We report a 1.432-kb DNA sequence at 59 min on the Escherichia coli chromosome that connects the published sequences of the pcm gene for the isoaspartyl protein methyltransferase and that of the katF or rpoS (katFirpoS) gene We are interested in the widely distributed L-isoaspartyl protein methyltransferase that can modify altered aspartyl residues arising from spontaneous chemical degradation of proteins (3, 6). The methylation reaction can initiate steps leading to the restoration of the normal L-configuration of such residues and may participate as such in a protein repair step (12,22,32). The pcm gene for this enzyme had been cloned, sequenced, and mapped to the 59-min region on the chromosome of Escherichia coli (11). It is about 1 kb upstream of the katF (or rpoS) (katFirpoS) gene whose product shows strong homology to the E. coli rpoD cr70 protein (34) and appears to be a sigma factor in the central regulation of stationary-phase gene expression (28,37,43). It is interesting that pcm mutant cells demonstrate two of the phenotypes of katF/rpoS mutant cells: reduced stationary-phase survival and lessened heatshock resistance (28,29,36). Genes with similar functions can be adjacent on the E. coli chromosome, even if they are not present in one operon. For example, four operons containing genes for chemotaxis and mobility are adjacent at 41.5 min; similarly, three operons containing genes for ribosomal proteins are adjacent at about 72.4 to 73.4 min on the E. coli chromosome (1). We were thus interested in exploring whether a polypeptide or polypeptides with a function potentially similar to that of pcm or katFirpoS might be encoded in the region between these genes.In this work, we have determined and analyzed the DNA

We determined the DNA sequence of a 2,232-bp region immediately upstream

Volume 176, no. 19, p. 6017, Fig. 1: The nucleotide sequence of the region upstream of the pcm gene contains several errors. Corrections include an insertion of a C at base pair position 230, insertion of a G at base pairs 894 and 1075, deletion of a G at base pair 1822, deletion of a C at base pair 1832, and inversion of a GC to CG at positions 2180 and 2181 of the originally published sequence. The corrections have been submitted to the GenBank Database (accession number L07942). The corrected sequence is now in accord with the sequence submitted under GenBank accession number U29579 by Guy P. Plunkett and changes the surE open reading frame described in our paper. The new, putative ATG start codon for the surE gene is located at nucleotides 1826 to 1828, 265 bp downstream from the original start codon. A putative Shine-Delgarno ribosomal binding site (AACAGG) is located 6 nucleotides upstream at nucleotides 1815 to 1820. This shortens the expected protein product to 26,903 Da but maintains the reading frame. A new open reading frame, designated orf1 (nucleotides 796 to 1845), now spans the region between orf0 and surE. The translated Orf1 product has a calculated molecular mass of 39,095 Da, which we suspect is the 37,000-Da product observed in the in vitro transcription/translation experiments shown in Fig. 3

No abstract