Worm Breeder's Gazette 11(2): 25
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.
To further study hsp-16 and ubq-1 expression we are introducing into the worm fusions of these genes to the E. coli -galactosidase gene and assaying for enzymatic activity. Two methods are being employed to study gene expression. Stable transgenic lines are established by microinjecting into hermaphrodite oocytes the -gal fusion vector of interest along with an antisense unc-22 construct (kindly provided by Andrew Fire) and selecting for twitching progeny. Alternatively, in a transient assay the -gal fusion vector is injected alone into nematode oocytes and the resulting progeny are directly screened for - galactosidase expression. The latter method has been extremely useful for initial testing of constructs, e.g. in deletion analysis, where only a positive or negative answer with respect to expression is required, and as a control, for comparison with stable lines in order to determine if a co-injected selectable plasmid alters the expression of the fusion vector. Deletion analysis of 930 bp of sequence upstream of the translational start of the ubq-1 (previously named ubiA) gene has yielded some surprising results. Nematodes carrying constructs containing all of this region (ubq delta0) express -galactosidase primarily in embryos and in the head and tail and along the body wall of all other stages. It is difficult to determine the exact cells which express the construct since the enzyme is readily diffusible. However, these cells are probably neuronal in nature since transformation with a larger upstream fragment (ubq 1.8 Eco) which also contains the SV40 nuclear localization signal directs expression to neuronal nuclei of the pharynx, head and tail, and along the ventral nerve cord. In addition, possibly some large intestinal nuclei are affected in the early larval stages. No difference in - galactosidase expression was observed in ubq delta0 stable lines when treated with heat shock. Progressively larger deletions extending from the 5' end of ubq delta0 do not significantly alter the pattern of expression until 830 bp of sequence has been removed, 330 bp downstream from the transcriptional start (5'SS), and about 120 bp upstream of the initial methionine. ubq delta 830 embryos stain blue intensely within one hour while ubq delta918 embryos stain marginally after overnight development. Embryos carrying ubq-pvu delta918, a construct of the first 550bp of ubq delta0 fused to ubq delta918 , show expression intermediate between that of ubq delta830 and ubq delta918. No expression in adults or larvae has been observed for ubq delta830, ubq delta918, or ubq-pvu delta918. The ubq delta918 break point is only 30 bp upstream of the 3' acceptor site (3'ss) for the leader (SL1) which is trans-spliced onto the hnRNA. To determine if the -gal message in ubq delta0 is trans-spliced as ubq-1 RNA normally is, we used PCR and oligonucleotides for SL1 and -gal to selectively amplify only a trans-spliced -gal transcript. The resulting band was eluted, cloned and is currently being sequenced to verify the presence of SL1. It will be interesting to see if -galactosidase expression in both ubq delta830 and ubq delta918 worms results from a trans-spliced transcript as well or if perhaps the low residual activity in ubiA delta918 is the result of a cis spliced mechanism. We are also studying the expression of constructs containing regions of the hsp-16 gene promoters fused to -galactosidase by both transient assay and stable transformation . When animals containing the intergenic region of the hsp16-48/1 locus are heat shocked and stained for -galactosidase activity, primarily embryos and neuronal cells are stained, with secondary staining occurring in the intestinal cells of some animals. Strains of the hsp16-41/2 locus stain very intensely in embryos and intestinal nuclei with some secondary staining in head and tail nuclei and occasionally in neurons of the ventral cord (Dennis Dixon has obtained similar results with this construct). Thus, while both loci seem capable of expression in the same tissues the priority of expression for each locus differs. No staining was detectable in any of these strains under nonshock conditions. Transient assays of a construct containing the -gal gene inserted into the first exon of the hsp16-1 gene results in extremely rapid and intense staining in embryos and worms. Some nematodes stain only in head and tail neurons while others appear to be saturated with stain in all tissues, except for the gonad. We are currently attempting to obtain some stable lines with this construct though initial results are disappointing. So far, 70 worms have been injected but no stable transformants have been obtained. The frequency of transmission of the unc-22 and staining phenotypes ranges from 5-95%. No strains to date have shown Mendelian inheritance. Southern analysis suggests the injected DNA forms relatively large arrays. Also, all strains for a particular construct show the same pattern of expression suggesting there are no positional effects. For these reasons, we believe the injected DNA probably forms extrachromosomal arrays which are heritable. This is in agreement with Andy Fire's results using unc-22 selection (personal communication ). [See Figure 1]