o con aining a epo e gene encoding o he g een luo escen p o ein (GFP) unde he con ol o he cy omegalo i us (CMV) p omo e (gene ously p o ided by D . Da id Macı ´as, Hospi al Uni e si a io Vi gen del Rocı ´o, Uni e si y o Se ille; len i i us i e : 3610 6 IFU/ml) wi h a mul iplici y o in ec ion o 3. Neu osphe e-de i ed cells we e incuba ed o 8 hou s wi h i us-added DM in T25 lasks con aining 250,000 cells. Vi us-con aining medium was hen emo ed and eplaced wi h esh DM. A e 48–72 hou s o i us incuba ion, cells om neu osphe es exp essed GFP as con i med by di ec isualiza ion in a Zeiss LSM 7 DUO con ocal mic oscope (Figu e 1B). The pheno ype o he neu osphe e- de i ed cells was no a ec ed by he in ec ion since hey main ained he same abili y o gene a e neu ons and glial cells in di e en ia ing condi ions (da a no shown). T ansec ion o he Medial Longi udinal Fascicle and Neu al P ogeni o Cell Implan a ion Unde gene al anes hesia (4% chlo al hyd a e, 1 ml/100 g, i.p.), adul a s we e placed on a s e eo axic ame and a window ephined in he occipi al bone. The igh MLF was ansec ed by means o a 750-mm wid h mic oblade aimed wi h a 45uan e io angle a he ollowing s e eo axic coo dina es: 3.5 mm caudal o lambda and 0 mm la e al [32]. Figu e 1C shows he si e o axo omy in a schema ic d awing o a a b ains em pa asagi al sec ion. Then, lesioned a s we e di ided in o wo g oups acco ding o he ype o p ocedu e pe o med a e he axo omy: he implan ed g oup and he axo omized g oup. Ra s o he implan ed g oup ecei ed an injec ion o GFP-exp essing neu o- sphe e-de i ed cells (one ml o a suspension o 50,000 iable cells/ ml p epa ed in DF-12). Cells we e injec ed wi h a Hamil on mic osy inge coupled o a glass capilla y in he same coo dina es o he lesion (Figu es 1C–E). Ra s om he axo omized g oup ecei ed only ehicle injec ion (one ml o DF-12) a e he MLF ansec ion. The unila e al ansec ion o he igh MLF dis up s he axons o he in e nuclea neu ons o he le abducens nucleus. As he le MLF was no ansec ed, igh side abducens in e nuclea neu ons emained unlesioned and we e used as con ol neu ons h oughou his s udy. Eigh weeks a e lesion ollowed by cell implan o ehicle injec ion, animals we e pe used ansca dially unde deep anes hesia (sodium pen oba bi al, 50 mg/kg, i.p.) wi h 100 ml o physiological saline ollowed by 250 ml o 4% pa a o maldehyde in 0.1 M sodium phospha e bu e , pH 7.4. The b ains em was VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 2 Janua y 2013 | Volume 8 | Issue 1 | e54519 emo ed and c yop o ec ed by imme sion in a solu ion o 30% suc ose in sodium phospha e-bu e ed saline (PBS) un il i sank. Then, o c yos a sec ioning, he b ains em was cu in wo pieces: he mo e os al one, con aining he si e o lesion in he MLF, was sec ioned sagi ally (Figu e 1C–E) and he caudal one, con aining he abducens nucleus, was sec ioned co onally (schema ically ep esen ed in Figu e 1F). All he sec ions we e 40-mm hick. Sec ions we e di ided in pa allel se ies o be p ocessed o di e en immunos ainings. Immunohis ochemis y A e washing in PBS, co onal sec ions con aining he abducens nucleus we e incuba ed wi h 1% sodium bo ohyd ide o 10 min o an igen e ie al. They we e insed again in PBS and incuba ed Figu e 1. Neu al p ogeni o cell cul u e, axo omy and cell implan . A. Floa ing neu osphe e ob ained om neu al p ogeni o s o he pos na al a sub en icula zone (SVZ). Scale ba : 25 mm. B. GFP-exp essing cells (in g een) in a SVZ-de i ed neu osphe e. Scale ba : 25 mm. C. Schema ic d awing o a a pa asagi al b ains em sec ion showing he loca ion o he medial longi udinal ascicle (MLF) ansec ion and cell implan . Abducens in e nuclea neu ons a e ep esen ed in ed and implan ed neu al p ogeni o cells in g een. Axons o abducens in e nuclea neu ons ( ed lines) cou se h ough he MLF owa ds he con ala e al oculomo o nucleus. The dis al s ump o dis up ed axons a e ep esen ed in ed dashed lines. D. Con ocal mic oscopy image o a pa asagi al b ains em sec ion showing he implan ed cells labeled wi h GFP (in g een) a he si e o axo omy. Dashed lines indica e he app oxima e do so- en al limi s o he MLF. Scale ba : 100 mm. E. Highe magni ica ion image o implan ed GFP-labeled cells. Do so- en al and os o-caudal o ien a ion as in D. Scale ba : 50 mm. F. Schema ic ep esen a ion o a a co onal sec ion h ough he pons showing he abducens nucleus loca ion. Abducens in e nuclea neu on soma a a e ep esen ed in ed. Abb e ia ions: ABD: abducens nucleus; C: caudal; CB: ce ebellum; D: do sal; GFP: g een luo escen p o ein; MLF: medial longi udinal ascicle; OCM: oculomo o nucleus; py: py amidal ac ; R: os al; V: en al; 7n: acial ne e. doi:10.1371/jou nal.pone.0054519.g001 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 3 Janua y 2013 | Volume 8 | Issue 1 | e54519 o 1 hou in a blocking solu ion consis ing o 5% no mal donkey se um in PBS wi h 0.01% T i on X-100 (PBS-T). Following he blocking ea men , sec ions we e incuba ed o e nigh a oom empe a u e wi h one o he ollowing p ima y an ibodies: abbi an i-VEGF (San a C uz Bio echnology, sc-507, 1:100), abbi an i- BDNF (San a C uz Bio echnology, sc-546, 1:400) o abbi an i- NT-3 (San a C uz Bio echnology, sc-547, 1:400), p epa ed in blocking solu ion. A e se e al inses in PBS, sec ions we e incuba ed wi h a bio inyla ed seconda y an ibody in PBS-T (bio inyla ed an i- abbi IgG Jackson ImmunoResea ch, 1:500), washed again and inally ans e ed o 45 minu es o a solu ion con aining s ep a idin-FITC (Jackson ImmunoResea ch, 1:800). Then, in o de o iden i y abducens in e nuclea neu ons, subsequen immunohis ochemis y o he de ec ion o cal e inin was pe o med since his p o ein is selec i ely exp essed wi hin he abducens nucleus by his popula ion o neu ons [33]. The p ima y an ibody was a goa an i-cal e inin om Swan (1:500) and he an ibody binding was made isible by incuba ion wi h an an i- goa -TRITC (Jackson ImmunoResea ch, 1:100). In o he expe i- men s, when he de ec ion o abducens mo oneu ons was equi ed, a goa an i-choline ace yl ans e ase (ChAT) an ibody (Millipo e, 1:500) ollowed by an i-goa -TRITC incuba ion was used. Sec ions we e hen washed o se e al imes, moun ed on glass slides and co e slipped wi h a 0.1 M solu ion o p opyl galla e p epa ed in glyce ol:PBS (9:1). The immunohis ochemical p o ocol o NGF a ied sligh ly since a di e en an igen e ie al me hod was needed, which consis ed o incuba ing he sec ions in 0.01 M sodium ci a e bu e , pH 6, a 74uC o 40 min. The emaining p o ocol was he same as desc ibed abo e excep ha incuba ion wi h he p ima y an ibody ( abbi an i-NGF; San a C uz Bio echnology, sc- 548, 1:100) las ed o 72 hou s a 4uC. Finally, cal e inin immunohis- ochemis y and moun ing on slides we e pe o med as desc ibed abo e. I is impo an o poin ou ha , in o de o a oid di e ences in immunos aining in ensi y due o me hodological p ocedu es, se ies o sec ions om implan ed and axo omized g oups we e always p ocessed simul aneously o each neu o ophic ac o . To analyze neu o ophic ac o exp ession by implan ed cells, pa asagi al sec ions con aining he lesion and implan si e we e p ocessed o single immunohis ochemis y o VEGF, BDNF, NT-3 o NGF, ollowing he same desc ibed p ocedu e bu a s ep a idin-DyLigh 649 was used (Jackson ImmunoResea ch, 1:800) ins ead o a FITC-coupled s ep a idin, since implan ed cells we e al eady labeled in g een by hei GFP exp ession. An ibody Cha ac e iza ion Fo posi i e con ols, he immunohis ochemis y p o ocol was pe o med in b ain sec ions ha con ained a eas p e iously epo ed as immunoposi i e o each o he neu o ophic ac o s used in his s udy. Nega i e con ols ca ied ou by omission o he p ima y an ibodies esul ed in absence o s aining in all cases. Addi ional nega i e con ols we e pe o med by p eadso p ion o he an ibodies wi h hei blocking pep ides. Speci ically, each an ibody and i s co esponding immunizing pep ide we e mixed o 2 hou s a a 1:20 p opo ion be o e being used o he immunohis ochem- ical p o ocol. Sec ions incuba ed wi h his p eadso bed solu ion showed no speci ic s aining. Fo he VEGF p ima y an ibody, a blocking pep ide was no a ailable by he p o ide . Then, as nega i e con ol, he p ima y an ibody incuba ion was eplaced by incuba ion wi h no mal abbi IgG (Jackson ImmunoResea ch) a he same concen a ion han he p ima y an ibody. Again, no speci ic labeling was ound. Con ocal Mic oscopy and Op ical Densi y Quan i ica ion Con ocal mic oscopy images o abducens neu ons we e cap u ed in o de o analyze di e ences be ween g oups in he in ensi y o he VEGF, BDNF, NT-3 and NGF immunos ainings. Fo each neu o ophic ac o , all abducens neu ons iden i ied in one ou o h ee co onal sec ions om each animal we e cap u ed wi h a con ocal lase scanning mic oscope (Zeiss LSM 7 DUO) a 63X magni ica ion and 8-bi esolu ion. Di e en ocal planes con aining he cell nucleus and sepa a ed 2 mm in he z axis we e scanned by exci ing he FITC and TRITC luo opho es wi h 488- nm a gon and DPSS 561-nm lase s, espec i ely, wi h a pinhole ape u e o 1 Ai y Uni . Be o e scanning, g ay scales we e adjus ed o maximize hei dynamic ange. Acquisi ion se ings we e kep cons an du ing he image cap u e o neu ons om he le (lesioned) and igh (unlesioned) abducens nucleus o he same sec ion. Cap u ed images we e analyzed wi h ImageJ (NIH). In o de o de e mine he in ensi y o he immunos aining, we calcula ed he mean g ay alue o he cell immunolabeling (i.e., he op ical densi y) by manually ou lining he cell cy oplasm in he co esponding scanned images. Op ical densi y measu emen s ha e been demons a ed o co ela e signi ican ly wi h he amoun o p o ein p esen in he b ain issue [34] and ha e been widely used by o he au ho s o compa e p o ein le els be ween g oups [35,36,37,38]. Op ical densi y alues o VEGF, BDNF, NT-3 and NGF immuno eac i ies we e de e mined o neu ons o he lesioned and he unlesioned (con ol) sides o bo h axo omized and implan ed animals and da a we e inally exp essed as pe cen ages ela i e o he con ol side. S a is ics Compa isons be ween g oups we e ca ied ou in Sigma Plo 11 (Sys a So wa e) by using one-way ANOVA ollowed by pos -hoc mul iple compa isons (Dunns me hod) a a signi ican le el o p,0.05. Resul s An ibody Cha ac e iza ion Posi i e and nega i e con ols we e ca ied ou o es he speci ici y o he an ibodies used in his s udy o he immunoi- den i ica ion o he neu o ophic ac o s BDNF, NT-3, NGF and VEGF. Ini ially, he immunohis ochemis y p o ocol and he an ibody i a ion we e es ablished in sec ions con aining b ain egions in which he exp ession o each ac o has been p e iously desc ibed in he li e a u e [30,39,40,41,42,43]. Thus, immuno eac i i y o BDNF was ound in a hippocampus and he immunos aining was no e iden when he an ibody was p eadso bed wi h i s immunizing pep ide (Figu es 2A and 2A’). Posi i e esul s we e ob ained o NT-3 immunos aining in ce ebella Pu kinje cells (Figu e 2B) and in hippocampus (no shown), and speci ic s aining was absen when sec ions we e incuba ed wi h he p eadso bed an ibody (Figu e 2B’). NGF immunode ec ion equi ed a mo e in ense an igen e ie al me hod a e which i s exp ession was e iden in ce ebellum (Figu e 2C) and in hippocampus (no shown). Again, p eincuba ion wi h he blocking pep ide led o nega i e s aining (Figu e 2C’). VEGF exp ession was conspicuous in Pu kinje cells o he ce ebellum (Figu e 2D) and in many o he b ain egions such as co ex o hippocampus (no shown). As he VEGF blocking pep ide was no a ailable (see Ma e ials and Me hods), sec ions we e p eincuba ed wi h no mal abbi IgG ins ead o he abbi - VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 4 Janua y 2013 | Volume 8 | Issue 1 | e54519 made p ima y an ibody, which also p oduced nega i e labeling in he ce ebellum (Figu e 2D’) and in he o he desc ibed egions. VEGF Immuno eac i i y is Inc eased in Lesioned Abducens In e nuclea Neu ons o Implan ed Animals As p e iously desc ibed, cal e inin cons i u es a good ma ke o he in e nuclea neu ons loca ed in he abducens nucleus [33]. The e o e, we used cal e inin immunolabeling o he iden i ica- ion o his cell ype. In all expe imen al g oups (i.e., con ol, axo omized and implan ed), he double immuno luo escence pe o med agains cal e inin and VEGF e ealed ha i ually all cal e inin-posi i e abducens cells we e also s ongly immuno- eac i e o VEGF (Figu e 3A). We u he e alua ed he in ensi y o he immunolabeling by quan i ying he op ical densi y wi hin he neu onal soma o abducens in e nuclea neu ons, as desc ibed in Ma e ials and Me hods. Thus, axo omy did no a ec he in ensi y o VEGF immuno eac ion, since inju ed neu ons showed simila alues o VEGF op ical densi y in compa ison wi h con ol cells (Figu e 3A and 3B). In e es ingly, abducens in e nuclea neu ons om animals ha ecei ed he NPC implan s showed a signi ican inc ease in VEGF op ical densi y, as compa ed wi h bo h con ol and axo omized cells (Figu e 3B; p,0.05, ANOVA es ; n = 60–67 neu ons in each g oup ob ained om 5 animals). BDNF Immuno eac i i y is Inc eased in Lesioned Neu ons o Axo omized Animals bu no o Implan ed Animals All cal e inin-immunoposi i e neu ons examined in he abdu- cens nucleus also showed a s ong immuno eac i i y agains BDNF, as obse ed in he h ee expe imen al g oups (Figu e 3A). Howe e , he in ensi y o BDNF immunolabeling in abducens in e nuclea neu ons changed eigh weeks a e MLF ansec ion. Thus, BDNF op ical densi y aised signi ican ly in axo omized neu ons in ela ion o con ol alues (Figu e 3C). In con as , axo omized neu ons om animals p o ided wi h NPC implan s did no exhibi any change in BDNF op ical densi y, as compa ed wi h he con ol g oup (Figu e 3C; n = 32–54 neu ons analyzed in each g oup om 4 animals). Changes in VEGF and BDNF Immuno eac i i y in he Abducens Nucleus a e Selec i e o he Inju ed Cell Popula ion In o de o es whe he he e ec s o cen al axo omy and NPC implan s on VEGF and BDNF exp ession we e es ic ed o he inju ed neu onal popula ion wi hin he abducens nucleus (i.e., he abducens in e nuclea neu ons) we also examined VEGF and BDNF immuno eac i i y in he mo oneu ons o he same nucleus, whose axons we e le in ac . Abducens mo oneu ons we e iden i ied by hei ChAT immunos aining (Figu e 4A and C). Bo h VEGF and BDNF we e exp essed by i ually all abducens mo oneu ons analyzed in he h ee expe imen al g oups (n = 58– 79 mo oneu ons pe g oup ob ained om 3 animals) (Figu e 4A– D), bu hei op ical densi y alues we e no a ec ed by ei he he MLF ansec ion o he cell implan a ion (Figu e 4E and 4F). The e o e, abducens mo oneu ons did no modi y hei VEGF o BDNF exp ession in esponse o he ansec ed ascicle and NPC implan s, so ha he changes ound in he immuno eac i i y o hese wo neu o ophic molecules we e speci ic o he popula ion o inju ed abducens in e nuclea neu ons. Implan ed Animals do no show Changes in NT-3 o NGF Immuno eac i i y wi h Respec o Axo omized Animals Abducens in e nuclea neu ons showed a weak immunos aining agains NT-3 (Figu e 5A). The analysis o op ical densi y a e NT- 3 immunolabeling e ealed a signi ican inc ease o his pa ame e in he cell body o abducens in e nuclea neu ons as a esul o hei axo omy (Figu e 5B). Abducens in e nuclea neu ons axo omized bu p o ided wi h he NPC implan s also showed a signi ican inc ease in NT-3 op ical densi y as compa ed wi h con ol, which was simila o ha exhibi ed by he axo omized g oup (Figu e 5B). The numbe o cal e inin-immunoposi i e Figu e 2. An ibody cha ac e iza ion. Con ocal mic oscopy images o a co onal sec ions h ough he hippocampus (A, A’) and ce ebellum (B, B’, C, C’) showing immuno eac i i y o BDNF, NT-3, and NGF in posi i e con ols (A, B and C, espec i ely) and in hei espec i e blocking pep ide- p eadso bed nega i e con ols (A’, B’ and C’). D and D’. Con ocal mic oscopy images o ce ebella co onal sec ions showing VEGF immuno eac i i y in a posi i e con ol (D) and a nega i e con ol consis ing o issue p e-incuba ion wi h no mal abbi IgG ins ead o VEGF an ibody (D’). Scale ba s: 50 mm. doi:10.1371/jou nal.pone.0054519.g002 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 5 Janua y 2013 | Volume 8 | Issue 1 | e54519 neu ons ha appea ed doubly labeled ollowing he NT-3 immuno luo escence was simila in he h ee expe imen al g oups and close o 100% (n = 25–38 neu ons in each g oup ob ained om 4 animals). NGF immuno eac i i y was mainly con ined o he neu opil o he abducens nucleus, showing ain ly-s ained cell bodies (Figu e 5A). Op ical densi y quan i ica ion e idenced no signi i- can change be ween g oups in he in ensi y o NGF immuno e- ac i i y, so ha axo omized abducens in e nuclea neu ons wi h o wi hou he NPC implan showed alues o NGF op ical densi y ha we e simila o con ol (Figu e 5C; n = 26–54 neu ons pe g oup ob ained om 3 animals). VEGF Exp ession by he Implan ed Neu al P ogeni o Cells We also aimed o explo e whe he implan ed cells in he lesion si e migh cons i u e a e og ade sou ce o neu o ophic ac o s o he axo omized abducens in e nuclea neu ons, due o he close p oximi y be ween he p oximal s ump o sec ioned axons and he implan ed NPCs. Fo ha , we pe o med VEGF, BDNF, NT-3 and NGF immunohis ochemis y in b ains em pa asagi al sec ions con aining he lesion and NPC implan a ion si e (Figu e 1C–E). Implan ed NPCs we e iden i ied by i s GFP labeling. App oxima ely 25% o he GFP-posi i e NPCs examined appea ed also labeled ollowing VEGF immuno eac i i y. Figu e 6 Figu e 3. VEGF and BDNF immuno eac i i y in abducens in e neu ons. A. Con ocal mic oscopy images o abducens in e nuclea neu ons (immunoposi i e o cal e inin, in ed) showing double immuno eac i i y agains ei he VEGF (in g een, second ow o images) o BDNF (in g een, ou h ow). Examples o neu ons om he di e en expe imen al condi ions a e shown: non lesioned neu ons (CONTROL column), lesioned neu ons om he axo omized g oup (AXOTOMIZED column) and lesioned neu ons om he implan ed g oup (IMPLANTED column). Ba : 20 mm. B. Op ical densi y quan i ica ion o VEGF immuno eac i i y in unlesioned neu ons (CONTROL), lesioned neu ons om he axo omized g oup (AXOTOMIZED) and lesioned neu ons om he implan ed g oup (IMPLANTED). Values a e exp essed as pe cen ages ela i e o con ol (unlesioned neu ons). Ba s ep esen he mean 6SEM o 60–67 neu ons analyzed om i e di e en animals in each g oup. * and #indica e signi ican di e ences wi h espec o he con ol and he axo omized g oup, espec i ely (ANOVA es ollowed by Dunn’s me hod o mul iple pai wise compa isons, p,0.05). C. Same as B, bu o BDNF immuno eac i i y. Ba s ep esen he mean 6SEM o 32–54 neu ons analyzed om ou di e en animals in each g oup. * and # indica e signi ican di e ences wi h espec o he con ol and he axo omized g oup, espec i ely (ANOVA es ollowed by Dunn’s me hod o mul iple pai wise compa isons, p,0.05). doi:10.1371/jou nal.pone.0054519.g003 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 6 Janua y 2013 | Volume 8 | Issue 1 | e54519 illus a es an example o a NPC implan ed a he lesion si e which showed double labeling o bo h GFP and VEGF immuno luo- escence. In con as , none o he analyzed GFP-posi i e cells in he implan was ound o be immunoposi i e o BDNF, NT-3 o NGF. The e o e, among all neu o ophic ac o s examined, VEGF was he only ha could be deli e ed e og adely om implan ed cells o axo omized neu ons. Discussion In his s udy, we ha e demons a ed ha VEGF and BDNF exp ession in axo omized neu ons can be modula ed di e en ially by NPCs implan ed in he si e o inju y. In addi ion, NPCs de i ed om he SVZ a e able o exp ess VEGF a e implan a ion in he damaged b ain. Based on he ecen ly-demons a ed and ele an unc ions o VEGF in neu op o ec ion and synap ic modula ion, we p opose ha his neu o ophic ac o could play an impo an ole in he bene i ial e ec s o NPC implan s on axo omized abducens in e nuclea neu ons. T ansec ion o he MLF lea es abducens in e nuclea neu ons disconnec ed om hei na u al a ge in he oculomo o nucleus. The axo omized neu ons exhibi a p o ound emo al o inhibi o y and exci a o y synapses and a educ ion in eye posi ion and eloci y sensi i i ies [2,11]. In a p e ious wo k, we ha e shown ha NPCs implan ed in he ca a e ansec ion o he MLF induce a no iceable eco e y in synap ic co e age and i ing p ope ies al e ed by he axo omy [19]. As desc ibed in o he lesion models, NPCs migh induce changes o p o ein exp ession in lesioned cells which could be de e minan in he mechanism o NPC-induced eco e y [21,22,23]. The e o e, we ques ioned whe he NPC implan s we e able o modi y neu o ophic ac o exp ession in axo omized abducens in e nuclea neu ons by means o immunohis ochemical ools. Fi s , we had o es ablish eliable immunohis ochemical p o ocols wi h speci ic an ibodies aised agains VEGF, BDNF, NT-3 and NGF. Posi i e con ols e ealed immunolabeling in b ain egions p e iously desc ibed o be immunoposi i e o he men ioned ac o s such as hippocam- pus, ce eb al co ex o ce ebellum [30,39,40,41,42,43]. Nega i e con ols we e pe o med by p eadso p ion o he an ibody wi h he blocking pep ide ( o BDNF, NT-3 o NGF) o wi h no mal IgG in subs i u ion o he an ibody ( o VEGF). I is no ewo hy ha we ne e ob ained speci ic labeling in hese cases which indica ed ha an ibodies used o his s udy we e eliable and speci ic. In he con ol si ua ion, we obse ed ha all analyzed abducens in e nuclea neu ons displayed an in ense immuno eac i i y o VEGF and BDNF, whe eas NT-3 and NGF e ealed a ain s aining. Howe e , his was no he case o o he b ains em neu ons which showed an in ense immuno eac i i y o NT-3 and NGF (da a no illus a ed). The e o e, he in ense labeling ound in abducens in e nuclea neu ons speci ically a e VEGF and BDNF immuno eac i i y sugges s ha hese wo neu o ophic ac o s likely ac as impo an molecules o he unc ional main enance o his neu onal popula ion, p obably h ough au oc ine o pa ac ine pa hways, al hough he e og ade and/o an e og ade ou es can no be disca ded. A e axo omy, abducens in e nuclea neu ons did no modi y hei immuno eac i i y o VEGF. Howe e , VEGF con en was signi ican ly inc eased in hese neu ons when NPCs we e implan ed a e axo omy. Two possible explana ions may accoun o his inding: i) VEGF migh ha e been e og adely anspo ed by he sec ioned axons and accumula ed in he soma, and ii) damaged neu ons migh ha e inc eased hei VEGF exp ession in implan ed animals. Since some o he implan ed cells we e VEGF- immuno eac i e and abducens in e nuclea neu ons exp ess he ecep o o VEGF, Flk-1 (unpublished da a), he i s possibili y is easible. Wi h ega d o he second possibili y, i is well es ablished ha he new mic oen i onmen c ea ed by he c oss- alk be ween implan ed and hos issue cells can induce changes in he neu onal exp ession o ce ain molecules [21,22,23,44]. The e o e, i is also possible ha an inc ease in VEGF exp ession was induced in axo omized cells as a esul o molecula in e ac ions de i ed om his new mic oen i onmen . Independen ly om he sou ce, damaged neu ons o implan ed animals migh bene i om he VEGF inc eased con en . Indeed, ecen e idence ela es in- c eased VEGF exp ession wi h neu op o ec i e ac ions. Fo ins ance, inju ed neu ons a e insul s such as ischemia o b ain seizu es inc ease hei VEGF exp ession and p obably his migh ope a e as an endogenous p o ec i e mechanism ha educes Figu e 4. VEGF and BDNF immuno eac i i y in abducens mo oneu ons. A o D. Con ocal mic oscopy images o con ol abducens mo oneu ons showing double immunolabeling o ChAT (A, in ed) and VEGF (B, in g een), o o ChAT (C, in ed) and BDNF (D, in g een). Scale ba : 20 mm. E and F. Op ical densi y (O.D.) quan i ica ion o VEGF immuno eac i i y (E) and BDNF immuno eac i i y (F) in abducens mo oneu ons om ei he he unlesioned side (CON), he lesioned side o he axo omized g oup (AX) o he lesioned side o he implan ed g oup (IMP). Values ep esen pe cen ages wi h espec o con ol (mo oneu ons om he unlesioned side). Ba s show he mean 6SEM o 58–79 mo oneu ons analyzed om h ee di e en animals in each g oup. No signi ican di e ences we e ob ained be ween g oups (ANOVA es ollowed by Dunn’s me hod o mul iple pai wise compa isons). doi:10.1371/jou nal.pone.0054519.g004 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 7 Janua y 2013 | Volume 8 | Issue 1 | e54519 neu onal exci abili y [27,29]. Also, o e exp ession o VEGF is known o delay mo o neu on degene a ion in animal models o amyo ophic la e al scle osis [28,45]. In he ca , NPC implan s a e able o es o e mos o he axo omy-induced al e a ions in i ing p ope ies and synap ic inpu s o abducens in e nuclea neu ons [19]. We p opose ha VEGF could be in ol ed in he neu op o ec i e ac ions exe ed by NPC implan s in axo omized animals al hough u he expe imen s a e needed in o de o elucida e his issue. Abducens in e nuclea neu ons o he axo omized g oup exhibi ed a signi ican inc ease in BDNF and NT-3 con en eigh weeks a e MLF ansec ion. We ha e p e iously demons a ed ha abducens in e nuclea neu ons exp ess he k ecep o s o BDNF and NT-3 ( kB and kC, espec i ely) [15]. The e o e, by inc easing BDNF and NT-3 exp ession, axo omized neu ons migh compensa e o he lack o a ge -de i ed ophic suppo ia au oc ine o pa ac ine pa hways [7,8,46]. Howe e , in he implan ed g oup, BDNF con en o abducens in e nuclea neu ons was no inc eased a e axo omy and emained in alues simila o heal hy neu ons. P obably, o he endogenously- eleased molecules in he new mic oen i onmen gene a ed by he NPC implan migh eplace BDNF-dependen neu o ophic ac ions, and conse- quen ly his neu o ophin migh be down- egula ed as compa ed o he axo omy si ua ion. This was no he case o NT-3, since his Figu e 5. NT-3 and NGF immuno eac i i y in abducens in e neu ons. A. Con ocal images o abducens in e neu ons (immunoposi i e o cal e inin, in ed) showing double immuno eac i i y agains ei he NT-3 (in g een, second ow o images) o NGF (in g een, ou h ow). Examples o neu ons om he di e en expe imen al condi ions a e shown: non lesioned neu ons (CONTROL column), lesioned neu ons om he axo omized g oup (AXOTOMIZED column) and lesioned neu ons om he implan ed g oup (IMPLANTED column). Scale ba : 20 mm B. Op ical densi y quan i ica ion o NT-3 immuno eac i i y in unlesioned neu ons (CONTROL), lesioned neu ons om he axo omized g oup (AXOTOMIZED) and lesioned neu ons om he implan ed g oup (IMPLANTED). Values a e exp essed as pe cen ages wi h espec o con ol (unlesioned neu ons). Ba s ep esen he mean 6SEM o 25–38 neu ons analyzed om ou di e en animals in each g oup. *, p,0.05 compa ed o he con ol g oup, no signi ican di e ences we e ob ained be ween he axo omized and he implan ed g oups (ANOVA es ollowed by Dunn’s me hod o mul iple pai wise compa isons, p,0.05). C. Same as B, bu o NGF immuno eac i i y. Ba s ep esen he mean 6SEM o 26–54 neu ons analyzed om h ee di e en animals in each g oup. No e ha he e we e no signi ican di e ences be ween g oups (ANOVA es ollowed by Dunn’s me hod o mul iple pai wise compa isons, p,0.05). doi:10.1371/jou nal.pone.0054519.g005 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 8 Janua y 2013 | Volume 8 | Issue 1 | e54519 neu o ophin inc eased signi ican ly in bo h he axo omized and implan ed g oups and, he e o e, nei he he NPCs no o he cells om he hos issue ha in e ac ed wi h he NPCs could modi y he axo omy-induced inc ease in NT-3 con en . Since NGF immuno eac i i y in abducens in e nuclea neu ons was low and, in addi ion, was no a ec ed by ei he he axo omy o he NPC implan s, i seems ha his neu o ophin is much less ele an in con ol as well as a e lesion and implan . A no iceably inding o he p esen s udy was ha he mo oneu ons o he abducens nucleus, which lie in e mingled wi h he in e nuclea neu ons, did no show any change in he in ensi y o hei immunos aining o ei he VEGF o BDNF. Since he same me hodological p o ocol was used o bo h cell ypes, hese da a clea ly alida e he modi ica ions ound in he popula ion o abducens in e nuclea neu ons a e NPC implan s and, mo eo e , demons a e ha changes in neu o ophic ac o s we e exclusi e o hose neu ons ha we e axo omized by he lesion p ocedu e. As s a ed be o e, implan ed cells migh cons i u e a new sou ce o ophic ac o s o axo omized cells. In ac , NPCs om he SVZ exp ess VEGF in he neu ogenic niche [24] and also when hey a e cul u ed in he o m o neu osphe es [47]. Fage lund e al. [18] ha e ecen ly epo ed ha NPCs de i ed om he SVZ and ansplan ed in he lesioned hypoglossal nucleus exp ess VEGF and p omo e mo o neu on su i al. In line wi h hose indings, we pu sued o in es iga e in ou CNS lesion model whe he SVZ neu osphe e-de i ed cells we e able o exp ess VEGF o o he neu o ophic ac o s eigh weeks a e implan a- ion in he hos lesioned issue. Ou esul s showed ha none o he analyzed implan ed cells we e immuno eac i e o BDNF, NT-3 o NGF, bu a mode a e pe cen age (app oxima ely 25%) was immunoposi i e o VEGF, which indica ed ha implan ed NPCs could elease VEGF in he p oximi y o he p oximal s umps o se e ed axons. In summa y, his s udy demons a es ha NPCs implan ed in he inju ed b ain exp ess VEGF eigh weeks a e implan a ion in he hos lesioned issue and a e able o modula e VEGF and BDNF exp ession in lesioned neu ons. We p opose ha he inc eased VEGF exp ession p omo ed by NPCs in inju ed neu ons migh be in ol ed in he es o a i e e ec s o NPC implan s. Suppo ing In o ma ion Figu e S1 Cha ac e iza ion o neu osphe e cul u es. A-B. Phase-con as mic oscopy images o loa ing neu osphe es (A) and dissocia ed neu osphe e cells g own as a monolaye on a poly- lysine subs a e (B). C, E, G. Fluo escence mic oscopy images o neu osphe es immunos ained wi h an ibodies agains nes in (C, ed), bIII- ubulin (E, a owhead, ed), GFAP (E, a ow, g een) and NG-2 (G, g een) and coun e s ained wi h DAPI (blue). D, F, H. Fluo escence mic oscopy images o adhe ed cells immunos ained wi h an ibodies agains nes in (D, ed), bIII- ubulin (F, a owhead, ed), GFAP (F, a ow, g een) and NG-2 (H, g een) and coun e - s ained wi h DAPI (blue). Scale ba s: 50 mm in A, B and D; 25 mm in C and F; 15 mm in E, G and H. (TIF) Suppo ing In o ma ion S1 Cha ac e iza ion o neu osphe e cul u es. (DOCX) Acknowledgmen s Some expe imen s we e pe o med in he Biology and Mic oscopy Cen al Se ices o he Uni e sidad de Se illa (CITIUS). We wish o hank D . Juan Luis Ribas o his help wi h he con ocal mic oscope and D . Da id Macı ´as o kindly p o iding he len i i us used in his s udy. Au ho Con ibu ions Concei ed and designed he expe imen s: ERM RRC AMP. Pe o med he expe imen s: RT ERM AMP. Analyzed he da a: RT ERM RRC. W o e he pape : ERM RRC AMP. Re e ences 1. Ti mus MJ, Fabe DS (1990) Axo omy-induced al e a ions in he elec ophys- iological cha ac e is ics o neu ons. P og Neu obiol 35: 1–51. 2. Pas o AM, Delgado-Ga cı ´a JM, Ma ı ´nez-Guija o FJ, Lo´pez-Ga cı ´aC,dela C uz RR (2000) Response o abducens in e nuclea neu ons o axo omy in he adul ca . J Comp Neu ol 427: 370–390. 3. Giehl KM, Te zla W (1996) BDNF and NT-3, bu no NGF, p e en axo omy- induced dea h o a co icospinal neu ons in i o. Eu J Neu osci 8: 1167–1175. 4. Kobayashi NR, Fan DP, Giehl KM, Beda d AM, Wiegand SJ, e al. (1997) BDNF and NT-4/5 p e en a ophy o a ub ospinal neu ons a e ce ical axo omy, s imula e GAP-43 and Talpha1- ubulin mRNA exp ession, and p omo e axonal egene a ion. J Neu osci 17: 9583–9595. 5. Da is-Lo´pez de Ca izosa MA, Mo ado-Dı ´az CJ, Tena JJ, Benı ´ ez-Temin˜o B, Pece o ML, e al. (2009) Complemen a y ac ions o BDNF and neu o ophin-3 on he i ing pa e ns and synap ic composi ion o mo oneu ons. J Neu osci 29: 575–587. 6. Da is-Lo´pez de Ca izosa MA, Mo ado-Dı ´az CJ, Mo cuende S, de la C uz RR, Pas o AM (2010) Ne e g ow h ac o egula es he i ing pa e ns and synap ic composi ion o mo oneu ons. J Neu osci 30: 8308–8319. 7. Vene o JL, Vizue e ML, Re uel a M, Va gas C, Cano J, e al. (2000) Up egula ion o BDNF mRNA and kB mRNA in he nig os ia al sys em and in he lesion si e ollowing unila e al ansec ion o he medial o eb ain bundle. Exp Neu ol 161: 38–48. Figu e 6. VEGF immuno eac i i y in neu al p ogeni o implan ed cells. Con ocal mic oscopy images o an implan ed cell a he lesion si e iden i ied by i s GFP exp ession (A, in g een), which was also immunoposi i e o VEGF (B, in blue). C shows he me ged image. Scale ba : 10 mm. doi:10.1371/jou nal.pone.0054519.g006 VEGF in Lesioned Neu ons a e NPC Implan s PLOS ONE | www.plosone.o g 9 Janua y 2013 | Volume 8 | Issue 1 | e54519