Why is motility so common in bacteria? An obvious answer to this ecological and evolutionary question is that in almost all habitats, bacteria need to go someplace and particularly in the direction of food. Although the machinery required for motility and chemotaxis (acquiring and processing the information needed to direct movement toward nutrients) are functionally coupled in contemporary bacteria, they are coded for by different sets of genes. Moreover, information that resources are more abundant elsewhere in a habitat would be of no value to a bacterium unless it already had the means to get there. Thus, motility must have evolved before chemotaxis, and bacteria with flagella and other machinery for propulsion in random directions must have an advantage over bacteria relegated to moving at the whim of external forces alone. However, what are the selection pressures responsible for the evolution and maintenance of undirected motility in bacteria? Here we use a combination of mathematical modeling and experiments with Escherichia coli to generate and test a parsimonious and ecologically general hypothesis for the existence of undirected motility in bacteria: it enables bacteria to move away from each other and thereby obtain greater individual shares of resources in physically structured environments. The results of our experiments not only support this hypothesis, but are quantitatively and qualitatively consistent with the predictions of our model.

A simple epidemiological model is used as a framework to explore the potential efficacy of measures to control antibiotic resistance in community-based self-limiting human infections. The analysis of the properties of this model predict that resistance can be maintained at manageable levels if: first, the rates at which specific antibiotics are used declines with the frequency of resistance to these drugs; second, resistance rarely emerges during therapy; and third, external sources rarely contribute to the entry of resistant bacteria into the community. We discuss the feasibility and limitations of these measures to control the rates of antibiotic resistance and the potential of advances in diagnostic procedures to facilitate this endeavor.

We postulate that the inhibition of growth and low rates of mortality of bacteria exposed to ribosome-binding antibiotics deemed bacteriostatic can be attributed almost uniquely to these drugs reducing the number of ribosomes contributing to protein synthesis, i.e., the number of effective ribosomes. We tested this hypothesis with Escherichia coli K-12 MG1655 and constructs that had been deleted for 1 to 6 of the 7 rRNA (rrn) operons. In the absence of antibiotics, constructs with fewer rrn operons have lower maximum growth rates and longer lag phases than those with more ribosomal operons. In the presence of the ribosome-binding “bacteriostatic” antibiotics tetracycline, chloramphenicol, and azithromycin, E. coli strains with 1 and 2 rrn operons are killed at a substantially higher rate than those with more rrn operons. This increase in the susceptibility of E. coli with fewer rrn operons to killing by ribosome-targeting bacteriostatic antibiotics is not reflected in their greater sensitivity to killing by the bactericidal antibiotic ciprofloxacin, which does not target ribosomes, but also to killing by gentamicin, which does. Finally, when such strains are exposed to these ribosome-targeting bacteriostatic antibiotics, the time before these bacteria start to grow again when the drugs are removed, referred to as the post-antibiotic effect (PAE), is markedly greater for constructs with fewer rrn operons than for those with more rrn operons. We interpret the results of these other experiments reported here as support for the hypothesis that the reduction in the effective number of ribosomes due to binding to these structures provides a sufficient explanation for the action of bacteriostatic antibiotics that target these structures.

In contrast to planktonic cells, bacteria imbedded biofilms are notoriously refractory to treatment by antibiotics or bacteriophage (phage) used alone. Given that the mechanisms of killing differ profoundly between drugs and phages, an obvious question is whether killing is improved by combining antibiotic and phage therapy. However, this question has only recently begun to be explored. Here, in vitro biofilm populations of Pseudomonas aeruginosa PA14 were treated singly and with combinations of two phages and bactericidal antibiotics of five classes. By themselves, phages and drugs commonly had only modest effects in killing the bacteria. However some phage-drug combinations reduced bacterial densities to well below that of the best single treatment; in some cases, bacterial densities were reduced even below the level expected if both agents killed independently of each other (synergy). Furthermore, there was a profound order effect in some cases: treatment with phages before drugs achieved maximum killing. Combined treatment was particularly effective in killing in Pseudomonas biofilms grown on layers of cultured epithelial cells. Phages were also capable of limiting the extent to which minority populations of bacteria resistant to the treating antibiotic ascend. The potential of combined antibiotic and phage treatment of biofilm infections is discussed as a realistic way to evaluate and establish the use of bacteriophage for the treatment of humans.

Streptococcus pneumoniae is a main cause of child mortality worldwide, but strains also asymptomatically colonize the upper airways of most children and form biofilms. Recent studies have demonstrated that ~50% of colonized children carry at least two different serotypes (i.e., strains) in the nasopharynx; however, studies of how strains coexist are limited. In this work, we investigated the physiological, genetic, and ecological requirements for the relative distribution of densities, and spatial localization, of pneumococcal strains within biofilm consortia. Biofilm consortia were prepared with vaccine type strains (i.e., serotype 6B [S6B], S19F , or S23F) and strain TIGR4 (S4). Experiments first revealed that the relative densities of S6B and S23F were similar in biofilm consortia. The density of S19F strains, however, was reduced to ~10% in biofilm consortia, including either S6B, S23F, or TIGR4, in comparison to S19F monostrain biofilms. Reduction of S19F density within biofilm consortia was also observed in a simulated nasopharyngeal environment. Reduction of relative density was not related to growth rates, since the Malthusian parameter demonstrated similar rates of change of density for most strains. To investigate whether quorum sensing (QS) regulates relative densities in biofilm consortia, two different mutants were prepared: a TIGR4ΔluxS mutant and a TIGR4ΔcomC mutant. The density of S19F strains, however, was similarly reduced when consortia included TIGR4, TIGR4ΔluxS, or TIGR4ΔcomC. Moreover, production of a different competencestimulating peptide (CSP), CSP1 or CSP2, was not a factor that affected dominance. Finally, a mathematical model, confocal experiments, and experiments using Transwell devices demonstrated physical contact-mediated control of pneumococcal density within biofilm consortia.

The rate at which mutations are generated is central to the pace of evolution. Although this rate is remarkably similar amongst all cellular organisms, bacterial strains with mutation rates 100 fold greater than the modal rates of their species are commonly isolated from natural sources and emerge in experimental populations. Theoretical studies postulate and empirical studies teort the hypotheses that these “mutator” strains evolved in response to selection for elevated rates of generation of inherited variation that enable bacteria to adapt to novel and/or rapidly changing environments. Less clear are the conditions under which selection will favor reductions in mutation rates. Declines in rates of mutation for established populations of mutator bacteria are not anticipated if such changes are attributed to the costs of augmented rates of generation of deleterious mutations. Here we report experimental evidence of evolution towards reduced mutation rates in a clinical isolate of Escherichia coli with an hyper-mutable phenotype due a deletion in a mismatch repair gene, (ΔmutS). The emergence in a ΔmutS background of variants with mutation rates approaching those of the normal rates of strains carrying wild-type MutS was associated with increase in fitness with respect to ancestral strain. We postulate that such an increase in fitness could be attributed to the emergence of mechanisms driving a permanent “aerobic style of life”, the negative consequence of this behavior being regulated by the evolution of mechanisms protecting the cell against increased endogenous oxidative radicals involved in DNA damage, and thus reducing mutation rate. Gene expression assays and full sequencing of evolved mutator and normo-mutable variants supports the hypothesis. In conclusion, we postulate that the observed reductions in mutation rate are coincidental to, rather than, the selective force responsible for this evolution.

There are both pharmacodynamic and evolutionary reasons to use multiple rather than single antibiotics to treat bacterial infections; in combination antibiotics can be more effective in killing target bacteria as well as in preventing the emergence of resistance. Nevertheless, with few exceptions like tuberculosis, combination therapy is rarely used for bacterial infections. One reason for this is a relative dearth of the pharmaco-, population- and evolutionary dynamic information needed for the rational design of multi-drug treatment protocols. Here, we use in vitro pharmacodynamic experiments, mathematical models and computer simulations to explore the relative efficacies of different two-drug regimens in clearing bacterial infections and the conditions under which multi-drug therapy will prevent the ascent of resistance. We estimate the parameters and explore the fit of Hill functions to compare the pharmacodynamics of antibiotics of four different classes individually and in pairs during cidal experiments with pathogenic strains of Staphylococcus aureus and Escherichia coli. We also consider the relative efficacy of these antibiotics and antibiotic pairs in reducing the level of phenotypically resistant but genetically susceptible, persister, subpopulations. Our results provide compelling support for the proposition that the nature and form of the interactions between drugs of different classes, synergy, antagonism, suppression and additivity, has to be determined empirically and cannot be inferred from what is known about the pharmacodynamics or mode of action of these drugs individually. Monte Carlo simulations of within-host treatment incorporating these pharmacodynamic results and clinically relevant refuge subpopulations of bacteria indicate that: (i) the form of drug-drug interactions can profoundly affect the rate at which infections are cleared, (ii) two-drug therapy can prevent treatment failure even when bacteria resistant to single drugs are present at the onset of therapy, and (iii) this evolutionary virtue of two-drug therapy is manifest even when the antibiotics suppress each other's activity.

Multi-drug therapy is the standard-of-care treatment for tuberculosis. Despite this, virtually all studies of the pharmacodynamics (PD) of mycobacterial drugs employed for the design of treatment protocols are restricted to single agents. In this report, mathematical models and in vitro experiments with Mycobacterium marinum and five antimycobacterial drugs are used to quantitatively evaluate the pharmaco-, population and evolutionary dynamics of two-drug antimicrobial chemotherapy regimes. Time kill experiments with single and pairs of antibiotics are used to estimate the parameters and evaluate the fit of Hill-function-based PD models. While Hill functions provide excellent fits for the PD of each single antibiotic studied, rifampin, amikacin, clarithromycin, streptomycin and moxifloxacin, two-drug Hill functions with a unique interaction parameter cannot account for the PD of any of the 10 pairs of these drugs. If we assume two antibiotic-concentration dependent functions for the interaction parameter, one for sub-MIC and one for supra-MIC drug concentrations, the modified biphasic Hill function provides a reasonably good fit for the PD of all 10 pairs of antibiotics studied. Monte Carlo simulations of antibiotic treatment based on the experimentally-determined PD functions are used to evaluate the potential microbiological efficacy (rate of clearance) and evolutionary consequences (likelihood of generating multi-drug resistance) of these different drug combinations as well as their sensitivity to different forms of non-adherence to therapy. These two-drug treatment simulations predict varying outcomes for the different pairs of antibiotics with respect to the aforementioned measures of efficacy. In summary, Hill functions with biphasic drug-drug interaction terms provide accurate analogs for the PD of pairs of antibiotics and M. marinum. The models, experimental protocols and computer simulations used in this study can be applied to evaluate the potential microbiological and evolutionary efficacy of two-drug therapy for any bactericidal antibiotics and bacteria that can be cultured in vitro.

Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR), together with associated genes (cas), form the CRISPR–cas adaptive immune system, which can provide resistance to viruses and plasmids in bacteria and archaea. Here, we use mathematical models, population dynamic experiments, and DNA sequence analyses to investigate the host–phage interactions in a model CRISPR–cas system, Streptococcus thermophilus DGCC7710 and its virulent phage 2972. At the molecular level, the bacteriophage-immune mutant bacteria (BIMs) and CRISPR–escape mutant phage (CEMs) obtained in this study are consistent with those anticipated from an iterative model of this adaptive immune system: resistance by the addition of novel spacers and phage evasion of resistance by mutation in matching sequences or flanking motifs. While CRISPR BIMs were readily isolated and CEMs generated at high rates (frequencies in excess of 10−6), our population studies indicate that there is more to the dynamics of phage–host interactions and the establishment of a BIM–CEM arms race than predicted from existing assumptions about phage infection and CRISPR–cas immunity. Among the unanticipated observations are: (i) the invasion of phage into populations of BIMs resistant by the acquisition of one (but not two) spacers, (ii) the survival of sensitive bacteria despite the presence of high densities of phage, and (iii) the maintenance of phage-limited communities due to the failure of even two-spacer BIMs to become established in populations with wild-type bacteria and phage. We attribute (i) to incomplete resistance of single-spacer BIMs. Based on the results of additional modeling and experiments, we postulate that (ii) and (iii) can be attributed to the phage infection-associated production of enzymes or other compounds that induce phenotypic phage resistance in sensitive bacteria and kill resistant BIMs. We present evidence in support of these hypotheses and discuss the implications of these results for the ecology and (co)evolution of bacteria and phage.

When growing populations of bacteria are confronted with bactericidal antibiotics, the vast majority of cells are killed, but subpopulations of genetically susceptible but phenotypically resistant bacteria survive. In accord with the prevailing view, these “persisters” are non- or slowly dividing cells randomly generated from the dominant population. Antibiotics enrich populations for pre-existing persisters but play no role in their generation. The results of recent studies with Escherichia coli suggest that at least one antibiotic, ciprofloxacin, can contribute to the generation of persisters. To more generally elucidate the role of antibiotics in the generation of and selection for persisters and the nature of persistence in general, we use mathematical models and experiments with Staphylococcus aureus (Newman) and the antibiotics ciprofloxacin, gentamicin, vancomycin, and oxacillin. Our results indicate that the level of persistence varies among these drugs and their concentrations, and there is considerable variation in this level among independent cultures and mixtures of independent cultures. A model that assumes that the rate of production of persisters is low and persisters grow slowly in the presence of antibiotics can account for these observations. As predicted by this model, pre-treatment with sub-MIC concentrations of antibiotics substantially increases the level of persistence to drugs other than those with which the population is pre-treated. Collectively, the results of this jointly theoretical and experimental study along with other observations support the hypothesis that persistence is the product of many different kinds of errors in cell replication that result in transient periods of non-replication and/or slowed metabolism by individual cells in growing populations. This Persistence as Stuff Happens (PaSH) hypothesis can account for the ubiquity of this phenomenon. Like mutation, persistence is inevitable rather than an evolved character. What evolved and have been identified are genes and processes that affect the frequency of persisters.