Worm Breeder's Gazette 11(2): 69
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.
Based on Northern analysis with 3'-end specific probes we have determined that C. elegans must possess at least two classes of ubiquitin genes. We have cloned the major ubiquitin gene, UbiA, which encodes worm polyubiquitin, and have undertaken various schemes for the isolation of the non-UbiA member(s) of the ubiquitin family. Unfortunately, exhaustive screening of available C. elegans cDNA and genomic libraries and attempts to directly clone size-fractionated genomic DNA have led only to the reisolation of UbiA clones ad nauseum. We then decided to devise a Polymerase Chain Reaction strategy for ubiquitin cloning, hopefully one which might provide a universal tool. PCR relies on the presence of unique oligonucleotide sequences in the gene of interest, or at least the presence of oligonucleotides which will define members of a set of related genes. Conceptually this is difficult for ubiquitin: ubiquitin is not only extremely highly conserved in DNA sequence but it tends to be found as part of a tandemly repeated gene structure (e.g. UbiA contains 11 tandem repeats of the ubiquitin sequence) -thus two oligonucleotide primers which are specific for ubiquitin could prime in any pairwise combination of repeat units, ultimately generating a PCR product which is the lowest common denominator, i.e. a mixed sequence representing all of the ubiquitin repeats. To overcome this problem we have used only one ubiquitin-specific degenerate oligonucleotide in combination with anchor-oligodT primed cDNA as illustrated: [See Figure 1] The use of only one ubiquitin oligonucleotide primer gives polarity to the cDNA amplification and hence the final product represents an 'accurate' copy of the final ubiquitin repeat and the 3'-UT of the represented locus, rather than a mixed sequence representing the coding sequence of all repeats at any given locus. The 3'-UT region may then be used as a probe to isolate the particular locus or may provide a locus-specific primer(s) for amplification of that region. It should be noted that the same strategy can be used to obtain the 5'- end 'non-ubiquitin' sequence of a ubiquitin locus if the first strand cDNA is homopolymer-tailed and a conserved ubiquitin oligonucleotide is then used in the opposite orientation. Using degenerate oligonucleotides corresponding to the conserved ubiquitin amino acid sequence we have been able to re-isolate the predicted truncated C. elegans UbiA cDNA clones as well as clones for a highly conserved UbiA homologue of C. briggsae. Refreshingly, we were also able to obtain a class of nonUbiA cDNAs from C. elegans. These clones were similar to the yeast Ubi2 gene in that they contained an amino terminal ubiquitin sequence (i.e. at least one) fused to a basic C-terminal extension region of 52 amino acids. This locus could account for the small ubiquitin mRNA (approximately 700 nt) seen on our Northern blots. This approach proved so successful with worms that we attempted to use it to isolate truncated ubiquitin cDNAs from close relatives of C. elegans -i.e. members of the plant kingdom which surround our building! At last check, these oligonucleotides have allowed us to amplify truncated ubiquitin cDNAs (with and/or without basic C- terminal extensions) from rhododendrons, Rhytidia, palms, mosses, Douglas Fir and Ginko trees.