In the last WBG (10#1:33) I reported that I had cloned a Tc1- containing fragment, eP52, which carries part of the
unc-5 gene from the
mut-6 induced allele
unc-5(
st1003). I have since shown that
st1003 is an inversion with I breakpoint at a Tc1 which is present at the
unc-5 locus in the
mut-6 parent strain RW7097 but not in N2. The presence of a Tc1 element at the
unc-5 locus in RW7097 does not by itself inactivate the gene but probably accounts for the frequent occurrence of
unc-5 alleles in this background. Five other
mut-6- induced that I have examined resulted from rearrangements involving the same Tc1. I cloned eP52 as a 4.9 kb EcoRI fragment. The non-Tc1 sequences from this fragment detect two EcoRI fragments, of 1.3 and 2.7 kb, in N2 DNA. The 1.3 kb fragment is replaced in RW7097 by a 2.9 kb Tc1-containing fragment, which is altered in every
mut-6-induced have examined . The 2.7 kb fragment is not altered in any allele except
st1003, nor does it contain Tc1. The figure summarizes the nature of the rearrangements affecting these fragments in
st1003 and its revertants. The observations upon which the figure is based are explained in more detail below. [See Figure 1] The 2.7 and 1.3 kb fragments detected by eP52 are not contiguous in the N2 genome. Genomic clones containing the 2.7 kb fragment do not contain the 1.3 kb fragment, and they map to the
msp-113 Contig on IV ( thanks to J. Sulston and A. Coulson for fingerprinting the clones), whereas those containing the 1.3 kb fragment map to the eP14 contig, which I described in the last WBG (see also my note on
fem-1 in this issue). They are located between eP14, 0.1 m.u. to the left of
unc-5, and eP60, 0.03 m.u. to the right of
unc-5. The 4.9 kb eP52 fragment therefore represents one ligation product of two breakpoints on LG IV, one in the immediate vicinity of
unc-5. A second Tc1-containing fragment, eP54, cosegregates with
unc-5(
st1003), and like eP52 is absent from revertants of that allele. That eP54 is the other ligation product of the two breakpoints which generated eP52 is suggested by the fact that the 1.3 kb N2 fragment detects both eP52 and eP54 in
st1003. These observations lead me to conclude that
st1003 is an inversion of the DNA between the 2.7 kb and 2.9 kb RW7097 fragments, as indicated in the diagram. Since the inversion accompanied the appearance of a second copy of Tc1, it is likely to have been generated during transposition of Tc1 from the 2.9 to the 2. 7 kb fragment. Reversion of
st1003 results in reversal of the inversion, and both breakpoint fragments retain a copy of Tc1. The 2. 9 kb fragment is regenerated and the 2.7 kb fragment is replaced by one of 4.3 kb which contains Tc1. Of the five other
mut-6 induced
unc-5 alleles I have examined, four (
st1000,
st1001, and
st1002, from D. Moerman, and
e2349) are likely to be inversions. All have one breakpoint within the 2.9 kb RW7097 fragment. The fifth,
ev447, is an
unc-5 utant kindly provided by J. Culotti. It is a deletion which extends rightward from a breakpoint in the 2.9 kb RW7097 fragment and spans the region in which I have found polymorphisms in
fem-1 alleles (see note in this issue). I have not yet located the other breakpoint, but it would be interesting to know if the deletion also accompanied transposition of the
unc-5 Tc1, or involved another mechanism, such as recombination between that Tc1 and a second already present in the same orientation on LG IV in RW7097. I am assuming for the time being that the Tc1 responsible for wreaking havoc at
unc-5 in RW7097 does not reside in the coding sequence of the gene, since its mere presence is immaterial to the Unc-5 phenotype. It could be in an intron or in flanking sequences proximal to a regulatory element. Tc1 insertion at a site 2-3 kb to the left of the 1.3 kb fragment in the Bergerac allele
bx8 (from S. Emmons) results in a weak
unc-5 phenotype. Cosmids spanning this region rescue the phenotype of
unc-5(
e53) animals when injected into oocytes (see note by Hope et al., this issue).