Worm Breeder's Gazette 12(4): 43 (October 1, 1992)
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.
We had previously reported (Aamodt et al., Science 252, 579-582, 1991) that a number of 5'-deletions of the normally gut-specific ges-1 gene, when transformed into the ges-1 (0)strain, caused esterase expression, not in the gut, but in the posterior part of the pharynx (along with a cell in the tail that we tentatively identified as mu Int R). These patterns were seen with both "transient" transformation and with heritable transformation (the latter, using either the unc-22 antisense or the rol-6 phenotypic markers). Our control against the possibility that it was vector sequences (pBluescript) that were actually directing this ectopic expression was to gel-purify insert from the appropriate ges-1 construct and to use transient transformation (in which marker plasmid could be omitted) to show that the purified insert still caused ges-1 expression in the pharynx. However, the spectre of cryptic enhancers or the like lurking in the vector is always a worry, especially since it has been reported that vector transcripts are present in strains carrying transforming arrays (Fire et. al, Development 113, 503-514, 1991) and since it has been found that a considerable fraction of C. elegans genomic DNA fragments are capable of causing b-galactosidase expression in various cells of the pharynx (Hope, Development 113, 399-408, 1991) . We have thus reconstructed one of the best pharynx-expressing constructs (with 521 base pairs upstream of the ATG) to allow the insert to be excised conveniently, have purified the insert over two gels, have purified the rol-6 insert from pRF4 also through two gels, and have made heritable strains with the combined inserts. The transformants still express in the pharynx/tail pattern and there seems little effect of whether vector is present or not. PCR on individual transformants still shows traces of vector present but, considering the sensitivity of PCR, we're not too worried; indeed, we sporadically detect the presence of vector even in untransformed strains. With "quantitative PCR", i.e. looking at intermediate cycles, we estimate that the vector sequences have been depleted at least 100-fold, compared to a strain transformed with intact plasmid. In other words, we see no evidence for any preferential amplification of vector sequences, the major source of concern. We conclude that, at least for this particular deletion, vector sequences are not the cause of the ectopic expression patterns. As far as narrowing down important sequences, we are now down to an internal deletion of 200 base pairs (1.1 to 1.3 kb upstream of the ATG, in a construct that contains a total of 3.3 kb of 5'-flanking sequence) that both abolishes gut-specific ges-1 expression and activates ges-1 expression in the pharynx/tail pattern.
We have also been doing inter-specific transformations between C. elegans and C. briggsae (whose ges-1 gene we have also cloned and sequenced). To a first approximation, each gene works in the other worm. In particular, the C. elegans pharynx-expressing construct is also expressed in the C. briggsae pharynx. Similarly, a deletion in the C. briggsae ges-1 gene leads to pharynx expression both in C. briggsae and in C. elegans. Thus, the molecular logic will probably turn out to be similar. The strange feature is that there are few obvious conserved sequences held in common in the two genes; overall similarity is close to that expected from random sequences; a few short conserved regions can be detected, but even these are in different positions in the two genes. This situation should help in narrowing down important sequences.
A final note is that we have managed to obtain ,b-galactosidase expression from a ges-1 fusion from which the ges-1 hydrophobic leader sequence has been removed. LacZ expression is indeed in the gut (with traces elsewhere) and we are now installing additional 5' and 3' sequences to achieve "perfect" gut expression before using the constructs in genetic screens for deranged ges-1 expression.
Fire et. al, Development 113, 503-514, 1991.
Hope, Development 113, 399-408, 1991.