The HSN motor neurons innervate the vulval muscles and drive egg laying in the hermaphrodite. After innervating the vulval muscles, the HSN processes enter the ventral cord and run anteriorly into the nerve ring where they synapse with other neurons. We previously proposed a model in which there is an inhibitory input onto the HSNs in the nerve ring that is active in the absence of bacteria (Desai et al., 1987 C. elegans CSH abstracts, p.74). In support of this model,
unc-34 and
unc76 mutants are egg-laying constitutive (Egl-c; i.e. unlike wild-type animals, they lay eggs in the absence of bacteria) and are defective in the anterior growth of the HSN processes, which terminate prematurely in the ventral cord before reaching the nerve ring. Presumably, these mutants are Egl-c because the HSN is dissociated from the inhibitory input in the nerve ring. To identify mutants abnormal in the postulated HSN inhibitory input, we have been making double mutants between
unc-76 and various egg-laying defective ( Egl-d) mutants. If our interpretation of the
unc-76 egg-laying phenotype is correct,
unc-76 should be epistatic to mutations that make the inhibitory neuron function constitutively (i.e. the double mutant should be Egl-c). In contrast, mutations that cause an HSN- defective phenotype should be epistatic to
unc-76 (i.e. the double mutant should be Egl-d).To verify that mutations that cause an HSN- defective phenotype are epistatic to
unc-76, we constructed double mutants between
unc-76(
e911) and
egl-1(
n487) (the HSNs undergo programmed cell death in the hermaphrodite),
egl-5(
n1439) (the HSNs are defective in migration and do not contain detectable serotonin, an HSN neurotransmitter),
egl-45(
n999) (the HSNs degenerate during the L4 stage at the time of HSN maturation) and
unc-86(
e1507) (the same as the
egl-45 phenotype). We were surprised to find that the
egl-45 e is partially suppressed in the
egl-45; utant. This suppression is specific to egl- 45, since the other double mutants are Egl-d (they have the HSN- defective phenotype of the egl mutant). Interestingly, the suppression of
egl-45 occurs by rescuing the HSN-degenerative phenotype. Only 25% (N=146) of
egl-45 adults contain detectable HSN serotonin as determined by immunohistochemical staining, whereas 57% ( N=195) of
egl-45; had at least one HSN that stained for serotonin. In addition,
egl-45; o stained more intensely than
egl-45 HSNs, and the HSN processes were easier to detect. Although there is some aberrant HSN process outgrowth in
egl-45; when the HSN process enters the ventral cord, it stops before entering the nerve ring, which is the Unc-76 HSN phenotype. An additional Egl- c mutant,
unc-42(
e270), also suppresses the HSN-defective phenotype of
egl-45. 82% (N=107) of
egl-45; had at least one HSN that stained for serotonin. In contrast to
unc-42 and
unc-76, 66) does not suppress the
egl-45 HSN phenotype (25% animals had a staining HSN(s); (N=124). This result was surprising, since
unc-34 mutants are Egl-c and like the
unc-76 mutant, have HSN processes that terminate in the ventral cord before entering the nerve ring. We are considering two general explanations that can account for the suppression of the
egl-45 HSN-defective phenotype by
unc-42 and
unc-76 mutations. One possibility is that the two gene products (or processes in which the products are involved) interact in the same cell, presumably the HSN. Alternatively, the HSN could require a synaptic or humoral input to express the
egl-45 phenotype, and this input could be defective or missing in
unc-42 and
unc-76 mutants. We are testing the second possibility in
egl-45 animals by killing with the laser neurons that have inputs onto the HSNs.