In this study, we show a correlation between synthesis of aberrant proteins and their oxidative modification. The level of aberrant proteins was elevated in Escherichia coli cultures by decreasing transcriptional or translational fidelity using specific mutations or drugs. Protein carbonylation, an oxidative modification, increased in parallel to the induction of the heat shock chaperone GroEL. As the protein turnover rates and level of intracellular oxidative stress remained unchanged, it appears that carbonylation results from the increased susceptibility of the misfolded proteins. These studies show that the cellular protein oxidation is not limited only by available reactive oxygen species, but by the levels of aberrant proteins. Thus, protein oxidation seen in aging cells may be the consequence also of reduced transcriptional/translational fidelity, and protein structures appear to have evolved to minimize oxidative damage. In addition, we discuss the possibility that carbonylation, being an unrepairable protein modification, may serve as a tagging system to shunt misfolded proteins between pathways of refolding by chaperones or the proteolytic apparatus.

A mutation in the Escherichia coli gene encoding the stationary phase‐inducible sigma factor (σ^s, RpoS) not only abolishes transcription of some genes in stationary phase, but also causes superinduction of other stationary phase‐induced genes. We have examined this phenomenon of repression by σ^s using as a model system the divergently transcribed stationary phase‐inducible genes, uspA and uspB. uspA is transcribed by σ⁷⁰‐programmed RNA polymerase and is superinduced in an rpoS mutant, while uspB induction is σ^s dependent. The data suggest that the superinduction of uspA is caused by an increased amount of σ⁷⁰ bound to RNA polymerase in the absence of the competing σ^s. Increasing the ability of σ⁷⁰ to compete against σ^s by overproducing σ⁷⁰ mimics the effect of an rpoS mutation by causing superinduction of σ⁷⁰‐dependent stationary phase‐inducible genes (uspA and fadD), silencing of σ^s‐dependent genes (uspB, bolAp1 and fadL) and inhibiting the development of σ^s‐dependent phenotypes, such as hydrogen peroxide resistance in stationary phase. In addition, overproduction of σ^s markedly reduced stationary phase expression of a σ⁷⁰‐dependent promoter. Thus, we conclude that sigma factors compete for a limiting amount of RNA polymerase during stationary phase. The implications of this competition in the passive control of promoter activity is discussed.

The emergence of microbial antibiotic resistance is a global health threat. In clinical settings, the key to controlling spread of resistant strains is accurate and rapid detection. As traditional culture-based methods are time consuming, genetic approaches have recently been developed for this task. The detection of antibiotic resistance is typically made by measuring a few known determinants previously identified from genome sequencing, and thus requires the prior knowledge of its biological mechanisms. To overcome this limitation, we employed machine learning models to predict resistance to 11 compounds across four classes of antibiotics from existing and novel whole genome sequences of 1936 E. coli strains. We considered a range of methods, and examined population structure, isolation year, gene content, and polymorphism information as predictors. Gradient boosted decision trees consistently outperformed alternative models with an average accuracy of 0.91 on held-out data (range 0.81–0.97). While the best models most frequently employed gene content, an average accuracy score of 0.79 could be obtained using population structure information alone. Single nucleotide variation data were less useful, and significantly improved prediction only for two antibiotics, including ciprofloxacin. These results demonstrate that antibiotic resistance in E. coli can be accurately predicted from whole genome sequences without a priori knowledge of mechanisms, and that both genomic and epidemiological data can be informative. This paves way to integrating machine learning approaches into diagnostic tools in the clinic. One of the major health threats of 21st century is emergence of antibiotic resistance. To manage its human health and economic impact, efforts are made to develop novel diagnostic tools that rapidly detect resistant strains in clinical settings. In our study, we employed a range of powerful machine learning tools to predict antibiotic resistance from whole genome sequencing data for E. coli. We used the presence or absence of genes, population structure and isolation year of isolates as predictors, and could attain average precision of 0.92 and recall of 0.83, without prior knowledge about the causal mechanisms. These results demonstrate the potential application of machine learning methods as a diagnostic tool in healthcare settings.

The response of exponentially growing cultures of Escherichia coli to abrupt shifts in hydrostatic pressure was studied. A pressure upshift to 546 atm (55,304 kPa) of hydrostatic pressure profoundly perturbed cell division, nucleoid structure, and the total rate of protein synthesis. The number of polypeptides synthesized at increased pressure was greatly reduced, and many proteins exhibited elevated rates of synthesis relative to total protein synthesis. We designated the latter proteins pressure-induced proteins (PIPs). The PIP response was transient, with the largest induction occurring approximately 60 to 90 min postshift. Fifty-five PIPs were identified. Many of these proteins are also induced by heat shock or cold shock. The PIP demonstrating the greatest pressure induction was a basic protein of 15.6 kDa. High pressure inhibits growth but does not inhibit the synthesis of stringently controlled proteins. Cold shock is the only additional signal which has been found to elicit this type of response. These data indicate that elevated pressure induces a unique stress response in E. coli, the further characterization of which could be useful in delineating its inhibitory nature.

The recent discovery that the protein DksA acts as a coregulator of genes controlled by ppGpp led us to investigate the similarities and differences between the relaxed phenotype of a ppGpp-deficient mutant and the phenotype of a strain lacking DksA. We demonstrate that the absence of DksA and ppGpp has similar effects on many of the observed phenotypes but that DksA and ppGpp also have independent and sometimes opposing roles in the cell. Specifically, we show that overexpression of DksA can compensate for the loss of ppGpp with respect to transcription of the promoters P _uspA , P _livJ , and P _rrnBP1 as well as amino acid auxotrophy, cell-cell aggregation, motility, filamentation, and stationary phase morphology, suggesting that DksA can function without ppGpp in regulating gene expression. In addition, ppGpp and DksA have opposing effects on adhesion. In the course of our analysis, we also discovered new features of the relaxed mutant, namely, defects in cell-cell aggregation and motility.