Wide diversity in structure and expression profiles among members of the Caenorhabditis elegans globin protein family

Abstract

Background: The emergence of high throughput genome sequencing facilities and powerful high performance bioinformatic tools has highlighted hitherto unexpected wide occurrence of globins in the three kingdoms of life. In silico analysis of the genome of C. elegans identified 33 putative globin genes. It remains a mystery why this tiny animal might need so many globins. As an inroad to understanding this complexity we initiated a structural and functional analysis of the globin family in C. elegans.

Results: All 33 C. elegans putative globin genes are transcribed. The translated sequences have the essential signatures of single domain bona fide globins, or they contain a distinct globin domain that is part of a larger protein. All globin domains can be aligned so as to fit the globin fold, but internal interhelical and N- and C-terminal extensions and a variety of amino acid substitutions generate much structural diversity among the globins of C. elegans. Likewise, the encoding genes lack a conserved pattern of intron insertion positioning. We analyze the expression profiles of the globins during the progression of the life cycle, and we find that distinct subsets of globins are induced, or repressed, in wild-type dauers and in daf-2(e1370)/insulin-receptor mutant adults, although these animals share several physiological features including resistance to elevated temperature, oxidative stress and hypoxic death. Several globin genes are upregulated following oxygen deprivation and we find that HIF-1 and DAF-2 each are required for this response. Our data indicate that the DAF-2 regulated transcription factor DAF-16/FOXO positively modulates hif-1 transcription under anoxia but opposes expression of the HIF-1 responsive globin genes itself. In contrast, the canonical globin of C. elegans, ZK637.13, is not responsive to anoxia. Reduced DAF-2 signaling leads to enhanced transcription of this globin and DAF-16 is required for this effect.

Conclusion: We found that all 33 putative globins are expressed, albeit at low or very low levels, perhaps indicating cell-specific expression. They show wide diversity in gene structure and amino acid sequence, suggesting a long evolutionary history. Ten globins are responsive to oxygen deprivation in an interacting HIF-1 and DAF-16 dependent manner. Globin ZK637.13 is not responsive to oxygen deprivation and regulated by the Ins/IGF pathway only suggesting that this globin may contribute to the life maintenance program.

Figure 1

Chromosomal distribution of the C. elegans globin genes.

Figure 2

Alignment of 9 representative globins and Sperm whale (SW) myoglobin. ZK637.13 is the canonical globin. It was the first reported globin species in C. elegans and features all characteristics of a standard globin. The next 6 globins have an unusual amino acid residue at CD1. Y75B7AL.1 is a chimeric protein consisting of a G-coupled receptor domain (not shown) and a globin domain. F21A3.6 and F19H6.2 show homology with mammalian cytoglobin and neuroglobin, respectively. Numbers between brackets indicate the number of amino acids preceding and following the globin domain. The residues at position CD1, E7 and F8 are marked in yellow, green and red, respectively.

Figure 3

Genomic structures.

Figure 4

Lack of conservation in the intron insertion positions. Phase 0 introns (55%) are inserted between 2 successive codons; phase 1 (9%) and phase 2 introns (36%) are inserted following the first and second base of a codon, respectively.

Figure 5

Levels of globin expression in third stage larvae (L3) and the alternative dauer (diapause) stage, relative to expression in young adult animals. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean. Only globins that showed a significant (borderline for C36E8.2) difference in one of the three comparisons are displayed.

Figure 6

Ten genes are upregulated in young adult wild-type animals following 12 h of anoxia. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean.

Figure 7

hif-1 mediates induction of the anoxia responsive genes. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean. F22B5.4 is an established hypoxia responsive gene of unknown function [43] and is included as a positive control. Globin ZK637.13 is not sensitive to anoxia and was included as a negative control.

Figure 8

Expression levels of globin genes in daf-2 young adults under normoxia or following 12 h exposure to anoxic conditions, relative to wild-type young adult worms under normoxia. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean. Only globins that showed a significant (borderline for T22C1.2) difference normoxia N2 vs. normoxia daf-2 are displayed.

Figure 9

Expression levels of globin genes in daf-16 and daf-2;daf-16 young adults relative to daf-2 young adult worms. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean.

Figure 10

Expression levels of globin genes in daf-16 young adults following 12 h exposure to anoxic conditions relative to normoxic conditions. The expression ratio's are the average values from 3 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean. Only globins that showed a significant (borderline for R13A1.8) difference are displayed.

Figure 11

Expression level of hif-1 under anoxic conditions in daf-16 young adults relative to N2 young adult worms. The expression ratio's are the average values from 4 replicate cultures (biological repeats). Bars indicate the 95% confidence interval of the mean.