David Rockabrand

Polypeptide flux through bacterial Hsp70: DnaK cooperates with trigger factor in chaperoning nascent chains.

A role for DnaK, the major E. coli Hsp70, in chaperoning de novo protein folding has remained elusive. Here we show that under nonstress conditions DnaK transiently associates with a wide variety of nascent and newly synthesized polypeptides, with a preference for chains larger than 30 kDa. Deletion of the nonessential gene encoding trigger factor, a ribosome-associated chaperone, results in a doubling of the fraction of nascent polypeptides interacting with DnaK. Combined deletion of the trigger factor and DnaK genes is lethal under normal growth conditions. These findings indicate important, partially overlapping functions of DnaK and trigger factor in de novo protein folding and explain why the loss of either chaperone can be tolerated by E. coli.

Heterogeneity of a Campylobacter jejuni Protein That Is Secreted through the Flagellar Filament

ABSTRACT Cj0859c, or FspA, is a small, acidic protein of Campylobacter jejuni that is expressed by a σ28 promoter. Analysis of the fspA gene in 41 isolates of C. jejuni revealed two overall variants of the predicted protein, FspA1 and FspA2. Secretion of FspA occurs in broth-grown bacteria and requires a minimum flagellar structure. The addition of recombinant FspA2, but not FspA1, to INT407 cells in vitro resulted in a rapid induction of apoptosis. These data define a novel C. jejuni virulence factor, and the observed heterogeneity among fspA alleles suggests alternate virulence potential among different strains.

Genetic Analysis of Lipooligosaccharide Core Biosynthesis in Campylobacter jejuni 81-176

We report isolation and characterization of Campylobacter jejuni 81-176 lgtF and galT lipooligosaccharide (LOS) core mutants. It has been suggested that the lgtF gene of C. jejuni encodes a two-domain glucosyltransferase that is responsible for the transfer of a beta-1,4-glucose residue on heptosyltransferase I (Hep I) and for the transfer of a beta-1,2-glucose residue on Hep II. A site-specific mutation in the lgtF gene of C. jejuni 81-176 resulted in expression of a truncated LOS, and complementation of the mutant in trans restored the core mobility to that of the wild type. Mass spectrometry and nuclear magnetic resonance of the truncated LOS confirmed the loss of two glucose residues, a beta-1,4-glucose on Hep I and a beta-1,2-glucose on Hep II. Mutation of another gene, galT, encoding a glycosyltransferase, which maps outside the region defined as the LOS biosynthetic locus in C. jejuni 81-176, resulted in loss of the beta-(1,4)-galactose residue and all distal residues in the core. Both mutants invaded intestinal epithelial cells in vitro at levels comparable to the wild-type levels, in marked contrast to a deeper inner core waaC mutant. These studies have important implications for the role of LOS in the pathogenesis of Campylobacter-mediated infection.

Multicopy plasmid suppression of stationary phase chaperone toxicity in Escherichia coli by phosphogluconate dehydratase and the N-terminus of DnaK

Overproduction of DnaK in Escherichia coli results in a bacteriocidal effect. This effect is most acute in stationary phase cells. A selection scheme was developed to isolate multicopy suppressors from an E. coli plasmid expression library, which overcome the stationary phase toxicity of excess DnaK. Two suppressor plasmids were recovered which contained inserts of 1.85 kb and 2.69 kb, respectively. Rearranged and deleted plasmid derivatives were constructed and used to further localize the suppressors. DNA sequence analysis demonstrated that one suppressor encoded phosphogluconate dehydratase (Edd) while the other suppressor encoded the N-terminal 237 amino acids of DnaK itself (DnaK′). Strains bearing the suppressor plasmids constitutively overproduced proteins with apparent masses of 66 kDa (Edd) and 37 kDa (DnaK′) as determined by gel electrophoresis. Western blot analysis using polyclonal antisera specific for either Edd or DnaK confirmed the identity of these overproduced proteins. Suppression of DnaK toxicity was eliminated by the introduction of a + 1 frameshift mutation early in the respective coding regions of either of the two suppressors. These results suggest that suppressor gene translation plays a role in the mechanism of DnaK suppression.