Worm Breeder's Gazette 10(3): 56

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.

Molecular Cloning of unc-93

Joshua Z. Levin and Bob Horvitz

Figure 1

unc-93 is a member of a group of interacting genes (unc-93, 
sup-10,  in 
muscle structure and function.  unc-93(e1500, n200) and sup-10(n983) 
are rare, altered-function mutations that confer a distinctive 
uncoordinated ('rubberband'), muscle-defective phenotype.  The 
rubberband phenotype suggests a defect in the regulation of muscle 
contraction.  Null alleles of unc-93, sup-10, and 
sup-18 result in a wild-type phenotype alone and are recessive 
suppressors of e1500 and n983.  The wild-type null phenotype of four 
of the five genes suggests they may be functionally redundant.  sup-10 
appears to encode a regulatory myosin light chain (see Cummins, Levin, 
Albertson, Horvitz, & Anderson, WBG, 10(1): 38-39).
We cloned the unc-93 gene by transposon tagging.  A 6.7 kb EcoRI 
fragment containing a Tc1 insertion about 0.3 kb from one end was 
cloned into a phage vector (see Figure).  The cloned DNA, with the Tc1 
removed, was used to probe Southern blots of total genomic DNA from 23 
unc-93 mutants.  Seven of thirteen mut-2-derived alleles showed 
polymorphisms and six of the seven were insertions the size of Tc1.  
Five of ten gamma ray-induced alleles showed polymorphisms; most of 
these were complex rearrangements.  [Additional polymorphisms could 
have been missed because the probe probably does not include the 
entire gene.]  Three restriction fragments that display size 
alterations in unc-93 mutants are shown in the Figure.  The data 
suggest that unc-93 sequences are located on both sides of the HindIII 
site closest to the Tc1 insertion shown.
Probing with the 5.1kb EcoRI fragment containing unc-93 DNA showed 
no cross-hybridizing bands on a genomic Southern at low stringency.  
This result fails to support unc-97s membership in a homologous gene 
family.  In addition, the cloned unc-93 DNA does not hybridize to the 
three known myosin light chain genes, mlc-1, 
low stringency.  [mlc-1 is sup-10;
bridizes with mlc-1; 
sential myosin light chain gene and does not 
cross-hybridize with mlc-1 or mlc-2.]Using the 5.1 kb EcoRI fragment 
to probe a phage library of N2 genomic DNA, we isolated and 
characterized two phage clones.  These two clones overlap and each 
contains about 13 kb of DNA.  We sent the phages to A.  Coulson and J. 
Sulston at the MRC, and they identified an unc-93 contig of about 530 
Using the 5.1 kb EcoRI fragment to probe a cDNA library (made by 
Stuart Kim in our laboratory), we found five hybridizing cDNA clones 
out of 165,000 screened.  Four of the five clones contain a 250 bp 
fragment and the fifth contains a 2650 bp fragment.  The low frequency 
of cDNA clones for unc-93 suggests that unc-93 may produce a low-
abundance RNA and perhaps a low-abundance protein.  Preliminary 
Northern analysis is consistent with this observation.  [In contrast, 
sup-10 encodes a high-abundance RNA.]  The 250 bp clone hybridizes to 
a 0.6 kb HindIII-HindIII genomic fragment (a fragment not found to 
change in any unc-93 mutants).  The 2650 bp clone hybridizes to the 5.
1 kb EcoRI fragment and to an adjacent 3.1 kb EcoRI fragment (to the 
right in the Figure).  Surprisingly, the two classes of cDNA clones do 
not cross-hybridize at any appreciable level.  The larger cDNA clone 
is likely to represent an unc-93 transcript because it localizes to 
the same region of the chromosome as unc-93 mutations.  The status of 
the smaller cDNA clone is unclear.  Northern analysis should resolve 
which cDNA clone(s) represent the unc-93 transcript(s).  We will then 
sequence the unc-93 gene and characterize it molecularly.
[See Figure 

Figure 1