Worm Breeder's Gazette 11(4): 65

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.

let-23 = kin-7: A Receptor Tyrosine Kinase Gene Necessary for Vulval Differentiation

Raffi Aroian, Makoto Koga, Jane Mendel, Yasumi Ohshima and Paul Sternberg

Figure 1

Genetic analysis suggests that let-23 is required in the vulva for 
both an inductive function necessary for vulval differentiation and an 
inhibitive function necessary for proper specification of vulval fates 
(WBG, 10(2), p.81).  Mutations in let-23 result in the Vulvaless and 
Hyperinduced phenotypes in the vulva, larval lethality, abnormal male 
spicules, sterility, and transformation of P12.p to P11.p fate.  Some 
of our let-23 alleles uncouple these phenotypes from one another, 
suggesting multifunctionality at the locus (CSH C.  elegans 1989 
Meeting Abstracts, p.  108).  The let-23 gene maps to the region 
around the cosmid T08E2, and microinjection of that cosmid rescues let-
23 mutant phenotypes (WBG, 11(2), p.  67).  Furthermore, T08E2 
contains a tyrosine kinase gene, kin-7, cloned by low-stringency 
hybridization with a v-ros probe (WBG, 11(1), p.  37).  We now show 
that let-23 and kin-7 are the same gene.
We inject using the rol-6 dominant selection scheme developed by 
Craig Mello.  We coinject DNA from the let-23 region (abbreviated LDNA)
along with rol-6(dom) DNA (abbreviated ROL) into the strain let-23(
mn23) unc-4(e120)/mnC1(dpy-10(e128) unc-52(e444)), where mn23 is a 
100% penetrant lethal let-23 allele and mnC1 is a balancer for the 
region.  We score rescue by two criteria which, in both positive and 
negative control experiments, always correlate with each other.  First,
if stable Roller lines, which should be mn23 e120/mnC1; array(
LDNA+ROL), segregate Rol Uncs, (genotypically mn23 e120/mn23 e120; 
array), then this suggests that the LDNA rescues the let-23 lethal 
phenotype (in all but one case the vulvaless and sterile phenotypes of 
mn23 are also rescued).  Second, rescue of the F1 progeny of the 
injected hermaphrodites is indicated by the presence of F1 Rollers 
which are Unc.  These Rol Uncs are often sterile (indicating transient 
rescue), though sometimes they are fertile.  We have shown that a few 
cosmids and plasmid constructs also rescue the Vulvaless allele sy97 
and the lethal allele mn216.The 15 kb genomic plasmid pK7-13.8, which 
contains the kin-7 gene and roughly 3 kb on either side, rescues let-
23 (see Figure).  This result suggests that let-23 is contained in pK7-
13.8.  The plasmid YO#NG213, which deletes about 4 kb of 3' genomic 
sequence from pK7-13.8 (including approximately 800 bp from the C-
terminus of cDNA sequence but not including the kinase domain), also 
rescues.  This observation suggests that no sequences 3' of the kinase 
domain are needed for rescue.  However, the constructs pIS6 and pK7-
DTK both of which delete an additional 1 kb (in genomic and cDNA 
sequence) from the C-terminus of YO#NG213 and lack the kinase domain, 
fail to rescue.  Thus, a deletion within kin-7 that lacks the kinase 
domain, but still maintains about half the N-terminal end of cDNA 
sequence, fails to rescue.  This observation suggests that the kinase 
domain is required for rescue and that no sequences 5' of kin-7 on pK7-
13.8 are needed for rescue.  Finally, the plasmid pK7-5.5 which 
contains the kinase domain and C-terminal sequences but lacks 10 kb of 
5' pK7-13.8 sequences fails to rescue.  Thus, the N-terminal half of 
kin-7 is needed for rescue.
By these rescue results, therefore, the let-23 locus and the kin-7 
tyrosine kinase gene are the same.  We can furthermore show that the 
two key negative results (pK7-DTK and pK7-5.5) are due to deletions in 
let-23 and not due to faulty plasmids (e.g.  gratuitous point 
mutations) since coinjection of pK7-5.5 with pK7-DTK results in rescue.
The one stable line we have segregates Rol Uncs, although at a lower 
frequency than seen with above constructs, and these are mostly 
sterile.  The coinjection also results in transient F1 Rol Uncs, 
although again at a lower frequency.
The presence of repeat elements in the region have stymied our 
attempts to detect allele specific polymorphisms on Southern blots.  
Nonetheless, we have localized a point mutation in the null allele sy5 
to the kinase domain using hydroxylamine modification of C mismatches (
the protocol kindly provided by Bob Barstead), further supporting our 
injection results.
Analysis of let-23 sequence suggests that let-23 encodes a receptor 
tyrosine kinase related to the EGF Receptor family.  We hypothesize 
based on genetic results and its inferred molecular structure that let-
23 might be the receptor in the vulva precursor cells for an 
intercellular signal required for vulval differentiation.
[See Figure 1]

Figure 1