Mup-2 is a gene required for muscle attachment and oviduct function in C. elegans. The muscle phenotype and temperature sensitive period of
mup-2(
e2346ts) have been described in WBG 10(3), 104-105 and CSH 1990 meeting abstracts, page 83. RFLP mapping around
mup-2: We mapped
mup-2(
e2346) between
sup-7 and
lev-9 on X. Cosmids from the contig between
sup-7 and
unc-4 were obtained from A. Coulson and John Sulston and used to probe genomic Southern blots of different wild type strains. Four RFLPs were identified between RC301 and N2. To map the RFLPs we constructed the strains RC301/mup-2
(e2346)
unc-6(
e78) and RC301/dpy-8
(e130)
mup-2(
e2346) and picked
unc-6 recombinants and
dpy-8 non
mup-2 recombinants. Southern blots of DNA of these homozygosed strains were probed to determine the segregation of the RFLP relative to the
mup-2 gene (see figure 1). Mup-2 maps between
sup-7 and the RFLP on F14G9. A 16 kb BglII RFLP in T15G10 was not separated from
mup-2 in a large number of recombinants and is therefore likely to be very close to
mup-2.[See Figure 1] The alignment of the physical and genetic maps between
sup-7 and unc- 4 (see figure 2) shows that recombination frequency is not linear between
sup-7 and
unc-4 and that there may be a recombination hotspot between K04G12 and
sup-7. Extrapolating distances from the genetic to the physical map is thus not a reliable method to choose cosmids for transformation rescue. [See Figure 2] F2 transformation rescue of
mup-2: We decided not to inject
e2346b homozygous because
e2346ts has a gonad defect at 25 C. Animals grown at 15 C have a small broodsize of about 40 or are sterile. Interestingly
mup-2(
e2346)
unc-6(
e78) hermaphrodites are about twice as fertile as
mup-2(
e2346) by itself. We co-injected the three overlapping cosmids T07F4, T15G10 and K04G12 and the
rol-6 roller plasmid (cfr. WBG 11(1)) into the syncitial gonads of +
mup-2(
e2346)
unc-6(
e78)/lon-2
(e678)++ hermaphrodites and grew the hermaphrodites at 25 C. Surviving
unc-4 would be either rescued
mup-2 unc-4 or +
mup-2 unc-6/lon-2 +
unc-6 recombinants. These recombinants segregate
lon-2 thus be easily excluded. We have not obtained rescued
mup-2 unc-4 in the F1 of injected hermaphrodites. However we have obtained F1 rollers of the genotype
mup-2 unc-4/lon-2 which segregate apart from the expected offspring many sterile unc's, fertile unc's and rollers. The fertile unc's breed true and are homozygous
mup-2 unc-4 (shown by outcrossing to N2). The sterile unc's are likely to be
mup-2 unc-6 with insufficient levels of wild type product to rescue the sterility of
mup-2. We know from the phenotype of
mup-2(
e2346); stDp2 that more than 33% wild type activity is required to rescue the gonad phenotype of
mup-2(
e2346) at 25 C. It is possible that we did not obtain rescue in the F1 because relatively high levels of wt product may be required for
mup-2 rescue. Cosmids T15G10, C32C12 and F36F9 injected separately or as a mixture do not rescue
mup-2 and serve as negative controls. We intend to narrow down the
mup-2 gene by transformation rescue with overlapping lambda clones from the
mup-2 region. General points: (1) The
mup-2 rescue and the
rol-6 phenotype need not segregate together in the rescued lines. (2) Compared to other wild type strains used, RC301 appears to be rich in RFLP's in both the
unc-6 and the
unc-53 region. (3) We have screened lambda 2001 genomic libraries directly with nick translated cosmid clones from A. Coulson and J. Sulston (lorist and PJB8 vector based) and had no problems recognizing positives over background although there is homology between cosmid vectors and lambda arms. (4) We have also obtained transformation rescue for
unc-53 by injection in the gonad arm adding to the list of genes for which this method works. Acknowledgement: This approach to cloning
mup-2 would not have been possible without the advice, physical map and cosmid clones from A. Coulson and J. Sulston.