Because the maternal germline produces oocytes, translational regulation plays a critical role in the control of messages that are stored for later use in embryogenesis. In C. elegans, the sex-determination gene
fem-3 has been a key model for this process, since mutations in the 3'-UTR increase FEM-3 activity in the germ line, causing all cells to differentiate as sperm. These mutations were thought to define a germline-specific site for regulation by the two FBF proteins, which are members of the PUF family of translational regulators. Recently, we identified a role for this translational regulation in the soma of XX animals. One of the strongest 3'-UTR mutations is
fem-3(
q96gf). We studied its interactions with
tra-2(mx) alleles, which affect the germ line by disrupting the physical interaction of TRA-2 with TRA-1. When we made double mutants between C. elegans
fem-3(
q96) and any of three different
tra-2(mx) mutations, we observed significant masculinization of tail-tip retraction, spicule formation and ray formation. Although C. briggsae lacks FBF proteins, the regulatory site in the
fem-3 3'-UTR is conserved, and might be targeted by other members of the PUF family. Thus, we made orthologous mutations in C. briggsae
fem-3, which had only a small effect on germ cells, by slightly increasing spermatogenesis in hermaphrodites. However, they caused masculinization of the tail when were combined with our C. briggsae
tra-2(mx) allele
v403. Indeed, this double mutant combination has such a severe effect on C. briggsae that homozygous strains are difficult or impossible to maintain. We conclude that the regulation of
fem-3 translation is not limited to the germ line. Furthermore, this regulation has been conserved during Caenorhabditis evolution. What somatic factors are controlled by FEM-3 and TRA-2? The simplest possibility is their known target TRA-1. Null mutations in
tra-1 cause the soma to become male. They probably act by eliminating TRA-1 repressor, which shuts off male genes. By contrast, we have some mutations, like cbr-
tra-1(
v48), which alter the C-terminus of TRA-1, a region that is cleaved off when the repressor is formed. Homozygous
tra-1(
v48) mutants make oocytes instead of sperm, but the XXanimals make normal female bodies, and the XO animals make normal male bodies. Furthermore, gene dosage tests show that
v48 prevents full-length TRA-1 from activating target genes. Surprisingly, TRA-1 activator mutations like
v48 suppress the
tra-2(mx);
fem-3(gf) masculinization in both C. elegans and C. briggsae. By contrast, they enhance the masculinization of
tra-2(null) mutant tails. Thus, we suspect that important aspects of the complex regulation of TRA-1 by TRA-2 and FEM-3 are designed to prevent aberrant sex-determination from causing developmental disorders.