The Caenorhabditis elegans rhy-1 gene inhibits HIF-1 hypoxia-inducible factor activity in a negative feedback loop that does not include vhl-1

Abstract

Hypoxia-inducible factor (HIF) transcription factors implement essential changes in gene expression that enable animals to adapt to low oxygen (hypoxia). The stability of the C. elegans HIF-1 protein is controlled by the evolutionarily conserved EGL-9/VHL-1 pathway for oxygen-dependent degradation. Here, we describe vhl-1-independent pathways that attenuate HIF-1 transcriptional activity in C. elegans. First, the expression of HIF-1 target genes is markedly higher in egl-9 mutants than in vhl-1 mutants. We show that HIF-1 protein levels are similar in animals carrying strong loss-of-function mutations in either egl-9 or vhl-1. We conclude that EGL-9 inhibits HIF-1 activity, as well as HIF-1 stability. Second, we identify the rhy-1 gene and show that it acts in a novel negative feedback loop to inhibit expression of HIF-1 target genes. rhy-1 encodes a multi-pass transmembrane protein. Although loss-of-function mutations in rhy-1 cause relatively modest increases in hif-1 mRNA and HIF-1 protein expression, some HIF-1 target genes are expressed at higher levels in rhy-1 mutants than in vhl-1 mutants. Animals lacking both vhl-1 and rhy-1 function have a more severe phenotype than either single mutant. Collectively, these data support models in which RHY-1 and EGL-9 function in VHL-1-independent pathway(s) to repress HIF-1 transcriptional activity.

Figure 1.—

nhr-57∷GFP, a reporter of HIF-1-dependent transcription in hypoxia and in various mutant backgrounds. (A–P) nhr-57∷GFP expression in wild-type and mutant animals, assayed by fluorescent microscopy. For each experimental condition, representative fluorescent images are paired with Nomarski images of the same animal. With the exception of C and D, all animals were incubated in room air. (A and B) Wild type. (C and D) Wild type incubated in 0.5% oxygen. (E and F) vhl-1(ok161). (G and H) hif-1(ia04); vhl-1(ok161). (I and J) egl-9(sa307). (K and L) egl-9(sa307); hif-1(ia04). (M and N) rhy-1(ok1402). (O and P) rhy-1(ok1402); hif-1(ia04).

Figure 2.—

EGL-9 acts via VHL-1-dependent and VHL-1-independent pathways to regulate HIF-1. (A) The levels of endogenous HIF-1 protein were assayed in wild type N2, vhl-1(ok161), and egl-9(sa307) L4 stage animals. The asterisk indicates an unidentified protein that is recognized by the HIF-1 antibody, but is also present in hif-1(ia04) animals, which have a large deletion in the hif-1 gene. (B) Real-time RT–PCR was used to quantitate mRNA levels of HIF-1 target genes in wild-type and mutant worms. The graph depicts average values from three independent experiments, and the error bars represent the mean standard error.

Figure 3.—

rhy-1 gene and mutant alleles. (A) Diagram of the rhy-1/W07A12.7 genomic region. Exons are indicated by boxes. Arrows indicate the positions of the start codon ATG and the position of the ia38 S157F point mutation. The endpoints of the ok1402 and ok1398 deletions are represented by parentheses. (B) The ia38 mutation is in a highly conserved region. RHY-1 is aligned with five other members of the gene family. Asterisks indicate amino acid identities. The arrow indicates the position of the ia38 S157F point mutation. The NCBI reference numbers for the mammalian and bacterial sequences are Unigene Hs.451560, Unigene Mm.183576, and accession no. Q6GIB3. The C. elegans genes rhy-1, W07A12.6, and nrf-6 are described at http://www.wormbase.org.

Figure 4.—

RHY-1∷GFP expression pattern. (A) Structures of two reporters, in which GFP is fused to rhy-1 regulatory sequences. pSC09 includes the entire rhy-1 coding region and can rescue the rhy-1 ia38, ok1402, and ok1398 mutant phenotypes. Boxes represent exons. (B–G) C. elegans carrying the pSC15 (B–E) and pSC09 (F and G) reporter constructs. Anterior is to the right. (B and C). Nomarski image (B) and fluorescent image (C) of a young adult hermaphrodite carrying the pSC15 reporter. GFP is strongly visible in the intestine and in certain head sensory neurons. (D) An L4 stage hermaphrodite. GFP is visible in hypodermal cells, the vulva (indicated by the arrow), and cells of the ventral nerve cord. (E) An L3 stage larva with strong hypodermal GFP fluorescence. (F and G) Nomarski image (F) and fluorescent image (G) of pSC09 expression in the head of an L2 stage animal.

Figure 5.—

RHY-1 has VHL-1-independent functions. (A and B) Expression of two HIF-1 target genes in various mutants. K10H10.2 and F22B5.4 mRNA levels were quantitated by real-time RT–PCR in three independent experiments. (C) Relative levels of nhr-57∷GFP as determined by immunoblots. (D) hif-1 mRNA levels were assessed by real-time RT–PCR. Numbers indicate the relative mRNA levels compared with wild-type worms. The following alleles were used for the above studies: rhy-1(ok1402), egl-9(sa307), hif-1(ia04), and vhl-1(ok161). Error bars represent the mean standard error from three independent experiments. (E) Protein blots probed with HIF-1 antisera. The position of HIF-1 is indicated by the arrow. The relative expression levels of HIF-1 are indicated below each lane. These are average values from three independent experiments. The asterisk indicates an unidentified protein that is recognized by the HIF-1 antibody, but is also present in hif-1(ia04) worms, which carry a large deletion in hif-1.

Figure 6.—

The egl-9 mutant phenotype, as assayed by expression of HIF-1 target genes, is not increased in severity by additional mutations in vhl-1 or rhy-1. (A) nhr-57∷GFP protein levels assayed on immunoblots. Total protein from 20 L4 stage animals was loaded into each lane. (B) Relative expression of the HIF-1 target genes K10H10.2 and F22B5.4, as determined by real-time RT–PCR. The following alleles were used: rhy-1(ok1402), egl-9(sa307), and vhl-1(ok161). In each case, an average of three independent experiments and the mean standard error are shown.