1207. Combining standard bacterial vaginosis treatment with cystine uptake inhibitors to block growth of <i>Lactobacillus iners</i> is a potential a target for shifting the cervicovaginal microbiota towards health-associated <i>Lactobacillus crispatus</i>-dominant communities

Abstract Background Cervicovaginal microbiota domination by Lactobacillus crispatus is associated with beneficial health outcomes, whereas L. iners dominance has more adverse associations. However bacterial vaginosis (BV) treatment with metronidazole (MTZ) typically leads to domination by L. iners rather than L. crispatus. L. iners differs from other lactobacilli by its inability to grow in MRS media. We hypothesized that exploring this growth difference would identify targets for selective L. iners inhibition. Methods Bacteria were grown anaerobically. Nutrient uptake and metabolism were assessed using UPLC-MS/MS and isotopically labeled substrates. Bacterial genome annotation employed Prodigal, Roary, and EggNOG. Competition experiments with mock mixed communities were analyzed by 16S rRNA gene sequencing. We confirmed result generalizability using a diverse collection of South African and North American strains and genomes. Results Supplementing MRS broth with L-cysteine (Cys) or L-cystine permitted robust L. iners growth, while L. crispatus grew without Cys supplementation. Despite their different growth requirements, neither species could synthesize Cys via canonical pathways. Adding the cystine uptake inhibitors S-methyl-L-cysteine (SMC, Fig 1) or seleno-DL-cystine (SDLC) blocked growth of L. iners but not other lactobacilli, suggesting L. iners lacks mechanisms other lactobacilli use to exploit complex exogenous Cys sources. Notably, cydABCD, an operon with Cys/glutathione transport and redox homeostasis activities, is absent from L. iners but present in non-iners Lactobacillus species. Consistent with possible roles for cydABCD in explaining the observed phenotypes, (1) L. iners failed to take up exogenous glutathione and (2) supplementing MRS with reducing agents permitted L. iners growth, which could be blocked by SMC or SDLC. In growth competitions testing L. iners and L. crispatus within mock BV-like communities, SMC plus MTZ outperformed MTZ alone in promoting L. crispatus dominance (Figs 2&amp;3). Figure 1: S-methyl-L-cysteine (SMC) selectively blocks growth of L. iners but not other cervicovaginal Lactobacillus species in cysteine-supplemented MRS broth. Growth was measured by optical density and inhibition calculated relative to Cys-supplemented no-inhibitor control during exponential growth. Values displayed are median (+/- maximum/minimum) for 3 replicates from a single experiment. In all panels, representative data are shown from 1 of &amp;gt;=2 independent experiments for each bacterial strain and media condition. Results are representative of multiple strains for L. iners (n = 16), L. crispatus (n = 7), and L. jensenii (n = 2). Figure 2: Relative abundance of L. crispatus, L. iners, or various BV-associated bacteria in mock bacterial communities grown in rich, non-selective media with or without metronidazole (MTZ) and/or SMC. Relative abundance was determined by bacterial 16S rRNA gene sequencing. Data are shown for three representative mock communities with 5 replicates per media condition. Figure 3: Ratio of L. crispatus to other species in the mock bacterial communities depicted in Figure 2. Statistical significance determined via 1-way ANOVA of log10-transformed ratios with post-hoc Tukey test; selected pairwise comparisons are shown (***, p &amp;lt; 0.001). Conclusion L. iners has unique requirements for exogenous cysteine/cystine or a reduced environment for growth. Targeting cystine uptake to inhibit L. iners is a potential strategy for shifting cervicovaginal microbiota towards L. crispatus-dominant communities. Disclosures Douglas S. Kwon, MD, PhD, Day Zero Diagnostics (Consultant, Shareholder, Other Financial or Material Support, co-founder)