Adaptive Processes Change as Multiple Functions Evolve.

Epistasis influences the gene-environment interactions that shape bacterial fitness through antibiotic exposure, which can ultimately affect the availability of certain resistance phenotypes to bacteria. The substitutions present within blaTEM-50 confer both cephalosporin and ß-lactamase inhibitor resistance. We wanted to compare the evolution of blaTEM-50 with that of another variant, blaTEM-85, which differs in that blaTEM-85 contains only substitutions that contribute to cephalosporin resistance. Differences between the landscapes and epistatic interactions of these TEM variants are important for understanding their separate evolutionary responses to antibiotics. We hypothesized the substitutions within blaTEM-50 would result in more epistatic interactions than for blaTEM-85 As expected, we found more epistatic interactions between the substitutions present in blaTEM-50 than in blaTEM-85 Our results suggest that selection from many cephalosporins is required to achieve the full potential resistance to cephalosporins but that a single ß-lactam and inhibitor combination will drive evolution of the full potential resistance phenotype. Surprisingly, we also found significantly positive increases in growth rates as antibiotic concentration increased for some of the strains expressing blaTEM-85 precursor genotypes but not the blaTEM-50 variants. This result further suggests that additive interactions more effectively optimize phenotypes than epistatic interactions, which means that exposure to numerous cephalosporins actually increases the ability of a TEM enzyme to confer resistance to any single cephalosporin.

Statistical Package for Growth Rates Made Easy.

Growth rates are an important tool in microbiology because they provide high throughput fitness measurements. The release of GrowthRates, a program that uses the output of plate reader files to automatically calculate growth rates, has facilitated experimental procedures in many areas. However, many sources of variation within replicate growth rate data exist and can decrease data reliability. We have developed a new statistical package, CompareGrowthRates (CGR), to enhance the program GrowthRates and accurately measure variation in growth rate data sets. We define a metric, Variability-score (V-score), that can help determine if variation within a data set might result in false interpretations. CGR also uses the bootstrap method to determine the fraction of bootstrap replicates in which a strain will grow the fastest. We illustrate the usage of CGR with growth rate data sets similar to those in Mira, Meza, et al. (Adaptive landscapes of resistance genes change as antibiotic concentrations change. Mol Biol Evol. 32(10): 2707-2715). These statistical methods are compatible with the analytic methods described in Growth Rates Made Easy and can be used with any set of growth rate output from GrowthRates.

Does Antibiotic Resistance Evolve in Hospitals?

Nosocomial outbreaks of bacteria are well documented. Based on these incidents, and the heavy usage of antibiotics in hospitals, it has been assumed that antibiotic resistance evolves in hospital environments. To test this assumption, we studied resistance phenotypes of bacteria collected from patient isolates at a community hospital over a 2.5-year period. A graphical model analysis shows no association between resistance and patient information other than time of arrival. This allows us to focus on time-course data. We introduce a hospital transmission model, based on negative binomial delay. Our main contribution is a statistical hypothesis test called the Nosocomial Evolution of Resistance Detector (NERD). It calculates the significance of resistance trends occurring in a hospital. It can inform hospital staff about the effects of various practices and interventions, can help detect clonal outbreaks, and is available as an R package. We applied the NERD method to each of the 16 antibiotics in the study via 16 hypothesis tests. For 13 of the antibiotics, we found that the hospital environment had no significant effect on the evolution of resistance; the hospital is merely a piece of the larger picture. The p-values obtained for the other three antibiotics (cefepime, ceftazidime, and gentamicin) indicate that particular care should be taken in hospital practices with these antibiotics. One of the three, ceftazidime, was significant after accounting for multiple hypotheses, indicating a trend of decreased resistance for this drug.

Adaptive Landscapes of Resistance Genes Change as Antibiotic Concentrations Change.

Most studies on the evolution of antibiotic resistance are focused on selection for resistance at lethal antibiotic concentrations, which has allowed the detection of mutant strains that show strong phenotypic traits. However, solely focusing on lethal concentrations of antibiotics narrowly limits our perspective of antibiotic resistance evolution. New high-resolution competition assays have shown that resistant bacteria are selected at relatively low concentrations of antibiotics. This finding is important because sublethal concentrations of antibiotics are found widely in patients undergoing antibiotic therapies, and in nonmedical conditions such as wastewater treatment plants, and food and water used in agriculture and farming. To understand the impacts of sublethal concentrations on selection, we measured 30 adaptive landscapes for a set of TEM ß-lactamases containing all combinations of the four amino acid substitutions that exist in TEM-50 for 15 ß-lactam antibiotics at multiple concentrations. We found that there are many evolutionary pathways within this collection of landscapes that lead to nearly every TEM-genotype that we studied. While it is known that the pathways change depending on the type of ß-lactam, this study demonstrates that the landscapes including fitness optima also change dramatically as the concentrations of antibiotics change. Based on these results we conclude that the presence of multiple concentrations of ß-lactams in an environment result in many different adaptive landscapes through which pathways to nearly every genotype are available. Ultimately this may increase the diversity of genotypes in microbial populations.

Growth rates made easy.

In the 1960s-1980s, determination of bacterial growth rates was an important tool in microbial genetics, biochemistry, molecular biology, and microbial physiology. The exciting technical developments of the 1990s and the 2000s eclipsed that tool; as a result, many investigators today lack experience with growth rate measurements. Recently, investigators in a number of areas have started to use measurements of bacterial growth rates for a variety of purposes. Those measurements have been greatly facilitated by the availability of microwell plate readers that permit the simultaneous measurements on up to 384 different cultures. Only the exponential (logarithmic) portions of the resulting growth curves are useful for determining growth rates, and manual determination of that portion and calculation of growth rates can be tedious for high-throughput purposes. Here, we introduce the program GrowthRates that uses plate reader output files to automatically determine the exponential portion of the curve and to automatically calculate the growth rate, the maximum culture density, and the duration of the growth lag phase. GrowthRates is freely available for Macintosh, Windows, and Linux. We discuss the effects of culture volume, the classical bacterial growth curve, and the differences between determinations in rich media and minimal (mineral salts) media. This protocol covers calibration of the plate reader, growth of culture inocula for both rich and minimal media, and experimental setup. As a guide to reliability, we report typical day-to-day variation in growth rates and variation within experiments with respect to position of wells within the plates.

Phylogenetic predictions of carbapenemase activity from the Guiana extended-spectrum (GES) family of ß-lactamases.

Objectives: We investigated the amino acid substitutions in the GES family of ESBLs that were most likely to be involved in the evolution of carbapenemase activity. Methods: To identify the substitutions that are functionally important, we analysed the evolutionary history of the GES ß-lactamases using an alignment and phylogeny to identify sites in GES that show evidence of positive selection and the selected phenotypes. Results and Conclusions: Data indicate that the substitutions G170S and G243A are associated with carbapenemase activity. The substitutions Q43E, E104K and T237A are most likely associated with ESBL activity.

The peaks and geometry of fitness landscapes.

Fitness landscapes are central in the theory of adaptation. Recent work compares global and local properties of fitness landscapes. It has been shown that multi-peaked fitness landscapes have a local property called reciprocal sign epistasis interactions. The converse is not true. We show that no condition phrased in terms of reciprocal sign epistasis interactions only, implies multiple peaks. We give a sufficient condition for multiple peaks phrased in terms of two-way interactions. This result is surprising since it has been claimed that no sufficient local condition for multiple peaks exist. We show that our result cannot be generalized to sufficient conditions for three or more peaks. Our proof depends on fitness graphs, where nodes represent genotypes and where arrows point toward more fit genotypes. We also use fitness graphs in order to give a new brief proof of the equivalent characterizations of fitness landscapes lacking genetic constraints on accessible mutational trajectories. We compare a recent geometric classification of fitness landscape based on triangulations of polytopes with qualitative aspects of gene interactions. One observation is that fitness graphs provide information that are not contained in the geometric classification. We argue that a qualitative perspective may help relating theory of fitness landscapes and empirical observations.

The Klebsiella pneumoniae carbapenemase (KPC) ß-Lactamase Has Evolved in Response to Ceftazidime Avibactam.

Klebsiella pneumoniae carbapenemase KPC is an important resistance gene that has disseminated globally in response to carbapenem use. It is now being implicated as a resistance determinant in Ceftazidime Avibactam (CAZ-AVI) resistance. Given that CAZ-AVI is a last-resort antibiotic, it is critical to understand how resistance to this drug is evolving. In particular, we were interested in determining the evolutionary response of KPC to CAZ-AVI consumption. Through phylogenetic reconstruction, we identified the variable sites under positive selection in the KPC gene that are correlated with Ceftazidime Avibactam (CAZ-AVI) resistance. Our approach was to use a phylogeny to identify multiple independent occurrences of mutations at variable sites and a literature review to correlate CAZ-AVI resistance with the mutations we identified. We found the following sites that are under positive selection: P104, W105, A120, R164, L169, A172, D179, V240, Y241, T243, Y264, and H274. The sites that correlate with CAZ-AVI resistance are R164, L169, A172, D179, V240, Y241, T243, and H274. Overall, we found that there is evidence of positive selection in KPC and that CAZ-AVI is the major selective pressure.

Distribution of ß-Lactamase Genes in Clinical Isolates from California Central Valley Hospital Deviates from the United States Nationwide Trends.

The evolution and dissemination of antibiotic resistance genes throughout the world are clearly affected by the selection and migration of resistant bacteria. However, the relative contributions of selection and migration at a local scale have not been fully explored. We sought to identify which of these factors has the strongest effect through comparisons of antibiotic resistance gene abundance between a distinct location and its surroundings over an extended period of six years. In this work, we used two repositories of extended spectrum ß-lactamase (ESBL)-producing isolates collected since 2013 from patients at Dignity Health Mercy Medical Center (DHMMC) in Merced, California, USA, and a nationwide database compiled from clinical isolate genomes reported by the National Center for Biotechnology Information (NCBI) since 2013. We analyzed the stability of average resistance gene frequencies over the years since collection of these clinical isolates began for each repository. We then compared the frequencies of resistance genes in the DHMMC collection with the averages of the nationwide frequencies. We found DHMMC gene frequencies are stable over time and differ significantly from nationwide frequencies throughout the period of time we examined. Our results suggest that local selective pressures are a more important influence on the population structure of resistance genes in bacterial populations than migration. This, in turn, indicates the potential for antibiotic resistance to be controlled at a regional level, making it easier to limit the spread through local stewardship.

Growth rate assays reveal fitness consequences of ß-lactamases.

NON-TECHNICAL SUMMARY: Antibiotic resistance is a global human health problem. We partnered with Dignity Health Mercy Medical Center to study antibiotic resistance in clinical isolates. We tested whether growth rates, a sensitive assay used to measure the fitness of bacterial samples, correlate with a clinical test to measure antibiotic resistance. We found a strong correlation between these two methods suggesting that growth rates could be reliably applied to evolutionary studies of clinically relevant problems. Moreover, the sensitivity of the growth rates assay enabled us to identify fitness effects of specific antibiotic resistance genes.

Evidence for recombination among the alleles encoding TEM and SHV beta-lactamases.

OBJECTIVES: The objective of this research was to determine whether recombination occurs in class A beta-lactamases. METHODS: We performed chi(2) analysis of the observed and expected numbers of times that beta-lactamases from the TEM, SHV and CTX-M groups co-occurred. Additionally, we performed phylogenetic analysis to detect independent occurrences of silent mutations in bla(TEM) and bla(SHV) variants. RESULTS: We found that the distribution of co-occurring bla(TEM), bla(SHV) and bla(CTX-M) alleles in clinical microbial populations is consistent with the regular occurrence of recombination among alleles within the groups. We also found that the distribution of silent mutations in bla(TEM) and bla(SHV) alleles is inconsistent with spontaneous point mutations. CONCLUSIONS: Our findings indicate that recombination has an important effect on the sequence evolution and population distribution of the alleles that encode class A beta-lactamases.

What antimicrobial resistance has taught us about horizontal gene transfer.

Horizontal gene transfer (HGT) has been responsible for the dissemination of numerous antimicrobial-resistance determinants throughout diverse bacterial species. The rapid and broad dissemination of resistance determinants by HGT, and subsequent selection for resistance imposed by the use of antimicrobials, threatens to undermine the usefulness of antimicrobials. However, vigilant surveillance of the emerging antimicrobial resistance in clinical settings and subsequent studies of resistant isolates create a powerful system for studying HGT and detecting rare events. Two of the most closely monitored phenotypes are resistance to beta-lactams and resistance to fluoroquinolones. Studies of resistance to these antimicrobials have revealed that (1) transformation occurs between different species of bacteria including some recipient species that were not previously known to be competent for natural transformation; (2) transduction may be playing an important role in generating novel methicillin-resistant Staphylococcus aureus (MRSA) strains, although the details of transferring the SCCmec element are not yet fully understood; (3) Resistance genes are probably moving to plasmids from chromosomes more rapidly than in the past; and (4) Resistance genes are aggregating upon plasmids. The linkage of numerous resistance genes on individual plasmids may underlie the persistence of resistance to specific antimicrobials even when use of those antimicrobials is discontinued. Further studies of HGT and methods for controlling HGT may be necessary to maintain the usefulness of antimicrobials.

High rate of mobilization for blaCTX-Ms.

We constructed a phylogenetic analysis of class A beta-lactamases and found that the blaCTX-Ms have been mobilized to plasmids approximately 10 times more frequently than other class A beta-lactamases. We also found that the blaCTX-Ms are descended from a common ancestor that was incorporated in ancient times into the chromosome of the ancestor of Kluyvera species through horizontal transfer. Considerable sequence divergence has occurred among the descendents of that ancestral gene sequence since that gene was inserted. That divergence has mainly occurred in the presence of purifying selection, which indicates a slow rate of evolution for blaCTX-Ms in the pre-antimicrobial drug era.

Fitness trade-offs in blaTEM evolution.

bla(TEM-1) expression results in penicillin resistance, whereas expression of many bla(TEM-1) descendants, called extended-spectrum beta-lactamases (ESBLs), results simultaneously in resistance to penicillins and extended-spectrum cephalosporins. Despite the expanded resistance phenotypes conferred by many ESBLs, bla(TEM-1) is still the most abundant bla(TEM) allele in many microbial populations. This study examines the fitness effects of the two amino acid substitutions, R164S and E240K, that have occurred repeatedly among ESBL bla(TEM-1) descendants. Using a single-nucleotide polymorphism-specific real-time quantitative PCR method, we analyzed the fitness of strains expressing bla(TEM-1), bla(TEM-10), and bla(TEM-12). Our results show that bacteria expressing the ancestral bla(TEM-1) allele have a fitness advantage over those expressing either bla(TEM-10) or bla(TEM-12) when exposed to ampicillin. This observation, combined with the fact that penicillins are the most prevalent antimicrobials prescribed worldwide, may explain why bla(TEM-1) has persisted as the most frequently encountered bla(TEM) allele in bacterial populations.

Recombination and selection can remove blaTEM alleles from bacterial populations.

We passaged cells expressing TEM-1 and TEM-12 from a single plasmid through either ampicillin or ceftazidime. We found that the combined effects of recombination and selection removed the bla(TEM-1) allele from the bacterial population when it was passaged through ceftazidime or the bla(TEM-12) allele when cultures were passaged through ampicillin.

Phylogenetic analysis as a tool in molecular epidemiology of infectious diseases.

Phylogenetics is a powerful tool for microbial epidemiology, but it is a tool that is often misused and misinterpreted by the field. Microbial epidemiologists are cautioned that in order to draw any inferences about the order of descent from a common ancestor it is necessary to correctly root a phylogenetic tree. Epidemiological samples of microbial populations typically include both ancestors and their descendants. In order to illustrate the relationships of those isolates, the phylogenetic method used must be able to detect zero-length branches. Unweighted Pair-Group Method (UPGMA) is the phylogenetic method that is most widely used in microbial epidemiology. Because UPGMA cannot detect zero length branches, and because it places the root of the tree based on a usually-false assumption, UPGMA is the worst possible choice among the several phylogenetic methods available. Because microbial epidemiology deals with relationships among strains within a species, rather than with relationships among species, recombination within those species can render phylogenetic trees meaningless and positively misleading. When there is evidence of significant recombination within the species of interest phylogenetic trees should not be used at all. Instead, alternative tools such as eBURST should be used to understand relationships among isolates.

Predicting evolutionary potential: in vitro evolution accurately reproduces natural evolution of the tem beta-lactamase.

To evaluate the validity of our in vitro evolution method as a model for natural evolutionary processes, the TEM-1 beta-lactamase gene was evolved in vitro and was selected for increased resistance to cefotaxime, cefuroxime, ceftazadime, and aztreonam, i.e., the "extended-spectrum" phenotype. The amino acid substitutions recovered in 10 independent in vitro evolvants were compared with the amino acid substitutions in the naturally occurring extended-spectrum TEM alleles. Of the nine substitutions that have arisen multiple times in naturally occurring extended-spectrum TEM alleles, seven were recovered multiple times in vitro. We take this result as evidence that our in vitro evolution technique accurately mimics natural evolution and can therefore be used to predict the results of natural evolutionary processes. Additionally, our results predict that a phenotype not yet observed among TEM beta-lactamases in nature-resistance to cefepime-is likely to arise in nature.

Experimental prediction of the natural evolution of antibiotic resistance.

The TEM family of beta-lactamases has evolved to confer resistance to most of the beta-lactam antibiotics, but not to cefepime. To determine whether the TEM beta-lactamases have the potential to evolve cefepime resistance, we evolved the ancestral TEM allele, TEM-1, in vitro and selected for cefepime resistance. After four rounds of mutagenesis and selection for increased cefepime resistance each of eight independent populations reached a level equivalent to clinical resistance. All eight evolved alleles increased the level of cefepime resistance by a factor of at least 32, and the best allele improved by a factor of 512. Sequencing showed that alleles contained from two to six amino acid substitutions, many of which were shared among alleles, and that the best allele contained only three substitutions.

Experimental prediction of the evolution of cefepime resistance from the CMY-2 AmpC beta-lactamase.

Understanding of the evolutionary histories of many genes has not yet allowed us to predict the evolutionary potential of those genes. Intuition suggests that current biochemical activity of gene products should be a good predictor of the potential to evolve related activities; however, we have little evidence to support that intuition. Here we use our in vitro evolution method to evaluate biochemical activity as a predictor of future evolutionary potential. Neither the class C Citrobacter freundii CMY-2 AmpC beta-lactamase nor the class A TEM-1 beta-lactamase confer resistance to the beta-lactam antibiotic cefepime, nor do any of the naturally occurring alleles descended from them. However, the CMY-2 AmpC enzyme and some alleles descended from TEM-1 confer high-level resistance to the structurally similar ceftazidime. On the basis of the comparison of TEM-1 and CMY-2, we asked whether biochemical activity is a good predictor of the evolutionary potential of an enzyme. If it is, then CMY-2 should be more able than the TEMs to evolve the ability to confer higher levels of cefepime resistance. Although we generated CMY-2 evolvants that conferred increased cefepime resistance, we did not recover any CMY-2 evolvants that conferred resistance levels as high as the best cefepime-resistant TEM alleles.

Rational design of antibiotic treatment plans: a treatment strategy for managing evolution and reversing resistance.

The development of reliable methods for restoring susceptibility after antibiotic resistance arises has proven elusive. A greater understanding of the relationship between antibiotic administration and the evolution of resistance is key to overcoming this challenge. Here we present a data-driven mathematical approach for developing antibiotic treatment plans that can reverse the evolution of antibiotic resistance determinants. We have generated adaptive landscapes for 16 genotypes of the TEM ß-lactamase that vary from the wild type genotype "TEM-1" through all combinations of four amino acid substitutions. We determined the growth rate of each genotype when treated with each of 15 ß-lactam antibiotics. By using growth rates as a measure of fitness, we computed the probability of each amino acid substitution in each ß-lactam treatment using two different models named the Correlated Probability Model (CPM) and the Equal Probability Model (EPM). We then performed an exhaustive search through the 15 treatments for substitution paths leading from each of the 16 genotypes back to the wild type TEM-1. We identified optimized treatment paths that returned the highest probabilities of selecting for reversions of amino acid substitutions and returning TEM to the wild type state. For the CPM model, the optimized probabilities ranged between 0.6 and 1.0. For the EPM model, the optimized probabilities ranged between 0.38 and 1.0. For cyclical CPM treatment plans in which the starting and ending genotype was the wild type, the probabilities were between 0.62 and 0.7. Overall this study shows that there is promise for reversing the evolution of resistance through antibiotic treatment plans.