Microbial virulence is a complex and often multifactorial phenotype, intricately linked to a pathogen’s evolutionary trajectory. Toxicity, the ability to destroy host cell membranes, and adhesion, the ability to adhere to human tissues, are the major virulence factors of many bacterial pathogens, including Staphylococcus aureus. Here, we assayed the toxicity and adhesiveness of 90 MRSA (methicillin resistant S. aureus) isolates and found that while there was remarkably little variation in adhesion, toxicity varied by over an order of magnitude between isolates, suggesting different evolutionary selection pressures acting on these two traits. We performed a genome-wide association study (GWAS) and identified a large number of loci, as well as a putative network of epistatically interacting loci, that significantly associated with toxicity. Despite this apparent complexity in toxicity regulation, a predictive model based on a set of significant single nucleotide polymorphisms (SNPs) and insertion and deletions events (indels) showed a high degree of accuracy in predicting an isolate’s toxicity solely from the genetic signature at these sites. Our results thus highlight the potential of using sequence data to determine clinically relevant parameters and have further implications for understanding the microbial virulence of this opportunistic pathogen.

Some of the most common infectious diseases are caused by bacteria that naturally colonise humans asymptomatically. Combating these opportunistic pathogens requires an understanding of the traits that differentiate infecting strains from harmless relatives. Staphylococcus epidermidis is carried asymptomatically on the skin and mucous membranes of virtually all humans but is a major cause of nosocomial infection associated with invasive procedures. Here we address the underlying evolutionary mechanisms of opportunistic pathogenicity by combining pangenome-wide association studies and laboratory microbiology to compare S. epidermidis from bloodstream and wound infections and asymptomatic carriage. We identify 61 genes containing infection-associated genetic elements (k-mers) that correlate with in vitro variation in known pathogenicity traits (biofilm formation, cell toxicity, interleukin-8 production, methicillin resistance). Horizontal gene transfer spreads these elements, allowing divergent clones to cause infection. Finally, Random Forest model prediction of disease status (carriage vs. infection) identifies pathogenicity elements in 415 S. epidermidis isolates with 80% accuracy, demonstrating the potential for identifying risk genotypes pre-operatively.

The bacterium Staphylococcus aureus is a major human pathogen for which the emergence of antibiotic resistance is a global public health concern. Infection severity, and in particular bacteraemia-associated mortality, has been attributed to several host-related factors, such as age and the presence of comorbidities. The role of the bacterium in infection severity is less well understood, as it is complicated by the multifaceted nature of bacterial virulence, which has so far prevented a robust mapping between genotype, phenotype and infection outcome. To investigate the role of bacterial factors in contributing to bacteraemia-associated mortality, we phenotyped a collection of sequenced clinical S. aureus isolates from patients with bloodstream infections, representing two globally important clonal types, CC22 and CC30. By adopting a genome-wide association study approach we identified and functionally verified several genetic loci that affect the expression of cytolytic toxicity and biofilm formation. By analysing the pooled data comprising bacterial genotype and phenotype together with clinical metadata within a machine-learning framework, we found significant clonal differences in the determinants most predictive of poor infection outcome. Whereas elevated cytolytic toxicity in combination with low levels of biofilm formation was predictive of an increased risk of mortality in infections by strains of a CC22 background, these virulence-specific factors had little influence on mortality rates associated with CC30 infections. Our results therefore suggest that different clones may have adopted different strategies to overcome host responses and cause severe pathology. Our study further demonstrates the use of a combined genomics and data analytic approach to enhance our understanding of bacterial pathogenesis at the individual level, which will be an important step towards personalized medicine and infectious disease management.

Bacterial virulence is a multifaceted trait where the interactions between pathogen and host factors affect the severity and outcome of the infection. Toxin secretion is central to the biology of many bacterial pathogens and is widely accepted as playing a crucial role in disease pathology. To understand the relationship between toxicity and bacterial virulence in greater depth, we studied two sequenced collections of the major human pathogen Staphylococcus aureus and found an unexpected inverse correlation between bacterial toxicity and disease severity. By applying a functional genomics approach, we identified several novel toxicity-affecting loci responsible for the wide range in toxic phenotypes observed within these collections. To understand the apparent higher propensity of low toxicity isolates to cause bacteraemia, we performed several functional assays, and our findings suggest that within-host fitness differences between high- and low-toxicity isolates in human serum is a contributing factor. As invasive infections, such as bacteraemia, limit the opportunities for onward transmission, highly toxic strains could gain an additional between-host fitness advantage, potentially contributing to the maintenance of toxicity at the population level. Our results clearly demonstrate how evolutionary trade-offs between toxicity, relative fitness, and transmissibility are critical for understanding the multifaceted nature of bacterial virulence. Global efforts to counter the growing problem of antibiotic resistance and develop alternative treatment strategies rely on a fuller understanding of when and why opportunistic pathogens cause disease. Recent advances in DNA sequencing technologies have opened up new opportunities to study infectious organisms, yet identifying the genetic variants that explain differences in disease remains challenging. Here we aimed to understand the complex relationship between toxicity—a known risk factor for disease in many bacteria—and infection severity for the major human pathogen S. aureus. Against expectations, we found that the bacteria that caused the most severe disease were the least toxic strains. We were able to determine the mutations responsible for the differences in toxicity and identified a number of novel toxicity-affecting genes. We further discovered that bacterial fitness in human serum could explain the unexpected association of low-toxicity isolates with severe, invasive disease. Invasive S. aureus infections are usually considered a dead end for these bacteria, as these infections are rarely transmitted to another person. Here we show using a simple mathematical model that this might in fact favour transmission of highly toxic bacteria on a population level and thus contribute to their global success. Our work therefore highlights the complexity of bacterial infection and should aid in devising new treatment and control strategies against this important pathogen.

The capacity for intracellular survival within phagocytes is likely a critical factor facilitating the dissemination of Staphylococcus aureus in the host. To date, the majority of work on S. aureus -phagocyte interactions has focused on neutrophils and, to a lesser extent, macrophages, yet we understand little about the role played by dendritic cells (DCs) in the direct killing of this bacterium. Using bone marrow-derived DCs (BMDCs), we demonstrate for the first time that DCs can effectively kill S. aureus but that certain strains of S. aureus have the capacity to evade DC (and macrophage) killing by manipulation of autophagic pathways. Strains with high levels of Agr activity were capable of causing autophagosome accumulation, were not killed by BMDCs, and subsequently escaped from the phagocyte, exerting significant cytotoxic effects. Conversely, strains that exhibited low levels of Agr activity failed to accumulate autophagosomes and were killed by BMDCs. Inhibition of the autophagic pathway by treatment with 3-methyladenine restored the bactericidal effects of BMDCs. Using an in vivo model of systemic infection, we demonstrated that the ability of S. aureus strains to evade phagocytic cell killing and to survive temporarily within phagocytes correlated with persistence in the periphery and that this effect is critically Agr dependent. Taken together, our data suggest that strains of S. aureus exhibiting high levels of Agr activity are capable of blocking autophagic flux, leading to the accumulation of autophagosomes. Within these autophagosomes, the bacteria are protected from phagocytic killing, thus providing an intracellular survival niche within professional phagocytes, which ultimately facilitates dissemination.