Carbapenem resistant Klebsiella (CRK) species, in particular carbapenem resistant Klebsiella pneumoniae (CRKP), are recognized as “urgent threat” pathogens globally. Bloodstream and other serious Klebsiella infections, including those due to CRK, usually are caused by a strain that colonizes the gastrointestinal (GI) tract. Emerging whole genome sequence (WGS) data show that the GI tract is colonized by clonal bacterial populations, in which genetic diversity develops over time. It is unknown how often bloodstream infections are caused by clonal but genetically diverse strains of a given bacterium, since clinical and research laboratories generally characterize a single colony from positive cultures. In preliminary studies, we performed WGS on 10 colonies from the first CRK-positive blood culture of 10 randomly selected patients. We demonstrated that bloodstream infections in 8 of 10 patients were caused by intraclonal, genetically diverse CRK, including strains with non-synonymous single nucleotide polymorphisms in the core genome, chromosomal gene deletions, insertions and other mutations, and plasmid or plasmid-borne gene loss. Genotypically distinct CRK strains exhibited significant differences in antibiotic susceptibility, capsular polysaccharide (CPS) content, and other virulence-associated phenotypes. Genetic variant strains from 2 patients displayed significant differences in virulence during bloodstream infections of mice. Our objectives in this study are to characterize the genetic and phenotypic diversity of CRK strains recovered from baseline and longitudinal blood cultures of patients in our hospital and to implicate specific genes, plasmids and mutations in antibiotic resistance and virulence. In specific aim 1, we will determine genetic diversity of CRK strains causing baseline and recurrent bloodstream infections by performing WGS, plasmid sequencing, and analyses of core genome phylogeny and gene mutations for individual colonies selected from positive blood cultures. For genetically distinct strains from each positive culture, we will determine antibiotic susceptibility and other phenotypes in vitro. For prioritized stains, we will assess antibiotic responses and virulence during mouse bloodstream infections. In specific aim 2, we will validate the impact of specific genes, plasmids and mutations on antibiotic treatment responses and virulence. We will infect mice intravenously with isogenic, lab-created mutant and complemented strains for genes in which diversity was observed in patients (e.g., those encoding CPS), and plasmid-curated and non-curated strains. We will measure outcomes in mice treated and not treated with antibiotics. Results from this study will affirm the extent and type of genetic diversity in CRK-positive blood cultures, afford new insights into CRK strain dynamics during bloodstream infections, and identify novel genetic determinants of antibiotic treatment responses and virulence. Insights gathered here will form a foundation for investigations of genetic diversity during bloodstream and other infections by non-CR Klebsiella and other bacteria, and for mechanistic studies of CRK genes or gene targets that dampen antibiotic responses and promote pathogenesis.