Bt Toxin Resistance from Loss of a Putative Carbohydrate-Modifying Enzyme Joel S. Griffitts, Johanna L. Whitacre, Daniel E. Stevens, and Raffi V. Aroian

To obtain the 3' end of the bre-5 transcript, total C. elegans RNA was reverse transcribed with oligo-dT and PCR amplified using oligo-dT and a primer predicted to be in the coding sequence, aatattcagatgcgagagtcagta. The 800 ntd. fragment was subcloned in pBluescript and sequenced. To obtain the 5' end, total C. elegans RNA was reverse transcribed with a bre-5 specific primer, tailed with dATP using terminal deoxynucleotide transferase, and PCR amplified using oligo-dT and a nested bre-5-specific primer. The resulting 800 base fragment was subcloned in pBluescript and sequenced. A complete cDNA was ligated and assembled using the 3' and 5' cDNA clones. The sequence of this cDNA is identical to the C. elegans genome consortium's genomic sequence for T12G3 in the region except for a single A-to-G alteration in the third base of a codon that would not alter the protein sequence. We believe this cDNA is likely to be complete since it contains a stop codon at the 3' end, an in-frame methionine close to the 5' end, and positions the transmembrane domain in about the same place as with other galactosyltransferases. We have also detected a shorter bre-5 transcript that may be less abundant and is trans-spliced (starting 27 nucleotides downstream of the predicted methionine) to the conserved SL1 leader. The closest in-frame methionine start codon would be at amino acid 92 (MRESVR...). Such a protein would lack the putative transmembrane domain; we do not know if such an alternative protein is made or functional.