CYSL-1 interacts with the O2-sensing hydroxylase EGL-9 to promote H2S-modulated hypoxia-induced behavioral plasticity in C. elegans

Abstract

The C. elegans HIF-1 proline hydroxylase EGL-9 functions as an O(2) sensor in an evolutionarily conserved pathway for adaptation to hypoxia. H(2)S accumulates during hypoxia and promotes HIF-1 activity, but how H(2)S signals are perceived and transmitted to modulate HIF-1 and animal behavior is unknown. We report that the experience of hypoxia modifies a C. elegans locomotive behavioral response to O(2) through the EGL-9 pathway. From genetic screens to identify novel regulators of EGL-9-mediated behavioral plasticity, we isolated mutations of the gene cysl-1, which encodes a C. elegans homolog of sulfhydrylases/cysteine synthases. Hypoxia-dependent behavioral modulation and H(2)S-induced HIF-1 activation require the direct physical interaction of CYSL-1 with the EGL-9 C terminus. Sequestration of EGL-9 by CYSL-1 and inhibition of EGL-9-mediated hydroxylation by hypoxia together promote neuronal HIF-1 activation to modulate behavior. These findings demonstrate that CYSL-1 acts to transduce signals from H(2)S to EGL-9 to regulate O(2)-dependent behavioral plasticity in C. elegans.

Figure 1. C. elegans displays O₂-associated locomotive behavioral plasticity

(A) Acute locomotive speed changes during O2-OFF and O2-ON responses. Average speed values with 2 standard errors of the means (indicated by light blue) of a population of animals (n>50) are shown with step changes of O₂ between 20% and 0% at the indicated times. The mean speed differences of all animals within 60 seconds before or after O2 restoration are highly significant (p<0.001, one-sided unpaired t-test; also see Supplemental Information). (B) Worm tracks showing locomotive patterns during basal state, O2-OFF and O2-ON responses. One-minute recordings were made under basal conditions (20% O2) or immediately following the initiation of O2-OFF or ON responses. (C) Schematic of behavioral paradigms used to test the effects of prior experience of hypoxia on the O2-ON response. (D) Behavior of hypoxia-experienced wild-type animals with suppressed O2-ON response, compared to that of naive animals as shown in (A). Graphs exclusively labeled “With prior experience of 24 hrs. hypoxia” show data for hypoxia-experienced but not naive animals. (E) Behavior of egl-9 mutants showing a lack of O2-ON response. (F) Behavior of egl-9; hif-1 mutants with a restored O2-ON response compared to that of egl-9 mutants. (G) Behavior of hypoxia-experienced hif-1 mutants with a suppressed O2-ON response, compared to that of hypoxia-experienced wild-type animals. (H) Behavior of hif-1 mutants with a normal O2-ON response.

Figure 2. RHY-1 modulates HIF-1 and the O2-ON behavioral response through EGL-9

(A) Fluorescence micrographs showing constitutive GFP signals in egl-9(−) mutants with the transgenic reporter P_K10H10.2::GFP (nIs470), indicating high HIF-1 transcriptional activity. GFP signals are absent in the wild type and in egl-9(sa307); hif-1(ia4) double mutants, except for weak GFP fluorescence in the pharynx. myo-2::mCherry expressed in pharyngeal muscles was used as the co-injection marker. (B) rhy-1(n5500) mutants with strong constitutive GFP expression that is suppressed by hif-1 mutations. (C) rhy-1(n5500) mutants show a defective O2-ON response. (D) rhy-1(n5500); hif-1(ia4) double mutants show a restored O2-ON response. (E) An extrachromasomal array containing rhy-1(+) genomic fragments rescues the behavioral defect in the O2-ON response of rhy-1(n5500) mutants. (F) Rescued K10H10.2::GFP ectopic expression of rhy-1(n5500) mutants by rhy-1(+) arrays. myo-3::mCherry expressed in body wall muscles was the co-injection marker. (G) rhy-1(ok1402) null mutants show a defective O2-ON response. Scale bar: 25 µm.

Figure 3. A modifier screen identified cysl-1 as a regulator of HIF-1 and behavior

(A) A schematic of the cysl-1 gene, indicating the seven alleles isolated from the rhy-1(n5500) suppressor screen. This drawing was generated by the Exon-Intron Graphic Maker (WormWeb.org). (B) K10H10.2::GFP expression in rhy-1(n5500); nIs470 mutants with myo-2::mCherry as the co-injection marker. Scale bar: 25 µm. (C) K10H10.2::GFP expression is absent in rhy-1(n5500); nIs470; cysl-1(n5515) mutants. (D–E) A defective O2-ON response of rhy-1(n5500) mutants is suppressed by the cysl-1(5515) mutation. (F) cysl-1(ok762) null mutants show a normal O2-ON response. (G) egl-9(sa307); cysl-1(ok762) double mutants with a defective O2-ON response. (H) Western blots of reporter GFP expression driven by the K10H10.2 promoter from single or multiple LOF or GOF mutants of rhy-1, cysl-1, egl-9 and hif-1. (I) Real-time PCR quantification of endogenous K10H10.2 mRNA levels (normalized to the control group of nIs470 animals) in various mutants as indicated in (H). *: p<0.01, one-way ANOVA, Bonferroni post-test.

Figure 4. Expression pattern and site-of-function of CYSL-1

(A) Fluorescence and Nomarski images showing the expression pattern of cysl-1 as visualized by the integrated transcriptional GFP reporter nIs500, which harbors a 2.8 kb promoter of cysl-1 fused to GFP. Head neurons are indicated by the arrow, and tail neurons are indicated by the arrowhead. (B) Enlarged view of fluorescence micrograph showing AVM and BDU neurons. (C) Confocal microscopic view of pharyngeal (I1 and M2 indicated by arrows) and head neurons. (D–E) Expression patterns of cysl-1 as visualized by the extrachromasomal array nEx1838 with a translational GFP reporter harboring the promoter and genomic coding regions of cysl-1 fused in-frame to GFP. (F–G) Rescue of cysl-1(n5515) phenotypes by neuronal expression of cysl-1(+) cDNA driven by the ric-19 promoter. (H) The unc-14 promoter-driven neuronal activation of HIF-1 causes defects in the O2-ON response of egl-9(sa307); hif-1(ia4) double mutants. (I) The dpy-7 promoter-driven hypodermal activation of HIF-1 does not cause defects in the O2-ON response of egl-9; hif-1 double mutants. Scale bars: 25 µm.

Figure 5. Evolutionary and enzymatic characteristics of CYSL-1

(A) Alignment of CYSL-1 homologs identifies a highly conserved glycine disrupted by cysl-1(n5515). This phylogenetic tree was generated using ClusterW2 and is displayed in a cladogram. The PLP-binding site is highlighted by the red box (Aitken et al., 2011). C. elegans cysl-1 paralogs are indicated in the blue box. (B) Endogenous CYSL-1 protein levels as determined by SDS-PAGE and western blots and a polyclonal antibody against CYSL-1. C. elegans protein lysates from various genetic backgrounds were analyzed. 50 µg protein samples were loaded per lane. (C, D) Purified recombinant CYSL-1 exhibits intrinsic cysteine synthase activity. Dependence of cysteine generation on varying OAS concentration with 20 mM sulfide and varying sulfide concentration with 20 mM OAS, respectively. Means of three measurements are shown and the Michaelis-Menten equation was used for curve fitting. Error bars represent SD. (E) K_M and k_cat values of CYSL-1 (determined from Figure 5C and 5D) and OAS sulfhydrylases of other species as established from previous studies (Bonner et al., 2005; Mino et al., 2000; Mozzarelli et al., 2011; Ono et al., 1994). K_M: Michaelis value, k_cat: turnover number.

Figure 6. CYSL-1 interaction with EGL-9 mediates H₂S signaling to HIF-1 and behavioral plasticity

(A) The gain-of-function mutation egl-9(n5535) fully suppresses the O2-ON defect of rhy-1(n5500) mutants. (B) Two dominant suppressors of rhy-1(n5500) alter the C-terminus of EGL-9. n5535 converts glutamic acid 720 to lysine, while n5539 and n5552 are splice-donor and acceptor mutations, respectively, that result in a C-terminally truncated EGL-9 protein. (C) Yeast two-hybrid assays with colony growth on selective plates (SD/-LEU/-TRP, or SD/-LEU/-TRP/-HIS3/-ADE) after co-transformation of cysl-1 prey constructs and various egl-9 bait constructs. The growth with vector-only control group, i.e. pGADT7 with egl-9-C-terminal, indicates non-specific reporter activation. All egl-9 constructs lack the N-terminal domain, which conferred moderate non-specific reporter activation. Note the specific association of CYSL-1 with EGL-9 with an intact C-terminus but not with n5535 or n5539 mutant EGL-9. (D) Exposure to low H₂S induced GFP fluorescence from the K10H10.2::GFP reporter in wild-type but not cysl-1 mutant animals. Scale bars: 25 µm. (E) QPCR measurements of the endogenous HIF-1 target K10H10.2 mRNA in the wild type and in hif-1(ia4), cysl-1(ok762) or egl-9(n5535) mutant backgrounds. p<0.001, two-way ANOVA, Bonferroni post-test. (F) QPCR measurements of the endogenous HIF-1 target nhr-57 in the wild type and in hif-1(ia4), cysl-1(ok762) or egl-9(n5535) mutant backgrounds. (G) Protein Co-IP experiments showing that the interaction between endogenous CYSL-1 and the GFP::EGL-9 fusion protein in vivo is markedly enhanced by H₂S exposure. GFP-bound protein complexes isolated from lysates of the strain nEx [P_ric-19::egl-9::gfp] using anti-GFP affinity beads were analyzed by SDS-PAGE and western blots. GFP levels served as internal loading controls. (H) Behavior of cysl-1 null mutants with decreased inhibition of the O2-ON response after 24 hr hypoxia experience as compared to hypoxia-experienced wild-type animals. (I) Behavior of egl-9(n5535) mutants with decreased inhibition of the O2-ON response after 24 hr hypoxia experience.

Figure 7. Models for behavioral plasticity mediated by the RHY-1/CYSL-1/EGL-9/HIF-1 pathway

(A) The genetic pathway in which rhy-1, cysl-1, and egl-9 act in a negative-regulatory cascade to control the activity of HIF-1, which functions in neurons to modulate the O2-ON response and in the hypoderm to regulate K10H10.2 expression. (B) Model depicting the major molecular interactions in neurons of naïve animals without experience of hypoxia, enabling the normal O2-ON response. EGL-9 promotes HIF-1 degradation (indicated by HIF-1 with dotted line) by O2-dependent hydroxylation and also inhibits HIF-1 transcriptional activity. (C) Model depicting the major molecular interactions in neurons of animals with experience of hypoxia leading to their suppressed O2-ON response. Decreased O₂ during hypoxia impairs HIF-1 hydroxylation, which stabilizes HIF-1 but alone is not sufficient to activate HIF-1 targets. Hypoxia also increases H₂S levels to promote CYSL-1 sequestration of EGL-9, thus alleviating inhibition of HIF-1 transcriptional activity. EGL-9 inactivation (indicated with dotted line) by the dual regulatory mechanisms drives HIF-1 activation. See text for details.