Genomic and metagenomic analysis of microbial agents causing infective endocarditis

Abstract

Infective endocarditis (IE) is an infection of the inner layer of the heart or the heart valves. Gram-positive cocci of the genera Staphylococcus, Streptococcus, and Enterococcus account for more than 80% of all IE cases. In rare cases, other microbial agents, such as Aerococcus spp. can cause the infection as well. IE is a relatively rare disease with two to ten incidents per 100,000 individuals per year in the general population world-wide. However, the mortality of IE is between 15-20%, and one-year mortality approaches 40%. Therefore, IE is a severe infectious disease, as it involves long term hospitalization and heart surgery in 50% of the cases. This thesis includes five studies, and its main focus is IE caused by Mitis group streptococci (MGS). The majority of the bacteria in this group are commensal colonizers of the human oral cavity, where they can have beneficial effects on the oral health of the host. However, this group of oral bacteria can escape their niche and in rare cases, cause infectious diseases such as IE. The MGS are closely related and naturally competent, which enables them to engage in recombination with closely related species. This complicates correct species identification, which is a crucial part of the diagnosis of IE. Additionally, species identification is important for correct antibiotic treatment, prognosis, and potential underlying predispositions. In cases of recidive of IE, it is also important to establish if the infection is due to treatment failure or whether it is caused by an infection by another species. The first study in this thesis compares different molecular phylogenetic approaches to identify the most suitable method for correct species identification using whole-genome sequencing of 80 MGS strains isolated from IE patient samples. MGS consist of several different species. Among those are the commensal Streptococcus mitis and Streptococcus oralis, which are closely related with the pathogen Streptococcus pneumoniae. S. pneumoniae is one of the leading causes of fatal infections in children world-wide. As the majority of MGS live as commensals in symbiosis with the host, the virulence potential of these species is not fully established. Study two and three in this thesis approaches this issue. Study two investigated genes that could be involved in the virulence of 40 S. mitis and S. oralis strains. All strains were isolated from IE patients and therefore had the ability to cause infections. In the third project, MGS genomes isolated from IE patients were compared with strains isolated from the oral cavity of healthy individuals with the aim to illuminate genetic differences that could explain the shift in pathogenicity. Study four investigated the virulence potential of MGS persevered in a 5,700-year-old chewed birch pitch, i.e., ancient ”chewing gum”. As the pitch had been chewed, the oral microbiota of a western hunter-gatherer woman had been preserved inside the lump. The ancient streptococcal DNA was identified and compared to DNA from modern streptococci. The pathogenic potential of the ancient Streptococcus DNA was investigated to see if the ancient MGS contained more or fewer virulence genes than the modern MGS. Study five in this thesis aimed to identify virulence genes in strains from another, though rare, IE-associated genus, Aerococcus spp. The two species Aerococcus sanguinicola and Aerococcus urinae can cause urinary tract infections, sepsis, and, more rarely, IE. No whole-genome comparisons and genomic characterizations have previously been performed on A. sanguinicola and A. urinae. In this study, we were able to identify several virulence genes associated with human disease. Collectively, the thesis’ first four studies provided an insight into the nature of MGS. For correct species identification, we showed that multi-locus sequence analysis involving seven genes provided enough genetic variability to generate distinct phylogenetic clusters. However, in cases where the hospitals have implemented whole-genome sequencing, species identification using CGI Phylogeny might be the fastest and most accurate approach. We furthermore identified genes that could be associated with virulence; genes associated with host colonization and modulation of the host immune system were identified in the S. mitis, S. oralis, S. sanguinins, S. gordonii, ancient Streptococcus DNA, and in the two Aerococcus species. The same level of virulence genes was identified in the ancient Streptococcus DNA as in modern oral samples. Interestingly, we found no genetic differences between strains isolated from patients with IE and the oral cavity of healthy individuals. Altogether, our findings illustrate the complexity of the role of these bacteria in IE etiology. While they all, most likely, carry a pathogenic potential, there might be other factors than the genetic composition that determines whether the bacterium can successfully establish an infection.