Mapping of Brain lipid binding protein (Blbp) in the brain of adult zebrafish, co-expression with aromatase B and links with proliferation

HAL Id: hal-01237088

https://hal-univ-rennes1.archives-ouvertes.fr/hal-01237088

Submitted on 6 Feb 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Mapping of Brain lipid binding protein (Blbp) in the brain of adult zebrafish, co-expression with aromatase B

and links with proliferation

Nicolas Diotel, Colette Vaillant, Olivier Kah, Elisabeth Pellegrini

To cite this version:

Nicolas Diotel, Colette Vaillant, Olivier Kah, Elisabeth Pellegrini. Mapping of Brain lipid binding protein (Blbp) in the brain of adult zebrafish, co-expression with aromatase B and links with pro- liferation. Gene Expression Patterns, Elsevier, 2016, 20 (1), pp.42-54. �10.1016/j.gep.2015.11.003�.

�hal-01237088�

Mapp ing of bra in l ip id b ind ing prote in (B lbp) in the bra in of adu lt zebrafish , co-express ion w ith aromatase B and l inks w ith pro l iferat ion

N ico las D iote l ^a

^,

^b

^,

^c

^,

^* , Co lette Va i l lant ^c , O l iv ier Kah ^c , E l isabeth Pe l legr in i ^c

aInserm, UMR 1188 Diabete atherothrombose Therapies Reunion Ocean Indien (DeTROI), plateforme CYROI, Sainte-Clotilde, F-97490, France

bUniversitedeLaReunion, UMR 1188, Sainte-Clotilde, F-97490, France

cInserm U1085, Universite de Rennes 1, Campus de Beaulieu, Research Institute in Health, Environment and Occupation, SFR Biosit, Rennes cedex, France

abstract

Adu lt fish exh ib it a strong neurogen ic capac ity due to the pers istence of rad ia l g l ia l ce l ls . In zebrafish , rad ia l g l ia l ce l ls d isp lay we l l-estab l ished markers such as the estrogen-synthes iz ing enzyme (AroB) and the bra in l ip id b ind ing prote in (B lbp) , wh ich is known to strong ly b ind omega-3 po lyunsaturated fatty ac ids such as docosahexaeno ic ac id (DHA) . Wh i le B lpb is ma in ly descr ibed in the te lencepha lon of adu lt zebrafish , its express ion in the rema in ing reg ions of the bra in is poor ly documented . The present study was des igned to further invest igate B lbp express ion in the bra in , its co-express ion w ith AroB , and its l ink w ith rad ia l g l ia l ce l ls pro l iferat ion in zebrafish . We generated a comp lete and deta i led mapp ing of B lbp express ion in the who le bra in and show its comp lete co-express ion w ith AroB , except in some tecta l and hypotha lam ic reg ions . By perform ing PCNA and B lbp immunoh isto-c hem istry on cyp19a1b-GFP (AroB-GFP) fish , we a lso demonstrated preferent ia l B lbp express ion in pro l iferat ive rad ia l g l ia l ce l ls in a lmost a l l reg ions stud ied . To our know ledge , th is is the first comp lete and deta i led mapp ing of B lbp-express ing ce l ls show ing strong assoc iat ion between B lbp and rad ia l g l ia l ce l l pro l iferat ion in the adu lt bra in of fish . G iven that zebrafish is now recogn ized mode ls for study ing neurogenes is and bra in repa ir , our data prov ide deta i led character izat ion of B lbp in the ent ire bra in and open up a broad fie ld of research invest igat ing the ro le of omega-3 po lyunsaturated fatty ac ids in neura l stem ce l l act iv ity in fish .

1 . Introduct ion

Bra in l ip id b ind ing prote in (B lbp) is a fatty ac id b ind ing prote in but its exact funct ions are poor ly understood . It is be l ieved to p lay a key ro le in l ip id storage , membrane synthes is , energy product ion v ia l ip id transport to the m itochondr ia , cho lestero l metabo l ism and ce l l s igna l ing . Its ro le in ce l l s igna l ing is thought to invo lve pa ir ing o f fatty ac ids and transcr ipt ion factors , such as members o f the perox isome pro l i ferator-act ivated receptor fam i ly (PPAR) (Kre l in et a l . , 2007) . Dur ing bra in deve lopment in rats , B lbp is detected in the nuc leus and cytop lasm o f rad ia l g l ia in both the embryon ic ventr icu lar zone and in Bergmann g l ia o f the postnata l cerebe l lum . Its express ion corre lates w ith neurona l d i f ferent iat ion notab ly in the postnata l cerebe l lum , the embryon ic sp ina l cord , and the ce- rebra l cortex (Feng et a l . , 1994) . In adu lt m ice , B lbp is a lso expressed

in rad ia l g l ia- l ike ce l ls in the dentate gyrus o f the h ippocampus , in Bergman g l ia in the cerebe l lum (Kurtz et a l . , 1994) and in a lmost a l l astrocytes or astrocyte- l ike ce l ls o f the subventr icu lar zone (Hart fuss et a l . , 2001) . S im i lar ly , in adu lt b ird bra ins , B lbp is detected in te lencepha l ic and cerebe l lar rad ia l g l ia l ce l ls as we l l as in astrocytes ( Rousse lot et a l . , 1997) . B ind ing a ffin ity assays strong ly suggested that in the centra l nervous system , docosahexaeno ic ac id (DHA ; 22 :6 ( u -3)) is the l igand for B lbp in the centra l nervous system (Xu et a l . , 1996) . DHA is the most abundant omega-3 po lyunsaturated fatty ac ids and is essent ia l for appropr iate ret ina l and neura l deve lopment . Moreover , severa l stud ies suggest that DHA has protect ive ro le in the bra in aga inst age-re lated d isorders and neurodegenerat ive d iseases such as A lzhe imer 's (Co le et a l . , 2010 ; Craw ford et a l . , 2013 ; Ho f fman et a l . , 2009) . DHA can be loca l ly metabo l ized in the organ ism to generate new ly descr ibed

*Corresponding author. Inserm U1188, UniversitedeLaReunion, Equipe DeTROI Plateforme CYROI, 2 rue Maxime Riviere, 97 490, Sainte-Clotilde, France. E-mail address:nicolas.diotel@univ-reunion.fr(N. Diotel).

l ip ids such as reso lv ins and neuropro tect ins , autaco ids w ith po- tent ia l ant i- inflammatory propert ies , (Gronert , 2008) . Knock -out (KO) exper iments have shown that B lbp d isrupted m ice d isp lay emo t iona l and behav iora l d isturbances , inc lud ing anx iety (Owada et a l . , 2006) . In the ir work ,Owada et a l . (2006) d id not observe any abnorma l ity at the leve l o f t issue organ izat ion and g l ia l morpho logy in the cerebra l and cerebe l lar cortex , the h ippocampus or the hypotha lamus in embryon ic and mature KO an ima ls . How- ever , more recent data ind icate that B lbp and DHA are invo lved in neurogen ic processes , such as neura l stem ce l l ma intenance , pro- l i ferat ion and d i f ferent iat ion ( Boneva et a l . , 2011 ; Matsumata et a l . , 2012 ; Schne l l et a l . , 2014 ; Watanabe et a l . , 2007) , suggest ing that DHA and B lbp are indeed important factors for bra in deve lopment and neurogenes is .

Adu lt neurogenes is is a comp lex and mu lt istep process regu- lated by intr ins ic and extr ins ic factors ( Hs ieh , 2012) . Among these factors , hormones , notab ly estrogens , exert p le iotrop ic act ions to modu late severa l aspects o f neurogenes is such as pro l i ferat ion , neurona l d i f ferent iat ion and surv iva l (Brock et a l . , 2010 ; D iote l et a l . , 2013 ; Ga lea et a l . , 2006 , 2008 ; Garc ia-Segura , 2008) . These estrogens are obta ined through the convers ion o f C19 androgens by the enzyme cytochrome P450 aromatase (aromatase) . Aromatase is expressed by numerous t issues , inc lud ing gonads and bra in o f a l l vertebrates spec ies stud ied so far (S impson et a l . , 1994) . Very interest ing ly , in contrast to mamma ls , the bra ins o f adu ltfish exh ib it an excep t iona l ly strong neurogen ic and aromatase act iv ity in d i f ferent reg ions such as the te lencepha lon , the preop t ic area and a lso the hypo tha lamus (D io te l et a l . , 2010a ; Gonca lves et a l . , 2008 ; Kah et a l . , 2009 ; Pasman ik and Ca l lard , 1985 ; Pe l legr in i et a l . , 2013 ; T immers et a l . , 1987) , suggest ing the strong l ink between estrogen synthes is and neurogenes is (D io te l et a l . , 2013) . The h igh leve l o f aromatase act iv ity in thefish bra in is dr iven by the intense tran- scr ipt ion o f thecyp19a1bgene cod ing for aromatase B (AroB) (D iote l et a l . , 2010a ; Lass iter and L inney , 2007 ; Mour iec et a l . , 2009 ; Pe l legr in i et a l . , 2015 , 2005 ; Tong et a l . , 2009) . Wh i le in mamma ls aromatase express ion is ma in ly observed in neurons and in react ive astrocytes fo l low ing in jury (Garc ia-Segura , 2008 ; Garc ia-Segura et a l . , 1999) , AroB infish is expressed on ly in rad ia l g l ia , spec ific ce l ls that behaves as neura l stem ce l ls (D io te l et a l . , 2010a ; For lano et a l . , 2001 ; Mour iec et a l . , 2008 ; Pe l legr in i et a l . , 2005 , 2007 , 2013 ; Tong et a l . , 2009) . Rad ia l g l ia l ce l ls have been re lat ive ly we l l- character ized in thefish bra in and part icu lar ly in the te lenceph- a lon . F irst , rad ia l g l ia l ce l ls are loca l ized in the ventr icu lar layers and exh ib it a typ ica l morpho logy character ized by a sma l l ce l l body loca l ized in the ventr icu lar layers , w ith one short process extend ing towards the ventr ic le and a long process cross ing the bra in pa- renchyma to contact the p ia l sur face w ith end feet , s im i lar to mamma ls (Bent ivog l io and Mazzare l lo , 1999 ; Rak ic , 1978) . Second , a number o f immunoh istochem ica l markers have confirmed the ident ity o f rad ia l g l ia l ce l ls in zebra fish . Among these are B lbp (bra in l ip id b ind ing prote in) , g fap (g l ia lfibr i l lary ac id ic pro te in) , S100 b (S100 ca lc ium b ind ing pro te in B) , cxcr4 (C-X-C chemok ine receptor type 4) , id1 ( inh ib itor o f DNA b ind ing 1 , a transcr ipt ion regu lator) and a lso the estrogen-synthes iz ing enzyme , aromatase B (AroB) , encoded by thecyp19a1bgene (D inare l lo , 2012 ; D io te l et a l . , 2010b ; D iote l et a l . , 2015 ; M€ arz et a l . , 2010 ; Menuet et a l . , 2005 ; Rodr iguez V ia les et a l . , 2015 ; Tong et a l . , 2009) . A lthough some stud ies reported the express ion o f B lbp in rad ia l g l ia l ce l ls in the bra in o f adu lt zebrafish (Ado l f et a l . , 2006 ; Ede lmann et a l . , 2013 ; M€ arz et a l . , 2010) , they ma in ly focus on B lbp express ion in the te lencepha lon , leav ing o ther bra in reg ions re lat ive ly unexp lored (Ado l f et a l . , 2006 ; Ede lmann et a l . , 2013 ; M€ arz et a l . , 2010) .

Numerous genes were dup l icated in te leost fish fo l low ing the genom ic dup l icat ion that occurred between 320 and 350 m i l l ion years ago (G lasauer and Neuhauss , 2014 ; S te inke et a l . , 2006) .

Among those , B lbp gene was a lso dup l icated intofabp7aandbin zebrafish .Fabp7a is strong ly expressed in the bra in o f adu lt zebrafish as shown by RT-PCR exper iments (L iu et a l . , 2004 , 2003) . In s i tuhybr id izat ion exper iments further showed thatfapb7ais strong ly expressed in the te lencepha l ic ventr icu lar zone and in the per ig lomeru lar grey zone o f the opt ic tectum (Ado l f et a l . , 2006 ; Ito et a l . , 2010) . Thus , thefabp7aiso form appears to be the ma in gene transcr ibed in the adu lt zebrafish bra in .

B lbp express ion was a lso descr ibed in g l ia l ce l ls from adu lt zebrafish cerebe l lum (Kas l in et a l . , 2009) . However , g iven the rea l lack o f data concern ing B lbp express ion in the who le bra in o f adu lt zebrafish , wefirst invest igated the deta i led d istr ibut ion o f B lbp- pos it ive ce l ls in the ent ire adu lt bra in us ing an a lready character - ized ant ibody . Next , we compared B lbp express ion w ith AroB one , a we l l-recogn ized marker o f rad ia l g l ia l ce l ls in fish , us ing a cyp19a1b-GFP transgen ic zebrafish l ine that fu l ly recap itu lates AroB prote in express ion in the bra in (Tong et a l . , 2009) . F ina l ly , g iven the h igh number o f neurogen ic hotspots descr ibed in the bra in o f adu lt zebrafish , we exam ined i f rad ia l g l ia l ce l l pro l i ferat ion was pre fer - ent ia l ly assoc iated w ith B lbp express ion and /or w ith AroB . 2 . Resu lts

2 .1 . B lbp is expressed in rad ia l g l ia in the who le bra in

B lbp was prev ious ly shown to labe l rad ia l g l ia l ce l ls in adu lt zebrafish and larvae (Ado l f et a l . , 2006 ; D io te l et a l . , 2010b ; M€ arz et a l . , 2010 ; Tong et a l . , 2009) .To comp lete the pattern o f B lbp express ion prev ious ly descr ibed in the adu lt te lencepha lon , we dec ided to further document its express ion by per form ing immu- noh istochem istry in the who le bra in o f adu lt ma le and fema le zebrafish (F igs . 1e4) . Our ana lys is o f B lbp express ion in the bra in o f adu lt zebrafish w i l l be descr ibed accord ing to the rostro-cauda l ax is , from the o l factory bu lbs /te lencepha lon junct ion , to the te lencepha lon , the d iencepha lon , the mes- and metencepha lon .

At the junct ion between the o l factory bu lbs and the te len- cepha lon , B lbp-pos it ive ce l ls were observed in the ventr icu lar layer (F igs . 1A , 2B and 4A) , and extend processes toward the p ia l sur face (F ig . 1A , D , F and H , see arrows) .

In the te lencepha lon , B lbp-express ing rad ia l g l ia l ce l ls were detec ted a long the ventr icu lar layer o f the ven tra l and dorsa l nuc le i o f the ven tra l te lencepha lon (Vv and Vd ,F igs . 1B , 2E and 4B) , the ventr ic le o f the dorsomed ian and dorso la tera l te lencepha lon (Dm and D l ,F igs . 1BeD , 2H and 4B and C) , as we l l as in the ventr icu lar layer o f the dorsa l and poster ior zone o f the dorsa l te lencepha l ic area (Dd and Dp ,F igs . 1B , D and 4C) . We a lso no t iced a ventr icu lar str ip devo id o f B lbp sta in ing , at the junct ion between the Vv and the Vd (F ig . 1B , as ter isks) . A lack o f B lbp express ion was prev ious ly documented in th is reg ion correspond ing to a h igh ly pro l i ferat ive area ca l led the ros tra l m igratory stream (RMS) . Th is reg ions is devo id o f rad ia l g l ia l ce l ls bu t con ta ins comm it ted progen itors ca l led neurob lasts (M€ arz et a l . , 2010) . More cauda l ly in the te lencepha lon , ventr icu lar ce l ls in the postcomm issura l nuc leus o f the ventr icu lar te lencepha l ic area (Vp) a lso express B lbp (F igs . 1D and 4C) .

In the an ter ior part o f the d iencepha lon , B lbp express ion was

de tec ted in rad ia l g l ia l ce l ls o f the an ter ior (PPa) and poster ior (PPp )

nuc le i o f the preop t ic area (F igs . 1D , E , 4C and D) and in the tha lamus

(A ,F ig . 4E) . In the anter ior , med iobasa l and cauda l part o f the hy-

po tha lamus (Hv and Hc) (F igs . 1F , H , 2N , 3E , H and 4EeG ) , numerous

B lbp-pos i t ive rad ia l g l ia l ce l ls were de tec ted and extend long cy to-

p lasm ic processes to the per iphery o f the bra in where they accu-

mu late at the p ia l sur face (F ig . 1F and H , see arrows) . In the dorsa l

zone o f the per iventr icu lar hypo tha lamus , rad ia l g l ia l ce l ls sur -

round ing the la tera l and poster ior recess (LR and PR ) o f the cauda l

hypo tha lamus exh ib it strong B lbp sta in ing (F igs . 1H , 3E and H) .

In the mesencepha lon , ven tr icu lar ce l ls o f the per iventr icu lar pretecta l nuc leus (PPv ) and o f the per ig lomeru lar gray zone o f the op t ic tectum (TeO ) a lso d isp lay B lbp sta in ing (F ig .1G ,I ,J ,3Kand4FeH . In the TeO , rad ia l g l ia l processes are c lear ly v is ib le at h igher magn i- fica t ion (F ig . 1J , see arrowhead ) . B lbp labe l ing was a lso de tected a t the per iphery o f the torus long itud ina l is (TL ,F ig . 1G and I) .

In the metencepha lon , b lbp sta in ing was a lso observed at the per iphery o f the va lvu la cerebe l l i (VCe) and more weak ly in the cerebe l lum (CCe) (F ig . 1 I , K , 3K and 4GeI) . It is a lso observed at the border o f the torus sem i-c ircu lar is and numerous fibers were detected through the nuc leus o f the med ia l long itud ina l fasc ic le (NMLF) , extend ing a long the vascu lar lacuna o f area postrema (Vas) as shown inF ig . 1I . More cauda l ly , B lbp was detected in ce l ls sur- round ing the rhombencepha l ic ventr ic le (RV) unt i l the end o f the medu l la ob longata (F ig . 1K and L) . In a l l these reg ions , B lbp was not co-expressed w ith the nuc lear neurona l marker HuC /D such as shown in the subpa l l ium (F ig . 1MeO) . We a lso d id not observe any obv ious d i f ferences in B lbp d istr ibut ion accord ing to sex .

2 .2 . B lbp and cyp19a1b-GFP express ion in rad ia l g l ia l ce l ls : co- express ion and he terogene i ty

As prev ious ly ment ioned , Aromatase B , the product o f the cyp19a1bgene , is a we l l-character ized marker o f rad ia l g l ia in the bra in o f zebrafish . In order to study the co-express ion o f B lbp and AroB in rad ia l g l ia , we per formed B lbp immunoh istochem istry on acyp19a1b-GFP transgen ic zebrafish l ine (F igs . 2e4) . Th is trans- gen ic l ine was prev ious ly shown to fu l ly recap itu late AroB express ion (Tong et a l . , 2009) . We observed the co-express ion o f B lbp w ithcyp19a1b-GFP pos it ive rad ia l g l ia in the ma in ventr icu lar reg ions o f the bra in (F igs . 2 and 4) . For instance , in the anter ior part o f the bra in , B lbp andcyp19a1b-GFP were co-expressed in rad ia l g l ia l ce l ls o f the dorsa l te lencepha l ic area (D) at the junct ion between the o l factory bu lbs and the te lencepha lon (F ig . 2AeC and 4A) . Such a co-express ion was a lso observed in the who le te len- cepha lon , notab ly in the Vv , Vd , Dm and D l (F ig . 2DeI and 4 B and C) . In the d iencepha lon , we not iced that rad ia l g l ia co-express both cyp19a1b-GFP and B lbp in the preop t ic area (F ig . 4C and F) , in the per iventr icu lar nuc leus o f the poster ior tubercu lum (TPp ,F ig . 2 JeL and 4E and F) and a lso in the anter ior part o f the hypotha lamus (F ig . 2MeO and 4E) . Interest ing ly , rad ia l g l ia l ce l ls o f the PPv strong ly d isp layed B lbp sta in ing wh i lecyp19a1b-GFPisnot or on ly bare ly expressed by few ce l ls (F ig . 3AeC e see arrowse and F ig . 4F) . In the dorsa l zone o f the med iobasa l hypo tha lamus , where the latera l recess (LR) starts to open , d icho tomy a lso appeared between B lbp andcyp19a1b-GFP express ion (F ig . 3DeF) . At th is leve l , on ly a few ce l ls were GFP-pos it ive wh i le numerous B lbp- express ing rad ia l g l ia l ce l ls were observed surround ing the ventr ic le (F ig . 3DeF , see arrows) . Some ce l ls were doub le labe led (F ig . 3DeF , see arrowheads) . However , the strength o f these reg iona l d i f ferences depends on the rostro-cauda l leve l (F ig . 4F) . Such d iscrepanc ies in prote in express ion were a lso detected more cauda l ly where the LR is we l l ind iv idua l ized and v isua l ized

(F ig . 3GeI) . Thus , among rad ia l g l ia l in ing the externa l part o f the LR , we observed a h igher number o f B lbp-pos it ive ce l ls compared to GFP-express ing ce l ls (F ig . 3GeIesee arrowseandF ig . 4G) . In contrast , rad ia l g l ia l ce l ls l in ing the inner part o f the LR ma in ly co- expressed both markers (F ig . 3GeI , see arrowheads) . D i f ferences were a lso not iced in rad ia l g l ia o f the per ig lomeru lar gray zone o f the opt ic tectum where rad ia l g l ia l ce l ls were more numerous and d isp layed a stronger sta in ing for B lbp than for cyp19a1b-GFP . S im i lar ly , we detected B lbp-pos it ive structures in the NMLF and the va lvu la o f the cerebe l lum wh i le no GFP was observed (F ig . 3J eL andF ig . 4G and H) .

2 .3 . Rad ia l g l ia pro l i fera t ion : is there a pre feren t ia l l ink w i th b lbp and /or AroB express ion?

In order to determ ine poss ib le or pre ferent ia l assoc iat ion be- tween B lbp and/or AroB express ion in pro l i ferat ive rad ia l g l ia l ce l ls , we per formed PCNA and B lbp immunoh istochem istry on cyp19a1b-GFP transgen ic l ine and focused on d i f ferent neurogen ic reg ions o f the bra in (F ig . 5,Tab le 1) . In order to avo id potent ia l d i f ferences assoc iated to sex and ages , we on ly used 6 fema les o f the same age . As i l lustrated inF ig . 5, we observed pro l i ferat ive ce l ls in the dorsomed ian te lencepha lon that are B lbp-pos it ive on ly (F ig . 5,A1andB1) ,cyp19a1b-GFP-pos it ive on ly (F ig . 5,A2 and B2) ,cyp19a1b-GFP and B lbp-pos it ive (F ig . 5,A 3andB3)and others that were ne ithercyp19a1b-GFP nor B lbp-pos it ive (F ig . 5A , A4 and B4) . Th is latest ce l l type is not rad ia l g l ia but correspond to further comm itted precursors as prev ious ly descr ibed ( M€ arz et a l . , 2010) . Interest ing ly , there was a s ign ificant ly h igher proport ion o f PCNA-pos it ive rad ia l g l ia l ce l ls that co-expressed both markers in a l l the reg ions stud ied (54 .9% for PPp , 55 .4% for Hd LR , 55 .6% for D l , 61 .2% for Vd , 66 .6% for Vv , 68 .6% for Dm , 79 .9% for Vs , 81 .8% for PPa , 85 .4% for Vp , seeF ig . 5Cand DandTab le 1) , except in the tha lamus and the PPv where respect ive ly 75 .1% and 90 .4% o f pro l i ferat ive rad ia l g l ia l ce l ls expressed on ly B lbp . Thus , in a l l bra in areas stud ied except for the dorsa l reg ions o f the per iven- tr icu lar hypotha lamus (Hd LR ,F ig . 5GandTab le 1) , the rema in ing pro l i ferat ive ce l ls that do not co-express both markers were ma in ly B lbp-pos it ive wh i le on ly few ce l ls d isp layed on ly cyp19a1b-GFP sta in ing (1 .2% for Vv , 0 .7% for Vd , 1 .4%e5 .3% for PPa , 0% for Vs , 5 .8% for Vp , 1 .7% for Dm , 0 .92% for D l , 14 .3% for PPp) . Indeed , there are s ign i ficant ly more PCNA-pos it ive rad ia l g l ia express ing on ly B lbp , except for in the PPa , PPp and the Vp (F ig . 5 andTab le 1) .

Indeed , 75 .1% and 90 .4% o f PCNA-pos it ive rad ia l g l ia l ce l ls expressed on ly B lbp in the tha lamus and PPv respect ive ly , wh i le the rema in ing ce l ls ma in ly co-expressed B lbp andcyp19a1b-GFP . On ly rare pro l i ferat ive rad ia l g l ia l ce l ls express on lycyp19a1b-GFP (F ig . 5E and F andTab le 1) . In contrast , in the Hd LR where the latera l recesses start to open , the d istr ibut ion o f B lbp and cyp19a1b-GFP in pro l i ferat ive rad ia l g l ia is changed . Around 55%

o f PCNA-pos it ive rad ia l g l ia l ce l ls co-expressedcyp19a1b-GFP and B lbp , fo l lowed by 34% that on ly expresscyp19a1b-GFP and 10%

Fig. 1.Blbp distribution in the brain of adult zebrafish. AeL: Blbp (red)immunohistochemistry on paraffin brain sections of adult zebrafish. DAPI counterstaining (blue) allows the visualization of cell nuclei. Blbp stainingis detectedin radial glial cells along the ventricularlayers of the dorsal telencephalic area at thejunction with the olfactory bulbs (A),in the ventral and dorsal nuclei of the ventral telencephalon, andin the dorsomedian, the dorsolateral, the dorsal and the posterior zone of the dorsal telencephalic area(BeD).Inthe RMS region (see asterisks in B), no Blbp staining was detected. Blbp-positive radial glial cells are also observedin the preoptic area and the hypothalamus (DeF). Strong Blbp staining is notably observed in cells surrounding the lateral and posterior recess of the hypothalamus (H). Radial glial cells from the periglomerular gray zone of the optic tectum are also positivefor Blbp and extendlong cytoplasmic process to the periphery of the brain (G and J, see arrowheadsinJ),in the periventricular pretectal nucleus (G). Blbp staining around the toruslongitudinalis, the torus semicircularis,in the nucleus of the mediallongitudinal fascicle (NMLF) and along the vascularlacuna of area postrema (Vas) correspond to Blbp- positivefibers (G and I). Few positive cells are observedin the valvula of the cerebellum and the cerebellum (I and K) andlining the rhombencephalic ventricle (KeL). Arrowheads highlight some long radial cytoplasmic processes crossing the brain parenchyma to reach the periphery of the brain (C, F and J). Arrows show the end of radial glial process at the pial surface (A, D, F and H).MeO: Blbp (red) and HuC/D (green, neuronal marker) doubleimmunohistochemistry reveals that Blbpis not expressedin mature neurons. Arrowheads indicate Blbp positive radial glial cells that do not express the HuC/D neuronal marker. Bar: 16mm (E, J, M, N and O); 50mm (C, F, G and L); 100mm (A, B, D, H, I and K).

on ly sta ined for B lbp (F ig . 5G) . Th is reg ion is the on ly one to show a s ign ificant ly h igher number o f pro l i ferat ive rad ia l g l ia express ing at least cyp19a1b-GFP compared to B lbp a lone (Tab le 1) .

Taken together , these data show that pro l i ferat ive rad ia l g l ia l ce l ls express at least B lbp in a l l the neurogen ic reg ions stud ied (except Hd LR) and that pro l i ferat ion is strong ly and s ign ifi- cant ly assoc iated w ith th is marker . Indeed , in the D l , Dm , Vd , Vv and Vs pro l i ferat ive rad ia l g l ia l ce l ls expressed at least B lbp (99 .1% for D l; 98 .3% for Dm ; 99 .3% for Vd; 98 .8% for Vv and 100%

for Vs) , wh i le on ly 56 .5% e79 .9% expressed at least AroB (56 .5%

for D l; 70 .3% for Dm; 61 .9% for Vd; 67 .8% for Vv and 79 .9% for

Vs) , that is s ign ificant ly d i f ferent (Tab le 1) . The large d i f ference

in the tha lamus and the PPv are l ike ly due to the low number o f

cyp19a1b-GFP rad ia l g l ia compared to B lbp-pos it ive ce l ls (F ig . 5E

and F andTab le 1) . Interest ing ly , we observed more pro l i ferat ive

rad ia l g l ia l ce l ls around the LR express ingcyp19a1b-GFP than

B lbp (89 .9% Vs 65 .5% , respect ive ly) . To conc lude , in most cere-

bra lreg ionsstud ied(Vv , Vd ,PPa , Vs , Vp , Dm , D l ,PPp , HdLR)

pro l i ferat ive rad ia l g l ia ma in ly co-expressedcyp19a1b-GFP/B lbp .

Fig. 2.cyp19a1b-GFP and Blbp co-expressionin the brain of adult zebrafish. AeO: Blbp (red)immunohistochemistry oncyp19a1b-GFP (green) transgenic zebrafish line shows that most radial glial cells co-express Blbp and GFP. Co-expression is obvious at the junction between the telencephalon and the olfactory bulbs (AeC), in the subpallium (DeF), and in the pallium (GeI). In the anterior part of the periventricular nucleus of the posterior tuberculum, GFP and Blbp are co-expressed (JeL) as well as in the anterior region of the hypothalamus (MeO). Arrows point to radial glial processes co-expressing GFP and Blbp, while arrowheads highlight co-expression in soma. Bar: 25mm(DeO); 100mm(AeC).

In some reg ions , d iv id ing g l ia l ce l ls expressed B lbp on ly and more rare ly , rad ia l g l ia l d iv id ing ce l ls expressed so le ly cyp19a1b- GFP .

3 . D iscuss ion

3 .1 . B lbp is w ide ly expressed in rad ia l g l ia l ce l ls ac t ing as neura l progen i tors

Th is study is thefirst to generate a deta i led and comp lete mapp ing o f B lbp express ion in rad ia l g l ia l ce l ls o f the adu lt zebra- fish bra in , add ing s ign ificant ly to prev ious work (Ado l f et a l . , 2006 ; M€ arz et a l . , 2010 ; Tong et a l . , 2009) . The ident ity o f B lbp-express ing ce l ls was confirmed (1) by ce l l morpho logy and re in forced (2) by co-express ion w ith AroB , a we l l character ized marker o f rad ia l g l ia in zebra fish , us ing acyp19a1b-GFP l ine (F igs . 2e4) and a lso (3) by the absence o f co-express ion w ith the HuC /D neurona l marker (F ig . 1MeO) . However , g iven the two B lbp iso forms in fish , we

cannot exc lude fabp7b prote in detect ion by immunoh istochem istry

due to cross-react ion o f the ant ibody between fabp7a and fabp7b .

Br iefly , B lbp-pos it ive rad ia l g l ia l ce l ls were detected in the te len-

cepha lon (subpa l l ium and pa l l ium) , the d iencepha lon , and in the

mes- and metencepha lon . The express ion o f B lbp in rad ia l g l ia l

ce l ls , act ing as neura l progen itors , is in l ine w ith to what has pre-

v ious ly been descr ibed in rodents and humans dur ing the deve l-

opment and the neonata l per iod (Anthony et a l . , 2005 ; Hart fuss

et a l . , 2001 ; Hebsgaard et a l . , 2009 ; Kurtz et a l . , 1994 ; L i et a l . ,

2011 ; Murre l l et a l . , 2005 ; Podgorny i and A leksandrova , 2009) ,

and in the bra in o f adu lt mamma ls (Hebsgaard et a l . , 2009 ; Marko

et a l . , 2011 ; S te iner et a l . , 2006 ; Takahash i and Yamanaka , 2006) ,

b irds (Rousse lot et a l . , 1997) and amph ib ians (D 'Am ico et a l . , 2011) .

In the h ippocampus o f adu lt monkeys , B lbp is expressed in neura l

progen itors but not in new ly generated neurons express ing PSA-

NCAM (po lys ia ly lated neurona l ce l l adhes ion mo lecu le) (Boneva

et a l . , 2011) . S im i lar ly , in the RMS o f zebrafish , neurob lasts ex-

press PSA-NCAM but not rad ia l g l ia l markers (M€ arz et a l . , 2010) .

Fig. 3.cyp19a1b-GFP and Blbp show differences of expression in more caudal regions of the brain of adult zebrafish. AeL: Blbp (red) immunohistochemistry oncyp19a1b-GFP (green) transgenic zebrafish line show discrepancies in expression of AroB and Blbp markers. In the pretectal periventricular nucleus, radial glia mainly expresses Blbp while the transgene is only expressed in few cells (AeC). In the dorsal zone of the periventricular hypothalamus, radial glial cells surrounding the lateral recess of the hypothalamus exhibit heterogeneous expression of both markers (DeI). When the lateral recess starts to open (DeF), numerous radial glial cell express Blbp along the LR, while GFP is rarely expressed. In more caudal sections, radial glial cells of the medial part of the LR nucleus exhibit GFP and Blbp staining (arrowheads in G-I), and the“external”ones mainly express Blbp and not GFP (arrows in G-I). Blbp staining is observed around the torus semicircularis and in numerous radial glial cells from the optic tectum. In contrast, GFP is almost not detected (JeL). Bar: 25mm(GeI); 50mm (A-F and JeL).

Thus , B lbp express ion appears to be most ly assoc iated to rad ia l g l ia and less to comm itted precursors (neurob lasts) in mamma ls and fish .

We a lso reported B lbp express ion in the per ig lomeru lar grey zone o f the opt ic tectum as prev ious ly shown forfabp7a(Ito et a l . , 2010) . We observed weak B lbp sta in ing in the cerebe l lum , probab ly in Bergmann g l ia , a spec i fic type o f rad ia l g l ia shown to express B lbp in mamma ls and b irds and more recent ly in zebrafish (Boneva et a l . , 2011 ; Feng et a l . , 1994 ; Feng and He intz , 1995 ; Kas l in et a l . , 2009 ; Kurtz et a l . , 1994 ; Rousse lot et a l . , 1997) . Interest ing ly , two

d i f ferent popu lat ions o f g l ia l ce l ls were prev ious ly reported in adu lt zebrafish cerebe l lum , inc lud ing one w ith rad ia l g l ia phenotype (Kas l in et a l . , 2009) . Taken together , these data show that B lbp express ion in rad ia l g l ia is evo lut ionar i ly conserved and suggest key ro les for B lbp , DHA and poss ib ly l ip id-der ived autaco ids in bra in deve lopment stem ce l l act iv ity .

3 .2 . Rad ia l g l ia l ce l ls are no t a homogeneous popu la t ion

In th is work , we a lso observed that the d istr ibut ion o f B lbp-

Fig. 4.Schematic distribution ofcyp19a1b-GFP/Blbp co-expressing radial glial cells and Blbp-expressing radial glial cells in the whole brain of adult zebrafish (adapted from Wullimann et al., 1996). The left hemisphere provides the anatomical regions of the zebrafish brain. The black lines schematized in the right hemisphere correspond to radial glia that co-expresses GFP and Blbp and extendlong radial glial processthroughthe brain. Redlines correspondto radial glial cellsthat mainly express Blbp and not GFP(or only barely). Note that in the anterior part of the brain (AeE), most radial glial cells co-express both markers. However,in posterior part of the brain, differencesin GFP and Blbp expression are more obvious, notably at thelevel of the optic tectum, the cerebellum and the nucleus of thelateral recess (FeI).In A, the redline highlighting the periphery of the glomerularlayer (GL) of the olfactory bulbs corresponds to a strong labeling of end feet at the pial surface. Bar: 350mm.

pos it ive and Aro-B express ing (orcyp19a1b-GFP) rad ia l g l ia l ce l ls is s im i lar (Menuet et a l . , 2005 ; Pe l legr in i et a l . , 2007 ; Tong et a l . , 2009) . Our data support prev ious conc lus ions from stud ies focus ing on the te lencepha lon ( D io te l et a l . , 2010a , 2010b ; M€ arz et a l . , 2010) . We showed that rad ia l g l ia l ce l ls co-express both markers in the te lencepha lon , the d iencepha lon , and the mes- and metencepha lon . However , obv ious reg iona l d i f ferences were observed betweencyp19a1b-GFP and B lbp express ion notab ly in the tha lamus , the opt ic tectum and in ce l ls surround ing the latera l recess (F igs . 3 and 4) . These d i f ferences are not due to transgene leak ing ascyp19a1b-GFP transgen icfish l ine fu l ly recap itu lates endogenous AroB as prev ious ly reported (Tong et a l . , 2009and data not shown) , and was part ia l ly reported in zebrafish for AroB , B lbp and cxcr4 , a chemok ine recep tor (D iote l et a l . , 2010b) .

Such a rad ia l g l ia heterogene ity was prev ious ly descr ibed in mamma ls and is current ly not comp lete ly understood (P into and Gotz , 2007) . Th is heterogene ity cou ld stem from the d ist inct neurogen ic versus g l iogen ic propert ies , pro l i ferat ion rate ( long-

term and short-term se l f-renew ing) or a lso d ist inct regenerat ive propert ies (Barry et a l . , 2014 ; Tack et a l . , 2012) .

3 .3 . Pro l i fera t ion is pre feren t ia l ly assoc ia ted w i th cer ta in markers accord ing to the n iche

We a lso looked for a potent ia l d i f ference in AroB or B lbp

express ion in pro l i ferat ive rad ia l g l ia l ce l ls . Our resu lts c lear ly show

that B lbp express ion is pre ferent ia l ly and s ign ificant ly more asso-

c iated w ith rad ia l g l ia pro l i ferat ion in reg ions d isp lay ing strong

cyp19a1b-GFP and B lbp co-express ion ( i .e . : A , D l , Dm , Vd , Vv , Vs ,

PPv) . There was no pre ferent ia l express ion assoc iated w ith pro l i f-

erat ion in the preopt ic area and the Vp , wh i lecyp19a1b-GFP ap-

pears to be pre ferent ia l ly assoc iated w ith pro l i ferat ion o f rad ia l g l ia

in the dorsa l zone o f the per iventr icu lar hypo tha lamus where the

LR starts to open . S ign ificant ly more pro l i ferat ive rad ia l g l ia express

at leastcyp19a1b-GFP compared to B lbp (89 .9% Vs . 65 .5% ;Tab le 1

andF ig . 5G) , suggest ing potent ia l d i f ferences in the regu lat ion o f

Fig. 5.Quantification of Blbp andcyp19a1b-GFP (AroB) expression in proliferative radial glial cells in different neurogenic regions.A: Blbp (blue) and PCNA (red) staining on cyp19a1b-GFP (green) transgenic zebrafishlinein the dorsomedian telencephalon. Proliferative cells (PCNA) can be Blbp-positive (inset 1), GFP-positive (inset 2), Blbp/GFP-positive (inset 3) or Blbp/GFP-negative (inset 4). The double negative group has been previously proposed to correspond to further committed precursors suggested to be neuroblasts (M€arz et al., 2010).B: Columns 1, 2, 3 and 4 corresponds tothe respectiveinsetsin Afor each channel.In 1, asterisks point to a proliferative cells displaying Blbp staining but not GFP.In 2, asterisks highlight a proliferative cells displaying GFP staining but not Blbp.In 3, asterisks show a proliferative cell displaying both GFP and Blbp staining, and neither stainingin 4. C-G: In each region of interest, proliferative (PCNA) radial glial cells were checked for the expression of Blbp andcyp19a1b-GFP (AroB). Counting was performed on 6 adult female zebrafish (n¼6), on 325 cells for the Vv, 270 for the Vd, 205 for the A, 158 for the PPv, 185 for the Hc Div and 205 for the nucleus of the LR. The selected regions correspond to the zebrafish brain atlas (Wullimann et al., 1996) sectionlevel number 71for the Vv and Vd; 131 for the A; 168 for the PPv and the Hd LR. Statistical analysis was performed using one- way ANOVA. Error bars correspond to standard errors (n ¼ 6 zebrafish). * 0,01<p<0,05; ** 0,01<p<0,0001; ***p<0,0001.

the neurogen ic n iche act iv ity . A lthough one can argue that some rad ia l g l ia l ce l ls cou ld be B lbp andcyp19a1b-GFP doub le-negat ive , lead ing to over-est imat ion o f our measurements , the deta i led stud ies on rad ia l g l ia l ce l ls in zebrafish confirm the cons istent co- express ion o f markers (Bau l ieu and Schumacher , 2000 ; D iote l et a l . , 2010b ; L indsey et a l . , 2012 ; M€ arz et a l . , 2010) , and such support the conc lus ion that our quant ificat ions are not overest imated .

3 .4 . B lbp , DHA and es trogens : l inks w i th neurogenes is

Dur ing embryogenes is in rats , the pre ferent ia l assoc iat ion o f B lbp w ith pro l i ferat ive zone is o f part icu lar interest as B lbp is downstream o f the neurogen ic transcr ipt ion factor Pax6 and en- hances ce l l pro l i ferat ion (Ara i et a l . , 2005) . New reports a lso suggest an invo lvement o f B lbp in neurogenes is in the SVZ and in the dentate gyrus o f rodents (Marko et a l . , 2011 ; Matsumata et a l . , 2012 ; Schne l l et a l . , 2014 ; Watanabe et a l . , 2007) . Other data ob- ta ined in adu lt monkeys show an up-regu lat ion o f B lbp in the h ippocampus o f post- ischem ic bra ins compared to contro ls and suggest that B lbp is part o f the mo lecu lar mach inery regu lat ing the act iv ity o f the neurogen ic n iche by prov id ing fatty ac ids to pro- gen itor ce l ls (Boneva et a l . , 2011) . B lbp was a lso reported to be pos it ive ly invo lved in pro l i ferat ion and invas ion o f human g l io- b lastoma (Chen et a l . , 2007) . B lbp l igand , DHA , is an omega-3 po lyunsaturated fatty ac id essent ia l for proper bra in deve lopment and funct ion ing (Craw ford et a l . , 2013 ; Ho f fman et a l . , 2009 ; Janssen and K i l iaan , 2014 ) . M ice fed w ith DHA and other n-3 po lyunsaturated fatty ac ids show better exp loratory behav ior and spat ia l memory but d isp lay decreased neurogenes is in the h ippo- campus , wh i le in rats DHA does not a f fect h ippocampa l ce l l pro- l i ferat ion and pos it ive ly impacts neurona l product ion and /or surv iva l (Janssen et a l . , 2015 ; Tokuda et a l . , 2014 ) . These data strong ly suggest important ro les for B lbp and DHA in neurogenes is and behav ior that deserve further invest igat ion . Interest ing ly , severa l stud ies suggest a l ink between hormones , espec ia l ly es- trogens , and DHA leve ls in the bra in . For instance , DHA leve ls are h igher in women than in men , and among women , those tak ing ora l contracept ive p i l ls d isp lay h igher DHA concentrat ions (G i ltay et a l . , 2004) . S im i lar sex d i f ferences and hormona l influences were a lso observed in rats and corre lated w ith ovar ian hormones (McNamara et a l . , 2009) . Interest ing ly , the bra ins o f embryon ic and

adu lt zebrafish expresse lov l4athat codes for the fatty ac id e longase 4 , an enzyme invo lved in DHA synthes is (Monro ig et a l . , 2010) and suggest that DHA can be loca l ly synthes ized from precursors in the bra in o ffish . Th is observat ion needs to be l inked to the capac ity o f the zebrafish bra in to synthes ize estrogensde novo(D iote l et a l . , 2011a , 2011b) . In the bra in , estrogens exert var ious e f fects on stem ce l l and neurona l act iv it ies by modu lat ing pro l i ferat ion o f neura l progen itors (Brock et a l . , 2010 ; Ga lea , 2008 ; Mart inez- Cerde~ no et a l . , 2006) , neurona l m igrat ion and d i f ferent iat ion , bra in p last ic ity (Br inton , 2009 ; Fan et a l . , 2006 ; Murashov et a l . , 2004 ; Wang et a l . , 2003) , as we l l as ce l l surv iva l (Garc ia-Segura , 2008 ; H i l l et a l . , 2009 ; Wang et a l . , 2001) .Inadu lt zebrafish , 17 b - estrad io l was shown to regu late bra in pro l i ferat ion , m igrat ion and surv iva l (Couma i l leau et a l . , 2015 ; D io te l et a l . , 2013 ; Makantas i and Dermon , 2014 ; Pe l legr in i et a l . , 2015) . G iven that rad ia l g l ia l ce l ls appear to be targeted by estrogens (Couma i l leau et a l . , 2015 ; D iote l et a l . , 2011b ; Menuet et a l . , 2005 ; Pe l legr in i et a l . , 2015 , 2013) , AroB- pos it ive rad ia l g l ia l ce l ls cou ld a f fect the ir own act iv ity as we l l as that o f the ir ne ighbors by produc ing estrogens in an autocr ine and / or paracr ine manner . Consequent ly , it is l ike ly that estrogen , B lbp and its l igand cou ld modu late neurogenes is in a comb inatory manner .

Taken together these data suggest key ro les for B lbp , and its l igand DHA in stem ce l l act iv ity and neurogenes is . Further in- vest igat ions us ing B lbp v ivo-morpho l ino and over -express ion techn iques cou ld prov ide important ins ide on the ro les o f B lbp and its l igand in homeostat ic and reparat ive neurogenes is infish . 3 .5 . Conc lus ions

To conc lude , we prov ide a deta i led mapp ing o f B lbp in the who le bra in o f adu lt zebrafish , show ing B lbp co-express ion w ith the rad ia l g l ia marker AroB and rad ia l g l ia l ce l l heterogene ity . Interest ing ly , pro l i ferat ive rad ia l g l ia l ce l ls are pre ferent ia l ly assoc iated w ith B lbp express ion in most bra in reg ions , suggest ing key ro les for th is b ind ing pro te in and its l igand in stem ce l l act iv ity and bra in func- t ions . G iven (1) the cont inuous growth o f Te leostfish dur ing the ir l i fespan , (2) the ma intenance o f bra in embryon ic propert ies in adu ltfish (D io te l et a l . , 2010a) and (3) the importance o f DHA for bra in deve lopment , these resu lts ident i fy B lbp and DHA as poten- t ia l factors med iat ing some aspects o f bra in homeostas is and funct ions dur ing adu lthood in zebra fish .

Table 1

Expression and statistical analysis ofcypa19a1b-GFP and/or Blbp-positive proliferative radial glial cellsin different brain regions. Zebrafish

brain atlas level

Brain

regions% of proliferative radial glia expressing: ANOVA (p-value) % of proliferative radial glia expressing: Cyp19a1b-GFP

and Blbp Cyp19a1b-GFP

only Blbp only Cyp19a1b-GFP/Blbp

vs.Cyp19a1b-GFP Cyp19a1b-GFP/Blbp

vs. Blbp Cyp19a1b-GFP

vs. Blbp Cyp19a1b-GFP Blbp Student's t-test Cyp19a1b-GFP vs. Blbp (p-value) 71 Vv 66.6(±8.1) 1.2(±0.7) 32.2(±8.5) <0.0001 0.0038 0.0072 67.8 98.8 0.0066 71 Vd 61.2(±7.7) 0.7(±0.7) 38.1(±8.1) <0.0001 0.0272 0.0016 61.9 99.3 0.0018 92 Ppa 92.9(±4.5) 1.4(±1.2) 5.7(±4.4) <0.0001 <0.0001 NS 94.3 98.6 NS 98 PPa 81.8(±6.88) 5.3(±3.1) 12.94(±4.7) <0.0001 <0.0001 NS 87.1 94.7 NS 92 Vs 79.9(±4) 0(±0) 20.1(±4) <0.0001 <0.0001 0.0002 79.9 100.0 0.0003 98 Vp 85.4(±5.5) 5.80(±4.3) 8.8(±3.4) <0.0001 <0.0001 NS 91.2 94.2 NS 114 Dm 68.6(±3.5) 1.7(±1.9) 29.7(±3.7) <0.0001 <0.0001 <0.0001 70.4 98.3 <0.0001 114 Dl 55.6(±5.2) 0.92(±0.7) 43.5(±5) <0.0001 0.0392 <0.0001 56.5 99.1 <0.0001 121/125 PPp 54.9(±7.2) 14.3(±6.6) 30.8(±8) 0.0007 0.0225 NS 69.2 85.7 NS 131 A 24.9(±6.9) 0(±0) 75.1(±6.6) 0.0079 <0.0001 <0.0001 20.7 83.6 0.006 158 PPv 8.9(±3.2) 0.7(±0.8) 90.4(±3) 0.0262 <0.0001 <0.0001 9.7 99.3 <0.0001 168 Hd (LR) 55.4(±3.5) 34.5(±5.5) 10.1(±4.4) 0.0027 <0.0001 0.0008 89,9 65,5 0.0035

Thetwofirst columns providethe brain nuclei and regions studied accordingtothe sectionlevelsfromthe zebrafish brain atlas(Wullimann et al., 1996). Columns 3, 4 and 5 correspondtothe percentage of proliferative radial glial cells expressing eithercyp19a1b-GFP/Blbp,cyp19a1b-GFP only or Blbp only. Standard errors are providedin brackets (n¼6 zebrafish). Columns 6, 7 and 8 correspondto one-way ANOVA performed on column 3, 4 and 5(asinFig. 5). Columns 9 and 10 providethe % of proliferativeradial glial cells expressing atleastcyp19a1b-GFP(columns 3 and 4) or atleast Blbp(columns 3 and 5). Statistical analysis(Student'st-test) was performedin orderto knowif one marker was preferentially associated with proliferation of radial glial cells (p-value are provided in column 11). A p-value greater than 0.05 was considered as statistically non- significant (NS).

4 . Exper imenta l procedure 4 .1 . An ima ls and e th ics

Th is work was approved by the eth ics comm ittee CREEA (Com ite Renna is d 'Eth ique en mat iere d 'Exper imentat ion An ima le) . Zebrafish were housed , hand led , and sacr ificed in accordance w ith the European Un ion regu lat ions concern ing the pro tect ion o f exper imenta l an ima ls . Three to s ix month-o ld ma le and fema le sexua l ly mature w i ld-type (Dan io rer io, AB stra in) andcyp19a1b- GFP transgen ic zebrafish (Tong et a l . , 2009) were housed in the zebrafish fac i l ity o f the IFR 140 ( INRA SCR IBE , Rennes) and ma in- ta ined under standard cond it ions o f temperature (28 .5 C) and photoper iod (14-h l ight /10-h) . F ish were anesthet ized w ith Tr ica ine (MS-222 ; S igmaeA ldr ich ; REF : A5040) and sacr ificed us ing an overdose o f anesthet ic . Next , bra ins were part ia l ly d issected and fixed over n ight at 4 C in 4% para forma ldehyde (4% PFA) in sa l ine phosphate bu f fer (PBS , pH 7 .4) . The next day , bra ins were extracted andfixed at 4 C in 4% PFA-PBS , be fore para ffin embedd ing . 4 .2 . Pr imary an t ibod ies

A rabb it po lyc lona l ant ibody ra ised aga inst the ~15kD rat /mouse B lbp was used (1 :2000 ; Chem icon Internat iona l Sero log ica l Com- pany ; REF : AB9558 ; RR ID : AB_2314014) . The spec ific ity o f th is ant ibody was prev ious ly assessed in zebrafish by Western B lott ing (D iote l et a l . , 2010b) and was ident ified as a marker o f rad ia l g l ia l ce l ls .

A mouse monoc lona l ant ibody ra ised aga inst the PCNA (pro l i f- erat ive ce l l nuc lear ant igen) was used for detect ing pro l i ferat ive ce l ls (1 :100 ; C lone PC10 ; Dako , G lostrup , Denmark ; REF : M0879 ; RR ID : AB_2160651) . Th is ant ibody reacts w ith pro l i ferat ing ce l ls in vertebrate spec ies , inc lud ing zebrafish (M€ arz et a l . , 2010 ; Pe l legr in i et a l . , 2007) .

Post-m ito t ic neurons were ident ified w ith a mouse monoc lona l ant ibody ra ised aga inst the human nuc lear pro te in HuC /D (1 :20 ; C lone 16A11 ; Mo lecu lar Probes , Eugen , OR ; REF : A21271 ; RR ID : AB_221448) . The spec ific ity o f th is ant ibody was tested by the supp l ier , and the ant i-HuC /D was prev ious ly used in zebrafish (Pe l legr in i et a l . , 2007) .

4 .3 . Immunoh is tochem is try

For B lbp mapp ing , immunoh istochem istry was per formed on para ffin bra in sect ions (6 m m) prepared w ith a m icro tome (M icrom HM 355 S) . Sect ions were depara ffin ised in xy lene , rehydrated through graded ethano l (100e30%) , r insed tw ice in PBS (pH 7 .4) fo l lowed by ant igen retr ieva l in sod ium c itrate bu f fer (pH 6 ; 80C) for 30 m in . S l ides were r insed tw ice in 0 .2% Tr iton PBS (PBS-T) and non-spec ific b ind ing was b locked in PBS-T conta in ing 1% m i lk powder for 45 m in at room temperature . F ina l ly , sect ions were incubated over n ight at room temperature in a hum id i fied chamber w ith the appropr iate first ant ibod ies d i luted in PBS conta in ing 0 .5%

m i lk powder . The next day , the sect ions were washed severa l t imes in PBS-T and incubated w ith A lexa F luor

^®

goat ant i-rabb it 594 (1 :200 ; Inv itrogen Mo lecu lar probes , Eugene , OR , REF : A-11037 ; RR ID : AB_10561549) for 1h30 at room temperature in a dark and hum id ified chamber . A fter three washes in PBS-T , s l ides were mounted w ith the ant i fad ing med ium Vectash ie ld (Vector) con- ta in ing 4 ,6-d iam ino-2-pheny l indo le (DAP I) , to v isua l ize ce l l nuc le i . B lbp mapp ing was per formed on three ma le and three fema le adu lt zebrafish .

For B lbp sta in ing on cyp19a1b-GFP zebrafish l ine , ant igen retr ieva l was per formed in a sod ium c itrate bu f fer (pH 7 .1 ; 64 C) for 1h15 in order to preserve GFPfluorescence .

For doub le B lbp and PCNA immunoh istochem istry oncyp19a1b- GFP zebrafish l ine , ant igen retr ieva l was per formed in a sod ium c itra te bu f fer (pH 7 .1 ; 64 C) for 1h15 and B lbp ant ibody was detected us ing a b iot iny lated goat ant i-rabb it IgG (1 /1500 ; Vector 's lab , REF : BA-1000 ; RR ID : AB_2313606) and revea led by an AMCA- strep tav id in comp lex (1 /500 ; Vector 's lab ; REF : SA-5008 ; RR I- D :AB_2336103) . PCNA ant ibody was revea led w ith A lexa fluor

^®

goat ant i-mouse 594 (1 :200 ; Inv itrogen Mo lecu lar probes , Eugene , OR , REF : A-11005 ; RR ID : AB_10561507) .

4 .4 . M icroscopy

Observat ions were per formed on an ep ifluorescence m icroscope O lympus Prov is equ ipped w ith a DP71 d ig ita l camera or w ith a Ze iss Apotome . Images were saved in“T IFF”format w ith Ana lys is so ftware to a l low image stack ing .

4 .5 . S ta t is t ica l ana lys is o f cyp19a1b-GFP (AroB) and b lbp express ion in pro l i fera t ive rad ia l g l ia l ce l ls

For each bra in nuc leus or reg ion , ident ified accord ing the Zebrafish bra in At las from Wu l l imann (Wu l l imann et a l . , 1996) , count ing was per formed on three success ive sect ions o f 6 adu lt fema le zebra fish . Count ing was per formed by two d i f ferent ex- per imenters who obta ined s im i lar data that were averaged . S ta- t ist ica l ana lys is was per formed us ing ANOVA or S tudent 's t-test . A p-va lue be low 0 .05 was cons idered stat ist ica l ly s ign ificant . 4 .6 . Nomenc la ture

The nomenc lature for bra in nuc le i and reg ions is accord ing to that estab l ished in the zebrafish at las by Wu l l imann and co l leagues (Wu l l imann et a l . , 1996) .

Abbrev iat ions o f anatom ica l terms used in th is study accord ing to Wu l l imann at las [84] .

A , anter ior tha lam ic nuc leus ; PN , accessory pretecta l nuc leus ; ATN , anter ior tubera l nuc leus ; CCe , corpus cerebe l l i ; Chab , habenu lar comm issure ; Chor , hor izonta l comm issure ; CM , corpus mam i l lare ;

CP , centra l poster ior tha lam ic nuc leus ; CPN , centra l pretecta l nuc leus ; Cpop , postopt ic comm issure ; Cpost , poster ior comm issure ; D , dorsa l te lencepha l ic area ;

Dc , centra l zone o f dorsa l te lencepha l ic area ; Dd , dorsa l zone o f dorsa l te lencepha l ic area ; D iV , d iencepha l ic ventr ic le ;

D l , latera l zone o f dorsa l te lencepha l ic area ; Dm , med ia l zone o f dorsa l te lencepha l ic area ; DOT , dorsomed ia l opt ic tract ;

Dp , poster ior zone o f dorsa l te lencepha l ic area ; DP , dorsa l poster ior tha lam ic nuc leus ;

DTN , dorsa l tegmenta l nuc leus ;

ECL , externa l ce l lu lar layer o f o l factory bu lb ; EG , em inent ia granu lar is ;

ENv , entopendoncu lar nuc leus , ventra l part ; FR , fasc icu lus retroflexus ;

GL , g lomeru lar layer o f o l factory bu lb ; Had , dorsa l habenu lar nuc leus ; Hav , ventra l habenu lar nuc leus ;

Hc , cauda l zone o f per iventr icu lar hypo tha lamus ;

Hd , dorsa l zone o f per iventr icu lar hypotha lamus ; Hv , ventra l zone o f per iventr icu lar hypo tha lamus ; ICL , interna l ce l lu lar layer o f o l factory bu lb ; IL , in fer ior lobe ;

LH , latera l hypo tha lam ic nuc leus ; LLF : latera l long itud ina l fasc ic le ;

LR , latera l recess o f d iencepha l ic nuc leus ; MLF , med ia l long itud ina l fasc ic le ;

N I I I , ocu lomo tor nuc leus ;

NMLF , nuc leus o f med ia l long itud ina l fasc ic le ; NLV , nuc leus latera l is va lvu lae ;

PG , preg lomeru lar nuc leus ;

PGa , anter ior preg lomeru lar nuc leus ; PG l , latera l preg lomeru lar nuc leus ; P it , p itu itary ;

PO , poster ior pretecta l nuc leus ;

PPa , parvoce l lu lar preop t ic nuc leus , anter ior part ; PPp , parvoce l lu lar preopt ic nuc leus , poster ior part ; PR , poster ior recess o f d iencepha l ic ventr ic le ; PSp , parvoce l lu lar superfic ia l pretecta l nuc leus ; PTN , poster ior tubera l nuc leus ;

PPv , per iventr icu lar pretecta l nuc leus ; R , rostro latera l nuc leus ;

RF , ret icu lar format ion ; RV : rhombencepha l ic ventr ic le ; SC , suprach iasmat ic nuc leus ; SO , secondary octava l popu lat ion ; TeO , tectum opt icum ;

TL , torus long itud ina l is ; TLa , torus latera l is ;

TPp , per iventr icu lar nuc leus o f poster ior tubercu lum ; TS , torus sem ic ircu lar is ;

V , ventra l te lencepha l ic area ; V3 , th ird ventr ic le ;

V I I , sensory roo t o f the fac ia l nerve ; V I I I , octava l nerve ;

VCe , va lvu la cerebe l l i ;

Vd , dorsa l nuc leus o f ventra l te lencepha l ic area ; VL , ventro latera l tha lam ic nuc leus ;

VM , ventromed ia l tha lam ic nuc leus ; VOT , ventro latera l opt ic tract ;

Vp , postcomm issura l nuc leus o f ventra l te lencepha l ic area ; Vv , ventra l nuc leus o f dorsa l te lencepha l ic area ;

ZL , zona l im itans . Conflicts of interest

None o f the authors have any compet ing interests . Authors ' contr ibut ions

OK , CV and EP des igned the exper iments and superv ised the work . ND and CV conducted the exper iments . ND , CV , OK and EP ana lyzed the data and wro te the manuscr ip t . A l l authors read and approved thefina l manuscr ip t .

Acknow ledgments

The Post-Grene l le grant NEMO (to Franço is Br ion and O l iv ier Kah) and a NSC Ta iwan-CNRS France exchange program (to Bon- Chu Chung and O l iv ier Kah) supported th is research . The outstand ing ass istance o f the sta f f o f the B IOS IT zebrafish fac i l ity ( INRA , Laborato ire de Phys io log ie et Genom ique des Po issons) was grea t ly apprec iated . We thank Dr . Mar ie-L ise Th ieu lant for her adv ice dur ing th is work , Sara Powers and Pr . Th ierry Char l ier for

proo fread ing the manuscr ip t . References

Adolf, B., Chapouton, P., Lam, C.S., Topp, S., Tannhauser, B., Str€ahle, U., G€otz, M., Bally-Cuif, L., 2006. Conserved and acquired features of adult neurogenesis in the zebrafish telencephalon. Dev. Biol. 295, 278e293.

Anthony, T.E., Mason, H.A., Gridley, T., Fishell, G., Heintz, N., 2005. Brain lipid- binding protein is a direct target of Notch signaling in radial glial cells. Genes Dev. 19, 1028e1033.

Arai, Y., Funatsu, N., Numayama-Tsuruta, K., Nomura, T., Nakamura, S., Osumi, N., 2005. Role of Fabp7, a downstream gene of Pax6, in the maintenance of neu- roepithelial cells during early embryonic development of the rat cortex. J. Neurosci. Off. J. Soc. Neurosci. 25, 9752e9761.

Barry, D.S., Pakan, J.M., McDermott, K.W., 2014. Radial glial cells: key organisers in CNS development. Int. J. Biochem. Cell Biol. 46, 76e79.

Baulieu, E., Schumacher, M., 2000. Progesterone as a neuroactive neurosteroid, with special reference to the effect of progesterone on myelination. Steroids 65, 605e612.

Bentivoglio, M., Mazzarello, P., 1999. The history of radial glia. Brain Res. Bull. 49, 305e315.

Boneva, N.B., Kaplamadzhiev, D.B., Sahara, S., Kikuchi, H., Pyko, I.V., Kikuchi, M., Tonchev, A.B., Yamashima, T., 2011. Expression of fatty acid-binding proteinsin adult hippocampal neurogenic niche of postischemic monkeys. Hippocampus 21, 162e171.

Brinton, R.D., 2009. Estrogen-induced plasticity from cells to circuits: predictions for cognitive function. Trends Pharmacol. Sci. 30, 212e222.

Brock, O., Keller, M., Veyrac, A., Douhard, Q., Bakker, J., 2010. Short term treatment with estradiol decreases the rate of newly generated cellsin the subventricular zone and main olfactory bulb of adultfemale mice. Neuroscience 166, 368e376.

Chen, C.J., Kono, H., Golenbock, D., Reed, G., Akira, S., Rock, K.L., 2007.Identification of a key pathway required for the sterile inflammatory response triggered by dying cells. Nat. Med. 13, 851e856.

Cole, G.M., Ma, Q.L., Frautschy, S.A., 2010. Dietary fatty acids and the aging brain. Nutr. Rev. 68 (Suppl. 2), S102eS111.

Coumailleau, P., Pellegrini, E., Adrio, F., Diotel, N., Cano-Nicolau, J., Nasri, A., Vaillant, C., Kah, O., 2015. Aromatase, estrogen receptors and brain develop- ment in fish and amphibians. Biochim. Biophys. Acta 1849, 152e162.

Crawford, M.A., Broadhurst, C.L., Guest, M., Nagar, A., Wang, Y., Ghebremeskel, K., Schmidt, W.F., 2013. A quantum theory for the irreplaceable role of docosa- hexaenoic acid in neural cell signalling throughout evolution. Prostagl. Leukot. Essent. Fat. Acids 88, 5e13.

D'Amico, L.A., Boujard, D., Coumailleau, P., 2011. Proliferation, migration and dif- ferentiationinjuvenile and adult Xenopuslaevis brains. Brain Res. 1405, 31e48.

Dinarello, C.A., 2012. Keep up the heat on IL-1. Blood 120, 2538e2539.

Diotel, N., Do Rego, J.L., Anglade, I., Vaillant, C., Pellegrini, E., Gueguen, M.M., Mironov, S., Vaudry, H., Kah, O., 2011a. Activity and expression of steroidogenic enzymes in the brain of adult zebrafish. Eur. J. Neurosci. 34, 45e56.

Diotel,N., Do Rego, J.L., Anglade, I., Vaillant, C., Pellegrini, E., Vaudry, H., Kah, O., 2011b. The brain of teleostfish, a source, and a target of sexual steroids. Front. Neurosci. 5, 137.

Diotel, N., Le Page, Y., Mouriec, K., Tong, S.K., Pellegrini, E., Vaillant, C., Anglade, I., Brion, F., Pakdel, F., Chung, B.C., Kah, O., 2010a. Aromataseinthe brain of teleost fish: expression, regulation and putative functions. Front. Neuroendocrinol. 31, 172e192.

Diotel, N., Rodriguez Viales, R., Armant, O., Marz, M., Ferg, M., Rastegar, S., Strahle, U., 2015. Comprehensive expression map of transcription regulators in the adult zebrafish telencephalon reveals distinct neurogenic niches. J. Comp. Neurol. 523, 1202e1221.

Diotel, N., Vaillant, C., Gabbero, C., Mironov, S., Fostier, A., Gueguen, M.M., Anglade,I., Kah, O., Pellegrini, E., 2013. Effects of estradiolin adult neurogenesis and brain repair in zebrafish. Horm. Behav. 63, 193e207.

Diotel, N., Vaillant, C., Gueguen, M.M., Mironov, S., Anglade, I., Servili, A., Pellegrini, E., Kah, O., 2010b. Cxcr4 and Cxcl12 expressionin radial glial cells of the brain of adult zebrafish. J. Comp. Neurol. 518, 4855e4876.

Edelmann, K., Glashauser, L., Sprungala, S., Hesl, B., Fritschle, M., Ninkovic, J., Godinho, L., Chapouton, P., 2013. Increased radial glia quiescence, decreased reactivation uponinjury and unaltered neuroblast behavior underlie decreased neurogenesis in the aging zebrafish telencephalon. J. Comp. Neurol. 521, 3099e3115.

Fan, X., Warner, M., Gustafsson,J.A., 2006. Estrogen receptor beta expressionin the embryonic brain regulates development of calretinin-immunoreactive GABAergic interneurons. Proc. Natl. Acad. Sci. U. S. A. 103, 19338e19343.

Feng, L., Hatten, M.E., Heintz, N., 1994. Brain lipid-binding protein (BLBP): a novel signaling system in the developing mammalian CNS. Neuron 12, 895e908.

Feng, L., Heintz, N., 1995. Differentiating neurons activate transcription of the brain lipid-binding protein gene in radial glia through a novel regulatory element. Development 121, 1719e1730.

Forlano, P.M., Deitcher, D.L., Myers, D.A., Bass, A.H., 2001. Anatomical distribution and cellular basis for high levels of aromatase activity in the brain of teleost fish: aromatase enzyme and mRNA expression identify glia as source. J. Neurosci. Off. J. Soc. Neurosci. 21, 8943e8955.

Galea, L.A., 2008. Gonadal hormone modulation of neurogenesis in the dentate gyrus of adult male and female rodents. Brain Res. Rev. 57, 332e341.

Galea, L.A., Spritzer, M.D., Barker, J.M., Pawluski, J.L., 2006. Gonadal hormone modulation of hippocampal neurogenesis in the adult. Hippocampus 16, 225e232.

Galea, L.A., Uban, K.A., Epp, J.R., Brummelte, S., Barha, C.K., Wilson, W.L., Lieblich, S.E., Pawluski, J.L., 2008. Endocrine regulation of cognition and neu- roplasticity: our pursuit to unveil the complex interaction between hormones, the brain, and behaviour. Can. J. Exp. Psychol. 62, 247e260.

Garcia-Segura, L.M., 2008. Aromatase in the brain: not just for reproduction anymore. J. Neuroendocrinol. 20, 705e712.

Garcia-Segura, L.M., Wozniak, A., Azcoitia, I., Rodriguez, J.R., Hutchison, R.E., Hutchison, J.B., 1999. Aromatase expression by astrocytes after brain injury: implications for local estrogen formation in brain repair. Neuroscience 89, 567e578.

Giltay, E.J., Gooren, L.J., Toorians, A.W., Katan, M.B., Zock, P.L., 2004. Docosahexae- noic acid concentrations are higher in women than in men because of estro- genic effects. Am. J. Clin. Nutr. 80, 1167e1174.

Glasauer, S.M., Neuhauss, S.C., 2014. Whole-genome duplication in teleost fishes and its evolutionary consequences. Mol. Genet. Genom. MGG 289, 1045e1060.

Goncalves, D., Teles, M., Alpedrinha, J., Oliveira, R.F., 2008. Brain and gonadal aro- matase activity and steroid hormonelevelsinfemale and polymorphic males of the peacock blenny Salaria pavo. Horm. Behav. 54, 717e725.

Gronert, K., 2008. Lipid autacoidsininflammation andinjury responses: a matter of privilege. Mol. Interv. 8, 28e35.

Hartfuss, E., Galli, R., Heins, N., Gotz, M., 2001. Characterization of CNS precursor subtypes and radial glia. Dev. Biol. 229, 15e30.

Hebsgaard,J.B., Nelander,J., Sabelstrom, H.,Jonsson, M.E., Stott, S., Parmar, M., 2009. Dopamine neuron precursors within the developing human mesencephalon show radial glial characteristics. Glia 57, 1648e1658.

Hill, R.A., Chua, H.K., Jones, M.E., Simpson, E.R., Boon, W.C., 2009. Estrogen defi- ciency results in apoptosis in the frontal cortex of adult female aromatase knockout mice. Mol. Cell. Neurosci. 41, 1e7.

Hoffman, D.R., Boettcher,J.A., Diersen-Schade, D.A., 2009. Toward optimizing vision and cognitionin terminfants by dietary docosahexaenoic and arachidonic acid supplementation: a review of randomized controlled trials. Prostagl. Leukot. Essent. Fat. Acids 81, 151e158.

Hsieh, J., 2012. Orchestrating transcriptional control of adult neurogenesis. Genes Dev. 26, 1010e1021.

Ito, Y., Tanaka, H., Okamoto, H., Ohshima, T., 2010. Characterization of neural stem cells and their progeny in the adult zebrafish optic tectum. Dev. Biol. 342, 26e38.

Janssen, C.I., Kiliaan, A.J., 2014. Long-chain polyunsaturated fatty acids (LCPUFA) from genesis to senescence: the influence of LCPUFA on neural development, aging, and neurodegeneration. Prog. Lipid Res. 53, 1e17.

Janssen, C.I., Zerbi, V., Mutsaers, M.P., deJong, B.S., Wiesmann, M., Arnoldussen,I.A., Geenen, B., Heerschap, A., Muskiet, F.A., Jouni, Z.E., van Tol, E.A., Gross, G., Homberg, J.R., Berg, B.M., Kiliaan, A.J., 2015. Impact of dietary n-3 poly- unsaturated fatty acids on cognition, motor skills and hippocampal neuro- genesis in developing C57BL/6J mice. J. Nutr. Biochem. 26, 24e35.

Kah,O.,Pellegrini, E., Mouriec, K., Diotel, N., Anglade,I., Vaillant, C., Thieulant, M.L., Tong, S.K., Brion, F., Chung, B.C., Pakdel, F., 2009. Oestrogens and neurogenesis: newfunctions for an old hormone. Lessons from the zebrafish.J. Soc. Biol. 203, 29e38.

Kaslin, J., Ganz, J., Geffarth, M., Grandel, H., Hans, S., Brand, M., 2009. Stem cells in the adult zebrafish cerebellum:initiation and maintenance of a novel stem cell niche. J. Neurosci. Off. J. Soc. Neurosci. 29, 6142e6153.

Krelin, Y., Voronov, E., Dotan, S., Elkabets, M., Reich, E., Fogel, M., Huszar, M., Iwakura, Y., Segal, S., Dinarello, C.A., Apte, R.N., 2007. Interleukin-1beta-driven inflammation promotes the development and invasiveness of chemical carcinogen-induced tumors. Cancer Res. 67, 1062e1071.

Kurtz, A., Zimmer, A., Schnutgen, F., Bruning, G., Spener, F., Muller, T., 1994. The expression pattern of a novel gene encoding brain-fatty acid binding protein correlates with neuronal and glial cell development. Development 120, 2637e2649.

Lassiter, C.S., Linney, E., 2007. Embryonic expression and steroid regulation of brain aromatase cyp19a1b in zebrafish (Danio rerio). Zebrafish 4, 49e57.

Li, H., Jin, G., Qin, J., Yang, W., Tian, M., Tan, X., Zhang, X., Shi, J., Zou, L., 2011. Identification of neonatal rat hippocampal radial glia cells in vitro. Neurosci. Lett. 490, 209e214.

Lindsey, B.W., Darabie, A., Tropepe, V., 2012. The cellular composition of neurogenic periventricular zones in the adult zebrafish forebrain. J. Comp. Neurol. 520, 2275e2316.

Liu, R.Z., Denovan-Wright, E.M., Degrave, A., Thisse, C., Thisse, B., Wright,J.M., 2004. Differential expression of duplicated genes for brain-type fatty acid-binding proteins (fabp7a and fabp7b) during early development of the CNS in zebra- fish (Danio rerio). Gene Expr. Patterns 4, 379e387.

Liu, R.Z., Denovan-Wright, E.M., Wright,J.M., 2003. Structure, mRNA expression and linkage mapping ofthe brain-typefatty acid-binding protein gene(FABP7)from zebrafish (Danio rerio). Eur. J. Biochem. FEBS 270, 715e725.

Makantasi, P., Dermon, C.R., 2014. Estradioltreatment decreases cell proliferationin the neurogenic zones of adult female zebrafish (Danio rerio) brain. Neurosci- ence 277, 306e320.

Marko, K., Kohidi, T., Hadinger, N., Jelitai, M., Mezo, G., Madarasz, E., 2011. Isolation of radial glia-like neural stem cells from fetal and adult mouse forebrain via selective adhesion to a novel adhesive peptide-conjugate. PLoS One 6, e28538.

Martinez-Cerde~no, V., Noctor, S.C., Kriegstein, A.R., 2006. Estradiol stimulates

progenitor cell division in the ventricular and subventricular zones of the embryonic neocortex. Eur. J. Neurosci. 24, 3475e3488.

M€arz, M., Chapouton, P., Diotel, N., Vaillant, C., Hesl, B., Takamiya, M., Lam, C.S., Kah, O., Bally-Cuif, L., Str€ahle, U., 2010. Heterogeneity in progenitor cell sub- types in the ventricular zone of the zebrafishadulttelencephalon. Glia 58, 870e888.

Matsumata, M., Sakayori, N., Maekawa, M., Owada, Y., Yoshikawa, T., Osumi, N., 2012. The effects of Fabp7 and Fabp5 on postnatal hippocampal neurogenesisin the mouse. Stem Cells 30, 1532e1543.

McNamara, R.K., Able, J., Jandacek, R., Rider, T., Tso, P., 2009. Gender differences in rat erythrocyte and brain docosahexaenoic acid composition: role of ovarian hormones and dietary omega-3 fatty acid composition. Psychoneur- oendocrinology 34, 532e539.

Menuet, A., Pellegrini, E., Brion, F., Gueguen, M.M., Anglade, I., Pakdel, F., Kah, O., 2005. Expression and estrogen-dependent regulation of the zebrafish brain aromatase gene. J. Comp. Neurol. 485, 304e320.

Monroig, O., Rotllant, J., Cerda-Reverter, J.M., Dick, J.R., Figueras, A., Tocher, D.R., 2010. Expression and role of Elovl4 elongasesin biosynthesis of verylong-chain fatty acids during zebrafish Danio rerio early embryonic development. Biochim. Biophys. Acta 1801, 1145e1154.

Mouriec, K., Lareyre, J.J., Tong, S.K., Le Page, Y., Vaillant, C., Pellegrini, E., Pakdel, F., Chung, B.C., Kah, O., Anglade, I., 2009 Oct. Early regulation of brain aromatase (cyp19a1b) by estrogen receptors during zebrafish development. Dev. Dyn. 238 (10), 2641e2651.http://dx.doi.org/10.1002/dvdy.22069. PMID: 19718764. Mouriec, K., Pellegrini, E., Anglade,I., Menuet, A., Adrio, F., Thieulant, M.L., Pakdel, F.,

Kah, O., 2008. Synthesis of estrogens in progenitor cells of adult fish brain: evolutive novelty or exaggeration of a more general mechanism implicating estrogens in neurogenesis? Brain Res. Bull. 75, 274e280.

Murashov, A.K., Pak, E.S., Hendricks, W.A., Tatko, L.M., 2004. 17beta-Estradiol en- hances neuronal differentiation of mouse embryonic stem cells. FEBS Lett. 569, 165e168.

Murrell, W., Feron, F., Wetzig, A., Cameron, N., Splatt, K., Bellette, B., Bianco, J., Perry, C., Lee, G., Mackay-Sim, A., 2005. Multipotent stem cells from adult ol- factory mucosa. Dev. Dyn. 233, 496e515.

Owada, Y., Abdelwahab, S.A., Kitanaka, N., Sakagami, H., Takano, H., Sugitani, Y., Sugawara, M., Kawashima, H., Kiso, Y., Mobarakeh, J.I., Yanai, K., Kaneko, K., Sasaki, H., Kato, H., Saino-Saito, S., Matsumoto, N., Akaike, N., Noda, T., Kondo, H., 2006. Altered emotional behavioral responsesin micelacking brain- type fatty acid-binding protein gene. Eur. J. Neurosci. 24, 175e187.

Pasmanik, M., Callard, G.V., 1985. Aromatase and 5 alpha-reductase in the teleost brain, spinal cord, and pituitary gland. Gen. Comp. Endocrinol. 60, 244e251.

Pellegrini, E., Coumailleau, P., Kah, O., Diotel, N., 2015. Aromatase and estrogens: involvement in constitutive and regenerative neurogenesis in adult zebrafish. In: Duncan Kelli, A. (Ed.), Estrogen Effects on Traumatic Brain Injury - Mecha- nisms of Neuroprotection and Repair.

Pellegrini, E., Menuet, A., Lethimonier, C., Adrio, F., Gueguen, M.M., Tascon, C., Anglade, I., Pakdel, F., Kah, O., 2005. Relationships between aromatase and estrogen receptors in the brain of teleostfish. Gen. Comp. Endocrinol. 142, 60e66.

Pellegrini, E., Mouriec, K., Anglade, I., Menuet, A., Le Page, Y., Gueguen, M.M., Marmignon, M.H., Brion, F., Pakdel, F., Kah, O., 2007.Identification of aromatase- positive radial glial cells as progenitor cells in the ventricular layer of the forebrain in zebrafish. J. Comp. Neurol. 501, 150e167.

Pellegrini,E., Vaillant, C., Diotel, N., Benquet, P., Brion, F., Kah, O., 2013. Expression, regulation and potential functions of aromatase in radial glial cells of thefish brain. In: Balthazart, J. Ball, Gregory, F. (Eds.), Brain Aromatase, Estrogens, and Behavior. Oxford, University Press, Oxford; New York.

Pinto, L., Gotz, M., 2007. Radial glial cell heterogeneityethe source of diverse progeny in the CNS. Prog. Neurobiol. 83, 2e23.

Podgornyi, O.V., Aleksandrova, M.A., 2009. BLBP-immunoreactive cells in the pri- mary culture of neural precursors from embryonic mouse brain. Bull. Exp. Biol. Med. 147, 125e131.

Rakic, P., 1978. Neuronal migration and contact guidance in the primate telen- cephalon. Postgrad. Med. J. 54 (Suppl. 1), 25e40.

Rodriguez Viales, R., Diotel, N., Ferg, M., Armant, O., Eich, J., Alunni, A., M€arz, M., Bally-Cuif, L., Rastegar, S., Str€ahle, U., 2015. The helix-loop-helix protein id1 controls stem cell proliferation during regenerative neurogenesis in the adult zebrafish telencephalon. Stem Cells 33, 892e903.

Rousselot, P., Heintz, N., Nottebohm, F., 1997. Expression of brain lipid binding protein in the brain of the adult canary and its implications for adult neuro- genesis. J. Comp. Neurol. 385, 415e426.

Schnell, A., Chappuis, S., Schmutz, I., Brai, E., Ripperger, J.A., Schaad, O., Welzl, H., Descombes, P., Alberi, L., Albrecht, U., 2014. The nuclear receptor REV-ERBalpha regulates Fabp7 and modulates adult hippocampal neurogenesis. PloS One 9, e99883.

Simpson, E.R., Mahendroo, M.S., Means, G.D., Kilgore, M.W., Hinshelwood, M.M., Graham-Lorence, S., Amarneh, B., Ito, Y., Fisher, C.R., Michael, M.D., et al., 1994. Aromatase cytochrome P450,the enzymeresponsiblefor estrogen biosynthesis. Endocr. Rev. 15, 342e355.

Steiner, B., Klempin, F., Wang, L., Kott, M., Kettenmann, H., Kempermann, G., 2006. Type-2 cells as link between glial and neuronal lineage in adult hippocampal neurogenesis. Glia 54, 805e814.

Steinke, D., Hoegg, S., Brinkmann, H., Meyer, A., 2006. Three rounds (1R/2R/3R) of genome duplications andthe evolution ofthe glycolytic pathwayin vertebrates. BMC Biol. 4, 16.