Injury-induced cooperation of InhibinβA and JunB is essential for cell proliferation in Xenopus tadpole tail regeneration

In animal species that have the capability of regenerating tissues and limbs, cell proliferation is enhanced after wound healing and is essential for the reconstruction of injured tissue. Although the ability to induce cell proliferation is a common feature of such species, the molecular mechanisms that regulate the transition from wound healing to regenerative cell proliferation remain unclear. Here, we show that upon injury, InhibinβA and JunB cooperatively function for this transition during Xenopus tadpole tail regeneration. We found that the expression of inhibin subunit beta A (inhba) and junB proto-oncogene (junb) is induced by injury-activated TGF-β/Smad and MEK/ERK signaling in regenerating tails. Similarly to junb knockout (KO) tadpoles, inhba KO tadpoles show a delay in tail regeneration, and inhba/junb double KO (DKO) tadpoles exhibit severe impairment of tail regeneration compared with either inhba KO or junb KO tadpoles. Importantly, this impairment is associated with a significant reduction of cell proliferation in regenerating tissue. Moreover, JunB regulates tail regeneration via FGF signaling, while InhibinβA likely acts through different mechanisms. These results demonstrate that the cooperation of injury-induced InhibinβA and JunB is critical for regenerative cell proliferation, which is necessary for re-outgrowth of regenerating Xenopus tadpole tails.


The expression of inhba and junb is induced by injury-activated signaling
Although injury-induced JunB is important for regenerative outgrowth/cell proliferation in the Xenopus tadpole tail, a CRISPR/Cas9-mediated junb KO in F0 or compound heterozygous tadpoles is not sufficient to block tail regeneration 15 .This prompted us to search for regulators that cooperate with JunB for regenerative outgrowth/ cell proliferation.Inhibition of TGF-β signaling in the regenerating tail after wound healing impairs cell proliferation in the regeneration bud 4 ; thus, in addition to a TGF-β/Activin family ligand responsible for the wound healing phase, other TGF-β/Activin family ligands possibly work together with JunB to promote regeneration during the proliferation phase.To identify candidate TGF-β/Activin family ligands that cooperate with JunB for regenerative cell proliferation, we examined the levels of expression of ligands (Tgfβs, Inhibins, Nodals, Gdfs, and Myostatins) during X. tropicalis tadpole tail regeneration using published RNA-seq data 16 (Supplementary Figure 1).Three genes belonging to the TGF-β/Activin family, namely inhba, tgfb1, and tgfb2, were identified as being highly expressed during the wound healing phase [~ 6 h post-amputation (hpa)] as is junb.However, only expression of inhba is upregulated after injury (Supplementary Figure 2).Since Smad2/3 and ERK are activated immediately after tail amputation 4,8 , we analyzed whether the injury-mediated increase in expression of inhba and junb is regulated by TGF-β/Smad and MEK/ERK signaling.As shown in Fig. 1, expression of inhba and junb during wound healing was downregulated by both SB-505124 (SB, TGF-β receptor inhibitor) and PD0325901 (PD, MEK/ERK inhibitor).These results suggest that, in addition to JunB, InhibinβA may also function downstream of injury-activated signaling to promote cell proliferation in Xenopus tail regeneration.
As injury-activated TGF-β/Smad and MEK/ERK signaling are important for the induction of inhba and junb expression, we investigated the relationship between the activation of Smad2/3 and ERK after tail amputation.In contrast to tgfb1 KO tadpoles 17 , phosphorylated Smad2/3 (pSmad2/3) expression was not downregulated by PD treatment at 2 hpa (Supplementary Figure 3).It has been shown previously that ERK phosphorylation does not require TGF-β/Smad signaling during wound healing after amputation of the Xenopus tail 8 .Taken together, these observations suggest that activation of TGF-β/Smad signaling and of MEK/ERK signaling are unlikely

InhibinβA is required for regenerative outgrowth/cell proliferation in cooperation with JunB
Since inhba was expressed in response to injury-activated signaling during tail regeneration, we investigated whether InhibinβA is essential for tail regeneration.We generated F0 inhba KO tadpoles using the CRISPR/Cas9 system: two different sgRNAs (sg 1 and sg 2) were designed using the genomic sequence corresponding to the pro-domain of InhibinβA; each sgRNA was injected into fertilized eggs (Fig. 3a).Both sgRNAs induced mutations in the inhba locus at moderate frequencies as determined by a T7 endonuclease I (T7E1) assay (Supplementary Figure 5).At 72 hpa, inhba KO tadpoles injected with each sgRNA showed a delay in tail regeneration, and the lengths of regenerating tails in inhba KO tadpoles were reduced compared to control tadpoles (tyrosinase (tyr) KO) (Fig. 3b and c).In addition, when classified according to the lengths of regenerating tails, a high proportion of inhba KO tadpoles showed delayed tail regeneration phenotypes (sg 1, 64.9%; sg 2, 74.0%) (Fig. 3d).The similarity of phenotypes of tadpoles injected with either sg 1 or sg 2 suggests that the retardation of tail regeneration in inhba KO tadpoles resulted from the specific inhibition of InhibinβA function.To increase the mutation efficiency, we injected a combination of sg 1 and sg 2 into fertilized eggs and found that 91.6% of alleles contained out-of-frame mutations with stop codons around the sgRNA target sites; 6.7% of alleles had in-frame mutations.In total, 98.3% of alleles had mutations in the inhba locus (Supplementary Figure 6).Moreover, the combination of sgRNAs increased the proportion of tadpoles displaying delayed tail regeneration (sg 1 + sg 2, 83.6%) compared with either sg 1 or sg 2 alone (Fig. 3d).Therefore, we used the combination of sg 1 and sg 2 in subsequent experiments.The delay in tail regeneration in inhba KO and junb KO tadpoles was specific as there was no significant delay in developing tadpole tails before tail amputation at stage 41/42; the delay in tail regeneration was observed even at 10 days post-amputation (dpa) (Supplementary Figure 7).In addition, the survival rate of these KO tadpoles was similar to that of the tyr KO control (Supplementary Figure 8).Finally, to further confirm the specificity of the phenotype observed in inhba KO, a rescue experiment was performed by overexpressing InhibinβA (Supplementary Figures 9 and 10).Overexpression of InhibinβA in inhba KO tadpoles partially but significantly rescued the shortened regenerating tail length and markedly reduced the proportion of tadpoles showing delayed tail regeneration.Taken together, we conclude that InhibinβA is required for regenerative outgrowth of the Xenopus tadpole tail.www.nature.com/scientificreports/Next, we examined whether InhibinβA and JunB cooperate for regenerative outgrowth/cell proliferation.Double KO (DKO) of inhba and junb was performed by combining two sets of sgRNAs (two each for inhba and junb).In the DKO tadpoles, the mutation efficiencies at inhba and junb loci were comparable to those observed in the respective single KO tadpoles based on the T7E1 assay.As shown in Fig. 4a, b, the lengths of regenerating tails were shorter in inhba/junb DKO tadpoles than in inhba KO or junb KO tadpoles.Furthermore, whole-mount immunostaining of phosphorylated Histone H3 (pH3) in regenerating cells at 36 hpa revealed that the number of proliferating cells was significantly lower in inhba/junb DKO tadpoles than in inhba KO or junb KO tadpoles (Fig. 4c, d).Notably, there were no significant differences in the rates of cell proliferation in the proximal (headward) tail region of amputation sites among experimental groups (Supplementary Figure 11; see also Fig. 5e, f described below), suggesting that InhibinβA and JunB function in regenerating tissue to regulate regenerative cell proliferation.These findings provide strong evidence that InhibinβA and JunB cooperate for regenerative outgrowth/cell proliferation in the Xenopus tadpole tail.

JunB regulates regenerative outgrowth/cell proliferation through FGF signaling
To elucidate the molecular mechanisms by which InhibinβA and JunB initiate regeneration processes, we searched for candidate downstream factors of InhibinβA and JunB during Xenopus tail regeneration.Inhibition of FGF signaling is known to affect Xenopus tail regeneration 9 .The FGF ligand fgf20 is highly upregulated after wound healing, and morpholino-mediated knockdown of Fgf20 causes a delay in tail regeneration 18 .Moreover, regenerative outgrowth/cell proliferation of the zebrafish caudal fin is inhibited in fgf20a mutants 19 .We therefore examined whether InhibinβA and JunB regulate tail regeneration through FGF signaling.First, expression of fgf20 and tbxt (also known as brachyury), a downstream effector of FGF signaling, were analyzed in regenerating cells of inhba KO and junb KO tadpoles at 36 hpa.A quantitative reverse transcription-PCR (qRT-PCR) analysis  showed that fgf20 and tbxt transcripts were downregulated to a significant extent in junb KO tadpoles (Fig. 5a), suggesting that JunB, but not InhibinβA, regulates the expression of fgf20 and tbxt at the time when regenerating cells were actively proliferating.A rescue experiment was performed to determine whether JunB facilitates tail regeneration through FGF signaling.We overexpressed Fgf20 or JunB in junb KO tadpoles and measured the lengths of regenerating tails in the tadpoles.Our results indicated that Fgf20 partially but significantly rescued the delay in tail regeneration of junb KO tadpoles to the same extent as JunB (Fig. 5b, c; Supplementary Figure 10).Moreover, when classified according to the lengths of regenerating tails, the proportion of junb KO tadpoles with severely delayed tail regeneration was greatly reduced by Fgf20 overexpression (Fig. 5d).Next, we analyzed cell proliferation in these tadpoles at 36 hpa.As shown in Fig. 5e, f, the reduction in proliferating cells in junb KO tadpoles was rescued by overexpression of Fgf20 or JunB.Together with the finding of downregulated expression of fgf20 and tbxt in junb KO tadpoles (described above), these data support the conclusion that JunB acts, at least in part, through FGF signaling to promote regenerative outgrowth/cell proliferation of the Xenopus tadpole tail.
FGF signaling in the proliferation phase is important for regenerative outgrowth/cell proliferation FGF signaling is well known to be essential for tissue/organ regeneration in regenerative animals 9,12,14,[18][19][20] .
Although inhibition of FGF signaling causes a delay in Xenopus tail regeneration 9 , the role of FGF signaling in regenerative cell proliferation is still unclear.In addition, the expression of fgf20, which functions downstream of injury-induced JunB, increases as outgrowth/cell proliferation progresses in the regenerating Xenopus tail 18 , and inhibition of FGF signaling does not affect wound healing 9 .Therefore, we speculated that FGF signaling in the proliferation phase might be required for outgrowth/cell proliferation in Xenopus tail regeneration.To suppress FGF signaling specifically in the proliferation phase, tail-amputated tadpoles were treated with SU5402 (SU, FGF receptor inhibitor) from 24 hpa (the time when regenerative outgrowth begins) and the lengths of regenerating tails were measured at 72 hpa (Fig. 6a).Consistent with a previous report 9 , inhibition of FGF signaling in the proliferation phase caused a delay in tail regeneration compared with DMSO-treated tadpoles (Fig. 6b, c).We then analyzed cell proliferation in SU-treated tadpoles and found that the number of proliferating cells in the www.nature.com/scientificreports/regenerating tail was considerably reduced by SU (Fig. 6d).Thus, our results demonstrate that FGF signaling in the proliferation phase is essential for regenerative outgrowth/cell proliferation of the Xenopus tadpole tail.Given the fact that JunB, but not InhibinβA, regulates tail regeneration via Fgf20 (Fig. 5), we conclude that FGF signaling downstream of JunB functions in parallel with InhibinβA signaling in the proliferation phase (Fig. 7; see Discussion).

Discussion
Injury-induced cell proliferation is a shared feature in animal species that have the ability to regenerate tissues.However, it is still not clear what factors are induced in response to injury to promote cell proliferation during tissue regeneration.Here, we show that inhba and junb are induced through injury-activated TGF-β/Smad and MEK/ERK signaling during the wound healing phase.Moreover, our study suggests the cooperative function of InhibinβA and JunB in injury-induced cell proliferation and tail regeneration.We propose that InhibinβA and JunB are important factors that mediate the transition from wound healing to regenerative cell proliferation during Xenopus tail regeneration (Fig. 7).The significance and possible mechanisms of the cooperative regulation of cell proliferation by InhibinβA and JunB are discussed in detail below.Upon injury, multiple cell signaling pathways (TGF-β/Smad, ERK, and ROS) are known to be activated in the wound healing phase of the regenerating Xenopus tail 4,8,18 .However, the molecular mechanisms by which these injury-activated signals control regenerative outgrowth/cell proliferation through the regulation of gene expression are not well understood.We found that at the beginning of wound healing (~ 2 hpa), the phosphorylation of Smad2/3 was not dependent on MEK/ERK signaling (Supplementary Figure 3).Conversely, a previous report showed that the phosphorylation of ERK is not affected by inhibition of TGF-β/Smad signaling 8 .These observations imply that TGF-β/Smad and MEK/ERK signaling are independently activated by injury and coordinately regulate the expression of inhba and junb during wound healing (Figs. 1 and 7).While the upstream factors that activate MEK/ERK have not been elucidated, Tgfβ1 is required for the injury-induced phosphorylation of Smad2/3 (Supplementary Figure 3) 17 .Interestingly, tgfb1 is already highly expressed in the uncut Xenopus tail 4,17 ; after tail amputation, Tgfβ1 stored in the extracellular matrix is expected to induce expression of downstream target genes, including inhba and junb, via the activation of Smad2/3.We postulate that MEK/ERK signaling may help to refine the expression of inhba and junb induced by the Tgfβ1-Smad2/3 pathway.The injury-induced burst of inhba and junb expression is fast and takes place within a few hours (~ 6 hpa).Therefore, it is possible that the coordination of TGF-β/Smad and MEK/ERK signaling is essential for a sharp and sustained increase in inhba and junb expression after tail amputation, ultimately contributing to the successful recovery from injury and the regeneration of damaged tissue.
The induction of inhba expression by the Tgfβ1-Smad2/3 pathway indicates the amplification of TGF-β/Smad signaling during Xenopus tail regeneration (Fig. 7).This is consistent with the observation that expression of TGF-β family ligands (inhba and gdf11) is sequentially induced during tail regeneration 4 .As described above, www.nature.com/scientificreports/ the activation of Smad2/3 is not suppressed by inhibition of MEK/ERK signaling at the beginning of wound healing.In contrast, at the end of wound healing, the phosphorylation of Smad2/3 is reduced by MEK/ERK inhibitor 8 .We speculate that the reduction of activated Smad2/3 at the late time point of wound healing may be due to the disruption of the amplification system of TGF-β/Smad signaling by downregulation of inhba after MEK/ERK inhibition (Fig. 1a).Furthermore, the core components of TGF-β signaling (ligands, receptors, and their transcriptional target genes) were found to be co-expressed during axolotl limb regeneration 21 .Given that TGF-β signaling plays critical roles in multiple steps of regeneration processes 4,[22][23][24] , it is tempting to suggest that the amplification of TGF-β/Smad signaling is conserved among animal species with a high regenerative capacity and is fundamental for maintaining TGF-β signaling throughout regeneration.Unlike urodeles, regenerating axial tissues (spinal cord, notochord, and somite) in the Xenopus tail have been shown to originate from the respective differentiated tissues in the tail stump, and metaplasia of tissue progenitors is not observed 25 .In addition, dedifferentiation phenotypes are not found and regenerating myofibers are derived from skeletal muscle tissue stem cells (satellite cells).These observations indicate that tissue renewal mediated by the induction of cell proliferation in differentiated cells is a critical step of Xenopus tail regeneration.Therefore, while InhibinβA and JunB could regulate differentiation directly or indirectly at some stages of regeneration, it is plausible that the cooperative function of InhibinβA and JunB in regenerative cell proliferation is important for successful tissue recovery.It will be interesting to analyze the contribution of this cooperation in other animal species with different regenerative capacities and cellular plasticities, particularly in urodeles.
During spinal cord regeneration in the axolotl, cell proliferation is accelerated by the progression of the cell cycle from the G0/G1 to S phase 26 , indicating that the release from the G0/G1 phase is a crucial step in regenerative cell proliferation.Notably, it has been shown that inhibition of InhibinβA or JunB in cultured cell lines increases the proportion of cells in the G0/G1 phase 27,28 .Moreover, in the present study, impairment of regenerative cell proliferation in inhba/junb DKO was greater than in either inhba KO or junb KO (Fig. 4d).Therefore, in response to tail amputation, InhibinβA and JunB may work together in Xenopus to promote cell cycle re-entry by facilitating the release from the G0/G1 phase.It has been reported that both Smad2/3 (signal transducers of InhibinβA) and JunB are essential for the transcription of cell cycle-related genes 29,30 and that Smad3 regulates target gene expression by binding to JunB [31][32][33][34] .These findings and our results suggest that the cooperative function of InhibinβA and JunB in regenerative cell proliferation might be mediated by a transcriptional complex composed of Smad2/3 and JunB during Xenopus tail regeneration.In this case, InhibinβA-activated Smad2/3 and JunB directly converge in the cell cycle regulatory system.In parallel with this potential mechanism, InhibinβA may also interact with downstream effectors of JunB in the regulation of cell proliferation.Intriguingly, we found that Fgf20 acted downstream of JunB in Xenopus tail regeneration (Fig. 5).In addition, Activin A, a homodimer of InhibinβA, requires FGF signaling in mesoderm formation during Xenopus embryogenesis and in the endodermal differentiation of mouse embryogenic stem cells [35][36][37] .As the expression of fgf20 and its target gene tbxt was not reduced in inhba KO tadpoles when regenerating cells were actively proliferating (Fig. 5a), it is possible that InhibinβA functions with FGF signaling downstream of tbxt expression to regulate tail regeneration.Tbxt is known to associate with Smad2/3, and the genomic binding sites of Tbxt overlap with those of Smad2/3 in differentiating human embryonic stem cells 38 .An alternative but not mutually exclusive possibility is that InhibinβA affects other signaling pathways involved in regeneration processes.During regeneration of the zebrafish fin and Xenopus tadpole tail, Activin/TGF-β signaling regulates the expression of the BMP target gene msx1 4,24 .Since BMP signaling is required for Xenopus tail regeneration 10 , InhibinβA might promote tail regeneration in cooperation with JunB by modulating the BMP pathway.Overall, the collaborative action of InhibinβA and JunB in regenerative outgrowth/cell proliferation may be mechanistically important for the robust patterning of regenerating tissue; this possibility will be investigated in future studies.
In conclusion, we showed that inhba and junb are coordinately induced upon tail amputation through TGF-β/ Smad and MEK/ERK signaling, and that injury-induced InhibinβA and JunB cooperate to drive regenerative outgrowth/cell proliferation in the Xenopus tadpole tail.These findings provide valuable insights into the molecular mechanisms that mediate the transition from wound healing to regenerative cell proliferation during tissue regeneration.

Animals, microinjection, and regeneration assay
All experiments were conducted in accordance with ARRIVE guidelines and with the Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology of Japan, and were approved by the Hiroshima University Animal Research Committee (Permit Number: G19-7-2).X. tropicalis embryos were cultured in 0.1X Marc's Modified Ringer solution (MMR) at 26 °C.Tadpoles at stage 41/42 were anesthetized in 0.05% MS-222/0.01XMMR and subjected to tail amputation at the mid-point (50%) of tail length using a surgical knife.The removal of as much as 75% of tail length does not affect tail regeneration or the survival of tadpoles 39 .The amputated tadpoles were maintained until 72 hpa or 10 dpa in tap water.For inhibitor experiments, tadpoles were treated with 25 μM SB-505124 (Cayman Chemical) or 0.25 μM PD0325901 (Cayman Chemical) from 1 h before tail amputation, or with 15 μM SU5402 (Sigma) from 24 hpa.

Cloning of X. tropicalis inhba and fgf20
Full-length cDNAs for inhba (GenBank Accession number XM_002933406.3)and fgf20 (GenBank Accession number NM_001143927.1)were PCR-amplified using the following primers: inhba, forward 5′-CGG GAT CCC ACC TGG TGA CAG GAT GC-3′ and reverse 5′-GCT CTA GAA ATT GCT GCA GGC TGG TAA C-3′; fgf20, forward 5′-CGG GAT CCC TTT TGG GGA TTT TGG GAC T-3′ and reverse 5′-AAG GCC TGC ACT GGG CRISPR/Cas9-mediated KO, rescue experiment, and genotyping CRISPR/Cas9-mediated mutagenesis was carried out as previously described 15,40 .For the KO experiment, 1000 pg of sgRNA was used.In the DKO experiment, 500 pg each of the two different sgRNAs for inhba and junb were used; the total amount of sgRNAs was 2000 pg.The injection solution was prepared by mixing sgRNAs and 1 ng of Cas9 protein (Integrated DNA Technologies) per 1000 pg of sgRNAs.For the rescue experiment, the capped junb, inhba, and fgf20 mRNAs were synthesized from pDH105-junb 41 , pDH105-inhba, and pDH105-fgf20, respectively, using the SP6 transcription kit (Invitrogen).mRNA was mixed with sgRNAs and Cas9 protein, and the mixture was injected into the fertilized eggs.The sequences of sgRNAs for tyr, junb, and tgfb1 were described in previous studies 15,17,40 , and primers for generating inhba sgRNAs were as follows: sg 1, 5′-ATT TAG GTG ACA CTA TAG GCC CCA ACT CCA GGA TCT GGT TTT AGA GCT AGA AAT AGC AAG-3′; sg 2, 5′-ATT TAG GTG ACA CTA TAG GCC ATG TCA CTC TGA GAA CGT TTT AGA GCT AGA AAT AGC AAG-3′.Genomic DNA was extracted from inhba KO tadpoles at 72 hpa and purified by the GenElute mammalian genomic DNA miniprep kit (Sigma).PCR amplification of the genomic region of inhba was performed using Q5 High Fidelity DNA polymerase (New England Biolabs).The PCR products were treated with T4 polynucleotide kinase (New England Biolabs) and cloned into the pUC57 vector, which was digested by EcoRV and dephosphorylated with Quick CIP (New England Biolabs).To determine the mutation types, single colonies were subjected to Sanger sequencing.T7E1 (New England Biolabs) assay was performed as previously described 15 .The following primers were used for PCR and sequencing: inhba, forward 5′-ACA GCC ACA AAT ACC CAC AG-3′ and reverse 5′-AAG GAG CCA GTG AAG CTT TG-3′.

Whole-mount in situ hybridization and whole-mount immunostaining
Whole-mount in situ hybridization and cryosectioning of hybridized tadpoles were performed as previously described [42][43][44] .Tadpoles were fixed in MEMFA solution for 2 h at room temperature and stored in 100% ethanol at − 20 °C.For anti-sense and sense probes, pDH105-inhba was linearized by BamHI and XbaI and subjected to in vitro transcription by T3 and SP6 RNA polymerases (Roche), respectively.For the whole-mount immunostaining, tadpoles were fixed in MEMFA solution for 30 min at room temperature and stored in 100% methanol at − 20 °C.Whole-mount immunostaining was carried out following our standard protocol for pH3 staining 15 and with a minor modification for pSmad2/3 staining 17 .The following antibodies were used at 1:500 dilution: anti-pH3 antibody (Upstate Biotechnology), anti-pSmad2/3 antibody (Cell Signaling Technology), Alexa Fluor 488 goat anti-rabbit antibody (Molecular Probes, Thermo Fisher Scientific), and Alexa Fluor 488 goat anti-mouse antibody (Molecular Probes, Thermo Fisher Scientific).

Microscopy, quantification, and statistical analysis
Fluorescent images were obtained with Axio Zoom.V16 (ZEISS), and the fluorescence intensity was quantified with ZEN blue software (ZEISS).Images of cryosections were captured using an Axio Observer.Z1 (ZEISS).Quantification of pSmad2/3 staining was performed as described previously 17 .The length of the regenerating tail was measured with cellSens Standard imaging software (Olympus) or ImageJ software (National Institutes of Health, USA).On the basis of the lengths of regenerating tails at 72 hpa, tadpoles were classified as showing normal regeneration (more than 80% of the average length of tyr KO), weakly delayed regeneration (60-80% of the average length of tyr KO), or severely delayed regeneration (less than 60% of the average length of tyr KO).The number of pH3-positive cells was manually counted in the regenerating tail, excluding the fin.Statistical analysis was performed by Student's t-test with or without Bonferroni correction, and error bars indicate the standard error of the mean.

Figure 1 .
Figure 1.The expression of inhba and junb is regulated by injury-activated signaling during wound healing.qRT-PCR analysis of inhba (a) and junb (b) expression in DMSO (control), SB-505124 (SB), and PD0325901 (PD)-treated tails.The regenerating tails were isolated at 1 and 2 hpa for the expression of junb and inhba, respectively.The data were normalized against expression of rps18, and then by the value of DMSO.*P < 0.05, **P < 0.01.

Figure 3 .
Figure 3. InhibinβA is required for tail regeneration.(a) Schematic drawing of sgRNA target sites (sg 1 and sg 2) in the inhba locus.Grey boxes, untranslated regions; orange boxes, coding regions; arrows, single-guide RNA target sites; bar, intron region.(b) Representative phenotypes of tyr KO (control), inhba KO sg 1, inhba KO sg 2, and inhba KO sg 1 + sg 2 tadpoles at 72 hpa.(c) The lengths of regenerating tails in KO tadpoles at 72 hpa.(d) Summary of phenotypes in KO tadpoles at 72 hpa.On the basis of the lengths of regenerating tails at 72 hpa, tadpoles were classified into three phenotypic groups (normal regeneration, weakly delayed regeneration, or severely delayed regeneration).Black arrowheads indicate amputation sites.Scale bar, 200 μm.NS, not significant; ***P < 0.001.

Figure 4 .
Figure 4. InhibinβA and JunB cooperate for regenerative outgrowth/cell proliferation.(a) Representative phenotypes of tyr KO (control), inhba KO, junb KO, and inhba/junb DKO tadpoles at 72 hpa.(b) The lengths of regenerating tails in KO and DKO tadpoles at 72 hpa.(c) Representative immunofluorescent images of pH3 staining (white dots) in KO and DKO tadpoles at 36 hpa.Whole-mount immunostaining was performed with the pH3 antibody; immunostaining without the pH3 antibody was used as a negative control (no 1st Ab).(d) Relative number of proliferating cells at 36 hpa.The number of pH3-positive cells was divided by the corresponding area.All values were normalized against the value of tyr KO.Black and white arrowheads indicate amputation sites.Scale bars, 200 μm.*P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5 .
Figure 5. JunB regulates regenerative outgrowth/cell proliferation via FGF signaling.(a) Relative expression of fgf20 and tbxt in inhba KO and junb KO tadpoles at 36 hpa.The data were normalized against the expression of rps18, and then by the value of tyr KO.(b) Representative phenotypes of tyr KO (control), junb KO, junb KO + junb mRNA, and junb KO + fgf20 mRNA tadpoles at 72 hpa.(c) The lengths of regenerating tails in KO tadpoles at 72 hpa.(d) Summary of phenotypes in KO tadpoles at 72 hpa.On the basis of the lengths of regenerating tails at 72 hpa, tadpoles were classified into three phenotypic groups (normal regeneration, weakly delayed regeneration, or severely delayed regeneration).(e) Representative immunofluorescent images of pH3 staining (white dots) in tyr KO (control), junb KO, junb KO + junb mRNA, and junb KO + fgf20 mRNA tadpoles at 36 hpa.(f) Relative number of proliferating cells at 36 hpa.The number of pH3-positive cells was divided by the corresponding area.All values were normalized against the value of tyr KO.Black and white arrowheads indicate amputation sites.Scale bars, 200 μm.NS, not significant; *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6 .
Figure 6.FGF signaling is required for regenerative outgrowth/cell proliferation.(a) Schematic of the experimental plan.Tadpoles were incubated in water containing 0.0375% DMSO (control) or 15 μM SU5402 (SU) from 24 hpa and analyzed at 72 hpa for regenerating tail length and proliferating cells.(b) Representative phenotypes of DMSO and SU-treated tadpoles at 72 hpa.(c) The lengths of regenerating tails at 72 hpa.(d) Relative number of proliferating cells at 72 hpa.The number of pH3-positive cells was divided by the corresponding area.In (b) and (c), the photograph and length of regenerating tails were obtained after wholemount immunostaining, which involves bleaching and methanol treatments.Due to these treatments, the measured regenerating tail length is relatively shorter than in other experiments.All values were normalized against the value of DMSO-treated tadpoles.Black arrowheads indicate amputation sites.Scale bar, 200 μm.**P < 0.01, ***P < 0.001.