Entropy of a bacterial stress response is a generalizable predictor for fitness and antibiotic sensitivity

Abstract

Current approaches explore bacterial genes that change transcriptionally upon stress exposure as diagnostics to predict antibiotic sensitivity. However, transcriptional changes are often specific to a species or antibiotic, limiting implementation to known settings only. While a generalizable approach, predicting bacterial fitness independent of strain, species or type of stress, would eliminate such limitations, it is unclear whether a stress-response can be universally captured. By generating a multi-stress and species RNA-Seq and experimental evolution dataset, we highlight the strengths and limitations of existing gene-panel based methods. Subsequently, we build a generalizable method around the observation that global transcriptional disorder seems to be a common, low-fitness, stress response. We quantify this disorder using entropy, which is a specific measure of randomness, and find that in low fitness cases increasing entropy and transcriptional disorder results from a loss of regulatory gene-dependencies. Using entropy as a single feature, we show that fitness and quantitative antibiotic sensitivity predictions can be made that generalize well beyond training data. Furthermore, we validate entropy-based predictions in 7 species under antibiotic and nonantibiotic conditions. By demonstrating the feasibility of universal predictions of bacterial fitness, this work establishes the fundamentals for potentially new approaches in infectious disease diagnostics.

Identifying the effect of vancomycin on health care–associated methicillin-resistant Staphylococcus aureus strains using bacteriological and physiological media

“…The increasingly ubiquitous RNA-seq data have been used to predict the transcriptional regulatory networks in S. aureus , where antibodies for most regulators are not readily available [ 36 ]. These data have also been used to predict fitness and sensitivities to different antibiotics in different pathogens [ 37 ]. In parallel to RNA-seq–based approaches, metabolic modeling has also come to the forefront of the effort to understand resistance mechanisms.…”

The Impact of Long-Term Antibiotic Therapy of Cutaneous Adverse Reactions to EGFR Inhibitors in Colorectal Cancer Patients

“…The long-term antibiotic therapy may alter not only the abundance of some species, but also the bacterial fitness, defined as the ability of bacteria to adjust their metabolism to adequately suit environmental conditions, thereby being able to survive and grow; this concept is considered a major physiological determinant [108][109][110]. Under certain conditions, including antibiotic therapy, bacteria with low fitness could be favorized, having significantly high and fluctuating gene expression changes that exceed bacteria with high level fitness [111][112][113]. Regarding possible modifications of gene expression of the gut bacteria in patients with CRC under antibiotic therapy, further research is needed to demonstrate these potential changes and also their impact on microbial virulence, antibiotic resistance and overall outcome of the patients.…”

Permutation Entropy as a Universal Disorder Criterion: How Disorders at Different Scale Levels Are Manifestations of the Same Underlying Principle

“…In a seminal study, Zhu et al showed that bacteria of different species respond in a similar fashion to antibiotic stressors [ 22 ]. Although bacterial stress responses include many different genetic pathways that depend on the type of stressor and the bacterial species involved, a generic stress response could nonetheless be observed when considering the whole system dynamics (i.e., when observing the whole gene transcription activity as measured in terms of differential mRNA expression in time).…”

“…Such ‘disorder’ can be expressed in terms of a statistical quantity called permutation entropy, which is a measure of the amount of randomness that can be observed in the covariance patterns the describe the relationships between the various components of a system ( Box 1 ). Zhu et al noted that the observed rise in disorder scores resulted from large-amplitude changes that were produced by independently responding genes, and that this independence may result from of a loss of regulatory connections that normally synchronize gene activity to produce order ( Figure 2 ) [ 22 ]. As it turns out, permutation entropy levels in the timeseries of bacterial gene expression predict bacterial fitness (defined as the growth and survival rates of bacteria).…”