Persistent Staphylococcus aureus infection through the selection of alternative lifestyles

Abstract

Staphylococcus (S.) aureus is an important human pathogen and is notorious for its ability to survive stressful environments. The classic mechanism of survival is through expression of virulence factors that can neutralise the immune response and the widespread transfer of antibiotic resistance that hinders clinical antibiotic treatments. However, cases occur where initially a clinical infection has cleared only for the infection to relapse. This can be via a new infection or as a result of changes in bacterial lifestyles of the original infection that can create sub-populations of S. aureus with varied mechanisms to survive different conditions of host and medically induced stressors. These lifestyles include the formation of Small Colony Variants (SCV), a cell type characterised by impeded metabolism and reduced virulence which confers evasion of the immune response and antibiotic tolerance. This phenotype is largely unstable and often reverts to the normal cell type which hinders research into the mechanism of SCV formation. To overcome this hurdle, stable SCVs (sSCV) have been generated by researchers through genetic mutations which cause impedance of the electron transport chain, create metabolic defects, and produces the SCV phenotype. Previous research has generated these within a laboratory setting and with a single mutation. This is not a complete representation of the complex, adaptative mechanisms which lead to the SCV phenotype within a clinical setting where adaptation occurs in response to nutritional limitations, immune response, antibiotic stress, and selection over long-term growth. Given there are multiple pathways which can lead to the SCV phenotype, our project investigates two different paradigms of SCV: firstly, that conditions of stress that cause changes within a cell and result in a SCV and when the stress is removed, the cells that revert to the normal cell type. Alternatively, SCVs continually form stochastically within a bacterial population and introduction of conditions of stress select for the SCVs and they dominate the population but when those stressors are removed, the normal cell types are selected. This project utilises alternative research models for investigating SCV formation and clinical isolates of S. aureus from diabetic foot infection (DFI) to study both the mechanisms which define the switch to SCV and the population dynamics which select for SCV. We have used physiological and genetic methods to assess the SCV generated through different methods. Continuous culture within a chemostat was used to grow S. aureus in a controlled, low growth rate over a prolonged duration. In these conditions, we have selected for alternative cell types with decreased metabolic demands within the population. In a similar theme, we have used long-term infection of the osteocyte cell line SaOS-2 provided an ex vivo model of long-term, persistent bone infection. Infection of SaOS-2 resulted in viable but non-culturable S. aureus cells that were able to silently persist within the osteocyte and over time, these cells switched to viable and culturable cells. Adaptive laboratory evolution was used to expose S. aureus to continual antibiotic stress at an inhibitory concentration over time. Over multiple generations of growth within these inhibitory conditions we observed changes in cell type, and development of cells with increased antibiotic resistance and antibiotic tolerance. We have undertaken a specific clinical study isolating S. aureus from patients with DFI. The complications of diabetes are notorious for establishing persistent infections within the foot ulcer and this can spread to the bone causing osteomyelitis. We have identified and extracted pairs of different S. aureus cell types within the same patient and notably we have isolated a pair of isolates comprising of a non-stable SCV and sSCV. From all these research models of S. aureus we have identified alternate cell types selected for in response to the respective stressful conditions. These cell types had phenotypic changes, with colonies of decreased pigmentation and size, decreased growth rate, increased biofilm formation, and increased resistance to antibiotics. Using whole genome sequencing, we identified a large variety of SNPs associated with these phenotypes, including carbohydrate metabolism, metal ion regulation, virulence regulation and cell wall biosynthesis. We found there was a lack of a common pathway associated with the introduced stress associated with our research models. This suggests the adaptations to these stressors are through selection of stochastic genetic variants with greater fitness instead of a defined pathway which lead to the observed cell types in response to the stress.