Applying at scale genomic and phenotypic approaches to study pathogens linked to nosocomial transmission and disease
Infectious diseases continue to pose major threats to human health. This manifests in different ways, from the incessant rise of antimicrobial resistance, to seasonal epidemics of respiratory viruses and global pandemics, all of which cause high levels of morbidity, mortality and burden on healthcare systems. In this thesis I have investigated three pathogens that are archetypes of these threats: the Gram-negative bacterium Klebsiella pneumoniae, the seasonal respiratory virus influenza and the pandemic coronavirus SARS-CoV-2.
Using isolates prospectively collected at Cambridge University Hospitals, I have utilised a combination of genomic, epidemiological and clinical datasets to characterise these pathogens. Phylogenetic analysis of K. pneumoniae bacteraemia isolates from a four year period show considerable genomic diversity and possible cryptic transmission events involving a lineage associated with hospital outbreaks. Using a novel method of high content imaging using confocal microscopy, I have developed high throughput pipelines that can quantify morphological changes induced in multiple bacterial species by a range of clinically relevant antimicrobials. I have used this approach to determine the phenotypic variation in a collection of 175 K. pneumoniae isolates that represent the diversity of the species. In paired imaging and RNAseq experiments using clinical isolates, I have identified large numbers of genes that are differentially expressed in the presence of antimicrobials. Among them are pathways associated with antimicrobial resistance, including mutagenesis, drug efflux and plasmid conjugation. Together, these data suggest that rather than passively selecting resistant isolates, antimicrobials may also actively drive resistance in K. pneumoniae.
Using large genomic epidemiology studies I have identified extensive networks of nosocomial transmission of influenza and SARS-CoV-2 in CUH, involving patients and healthcare workers with an associated high inpatient mortality. For hospitalised influenza patients, I have shown the protective effect of the antiviral oseltamivir in multivariable analyses, and how rates of hospital acquired infection and mortality have fallen in response to multidisciplinary interventions over three winters. During the COVID-19 pandemic, real-time genomic-epidemiology analyses have been integrated with clinical services to understand and reduce viral spread.
Finally, I have applied these approaches to a community setting, identifying SARS-CoV-2 transmission dynamics in the University of Cambridge and how they have been interrupted by local and national measures to control the COVID-19 pandemic. In particular, I have detailed the implementation and evaluation of a large regular screening programme for asymptomatic SARS-CoV-2 infection using pooled sampling, a programme that has provided insights into the efficacy of mass testing approaches.