Abstract
The web application PrimerPair at ecogene.org generates large sets of paired DNA sequences surrounding all protein and RNA genes of Escherichia coli K-12. Many DNA fragments, which these primers amplify, can be used to implement a genome reengineering strategy using complementary in vitro cloning and in vivo recombineering. The integration of a primer design tool with a model organism database increases the level of quality control. Computer-assisted design of gene primer pairs relies upon having highly accurate genomic DNA sequence information that exactly matches the DNA of the cells being used in the laboratory to ensure predictable DNA hybridizations. It is equally crucial to have confidence that the predicted start codons define the locations of genes accurately. Annotations in the EcoGene database are queried by PrimerPair to eliminate pseudogenes, IS elements, and other problematic genes before the design process starts. These projects progressively familiarize users with the EcoGene content, scope, and application interfaces that are useful for genome reengineering projects. The first protocol leads to the design of a pair of primer sequences that were used to clone and express a single gene. The N-terminal protein sequence was experimentally verified and the protein was detected in the periplasm. This is followed by instructions to design PCR primer pairs for cloning gene fragments encoding 50 periplasmic proteins without their signal peptides. The design process begins with the user simply designating one pair of forward and reverse primer endpoint positions relative to all start and stop codon positions. The gene name, genomic coordinates, and primer DNA sequences are reported to the user. When making chromosomal deletions, the integrity of the provisional primer design is checked to see whether it will generate any unwanted double deletions with adjacent genes. The bad designs are recalculated and replacement primers are provided alongside the requested primers. A list of all genes with overlaps includes those expressed from the translational coupling motifs 5′-UGAUG-3′ and 5′-AUGA-3′. Rigid alignments of the 893 ribosome binding sites (RBSs) linked to the AUG codons of this coupled subset are assessed for information content using WebLogo 3.0. These specialized logos are missing the G at the prominent information peak position normally seen in the rigid alignment of all genes. This novel GHOLE motif was apparently masked by the normal RBSs in two previously published rigid alignments. We propose a model constraining the distance between the ATG and the RBS, obviating the need for a flexible linker model to reveal a Shine–Dalgarno-like sequence.
