Description

Regulation of the essential trace element zinc is necessary to avoid the toxic consequences caused by too little or too much of this metal (Vallee and Falchuk 1993; Rosen 2006). The zinc-response pathway has been extensively studied in the nematode roundworm Caenorhabditis elegans and several genes have been discovered that function to modulate sensitivity to both high and low zinc concentrations (Dietrich et al. 2016). Recently, we identified a quantitative trait locus (QTL) on the center of chromosome V, indicating that natural genetic variation between the laboratory strain, N2, and a genetically divergent wild isolate from Hawaii, CB4856, contributes to differential responses to excess zinc (Evans et al. 2020).

The confidence interval for this QTL is 1.6 Mb and contains 629 genes (WS263). Of these genes, 113 have one or more genetic variants predicted to modify the amino-acid sequence of the protein. However, protein-coding variation is just one of the ways that genetic variation can cause phenotypic variation. Another is variation in gene expression, which is hypothesized to be important in the majority of complex traits (Hindorff et al. 2009). We identified 83 expression QTL (eQTL) within this 1.6 Mb region using the eQTL dataset that mapped expression differences among a panel of recombinant inbred advanced intercross lines (RIAILs) also derived from N2 and CB4856 (Rockman et al. 2010; Evans and Andersen 2020). The most significant eQTL in this region caused a change in expression of the gene cdr-6 (Figure 1A). This gene, a homolog of the cadmium-response gene cdr-1, is highly expressed in intestinal and pharynx muscle cells during larval stages and is downregulated after treatments with arsenic, cadmium, or zinc (Dong et al. 2008). Furthermore, cdr-1 was previously shown to mitigate cadmium toxicity in C. elegans (Dong et al. 2005; Hall et al. 2012). Together, these data suggest that expression of cdr-6 might be toxic to C. elegans in the presence of heavy metals. Additionally, RIAILs with the CB4856 allele on chromosome V naturally express higher levels of cdr-6 (Figure 1B) and are also more sensitive to zinc than RIAILs with the N2 allele (Evans et al. 2020). This result indicates that strains with naturally low expression of cdr-6 are more resistant to excess zinc than strains with naturally high expression of this gene.

To test this hypothesis, we used CRISPR-Cas9 genome editing to create strains with large deletions of cdr-6 in both the N2 and CB4856 genetic backgrounds (Figure 1C,D). Because expression of cdr-6 was higher in RIAILs with the CB4856 allele (associated with zinc sensitivity) than in RIAILs with the N2 allele (associated with zinc resistance) (Figure 1B), we expected that a knockout of cdr-6 in the CB4856 genetic background would cause increased resistance to excess zinc. Alternatively, if variation in expression of cdr-6 underlies the zinc-response QTL on chromosome V, a knockout of cdr-6 in the N2 genetic background should not cause an increase in zinc resistance. We exposed N2, CB4856, and two strains with independently derived cdr-6 deletion alleles in each genetic background to elevated zinc and measured their optical densities using a high-throughput assay with the COPAS BIOSORT (Andersen et al. 2015; Evans and Andersen 2020; Evans et al. 2020). We found that strains with a deletion of cdr-6 phenocopied the strain with the same genetic background (Figure 1E), suggesting that differences in expression of cdr-6 do not underlie zinc responses.

This study not only provides evidence against cdr-6 as the causal gene underlying differences in zinc resistance between the N2 and CB4856 strains, but also indicates that cdr-6 does not influence zinc resistance in C. elegans. Although a previous study showed that cdr-6 was downregulated in response to zinc (Dong et al. 2008), our results do not contradict theirs. In fact, the authors also showed that the accumulation of fluid-filled droplets in the pseudocoelom, a phenotype observed in strains with inhibited CDR-6 expression using RNAi, is not increased in response to zinc or cadmium exposure (Dong et al. 2008). Taken together, we conclude that expression of cdr-6 decreases in response to zinc but animal development in the presence of zinc is not affected by cdr-6 function. It is likely that cdr-6 does not function in the nematode zinc response but rather is downregulated as an indirect effect of zinc exposure.