ABSTRACT Streptococcus sanguinis is an abundant member of the oral microbiota and a promising oral
probiotic since it has been associated with oral health in humans and shown to compete with oral pathogens.
Conversely, S. sanguinis is also an important cause of bacterial endocarditis, a serious infection of the heart
that affects >40,000 people in the U.S each year and has mortality rates exceeding 20%. Endocarditis is of
critical concern to U.S dentists, who typically provide antibiotic prophylaxis to high-risk patients prior to invasive
dental procedures. The underlying cause of the duality in S. sanguinis lifestyle as a health-associated
commensal or a serious pathogen is poorly understood. Delineation of the S. sanguinis factors that are
important for oral health versus endocarditis virulence would facilitate the development of efficacious and safe
probiotics, and more targeted prophylactics against endocarditis during dental procedures. A significant barrier
to progress in the field is the paucity of genetically-, epigenetically-, or phenotypically- characterized isolates.
Numerous studies have employed a variety of poorly characterized strains, many of which are no longer
classified as S. sanguinis, making it difficult to assess findings. Moreover, related Streptococcus species are
notoriously genetically diverse, and additional epigenetic mechanisms have been demonstrated to drive
phenotypic diversity and virulence even within clonal populations. Although the complete genome of SK36 has
been published, its epigenetic profile (i.e. methylome) is unknown and, regardless, no single strain can capture
the true capabilities of the species. This proposal will surmount these barriers by first using single-molecule
real-time (SMRT) sequencing to generate complete genomes/epigenomes of 40 bona fide S. sanguinis strains,
representing both oral and endocarditis isolates. This will reveal the core and accessory genomes related to
oral and endocarditis environmental niches, in addition to the methylome of each isolate. The central
hypothesis of this proposal is that the dual lifestyle of S. sanguinis is either A) facilitated by the presence or
absence of specific genetic loci (i.e. virulence genes, islands, or plasmids) or B) epigenetically encoded and
induced by transcriptomic alterations within permissive environmental niches. An in-vivo endocarditis model
will be used to screen for infective endocarditis virulence and test whether targeted reengineered isogenic
mutants are deficient for virulence, making them suitable candidates for development as oral probiotics. The
immediate outcomes will be an S. sanguinis strain collection that is genetically-, epigenetically-, and
phenotypically defined and which will be made available to the research community, and the ascertainment of
whether S. sanguinis endocarditis virulence is genetically or epigenetically encoded. The long-term outcome
will be the creation of strains engineered to have low endocarditis virulence. If virulence is controlled
epigenetically in S. sanguinis as it is in some of its closest relatives, the loss of endocarditis virulence will be
accompanied by an increase in oral competitiveness, making these strains ideal for oral probiotic development.