The
pat-3 gene encodes a 1 integrin (Gettner et al., in preparation), and is defined by sterile (
rh151 ),lethal (
rh54 ),and mild (
rh96 )alleles (Hedgecock et al, in preparation). This last allele has several impenetrant migration defects, including the CAN, HSN, Q cells, and a peculiar distinctive looping of the posterior gonadal arm. In all animals, however, the excretory canal process extends only to V2 at hatch, and grows out to varying extents thereafter. In the course of our studies, we constructed the double mutant
rh96 dpy-17 (
e164).To our surprise, this strain has very short canal processes, is Egl with a small brood size, and is dumpier than
e164 alone. We also saw that
e164 alone has occasional defects in the posterior canal processes; on rare occasions they branch inappropriately, and in other cases (5-10% penetrance) turn dorsally and proceed anteriorly into the head. It is not clear if these defects are caused by defective specific guidance cues, or simply by the extreme dumpy shape in the early larval stages, when the posterior canal processes complete outgrowth.
rh1019 is a mutation between
unc-36 and
unc-69 on LGIII. This recessive Tc1 -inducedallele also causes impenetrant defects in migration of the CAN, HSN, ALM neurons, and a 100% penetrant defect in the canal cell. Its processes are always short and wide and meander considerably; the length varies from animal to animal. The
rh96 rh1019 double is much more severe than either mutant alone. The canal processes are virtually non-existant, CAN and HSN are not at their correct positions in almost every animal (the adults are Egl with withered tails as a result), the distal tip cells stop migrating early in some cases; the alae have gaps and are twisted, and the animals appear lumpy, with mispositioned hypodermal cells. In addition, the lamina defining the gonad is weakly defined by DIC microscopy, and could be missing in some animals. These phenotypes and the map position suggest that rh 1019 could be allelic to
emb-9 ,which encodes collagen type IV (Guo et al., Nature 349, 707-709, '89). 1 integrin is known to bind to this collagen in other animals. We are currently complementing
rh1019 to
emb-9 see if this is the case.