Skip to main content
PMC home page
eLife logoLink to eLife
. 2021 May 6;10:e61804. doi: 10.7554/eLife.61804

Non-canonical Hedgehog signaling regulates spinal cord and muscle regeneration in Xenopus laevis larvae

Andrew M Hamilton 1,✉, Olga A Balashova 1, Laura N Borodinsky 1,✉
Editors: Paola Bovolenta2, Marianne E Bronner3
PMCID: PMC8137141  PMID: 33955353

Abstract

Inducing regeneration in injured spinal cord represents one of modern medicine’s greatest challenges. Research from a variety of model organisms indicates that Hedgehog (Hh) signaling may be a useful target to drive regeneration. However, the mechanisms of Hh signaling-mediated tissue regeneration remain unclear. Here, we examined Hh signaling during post-amputation tail regeneration in Xenopus laevis larvae. We found that while Smoothened (Smo) activity is essential for proper spinal cord and skeletal muscle regeneration, transcriptional activity of the canonical Hh effector Gli is repressed immediately following amputation, and inhibition of Gli1/2 expression or transcriptional activity has minimal effects on regeneration. In contrast, we demonstrate that protein kinase A is necessary for regeneration of both muscle and spinal cord, in concert with and independent of Smo, respectively, and that its downstream effector CREB is activated in spinal cord following amputation in a Smo-dependent manner. Our findings indicate that non-canonical mechanisms of Hh signaling are necessary for spinal cord and muscle regeneration.

Research organism: Xenopus

Introduction

Injury in low regenerative capacity tissues such as spinal cord represents a persistent challenge in modern medicine, with millions of cases of spinal cord-related disability worldwide (Lee et al., 2014). There are currently no therapies for repairing severe damage in the mammalian central nervous system, but a variety of paradigms have shown great promise in diverse model systems, including stem cell transplantation (Dasari et al., 2014; Lu et al., 2012; Méndez-Olivos et al., 2017), pharmacological and genetic manipulation of apoptosis, reactive oxygen species, ion channel function, growth factor and morphogenetic protein signaling, and the immune response (Akazawa et al., 2004; Beck et al., 2003; Fukazawa et al., 2009; Love et al., 2013; Tapia et al., 2017; Tseng et al., 2007; Tseng et al., 2010), and direct electrical stimulation (Gomes-Osman et al., 2016). Although diverse cellular processes are involved in neural regeneration, one pathway that has consistently proven to be pro-regenerative is Hedgehog (Hh) signaling. Treatment with Sonic hedgehog (Shh) or with agonists of its main effector, Smoothened (Smo), enhances neurological recovery to nerve damage in rats (Bambakidis et al., 2012) possibly by enhancing neural cell proliferation (Bambakidis et al., 2009), while loss of Shh in zebrafish impairs retinal regeneration (Sherpa et al., 2014). Hh signaling is also essential for injury response in a wide variety of tissues, including cardiac muscle (Kawagishi et al., 2018; Singh et al., 2018a; Sugimoto et al., 2017), liver (Grzelak et al., 2014), limb (Singh et al., 2012; Yakushiji et al., 2009), and tail (Romero et al., 2018; Singh et al., 2018b; Taniguchi et al., 2014), making this pathway an attractive therapeutic target.

The Hh pathway is classically known as a primary regulator of organogenesis and tissue homeostasis (Briscoe and Thérond, 2013). Hh binding represses its receptor Patched (Ptch), thereby de-repressing the G-protein coupled receptor Smo. Canonically, Smo activates the transcription factor Gli2, which drives transcription of both Ptch and the positive feedback regulator Gli1, as well as a wide variety of genes controlling cell migration, differentiation, and especially proliferation (Briscoe and Thérond, 2013). In addition to the canonical, Gli-dependent cascade, a number of non-canonical, Gli-independent pathways exist, operating via Src family kinase (Sloan et al., 2015; Yam et al., 2009), the small GTPases Rac1 and RhoA (Ho Wei et al., 2018; Polizio et al., 2011), NF-ĸB activation via PKC (Qu et al., 2013), and the Ca2+-Ampk axis (Teperino et al., 2012), as well as other non-canonical signaling cascades (Belgacem et al., 2016; Teperino et al., 2014). Our own work has shown that Shh drives spontaneous Ca2+ spike activity in immature neurons, regulating neuronal differentiation during spinal cord development (Belgacem and Borodinsky, 2011). Moreover, neural tube formation in mouse and frog is associated with a deactivation of canonical Shh signaling (Balaskas et al., 2012; Belgacem and Borodinsky, 2015; Lee et al., 1997), and a concomitant switch to a non-canonical, PKA and Ca2+-dependent pathway, which itself contributes to the repression of the canonical, Gli-dependent cascade (Belgacem and Borodinsky, 2015).

To better understand its role in tissue repair, we examined the Hh signaling pathway during tail regeneration in Xenopus laevis larvae. We found that Hh signaling is necessary for the regeneration of the spinal cord and skeletal muscle, primarily through Gli-independent pathways, and our results implicate PKA/CREB as an interacting signaling cascade. These findings offer the possibility of enhancing regeneration by differentially targeting canonical and non-canonical Hh pathways.

Results

Hedgehog signaling regulates regeneration of muscle and spinal cord

Xenopus larvae exhibit remarkable regenerative capacity to replenish all the tissues of the tail when amputated. This ability to regenerate the tail’s spinal cord, muscle, notochord, and skin is present throughout larval stages (stages 39– 46) and before metamorphosis (stages 49–54), with a refractory period during stages 47–48, during which regeneration is blocked (Aztekin et al., 2019; Aztekin et al., 2020; Beck, 2012; Ferreira et al., 2018; Kakebeen et al., 2020; Love et al., 2013). Both regenerative periods are considered as such, rather than merely developmental growth phases, because during both periods tail amputation leaves a stump containing tissues that have already matured in the larva, unlike the growing tip of the tail that is removed by amputation. Hence, rather than a recapitulation of tail growth, this represents a true regenerative process that requires reactivating mature tissues to replace the missing tail structures.

Our previous studies have shown that electrical activity is important for spinal cord and muscle regeneration in tail-amputated stage 39–41 X. laevis larvae (Tu and Borodinsky, 2014). In addition, Ca2+-dependent activity is necessary for neuronal and muscle cell differentiation during embryonic skeletal muscle and spinal cord development (Borodinsky et al., 2004; Ferrari et al., 1996; Ferrari and Spitzer, 1999; Gu and Spitzer, 1995). This activity interacts with non-canonical, Ca2+-dependent Shh signaling during spinal cord neuron differentiation (Belgacem and Borodinsky, 2011), positioning non-canonical Hh signaling as an excellent candidate for regulating regeneration in muscle and neural tissues. Furthermore, Hh isoforms are expressed during these larval stages in several tissues, including somites, floor plate, and notochord (Ekker et al., 1995; Howell et al., 2002; Koide et al., 2006; Yin et al., 2010) and transcripts for Hh ligands are detected through single-cell RNA seq in notochord, floor plate, and various subsets of spinal cord cells in stumps and regenerating tails of stage 40 larvae from 0 through 3 days post amputation (Aztekin et al., 2019). We also find Hh protein during these stages in the axial musculature and spinal cord, and the protein level increases after 24 h post amputation (hpa) in the regenerating tail (Figure 1—figure supplement 1). Therefore, to directly examine the role of Hh-Smo signaling in tissue regeneration, tails of X. laevis larvae were amputated and larvae allowed to regenerate in Smo antagonist cyclopamine, Smo agonist SAG, or vehicle control solution, then fixed at 72 hpa and stained for mitotic activity (phospho-histone H3 [P-H3]), spinal cord (Sox2+ neural stem cells [NSCs] lining the spinal cord central canal), and skeletal muscle (12/101+ differentiated skeletal muscle cells) in the regenerate and vicinity (Figure 1A–C).

Figure 1. Hedgehog signaling regulates spinal cord and muscle regeneration.

Stage 39–40 Xenopus laevis larvae were incubated for 24 or 72 h after tail amputation in vehicle (0.1% DMSO, Control), antagonist (10 µM cyclopamine, cyc), or agonist (1 µM SAG) of Smoothened (Smo), and immunostained at 72 h post amputation (hpa). (A) Schematics of tail amputation and 24 vs. 72 h treatments. (B) Measurement of outgrowth for regenerated spinal cord and muscle (solid lines) from amputation plane (dashed line), identified by the posterior-most close-packed band of muscle fibers (arrow). (C) Representative z-projections of whole-mount immunostained samples for each group at 72 hpa. Transverse red dashed line indicates amputation plane. Scale bars in (B, C), 100 µm. (D–I) Graphs show mean ± SEM regenerated muscle outgrowth (D) and total new muscle volume (E), regenerated spinal cord outgrowth (F), total number (G), and per length density (H) of Sox2+ cells in the regenerated spinal cord, and number of phospho-histone H3+ (P-H3+) cells (I) in the regenerate at 72 hpa as % of cohort-matched control. n of larvae: 25–42 per group, N of experiments ≥ 3. *p<0.05, ***p<0.001, ****p<0.0001, ns: not significant, ordinary one-way ANOVA, Brown–Forsythe and Welch ANOVA, or Kruskal–Wallis test, followed by Tukey’s, Dunnett’s T3, or Dunn’s multiple comparisons test, respectively, according to prior normality and equality of SDs tests within and between groups.

Figure 1—source data 1. Hedgehog signaling regulates spinal cord and muscle regeneration.

Figure 1.

Figure 1—figure supplement 1. Expression of Hedgehog (Hh) ligands is apparent in stage 40 larvae and increases after amputation in the regenerating tail.

Figure 1—figure supplement 1.

(A) Representative image of transverse section of tail from 4 h post amputation (hpa) stage 40 larva, stained for Sox2 and Hh (5E1 antibody that recognizes Xenopus Sonic hedgehog [Shh] and Ihh). Scale bar, 10 μm. (B) Representative western blot assay of whole-cell lysates from 500-μm-long most posterior tail from stage 40 larvae right after (0 hpa) and 24 hpa. Predicted full-length Hh MW 45–47 kD. GAPDH (bottom) was used as loading control. Graph shows mean ± SEM% of 0 hpa Hh/GAPDH optical density (O.D.), n = 18 larvae per group, N = 3 experiments, *p<0.05, paired t-test.
Figure 1—figure supplement 1—source data 1. Treatment with Smoothened (Smo) antagonist vismodegib reduces muscle and spinal cord regeneration.
Figure 1—figure supplement 2. Treatment with Smoothened (Smo) antagonist vismodegib reduces muscle and spinal cord regeneration.

Figure 1—figure supplement 2.

Stage 39–40 X. laevis larvae were incubated with vehicle (0.1% DMSO, Control) or 20 µM vismodegib for 72 hr, then immunostained. (A) Representative z-projections of whole-mount immunostained samples for both groups at 72 h post amputation (hpa). Transverse red dashed line indicates amputation plane. Scale bar 100 µm. (B–G) Graphs show mean ± SEM regenerated muscle outgrowth (B) and total new muscle volume (C), regenerated spinal cord outgrowth (D), total number (E) and per length density (F) of Sox2+ cells in the regenerated spinal cord, and number of phospho-histone H3+ (P-H3+) cells (G) in the regenerate at 72 hpa as % of cohort-matched control. N of larvae: 8–10 samples per group. ***p<0.001, ****p<0.0001, ns: not significant, two-tailed t-test.
Open in a new tab

Smo modulation alters both muscle and spinal cord regeneration. Inhibition of Smo for 72 hpa with cyclopamine reduces muscle outgrowth into the regenerating tail (Figure 1D) and total regenerated muscle volume (Figure 1E). Cyclopamine also impairs spinal cord outgrowth (Figure 1F) and reduces the total number of NSCs in the newly regenerated spinal cord (Figure 1G), resulting in spinal cord with a lower NSC density than in control larvae (Figure 1H), which suggests that inhibiting Smo interferes with activation/proliferation of NSCs. Furthermore, treatment of amputated larvae with vismodegib, a structurally distinct Smo antagonist that binds to a different domain than cyclopamine, results in very similar reductions in regeneration (Figure 1—figure supplement 2).

In contrast, enhancing Smo activity with SAG has no effect on total muscle regeneration (Figure 1E), and even shows a modest inhibitory effect on muscle outgrowth (Figure 1D). In addition, SAG treatment universally results in abnormal muscle regeneration consisting in muscle cells that do not follow a longitudinal alignment parallel to the anteroposterior axis and ectopic outgrowth of new muscle fibers dorsally and ventrally from intact axial musculature in the tail stump and regenerated tail, in contrast to all other groups where this phenotype was not observed (26 out of 26 SAG-treated samples; 0 out of 41 controls; 0 out of 31 cyclopamine-treated samples; Figure 1C), indicating dysregulated muscle morphogenesis. SAG also reduces the outgrowth of the spinal cord (Figure 1F) without altering the total number of new NSCs (Figure 1G), resulting in a truncated, wide spinal cord with higher NSC density than in control larvae (Figure 1C, H). Similarly, overall mitotic activity at 72 hpa is not affected by SAG but is reduced by cyclopamine (Figure 1I). These results argue that Smo activity is essential for NSC proliferation, and that ectopically elevated Smo signaling interferes with the normal progression from NSC proliferation to spinal cord outgrowth. Altogether these results suggest that Smo-mediated signaling is necessary for muscle and spinal cord regeneration.

In addition, we examined the interval during which Hh signaling is necessary for regeneration by comparing 72 h exposure with treatment for only the first 24 h of 72 h total (Figure 1A; 24 h vs. 72 h). We found that the effects of 24 h exposure to SAG were comparable to the full 72 h treatment for all metrics of both muscle (Figure 1D, E) and spinal cord (Figure 1F–H) regeneration, as well as overall cell proliferation in the regenerate (Figure 1I), and gave the same abnormal muscle regeneration phenotype with ectopic dorsoventral muscle outgrowth and shortened spinal cord in all samples, in contrast to all other groups where this phenotype was not observed (27 of 27 SAG-treated samples; 0 out of 41 controls; 0 out of 35 cyclopamine-treated samples; Figure 1C). Cyclopamine treatment for only 24 hr, however, showed weaker effects than the full 72 hr; the number and density of NSCs in the newly formed spinal cord, as well as the number of mitotic cells in the regenerate, are unchanged by 24 h cyclopamine (Figure 1G–I), while reductions in muscle outgrowth and volume are smaller than with 72 h cyclopamine incubation (Figure 1D, E). In contrast, the reduction in spinal cord outgrowth is comparable between 24 and 72 h cyclopamine treatments (Figure 1F).

These results suggest that endogenous Hh signaling plays a critical role in the early stages of muscle regeneration that involve activation of muscle stem cells, as well as an ongoing role in maintaining the proliferation necessary for both spinal cord and muscle growth, and indicate that balanced Hh signaling is necessary for spinal cord and muscle regeneration.

Canonical Hh signaling is rapidly repressed following tail amputation

Next, we examine the endogenous activity of canonical Hh signaling following amputation using a dual-cassette transcriptional activity reporter plasmid with simultaneous Gli1/2-dependent expression of enhanced green fluorescent protein (EGFP), and constitutive expression of near-infrared fluorescent protein as a normalizing factor for reporter expression (iRFP670; Gli reporter; Figure 2A). When injected into 2–4 cell-stage embryos, this construct is distributed in a mosaic fashion throughout the larva, giving an expression-normalized readout of Gli1/2 transcriptional activity (EGFP:iRFP670 ratio, Figure 2A’). We validated this construct by treating Gli reporter-injected embryos with SAG, resulting in a significant increase in EGFP:iRFP670 signal during neural plate development, a period known to exhibit high canonical Hh signaling activity (Lee et al., 1997; Belgacem and Borodinsky, 2015; Figure 2—figure supplement 1).

Figure 2. Canonical, Gli-dependent Hedgehog signaling is downregulated following tail amputation.

Tails of stage 39–40 larvae expressing the Gli transcriptional activity reporter were amputated, and larvae imaged live at intervals from 0 to 28 h post amputation (hpa). (A) Schematic of bicistronic reporter plasmid. Constitutive promoter CMV drives expression of iRFP670, and minimal δ-crystallin promoter juxtaposed to eight Gli-binding sites drives expression of enhanced green fluorescent protein (EGFP). (A’) Representative z-projections of normalizing factor (iRFP670) and Gli transcriptional activity reporter (EGFP) in amputated larval tail at 12 hpa. White outline represents the edge of the tail. Scale bar, 100 µm. (B–F) 2D composite of EGFP:iRFP670 (G:iR ratio) intensity displayed by heat scale at indicated hpa from combined iRFP670+ cells. Cell location displayed by position relative to amputation plane (x) and dorsal or ventral to the spinal cord (y). (G) Cumulative distribution of EGFP:iRFP670 ratios by time point. All cumulative distribution curves significantly different from control (0 hpa); N of larvae ≥12; p<0.001, all time points compared to 0 hpa, Kolmogorov–Smirnov post-hoc test.

Figure 2—source data 1. Canonical, Gli-dependent Hedgehog signaling is downregulated following tail amputation.

Figure 2.

Figure 2—figure supplement 1. Positive control for Gli transcriptional activity reporter.

Figure 2—figure supplement 1.

Neural plate stage embryos (18 h post-fertilization) expressing Gli transcriptional activity reporter were incubated in the presence or absence of 10 nM SAG for 5 h followed by live confocal imaging with 488 and 647 nm lasers to acquire the signal corresponding to Gli transcriptional activity (enhanced green fluorescent protein [EGFP]) and expression level for the reporter (iRF670). Individual cells displayed by iRFP670 (x) vs. EGFP (y) fluorescence intensities in control and experimental embryos. Distribution of control vs. SAG EGFP:iRFP670 values significantly different by Kolmogorov–Smirnov test (p<0.0001; N of embryos per condition: 8).
Figure 2—figure supplement 1—source data 1. Positive control for Gli transcriptional activity reporter.
Open in a new tab

When each iRFP670+ cell is assigned a heat scale color by EGFP:iRFP670 intensity and displayed in two-dimensional space by its position relative to the spinal cord (Y) and the amputation plane (X), we find that relative to 0 hpa (Figure 2B), by 4 hpa there is a broad reduction in Gli-reporter signal which appears 300–800 μm anterior to the amputation plane (Figure 2C). This decrease expands and deepens through 8 hpa, until reaching a minimum at 12 hpa, before showing partial recovery at 28 hpa (Figure 2D–F). When the cumulative distribution of EGFP:iRFP670 ratios is compared by time point, we find significant reductions in the proportion of higher Gli activity cells at all time points post amputation (Figure 2G).

These data indicate that amputation rapidly induces widespread inhibition of canonical Gli-dependent signaling.

Hedgehog-dependent spinal cord and muscle regeneration are primarily non-canonical

Our data suggest that while Smo-dependent Hh signaling is necessary for proper spinal cord and muscle regeneration, canonical, Gli-dependent activity is downregulated immediately following tail amputation; therefore, we directly addressed the necessity of Gli1/2 activity during tail regeneration. GANT61 is a small-molecule inhibitor of the transcriptional activity of Gli1 and Gli2, the primary activators of downstream canonical Hh signaling (Lauth et al., 2007). Using our Gli reporter, we demonstrate that treatment with 10 µM GANT61 significantly reduces Gli transcriptional activity in X. laevis larvae (Figure 3—figure supplement 1).

Treatment of tail-amputated larvae with GANT61 for 72 h (Figure 3A) results in only a modest reduction in the number of new NSCs (Figure 3E), reducing Sox2 density in the regenerating spinal cord (Figure 3F), but does not significantly alter any other regeneration or proliferation metrics in either muscle or spinal cord (Figure 3B–D, G). This contrasts sharply with cohort-paired cyclopamine treatment, which shows reductions in all regeneration metrics for muscle and spinal cord (Figure 3). Furthermore, when GANT61 treatment is combined with cyclopamine, the results are indistinguishable from cyclopamine treatment alone for all regeneration metrics (Figure 3).

Figure 3. Gli1/2 transcriptional activity is not necessary for spinal cord and muscle regeneration.

Stage 39–40 larvae were incubated for 72 h after tail amputation in vehicle (0.1% DMSO, Control) or Gli1/2 antagonist (10 µM GANT61, GANT) and/or 10 µM cyclopamine (cyc), then whole-mount immunostained. (A) Images show representative z-projections for each group at 72 h post amputation (hpa). Transverse red dashed line indicates amputation plane. Scale bar, 100 µm. (B–G) Graphs show mean ± SEM regenerated muscle outgrowth (B) and total volume (C), regenerated spinal cord outgrowth (D), total number (E) and per length density (F) of Sox2+ cells in the regenerated spinal cord, and overall number of phospho-histone H3+ (P-H3+) cells (G) in the regenerate at 72 hpa as % of cohort-matched control, n of larvae: 13–31 per group, N of experiments ≥ 3. ***p<0.001, ****p<0.0001, ns: not significant, ordinary one-way ANOVA, Brown–Forsythe and Welch ANOVA, or Kruskal–Wallis test, followed by Tukey’s, Dunnett’s T3, or Dunn’s multiple comparisons test, respectively, according to prior normality and equality of SDs within and between groups.

Figure 3—source data 1. Gli1/2 transcriptional activity is not necessary for spinal cord and muscle regeneration.

Figure 3.

Figure 3—figure supplement 1. Treatment with GANT61 inhibits Gli1/2 transcriptional activity.

Figure 3—figure supplement 1.

Stage 39 larvae expressing the Gli transcriptional activity reporter were incubated with 10 µM GANT61 and confocally live-imaged with 488 and 647 nm lasers to acquire the signal corresponding to Gli transcriptional activity (enhanced green fluorescent protein [EGFP]) and expression level for the reporter (iRF670), at 0 (A), 6 (B), and 18 h (C) of incubation. Shown are paired EGFP and iRFP fluorescence intensities for individual cells at the indicated time points. Both 6 h (p=0.015) and 18 h (p<0.0001) time points showed significantly different distribution of EGFP:iRFP670 values from 0 h (Kolmogorov–Smirnov test; N of larvae: 8).
Figure 3—figure supplement 1—source data 1. Treatment with GANT61 inhibits Gli1/2 transcriptional activity.
Open in a new tab

In addition to pharmacological inhibition of Gli1/2 activity, we used a translation-blocking morpholino to downregulate protein abundance of Gli2, the primary activator of canonical Hh signaling. As Gli2 signaling is indispensable for early embryogenesis, we coinjected a complementary, UV photolabile blocker morpholino (Photo-MO) which binds to and disables the matched Gli2 morpholino (Gli2-MO), allowing UV-dependent uncaging of Gli2-MO at later developmental periods (Figure 4A). Exposure of uninjected embryos to UV does not reduce any regeneration metrics (Figure 4—figure supplement 1A–D), while activation of Gli2-MO has no significant effect on regeneration of muscle or spinal cord, with the exception of a small reduction in spinal cord outgrowth (Figure 4C–H), despite downregulation of full-length Gli2 protein levels (Figure 4—figure supplement 1E).

Figure 4. Genetic downregulation of canonical Hedgehog-Smoothened (Hh-Smo) signaling does not affect spinal cord and muscle regeneration.

Larvae containing Gli2 morpholino (Gli2-MO) bound to photo-morpholino were UV illuminated (+UV) or not (control, no UV) for 30 min, 28 h before (stage 28) amputation (stage 39–40), to uncage morpholino and induce downregulation of Gli2 expression. (A) Schematic of time course of morpholino activation and amputation. (B) Images show representative z-projections of samples for each group at 72 h post amputation (hpa). Transverse red dashed line indicates amputation plane. Scale bar, 100 µm. (C–H) Graphs show mean ± SEM regenerated muscle outgrowth (C) and total volume (D), regenerated spinal cord outgrowth (E), total number (F) and per length density (G) of Sox2+ cells in the regenerated spinal cord, and overall number of phospho-histone H3+ (P-H3+) cells (H) in the regenerate at 72 hpa as % of cohort-matched control, n of larvae: 16–35 per group, N of experiments ≥ 3. ***p<0.001, ns: not significant, unpaired t-test, Welch’s t-test, or Kolmogorov–Smirnov test, according to prior normality and equality of SDs tests within and between groups.

Figure 4—source data 1. Genetic downregulation of canonicalHedgehog-Smoothened(Hh-Smo)signaling does not affect spinal cord and muscle regeneration.

Figure 4.

Figure 4—figure supplement 1. UV treatment does not reduce regeneration and uncages Gli2-MO, knocking down Gli2 expression.

Figure 4—figure supplement 1.

Uninjected larvae (A–D) or larvae containing Gli2-MO bound to photo-morpholino (E) were UV illuminated (+UV) or not (control, no UV) for 30 min, 28 h (stage 28) before amputation (stage 39–40). Amputated samples were processed at 72 h post amputation (hpa) for whole-mount immunostaining (A–D) or frozen immediately for western blot assays (E). (A–D) Graphs show mean ± SEM regenerated muscle outgrowth (A) and total volume (B), regenerated spinal cord outgrowth (C), and overall number of phospho-histone H3+ (P-H3+) cells (D) in the regenerate at 72 hpa as % compared to control, n of larvae: 12–19 per group. *p<0.05, ns: not significant, unpaired t-test. (E) Representative western blot assay of whole-larva nuclear fraction lysates. Predicted full-length Gli2 MW ≥168 kD indicated by arrow. Lamin B1 (bottom) was used as loading control for nuclear fraction. Graph shows mean ± SEM% of control (no UV) Gli2/Lamin-B1 optical density (O.D.), n = 18 larvae per group, N = 3 separate experiments, *p=0.01, two-tailed t-test, resulting in a mean ± SEM% knockdown of 49 ± 6%.
Figure 4—figure supplement 1—source data 1. UV treatment does not reduce regeneration and uncages Gli2-MO, knocking down Gli2 expression.
Open in a new tab

These findings indicate that canonical Hh signaling is, for the most part, not necessary for spinal cord and muscle regeneration, and suggest that Smo-dependent regeneration recruits a non-canonical signaling pathway.

PKA signaling is necessary for regeneration

In previous studies, we found that Ca2+ activity is associated with both muscle regeneration (Tu and Borodinsky, 2014) and PKA-dependent non-canonical Shh signaling through Smo during embryonic spinal cord development (Belgacem and Borodinsky, 2011; Belgacem and Borodinsky, 2015). To examine a potential role for PKA activity during tail regeneration, we treated amputated larvae with the PKA inhibitor KT5720 (Figure 5A). Inhibiting PKA diminishes all regeneration metrics, reducing outgrowth of muscle (Figure 5B) and spinal cord (Figure 5D), total muscle regeneration (Figure 5C), and spinal cord NSC (Figure 5E) and mitotic cell (Figure 5G) counts. The reduction in NSC number when PKA is inhibited is approximately proportional to the reduction in spinal cord outgrowth, resulting in a slight increase in NSC density (Figure 5F), likely due to the fact that PKA inhibition leads to a larger overall decrease in spinal cord regeneration than when inhibiting Smo (Figure 5D). This suggests that PKA participates in spinal cord regeneration beyond the initial activation and proliferation of NSCs, while Smo signaling is particularly important for NSC proliferation.

Figure 5. Tissue-specific interaction between PKA and Smoothened (Smo) signaling in regulation of spinal cord and muscle regeneration.

Figure 5.

Open in a new tab

Stage 39–40 larvae were incubated for 72 h after tail amputation in vehicle (0.1% DMSO, Control), PKA antagonist (10 µM KT5720, KT), or/and 10 µM cyclopamine (cyc). (A) Images show representative samples for each group at 72 h post amputation (hpa). Transverse red dashed line indicates amputation plane. Scale bar, 100 µm. (B–G) Graphs show mean ± SEM regenerated muscle outgrowth (B) and total volume (C), regenerated spinal cord outgrowth (D), total number (E) and per length density (F) of Sox2+ cells in the regenerated spinal cord, and overall number of phospho-histone H3+ (P-H3+) cells (G) in the regenerate at 72 hpa as % of cohort-matched control, n of larvae: 11–29 per group, N of experiments ≥ 3. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns: not significant, ordinary one-way ANOVA, Brown–Forsythe and Welch ANOVA, or Kruskal–Wallis test, followed by Tukey’s, Dunnett’s T3, or Dunn’s multiple comparisons test, respectively, according to prior normality and equality of SDs within and between groups.

Figure 5—source data 1. Tissue-specific interaction between PKA and Smoothened(Smo)signaling in regulation of spinal cord and muscle regeneration.

Simultaneous treatment with KT5720 and cyclopamine reveals tissue-specific epistasis: inhibition of Smo is epistatic over PKA inhibition for muscle outgrowth (Figure 5B) and total new muscle formation (Figure 5C), suggesting that PKA acts upon muscle regeneration downstream of Smo. In contrast, dual KT5720/cyclopamine treatment results in additive inhibition of spinal cord outgrowth (Figure 5D), but no significant additivity in reduction of NSC (Figure 5E) or mitotic cell counts (Figure 5G). Interestingly, simultaneous inhibition of PKA and Smo slightly ameliorates the reduction in NSC density observed with cyclopamine alone (Figure 5F), reinforcing the concept that PKA acts upon spinal cord regeneration in a mechanistically distinct manner from Smo. This suggests that PKA and Smo act on spinal cord regeneration in a parallel but coordinated manner. It should be noted that KT75720 has been shown to also inhibit other kinases in cell line and cell-free studies, including phosphorylase kinase, PDK1, and MEK (Davies et al., 2000; Murray, 2008). Hence, further investigation will be necessary to sort out the participation of specific alternative kinase pathways in tissue regeneration.

Since kinases such as PKA are known to be potent regulators of a wide variety of signaling pathways, we directly examined one of PKA’s primary downstream effectors, the transcription factor CREB, which is activated by non-canonical Hh signaling in the embryonic spinal cord (Belgacem and Borodinsky, 2015). To examine endogenous CREB activity following amputation, we fixed larvae pre-amputation (0 hr) and at intervals from 4 to 48 hpa and stained for activated phospho-Ser133-CREB (P-CREB; Belgacem and Borodinsky, 2015; Gonzalez and Montminy, 1989). We found that P-CREB signal is strong in the skin around the regenerating tail tip, making isolation of P-CREB signal in muscle unfeasible (Figure 6—figure supplement 1). However, we were able to isolate P-CREB staining from the Sox2-labeled region of the spinal cord (Figure 6A). We find that the density of P-CREB+ cells in the spinal cord within the amputated tail stump, 100 μm or more anterior from the amputation plane, is similar across individual 100 μm bins of spinal cord in all time points post amputation, comparable to pre-amputation controls (Figure 6B). In contrast, within the amputation region and regenerating spinal cord, we observe a significant increase in P-CREB+ cells compared to the intact stump as early as 4 hpa and throughout the first 24 hpa (Figure 6B). By 48 hpa, this increase in P-CREB+ cell density is no longer apparent at any region in the regenerating spinal cord (Figure 6B). This suggests that CREB activity is differentially regulated spatially and temporally in the spinal cord, with a rapid, transient increase in CREB activation in the spinal cord proximal to the amputation site.

Figure 6. Spatiotemporal activation of CREB in the injured and regenerated spinal cord.

Stage 39–40 larvae were amputated and processed for whole-mount immunostaining at the indicated hour post amputation (hpa), with the exception of the 0 h group, which was first fixed and then amputated to represent the pre-amputation group. (A) Images show representative z-projections of immunostained samples for each group that were digitally processed to isolate spinal-cord associated P-CREB. Transverse red dashed line indicates amputation plane. Scale bar, 100 µm. (B) Graph shows mean ± SEM number of total P-CREB+ cells normalized to number of Sox2+ cells in 100 µm sections of spinal cord anterior (positive) and posterior (negative) to the amputation plane (0). N of larvae ≥5 per group. Open circles denote p<0.05 vs. time-matched average number of P-CREB+/Sox2+ cells within the region 100–200 µm anterior (100 µm, dashed black line) to the amputation plane (0 µm), one-way ANOVA.

Figure 6—source data 1. Spatiotemporal activation of CREB in the injured and regenerated spinal cord.

Figure 6.

Figure 6—figure supplement 1. Unedited images from samples featured in Figure 6.

Figure 6—figure supplement 1.

Stage 39–40 larvae were amputated and processed for whole-mount immunostaining at the indicated hour post amputation (hpa), with the exception of the 0 h group, which was first fixed and then amputated to represent the pre-amputation group. Images show representative total z-projections of whole-mount immunostained samples for each group corresponding to the edited images shown in Figure 6. Transverse red dashed line indicates amputation plane. Scale bar, 100 µm.
Open in a new tab

Moreover, phosphorylation of P-CREB immediately post-amputation is bidirectionally dependent on Smo activity, as shown in transverse sections of paraffin-embedded larvae treated with SAG or cyclopamine for 4 hpa. Within 400 µm of the amputation site, cyclopamine-treated larvae show significantly fewer P-CREB+ cells in the spinal cord relative to vehicle controls, while SAG treatment increases this metric (Figure 7A, B). Analysis of 24 hpa whole mounts reveals a similar effect, with cyclopamine treatment reducing P-CREB+ cell counts in regenerated spinal cord, while treatment with SAG increases CREB activation compared to controls (Figure 7C, D). These data indicate that Hh signaling regulates CREB activation during regeneration.

Figure 7. Post-amputation activation of CREB is regulated by Hedgehog (Hh) signaling.

Figure 7.

Open in a new tab

Stage 39–40 larvae were amputated in vehicle (0.1% DMSO, control), 10 µM cyclopamine, or 1 µM SAG, then fixed at either 4 or 24 h post amputation (hpa) and processed for immunostaining in either paraffin sections (A, B), or whole-mounts (C, D), respectively. (A) Representative longitudinal sections from 4 hpa. Sample orientation displayed in schematic. Scale bars, 20 µm. (B) Graph shows mean ± SEM P-CREB+ cell count in 100-µm-long spinal cord within 400 µm of the amputation plane as % of control, n of larvae ≥5 per group from N = 3 independent experiments. (C) Representative whole-mount sections from 24 hpa. Sample orientation displayed in schematic. Scale bar, 50 µm. (D) Graph shows mean ± SEM P-CREB+ cell count in the regenerated spinal cord, normalized to cohort matched controls, n of larvae ≥12 per group from N = 3 separate experiments. In (B, D), *p<0.05, ***p<0.001, ****p<0.0001, one-way ANOVA + Holm–Sidak’s multiple comparison.

Figure 7—source data 1. Post-amputation activation of CREB is regulated by Hedgehog(Hh)signaling upon injury.
Figure 7—source data 2. Post-amputation activation of CREB is regulated by Hedgehog(Hh)signaling in the regenerating spinal cord.

To further examine the potential for non-canonical Hh signaling-dependent interplay between regeneration of spinal cord and muscle, we compared the correlation in the regeneration of these tissues when either Smo or PKA signaling has been altered. Under control conditions, we observe a close correlation between muscle and spinal cord outgrowth when examined on a per-sample basis. This correlation is decoupled by enhancing Smo signaling, which allows muscle outgrowth in the absence of normal spinal cord outgrowth, and by inhibiting it, which blocks muscle regeneration almost entirely, despite substantial spinal cord outgrowth (Figure 8). In contrast, PKA inhibition reduces all regeneration metrics without altering the interrelationship between the extent of regeneration of both tissues. Inhibiting Gli1/2 also has no discernible effect on correlated outgrowth between the two tissues (Figure 8). This suggests that balanced non-canonical Smo-dependent Hh signaling is necessary for coordinated regeneration of spinal cord and muscle (Figure 9).

Figure 8. Coordination of spinal cord and muscle outgrowth in the regenerating tail is dependent on Smoothened (Smo) signaling.

Figure 8.

Open in a new tab

Individual samples for each treatment are displayed by total outgrowth for spinal cord (X) and muscle (Y). Simple linear regression lines are fit for each treatment, and R2 goodness-of-fit values displayed by each line. N = 13–39 samples per group.

Figure 8—source data 1. Coordination of spinal cord and muscle outgrowth in the regenerating tail is dependent on Smoothened(Smo)signaling.
elife-61804-fig8-data1.xlsx (148.1KB, xlsx)

Figure 9. Model for Hedgehog (Hh)-dependent regulation of spinal cord and muscle regeneration.

Figure 9.

Open in a new tab

Injury recruits non-canonical Hh signaling in the spinal cord to activate neural stem cells for replenishing the regenerating spinal cord. Injury also activates PKA which acts independently from Smoothened (Smo) on neural stem cell proliferation to promote spinal cord regeneration. Non-canonical Hh signaling is also essential for muscle regeneration starting from the initial stages post injury, where PKA appears to be downstream of Smo activation. The coordination between the magnitude of regenerating spinal cord and muscle outgrowth is dependent on non-canonical Hh/Smo signaling.

Discussion

Our work demonstrates that Hh signaling is necessary for regeneration of muscle and spinal cord in Xenopus larvae. These findings are in keeping with previous discoveries showing Hh signaling involvement in the regeneration of a variety of tissues including liver, heart, and limb (Grzelak et al., 2014; Singh et al., 2012; Singh et al., 2018b), in addition to Xenopus larval tail (Taniguchi et al., 2008; Taniguchi et al., 2014), as well as the fact that Hh pathway activation is enhanced at the site of tail amputation in X. laevis (Taniguchi et al., 2014) and zebrafish (Romero et al., 2018). Recent studies have also shown that Smo and Ptch1 experience a transient increase in transcriptional availability immediately after tail amputation in Xenopus tropicalis (Kakebeen et al., 2020), and shh transcript level increases in 1 dpa X. laevis larvae (Aztekin et al., 2019), further emphasizing the importance of this pathway during regeneration.

The regenerated spinal cord under Smo inhibition exhibits around half of the number of NSCs at 72 hpa compared to control animals. This suggests that with regards to spinal cord regeneration Hh-Smo signaling is particularly important for NSC proliferation. Moreover, pharmacological enhancement of Smo signaling leads to simultaneous NSC proliferation and blockade of spinal cord morphogenesis, further supporting the model of Smo signaling acting predominantly on NSC activation and proliferation upon injury (Figure 9).

Smo signaling is also essential for muscle regeneration, and arguably for the initial stages of muscle stem cell activation and proliferation since inhibiting Smo signaling during the first 24 hpa is enough to strongly decrease the replenishing of skeletal muscle (Figure 9). Moreover, it appears that the first 24 h post injury are a critical period for Smo-dependent muscle stem cell activation, unlike NSCs, which seem to be able to be activated and proliferate even after an initial 24 h of Smo inhibition. It remains to be determined whether Smo-dependent muscle regeneration is acting directly on muscle stem cells or indirectly through signals from the spinal cord. Nevertheless, the evidence that in the presence of overactive Smo both the spinal cord and muscle exhibit aberrant outgrowth, along with published work from others demonstrating the necessity of the spinal cord for proper tail regeneration (Taniguchi et al., 2008), supports a potential interaction between Smo-mediated spinal cord and muscle regeneration (Figure 9).

Additionally, we demonstrate that canonical, Gli-dependent Hh signaling is endogenously downregulated after tail amputation, and further inhibiting this pathway either pharmacologically or genetically only marginally affects muscle and spinal cord regeneration, suggesting that Hh-Smo is acting primarily through a Gli-independent pathway (Figure 9). In contrast, CREB activity is recruited immediately after injury, at the amputation site and within the regenerating spinal cord, and fades within the first 24 h of spinal cord regeneration. Moreover, PKA activity, unlike Gli, is necessary for efficient regeneration of both muscle and spinal cord, and CREB is activated upon injury in amputation-proximal and regenerating neural tissue in a Smo-dependent manner. This distinct recruitment of a non-canonical Hh signaling pathway resembles the switch in Hh signaling observed during embryonic stages of early spinal cord development, when canonical, Gli-activating Shh signaling is restricted to the early neural plate (Balaskas et al., 2012; Belgacem and Borodinsky, 2015; Lee et al., 1997), but is later repressed through a Shh-Ca2+-PKA-CREB signaling axis that mediates Shh-dependent spinal cord neuron differentiation (Belgacem and Borodinsky, 2011; Belgacem and Borodinsky, 2015).

PKA recruitment is not the only potential mediator of Hh non-canonical, Smo-dependent regeneration. The effects of inhibiting PKA on muscle regeneration are lower in magnitude compared with Smo inhibition, with the latter resulting in almost complete blockade of muscle replenishment in the regenerated tail. This may be the reason for apparent epistasis when simultaneously inhibiting Smo and PKA pathways in regulating muscle regeneration. Alternatively, PKA may be one of multiple non-canonical Hh signaling pathways downstream of Smo acting on muscle regeneration (Figure 9). It should also be noted that KT5720 has been demonstrated to effectively inhibit other kinases in purified, cell-free preparations (Davies et al., 2000); among these is GSK3β, which is known to regulate Hh signaling through both Gli2 (Riobó et al., 2006) and its inhibitor SUFU (Chen et al., 2011). The involvement of non-canonical Hh signaling and CREB supports the involvement of PKA (Belgacem and Borodinsky, 2015), but further experiments would be necessary to rule out or include contributions by other kinases.

The link between Smo and regeneration must lie in downstream transcription factors other than Gli, such as NFĸB (Qu et al., 2013), MycN (Mani et al., 2020; Singh et al., 2018b), CREB (Belgacem and Borodinsky, 2015), or JAK/STAT (Tian et al., 2015), which has previously been shown to regulate regeneration in X. laevis (Tapia et al., 2017). If specific signaling pathways downstream of Hh-dependent regeneration can be targeted, it may help limit the pleiotropic effects observed with direct manipulation of Smo (Wu et al., 2017).

Finally, PKA activity in spinal cord regeneration does not appear to be linearly downstream to Smo-dependent signaling (Figure 9) since we observed additive effects when simultaneously inhibiting Smo and PKA. The exact mechanisms by which PKA modulates regeneration, however, remain unclear. PKA may act as an inhibitor of Gli, which would be in keeping with both our data showing Gli downregulation following amputation and its developmental role as a repressor of canonical Hh signaling (Hammerschmidt et al., 1996), as well as PKA participation in the non-canonical, Ca2+ activity and CREB-dependent repression of Gli during X. laevis spinal cord development (Belgacem and Borodinsky, 2015; Figure 9). Since we have also found that Ca2+ activity is necessary for Xenopus larva tail regeneration following amputation (Tu and Borodinsky, 2014), our results showing an early reduction in Gli activity and activation of CREB, as well as the dependence of spinal cord and muscle regeneration on PKA activity, suggest that the aforementioned non-canonical pathway may be at least partially responsible for tissue regeneration.

Manipulation of Hh signaling has already shown great promise in treating a variety of conditions, including cancer (Ruat et al., 2014), neural injury and stroke (Bambakidis and Onwuzulike, 2012), cardiac ischemia (Dunaeva and Waltenberger, 2017), appendage regeneration (Chen et al., 2014; Singh et al., 2015), and even osteoporosis and obesity (Hadden, 2014). Based on the results from this study, we predict that selectively enhancing regeneration-specific, non-canonical Hh signaling in spinal cord and muscle during a critical period following injury might promote the repair and replenishment of functional tissues.

Materials and methods

Key resources table.

Reagent type
(species)
or resource		 Designation Source
or reference		 Identifiers Additional
information
Antibody 5E1
(mouse monoclonal)		 Developmental Studies Hybridoma Bank RRID:AB_528466 Shh; (1:50)
Antibody GAPDH
(goat polyclonal)		 SICGEN RRID:AB_0049-200 Loading control whole-cell lysates, western blots (1:1000)
Antibody 12/101 (mouse monoclonal) Developmental Studies Hybridoma Bank RRID:AB_531892 Skeletal muscle;
(1:100)
Antibody Sox2
(goat polyclonal)		 R&D AF2018, RRID:AB_355110 Neural stem cells; (1:300-1:400)
Antibody P-H3
(rabbit polyclonal)		 Millipore 06-570, RRID:AB_310177 Mitotic marker; (1:400)
Antibody P-CREB
(rabbit polyclonal)		 Cell Signaling 9198 Phosphorylated transcription factor; (1:800-1:1500)
Antibody Gli2
(goat polyclonal)		 R&D AF3635; RRID:AB_211902 Transcription factor; (1:800)
Antibody Lamin-B1
(rabbit monoclonal)		 Cell Signaling 9087; RRID:AB_10896336 Nuclear protein for loading control in western blot assays; (1:500)
Recombinant DNA reagent p8xGli-EGFP_CMV-iRFP670 This paper Gli activity reporter; design described in Materials and methods/Gli activity reporter section of this paper
Sequence-based reagent Gli2-morpholino Gene Tools GCACAGAACGCAGGTAATGCTCCAT
Sequence-based reagent Gli2-photo-morpholino Gene Tools ATGGAGCATTACPTGCGTTCT
Chemical compound, drug Cyclopamine Sigma-Aldrich C4116 10–20 mM stock in DMSO
Chemical compound, drug SAG Calbiochem 566660 5 mM stock in H2O
Chemical compound, drug Vismodegib Sigma 879085-55-9 50 mM in DMSO
Chemical compound, drug KT5720 Tocris 1288 10 mM stock in DMSO
Chemical compound, drug GANT61 Tocris 3191 10 mM srock in DMSO
Chemical compound, drug Tricaine-S Syndel MS 222 Anesthetic
Open in a new tab

Animals

X. laevis females were primed (50 units) and injected (350–400 units) with human chorionic gonadotropin to induce egg laying. Eggs were squeezed into 1× Marc’s Modified Ringer solution (MMR in mM: 110 NaCl, 2 KCl, 1 MgSO4, 2 CaCl2, 5 HEPES, 0.1 EDTA; pH to 7.8), fertilized with minced testis, and allowed to develop at a controlled temperature in 10% MMR. Mixed sex specimens were used at either 12–18 h (Nieuwkoop-Faber [NF] stage 11–16 embryos) or 60 h (NF stage 39–40 larvae) post fertilization.

Tail amputation and pharmacological treatments

NF stage 39–40 larvae were anesthetized with 0.02% tricaine methanesulfonate (TMS, Syndel) with or without the specified drugs until non-responsive. Larvae were then amputated under a dissection stereoscope using a scalpel blade at approximately 1/4 of the length from the tail tip, where the tail begins to taper. The anesthetic was then washed out and the amputated larvae were incubated at 21–23°C in 10% MMR with vehicle or 10 μM cyclopamine (from 10 to 20 mM stock in DMSO; Sigma C4116), 20 μM vismodegib (from 50 mM stock in DMSO; Sigma 879085-55-9), 1 μM SAG (from 5 mM stock in H2O; Calbiochem 566660), 10 μM GANT61 (from 10 mM stock in DMSO; Tocris 3191), and/or 10 μM KT5720 (from 10 mM stock in DMSO; Tocris 1288). For multi-day treatments, solutions were replaced daily.

Regeneration analysis

The amputation plane was defined by the most posterior region of close-packed, organized bands of 12/101 stained muscle cells as regenerated muscle lacks this structure. Outgrowth was defined as the distance between the amputation plane and either the furthest posterior Sox2+ cells associated with the regenerating spinal cord (spinal cord outgrowth) or the furthest posterior 12/101+ muscle cells (muscle outgrowth). Total muscle regeneration was defined as the volume of muscle cells in the regenerate posterior to the amputation plane, not including dorsal/ventral outgrowth of muscle from the intact chevrons in SAG treatment conditions. Sox2 and P-H3 counts were performed respectively on spinal cord and total regenerate posterior to the amputation plane. All these measurements were done using the image analysis software Imaris.

Whole-mount immunostaining

Larvae were anesthetized in 0.02% TMS in 10% MMR, then fixed in 3.7% formaldehyde in 1× MEMFA saline (100 mM MOPS, 2 mM EGTA, 100 mM MgSO4) overnight at 4°C. Samples were bleached overnight in H2O2/Dent’s fixative, then permeabilized with 0.5% Triton X100 in 1× PBS (PBT) and blocked in 0.5% PBT + 2% BSA. Primary and secondary antibody incubations were performed in 0.1% PBT overnight at 4°C. Samples were washed in 0.5% PBT, mounted in 90% glycerol in 1× PBS, and imaged within 1–4 days. Antibodies were obtained and used as follows: 1:300 Sox2 (R&D AF2018, RRID:AB_355110; neural stem cells), 1:100 12/101 (DSHB, RRID:AB_531892; skeletal muscle), and 1:400 phospho-Serine10-histone-H3 (P-H3; Millipore 06-570, RRID:AB_310177, mitotic marker). All donkey secondary antibodies were used at 1:1500–1:2000 from Thermo Fisher: anti-Goat-Alexa-647 (A21447, RRID:AB_141844), anti-Goat-Alexa-594 (A11058, RRID:AB_2534105), anti-Mouse-Alexa-488 (A21202, RRID:AB_141607), anti-Rabbit-Alexa-594 (A21207, RRID:AB_141637), and anti-Rabbit-Alexa-647 (A31573, RRID:AB_2536183). Phospho-CREB (P-CREB) immunostaining required conditions as follows: samples were fixed as above for 3 h at 4°C with gentle agitation, then washed in 0.1% PBT, dehydrated in methanol, and kept at −20°C overnight in methanol. Samples were then bleached at room temperature for 3 hr, rehydrated, permeabilized as above, blocked in 10% BSA + 1.5% normal donkey serum, and placed in primary antibody solution: 1:400 Sox2, 1:100 12/101, and 1:1500 P-CREB (Cell Signaling 9198) for 4–5 days at 4°C, then treated as above for secondary antibody incubation, washes, and mounting.

Immunostaining in tissue sections

Larvae were fixed in 4% PFA for 1 h at room temperature. Samples were then processed for paraffin embedding and sectioned transversely in 10-μm-thick sections. Slides were processed for immunostaining by incubating overnight with the primary antibodies Sox2 (1:300) and 5E1 (1:50, recognizes X. laevis Shh and Ihh) at 4°C and for 2 h at room temperature with AlexaFluor conjugated secondary antibodies (1:300) in 1% BSA, 0.1% Tween in PBS. Immunostained samples were mounted and imaged in an epifluorescence microscope.

Gli activity reporter

p8xGli-EGFP_CMV-iRFP670 was constructed as follows: the 8xGli-EGFP reporter plasmid was a gift from Prof. James Chen of Stanford University (Sasaki et al., 1997). The 8xGli-δcrystallin promoter was removed and spliced into the pXreg4-FireflyLuciferase_keratin-EGFP_CMV-RenillaLuciferase (courtesy of Dr. Yesser Hadj Balgacem, UC Davis, Belgacem and Borodinsky, 2015) in place of the Xreg4-FL_keratin cassette to make p8xGli-EGFP_CMV-RenillaLuciferase. iRFP670 was then PCR-amplified from iRFP670-N1 (Shcherbakova and Verkhusha, 2013, Addgene 45457) and swapped with Renilla Luciferase to make p8xGli-EGFP_CMV-RFP670. 150–200 pg p8xGli-GFP_CMV-iRFP670 were injected per embryo at the 2–4 cell stage. Individual cells were selected in Imaris (Bitplane) from greater than twice background intensity iRFP670+ cells and analyzed for mean EGFP and iRFP670 intensity. The construct was validated in 18 h post-fertilization (hpf), neural plate stage embryos ± 10 nM SAG. Imaging took place at 18 or 60–88 hpf embryos or larvae, respectively. Neural plate stage validation is presented as scatter plots, by iRFP670 (X) and EGFP (Y) signal intensity, with all iRFP670+ cells from all samples pooled (Figure 1—figure supplement 1). Post-amputation Gli activity is presented as two-dimensional reconstructions of all iRFP670+ cells from all samples pooled, presented by each spot’s relationship to the amputation plane (x axis) and the spinal cord (y axis), with each iRFP670+ cell assigned a heat map intensity by EGFP:iRFP670 ratio (Figure 2).

Gli2 knockdown

Morpholino antisense oligonucleotides targeted to block Gli2 translation (Belgacem and Borodinsky, 2015, 5′-GCACAGAACGCAGGTAATGCTCCAT-3′, Gli2-MO) and matched Gli2-PhotoMO (5′-ATGGAGCATTACPTGCGTTCT-3′) were ordered from Gene Tools. During and after injection, all steps were performed in the dark, using only >580 nm light for illumination. Embryos were injected at 4-cell stage with 4 nl of solution containing a total of 2 pmol each Gli2-MO and PhotoMO blocker with Cascade Blue-Dextran tracer and grown at 20–22°C until 32 hpf. Larvae were then split into two separate Petri dishes in 10% MMR, and either kept in the dark (inactive control) or exposed to 30 min of 365 nm UV transillumination, low-output setting on a MaestroGen 240 V UV transilluminator (active MO). Larvae were then left at 20–22°C until 28 h post-UV, then amputated, and kept at 22–23°C until anesthesia, screening for tracer and fixation at 72 hpa.

P-CREB+ cell quantification

In whole mounts

Sox2 and P-CREB immunopositive cells within the spinal cord of immunostained whole mounts were separately analyzed on Imaris software (Bitplane). All cells were filtered for minimum 3× background Sox2 or P-CREB mean signal intensity, and background subtracted for mean Sox2 or P-CREB intensity. P-CREB+ cell counts are presented as the ratio of numbers of P-CREB+ cells per total number of Sox2+ nuclei for each 100 μm bin along the spinal cord posterior and anterior to the amputation plane for controls (data presented in Figure 6B), and as raw P-CREB count values for conditions treated with vehicle, SAG, or cyclopamine (data presented in Figure 7D).

In sections

Larvae were amputated, incubated for 4 h with DMSO (control), 10 μM cyclopamine or 1 μM SAG, then fixed in 4% PFA for 1 h at room temperature. Samples were then processed for paraffin embedding and sectioned longitudinally in 10-μm-thick sections. Slides were processed for immunostaining by incubating overnight with the primary antibodies Sox2 (1:300), P-CREB (1:800) at 4°C and for 2 h at room temperature with AlexaFluor conjugated secondary antibodies (1:300) in 1% BSA, 0.1% Tween in PBS. Immunostained samples were mounted and imaged in an epifluorescence microscope. Number of P-CREB and Sox2 immunopositive cells was quantified by thresholding the signal to twice the background and normalized per 100 μm extension of the spinal cord, determined by the length of Sox2-labeled tissue. Data was collected considering as 0 μm the tip of the tail in 4-hpa larvae.

Live and fixed sample imaging

Live larvae anesthetized with 0.02% TMS in 10% MMR were imaged under a Nikon swept-field confocal microscope using 488 nm (EGFP) and 647 nm (iRFP670) lasers. Fixed samples were imaged on Nikon-A1 or C2 point laser-scanning confocal microscopes using 488 nm (Alexa488), 561 nm (Alexa594), and 640 nm (Alexa647) lasers or using an Olympus epifluorescence microscope for immunostained sections.

Western blot assay

Nuclear fraction was obtained from stage 39–40 Gli2MO + PhotoMO larvae (three larvae for each group) to assess endogenous expression of Gli2. Briefly, larvae injected with Gli2MO + PhotoMO +/– UV (as described above) were frozen in liquid nitrogen, stored at −80°C, then homogenized in 25 mM HEPES pH 7.4, 50 mM NaCl, 2 mM EGTA, 5 mM MgCl2, protease inhibitors cocktail (784115, Thermo Fisher Scientific) on ice for 30 min and centrifuged for 10 min at 1000 g. Nuclear pellets were resuspended in 2× protein loading buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% (w/v) glycerol, 0.005% Bromophenol Blue, 5% β-mercaptoethanol) and boiled for 5 min. Samples were run in 10% SDS-PAGE and transferred to PVDF membrane. PVDF membrane was probed with anti-Gli2 goat polyclonal (AF3635; RRID:AB_211902), 1:800 in 5% BSA at 4°C, followed by incubation with horseradish peroxidase (HRP)-conjugated secondary antibody (711-035-152, Jackson ImmunoResearch; 1:10,000) and visualized by Western Lightning Plus-ECL, Enhanced Chemiluminescence Substrate (NEL103E001, Perkin Elmer). PVDF membranes were stripped in 0.2 M glycine HCl buffer, pH 2.5, 0.05% Tween for 20 min and re-probed with 1:500 anti-LaminII/III (Cell Signaling 9087; RRID:AB_10896336) for nucleus-specific loading control in 5% BSA.

For assessment of Hh ligand expression, whole-cell lysates were obtained from 500-μm-long stump/regenerate from 0 and 24 hpa larvae (35/group) by freezing samples in liquid nitrogen, then solubilizing in 1× Laemmli buffer and centrifuging for 5 min at 17,900 g. Supernatant containing 2.5% β-mercaptoethanol was boiled for 4 min. Samples were run in 4–20% gradient gel SDS-PAGE and transferred to PVDF membrane, which was probed with anti-Hh antibody 5E1 (DSBH), 1:50 in 5% BSA at 4°C overnight, followed by incubation with HRP-conjugated goat anti-mouse secondary antibody (Millipore, 12-349) and visualized by ECL2. PVDF membranes were stripped in 0.2 M glycine HCl buffer, pH 2.5, 0.05% Tween for 20 min and re-probed with 1:1000 anti-GAPDH (SICGEN; RRID:AB_0049-200) for whole-cell loading control in 5% BSA, followed by incubation with HRP-conjugated rabbit anti-goat secondary antibody (R&D, HAF109).

All membranes were imaged with ChemiDoc-MP imaging instrument and optical density of bands of interest measured with associated software (Bio-Rad Laboratories).

Experimental design and statistical analyses

All data were analyzed with Prism software (GraphPad). Data were first analyzed for normality, followed by parametric (normally distributed) or non-parametric tests (not normally distributed). When normally distributed and SDs were equal among groups, unpaired two-tail t-test or ordinary one-way ANOVA followed by Tukey’s multiple comparisons test was used, when two or more groups were compared, respectively. When normally distributed and SDs were not equal among groups, Welch’s t-test or Brown–Forsythe and Welch ANOVA followed by Dunnett’s T3 multiple comparisons test was used, when two or more groups were compared, respectively. When data were not normally distributed, Mann–Whitney test or Kruskal–Wallis followed by Dunn’s multiple comparisons test was used, when two or more groups were compared, respectively. Graphed values are presented as a normalized percent of the experiment-matched control treatment, pooled across experiments. Error bars for the pooled control groups coming from different experiments in Figures 1, 3, and 4, Figure 4—figure supplement 1, Figures 5 and 7D represent the largest error value for the control group in any of the experimental groups included in each graph. Significance was set to p<0.05. Number of samples and experiments is indicated in figure legends.

Acknowledgements

We thank Dr. Yesser Hadj Belgacem for comments on the manuscript. This work was supported by NSF 1120796 and 1754340, NIH-NINDS R01NS073055, R01NS105886, R01NS113859, and Shriners Hospital for Children 86700-NCA grants to LNB and Shriners Hospital for Children Postdoctoral Fellowship to AMH. We thank Christopher Hom for assistance with data acquisition and analysis and Jacqueline Levin for technical advice and assistance.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Andrew M Hamilton, Email: andrewmichaelhamilton@gmail.com.

Laura N Borodinsky, Email: lnborodinsky@ucdavis.edu.

Paola Bovolenta, CSIC-UAM, Spain.

Marianne E Bronner, California Institute of Technology, United States.

Funding Information

This paper was supported by the following grants:

  • National Science Foundation 1120796 to Laura N Borodinsky.

  • National Institute of Neurological Disorders and Stroke R01NS073055 to Laura N Borodinsky.

  • Shriners Hospitals for Children 86700-NCA to Laura N Borodinsky.

  • Shriners Hospitals for Children 84303-NCA to Andrew M Hamilton.

  • National Science Foundation 1754340 to Laura N Borodinsky.

  • National Institute of Neurological Disorders and Stroke R01NS105886 to Laura N Borodinsky.

  • National Institute of Neurological Disorders and Stroke R01NS113859 to Laura N Borodinsky.

Additional information

Competing interests

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing.

Data curation, Formal analysis, Validation, Investigation, Methodology.

Conceptualization, Supervision, Funding acquisition, Writing - original draft, Project administration, Writing - review and editing.

Ethics

Animal experimentation: This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of the animals were handled according to approved institutional animal care and use committee (IACUC) protocols (#20537) of the University of California Davis. This institution is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care, International (AAALAC). This institution has an Animal Welfare Assurance on file with the Office of Laboratory Animal Welfare (OLAW). The Assurance Number is A3433-01. All procedures were performed under tricaine anesthesia, and every effort was made to minimize suffering.

Additional files

Transparent reporting form

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files.

References

  1. Akazawa C, Tsuzuki H, Nakamura Y, Sasaki Y, Ohsaki K, Nakamura S, Arakawa Y, Kohsaka S. The upregulated expression of sonic hedgehog in motor neurons after rat facial nerve axotomy. Journal of Neuroscience. 2004;24:7923–7930. doi: 10.1523/JNEUROSCI.1784-04.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aztekin C, Hiscock TW, Marioni JC, Gurdon JB, Simons BD, Jullien J. Identification of a regeneration-organizing cell in the Xenopus tail. Science. 2019;364:653–658. doi: 10.1126/science.aav9996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aztekin C, Hiscock TW, Butler R, De Jesús Andino F, Robert J, Gurdon JB, Jullien J. The myeloid lineage is required for the emergence of a regeneration-permissive environment following Xenopus tail amputation. Development. 2020;147:dev185496. doi: 10.1242/dev.185496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Balaskas N, Ribeiro A, Panovska J, Dessaud E, Sasai N, Page KM, Briscoe J, Ribes V. Gene regulatory logic for reading the sonic hedgehog signaling gradient in the vertebrate neural tube. Cell. 2012;148:273–284. doi: 10.1016/j.cell.2011.10.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bambakidis NC, Horn EM, Nakaji P, Theodore N, Bless E, Dellovade T, Ma C, Wang X, Preul MC, Coons SW, Spetzler RF, Sonntag VKH. Endogenous stem cell proliferation induced by intravenous hedgehog agonist administration after contusion in the adult rat spinal cord. Journal of Neurosurgery: Spine. 2009;10:171–176. doi: 10.3171/2008.10.SPI08231. [DOI] [PubMed] [Google Scholar]
  6. Bambakidis NC, Petrullis M, Kui X, Rothstein B, Karampelas I, Kuang Y, Selman WR, LaManna JC, Miller RH. Improvement of neurological recovery and stimulation of neural progenitor cell proliferation by intrathecal administration of sonic hedgehog. Journal of Neurosurgery. 2012;116:1114–1120. doi: 10.3171/2012.1.JNS111285. [DOI] [PubMed] [Google Scholar]
  7. Bambakidis NC, Onwuzulike K. Sonic hedgehog signaling and potential therapeutic indications. Vitamins and Hormones. 2012;88:379–394. doi: 10.1016/B978-0-12-394622-5.00017-1. [DOI] [PubMed] [Google Scholar]
  8. Beck CW, Christen B, Slack JM. Molecular pathways needed for regeneration of spinal cord and muscle in a vertebrate. Developmental Cell. 2003;5:429–439. doi: 10.1016/S1534-5807(03)00233-8. [DOI] [PubMed] [Google Scholar]
  9. Beck CW. Studying regeneration in Xenopus. Methods in Molecular Biology. 2012;917:525–539. doi: 10.1007/978-1-61779-992-1_30. [DOI] [PubMed] [Google Scholar]
  10. Belgacem Y, Hamilton A, Shim S, Spencer K, Borodinsky L. The many hats of sonic hedgehog signaling in nervous system development and disease. Journal of Developmental Biology. 2016;4:35. doi: 10.3390/jdb4040035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Belgacem YH, Borodinsky LN. Sonic hedgehog signaling is decoded by calcium spike activity in the developing spinal cord. PNAS. 2011;108:4482–4487. doi: 10.1073/pnas.1018217108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Belgacem YH, Borodinsky LN. Inversion of sonic hedgehog action on its canonical pathway by electrical activity. PNAS. 2015;112:4140–4145. doi: 10.1073/pnas.1419690112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Borodinsky LN, Root CM, Cronin JA, Sann SB, Gu X, Spitzer NC. Activity-dependent homeostatic specification of transmitter expression in embryonic neurons. Nature. 2004;429:523–530. doi: 10.1038/nature02518. [DOI] [PubMed] [Google Scholar]
  14. Briscoe J, Thérond PP. The mechanisms of hedgehog signalling and its roles in development and disease. Nature Reviews Molecular Cell Biology. 2013;14:416–429. doi: 10.1038/nrm3598. [DOI] [PubMed] [Google Scholar]
  15. Chen Y, Yue S, Xie L, Pu X, Jin T, Cheng SY. Dual phosphorylation of suppressor of fused (Sufu) by PKA and GSK3β regulates its stability and localization in the primary cilium*. Journal of Biological Chemistry. 2011;286:13502–13511. doi: 10.1074/jbc.M110.217604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chen Y, Love NR, Amaya E. Tadpole tail regeneration in Xenopus. Biochemical Society Transactions. 2014;42:617–623. doi: 10.1042/BST20140061. [DOI] [PubMed] [Google Scholar]
  17. Dasari VR, Veeravalli KK, Dinh DH. Mesenchymal stem cells in the treatment of spinal cord injuries: a review. World Journal of Stem Cells. 2014;6:120–133. doi: 10.4252/wjsc.v6.i2.120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Davies SP, Reddy H, Caivano M, Cohen P. Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochemical Journal. 2000;351:95–105. doi: 10.1042/bj3510095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dunaeva M, Waltenberger J. Hh signaling in regeneration of the ischemic heart. Cellular and Molecular Life Sciences. 2017;74:3481–3490. doi: 10.1007/s00018-017-2534-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ekker SC, McGrew LL, Lai CJ, Lee JJ, von Kessler DP, Moon RT, Beachy PA. Distinct expression and shared activities of members of the hedgehog gene family of Xenopus laevis. Development. 1995;121:2337–2347. doi: 10.1242/dev.121.8.2337. [DOI] [PubMed] [Google Scholar]
  21. Ferrari MB, Rohrbough J, Spitzer NC. Spontaneous calcium transients regulate myofibrillogenesis in embryonic Xenopus myocytes. Developmental Biology. 1996;178:484–497. doi: 10.1006/dbio.1996.0233. [DOI] [PubMed] [Google Scholar]
  22. Ferrari MB, Spitzer NC. Calcium signaling in the developing Xenopus myotome. Developmental Biology. 1999;213:269–282. doi: 10.1006/dbio.1999.9387. [DOI] [PubMed] [Google Scholar]
  23. Ferreira F, Raghunathan V, Luxardi G, Zhu K, Zhao M. Early redox activities modulate Xenopus tail regeneration. Nature Communications. 2018;9:4296. doi: 10.1038/s41467-018-06614-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fukazawa T, Naora Y, Kunieda T, Kubo T. Suppression of the immune response potentiates tadpole tail regeneration during the refractory period. Development. 2009;136:2323–2327. doi: 10.1242/dev.033985. [DOI] [PubMed] [Google Scholar]
  25. Gomes-Osman J, Cortes M, Guest J, Pascual-Leone A. A systematic review of experimental strategies aimed at improving motor function after acute and chronic spinal cord injury. Journal of Neurotrauma. 2016;33:425–438. doi: 10.1089/neu.2014.3812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Gonzalez GA, Montminy MR. Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133. Cell. 1989;59:675–680. doi: 10.1016/0092-8674(89)90013-5. [DOI] [PubMed] [Google Scholar]
  27. Grzelak CA, Martelotto LG, Sigglekow ND, Patkunanathan B, Ajami K, Calabro SR, Dwyer BJ, Tirnitz-Parker JE, Watkins DN, Warner FJ, Shackel NA, McCaughan GW. The intrahepatic signalling niche of hedgehog is defined by primary cilia positive cells during chronic liver injury. Journal of Hepatology. 2014;60:143–151. doi: 10.1016/j.jhep.2013.08.012. [DOI] [PubMed] [Google Scholar]
  28. Gu X, Spitzer NC. Distinct aspects of neuronal differentiation encoded by frequency of spontaneous Ca2+ transients. Nature. 1995;375:784–787. doi: 10.1038/375784a0. [DOI] [PubMed] [Google Scholar]
  29. Hadden MK. Hedgehog pathway agonism: therapeutic potential and small-molecule development. ChemMedChem. 2014;9:27–37. doi: 10.1002/cmdc.201300358. [DOI] [PubMed] [Google Scholar]
  30. Hammerschmidt M, Bitgood MJ, McMahon AP. Protein kinase A is a common negative regulator of hedgehog signaling in the vertebrate embryo. Genes & Development. 1996;10:647–658. doi: 10.1101/gad.10.6.647. [DOI] [PubMed] [Google Scholar]
  31. Ho Wei L, Arastoo M, Georgiou I, Manning DR, Riobo-Del Galdo NA. Activation of the gi protein-RHOA Axis by non-canonical hedgehog signaling is independent of primary cilia. PLOS ONE. 2018;13:e0203170. doi: 10.1371/journal.pone.0203170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Howell M, Inman GJ, Hill CS. A novel Xenopus Smad-interacting forkhead transcription factor (XFast-3) cooperates with XFast-1 in regulating gastrulation movements. Development. 2002;129:2823–2834. doi: 10.1242/dev.129.12.2823. [DOI] [PubMed] [Google Scholar]
  33. Kakebeen AD, Chitsazan AD, Williams MC, Saunders LM, Wills AE. Chromatin accessibility dynamics and single cell RNA-Seq reveal new regulators of regeneration in neural progenitors. eLife. 2020;9:e52648. doi: 10.7554/eLife.52648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kawagishi H, Xiong J, Rovira II, Pan H, Yan Y, Fleischmann BK, Yamada M, Finkel T. Sonic hedgehog signaling regulates the mammalian cardiac regenerative response. Journal of Molecular and Cellular Cardiology. 2018;123:180–184. doi: 10.1016/j.yjmcc.2018.09.005. [DOI] [PubMed] [Google Scholar]
  35. Koide T, Hayata T, Cho KW. Negative regulation of hedgehog signaling by the cholesterogenic enzyme 7-dehydrocholesterol reductase. Development. 2006;133:2395–2405. doi: 10.1242/dev.02393. [DOI] [PubMed] [Google Scholar]
  36. Lauth M, Bergström A, Shimokawa T, Toftgård R. Inhibition of GLI-mediated transcription and tumor cell growth by small-molecule antagonists. PNAS. 2007;104:8455–8460. doi: 10.1073/pnas.0609699104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lee J, Platt KA, Censullo P. Gli1 is a target of Sonic hedgehog that induces ventral neural tube development. Development. 1997;124:2537–2552. doi: 10.1242/dev.124.13.2537. [DOI] [PubMed] [Google Scholar]
  38. Lee BB, Cripps RA, Fitzharris M, Wing PC. The global map for traumatic spinal cord injury epidemiology: update 2011, global incidence rate. Spinal Cord. 2014;52:110–116. doi: 10.1038/sc.2012.158. [DOI] [PubMed] [Google Scholar]
  39. Love NR, Chen Y, Ishibashi S, Kritsiligkou P, Lea R, Koh Y, Gallop JL, Dorey K, Amaya E. Amputation-induced reactive oxygen species are required for successful Xenopus tadpole tail regeneration. Nature Cell Biology. 2013;15:222–228. doi: 10.1038/ncb2659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lu P, Wang Y, Graham L, McHale K, Gao M, Wu D, Brock J, Blesch A, Rosenzweig ES, Havton LA, Zheng B, Conner JM, Marsala M, Tuszynski MH. Long-distance growth and connectivity of neural stem cells after severe spinal cord injury. Cell. 2012;150:1264–1273. doi: 10.1016/j.cell.2012.08.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mani S, Radhakrishnan S, Cheramangalam RN, Harkar S, Rajendran S, Ramanan N. Shh-Mediated increase in β-Catenin levels maintains cerebellar granule neuron progenitors in proliferation. The Cerebellum. 2020;19:645–664. doi: 10.1007/s12311-020-01138-2. [DOI] [PubMed] [Google Scholar]
  42. Méndez-Olivos EE, Muñoz R, Larraín J. Spinal cord cells from Pre-metamorphic stages differentiate into neurons and promote axon growth and regeneration after transplantation into the injured spinal cord of Non-regenerative Xenopus laevis Froglets. Frontiers in Cellular Neuroscience. 2017;11:398. doi: 10.3389/fncel.2017.00398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Murray AJ. Pharmacological PKA inhibition: all may not be what it seems. Science Signaling. 2008;1:re4. doi: 10.1126/scisignal.122re4. [DOI] [PubMed] [Google Scholar]
  44. Polizio AH, Chinchilla P, Chen X, Manning DR, Riobo NA. Sonic hedgehog activates the GTPases Rac1 and RhoA in a Gli-Independent manner through coupling of smoothened to gi proteins. Science Signaling. 2011;4:pt7. doi: 10.1126/scisignal.2002396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Qu C, Liu Y, Kunkalla K, Singh RR, Blonska M, Lin X, Agarwal NK, Vega F. Trimeric G protein-CARMA1 Axis links smoothened, the hedgehog receptor transducer, to NF-κB activation in diffuse large B-cell lymphoma. Blood. 2013;121:4718–4728. doi: 10.1182/blood-2012-12-470153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Riobó NA, Lu K, Ai X, Haines GM, Emerson CP. Phosphoinositide 3-kinase and akt are essential for sonic hedgehog signaling. PNAS. 2006;103:4505–4510. doi: 10.1073/pnas.0504337103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Romero MMG, McCathie G, Jankun P, Roehl HH. Damage-induced reactive oxygen species enable zebrafish tail regeneration by repositioning of hedgehog expressing cells. Nature Communications. 2018;9:4010. doi: 10.1038/s41467-018-06460-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ruat M, Hoch L, Faure H, Rognan D. Targeting of smoothened for therapeutic gain. Trends in Pharmacological Sciences. 2014;35:237–246. doi: 10.1016/j.tips.2014.03.002. [DOI] [PubMed] [Google Scholar]
  49. Sasaki H, Hui C, Nakafuku M, Kondoh H. A binding site for gli proteins is essential for HNF-3beta floor plate enhancer activity in transgenics and can respond to shh in vitro. Development. 1997;124:1313–1322. doi: 10.1242/dev.124.7.1313. [DOI] [PubMed] [Google Scholar]
  50. Shcherbakova DM, Verkhusha VV. Near-infrared fluorescent proteins for multicolor in vivo imaging. Nature Methods. 2013;10:751–754. doi: 10.1038/nmeth.2521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Sherpa T, Lankford T, McGinn TE, Hunter SS, Frey RA, Sun C, Ryan M, Robison BD, Stenkamp DL. Retinal regeneration is facilitated by the presence of surviving neurons. Developmental Neurobiology. 2014;74:851–876. doi: 10.1002/dneu.22167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Singh BN, Doyle MJ, Weaver CV, Koyano-Nakagawa N, Garry DJ. Hedgehog and wnt coordinate signaling in myogenic progenitors and regulate limb regeneration. Developmental Biology. 2012;371:23–34. doi: 10.1016/j.ydbio.2012.07.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Singh BN, Koyano-Nakagawa N, Donaldson A, Weaver CV, Garry MG, Garry DJ. Hedgehog signaling during appendage development and regeneration. Genes. 2015;6:417–435. doi: 10.3390/genes6020417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Singh BN, Weaver CV, Garry MG, Garry DJ. Hedgehog and wnt signaling pathways regulate tail regeneration. Stem Cells and Development. 2018a;27:1426–1437. doi: 10.1089/scd.2018.0049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Singh BN, Koyano-Nakagawa N, Gong W, Moskowitz IP, Weaver CV, Braunlin E, Das S, van Berlo JH, Garry MG, Garry DJ. A conserved HH-Gli1-Mycn network regulates heart regeneration from newt to human. Nature Communications. 2018b;9:4237. doi: 10.1038/s41467-018-06617-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sloan TF, Qasaimeh MA, Juncker D, Yam PT, Charron F. Integration of shallow gradients of shh and Netrin-1 guides commissural axons. PLOS Biology. 2015;13:e1002119. doi: 10.1371/journal.pbio.1002119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sugimoto K, Hui SP, Sheng DZ, Kikuchi K. Dissection of zebrafish shha function using site-specific targeting with a Cre-dependent genetic switch. eLife. 2017;6:e24635. doi: 10.7554/eLife.24635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Taniguchi Y, Sugiura T, Tazaki A, Watanabe K, Mochii M. Spinal cord is required for proper regeneration of the tail in Xenopus tadpoles: spinal cord ablation in regeneration. Development, Growth & Differentiation. 2008;50:109–120. doi: 10.1111/j.1440-169X.2007.00981.x. [DOI] [PubMed] [Google Scholar]
  59. Taniguchi Y, Watanabe K, Mochii M. Notochord-derived hedgehog is essential for tail regeneration in Xenopus tadpole. BMC Developmental Biology. 2014;14:27. doi: 10.1186/1471-213X-14-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Tapia VS, Herrera-Rojas M, Larrain J. JAK-STAT pathway activation in response to spinal cord injury in regenerative and non-regenerative stages of Xenopus laevis. Regeneration. 2017;4:21–35. doi: 10.1002/reg2.74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Teperino R, Amann S, Bayer M, McGee SL, Loipetzberger A, Connor T, Jaeger C, Kammerer B, Winter L, Wiche G, Dalgaard K, Selvaraj M, Gaster M, Lee-Young RS, Febbraio MA, Knauf C, Cani PD, Aberger F, Penninger JM, Pospisilik JA, Esterbauer H. Hedgehog partial agonism drives Warburg-like metabolism in muscle and Brown fat. Cell. 2012;151:414–426. doi: 10.1016/j.cell.2012.09.021. [DOI] [PubMed] [Google Scholar]
  62. Teperino R, Aberger F, Esterbauer H, Riobo N, Pospisilik JA. Canonical and non-canonical hedgehog signalling and the control of metabolism. Seminars in Cell & Developmental Biology. 2014;33:81–92. doi: 10.1016/j.semcdb.2014.05.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Tian A, Shi Q, Jiang A, Li S, Wang B, Jiang J. Injury-stimulated hedgehog signaling promotes regenerative proliferation of Drosophila intestinal stem cells. Journal of Cell Biology. 2015;208:807–819. doi: 10.1083/jcb.201409025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Tseng AS, Adams DS, Qiu D, Koustubhan P, Levin M. Apoptosis is required during early stages of tail regeneration in Xenopus laevis. Developmental Biology. 2007;301:62–69. doi: 10.1016/j.ydbio.2006.10.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Tseng AS, Beane WS, Lemire JM, Masi A, Levin M. Induction of vertebrate regeneration by a transient sodium current. Journal of Neuroscience. 2010;30:13192–13200. doi: 10.1523/JNEUROSCI.3315-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Tu MK, Borodinsky LN. Spontaneous calcium transients manifest in the regenerating muscle and are necessary for skeletal muscle replenishment. Cell Calcium. 2014;56:34–41. doi: 10.1016/j.ceca.2014.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Wu F, Zhang Y, Sun B, McMahon AP, Wang Y. Hedgehog signaling: from basic biology to Cancer therapy. Cell Chemical Biology. 2017;24:252–280. doi: 10.1016/j.chembiol.2017.02.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Yakushiji N, Suzuki M, Satoh A, Ide H, Tamura K. Effects of activation of hedgehog signaling on patterning, growth, and differentiation in Xenopus froglet limb regeneration. Developmental Dynamics. 2009;238:1887–1896. doi: 10.1002/dvdy.22011. [DOI] [PubMed] [Google Scholar]
  69. Yam PT, Langlois SD, Morin S, Charron F. Sonic hedgehog guides axons through a noncanonical, Src-family-kinase-dependent signaling pathway. Neuron. 2009;62:349–362. doi: 10.1016/j.neuron.2009.03.022. [DOI] [PubMed] [Google Scholar]
  70. Yin A, Winata CL, Korzh S, Korzh V, Gong Z. Expression of components of wnt and hedgehog pathways in different tissue layers during lung development in Xenopus laevis. Gene Expression Patterns. 2010;10:338–344. doi: 10.1016/j.gep.2010.07.005. [DOI] [PubMed] [Google Scholar]

Decision letter

Editor: Paola Bovolenta1
Reviewed by: Henry Roehl2

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This study shows that Hh signaling contributes to the regeneration of the Xenopus to spinal cord and muscle, after tail amputation at larva stages. The study confirms previous reports and further adds the notion that this function is mediated by a non-canonical branch of the pathway involving PKA and CREB underlies this process.

Decision letter after peer review:

Thank you for submitting your article "Non-canonical Hedgehog signaling regulates spinal cord and muscle regeneration" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Marianne Bronner as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Henry Roehl (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

As the editors have judged that your manuscript is of interest, but as described below that additional experiments are required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

Summary:

Hh signalling contributes to the generation of a number of tissues. The present manuscript investigates how this signalling contributes to the regeneration of the Xenopus to spinal cord and muscle, after tail amputation at larva stages. The manuscript suggests that a non-canonical branch of the pathway involving PKA and CREB underlies this process.

The manuscript is well written and easy to follow. However the finding that distinguishes this manuscript from previous related reports -namely the involvement of non-canonical signalling in Hh-mediated regeneration- is poorly supported by the current data and needs to be strengthened by further experimentation.

Essential revisions:

1. The authors need to provide the rationale for using st 39-40 larvae instead older stages, commonly accepted as regenerative stages (i.e. st 49 to 54). This should be included in the method/results, further discussing the potential advantages/ disadvantages. More information on the timing of full regeneration at used stages should also be provided.

2. The authors should address the question of whether the activation of the non canonical pathway is ligand dependent or independent, among other by analysing whether HH ligands are in the initial 24 hr of regeneration.

3. The chemical inhibitors used in the study are not fully specific and therefore additional approaches should be used to validate the findings. One possibility is to use Photo-MO for Smoothened and CREB. Additional methods to support the involvement of PKA are also needed.

4. A direct link between pCREB and non-canonical HH signalling is needed. This could be provided by testing pCREB reduced after Smoothened inhibition during regeneration.

Reviewer #1:

This manuscript shows that spinal cord and muscle regeneration in the Xenopus larvae is mediated by non-canonical Hh signaling through the activity of PKA. This conclusion is based on a series of pharmacological treatments and a careful analysis of a number of quantified parameters.

The study adds to the idea that Hh signaling is important for tissue regeneration, with the additional interesting idea that the process is mediated, at least at early stages, by non-canonical signaling. The enthusiasm for these results is however undermined by a series of concerns:

According to the literature, st 49 to 54 are considered as the regenerative stages in Xenopus laevis (Edwards-Faret et al. 2017 Nat Protocols). Why have the authors used larvae at st 39-40? Can these earlier stages entail a contribution from developmental related aspects? How long does it take to reach full regeneration in these larvae? This information is missing and needs to be provided for non-specialists. This is important also because Figure 2 shows that Gli transcriptional activity begins to recover at 28 hr. When does it get back to normal levels? Is it before than 72 hrs (time used for most of the analysis)? Can Gli activity account for the differences observed between KT5720 and KT5720+ cyclop treatments?

On the same line: does treatment with GANT61 beginning 48hr (or so) after amputation has any effect on regeneration parameters?

The experiments reported in Figure 6 do not really prove that P-CREB mediates regeneration. Can the authors attempt to down-modulate its activity?

Reviewer #2:

In this manuscript the authors examine how Hh signalling is involved in spinal cord and muscle regeneration in the Xenopus laevis. The important finding is that a non-canonical branch of the Hedgehog pathway involving Protein Kinase A and CREB is acting early during Xenopus larval tail regeneration. The paper is well-written and the findings are very interesting, but further work is needed to warrant publication in eLife. We feel that these experiments fall within the authors expertise and could be accomplished within a reasonable timeframe.

1. Using the same regeneration model, Taniguchi et al. (2014) showed that all HH ligands become expressed in the notochord about 24 hours after injury. The analysis presented here is focused before this time. Is the hypothesis that non-canonical HH signalling is ligand independent? If so the Authors should confirm that HH ligands are not expressed during the period of their analysis (i.e. between 0 and 24 hours after injury).

2. The paper relies very heavily on chemical inhibitors which may not be entirely specific (see also point 3). The authors should consider using Photo-MO for Smothened to confirm some key results. This would be especially convincing when compared to the Gli MO (which did not have a strong effect).

3. KT5720 is used as an inhibitor of PKA. There are reports of potential off target effects including inhibiting MAPK (Pharmacological PKA Inhibition: All May Not Be What It Seems, Andrew J. Murray, 2008). MAPK can act upstream of CREB too and is known to be activated immediately after wounding (Christen and Slack, 1999). Could another method be used to strengthen the PKA hypothesis?

4. The Authors use pCREB as a read-out for non-canonical HH signalling in Figure 6. They need to show that pCREB is reduced by Smoothened inhibition during regeneration. What is the significance of its presence in skin after wounding, is that also dependent upon non-canonical HH signalling?

Reviewer #3:

This manuscript demonstrated the role of non-canonical Hh signaling in spinal cord and muscle regeneration in Xenopus tadpoles. The downstream effectors of this non-canonical Hh signaling involves PKA and CREB activation. The finding provides an explanation of the mechanisms of Hh signaling mediated tissue regeneration. Thus, it is of some interest for regenerative biology.

I have two concerns about this study.

First, for tadpole tail amputations, the authors used NF stages 39-40 embryos. While there is no consensus in the community on the Xenopus tadpole stages used for regeneration analysis, the tail buds of stage 39-40 embryos are not typically considered as differentiated tail, and the cells in the tail buds are precursor cells, rather than differentiated tissue cells. This is particularly true for the cells in the spinal cord. As has been well established, there is ontogenetic decline in the regeneration capacity of Xenopus tadpoles. Thus the stages of tadpoles used for tail amputation is an important concern when performing regeneration analysis.

Second, for Gli activity reporting, the authors injected EGFP:iRFP670 plasmid in the early embryos. This is not ideal, as transgenic tadpoles can be easily generated now.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Non-canonical Hedgehog signaling regulates spinal cord and muscle regeneration in Xenopus laevis larvae" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Marianne Bronner as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Henry Roehl (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

The manuscript is much improved but there are a couple of key issues that still need to be carefully addressed and included in the manuscript to assure its publication in eLife.

Essential revisions:

1. The first relates to the question of whether Hh ligand expression is activated upon amputation at stage 40. Staining of Hh ligands needs to be analyzed and included in the manuscript. This is an important support for the model presented in Figure 9.

2. Graphs should be better presented. Error bars for controls are not treated consistently.

3. The method section related to how regeneration is measured in Figure 1—figure supplement should be expanded and include an explanation on what is the difference between the data presentation of Figure 1 and figure supplement 1.

Reviewer #1:

The manuscript has been revised addressing the comments raise during the first round of review. The manuscript will add new information of the role of Hh signalling in regeneration

Reviewer #2:

The responses to the reviewers' comments are on the whole satisfactory, except around the question of whether there is evidence that Hedgehog ligand expression is activated by excision at stage 40.

The authors state that "Taniguchi et al. reported an upregulation in shh expression upon amputation in the regenerating tail, but they do not argue that Hh is not expressed prior to amputation or during the first 24 hours, nor do their findings support this position." This is not accurate. Figure 1A of Taniguchi et al. analyses expression of shh and bhh over time. At 0hrs post amputation expression of shh is barely detectable, and bhh is not detected. At 24hrs post amputation there is no expression of either ligand. Starting at 48 hours post amputation, shh is induced strongly and bhh to a lesser degree. This expression is stated as being exclusively in the notochord. This has led to the model that Hedgehog signalling becomes slowly activated peaking four to five days after amputation. The authors should familiarise themselves with the literature on the role of Hedgehog signalling during tail regeneration. The comparison with other published analysis of tail regeneration is important even if it is at a different stage.

The authors show a staining of Hedgehog ligands in their rebuttal but do not present analysis of this data and have not added the data to their manuscript. It appears to show that Hedgehog ligands are strongly upregulated within 4 hours after amputation, and that this activation is primarily in the neural tube (not notochord?). This finding would be consistent with the author's model that Hedgehog signalling is activated by wounding and stimulates muscle and neural tube regeneration (Figure 9). Importantly this data would indicate that Hedgehog expression becomes active much earlier than has been previously described for stage 49. This data should be properly analysed and presented in the manuscript. It would provide important support for their model. This is not a labour intensive request.

Reviewer #3:

The revised manuscript added new data from additional experiment, explained the reasoning of using stage 39-40 tadpoles for tail amputation experiment, and replaced the western blot images showing the efficiency of Gli2-mo. The revision adequately addressed my concerns.

However, I find it confusing in understanding the graph presentations, especially with the new data. Error bars for controls are not treated consistently. For example, in Figure 1D-I, there are no error bars shown for the control groups, but in Figure 1—figure supplement 1 B-G, there are error bars shown for the control. Yet, the means for the controls in Figure 1—figure supplement 1 B-G are all 100. Could you please explain a bit more in the method section, how the regeneration is measured in Figure 1—figure supplement? What is the difference between the data presentation of Figure 1 and figure supplement 1?

eLife. 2021 May 6;10:e61804. doi: 10.7554/eLife.61804.sa2

Author response

Essential revisions:

1. The authors need to provide the rationale for using st 39-40 larvae instead older stages, commonly accepted as regenerative stages (i.e. st 49 to 54). This should be included in the method/results, further discussing the potential advantages/ disadvantages. More information on the timing of full regeneration at used stages should also be provided.

A review of the literature on studies using Xenopus larvae as a model for tail regeneration results in a total of 63 original research articles, with 37 in which larvae of stages 39-42 are used and 26 in which stages older than 48 are used. Hence, the earlier stages that we used in this study are well accepted in the field as a good model for tail regeneration in this system. The reasoning behind this is that unlike the tip of the tail that is still developing and growing, the tissues of the stump that become exposed after amputation correspond to mature larval tissues, thus the replenishing of these tissues represents a regenerative process and not the mere developmental progression of tissue growth.

Nevertheless, this is an important point that we are thankful the reviewers brought to our attention so that we can better introduce the readers to this model system. We revised the Results section accordingly (Page 5, new paragraph 1).

2. The authors should address the question of whether the activation of the non canonical pathway is ligand dependent or independent, among other by analysing whether HH ligands are in the initial 24 hr of regeneration.

Expression of Shh has been documented during these larval stages mostly through in situ hybridization assays (Howell et al., 2002; Koide et al., 2006; Yin et al., 2010; Ekker et al., 1995, Kazanskaya et al. 2000). From the 4 hedgehog genes identified in Xenopus laevis, Shh, expressed in notochord and/or floor plate, and banded hedgehog (bhh), expressed in somites (Ekker et al., 1995), are likely the potential ligands of Patched that activate Smo signaling during regeneration at these larval stages. Moreover, immunostaining of transverse sections of non-amputated and 4 hpa-amputated larvae shows immunolabeling for Hh (5E1 antibody that detects both Xenopus shh and bhh, Developmental Studies Hybridoma Bank) in somites of larvae at these stages and in spinal cord of regions proximal to the amputation plane (see Author response image 1).

Author response image 1.

Author response image 1.

Open in a new tab

We revised the manuscript to include references that report expression of Hh ligands at these stages of development (page 5, last paragraph).

3. The chemical inhibitors used in the study are not fully specific and therefore additional approaches should be used to validate the findings. One possibility is to use Photo-MO for Smoothened and CREB. Additional methods to support the involvement of PKA are also needed.

Cyclopamine and SAG have been extensively validated and utilized as both highly specific and efficient antagonist and agonist respectively of Smoothened, Hh signaling major effector. The main focus of this study is to demonstrate the necessity for non-canonical Hh, Smoothened-dependent, Gli-independent signaling for muscle and spinal cord regeneration. Hence, we complemented the pharmacological loss and gain of Smoothened function approaches with both pharmacological and genetic loss of Gli function by treating amputated larvae with GANT61, Gli transcriptional activity inhibitor, and by microinjecting Gli2-photo-morpholino. In addition, we also report the decrease in Gli activity upon amputation. We employed multiple approaches to demonstrate the main scope of this study.

Nevertheless, we performed additional experiments that further support the conclusion of Smoothened-dependent regeneration of muscle and spinal cord by incubating amputated larvae with another highly specific Smoothened inhibitor, vismodegib, that binds to a different domain than cyclopamine, and obtained similar effects on regeneration of muscle and spinal cord as with cyclopamine. These new data are included as a new supplement to Figure 1 in the revised manuscript (Figure 1 – supplement 1) and described in the Results section (page 6, second paragraph, lines 101-104).

We chose pharmacological means to manipulate signaling because this study requires temporal control so that the inhibition of the signaling is restricted to the regeneration period. In addition, Gene Tools, the company that develops morpholinos, has discontinued the production of photo-morpholinos. We have attempted to utilize Smo vivo-morpholino, that, unlike the regular morpholinos, permeate into cells, thus, do not need to be injected in the fertilized embryo. However, high variability in regeneration metrics observed in pilot experiments and difficulty in successfully staining for Smo in whole mounts led to inconsistent results and uncertainty of the efficiency of the intended manipulation.

We agree with the reviewers that further investigation is needed to identify the non-canonical Smo-dependent signaling involved in the regeneration of muscle and spinal cord. However, we consider this belongs to a follow up study and believe the findings presented in this manuscript contribute with significant and relevant understanding on Hh-dependent tissue regeneration.

4. A direct link between pCREB and non-canonical HH signalling is needed. This could be provided by testing pCREB reduced after Smoothened inhibition during regeneration.

We have performed additional experiments to directly address the link between the non-canonical HH signaling and pCREB during regeneration by incubating amputated larvae in the presence or absence of SAG or cyclopamine followed by immunostaining of samples for pCREB. Results show that cyclopamine or SAG incubation for the first 4 hpa respectively decreases or increases the number of pCREB+ nuclei in the spinal cord proximal to the tip of the amputated tail. These results suggest that HH signaling modulates CREB activity in the regenerating spinal cord. Moreover, we find that 24-h incubation with cyclopamine or SAG decreases or increases respectively the number of pCREB immunopositive nuclei in the regenerating spinal cord. We revised the manuscript to include these new sets of data (Results, page 12, last paragraph; Figure 7, in current revised manuscript).

Reviewer #1:This manuscript shows that spinal cord and muscle regeneration in the Xenopus larvae is mediated by non-canonical Hh signaling through the activity of PKA. This conclusion is based on a series of pharmacological treatments and a careful analysis of a number of quantified parameters.

The study adds to the idea that Hh signaling is important for tissue regeneration, with the additional interesting idea that the process is mediated, at least at early stages, by non-canonical signaling. The enthusiasm for these results is however undermined by a series of concerns:

According to the literature, st 49 to 54 are considered as the regenerative stages in Xenopus laevis (Edwards-Faret et al. 2017 Nat Protocols). Why have the authors used larvae at st 39-40? Can these earlier stages entail a contribution from developmental related aspects? How long does it take to reach full regeneration in these larvae? This information is missing and needs to be provided for non-specialists.

A review of the literature on studies using Xenopus larvae as a model for tail regeneration results in a total of 63 original research articles, with 37 in which larvae of stages 39-42 are used and 26 in which stages older than 48 are used. Hence, the earlier stages that we used in this study are well accepted in the field as a good model for tail regeneration in this system. The reasoning behind this is that unlike the tip of the tail that is still developing and growing, the tissues of the stump that become exposed after amputation correspond to mature larval tissues, thus the replenishing of these tissues represents a regenerative process and not the mere developmental progression of tissue growth.

Nevertheless, this is an important point that we are thankful the reviewers brought to our attention so that we can better introduce the readers to this model system. We revised the Results section accordingly (Page 5, new paragraph 1).

Studies have shown that full regeneration is achieved after 7-10 days post tail amputation in these larval stages. Nonetheless, the current study is focused on the first stages of regeneration and the signaling mechanisms responding to the acute injury. Hence, the latest end point for the assays presented is 3 days post amputation, when regeneration of muscle and spinal cord are appreciable and measurements of extent of regeneration are reproducible, to allow comparison across controls and experimental groups.

This is important also because Figure 2 shows that Gli transcriptional activity begins to recover at 28 hr. When does it get back to normal levels? Is it before than 72 hrs (time used for most of the analysis)? Can Gli activity account for the differences observed between KT5720 and KT5720+ cyclop treatments?

Because cyclopamine effect in 24-h only group is already significant, this suggests that Smo activity during these first 24 h is important. In contrast, GANT61 and Gli2-MO incubated for the full 72 h post amputation are almost inactive upon the process of spinal cord and muscle regeneration. Hence, we conclude that the canonical, Gli-dependent Hh signaling does not contribute significantly to this process during the first 3 days of regeneration.

We do not think that Gli activity explains the differences in regeneration between inhibiting PKA alone or along with Smo inhibition because inhibiting Gli directly either pharmacologically with GANT61 or genetically by activating Gli2-MO does not result in any appreciable effect on regeneration of muscle or spinal cord. Instead, we think that other non-canonical Smo-dependent signaling, independent of PKA is triggered; and conversely some PKA-dependent regeneration may operate independently from Smo pathway.

On the same line: does treatment with GANT61 beginning 48hr (or so) after amputation has any effect on regeneration parameters?

Treatments for 72 h were done so that the drugs were replenished in the saline that larvae were incubated in every day. Hence, the 72-h treatment implemented for GANT61 consisted on replenishing the solution with GANT61 each day of the 3-day period. Any effect of inhibiting Gli1/2 transcriptional activity during the last day of regeneration (48-72 h) should have been observed. We interpret the lack of significant effect on almost every parameter of regeneration as a lack of participation/necessity of Hh Gli-dependent canonical pathway in the early stages (first 3 days) of the regenerative process.

We do not expect a different result from starting incubation with GANT61 at 48 hpa, compared to 0 hpa.

The experiments reported in Figure 6 do not really prove that P-CREB mediates regeneration. Can the authors attempt to down-modulate its activity?

We believe that inhibiting PKA inhibits CREB activation, as we and many others have demonstrated in previous studies (Belgacem and Borodinsky 2015). Thus, the effect of PKA inhibitor on spinal cord and muscle regeneration may be revealing a role for P-CREB on the regeneration of these tissues.

We agree with the reviewers that assessing more specifically the role of P-CREB will be an interesting future direction for this investigation. However, we believe it belongs to a follow-up study beyond the scope of the present manuscript.

Reviewer #2:

In this manuscript the authors examine how Hh signalling is involved in spinal cord and muscle regeneration in the Xenopus laevis. The important finding is that a non-canonical branch of the Hedgehog pathway involving Protein Kinase A and CREB is acting early during Xenopus larval tail regeneration. The paper is well-written and the findings are very interesting, but further work is needed to warrant publication in eLife. We feel that these experiments fall within the authors expertise and could be accomplished within a reasonable timeframe.

1. Using the same regeneration model, Taniguchi et al. (2014) showed that all HH ligands become expressed in the notochord about 24 hours after injury. The analysis presented here is focused before this time. Is the hypothesis that non-canonical HH signalling is ligand independent? If so the Authors should confirm that HH ligands are not expressed during the period of their analysis (i.e. between 0 and 24 hours after injury).

Taniguchi et al. (2014) used stage 49 tadpoles (12 day-old) instead of stage 40 larvae (2.75 day-old) as in this study. Hence it is difficult to compare. Nevertheless, Taniguchi et al. reported an upregulation in shh expression upon amputation in the regenerating tail, but they do not argue that Hh is not expressed prior to amputation or during the first 24 hours, nor do their findings support this position.

In addition, expression of Shh has been documented during these larval stages mostly through in situ hybridization assays (Howell et al., 2002; Koide et al., 2006; Yin et al., 2010; Ekker et al., 1995, Kazanskaya et al. 2000). From the 4 hedgehog genes identified in Xenopus laevis, Shh, expressed in notochord and/or floor plate, and banded hedgehog (bhh), expressed in somites (Ekker et al., 1995), are likely the potential ligands of Patched that activate Smo signaling during regeneration at these larval stages. Moreover, immunostaining of transverse sections of non-amputated and 4 hpa-amputated larvae shows immunolabeling for Hh (5E1 antibody that detects both Xenopus shh and bhh, Developmental Studies Hybridoma Bank) in somites of larvae at these stages and in spinal cord of regions proximal to the amputation plane (see Author response image 1).

We revised the manuscript to include references that report expression of Hh ligands at these stages of development (page 5, last paragraph).

2. The paper relies very heavily on chemical inhibitors which may not be entirely specific (see also point 3). The authors should consider using Photo-MO for Smothened to confirm some key results. This would be especially convincing when compared to the Gli MO (which did not have a strong effect).

Every approach has its pros and cons. The pharmacological approach offers a higher temporal resolution of the down or upregulation of a signaling pathway, which is particularly important in this case when we want to impose the loss or gain of function of a certain pathway strictly during injury and regeneration.

We agree with the reviewer that showing the same outcome with multiple approaches strengthens the relevance and significance of the finding. Nevertheless, Cyclopamine and SAG have been extensively validated and utilized as both highly specific and efficient antagonist and agonist respectively of Smoothened, Hh signaling major effector. The main focus of this study is to demonstrate the necessity for non-canonical Hh, Smoothened-dependent, Gli-independent signaling for muscle and spinal cord regeneration. Hence, we complemented the pharmacological loss and gain of Smoothened function approaches with both pharmacological and genetic loss of Gli function by treating amputated larvae with GANT61, Gli transcriptional activity inhibitor, and by microinjecting Gli2-photo-morpholino. In addition, we also report the decrease in Gli activity upon amputation. We employed multiple approaches to demonstrate the main scope of this study.

We also performed additional experiments that further support the conclusion of Smoothened-dependent regeneration of muscle and spinal cord by incubating amputated larvae with another highly specific Smoothened inhibitor, vismodegib, that binds to a different domain than cyclopamine, and obtained similar effects on regeneration of muscle and spinal cord as with cyclopamine. These new data are included as a new supplement to Figure 1 in the revised manuscript (Figure 1 – supplement 1) and described in the Results section (page 6, second paragraph, lines 101-104).

In addition, Gene Tools, the company that develops morpholinos, has discontinued the production of photo-morpholinos. We have attempted to utilize Smo vivo-morpholino, that, unlike the regular morpholinos, permeate into cells, and thus do not need to be injected. However, high variability in regeneration metrics observed in pilot experiments and difficulty in successfully staining for Smo in whole mounts led to inconsistent results and uncertainty of the efficiency of the intended manipulation.

3. KT5720 is used as an inhibitor of PKA. There are reports of potential off target effects including inhibiting MAPK (Pharmacological PKA Inhibition: All May Not Be What It Seems, Andrew J. Murray, 2008). MAPK can act upstream of CREB too and is known to be activated immediately after wounding (Christen and Slack, 1999). Could another method be used to strengthen the PKA hypothesis?

We agree with the reviewer and appreciate the reviewer’s suggestion of strengthening the results from pharmacologically inhibiting PKA. However, the main focus of this manuscript is to demonstrate that a non-canonical, Hh signaling is necessary for spinal and muscle regeneration, while the canonical Gli-dependent signaling is dispensable and suppressed during the early stages of tissue regeneration.

We believe we convincingly demonstrate that and provide further evidence of potential partners in the Hh non-canonical signaling without being exhaustive or definitive about the identity of these signaling mechanisms.

Although we used in previous studies overexpression of dominant negative and constitutively active forms of PKA, by microinjecting these constructs in 2-cell stage embryos (Belgacem and Borodinsky, 2011), for the current study focused on later larval stages and restricted to the regenerative process after tail amputation, this approach is not adequate because it interferes with embryonic development.

We present the inhibition of PKA as a potential candidate pathway, but we agree with the reviewer that other pathways may be affected by the PKA inhibitor and participating in the regeneration of muscle and spinal cord. Further investigation is needed and will be pursued in a follow up study. We have revised the manuscript to indicate that the drug may affect other kinases that may participate in this process, that are also known to activate CREB (page 11, lines 217-220).

Nonetheless, we now provide further evidence in the revised manuscript of Smo-dependent activation of P-CREB in the spinal cord proximal to the site of injury (Figure 7), strengthening the link between Smo and the non-canonical signaling, similar to what we have identified in the embryonic spinal cord (Belgacem and Borodinsky, 2015).

4. The Authors use pCREB as a read-out for non-canonical HH signalling in Figure 6. They need to show that pCREB is reduced by Smoothened inhibition during regeneration. What is the significance of its presence in skin after wounding, is that also dependent upon non-canonical HH signalling?

We have performed additional experiments to directly address the link between the non-canonical HH signaling and pCREB during regeneration by incubating amputated larvae in the presence or absence of SAG or cyclopamine followed by immunostaining of samples for pCREB. Results show that cyclopamine or SAG incubation for the first 4 hpa respectively decreases or increases the number of pCREB+ nuclei in the spinal cord proximal to the tip of the amputated tail. These results suggest that HH signaling modulates CREB activity in the regenerating spinal cord. Moreover, we find that 24-h incubation with cyclopamine or SAG decreases or increases respectively the number of pCREB immunopositive nuclei in the regenerating spinal cord. We revised the manuscript to include these new sets of data (Results, page 12, last paragraph; Figure 7, in current revised manuscript).

Reviewer #3:

This manuscript demonstrated the role of non-canonical Hh signaling in spinal cord and muscle regeneration in Xenopus tadpoles. The downstream effectors of this non-canonical Hh signaling involves PKA and CREB activation. The finding provides an explanation of the mechanisms of Hh signaling mediated tissue regeneration. Thus, it is of some interest for regenerative biology.

I have two concerns about this study.

First, for tadpole tail amputations, the authors used NF stages 39-40 embryos. While there is no consensus in the community on the Xenopus tadpole stages used for regeneration analysis, the tail buds of stage 39-40 embryos are not typically considered as differentiated tail, and the cells in the tail buds are precursor cells, rather than differentiated tissue cells. This is particularly true for the cells in the spinal cord. As has been well established, there is ontogenetic decline in the regeneration capacity of Xenopus tadpoles. Thus the stages of tadpoles used for tail amputation is an important concern when performing regeneration analysis.

A review of the literature on studies using Xenopus larvae as a model for tail regeneration results in a total of 63 original research articles, with 37 in which larvae of stages 39-42 are used and 26 in which stages older than 48 are used. Hence, the earlier stages that we used in this study are well accepted in the field as a good model for tail regeneration in this system. The reasoning behind this is that unlike the tip of the tail that is still developing and growing, the tissues of the stump that become exposed after amputation correspond to mature larval tissues, thus the replenishing of these tissues represents a regenerative process and not the mere developmental progression of tissue growth.

Nevertheless, this is an important point that we are thankful the reviewers brought to our attention so that we can better introduce the readers to this model system. We revised the Results section accordingly (Page 5, new paragraph 1).

Second, for Gli activity reporting, the authors injected EGFP:iRFP670 plasmid in the early embryos. This is not ideal, as transgenic tadpoles can be easily generated now.

We appreciate the reviewer’s suggestion. However, we consider that the mosaic expression of the reporter actually plays to our advantage in allowing us to better assess in sparsely labeled cells Gli activity through live imaging. If reporter expression is extended to all cells as in transgenic larvae, the ability to distinguish between individual cells would have been difficult, if not impossible. Additionally, transgenesis is not an established methodology in the lab and currently we do not have the capability of breeding transgenic animals. Moreover, although transient F0 transgenic larvae would be doable in the lab, the efficiency of transgenesis for big constructs is low and the reporter we developed for this study is 6,378 bp, on the large side of constructs.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

The manuscript is much improved but there are a couple of key issues that still need to be carefully addressed and included in the manuscript to assure its publication in eLife.

Essential revisions:

1. The first relates to the question of whether Hh ligand expression is activated upon amputation at stage 40. Staining of Hh ligands needs to be analyzed and included in the manuscript. This is an important support for the model presented in Figure 9.

To assess Hh ligand protein expression upon amputation, we performed Western blot assays from 500 mm-long regions of the tail stump at 0 and 24 h after amputation. We probed membranes with whole-cell lysates from these samples with the Hh 5E1 antibody (DSHB). Results show that Hh ligands are detected before amputation and during early regeneration. Moreover, we find an increase in Hh protein levels in regenerating tail after 24 h post amputation compared to 0 h. We included these additional results in a new Figure 1- supplement 1, along with an image of 4 hours post amputation larvae stained with the 5E1 antibody to illustrate the spatial distribution of Hh ligands during the first hours post amputation.

We attempted to do a quantitative analysis of 5E1 immunostaining in tissue sections of non-amputated compared to amputated larvae, but we were not satisfied with the reliability of this approach. Due to the pattern of the staining, which mostly localizes to cell membrane, we were not able to count cells or threshold the signal in a consistent manner across samples. Nevertheless, data on the expression of Hh ligand transcripts in the same model system at the same developmental and post amputation times that we used for our study have been recently published using single-cell RNA seq (Aztekin et al., Science 2019). We browsed the data shared (https://marionilab.cruk.cam.ac.uk/XenopusRegeneration/) by the authors and found that Hh ligands are indeed expressed in stage 40 larvae. Transcripts for shh are detected in posterior notochord and floor plate in pre-amputated and in 1 through 3 days post amputation of the regenerating tail, with an increase in levels of shh transcript as early as 1 day post amputation, which is the earliest time point presented by the authors. Transcripts for shh are also expressed in neural tissue, including spinal cord progenitors and subtypes of spinal cord neurons, while dhh is detected in neural tissue and floor plate.

It should be noted that the pattern of immunohistochemistry that we included in our prior responses to reviewers and in the new Figure 1- supplement 1 may detect not only cells/tissues that express the Hh ligands but also target cells that receive it. Indeed, Aztekin and colleagues reported expression of ptch1 and ptch2 genes in the myotome, neural tissue and floor plate.

2. Graphs should be better presented. Error bars for controls are not treated consistently.

We have revised graphs in Figures 1, 3, 4, 5 and 7D to keep a consistent presentation of error bars for control groups.

3. The method section related to how regeneration is measured in Figure 1—figure supplement should be expanded and include an explanation on what is the difference between the data presentation of Figure 1 and figure supplement 1.

We have revised the presentation of the data so that is consistent across figures and figure supplements and revised the text in the Methods section (page 27, first paragraph) to explain how control groups and error bars were treated for each set of data.

Reviewer #2:

The responses to the reviewers' comments are on the whole satisfactory, except around the question of whether there is evidence that Hedgehog ligand expression is activated by excision at stage 40.

The authors state that "Taniguchi et al. reported an upregulation in shh expression upon amputation in the regenerating tail, but they do not argue that Hh is not expressed prior to amputation or during the first 24 hours, nor do their findings support this position." This is not accurate. Figure 1A of Taniguchi et al. analyses expression of shh and bhh over time. At 0hrs post amputation expression of shh is barely detectable, and bhh is not detected. At 24hrs post amputation there is no expression of either ligand. Starting at 48 hours post amputation, shh is induced strongly and bhh to a lesser degree. This expression is stated as being exclusively in the notochord. This has led to the model that Hedgehog signalling becomes slowly activated peaking four to five days after amputation. The authors should familiarise themselves with the literature on the role of Hedgehog signalling during tail regeneration. The comparison with other published analysis of tail regeneration is important even if it is at a different stage.

We have revised the text to include more recent studies that have addressed the expression of Hh ligands during regeneration of the tail at the developmental stages used in this study (Results section, page 5 last sentence and page 6 first paragraph). We browsed the single-cell RNA seq data shared (https://marionilab.cruk.cam.ac.uk/XenopusRegeneration/) by Aztekin et al. (Science 2019), given that they used X. laevis larvae at the same developmental and post amputation times that we used for our study, and found that Hh ligands are indeed expressed in stage 40 larvae. Transcripts for shh are detected in posterior notochord and floor plate in pre-amputated and in 1 through 3 days post amputation of the regenerating tail, with an increase in levels of shh transcript as early as 1 day post amputation, which is the earliest time point presented by the authors. Transcripts for shh are also expressed in the neural tissue, including spinal cord progenitors and subtypes of spinal cord neurons, while dhh is detected in neural tissue and floor plate.

Comparing patterns of expression in the regenerating tail between stage 40 and stage 49 larvae is problematic, and it is not that surprising that there are differences between the findings of Taniguchi et al. in stage 49 larvae and Aztekin et al. in stage 40 animals. It is also important to consider the methodological approaches used to assess expression, which exhibit different sensitivities and spatiotemporal resolution of expression levels. In addition, these studies report on Hh ligand transcript expression; whether transcripts correlate directly with protein levels is doubtful. For instance, Author response image 2 shows a representation of shh mRNA and protein from Xenbase, where the developmental changes in transcript and protein levels strikingly diverge. This suggests that reporting on transcript levels of a signaling molecule may not predict the temporal course of activation of a certain signaling pathway.

Author response image 2. X. laevis embryonic protein expression.

Author response image 2.

Open in a new tab

This graph shows protein expression in Xenopus laevis. Optionally the mRNA expression is also charted for comparison. The number in brackets after the gene symbol indicates the number of peptides. (Peshkin et al., 2019)

The authors show a staining of Hedgehog ligands in their rebuttal but do not present analysis of this data and have not added the data to their manuscript. It appears to show that Hedgehog ligands are strongly upregulated within 4 hours after amputation, and that this activation is primarily in the neural tube (not notochord?). This finding would be consistent with the author's model that Hedgehog signalling is activated by wounding and stimulates muscle and neural tube regeneration (Figure 9). Importantly this data would indicate that Hedgehog expression becomes active much earlier than has been previously described for stage 49. This data should be properly analysed and presented in the manuscript. It would provide important support for their model. This is not a labour intensive request.

To assess Hh ligand protein expression upon amputation, we performed Western blot assays from 500 mm-long regions of the tail stump at 0 and 24 h after amputation. We probed membranes with whole-cell lysates from these samples with the Hh 5E1 antibody (DSHB). Results show that Hh ligands are detected before amputation and during early regeneration. Moreover, we find an increase in Hh protein levels in regenerating tail after 24 h post amputation compared to 0 h.

We included these additional results in new Figure 1- supplement 1, along with an image of 4 hours post amputation larvae stained with the 5E1 antibody to illustrate the spatial distribution of Hh ligands during the first hours post amputation.

We attempted to do a quantitative analysis of 5E1 immunostaining in tissue sections of non-amputated compared to amputated larvae, but we were not satisfied with the reliability of this approach. Due to the pattern of the staining, which mostly localizes to cell membrane, we were not able to count cells or threshold the signal in a consistent manner across samples. Nevertheless, data on the expression of Hh ligand transcripts in the same model system at the same developmental and post amputation times that we used for our study have been recently published using single-cell RNA seq (Aztekin et al., Science 2019). We browsed the data shared (https://marionilab.cruk.cam.ac.uk/XenopusRegeneration/) by the authors and found that Hh ligands are indeed expressed in stage 40 larvae. Transcripts for shh are detected in posterior notochord and floor plate in pre-amputated and in 1 through 3 days post amputation of the regenerating tail, with an increase in level of shh transcript as early as 1 day post amputation, which is the earliest time point presented by the authors. Transcripts for shh are also expressed in neural tissue, including spinal cord progenitors and subtypes of spinal cord neurons, while dhh is detected in neural tissue and floor plate.

It should be noted that the pattern of immunohistochemistry that we included in our prior responses to reviewers and in the new Figure 1-supplement 1 may detect not only cells/tissues that express the Hh ligands but also target cells that receive it. Indeed, Aztekin and colleagues reported expression of ptch1 and ptch2 genes in the myotome, neural tissue and floor plate.

Reviewer #3:

The revised manuscript added new data from additional experiment, explained the reasoning of using stage 39-40 tadpoles for tail amputation experiment, and replaced the western blot images showing the efficiency of Gli2-mo. The revision adequately addressed my concerns.

However, I find it confusing in understanding the graph presentations, especially with the new data. Error bars for controls are not treated consistently. For example, in Figure 1D-I, there are no error bars shown for the control groups, but in Figure 1—figure supplement 1 B-G, there are error bars shown for the control. Yet, the means for the controls in Figure 1—figure supplement 1 B-G are all 100. Could you please explain a bit more in the method section, how the regeneration is measured in Figure 1—figure supplement? What is the difference between the data presentation of Figure 1 and figure supplement 1?

Error bars in control groups were used in Figure 1 supplement 1 (current Figure 1-supplement 2 and several others) because the control conditions were perfectly cohort-matched with treatment conditions, such that the error bars for controls would apply exactly to the treatments depicted; i.e. all of the data that went into said bar graphs came from the same experimental cohorts. In contrast, the control bars in Figure 1 and others did not have error bars because some independent experimental sets were combined for presentation: i.e. the cohort-matched controls for each treatment bar were not the same samples. For example, some cyclopamine treatments did not have matched SAG treatments in their cohort, and hence the controls for the SAG and cyclopamine bars were not the same samples. All metrics presented were pooled for analysis as a percent of the average, cohort-paired control values for each parameter in question.

In the interest of consistency in the presentation of the data and to report on the variability inherent to control groups in all figures, we have added error bars to the controls corresponding to the largest standard error values in the control groups of the different independent experiments that were included in the given bar graph.

Hence, we revised graphs in Figures 1, 3, 4—figure supplement 1, 5 and 7 to keep a consistent presentation of error bars for control groups.

We revised the text in the Methods section (page 27, first paragraph) to explain how control groups and error bars were treated for each set of data presented.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    Figure 1—source data 1. Hedgehog signaling regulates spinal cord and muscle regeneration.
    elife-61804-fig1-data1.xlsx (56.5KB, xlsx)
    Figure 1—figure supplement 1—source data 1. Treatment with Smoothened (Smo) antagonist vismodegib reduces muscle and spinal cord regeneration.
    elife-61804-fig1-figsupp1-data1.xlsx (19.8KB, xlsx)
    Figure 2—source data 1. Canonical, Gli-dependent Hedgehog signaling is downregulated following tail amputation.
    elife-61804-fig2-data1.xlsx (1.3MB, xlsx)
    Figure 2—figure supplement 1—source data 1. Positive control for Gli transcriptional activity reporter.
    elife-61804-fig2-figsupp1-data1.xlsx (119.7KB, xlsx)
    Figure 3—source data 1. Gli1/2 transcriptional activity is not necessary for spinal cord and muscle regeneration.
    elife-61804-fig3-data1.xlsx (28.9KB, xlsx)
    Figure 3—figure supplement 1—source data 1. Treatment with GANT61 inhibits Gli1/2 transcriptional activity.
    elife-61804-fig3-figsupp1-data1.xlsx (163.3KB, xlsx)
    Figure 4—source data 1. Genetic downregulation of canonicalHedgehog-Smoothened(Hh-Smo)signaling does not affect spinal cord and muscle regeneration.
    elife-61804-fig4-data1.xlsx (24.5KB, xlsx)
    Figure 4—figure supplement 1—source data 1. UV treatment does not reduce regeneration and uncages Gli2-MO, knocking down Gli2 expression.
    elife-61804-fig4-figsupp1-data1.xlsx (19.5KB, xlsx)
    Figure 5—source data 1. Tissue-specific interaction between PKA and Smoothened(Smo)signaling in regulation of spinal cord and muscle regeneration.
    elife-61804-fig5-data1.xlsx (31.7KB, xlsx)
    Figure 6—source data 1. Spatiotemporal activation of CREB in the injured and regenerated spinal cord.
    elife-61804-fig6-data1.xlsx (22.5KB, xlsx)
    Figure 7—source data 1. Post-amputation activation of CREB is regulated by Hedgehog(Hh)signaling upon injury.
    elife-61804-fig7-data1.xlsx (15.8KB, xlsx)
    Figure 7—source data 2. Post-amputation activation of CREB is regulated by Hedgehog(Hh)signaling in the regenerating spinal cord.
    elife-61804-fig7-data2.xlsx (13.5KB, xlsx)
    Figure 8—source data 1. Coordination of spinal cord and muscle outgrowth in the regenerating tail is dependent on Smoothened(Smo)signaling.
    elife-61804-fig8-data1.xlsx (148.1KB, xlsx)
    Transparent reporting form
    elife-61804-transrepform.docx (111KB, docx)

    Data Availability Statement

    All data generated or analysed during this study are included in the manuscript and supporting files.

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES