The fem--3(+) gene specifies male development. In XX hermaphrodites,
fem-3 is required only in the germline: it must first be active to direct spermatogenesis and then it must be packaged into oocytes for a maternal contribution to the embryo. This maternal product must be negatively regulated so that it will not transform oocyte precursors into sperm. Nine temperature sensitive gain-of-function (gf) alleles of
fem-3 cause the aberrant production of a vast excess of sperm in hermaphrodites; no oocytes are made, and no alteration is observed in the hermaphrodite soma. Genetic studies suggest that
fem-3(gf) mutations do not increase
fem-3(+) activity, but instead produce a novel unregulated product (1). The germline-specific masculinization of
fem-3(gf) is consistent with the finding that
fem-3 RNA's are limited to the germline in hermaphrodites (2). The wild-type
fem-3 gene has been cloned, and its sequence and gene structure elucidated ( 2; Rosenquist, unpublished). We have cloned all nine
fem-3(gf) genes. We sequenced one
fem-3(gf) allele,
q20, in its entirety and found a single base pair change in the 3' untranslated region (3'-UTR). Partial sequencing of the rightmost eight
fem-3(gf) alleles similarly reveal changes in the 3'-UTRs. The two weak
fem-3(gf) alleles are both T to C changes of the same base pair; the five intermediate
fem-3( gf) alleles are all C to T changes of an adjacent base pair; and the one strong
fem-3(gf) allele is a G to T change a few bases pairs 3' of the other mutations (see figure). One super strong
fem-3(gf),
q95, which is barely temperature sensitive, is a 114bp deletion centered around the region of the point mutants and contained within the 3'-UTR, It may be significant that most of the
fem-8(gf) mutants lie in a CTT repeat. The short sequence in the 3'-UTR that has been shown by transformation experiments to be important to
unc-54 function also contains a CTT repeat (Fire, WBG 10 No. 2); furthermore, this region is conserved in the 3'-UTR of the
mlc-1 gene (Cummins and Anderson, pers . comm. ) . The clustering of these
fem-3(gf) base pair changes within a stretch of five base pairs indicates that this small region is centrally involved in the regulation of
fem-3. The temperature sensitivity of the mutations suggests the involvement of an RNA/RNA or RNA/protein interaction. In searches for secondary structure in the region, there is no striking stem including or near the base pairs identified by mutation. The negative regulation of fem- 3 activity identified by the mutations may act through RNA stability, efficiency of translation, or localization of the RNA within the germ line.