HvgA enables GBS persistent intestinal colonization and promotes its crossing of the intestinal barrier. (a–c) Groups (n = 6) of 3–4-wk-old Swiss female mice were infected orally with 10¹⁰ CFUs WT GBS or mutant strains, and fecal shedding was assessed 3, 7, and 14 d after inoculation by CFUs enumeration. Values represent fecal shedding of the six mice in a cage. (d) Infection of Swiss mice (n = 4) ligated cecum with 10¹⁰ GBS WT ST-17 or ΔhvgA mutant strain. Bacteria were recovered 1 h after infection. Values are expressed as the percentage of adhesion relative to the inoculum. (e and f) Competition assays between WT ST-17 (BM110) and WT non–ST-17 (NEM316) or ΔhvgA BM110 mutant. 3–4-wk-old Swiss mice (n = 6) were infected orally with a 1:1 mixture of the two strains (total dose 10¹⁰ CFUs). Bacteria were enumerated from the feces collected 3, 7, and 10 d after infection. (g) Mortality rate 12 h after infection of 15–21 d old BALB/c mice (n = 10) infected orally with the WT ST-17 or ΔhvgA mutant strains. (h) Groups of 4-wk-old BALB/c mice (n = 10) were inoculated orally with 10¹⁰ CFUs WT ST-17 or ΔhvgA mutant strains. 8 h after infection, animals were sacrificed. Cecum were collected and divided longitudinally in two parts. One half was directly homogenized: extra- and intratissular bacteria associated with the cecum were enumerated (no genta.). The second half was incubated for 3 h in DMEM containing 250 µg/ml gentamicin to kill extratissular bacteria and homogenized. Intratissular invading bacteria were then enumerated (genta.). Animal experiments represented in this figure were repeated at least two times and groups of mice contained at least five animals. Error bars represent the SD of depicted variable performed in triplicate. Asterisks indicate significant differences as assessed by the Mann-Whitney test (*, P < 0.05; **, P < 0.01; ***, P < 0.001).