Functional distinction in oncogenic Ras variant activity in Caenorhabditis elegans

Abstract

Ras genes are important oncogenes that are frequently mutated in cancer. Human oncogenic variants exhibit functional distinctions in terms of their representation in different cancer types, impact on cellular targets and sensitivity to pharmacological treatments. However, how these distinct variants influence and respond to the cellular networks in which they are embedded is poorly understood. To identify novel participants in the complex interplay between Ras genotype and cell interaction networks in vivo, we have developed and tested an experimental framework using a simple vulva-development assay in the nematode C. elegans. Using this system, we evaluated a set of Ras oncogenic substitution changes at G12, G13 and Q61. We found that these variants fall into distinct groups based on phenotypic differences, sensitivity to gene dosage and inhibition of the downstream kinase MEK and their response to genetic modulators that influence Ras activity in a non-autonomous manner. Together, our results demonstrated that oncogenic C. elegans Ras variants exhibit clear distinctions in how they interface with the vulva-development network and showed that extracellular modulators yield variant-restricted effects in vivo.

Keywords: C. elegans vulva development; Oncogenic Ras variants; Ras/MEK signaling pathway.

Fig. 1.

C. elegans let-60 alleles that code for the oncogenic variant G13D exhibit normal vulva development. (A-C) Adult wild-type (A) and let-60(gu287[G13D]) (C) animals exhibit a smooth ventral surface, whereas many let-60(n1046[G13E]) animals (B) exhibit notable ventral bumps (black arrows) corresponding to the Muv phenotype. Animals are anterior to the right, ventral down. (G) Quantification of experiments described for A-C. Sample sizes in G: n=41 for let-60(+); n=52 for let-60(n1046[G13E]); n=45 for let-60(gu287[G13D]). (D-F) Vulva structure in animals at larval stage 4 (L4). Wild-type (D) and let-60(gu287[G13D]) (F) animals exhibit formation of a single vulval opening (white arrow) produced when three of six vulval precursor cells (VPCs) are induced to produce vulval tissue. let-60(n1046[G13E]) animals (E) frequently exhibit additional invaginations (black arrow) that result when VPCs in addition to the normal three are induced to produce vulval tissue. (H) Quantification of experiments described for D-F. Sample sizes in H: n=30 for let-60(+); n=55 for let-60(n1046[G13E]); n=34 for let-60(gu287[G13D]). Error bars correspond to standard error of the proportion (G) or standard deviation (H). Asterisks indicate statistical differences compared with wild-type (control) (P<0.05); two proportion Z-test for G, t-test for H.

Fig. 2.

Activity associated with let-60 variants can be evaluated in a controlled manner using MosSCI-based transgenes. (A). Animals homozygous for a let-60 gene copy at the ttTi5605 landing site on LG II – i.e. let-60(+); Si(let-60(+)) – and corresponding to four rather than the wild-type two gene copies in the genome, exhibit a wild-type phenotype. Introduction of the let-60(G13E) variant at this landing site can weakly interfere with wild type, producing a low frequency of Muv animals. Likewise, transposon-introduced wild-type gene copies can modulate let-60(G13E) at the endogenous locus, reducing the frequency of Muv animals. Finally, presence of let-60(G13E) at all four sites (endogenous and transposon site) confers a strong Muv phenotype. Sample sizes: n=40 for let-60(+); n=42 for let-60(G13E); n=44 for let-60(+); Si(let-60(+)); n=40 for let-60(+); Si(let-60(G13E)); n=41 for let-60(G13E); Si(let-60(+)); n=42 for let-60(G13E); Si(let-60(G13E)). (B) In contrast to let-60(G13E), increasing the dosage of let-60(G13D) by adding transposon-based copies at ttTi5605 does not alter the phenotype. Sample sizes: n=43 for let-60(G13D); n=45 for let-60(+); Si(let-60(G13D)); n=47 for let-60(G13D); Si(let-60(G13D)). Error bars correspond to standard error of the proportion. Asterisks in A indicate statistical differences (P<0.05); two proportion Z-test and Bonferroni correction, comparing each G13E-containing strain to let-60(G13E) (control).

Fig. 3.

Introduction of some mutations orthologous to human oncogenic Ras variants can produce the Muv phenotype in C. elegans. Adult animals homozygous for transposons bearing wild-type or different let-60 oncogenic variants. (A-G) Animals in each panel are oriented as in Fig. 1, with their anterior to the right and ventral down. Animals homozygous for wild-type (A), let-60(G13D) (C) or let-60(G12C) (F) transposons exhibit a smooth ventral surface. Most let-60(G13E) (B) transposon-bearing animals are similarly nonMuv. In contrast, the Muv phenotype is observed among animals bearing the let-60(G13R) (D), let-60(G12D) (E) or let-60(Q61R) (G) variants. Ventral bumps indicative the Muv phenotype are indicated with black arrows. (H) Quantification of the Muv phenotype. Sample sizes: n=53 for let-60(+); Si(let-60(+)); n=45 for let-60(+); Si(let-60(G13E); n=41 for let-60(+); Si(let-60(G13D)); n=43 for let-60(+); Si(let-60(G13R)); n=37 for let-60(+); Si(let-60(G12D)); n=37 for let-60(+); Si(let-60(G12C); n=42 for let-60(+); Si(let-60(Q61R)). Error bars indicate the standard error (±s.e.) of the proportion. Asterisks indicate statistical differences (P<0.05); two proportion Z-test and Bonferroni correction, comparing each single-copy oncogenic variant transposon to the wild-type transposon. n.s., not significant.

Fig. 4.

let-60 oncogenic variants differ in dosage sensitivity. A single copy of the let-60(G12D) or let-60(Q61R) transposon (in an otherwise wild-type background) is sufficient to produce a strong Muv phenotype, whereas let-60(G13R) produces only a weak Muv phenotype. The dominance test for let-60(n1046[G13E]) was completed in parallel, as a comparison. Sample sizes: n=30 for let-60(G13E)/let-60(+); n=31 for let-60(+); Si(let-60(+))/−; n=41 for let-60(+); Si(let-60(G13R))/−; n=32 for let-60(+); Si(let-60(G12D))/−; n=35 for let-60(+): Si(let-60(Q61R))/−. Error bars indicate the standard error (±s.e.) of the proportion. Asterisks indicate statistical differences (P<0.05); two proportion Z-test and Bonferroni correction, comparing each single-copy oncogenic variant transposon to the wild-type transposon. n.s., not significant.

Fig. 5.

let-60 oncogenic variants exhibit distinct dosage sensitivity to MEK activity. (A) The Muv phenotype associated with each oncogenic variant is blocked when animals are grown on plates containing 60 µM of the MEK inhibitor U0126. Sample sizes were as follows. let-60(G13E): n=163 (DMSO) and n=132 (U0126). let-60(+); Si(let-60(+)): n=139 (DMSO) and n=102 (U0126). let-60(+); Si(let-60(G13R)): n=152 (DMSO) and n=114 (U0126). let-60(+): Si(let-60(G12D)): n=54 (DMSO) and n= 72 (U0126). let-60(+); Si(let-60(Q61R)): n=80 (DMSO) and n=81 (U0126). Error bars correspond to the standard error of the proportion. Asterisks indicate statistical differences; P<0.05; two proportion Z-test and Bonferroni correction, comparing each U0126 treatment condition to its DMSO control. (B) Plotted is the response (left to right) of strains let-60(G13E) (blue); let-60(+); Si(let-60(G13R)) (red); let-60(+); Si(let-60(Q61R)) (yellow) and let-60(+); Si(let-60(G12D)) (green) to different doses of the MEK inhibitor U0126. Plotted are the results of eight trials, with at least 30 animals per condition per trial (exact sample sizes are included in Table S4). Pairwise comparison of EC₅₀ values from eight trials for each strain was carried out using one-way ANOVA and Tukey's HSD test, indicating that all conditions are statistically different (P<0.05) except for let-60(G13E) with let-60(+); Si(let-60(G13R)) as well as let-60(+); Si(let-60(G13R)) with let-60(+); Si(let-60(Q61R)). Error bars indicate the standard error (±s.e.). The graph in B was generated using the Quest Graph calculator (AAT Bioquest, Inc.).

Fig. 6.

let-60 oncogenic variants exhibit differential sensitivity to genes identified in the let-60(G13E) genetic background as non-autonomous modulators of activated Ras activity. (A) Schematic representation of the genetic chimera method, as described by Artiles et al., (2019). GPR-1 is overexpressed [GPR-1(oe)] in the maternal germline and oocytes, causing the pronuclei to segregate to daughter cells without fusing in the zygote. The chimera class evaluated in this experiment is that in which replicated maternal chromosomes segregate to the AB blastomere, i.e. the precursor to the VPCs, whereas replicated paternal chromosomes segregate to the P1 blastomere, i.e. the precursor to the somatic gonad and most body wall muscle cells. (B) Chimeric animals with paternal (P1) DNA and let-60(+) in the P1 cell lineage, and maternal let-60 oncogenic genotypes – with wild type for each paternal gene – in the AB cell lineage. Paternally introduced alleles of szy-5 and hpo-18 confer a strong suppression of the Muv phenotype in let-60(n1046[G13E]) mutants (Corchado-Sonera et al., 2022). Animals homozygous for Si(let-60(G13R)) transposons in AB-derived cells are moderately sensitive to disruption of szy-5 and hpo-18, whereas Si(let-60(G12D)) or Si(let-60(Q61R)) are not. Sample sizes were as follows. let-60(G13E): n=41 for control (+), n=45 for hpo-18(ok3436) and n=41 for (szy-5(tm810). let-60(+); Si(let-60(G13R)): n=35 for control (+), n=44 for hpo-18(ok3436) and n=56 for szy-5(tm810). let-60(+); Si(let-60(G12D)): n=32 for control (+), n=30 for hpo-18(ok3436) and n=33 for szy-5(tm810); let-60(+); Si(let-60(Q61R)): n=50 for control (+), n=30 for hpo-18(ok3436) and n=36 for szy-5(tm810). Error bars indicate the standard error (±s.e.) of the proportion. Asterisks indicate statistical differences (P<0.05); two proportion Z-test and Bonferroni correction, comparing each variant with mutant paternal contribution to its wild-type control. n.s., not significant.