Natural Variation in plep-1 Causes Male-Male Copulatory Behavior in C. elegans

Abstract

In sexual species, gametes have to find and recognize one another. Signaling is thus central to sexual reproduction and involves a rapidly evolving interplay of shared and divergent interests [1-4]. Among Caenorhabditis nematodes, three species have evolved self-fertilization, changing the balance of intersexual relations [5]. Males in these androdioecious species are rare, and the evolutionary interests of hermaphrodites dominate. Signaling has shifted accordingly, with females losing behavioral responses to males [6, 7] and males losing competitive abilities [8, 9]. Males in these species also show variable same-sex and autocopulatory mating behaviors [6, 10]. These behaviors could have evolved by relaxed selection on male function, accumulation of sexually antagonistic alleles that benefit hermaphrodites and harm males [5, 11], or neither of these, because androdioecy also reduces the ability of populations to respond to selection [12-14]. We have identified the genetic cause of a male-male mating behavior exhibited by geographically dispersed C. elegans isolates, wherein males mate with and deposit copulatory plugs on one another's excretory pores. We find a single locus of major effect that is explained by segregation of a loss-of-function mutation in an uncharacterized gene, plep-1, expressed in the excretory cell in both sexes. Males homozygous for the plep-1 mutation have excretory pores that are attractive or receptive to copulatory behavior of other males. Excretory pore plugs are injurious and hermaphrodite activity is compromised in plep-1 mutants, so the allele might be unconditionally deleterious, persisting in the population because the species' androdioecious mating system limits the reach of selection.

Figure 1. Excretory system anatomy and the genetic architecture of excretory pore plugging

(A) A male worm with a copulatory plug on its excretory pore (QG71, right lateral view). (B) Schematic of the excretory system. Paired bilaterally symmetric canals are drained through the excretory, duct and pore cells, immediately posterior and ventral to the pharyngeal bulb. The function of the gland cells is unknown. (C) Ectopic copulatory plugging exhibits transgressive segregation. Of the 82 recombinant inbred lines, ordered by Plep phenotype, many have higher Plep and Plob frequencies than either parental strain (blue and green). The plotted values and prediction intervals are derived from a mixed-effect generalized linear model (see Methods). Colors at the bottom represent strain genotype at the major-effect Plep QTL. (D) QTL mapping of ectopic copulatory plugging in the recombinant inbred lines. Single QTL scans for excretory pore (black) and body (gold) plugging show one and two significant linkages. After conditioning on the major chromosome II QTL, three additional Plep QTL are significant (gray). Horizontal lines show permutation-based thresholds for genome-wide significance at P = 0.05.

Figure 2. Identification of plep-1

(A) The Chr II QTL confers recessive receptivity to excretory pore plugs in heterogeneous F₂ population assays. Proportions of Plep males and their binomial confidence intervals are plotted by genotype at a QTL-linked marker (C for the CB4856 allele, A for AB2). (B) The frequency of Plep QG71 males is significantly reduced by the presence of a single vulvaless hermaphrodite and (C) by aqueous supernatant from hermaphrodites. (D) Excretory pore plugging is not mediated by ascaroside pheromones. QG71 males with an introgression of the dhs-28(hj8) loss of function allele are plugged at higher rates than those with a matched N2 introgression. (E) A 27kb interval defined by NILs behaves as a recessive locus. Strains QG1451, QG1448, and their F₁ are depicted as bars representing chromosome II, with CB4856 DNA in blue and AB2 DNA in green, as in (F), which details the NIL-defined interval. The interval encompasses 10 protein-coding genes and 217 AB2-CB4856 nucleotide variants, of which 5 perfectly associate with Plep phenotype across 11 wild strains (blue: synonymous or intronic, red: non-synonymous). (G) In complementation tests for Mos1-insertion mutants of Y52E8A.4 and Y52E8A.6 (N2 background: orange), Y52E8A.4 fails to complement the Plep NIL QG1448. (H-J) RNAi against Y52E8A.4 induces headplugging when administered to L4 males of non-Plep NIL QG1451 and CB4856 him-5, and to CB4856 him-5 adult males.

Figure 3. plep-1 expression and phylogeny

(A) Maximum intensity projection along the dorsoventral axis showing p_plep-1∷GFP expression in the excretory cell body, just posterior to the terminal pharyngeal bulb, and symmetric canals in a young adult hermaphrodite. (B) Lateral view of a young adult male. In addition to consistent expression in the excretory cell, fluorescence in two dorsal cells in the male tail, shown here, was observed inconsistently. (C) Phylogeny for UNC-93 domain (IPR010291) homologs. UNC-93, MFSD11 and PLEP-1 clades are highlighted (red, blue and yellow). Unboxed genes are MFSD11 paralogs in Caenorhabditis and Pristionchus nematodes.