[
Worm Breeder's Gazette,
1990]
One is (alas) frequently interested in cloning genes that lie in regions not yet included in the contig map, or in regions spanned by YAC DNA but not by cosmid DNA. Since cloning strategies often involve rescue of mutant phenotypes by injection of cosmid DNAs, both the above situations provide the necessity of chromosomal walking. There are several cosmid vectors designed for chromosomal walking including the lambda-origin vectors such as Lorist6 [Gibson et al., Gene 53:283-286 (1987)] and the ColE1 replicon-based cosmid pWE [Wahl, et al., PNAS 84:2160-2164 (1987)]. These vectors have bacteriophage promoters flanking the cloning site. Chromosomal walking using these vectors consists of making labeled RNA probes from each end of the starting cosmid, determining which probe goes in the desired direction, and using that probe to take the next step. This procedure is facile, but the production of RNA probes can be unreliable. The presence of NotI sites flanking the cloning site in a cosmid vector provides an inexpensive, low-tech way to identify the terminal restriction fragments of a cosmid insert using restriction digests. The terminal fragments, once identified, can be gel isolated and labeled by standard methods. The cosmid vector pWE has NotI sites immediately upstream from the bacteriophage promoters. However, pWE will cross-hybridize with the YAC vector and is therefore not ideal for all applications. Accordingly, we decided to make a lambda-origin vector with the features of pWE. The construction of this vector was straightforward. In brief, Lorist6 has four EcoRI sites, (see below) two of them in essential genes. To protect these sites, Lorist6 was digested with NcoI, end-filled, then redigested with EcoRV. A 2.7 kb restriction fragment containing the cloning site was subcloned into SmaI-EcoRV digested Bluescript regenerating the NcoI site. A 700 bp EcoRI fragment containing the cloning site was removed and replaced with the cloning region of pWE 15, contained in a 100 bp EcoRI fragment. The NcoI-EcoRV fragment, now 2.1 kb, was removed and religated to the 2.5 kb NcoI-EcoRV fragment of Lorist6. The resulting vector was named WEkan, WE from pWE meaning walking easily, and kan because the vector confers resistance to kanamycin. The construction of WEkan resulted in the removal of transcription terminators present in Lorist6 to protect vector genes from transcriptional interference from insert DNA. We do not know what effect this has on representation in libraries prepared in WEkan. It does not seem to affect viability in a general way. [See Figure 1] We prepared a genomic library in WEkan using as insert N2 DNA partially digested with XhoII and size-selected by reverse field gel electrophoresis. Cosmid arms were prepared by digesting with BstEII/BamHI and XbaI/BamHI. DNA prepared from 18 random clones showed that the library contains 10-20% non-recombinant clones and that insert size ranges from 35 to 50 kb. 11,500 members of this library were gridded into micro-titer dishes. The library has been probed with several probes and has been shown to contain three to four genomes. The clones appear to be stable and to grow well.