Endothelin 3 and T-type Ca 2+ channels drive enteric neural crest cell calcium activity, contractility and migration

Enteric neural cre st cells (ENCCs) colonize the gut during embryogenesis and migration defects give rise to Hirschsprung disea se (HD). Muta tions in GDNF/RET and EDN3/EDNRB are known to be causal in HD. Here, we show that migrating ENCCs in mice exhibit endogenous EDN3/EDNRB-ga ted calcium activit y, mediated by chloride channels, T-type Ca 2+ channel s and inositol trisphosphate-sensitive intracellular-store relea se. We find that inhibiting Ca 2+ activity results in ENCC migration defects, while exciting it promo tes migration by increa sing ENCC contractility and traction force to the extracellular matrix. Our study demonstrate s that embryonic endothelin-mediated neural crest migration and a dult endothelin-mediated vasoconstric tion is one and the same phenomenon, taking p l a c e i n d i f f e r e n t c e l l t y p e s . O u r r e s u l t s s u g g e s t a f u n c t i o n a l l i n k b e t w e e n r a r e mu t a t i o n s o f CACNA1H (the gene encoding CaV3.2) and HD, and pave the way for understanding neurocristopa thies in terms of neural crest cell bioelectric activity deficits. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Main text Neural crest cell s (NCC) are a popul ation of mul tipotent, highly migratory cells common to all vertebrate s. They colonize developing organs during embryonic development to give rise to a variety of struc ture s including cranio-facial bone, chromaffin cells of the adrenal medulla, Schwann cells of the peripheral nervous sy stem and the neurons and glia of the enteric nervous system (ENS) 1 . NCCs have attracted much scientific attention because of their paramount contributions to vertebrate physiological and pathological 2,3 development, to dome stication-induced phenotypes 4,5 a n d b e c a u s e aberrantly expressed NCC developmental programs can lead to cancer tumor invasiveness 6,7 . Vagal- derived 8 enteric neural crest cells (ENCCs ) invade the gut rostr o-caudally between 9.5 and 13.5 days of development in mice to form the ENS. ENCC migration defects result in colonic aganglionosis a t birth, a syndrome known as Hirschsprung di sea se (HD). HD is one of the most frequent neurocristopa thies, affecting ~1:5000 births 9 . The mutations hitherto identified, mostly in the GDNF/RET 10–12 a n d E D N 3 / E D N R B 13,14 signaling ca scades, are of incomplete penetrance and account for only 50 % of the familial and 15-20 % of the spo radic cases 15 : the cause of mos t HD cases is today not understood. Bioelectric activity 16–18 , as measured by Ca 2+ imaging or membrane potential changes, is increasingly recognized as a key player of stem cell 16–18 and NCC 19–22 b e h a v i o r . Spontaneous propagating Ca 2+ wave s in ENCCs 23 were found to depend on purinergic signaling, but their implication for migration and HD remained uncertain. Here, we show that most of the Ca 2+ activity in ENCCs occurs as unsynchronized single-cell events and not a s waves, that ENCC Ca 2+ activity is driven by EDN3/EDNRB, via the openi ng of Cl - and T-type Ca 2+ channels, and that Ca 2+ activity is intimately correlated with ENCC migration potential through a simple mechanism: elevated int racellular Ca 2+ leads to incre ased ENCC contractility and traction force to the extracellular matrix that allows them to migrate down the gut. This mechanism renew s our understanding of how edn3/EDNRB affects ENCC migration. Disruption of any components of this mechanism can result in a NCC migration defec t, opening-up new perspec tives for neurocristopathy and collective cell migration research. Enteric neural crest cells display spontaneous Ca 2+ activity during gut invasion We monitored Ca 2+ a c t i v i t y ( C A ) o n e x - v i v o mo u s e e mb r y o n i c g u t s e x p r e s s i n g t h e i n t r a c e l l u l a r C a 2+ reporter G CaM P6f specifically in NCCs (Fig.1a, Video S1), and meas ured its spatial dis tribution and characte ristics with an automated analysis pipeline (Fig.S1). CA in migrating ENCCs was spontaneou s and occurred mostly a s asynchronous, non-propagating, single-cell events. Ca 2+ trans ients could sometimes propagate radially to neighboring cells, but such wave-like events 23 oc c ur r e d , at E1 1 . 5 , at an act ivity of CA wave = 7.5x10 -4 ± 2x10 -4 events/min/100 µm² (± SEM , n =15), i.e. ~100-1000 times les s frequently than singl e-cell events (Fig.1b). Endogeno us CA at the ENCC migration front decreased from E10.5 to E12.5 (Fig.1b, Fig.S2, Video S1). At E11.5, it was significantly higher at the ENCC wavefront located at the il eo-cecal junction (ilcc) than in trailing ENCCs (Fig.1c ), concentrating at the cecum and antimesenteric ileum border (Fig.1a, Fig.S3). CA was also present in transmesenteric ENCCs (Fig.S3). CA was insensitive to tetrodotoxin (Fig.1d, Fig.S4), and the refore did not stem from neural activity; the latter could however be elicited by veratridine in proximal gut segments (Fig.1e, Fig.S4). Endogenous CA co-localized with the ENCC specific marker Sox10, but only very partially with the neuronal marker Tuj1 (Video S2). These results indicate that spontaneous CA originated in ENCC s and was most intense during early gut colonization stages, decreasing a s ENCCs gradually differentiated to neurons and glia. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 1. Endoge nous C a 2+ ac t iv i t y in e n t e r i c n e u r a l c r e st ce ll s d ur i n g g ut c o l o n i z at i o n . (a ) Gre e n : a v erag e ti m e -pro jec ti on of G C a M P t i m e - la ps e at t h e E 1 1. 5 i l e o - c e c al ju n c t io n ( Vi de o S 1 ) ; E N C C s ha v e r e ac he d t h e c e c um an d a r e ab ou t t o e nt e r t he col o n (white arro w). R ed: C A fre q uency h e a tm a p sho w s con c entra ti on of act ivity a t t h e cecu m an d the i leum a nti- m e s e n t e r i c b o r d e r ( s e e a l s o F i g . S 3 ) . ( b ) C A a t t h e E N C C m i g r a t i o n f r o n t a t s t a g e s E 1 0 . 5 ( j e j u n u m , n = 1 1 ) , E 1 1 . 5 ( i l e o - c e c a l ju nct io n , n=2 3 ), E1 2 . 5 (co lo n , n=1 5 ), K ruskall Wall is test. (c) CA in the d uodenu m ( duo d), jeju num (j e j ) an d i leo-cec a l j un c t i o n ( i l c c ) a t E 1 1 . 5 . L i n e s l i n k p o i n t s m e a s u r e d f r o m t h e s a m e s a m p l e ( n = 1 1 ) , W i l c o x o n m a t c h e d p a i r s s i g n e d r a n k ( W M P ) t e s t . (d ) Tetrod oto x i n (TTX, 1 µM ) di d not mod ify CA i n the du od e n um (n = 8 ) or i lcc (n=7) a t E11.5. ( e ) Vera trid i ne (V era, 10 µM) in duc ed CA in the d uodenu m (n=8), bu t n o t in t h e i lcc (n = 7) a t E 1 1 .5, W MP tes t. Endogenous Ca 2+ activity is induced by EDN3/EDNRB We nex t i nve s t i gat ed w he ther the l igan d/rec ep t o r pai r EDN3 /EDNRB i nv olved i n EN CC d evelo pm ent wa s r e la ted to ENCC ele c tr i c a ctivi t y . Appli catio n o f 1 nM E DN3 at E 11.5 l e d to a thr ee fold CA inc rea se ( Vide o S3 , Fig.2 a , Fig .S5 ), w hile f urt her addi tion of E DN3 ha d more variabl e eff ec t s , pre sumably c aus ed b y the ex haus t i on of C a 2+ res ervoir s up on r epea t e d stimul ation wi t h EDN3 . I n sta ge E12 .5 ile um, CA a l s o i ncr ea sed th r e e fold a t 1 nM E DN3 (Fig.S 5) , and the a ppl ica t io n o f 10 n M EDN3 ind uc ed an imme di ate inc rea s e of in t r ac ellula r C a 2+ ac ro ss a ll E NC Cs (Fig .2 b , Vi deo S3, F ig .S 5) . U n l i k e E D N 3 , G D N F , t h e o t h e r m a j o r l i g a n d i n v o l v e d i n E N C C d e v e lo p m e n t , d i d n o t m o d i fy C A a t t h e 10 ng/mL c oncent rati on know n to d rive EN C C ch emota xi s 24 (Fig .2c, Fig .S6 ). E D N 3 i s end ogeno u s ly ex pr e s s ed b y th e gut me senc hyme, co n centra ting a t th e cec um an d a t th e a nti- m es ent eric ile al border 25,26 . T his pat ter n mir rored the C A h ea tmap s we record ed (Fig.1 a , Fi g.S3). We the re fore bloc ked endog e nou s ED N 3/E DNRB s i g naling wi t h t h e ED NRB an tago ni st BQ788 : it almo st ex t i nguish ed C A ( - 77±14 % , n =14 ) (Fig .2 d, Fig .S5, Vid eo S 4 ) for a t le a st 24 h (Fig .2e) , whil e t h e dr u g ve hicle , DMSO , di d n ot aff ec t C A (Fig .S 7). In a ddi tion , B Q 788 sy stem atical ly l e d to a r o u nding o f ENC Cs and a r et rac t i on o f th ei r cel l p r oce ss e s (Fig .2 f, Vid eo S4) . W e f i nally te s t ed whe the r C A diff er ed in an E D N3 mi s s en s e mu ta tion mode l t ha t de vel op s H i rsch sprung di s ea se , th e ls/l s mou s e . Bec au se t h i s mou s e didn’t e xpre s s G C a MP , we re sort ed to 1 d ay on -s ub str at e-culture o f int e s tina l ex plant s in medi um with GDN F and lo ad ed the prep a r a t io n with F luo4 - AM . The a verage freq uency o f E N CC Ca 2+ t ran sie nt s wa s low er by ~25 % in t he l s /l s homozygo te t h a n in c ontrol s ( Fi g.2g- i) . (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 2. EDN3/EDNRB d r ive s C a 2+ a c ti v ity in E NC Cs . ( a ) E f f e c t o f E D N 3 1 a n d 1 0 n M o n C a 2+ a c t i v i t y ( C A ) i n E 1 1 . 5 i l c c , n = 1 2 , W MP test. (b ) S h arp Ca 2+ ri s e spa nn ing a l l G Ca M P posi t ive c e l ls after 10 nM E DN3 ad m i nistr a tio n dete cted i n 11 /13 sampl e s i n E12 . 5 ileum ( Vid e o S3). R ight p an e l: G C aM P sig nal i ntensity in the d ash e d r e d r ec ta ng le up on E D N3 i n trod u cti o n. The intra c ell u la r Ca 2+ rise co uld no t be s ti m u l a t ed a gain upo n renewe d ap pl i c a t ion of 10 nM ED N3 (n = 6 /6). (c) Effe ct o f GDNF 10 n g/mL in E 11.5 ilcc , n=11, WMP tes t. (d) E ff ect o f the EDNR B bl o cke r B Q78 8 10 µM, E 1 1 . 5 ilcc , n=14, W MP tes t. (e ) Ev o lution of CA i n c on tro l (n=6) and BQ 788 (n = 6 ) tre a t e d sampl es a ft e r 2 4h in cult u re , Ma nn Wh i t ney t e st. C A in cre a se o f con tro l samples af te r 2 4h cu ltu re do e s n ot reflect the p hys i ologi c a l , sta ge-b y - sta ge decr e a se o f CA ( Fig.1b ) : this d i sc repa ncy cou l d a rise f rom c o m p on e n ts of the m ed ium, po ssibl y se rum. (f) Left: ENCC proc ess re tra c ti on and c ell ro undi n g (yello w a rr o whead s) o bse rv ed after 15 min i n cubat io n in BQ 7 88 (Vid e o S4). R igh t : mag n i fied a p p e a ran ce of a cell gro up be f or e / af t e r B Q 78 8 a p p l i c at i o n . ( g, h ) L e f t p an e l s : S o x 1 0 an d α- S M A im m u no hi s t o c h e m i s t r y of E N C C s a nd m e se n c hy m a l ce l ls i n c ontrol (hete ro z yg ote) a nd ls/ls sample. R i g ht pa nels: re g iste red heatmap o f Ca 2+ t r a ns i e nt fr e qu e n c y , E N C C a ggrega te bo rders ar e sho wn i n yello w as drawn fro m t h e IHC imag e s. E NC Cs d isp l ay low e r fr e q u ency tha n the mes enchym e . (i) Fre q uency o f tra nsients for n= 7 c ontrols a nd n = 5 l s/ l s, M a n n Whitney tes t. Each do t i s a differe n t sampl e / e m b ry o a nd a l in e c o nn e cts t h e s a m e sa mp le before a nd af te r d rug applica tion in a l l figu r es o f this report. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint T-type Ca 2+ channels and Cl - channels mediate Ca 2+ oscillations We next investiga ted the ionic transport mechanisms leading to Ca 2+ oscillations downstream of EDNRB. Removing extrac ellular Ca 2+ w i t h E D T A l e d t o a c o mp l e t e c e s s a t i o n o f C A ( F i g . 3 a , V i d e o S 5 ) . Consistent with the requirement for extracellular Ca 2+ , GdCl 3 , a cation channel blocker, signific antly reduced CA (Fig.3a, Fig.S8). Interestingly , transients could still be elicited by acute administration of EDN3 10 nM in the presence of EDT A (Fig.3e, Fig.S9, Video S5), indicating a contribution of int racellular Ca 2+ stores. 2 -APB halted activity (Fig.3a, Fig.S8) while ryanodine did not affect CA (Fig.3a, Fig.S8). These findings show that both extracellula r Ca 2+ and intracellular IP3-, but not ryanodine-sensitive Ca 2+ stores contribute to CA. We further questioned the entry pathway of extracellular Ca 2+ , first considering voltage-gated Ca 2+ channels (VGCCs). The L-type Ca 2+ channel (CaV1.2) blocker nifedipine did not significantly alter CA (Fig.3b, Fig.S10), consistent with previous observations 27 . Ruthenium red, a non-selective blocker of P/Q-type (CaV2.1) and N-type (CaV2.2) channels as well as of cationic TRPV cha nnels, did not affect CA either (Fig.3b, Fig .S10). Blockade of all 3 T-type Ca 2+ channels (CaV3.1,2,3) with Z944 induced a dose- a nd time-dependent dec rease of CA (Fig.3b, Fig.S10,S12, Video S6). CaV3.2 (encoded by Cacna1h ) is strongly expressed in E11.5 ENCCs 28 ; we found that s pecific blockade of CaV3.2 with a s c o r b i c a c i d ( A A ) 1 mM 29 also induced a dose- and time-dependent decrease of CA (Fig.3b, F i g . S 1 0 , S 1 2 , V i d e o S 6 ) . T h e o n l y c o mme r c i a l l y a v a i l a b l e T - t y p e C a 2+ channel agonist, SAK3, is specific for CaV3.1 and CaV3.3 30 . This molecule induced a significant CA increa se at 10 µM (Fig.3b, Fig.S10, Video S6), suggesting that all three T-type Ca 2+ channels are involved in ENCC CA. IHC for CaV3.1, Ca V3.2 and CaV3.3 (Fig.3d-f) showed that all three channel types were indeed expressed by ENCCs (Sox10+ cells), but not by the surrounding mesenchyme. SAK3 also led to a significant CA increase post-EDNRB blockade with BQ788, but CA remained well below physiol ogical levels, even after 1 day culture (Fig.S14). We next t ar geted Cl - channels by applying three different classes of inhibitors: 5-Nitro-2-(3- phenylpropylamino)benzoic acid (NPPB), niflumic acid (NFA) and 4,4'-Diisothiocyanato-2,2'- stilbene disulfonic acid (DIDS ). These compounds led to a drastic CA decrease (Fig.3c, Fig.S11, Video S7) and a retraction of cell proc esses (Video S7). Transient CA could also be elicited by E DN3 10 µM after 15 min incubation in 100 µM NPPB, but the transients had low amplitude (Fig.S9). NPPB and DIDS sy stematically induced ENCC death after 24 h (Fig.S15). NFA 50 µM inhi bition kinetic s were similar to Z944 50 µM (Fig.S12): CA had recovered after 24h, ENS morphology was normal, and the frac tion of dead ENCCs similar to control samples (Fig.S16). Cl - channels likely contribute to Ca 2+ transients by depolarizing the c ell membrane through Cl - efflux, ther eby opening VGCCs 31,32 . This idea is in line with the fac t that the membrane depolarizer 4-AP mark edly increased CA (Fig.S13 ). 4-AP could not however re-stimulate CA post-EDNRB block (Fig.S14c), and also induce d cell death after 24 h (Fig.S15). The purinergic pathway is an important actor of ENCC multicellular Ca 2+ w a v e s 23 . We found that the ATP e enzymatic degradation inhibitor ARL 67156 and exogenous ATP e could transiently increase wave activity after EDNRB blockade (Fig.S14d,e). Higher ATP e c o n c e n t r a t i o n s c o u l d e v e n r e s u l t i n a network-spanning intracellular Ca 2+ rise (Fig.S14d, Video S8). These effects were howeve r not sustained in time (Fig.S14e), indicating that the purinergic pathway could not bypass the main regulator EDN3/EDNRB. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 3 . T-ty pe C a 2+ cha n n e ls and C l- c hannels are c ri tic al f or Ca 2+ os ci llati ons. A l l e x p e r i m e n t s i n ( a - c ) a r e p e r f o r m e d a t E11 . 5 a t th e i leo-ce ca l ju nc ti o n. (a) CA ch ange re l ative to co ntrol befo re d rug ap pl ic ati o n f or EDT A (2 m M , n = 6 ), GdCl 3 (10 0 µM , n = 6 ), 2 -APB (100 µM, n = 5 ), r ya n o d in e (10 µM, n = 5 ). (b ) C A cha n g e for n ifedi pin e (Nif, 1 0 µM, n= 5 ), ru theniu m re d (R uR, 1 0 0 µM , n=6), Z 94 4 at 1 µ M (n = 7 ), 10 µ M (n = 7 ), 5 0 µM (n = 9 ), a sc o rb i c acid (AA , 1 mM , n=13 ) a n d SAK 3 a t 1 µ M (n=6), 1 0 µM (n=6). (c) CA c h an ge for NPP B (1 0 0 µ M, n= 1 1), NFA (5 0 µM, n=9) a nd D IDS ( 500 µM , n=5). p -valu e s fo r (a -c) ar e cal c u late d from the Stud e n t t-test. ( d-f) So x1 0 a n d Ca V3. 1 (d ), CaV3.2 ( e ) a nd CaV3 . 3 (f) IH Cs o f ENC Cs mig rating f rom an in te sti na l e xpl an t on a sub strate. A ll three T-typ e Ca 2+ ch annels are expre ss e d by E NC Cs. The sma ll b ri gh t green do ts are AF647 µ -bead s fl uorescing in the sam e ra ng e a s the T -typ e Ca 2+ ch ann el secon dary an tibody. They wer e ad ded in the o v erlying c ollag en g e l for anoth e r experim e n t (s ee Fi g ure 6) . (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Ca 2+ activity correlates with ENCC migration speed We next investigated how pharmacological upregulation of CA by EDN3 or SAK3 affected ENC C migration from E11.5 midgut explants embedded in a 3D collagen gel. We found that both EDN3 1 nM and GDNF 10 ng/mL increased migration distanc es from the explant compared to con trol conditions, and that they ac ted synergistically when combined (Fig.4). Strikingly, we found tha t the C A a c t i v a t o r S A K 3 w a s a b l e t o mi mi c t h e e f f e c t o f E D N 3 , i . e . mi g r at i o n d i s t a n c e s w e r e s i mi l a r l y h i g h in GDNF + SAK3 10 µM and GDNF + EDN3 1 nM conditions (Fig.4). We next assessed the effect of CA inhibitors on ENCC migration in the colon in full E11.5 guts cultured for 1 day in agarose, a gel not permissive to migration. We found tha t the four condition s that significantly and long -lastingly (Fig.S12, Fig.2e) lowered CA - Z944 50 µM, NFA 50 µM, AA 1 mM and BQ788 10 µM - all decreased E NCC migration distance s in the colon (Fig.5a,b). These effec ts could not be attributed to toxicity, as cell death in most (n=49/54) samples was on the same level a s control sample s (Fig.S16). Addition of SAK3 to BQ788 did not improve ENCC migration down the colon, consi stent with the fact that CA was still very low in these conditions (Fig.S14). The se result s show that, in addition to the well-known inhibitory effec t of BQ788 on ENCC migration 33,34 , Cl - channel and T-type VGCC inhibi tion also result in a migration defect leading to colonic aganglionosis. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 4 . C a 2+ activity enh a nce rs ED N3 a nd SAK 3 pr omote mi gr a ti o n in a coll ag e n gel mi gr a ti o n as say. (a ) E 1 1 . 5 m i dg ut e x p la nts em b e d d ed in a coll a gen matri x an d cu l t u red for 3 d ays in con tro l , E DN3 1 nM, GD NF 1 0 ng/m L, GDNF + E DN3 1 nM a nd GDNF + SAK 3 10 µM c o ndit io ns. ENCC migra t io n i n t h e c o lla gen i s v i sibl e i n b righ tf ield Sc h lieren -type li ghting as a ha lo o f rad ially ori e n te d ce l ls surro un din g the da rk er e x p la nt. (b) A v erag e m i grati o n di st a n ce fro m th e expl an t after 3 da y s in the d iffer e n t dru g c o ndit io ns, M an n-Whitney te st. Sa m p les nu m b e rs i n each gro up, fro m left t o r ig h t, a re n= 6 , 10 , 1 5 , 9, 6, 6, 11. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 5 . I n hi b i ti on o f C a 2+ a ctiv i ty sl ow s d o w n ENCC m igr ation i n t h e c e cum and col on. Maximum pro jec ti on of G Ca M P z - stacks of E11 . 5 guts cultured f or 1 day in co n tro l c o nd i ti o ns (with D MSO vehicle al on e ) a nd in v a ri ou s chem i c a l con d i t ion s th at alter CA . The d ashed yello w l i ne mark s th e p ositio n of the the i lcc jun c ti on , and the yello w arro w head the p ositi on of the migra ti o n fro nt a fte r 1 d ay cultu re . AA= ascorb ic acid. (b ) M igra t ion d istanc e in the col on from the il c c j unc tion for con tro l (n = 1 0), Gd C l 3 1 00 µ M (n=6), Z 944 1 0 µM (n = 7 ), Z 944 50 µM (n=7) , NF A 5 0 µ M (n = 9 ), AA 1 m M (n=7) , BQ78 8 10 µ M (n = 10), BQ78 8 10 µM + SAK3 10 µ M ( n = 5 ). P -valu e s a re sh own for the Ma n n -Whitney te st co m p ared to the con tro l gr o up. Z9 44 5 0 µM , NFA 50 µM, A A 1 mM and BQ 788 1 0 µ M si gn ifi ca nt ly slowe d do w n migr a ti o n. B Q + SAK 3 di d n o t i m p rove m i grati o n compa re d to BQ a lone. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Ca 2+ activity drives cell contractility and traction force to the extracellular matrix ENCCs exert a traction force on the tissue as they migrate 24 , visible in-situ as a backflow of mesenchyme (Video S9) . Several observations indicated that Ca 2+ oscillations could modulate thi s tracti on force by influencing cell contractility, a s in smooth muscle: 1°) CA blockers led to process retraction and rounding (Fig.2f, Video S4,7), indicating a loss of ENCC-extrace llular matrix (ECM) tracti on; 2°) In a 2D migration assay at x60 magnification, we noticed that Ca 2+ transient could induce abrupt lamellipodium de tachment from the substrate ( n =4 such events observ ed, Fig.S17a, Video S10). This could be driven by actomyosin or by calpain activity 35 ; 3°) Most Ca 2+ transients (79/102) in this assay were correlated with an a brupt change of nucleus speed, as measured by deep-l earning assisted tracking and by kymographs (Fig .S17b-c, Video S11). To quantify the force exerted by ENCCs on the ECM, we let ENCCs migrate from E11.5 midgut explants in a collagen gel seede d with fluorescent be ads that served a s fiducial markers of gel deformation (Fig.6a inset). The traction force on the collagen sca ffold induced bead displacements towards the migration front (Fig.6, Video S12). Immediately after induction of CA by addition of EDN3 1 nM, the bead speed abruptly increased (Fig.6, Video S12), and the tip of the migration “fingers” accelerated (Video S12). When EDNRB w as blocked by BQ788, the beads suddenly relaxed, indicating a sudden loss of ENCC-ECM traction forc e (Fig.6a,b,d, Video S12), and the migration fingers retracted (Video S12). The effect of t he T-type Ca 2+ channel blocker Z944 was similar to that of BQ788, although the decrease in traction force was less pronounced than for BQ788 (Fig.6c,e). This experiment show s that, similarly to endothelin-induc ed smooth muscle contraction, E DN3 could stimulate Ca 2+ -dependent actomyosin, increa sing the ENCC-ECM traction force necessary for migration, whereas blocking EDNRB or T-type Ca 2+ channels relaxed it. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 6 . C a 2+ tra nsi e nts prom ote migrat ion by in c rea s i ng ENCC tracti on f o rce to the e xtracellular mat rix . (a) ENCC 3D migra ti o n fro m a E11.5 mid gut e xpl an t (to th e l eft, not i n th e field o f view ), after 24 h in a c o l l a gen m a trix with 10 n g/mL GDNF and fiduc ial marke r b e a ds. M aximum proj e ctio ns of z-sta c ks shows ENC Cs t o t h e l e ft (g reen, GCaM P sig n a l) , and AF647 bead s (re d ) i nse rt ed in the col l a g e n matrix. T h e veloci t y vec to rs (cyan) are d e d uce d fro m bead track i ng in 1 sta ck - per- m i n ut e v id e o s ( Vi d e o S 1 2 ) . L e f t p an e l : i n i t ia ll y , t h e be ad s m ov e t o w ar d t he E N CC s be c au s e o f t he t r a c t i o n fo r c e t h e y e xe r t o n th e co llagen g e l . The in s et sh ows the ex p e ri menta l setup . Mi d d le pa nel: t h e traction fo rc e sh arpl y in creas es w h e n Ca 2+ r i s es a fte r EDN3 1 nM admin istrat i o n . Right p an e l: th e tractio n fo rc e in v erts ( re l ax es) a ft e r a pp lica tio n of BQ788. (b ) Ave ra g e b e a d x d i s p l a c e m e n t f r o m t h e i r i n i t i a l x p o s i t i o n v e r s u s t i m e ( r e d c u r v e ) , a n d a v e r a g e G C a M P s i g n a l m e a s u r e d i n a n R O I surro un di n g the E N CCs ( gree n ), in a con t rol, E DN 3 1 nM , EDN3 10 n M a nd BQ 7 8 8 1 0 µ M sequ e n c e a nd (c) a c o ntrol, E DN 3 1 n M and Z 94 4 50 µ M s eq uence . T h e av e ra ge b e a d d ispl a cement wa s performed o v e r a ll bead s that we r e n ot i n d irec t c ontact with the ENCCs. T h e a ve ra ge b ead spe e d for e a c h condition is defin e d as th e slop e o f th e d ash e d lin e s. Note tha t the time- resolu tion o f th i s e xp erim ent (1 z-stack per minu t e ) did n ot a l low to r esol v e in d e ta il th e Ca 2+ tran s ie n ts, b ut only occ a si on a l Ca 2+ bu r s t s ( e d n3 ) or t h e ab s e nce of ac t iv it y ( BQ 7 88 , Z 94 4) . ( d) A v e r a g e b e ad s p e e d fo r n = 9 c o n t r ol - E D N 3 1 nM – E D N 3 1 0 n M – BQ78 8 10 µM e xperim e n ts an d (e ) n=4 contr o l - E DN3 1 nM – Z944 50 µM e xpe riments, p aired Student t-te st. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint

Endothelins were identified in 1987 as potent modulators of vascular smooth muscle (vSMC) contractility 36 . When binding to EDNRs on vSMCs, endothelins drive an increase in cytosolic Ca 2+ mediated both by the release of Ca 2+ f r o m i n t r a c e l l u l a r s t o r e s a n d b y t h e e n t r y o f e x t r a c e l l u l a r C a 2+ via CaV1.2 L-type VGCCs 32,37,38 . The opening of CaV1.2 is favored by the efflux of Cl - f r o m t h e c y t o s o l to the extracellular medium, res ult ing in membr ane depolar iz ation 31,32 . Increased cytosolic Ca 2+ leads to calmodulin activation and acto-myosin cross-bridges responsible for vSMC contrac tion. As the role s of endothelins in blood pressure regulation became more prominent, mutant mice for EDN3 39,40 and EDNRB 41 mouse were re-examined. The phenotype of these mice came as a surprise: they presented with colonic aganglionosis at birth, a phenotype known as Hirschsprung disea se (HD). E D N 3 / E D N R B h a s l a t e r b e e n i d e n t i f i e d a s a k e y p a t h w a y p r o mo t i n g E N C C p r o l i f e r a t i o n a n d d e l a y i n g their differentiation to enteric neurons and glia 10,42,43 . The EDN3/EDNRB pathway has since then been attributed a dual, “moonlighting” role 14,44 : regulator of vSMC tone and blood pressure in adult physiology, driver of ENCC proliferation in the embryo. We reveal here that the immediate action of ED N 3 o n EN C Cs i s i n f a c t n ear ly i d e nt i ca l t o its eff e c t o n s moo t h mu s c le to n e : it tr i g ge r s Ca 2+ a ct iv i t y ( F i g . 1 , 2 ) , v i a a v e r y s i mi l a r mo l e c u l a r r o u t e ( F i g . 3 , 7 ) t o v S M C , a n d t h e i n c r e a s e d c y t o s o l i c C a 2+ oscillations enhance migration (Fig.4,5) by inducing cell contractility (Fig.6). vSMC contraction leads to vessel constric tion; ENCC contraction leads to a n increased traction force to the extracellular- mat r i x, t h a t is n e c es s ar y f o r t he i r mi g r a t io n i ns i d e t h e g u t mes e n c hy me . T his f o r ce i s t h e r e as o n wh y investigators have found it indi spensable to pin 45 or embed the gut tract during ex-vivo migration assays 46 , as it otherwise leads to tissue shrinkage and improper migrati on. The traction force i s transmitted via β1-integr ins ( Fig.7), and EN CC-specific β1- integrin mutants have been s hown to present with a HD phenotype 47–49 . Ca 2+ signaling defects reduce the traction force of the ENCCs to the extracellular matrix, lowering their migration speed, resulting in colonic aganglionosis. Our in v e s ti g a t i o n sh o w s t h at E NC Cs ar e a k i n t o min i at u r e mu s c les t ha t c o nt r a ct an d c r a wl i n r es po ns e to a constrictor peptide, endothelin 3. It i s likely that cytosolic Ca 2+ oscillations are also involved in the long-term effects of EDN3/EDNRB on ENCC prolifera tion. ENCC proliferation and migration a re inexorably linked – ENCC cannot invade the colon or a collagen gel if th ey do not also proliferate 50 . Here, we reported macroscopic migration distances (linear in the colon, spherical in the gel) rather than local proliferation rates. EDN3/EDNRB also delays ENCC differenti ation 10,42,43 . Higher CA is generally associated with an undifferentiated state in stem cell cultures 16,17 . The mitogen activat ed pr otein kinase (MAPK) and the phosphoinositide 3-kinase (P I3K) pathways can both be induced downstream of EDNRB 51 , and are dependent on Ca 2+ a c t i v i t y 52,53 (Fig.7). MAPK family members extracellul ar signal-regulated kina se (ERK) and c-Jun N-terminal kinase (JNK) both promote ENCC migration 54 , and likely also regulate ENCC proliferation and differentiation. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig u re 7. S ynth e tic s cheme o f the path w a y s le adi ng t o Ca 2+ transient generati on and ENCC migrat ion. P R : p ur i n e r gi c recepto r, IP3 : ino si tol trisphosph ate, CaM : ca lmo d ulin, ML CK: my o sin lig h t c ha i n kin ase, MA PK: m i tog e n a ctiva ted p rotein k i n as e , PI 3K : p ho s ph oi n os i t i de 3 -k in a s e . D as he d a r r ow s i nd ic at e m e c ha ni s t i c l i nk s f or w hi c h t h e r e i s e v i d e nc e f r o m t he li te ra ture in re l ated systems. EDNR B act iva t ion ca n tri g ger a d irect re l e a se of C a 2+ f r om i n t e r c e ll u la r s t or e s , b u t C l - c ha nn e ls, T-ty p e Ca 2+ ch a nnels and ex tra ce ll ula r Ca 2+ a re nec e s sary fo r susta ined Ca 2+ osci llations. T- type re cept or bloc kad e by mibe fr adi l 1 µM wa s p rev iou s l y repo rt ed not to a f f ec t E N CC mi gratio n 28 . I t i s p ro b a b l e t h a t t hi s c o n cen t ra t i o n wa s ins uffi ci en t t o c aus e a ga n g l i onos is , be c aus e t h e IC 5 0 o f mibe fr adi l on i sola ted cel l s i s 3 µ M 55 ; i n o ur exp er i ence, e ff ec t i ve, s u stain ed bloc king of c hanne l ac tivity in wh ole - gut c ul tur e s requi re s c oncen t r ation s fa r ab ove d rug I C 50 t o y ield a mig r a t i o n p h e no t ype . H ir s t et a l. 28 al so block ed C l - chan nel s w ith NPP B 100 µM fr om an only 10 0-fo ld s tock solu tion in a no n - spec i fie d s ol ve nt, a nd found n o e f fec t o n migration o r o n E NCC s u rvi val. In our ex periment s, NPPB 100 µM di s r up ted C A a nd migra tion by sy ste matica lly in ducin g ENCC de at h (n=11 /11, Fig.S 15) . The choi ce o f N PP B s olve nt or incon si s t e ncie s in t he ac tua l co nc entra tion s ap plie d by Hirst et a l . ma y ex plain the di scre pa ncy. Mo re ge ne rally, we wer e guide d in o ur cho ic e s o f pha r ma colog ical c ompound s and co nc e ntr a t i on s by th e C a 2+ r e s pon s e, wh e r e a s Hi r s t et a l . a pplie d the se c om pound s “blind ly” t o wild type g uts. O ur s t ud y differ s m e t h odologic ally fr om t he pion eering w o r k o f H a o e t a l . 23 . The se inve s tigato r s focu sed on mul tice llula r Ca 2+ w aves , with fr equ enci e s of ~10 -4 wav es /min /100 µm² . Th is i s c on s i st ent with th e w ave a ctivi t y we me a s ure , b ut ~ 100- 1000 tim e m o r e fr e q u e n t e v e n t s o c c u r a s s i n g l e - c e l l t r a n s i e n t s a n d a r e t h e m a i n fo r m o f E N C C C A . W e fo u n d (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint that, although ATP e and ARL 67156 can indeed promote multicellular waves 23 , they only play a modulatory role in a mechani sm that i s dominated by EDNRB, Cl - and CaV3 channels. Ca V3.1 and CaV3.2 expression in E11.5 ENCCs ha s been previously measured by PCR 28 ; CaV3.2 was found to be ~640 time s more expressed than CaV3.1. Our IHC, CA and migration assay s indicate that all three CaV3 channel types are present and functi onal in ENCCs. The CaV3.1 & CaV3.3 agonist SAK3 promoted CA (Fig.3) and ENCC migration (Fig.4); th e CaV3.2 antagonist ascorbic acid reduced CA (Fig.3) and ENCC migration (Fig.5). Very interestingly, a single-nucleotide polymorphism (SNP) of the CACNA1H gene encoding CaV3.2 has been uncovered in a rec ent HD exome-wide associa tion study 15 . Although i t is not known whether this point mutation altered CaV3.2 expression or func tion, our findings suggest it induces HD by al te ring Ca 2+ signaling. This motivates further res ear ch using dedicated mouse models to better understand the causes of neurocristopathie s. We note that the Ca V3.2 -/- mouse is viable 56,57 , although the mutati on induce s pre-natal lethality 58 . It is possible that the surviving Ca V3.2 -/- embryos devel op compensatory Ca 2+ influx mechanisms, a common behavior when only one of several protein isoforms is knocked-out 59 . We found tha t GDNF and EDN3 (or SAK3) have a synergistic action on ENCC migration from explants (Fig.3). Our results a re in agreement with the findings of Bergeron et al. 60 . We did not observe that EDN3 and GDNF we re antagonistic, as has been reported in other studies with mouse 25 , rat 61 and chicken embryos 46 . These studies were performed respectively a t stages E10.5-E11, E13 and E8, at EDN3 concentrations of 100 , 20 and 100 nM and after 16h, 2-3 days and 3 days of migration. It i s d if f ic u lt to f a t ho m t he r eas o ns of t h is d is c r ep an c y b u t we s t ress t ha t : 1°) mi gr at ion a ft er 1 6 h 25 i s t o o scarc e to be precisely quantifie d, 2°) the EDN3 concentrations applied by other investig ators are 1-2 orders of magnitude higher t han the maximum pro-CA and pr o-migrator y effect we report at 1 nM. 1 nM likely reflects the concentration encountered physiologically by ENCCs as they migrate down the gut mesenchyme. We found that CA was not significantly increa sed at 10 nM EDN3 compa red to 1 nM (Fig.2a, Fig.S5c), indicating a saturation effect. Administration of EDN3 10 nM at E12.5 gave rise to a single pan-ENCC transient, which, although impre ssive, most likely never occurs physiologically. The pan-ENCC Ca 2+ surge induced by exogenous EDN3 may occur only at E12.5 (and not at E11.5 ) because of a reduced saturation of EDNRB receptors by endogenous EDN3 a t this stage, making it more sensitive to exogenous application. Edn3 expression is highest in the cecum, with little expression in the hindgut at E12 26 , which c o r r e l a t e s w i t h r e d u c e d C A a t t h e E N C C w a v e f r o n t a t E 1 2 . 5 ( F i g . 1 ) . T h e s e o b s e r v a t i o n s s u g g e s t t h a t EDN3 may be most necessary up until the cecum is traversed; slowing down o f E NCCS by CA activity inhibitors in our ex-vivo assay (Fig.5) may have primarily affected migration through the cecum, resulting in a paucity of ENCCs in the colon. Colonic aganglionosis can result from slowed-down ENCC migration at any point of their journey down the gut, not necessarily from slower migration in the colon. We found that ls/ls ENCCs only displayed a ~25% reduction in CA compared to controls. This experiment was performed in the presence of GDNF, serum, and after 1 day culture, all of which can ar tificially hike CA (Fig.2e) compared to its in-vivo state. A more refined appr oach based on GCaMP expressing ls/ls mutants would, given our results on the effect of EDNRB blockade, likely reveal a much more drastic CA d eficit. Testing the hypothesis that Cl - channels drive an efflux of Cl - ions and a concomitant depolariza tion (Fig.5) a s in smooth muscle 31,32 will require electrophysiological measur ements . The identificat ion of the Cl - channels involved hinges on the future development of more specific antagonists and agonists: ENCCs express many different Cl - channels 28 and several types may be involved. We have not found a correlation betw een mutations a ffecting Cl - channels and HD pathogenesis in the literature. This of course does not pre empt the fact that they are (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint important for ENCC migration, as such mutations may plainly be le thal. Although our pharmacological approach allowed us to specify CaV3 channels as an important gateway of extracellular Ca 2+ , we cannot at this stag e exclude oth er non-VGCC entry mechanisms, like TRPC 62 . We deciph ered an important mechanism by which endothelin 3 triggers Ca 2+ activity and contrac tility necessary for the migra tion of enteric neural crest cells. This mechanism is likely to play a more general role in NCC de rived melanocytes or Schwann cells and EDNRA-expre ssing cranial NCCs 14 , potentially linking Ca 2+ activity to other neurocristopathy syndromes 3 . Melanoma 7 a n d neuroblastoma 63 are known to recapitulate many NCC traits. T-type Ca 2+ channel upregulation is associated with melanoma aggressiveness 64 w h i l e C a 2+ signaling is altered in neurobla stoma 65 . The endothelin axis is more generally aberrantly expressed in tumors 66 , promoting invasion and meta stasis: the mechanism we describe here for neural crest cells may also be at play in cancer cell migration.

Et h i cs Mice were hosted at the Institut Jacques Monod (GCaMP) and at the Institut Imagine LEAT (ls/+ hete rozygotes) animal husbandries. They had access to housing, food and water ad-libitum. For mating, the male is introduced in the ev ening and removed in the morning after detection of plugs in the morning. Pregnant mice were killed by cervical dislocation to re trieve e mbryos age E10.5 to E 1 2 . 5 . T h e e mb r y o s w e r e s e p a r a t e d a n d i mme d i a t e l y b e h e a h e d w i t h s u r g i c a l s c i s s o r s . T h e me t h o d s used to kill the mice conform to the g uideline s of CNRS and INSERM animal welfare committees. Killing of mice for retri eval of embryos is a terminal proc edure for which neither CNRS or INSERM assign ethics approval codes hence none are given here. Mouse gut samples The Cre reporter mice C57BL/6N-Gt(ROSA)26Sor tm1 (CAG-GCaMP6f)Khakh/J mice is referred to a s Gcamp6fl/fl. A transgenic mouse line in which the transgene is under the control of the 3-kb fragment of the human ti ssue pl asminogen activator (Ht-PA) promoter Tg(PLATcre) 116Sdu16 is referred to as Ht-PA::Cre. GCamp6fl/fl male s were crossed with Ht-PA::Cre females to generate embryos carrying the calcium fluorescent reporter in neural crest cells and their derivatives. This model was previously described 27 . The gut of each embr yo was dissected in PBS wit h 1 mM Ca 2+ , 0.5 mM Mg 2+ and 1% penicillin-streptomycin, from stomach (cut at the oesophagus-stomach junction) to colon (cut at the colon-anus junction). The mesente ry of E11.5 guts was kept intact. 57 % (118/207 for which the full count was performed) of the embryos expressed GCaMP specifically in NCCs, while 43 % expressed GCaMP in all cells, with the mesenchymal signal being dominant 67 ; thes e phenotypes could be ea sily recognized during confocal imaging. We do not know the reason of this partial s p e c i f i c i t y o f G C a M P . O n l y e mb r y o s e x p r e s s i n g G C a M P i n N C C s w e r e u s e d f o r C a 2+ imaging. Ubiquitously GCaMP expressing embryos, which were otherwise morphologically normal, were used for experiments where Ca 2+ imaging was not r equir ed. H e t e r o z y g o u s l s / + a n i ma l s w e r e c r o s s e d a n d e mb r y o s r e t r i e v e d a t E 1 1 . 5 . D N A e x t r a c t i o n a n d subsequent genotyping were performed from head biopsie s of each embryo, using the direct PCR (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint lysis reagent (Viagen). Genotyping was performed using primers (Eurogentec, Belgium) as previously desc ribed 68,69 . Organ c ultur e and Ca 2+ ima ging After dissection, each gut was placed on a pre-solidified 0.5 mL layer of 1% low-melting point agarose (Condalab 8050.11, dissolved in PBS with Ca 2+ 1 mM a n d M g 2+ 0 . 5 mM ) , i n a 3 5 mm d i a me t e r P e t r i dish (Greiner). This gel layer prevents a dhesion of the gut and subsequent E NCC migration on the dish bottom, c onfining it to the organ. The gut was then immobilized by pouring additional 0.5 mL of liquid agarose on top and letting it gel at 4°C for 10 minutes. Capillary forces of the shallow liquid agarose layer gently pre ss down the gut, most of the time positioning the stomach, midgut and c olon in a horizontal plane that was optimal for microscopy. Immobilization is crucial both for Ca 2+ imaging and to prevent shrinking / retraction of the sample during culture. The 1 mL gut & gel were covered by 2mL of complemented DME M:F12 Glutamax (Gibco 31331-028). Final concentrations a fter diffusion and dilution of the medium in the PBS-based agarose were Ca 2+ 1 mM , M g 2+ 0 . 5 mM , penicillin-streptomycin 1%, Fetal Veal S erum 6.6 %, while all other molecule s that are present in DMEM:F12 but not in PBS (e.g. glucose, amino acids etc.) had their concentration divided by a facto r 2/3 compared to DMEM:F12 alone. Samples were incubated at 37 °C in a 5% CO 2 95% air atmosphere for at least 45 min before imaging. Calcium imaging was performed on an inverted spinning-di sk microscope (Olympus IX-81, Yokogawa CSU-X1) equipped with an ILE laser-base (Andor) and a Zyla camera (Andor, resolution 1392x1040 pixel). The GCaMP signal was excited at 488 nm and the emission filtered a t 497-527 nm. Time-lapse videos we re recorded at x10 magnification, 400 ms exposure time, 1 Hz acquisiti on rate, for at lea st 3 min, and often longer depending on the type of experiment. We did not observe any illumination- induced bleaching or alteration of the sample. Sample basal (endogenous) Ca 2+ a c t i v i t y w a s f i r s t recorded, and the position of the ENCC wavefront determined by a z-stack at the level of the ileo- caecal junction (E11.5). Drugs were the n added from stock solutions directly in the Petri dish, fa r from the sample, without displacing it, and homogenized by up & down move ments with a 1 mL pipette. In preliminary experiments, addition of the drug was performed during the time-lapse to capture potential immediate effects. For most drugs, CA was then recorded 10-15 min post- administra tion. After imaging, samples were placed back in the incubator for further culture. Time- lapse and z-stack imaging was performed after 24 h to a ssess ENCC wavefront position and CA. Some sampl es were imaged multiple times during the culture period to a ssess drug kinetics (Fig.S12 ). Some sampl es were fixed and processed fo r IHC a s de scribed below. The experiment with ls/ls embryos required that the ENCCs migra te out of the intestine becau se Fluo4AM does not penetrate inside the tissue. We achieved this by placing E11.5 explants on 35 mm Greiner Petri dishe s and covering them with a thin meniscus of collagen gel (0.5 mL, 1 mg/mL, a s desc ribed below) to hold them down a gainst the Petri surface. After gelling, the preparation wa s topped with 2 mL complemented medium with GDNF 10 ng/mL, cul tured for 1 day, loaded with 1.8 µM Fluo4-AM for 5 min, and 3 min timelapse movies were recorded in 3 different loc ations per sampl e. The samples were immediately fixed after calcium imaging for IHC. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Ph arma cology Products used in this report a s well a s solvent and stock concentra tion are listed in the table below: Full na me Short na me S upplier / Refer e nc e Sol vent Stoc k conc e ntr a t ion Tetrodotoxin citrate TTX Abcam 120055 H 2 O 1 mM Veratridine Vera Sigma V5754 DMSO 20 mM Endothelin 3 EDN3 Calbiochem 05-23-3801 H 2 O 1-100 µM Glial-derived neurotrophic factor GDN F Alomone Labs G-240 H 2 O 10 µg/mL BQ788 BQ Alomone Labs B-176 DMSO 5 mM Fluo4-AM Invitrogen F1420 DMSO 4.6 mM Ethylenediaminetetraacetic acid EDTA Sigma E6511 H 2 O 1 M Gadolinium chloride GdCl 3 Biotechne 4741 H 2 O 10 mM 2-Aminoethoxydiphenyl borate 2-APB Biotechne 1224 DMSO 100 mM Ryanodine Ryad Biotechne 1329 DMSO 10 mM Nifedipine Ni f Sigma N7634 DMSO 10 mM Ruthenium red RuR Biotechne 1329 H 2 O 10 mM Z944 Biotechne 6367 DMSO 100 mM L-Ascorbic acid AA Duchefa A0602 H 2 O 1 M 5-Nit ro-2-(phenylpropylamino)- benzoate NPPB Biotechne DMSO 100 mM Niflumic acid NFA Sigma N0630 DMSO 150 mM DIDS Sigma 309795 H 2 O 100 mM 4-Aminopyridine 4-AP Sigma 275875 H 2 O 100 mM ARL 67156 ARL Biotechne H 2 O 20 mM Adenosine t riphosphate A TP e Biotechne H 2 O 10 mM SAK3 Sigma SML2039 DMSO 100 mM Table 1. Drugs used in this report, with solvent and stock concentration Ca 2+ a ctivity a nalysis Analysis of CA was performed using custom-written ImageJ macro and Matlab sc ripts, and is synthesized in Fig. S1. Briefly, a fixed mesh comprised of 2745 square regions-of-interes t (ROIs), each 2 5 0 µ m² , w a s o v e r l a y e d o n t h e 8 - b it v i d e o . A ma t r i x o f t h e a v e r a g e p i x e l i n t e n s i t y o v e r t i me i n e a c h square was retrieved and input in to Matlab for peak detection. The “findpeaks” function wa s applied, specifying, to filter out noise, a minimum peak prominence of 2 (8-bit pixel intensity unit), a minimum peak width of 3 sec, a minimum inter-peak time of 3 sec. Illumination and peak filtering settings were identical for all videos acquired, allowing quantitative compa rison of GCaMP signal intensity ac ross samples. We measured in each ROI the number of peak s, and, for ROIs in which there was at least one peak, the frequency (number/acquisition time), average duration (width at half-maxima) and average intensity ratio I/I 0 ; the data of all ROIs was collected to generate heat maps of these vari ables, assess their spatial distribution, and compute thei r spatial average . To reflect the general activity of the region imaged, we defined the Ca 2+ a c t i v i t y /g1829/g1827 /g3404 /g1840 /g3043/g3032/g3028/g3038/g3046/g4666/g1827 /g3030/g3032/g3039/g3039/g3046 /g1846/g4667⁄ , where /g1840 /g3043/g3032/g3028/g3038/g3046 i s t h e t ot a l n u mbe r of t r a ns ie n ts in a g ive n v i d eo , /g1827 /g3030/g3032/g3039/g3039/g3046 the total area of the GCaMP positive cells within the field o f view and /g1846 the duration of the acquisition. /g1827 /g3030/g3032/g3039/g3039/g3046 w a s d e t e r mi n e d b y f i r s t time-projecting the average intensity to obtain a crisp image of the ENCC network, and by then applying iteratively stricter Bern sen local thresholds under ImageJ, stopping just before it diverged (Fig.S1e). /g1827 /g3030/g3032/g3039/g3039/g3046 should remain consta nt for an experiment with a fixed z and field of view. Because our proc edure yielded some variability depending on Ca 2+ b a s a l l e v e l a n d a c t i v i t y , w e s e l e c t e d f o r a (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint given experiment the biggest /g1827 /g3030/g3032/g3039/g3039/g3046 and applied it to all other conditions of this experiment (e.g. before /after drug). CA wa s expressed in units of events/100 µm²/min; because a cell has typically an area of 100 µm², CA has values lying in the same range as frequencies expressed in cycles-per-minute (cpm). For ls/ls samples, we followed the same procedure but used a finer mesh of 86 µm² (9890 ROIs in total) to gain resolution to differentiate EN CC fr om mes enchymal CA. After r egistration of Sox10 & α- SMA IHCs with freque ncy heatmaps, we drew ROIs around Sox10 positive aggregates using the ImageJ ROI manager, a nd then measure d the average pixel intensity of these ROIs on the grey scale frequency heatmap. The average frequency was obtained by weighing the measured frequency o f each aggregate by its area (i.e. bigger aggregates contributed more to the average). Area-weighted frequency and CA are proportional; we present the former in Fig.2i because the finer mesh size used

in a higher event count compared to the standa rd mesh used in the remainder of the study. Co llagen g el organ culture Collagen gels (Fig.4,6) were prepared from Cultrex Rat Collagen I (Bio-Techne) at 1 mg/mL on ice, following the manufacturer’s guidelines. For 3D contractility experiments (Fig.6), fluorescent deep- red 0.2 µm diameter spheres (Thermofisher, F8807) were added to the mixture at 1:100000 dilution to serve a s fiducial trac ers. Midgut segments from E11.5 embryos were cut in 2-3 pieces with micro- sci ssors and each segment was embedded in 1 mL of liquid collagen, that gelled after incubating at 37°C for 45 min. 2 mL of complemented medium with 10 ng/mL GDNF were then added. Collagen gel migration wa s assessed after culture for 3 days at 37°C, in a 5% CO 2 – 9 5 % a i r h u mi d i f i e d atmosphere. We mea sured the area S 1 occupied by the gut explant and the ENCC halo, the area S 2 occupied by the gut explant alone, and computed the average migration distance from the explan t a s /g3493 /g1845 /g2869/g2024⁄ /g3398 /g3493 /g1845 /g2870/g2024⁄ . This approach yielded an average radius which took into account halo as ymmetr ies . Im m unohis toche mis try For regi stration of CA with IHC (Video S2, Fig.2g,h), guts were fixed in 4% PFA in PBS for 20 min, wa s h e d 3 t i me s, t he n b lo ck e d an d p e r me a te d in 1 % B SA an d 0 . 1 % t r i t o n i n PB S fo r 1 h , i mme rs e d in 1:200 mouse monoclonal Anti-SOX10 (Sigma, AMAB91297) for 24h, washed 3 times, immersed in 1:500 anti-mouse AF647 secondary antibody with 1:500 βIII-tubulin FITC conjugated antibody (Abcam 224978) or 1:500 α-SMA Cy3 conjugated antibody (Sigma C6198) for 24h, washed 3 times in PBS, and imaged in the same location as during calcium imaging. Slight translation and rotation registration corrections between the CA time lapse movie and the IHC were performed manually using the GIMP software. For IHC of T-type Ca 2+ channels (Fig.3d-f), we use d explants from the 3D contractility assay (see below). Some of t hese explants t ouched the bottom of the Petr i-dis h and EN CCs can, in addition to t h e 3 D ha lo , a ls o mi gr at e o n t h e d is h s u r f a ce , as a 2D s he et , wh i c h f ac i l i t at e d i mag i n g. A f t er f ix at i o n, blocking and permeation, samples were incubated for 1 day in 1:200 mouse Anti-S OX10 and 1:100 rabbit anti CaV3.1 (Alomone Labs #ACC-021) or 1:100 rabbit anti CaV3.2 (Alomone Labs #ACC-025) or 1:100 rabbit anti Ca V3.3 (Alomone Labs #ACC-009) for 24h, washed 3 time s, incubated in secondary 1:500 anti-mouse Cy3 and 1:500 anti-rabbit AF647 antibody for anothe r day, wa she d and imaged. T- type Ca 2+ channel IHC was only succe ssful when performed on fixed cells that had been induced to emigrate from an explant, without embedding or cutting; we did not observe any specific signal when native E11.5 guts were fixed, cryo-sectioned and labeled, most probably owing to degradation of t he prot eins when per forming these steps. (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint 2D m igr a t io n a s s a y For the x60 experiments (Fig .S17), a E11.5 GCa MP midgut explant wa s cultured on a Petri dish with a culture surfac e-treated polymer coverslip bottom (Ibidi 81156), immobilized with a 500 µL meniscu s of 1% low-melting point agarose gel, and cultured one day in complemented medium with 10 ng/mL GDNF. Numerous EN CCs migr ated out on the covers lip and were imaged at x60 magnification at 1Hz, for 10 to 30 min. Cell nuclei appeared as dark spots surrounded by brighter cytopla sm. They were tracked using the deep-learning Segment Any thing Model 2 (SAM2) developed by Meta and implemented under QuPa th 70 . Nucleus boundary and centroid tracking were very precise (Video S11). Nucleus centroid position was plotted together with the GCaMP signal obtained from ROI intensity me asurements of the cytoplasm of the tracked cell. Kymographs were obtained with the ImageJ Reslic e func tion performed along a rectangle encompassing the cell trajectory. 3 D c ontr a ctility assay 3D contrac tility experiments were performed after 1-2 day of culture in collagen gel seeded with beads (see above), at x20 magnification, acquiring z-stacks in the bulk of the g el in GFP (cell s) and AF647 (beads) with a step of 2 µm over a thickness of 30-50 µm every minute for up to 120 min. Drugs were added during the time-lapse and mixed very carefully to not perturb bead positions. The z&t stacks of the cells (GFP) and beads (AF647) were first z max-projected to yield t-stacks (time- lapse). The bead t-stack was flat-field c orrec ted (Biovoxxel toolbox under Fiji), smoothed (Gaussian blur radius 2), thresholded (Li Auto Threshold) and the initial bead positi ons were measured using the ImageJ Analyze Particles tool. These initial coordinated were fed into the Tracker plugin (developed by our colleague O. Cardoso) to yield the coordinates of each bead at each time point. Inconsistent tracking was filtered ou t by applying conditions on the maximum rate of displac ement. The average displacement of bead s from their initial position towards the ENCC migration front wa s computed. S t a t istics and Re pr oduc ibility All sample numbers indicated in this report correspond to different embryos (guts, biological replicates). Except for ls/ls experiments which were performed on 2 litters, data for all other experiments was collected from at least 3 different litters and technically replicated at least 3 time s, i.e. they were performed on 3 different days with fresh samples following the sa me procedure each time. The sample size range is n =5-23, depending on the type and variability of experiments. p-value s reported correspond to statistical tests mentioned in the figure legends. Supplementary Figures Fig.S2,4-8,10-11,13 present the effects of the di fferent drugs applied in this study on all parameters (calcium activity CA, frequency F, width W, intensity ratio IR, ave rage calcium AC). Fig.S1: Methodology of Ca 2+ activity analysis Fig.S2: CA characteristics across stages at the migration front Fig.S3: Calcium events concentra te at the cecum and ileum anti-mesenteric border. Fig.S4: Role of Na + channels in CA Fig.S5: Endothelin 3 and receptor EDNRB are critical for CA (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Fig.S6: Effects of GDN F 10 ng/ mL on CA Fig.S7: Effect of DMSO vehicle on CA Fig.S8: Extracellula r and intracellular sources of Ca 2+ Fig.S9: Induction of CA by EDN3 10 nM in control, EDTA and NPPB conditions. Fig.S10: Ca 2+ channel dependence of CA Fig.S11: Cl - channel depe ndence of CA Fig.S12: 24 h kinetic s of Ca 2+ act ivity inhibitors Fig.S13: K + channel dependence of CA Fig.S14: Stimulation of T-type Ca 2+ channels, K + channels or purinergic receptors after EDNRB blockade does not allow to recover physi ological CA levels. Fig.S15: 2-APB 100 µM, NPPB 100 µM, DIDS 500 µM, 4 -AP 1 mM induced massive ENCC death after 24h application. Figure S16: Quantification of cell death in E11.5+1 guts in different culture conditions Figure S17: Ca 2+ flashes induced di sconti nuities in cell migration. Supplementary Videos Video S1: CA across stage s E10.5, E11.5, E12.5 at the migration front Video S2: Registration of CA with Sox10 (ENCCs) and Tuj1 (neurons), E11.5 ileum Video S3: Effect of EDN3 1 nM & 10 nM in E11.5 ilcc and effect of EDN3 10 nM in E12.5 ileum Video S4: Effect of EDNRB blocker BQ788 at E11.5 on CA, and long-term morphological effects Video S5: Effect of extracellular Ca 2+ removal by EDTA 2 mM, and subsequent stimulation by EDN3 Video S6: Effects of T-type Ca 2+ channel blocker Z944, CaV3.2 specific inhibitor a scorbic acid and Ca V3.1 & CaV3.3 agonist SAK3. Video S7: Effects of Cl - channel blocker NFA and NPPB. Video S8: Network-spanning rise in intracellular Ca 2+ induced by ATP e 1 mM Video S9: Mesenchyme backflow during ENCC invasion of the colon, E11.5 followed for 24 h. The video needs to be loaded in ImageJ and the time-cursor tracked fast-forward & backward repeatedly between frames 70 and 89, focusing on the areas indicated by the arrows. Video S10: Lamellipodium detachment after Ca 2+ trans ient Video S11: Deep learning assisted tracking of cell nucleus boundary and centroid of an ENCC chain with calcium activity. Video S12: ENCC 3D traction force on collagen gel is stimulated by EDN3 and relaxed by BQ788 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint Acknowledgments This research wa s funded by the Agence Nationale de la Recherche ANR GASTROMOVE - ANR-19- CE30-0016-01, by the Université de Paris IDEX Emergence en Recherche CHE VA19RDX-MEUP1, by the CNRS PEPS INSIS “COXHAM” grant, by the Labex “Who AM I ?” ANR-11-LABX-0071, and by the Imaging platform BioEmergences-IBiSA, ANR-10-INBS-04 and ANR-11-EQPX-0029. We thank Sylvie Dufour for providing the Ht-PA::Cre mouse line, Ko Sugarawa for help with the Segment Anything Model under QuPath, Vincent Fleury, Michael Levin, Alexandre Ayed, Nathalie Rouach, Isabelle Arnoux, Olivier Romito and Master 2 students of the Université Paris Cité Biomedical Engineering 2024-2025 program for thoughtful discussions and/or performing experiments together. Author contributions NRC led the projec t, obtained funding, performed experiments, analyzed data, synthesi zed data, wrote the dra ft and revised the paper; TS implemented new analysis methods and analyzed data; ZC performed experiments and analyzed data; NB, FG, MF , AEM, LC, MD, ILP, LZ performed experiments; LZ critically discussed the data; NB revised the draft. Data and code availability Essential data genera ted or analyzed during this study are included in the manuscript and supporting files . Source data as well as essential codes for calcium imaging analysis are provided with this paper. Other data are available from the corresponding author upon request. Competing interests The authors declare no competing intere sts.

1 . L e D o ua ri n, N. M. & D u p i n, E. The N eural Cre st, a Fourt h G erm L ay er of t h e V erte bra t e Em bryo: Si g n ificanc e i n C hor date Ev olut io n. Neural Crest Cells Evol. Dev. Dis. 3 –26 ( 20 14). d o i:1 0 .1 0 16/B 9 78 - 0- 12 -4 017 30 -6 . 000 01 -6 2 . Bo lan de , R. T he n e u roc rist op a t hi e s : u nif yin g conce pt o f dise a s e ari s ing in n e ura l c rest mal d e velo p m e nt. Hum. Pathol. 5 , 40 9 –4 29 ( 1974 ) . 3 . Ve g a-Lo p ez, G. A. , Cerrizu ela , S ., T r i b u lo , C. & A y b ar, M . J. N e u r o c rist op a t hi es: N e w i nsi g h t s 15 0 yea rs after the neura l c rest dis co v er y. D e v. Biol. 444 , S 110 – S 1 43 ( 20 1 8). 4 . W i lk in s , A . S . , W r an g ha m , R . W . & T e cu m s eh F i t c h, W . Th e ‘d om es t i c a ti o n syn d ro me ’ i n ma m m a ls: A u nified e x p l anati o n b ase d o n n eural crest c e l l be havi o r a n d g enet ics. Genetics 197 , 7 95– 80 8 ( 2 0 14). 5 . L i bra do, P. et al. A ncient g e n o m ic c h anges as soc i a te d w i t h d omes ticati o n of t he h o r se. Science (80-. ). 356 , 44 2 –4 45 ( 2017 ) . 6 . B a i l ey , C . M . , M o r ri so n , J . A . & Ku le s a , P . M . M e l a no m a r e viv e s a n e m b ry o n i c m ig r a ti o n pr o g r am to pro mot e pl as tic i t y an d in vas ion. Pigment Cell Melanom a Res. 25, 57 3–5 8 3 ( 2 0 12). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint 7 . Die n e r, J. & S o m m er, L. R e emerg e nc e of n e u ra l cr e s t s tem ce ll-l i ke s tat es in me l an oma d u ring di s e as e pro gre s si on a nd trea t m e n t. Stem Ce lls T ransl. Med. 10 , 522 – 533 ( 20 21 ) . 8. Y u , Q . et al. Sacral N eura l Cre st-I n depe nde nt O rig i n of t he En te r i c Ne r v o us Syst em i n Mo u se. Gastro ente ro logy 166 , 10 85– 10 9 9 ( 2 0 24). 9 . L a ke, J. I. & He uck e rot h, R . O. E n teric ner v o u s s ys te m deve l o p ment : M igrati on, d if fere n t ia tio n , an d dis ea s e . Am. J. P hysio l. - Ga strointest. Live r Physiol. 305 , 1– 24 ( 2 0 1 3 ). 10. M w i z e rw a , O. et al. G d nf is m i to ge nic, n e uro tro p h ic, a nd c he m o attract ive to ent eric n eural c re s t cel ls i n th e em bryo nic co lon. D ev. D yn. 240 , 1 40 2–1 4 11 ( 2 01 1). 11. Y oun g, H. M. et al . G DNF i s a che moat tracta nt f or e n t e ric n eural cel l s . Dev . Bio l . 229 , 503 –51 6 (2 001 ) . 12. A nd e r so n , R. B., Ne w g r e e n, D. F . & You n g , H. Neu ra l C r est a n d t h e D evelo p m e nt of t h e E nt e r i c Ne r vo us Sys tem. (20 1 3 ) . 1 3 . B ond u r and , N . , Du f ou r , S . & P i ng a u lt , V . Ne ws f r om th e e ndo th e l in - 3 / ED NRB si g na l ing pa th wa y : R ol e d uring ent eric n ervo us s yst em d evelo pm ent a n d inv olv e m ent in n e ural cr e s t-a s soci a t ed di sord ers. Dev. Biol . 444 , S 1 56 – S 169 (201 8 ) . 1 4 . Ka n a i , S . M . & Cl out hi e r , D . E . E n do th e l in si gn a lin g in de v e l op me nt . Dev. 150 , ( 2 0 23). 15. Ta ng, W. et al. E x o m e-W id e Ass ocia ti o n St ud y I den tifi ed Ne w Ris k L o c i f or H irsc h s p r un g’ s Di s e as e. Mol. Neurobiol. 54 , 1 77 7 –1 785 (20 17 ) . 16. Su nde la c ru z , S., L e v i n , M. & Ka plan , D. L. R o l e o f membra ne p o te n tia l i n t he r eg ulati on of c el l prol ife rat io n a nd differe nti a ti o n. Stem Ce ll Rev . Reports (2 0 09 ) . do i:1 0 .1 007 / s 12 015 -0 09 -9 08 0 -2 17. T orre, E. C., Bi c e r , M ., Co ttrel l , G. S ., W i d era, D. & Tamag nini, F . T im e - de pen dent re du c t ion of c alc i u m osc il l ati ons i n a d ipos e-d e r ived ste m cells d iff e ren tiat in g t ow a rds a dip o g e nic a n d os teo genic l inea g e. Biomolecules 11, 1 40 0 (202 1) . 18. Ya ng, M. & Brack e n b u ry , W . J. M e m brane p o te nti a l and can c e r pr ogres si on. F r ont ie rs i n Physio l ogy (2 0 13). doi:1 0 . 338 9/f ph y s .2 0 13. 001 8 5 1 9 . C a r e y , M . B . & M a t sumo to , S . G . S po nt a n eo us ca l c i um tr a n s ie n ts a r e r eq u i r ed f o r ne u ron a l diff ere n tia tio n o f m urine neural c re s t. D e v. B iol. 215 , 298 – 313 ( 19 99 ) . 20. Mor oku ma , J. et al. M o dula tio n of p otas si u m cha nnel f u nctio n co n fers a hy pe r prolif e r a t iv e i nvas i v e ph e no typ e on e m b ry on i c st e m c el l s. Proc. Na tl. Aca d. Sci. U. S. A. 105 , 1 66 0 8– 16 613 (20 08) . 21. Follm e r , M . L., I s n er, T ., Oz e kin, Y . H., L e v it t, C. & B a te s, E. A. Depo la riz ati o n i n d u c e s c a lc ium - d e p e n d e nt BM P4 releas e fro m m ous e e m br y o n ic pa l ate mes en c hym e . bi oRxiv 2 024. 0 6 . 1 1. 5 9 83 3 3 ( 2 0 24). do i :10 . 1 038 /s4 1 467 - 024 - 5 364 2 -2 22. Ferre ira, F ., M oreira, S. , Z hao , M . & B a rriga, E . H. S t re t c h-in d uced e n d o g e n o us e l ectric fiel d s dri v e directe d c ol l ecti ve cell mi g ra tio n i n vi vo. Nat. Mater. 2025 243 24 , 4 62– 470 (2 0 25 ) . 23. Hao , M. M. et a l. Spo nt ane ous c alciu m w aves in th e dev elop i n g ent eric n e r vo us s y s t em . Dev. B iol. 428 , 74 – 87 (2 017 ) . 2 4 . C h e v a li e r , N . R . et al. Ho w t iss u e m e c han i c a l pro p ert ies a ff ect e nt eric ne ural c re s t cel l m ig rati o n. Sci. Rep. 6 , 209 27 ( 2016 ) . 25. Barl o w , A., De Graaff, E. & P ac hn is, V. E nteric Nerv ous S y s tem Pr oge n it ors Are Co ordi nate l y Con troll ed by t he G P r ot e in - Coup l ed Re c e pto r E DN RB a nd th e R e ce p to r T y r os in e Kin a se R ET . N euron 40 , 90 5–9 16 (2 0 03). 26. L e ibl, M . a e t al. E x p r e s s ion of e n dot helin 3 b y me se n c hy ma l cell s of e mbry onic mo u s e ca ecum. Gut 44, 246 –25 2 (19 9 9) . 2 7 . A med zr ov i A g be si , R . J . , E l M e rh i e , A . , S p en c er , N. J . , H ib b e rd , T . & C hev a l i e r , N . R . T et r odot o x in - resis t a n t me c h a nos ensi ti vit y an d L-t y p e calci u m ch ann el-me di a te d s po nt a n e ou s cal c i um a c tivit y in (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint en teri c n e uro n s . Exp. Physiol. 109 , 1 5 45– 155 6 ( 2 0 24). 28. Hirs t, C. S . et al. Io n c han nel ex pres si on i n t he d e velo pin g e nteric n er v ous s yst em. PLoS O ne 10, (2 015 ) . 29. Nels o n, M . T . et al. Mol ecular m e c h a n ism s o f s ub t ype-s pecific i nhi biti o n of neur on a l T- t y p e c a lc ium ch an ne l s b y a s co r ba t e . J. Neurosc i . 27 , 12 57 7 – 12 583 (200 7 ) . 30. Ya bu ki , Y. et a l. P ha r m a c o log i ca l p r o pe r t i e s o f SA K3 , a no v e l T - ty p e vo lt a g e - ga t e d C a 2+ cha nn e l en hancer. N europharmacology 117 , 1 –13 (201 7 ) . 31. Ta k e nak a , T . et al . At ten uat ion of e nd ot he l i n e ffects b y a c hl ori d e c han nel i nhi bit or, i n dan y l ox ya c et i c acid. Am. J. Physi ol . 262 , ( 1 99 2). 32. W ray , S., Pre nder gast , C. & Ar rows mi th , S. C a l c i um- Acti v a t ed Chl ori d e Cha n n el s i n M yom et ria l and Vasc ular S m oot h M u s cle. Front. Physiol. 12 , ( 2 021 ). 33. Y oun g, H. M. et al . C olo nizi ng w hil e mi gratin g: h ow do i n divi dual e nt e r i c neural cr e s t c e lls b e h a v e ? BMC Biol . 12 , 2 3 ( 201 4). 34. W o o dwar d , M . N . , K e n n y , S. E., Vailla n t , C., L lo yd, D. a & E dg ar, D. H. T i m e - de pen den t eff ec ts of en dot he lin-3 on ent eric n e r vo us s yst em deve l o pm ent in a n or gan c ult ure m o d el of H irsc hs prun g’s dis ea s e . J. Pediatr. Surg. 35 , 25– 29 ( 20 0 0). 35. Fran c o, S. J. et al. Cal p a in - me d i at ed prot e o l y sis of tal in re gula te s a d h es io n d y namics. Nat. Cel l Biol. 6 , 977 –98 3 (20 0 4) . 3 6 . B a r ton , M . & Y an a g i sa w a , M . End o th e li n: 3 0 y ea r s f r om d is c ov e ry t o th e r ap y . Hy pe rte nsi on 74 , 1 2 32– 126 5 (2 01 9) . 37. Su djarw o, S . A., H ori, M . & Ka ra ki, H. E ffect o f end o theli n - 3 o n c y t os olic calci um l eve l i n vas cula r en dot he lium a n d o n s m o o t h musc le c on tract io n. Eur. J. Pharmacol . 229 , 13 7–1 4 2 ( 1 9 92). 38. Ha ynes, W . G., S t ra c ha n, F. E. & W eb b, D . J. En do t he li n ETA a nd E TB rece p t o rs ca u s e vas oc onst ri c ti on of h uman re s is tanc e a nd ca pa c i t ance v es s e l s i n viv o. Circulation 92, 3 57– 363 (1 99 5 ) . 39. L a ne, P. W . A ss o c ia ti o n of m e g a c olo n wit h t w o r e c es si v e s p o t t i n g ge n es i n t he mo use. J. Hered. 57, 29– 31 (196 6 ) . 40. Ba yna sh, A. G. et al. I nt e r a c ti o n o f e n dot heli n- 3 w i t h en dot h el i n- B r e c ept or i s es se n t ial f o r de v e lop ment of epi d e rma l melan o c y te s a nd ent e ric ne u r ons. Cell 79, 12 7 7 – 1 285 (1 9 94). 41. Gariepy, C. E ., C a s s , D . T. & Ya na g is aw a , M . N ull m u ta t i o n o f e nd ot helin re c ept or t y p e B ge n e i n s pot ti n g l et ha l ra t s ca u s es a ga ngli onic m egacol on a nd w hit e c oat c ol or. Proc. Na tl. Ac ad. Sci. U. S. A. 93 , 867 –87 2 (19 9 6) . 42. L a ha v , R. et a l. E ndo th e lin 3 s e l e c ti v e ly p ro mo t e s s u rv iv al a nd p ro l i fe r a t ion o f n e u r al cr e st- d e r iv e d g l i a l an d mela n o c y t ic prec urs o rs in v i tro. Proc . Na tl. Aca d. Sc i. U. S. A. 95 , 1421 4–1 421 9 ( 1 998 ) . 43. Natara j a n, D., Gri gorio u, M ., M arcos- G u t ierrez, C. V, At k i ns, C . & P a c hni s, V. Mu lti pot en tial p r oge nit ors of th e mamm a li an e nteric ner v o us s ys te m c apabl e of c olo nisi n g aga ngli o n ic bo wel i n o r ga n c u lt ure. Developm ent 126 , 15 7–1 6 8 ( 1 9 99). 44. Ch eval i er, N. P hys i c a l or g a n oge n es is of t he g u t . Devel opment 149 , d ev 20 0 765 ( 202 2 ) . 45. Hear n, C. J. , You n g , H. M . , Ciam p o li, D., Lo max , A. E. G. & N ew gre e n, D . Cat enar y c ult ures o f embry o n ic gas tro intes t inal tract s upp ort o r g a n m orp ho g e n esis, m o tilit y , n e ura l c re s t cel l migra tion, an d cel l diff ere n tia tio n. Dev. D yn. 214 , 239 –24 7 ( 1 999). 46. Nag y , N. & G o l dst ei n , A. M . E nd ot heli n- 3 re g u la tes ne ura l crest c e l l p r oliferati on an d diff e re nti atio n in t he h indg ut e nt e r i c n e rv ous sy st em . Dev. B iol. 293 , 2 03– 21 7 (2 0 06). 47. Brea u , M. a et al. L a c k of bet a1 i nt egrins in ent eric n eu ra l crest cells l e a d s t o a H i rs c hs pru n g -li k e p hen ot y p e. Devel opmen t 133 , 1 72 5–1 7 34 ( 2 00 6). (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint 48. Gazq uez , E. et al. En dot h el i n- 3 s tim ulates c e ll a dh e s io n and c o operat e s w it h β 1-int e g r i n s d uri n g ent eric n e rv ous sy st em on tog en e s i s. Sci. Rep. 6 , 3787 7 (2 016 ) . 49. Ch eval i er, N. R. et al. A n e ur al c rest cell is otr o pi c -to- n e m a t i c p has e tr ansit io n i n t he d e v e l o ping ma m m al i an gut . C o mmu n. Biol. 4 , 770 ( 20 21) . 50. Simps o n, M . J., Z h a n g, D. C., M a r i a ni, M ., L an dm an, K. a . & N ew green, D. F. Ce ll pr oli fer ati on drives ne ura l crest c e l l in v a si o n of th e i n tes ti ne . D ev. B iol. 302 , 5 53 – 56 8 (200 7) . 51. Halak a , M. et al . D ifferenc e s i n E nd ot helin B Rece pt or I s oforms Expre ssi o n an d F u n c ti o n i n Breas t Ca n c er Ce ll s . J. C ancer 11 , 2 688 ( 2 0 20). 52. W h ite , C. D . & Sa c k s , D. B. Re gul a ti o n of M AP ki nase si gna l i ng b y c a lci um. Methods Mol. Biol. 661 , 151 – 165 (20 10 ) . 53. Ghi go, A., L a ffar gue, M ., Li, M . & Hirsc h, E. PI 3 K a n d C a lci um Si gna l i n g i n C ar di ova s c ula r Di seas e. Circ. Res. 121 , 282 –29 2 (2 017 ) . 54. Got o, A. et al. G DNF a n d e ndot h el i n 3 re g u lat e m igrati on of enteric neural c re s t- de r ived ce lls v i a pro tei n kin ase A a nd R a c 1 . J. N eurosci. 33, 49 01 – 12 (2 013 ) . 55. Me hrk e , G. , Zon g, X ., Fl o c k erz i , V. & Hof man n, F . The Ca(++)-c h a n n e l b l ocker R o 40- 5 9 67 b l ocks diff ere n tl y T- typ e an d L-ty pe Ca+ + c han nels - P u bM ed. J Pharmac ol Exp Ther. 271 , 1 483 –8 (1 994 ) . 56. Wa n g , R. & L e w i n , G. R. T h e C a v3. 2 T-t y pe cal cium cha n n e l r eg ul a tes t emp oral c o di n g i n m ous e m e c ha n o r e c ep to rs . J. Phy si ol . 589 , 2 2 29– 224 3 (2 011 ) . 57. Ch en, C. C. et a l. Ab no r ma l Co r on a ry F un c t ion i n Mi c e D e f i ci ent i n α1 H T - typ e C a 2+ Cha nne l s. Scienc e (8 0-. ). 302 , 14 16– 141 8 (20 0 3 ). 58. Al p dog an, S ., Cl e mens, R., Hesc h el e r , J., Neu ma i er, F . & S c hne ider, T . N o n-M end e li an i nh erita n c e du r i n g in b r e ed ing o f C av3 .2 a nd C av 2 .3 de f i cie nt m i ce . Sci. R ep. 10 , ( 2 02 0). 59. Ba gh dadi, M . B. et al . PI E Z O-d epe nde nt m echa nose ns i ng is e s se n t ial for i ntes ti n a l s tem c ell f at e decisi o n a nd ma i nt e n a nce. Sci enc e ( 8 0- . ). 386 , (2 0 24). 60. Ber ge ro n, K. F . et a l. M ale - bi a s ed aga nglio nic meg a c olo n in t he Ta s hT mo use l i ne du e to p e rt u r bat ion of s i le ncer ele m e nts i n a larg e gene des ert of c hr o m os om e 1 0 . PLoS Gen et. 11, (20 1 5 ). 61. Kr ug e r, G. M . et al. T e mporal ly Di s ti n c t R e q uirements f or En dot h el i n Recep tor B i n t he Ge n era ti o n an d M i gra ti o n of G u t N e ura l C re st St em C e lls. Neu r on 40 , 91 7–9 29 ( 2 003 ). 62. Sh i m, S . et a l. C al c i u m d y n a m ics a t t h e neural c e ll prim ary c i li u m re gula te He d g e ho g s i g n alin g – de pen den t neuro g e n esi s i n t he em b r yon i c n eura l t u be . Proc. N atl. Acad. Sci. U . S. A. 120 , (2 0 23). 63. Jo h nse n , J. I ., D y b erg, C. & W i c ks trö m , M . Ne uro b l ast o ma— A N eural Cres t D e ri ved Embr y onal Malig nanc y. Front. Mol. N eurosci. 12, 9 ( 2 01 9 ). 64. Barcel ó , C. et al . T- Ty pe C a lciu m C han nels : A Pot enti al N o vel T ar get i n M elan oma. Cancers (Basel). 12, 391 (20 20 ) . 65. L a n ge , I., K os ter, J . & Ko omo a , D. L. T . C a lci u m s i g n a li n g re gul a t es f u n dament al proc ess e s i nv o l v e d in Neur ob l as t om a pro g r e s s ion. Cell Cal cium 82, (20 1 9 ). 66. R osa n ò , L., Spi nella , F . & Bag nat o, A. E n dot h el i n 1 in c anc e r: Biolo g i c a l i mplica tion s and t he r ape utic op po rtu n i ti e s. N at. Rev. Cancer 13 , 6 3 7 – 6 51 (2 01 3). 6 7 . C h e v a li e r , N . R . et al. Cal c i u m w ave dy namics i n t h e em bryo nic m o us e gut m e s enc hy m e: imp a c t on s moo t h mu sc l e di ff ere ntiati o n . C om mun. Bio l. 7 , 12 77 ( 2 024 ). 68. Sta n c hin a, L . et al. Int eractio n s b e tw een S ox 1 0, E dn 3 an d Ed nr b d uring ent eric n e r vo us s ys te m a nd m e l a no cy te d e v e l opm en t . D e v . B io l. 295, 232 – 2 4 9 ( 2 0 06). 69. Rice, J. et al. T r a n s genic resc u e of a gan g li o no sis an d pieba l d is m in leth al s p o tte d mice. Dev. Dyn. 217 , (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint 120 –13 2 (20 0 0) . 70. Su gawara, K. Tra inin g deep learnin g m o de ls f or cell im age s eg m e nt atio n wit h s p ars e a nno tatio ns. bioRxiv 2 0 23 . 06 .1 3 . 5 4478 6 (2 023 ) . do i :10 . 110 1 / 202 3 . 06 .13 . 54 47 86 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted October 24, 2025. ; https://doi.org/10.1101/2025.10.23.684245doi: bioRxiv preprint