We have attempted to 'Tc1 tag' the
sqt-1 gene using a complementation screen with the recessive left roller allele
sc13. Heterozygous
unc-4(
e120) 3) males were mated to TR679 hermaphrodites and the Fl progeny were screened for left rollers (putatively
e120 sc13 /
sqt-1 Tc1). Five left roller animals were detected among 16,000 Fl progeny. So far, four of these animals have been examined. All four of these animals segregate only Rol-Unc and Rol progeny. We had expected to see 1/4 wild-type progeny, since the null phenotype for
sqt-1 is believed to be wild type. Instead, the putative
sqt-1 Tc1 alleles all behave as recessive left roller mutations. Other labs have reported that Tc1 insertion mutations do not necessarily produce null phenotypes. Two independent lines (mutant pairs) were established from each of the four original mutants. Each line was backcrossed to N2 five times, and the flanking markers
unc-4(
e104) and
lin-29(
n836) were crossed in and out. The backcrossed lines were examined for the presence of new Tc1 elements. Each of the lines contained only 0-3 new Tc1's. For three of the mutant pairs there was no new Tc1 element present in both members of the pair. Therefore, the left roller phenotype in these animals is not caused by a Tc1 insertion. It is possible that these mutations are the result of a Tc1 insertion and imprecise excision, or they may be due to other mutational events. In one of the mutant pairs, BE143-BE144, there is a new Tc1 element present in both members of the pair. This Tc1 element is on a 3kb EcoRI fragment, and has been isolated from a size-selected sub-library of BE143 DNA cloned in lambda
gt10. The insert from the phage clone has been subcloned into pUC18 and named pJK104. DNA flanking the Tc1 insert in pJK104 has been probed against genomic DNA's from N2, EM1002, and BE145 (a strain congenic for the
sqt-1 region from TR679). It is clear that the Tc1 element in pJK104 is not present in any of these strains, and therefore it has transposed in the BE143-BE144 strains. pJK104 was used to isolate four overlapping cosmid clones, and Alan Coulson has kindly provided three more overlapping cosmids. The cosmids cover approximately 80kb. Since we have not isolated revertants of either BE143 or BE144, we do not know that this Tc1 element is actually inserted into
sqt-1. We would not be greatly surprised if
sqt-1 were a member of the collagen multigene family. However, probing of the cosmid clones has shown that there are no collagen genes detectable within the 80kb covered by the cosmids. Either
sqt-1 is not contained in these clones or it is not a collagen. We are also attempting to localize these clones relative to the deficiencies in the region of
sqt-1. So far, various mnDf DNA's have been probed with pJK104 and one of the cosmid clones, with confusing results. Both the plasmid and cosmid clone repeatably show reduced hybridization intensity when probed to mnDf77 and mnDf87 DNA's, but show normal intensity with mnDf83 and mnDf89 DNA's. This is inconsistent with the fact that the latter two deficiencies extend beyond the former two deficiencies in both directions, according to the genetic mapping data. Internal controls have been used to equalize for the total amounts of DNA in different lanes. We are trying to clear up this discrepancy. Also, DNA's from strains carrying various
sqt-1 alleles are being analyzed to determine if any alterations in restriction pattern are detectable by probing with the cosmid clones. We have recently begun another screen for Tc1 transposition into sat- l using Mike Finney's HH*6 mutator strain. Left roller-mutator animals are being constructed and will be screened for non-roller progeny.