Axotomy-induced HIF-serotonin signalling axis promotes axon regeneration in C. elegans

Abstract

The molecular mechanisms underlying the ability of axons to regenerate after injury remain poorly understood. Here we show that in Caenorhabditis elegans, axotomy induces ectopic expression of serotonin (5-HT) in axotomized non-serotonergic neurons via HIF-1, a hypoxia-inducible transcription factor, and that 5-HT subsequently promotes axon regeneration by autocrine signalling through the SER-7 5-HT receptor. Furthermore, we identify the rhgf-1 and rga-5 genes, encoding homologues of RhoGEF and RhoGAP, respectively, as regulators of axon regeneration. We demonstrate that one pathway initiated by SER-7 acts upstream of the C. elegans RhoA homolog RHO-1 in neuron regeneration, which functions via G12α and RHGF-1. In this pathway, RHO-1 inhibits diacylglycerol kinase, resulting in an increase in diacylglycerol. SER-7 also promotes axon regeneration by activating the cyclic AMP (cAMP) signalling pathway. Thus, HIF-1-mediated activation of 5-HT signalling promotes axon regeneration by activating both the RhoA and cAMP pathways.

Figure 1. Serotonin activates axon regeneration.

(a) The synthesis of serotonin from tryptophan catalyzed by TPH-1. (b) Representative PLM sensory neurons in wild type and tph-1 mutant animals 24 h after laser surgery. Arrowheads indicate cut sites. Scale bar=20 μm. (c) Regeneration of PLM sensory neurons. Lengths of PLM regrowth at 24 h after laser surgery are shown. Error bars indicate s.e.m. ***P<0.001 (n≥30; unpaired t-test, two-tailed and unequal variances). (d,f) Regeneration of D-type motor neurons. Percentages of axons that initiated regeneration 24 h after laser surgery are shown. Error bars indicate 95% CI. ***P<0.001; **P<0.01; *P<0.05; NS, not significant (Fisher's exact test, two-tailed, n≥50). (e) Cell-specific deletion of tph-1 using a Cre-loxP recombination system. CI, confidence interval.

Figure 2. HIF-1 positively regulates axon regeneration.

(a) Induction of Ptph-1::gfp expression in D-type motor neurons by laser surgery. Expression of fluorescent proteins in D-type motor neurons 30 min after laser surgery is shown. A schematic representation of D-type motor neurons is shown on the bottom. Arrowheads and arrows indicate axons and cell bodies of D-type neurons with (white) or without (blue) laser surgery, respectively. D-type neurons are visualized by mCherry under control of the unc-47 promoter. Cell bodies of D-type neurons are magnified (blue boxes). Scale bar=20 μm. (b) Percentages of animals expressing tph-1. Induction of Ptph-1::gfp expression in D-type motor neurons with (+) or without (−) laser surgery was assayed as described in Methods section. Twenty neurons were examined for each condition. Error bars indicate 95% CI. ***P<0.001; *P<0.05 (Fisher's exact test, two-tailed, n=20). (c) Regeneration of D-type motor neurons. Percentages of axons that initiated regeneration 24 h after laser surgery are shown. Error bars indicate 95% CI. ***P<0.001; *P<0.05 (Fisher's exact test, two-tailed, n≥50). (d) Regeneration of PLM sensory motor neurons. Lengths of PLM regrowth at 24 h after laser surgery are shown. Error bars indicate s.e.m. ***P<0.001 (n≥30; unpaired t-test, two-tailed, unequal variances). CI, confidence interval.

Figure 3. 5-HT-SER-7 signalling activates axon regeneration via GPA-12 and RHGF-1.

(a,d) Percentages of axons that initiated regeneration 24 h after laser surgery. Error bars indicate 95% CI. ***P<0.001; **P<0.01; *P<0.05; NS, not significant (Fisher's exact test, two-tailed, n≥50). (b) SER-7 signalling pathway. SER-7 activates the GPA-12-RHGF-1-RHO-1 pathway. Activation of RHO-1 inhibits DGK-1, leading to a stabilization of DAG levels. (c) Structure of RHGF-1. Schematic diagrams of RHGF-1 and its human counterpart are shown. Domains are shown as follows: a PDZ domain, a regulator of G protein signalling domain (RGS), a Dbl homology domain (DH) and a Pleckstrin homology domain (PH). The bold lines underneath indicate the extent of the deleted regions in the rhgf-1(ok880) and rhgf-1(gk217) mutants.

Figure 4. Effect of RhoGAP on axon regeneration.

(a) Structure of RGA-5. Schematic diagrams of RGA-5 and its human counterpart are shown. Domains are shown as follows: a GTP binding domain (GBD) and a Rho GAP domain (GAP). The bold line underneath indicates the extent of the deleted region in the rga-5(ok2241) mutant. (b) A two-hybrid assay for the interaction of RGA-5 with RHO-1. The reporter strain PJ69-4A was co-transformed with expression vectors encoding GAL4 DBD-RHO-1 (T19N) (GDP), GAL4 DBD-RHO-1 (G14V) (GTP), GAL4 DBD-CED-10 (G12V) (GTP), GAL4 DBD-CDC-42 (G12V) (GTP) and GAL4 AD-RGA-5 (GAP domain) as indicated. (c) Representative D-type motor neurons in wild-type animals and wild-type animals overexpressing RGA-5 GAP domain 24 h after laser surgery. In wild-type animals, severed axons have regenerated growth cones (orange arrowhead). In wild-type animals overexpressing the RGA-5 GAP domain, the proximal end of axon failed to regenerate (white arrowhead). Scale bar=20 μm. (d) Percentages of axons that initiated regeneration 24 h after laser surgery. Error bars indicate 95% CI. ***P<0.001; **P<0.01; *P<0.05; NS, not significant (Fisher's exact test, two-tailed, n≥50). CI, confidence interval.

Figure 5. SER-7 promotes axon regeneration via the cAMP pathway.

(a,c) Percentages of axons that initiated regeneration 24 h after laser surgery. Error bars indicate 95% CI. ***P<0.001; **P<0.01; *P<0.05; NS, not significant (Fisher's exact test, two-tailed, n≥50). (b) SER-7 signalling pathways. SER-7 activates both GPA-12-RHGF-1-RHO-1 and GSA-1-ACY-1-PKA pathways.