With the aim of optimizing the cloning of novel genes from a genomic pool containing many previously identified homologous genes, we designed a redundant exclusion PCR (RE-PCR) technique. In RE-PCR, a pair of generic amplification primers are combined with additional primers that are designed to specifically bind to redundant, unwanted genes that are a subset of those copied by the amplification primers. During RE-PCR, the specific primer blocks amplification of the full-length redundant gene. Using this method, we managed to clone a number of cry8 or cry9 toxin genes from a pool of Bacillus thuringiensis genomic DNA while excluding amplicons for cry9Da, cry9Ea, and cry9Eb. The method proved to be very efficient at increasing the number of rare genes in the resulting library. One such rare (and novel) cry8-like gene was expressed, and the encoded toxin was shown to be toxic to Anomala corpulenta.
IMPORTANCEProtein toxins from the bacterium Bacillus thuringiensis are being increasingly used as biopesticides against a wide range of insect pests, yet the search for new or improved toxins is becoming more difficult, as traditional methods for gene discovery routinely isolate previously identified clones. This paper describes an approach that we have developed to increase the success rate for novel toxin gene identification through reducing or eliminating the cloning of previously characterized genes.A s a result of the proteinaceous insecticidal toxins produced by Bacillus thuringiensis (Bt), this bacterium has become a commercially successful biopesticide (1). Products based on Bt include formulations of the bacterium itself or the toxin expressed in an alternative host, in particular genetically modified crops (2). Despite the increasing use of these products, there remains a need to discover new toxins with desirable properties; such properties include an increased activity against a given target, activity against a new target pest, or the ability to control a pest that has developed resistance to an existing toxin. A number of different approaches can be used to identify novel toxins, with the traditional one being to screen strains for a desired activity and then isolate the active ingredient. In recent times molecular approaches have been increasingly used, including genome sequencing (3) and PCR techniques. The latter rely on there being conserved regions present in toxin gene families as well as the more variable regions that give toxins their individual characteristics (4). Improved PCR procedures have allowed the successful cloning of Bt toxin genes from complex DNA mixtures prepared from pooled samples (5, 6). A problem with this sort of approach, however, is the high ratio of known or undesired toxin genes in libraries made from these pooled samples, which has made the discovery of new genes increasingly difficult.The B. thuringiensis Cry8 and Cry9 proteins have significantly different insecticidal spectra despite phylogenetic analyses indicating that they share high sequence similarity in domain...