Fig. 2: S. Typhimurium mutants deficient in D-fructose, D-mannose, and D-galactose are attenuated.

Panels (a–e) show the competitive index for ΔfruK (D-fructose; white), ΔgalK (D-galactose; red), and ΔmanA (D-mannose; blue) across five mouse models, statistically compared to the SL1344 wild type. The upper black line indicates a wild-type competitive index of 1, and the lower line indicates the detection limit (10⁻³). Sample sizes for each day post-infection are indicated in the subpanels. Panel (f) shows the competitive index for ΔnagB (N-acetylglucosamine), and panel (h) shows the competitive index for ΔkdgK (hexuronate) at 2 days post-infection across all models. Panel (g) depicts a schematic of hexuronate degradation, highlighting KdgK as a key enzyme. Panel (i) presents SL1344 CFUs/g feces for WISH-barcoded SL1344 pool in competition with ATCC14028s at 1:50 (n = 9) and 1:1000 (n = 9) ratios and in the absence of ATCC14028s (n = 12), from at least independent experiments. Panel (j) shows Shannon evenness scores (SES) for the SL1344 mutant pool, with the number of datasets exceeding SES > 0.9 indicated. Panel (k) shows the ΔfruK competitive index in the presence of ATCC14028s at 1:50 ratios (n = 7 for day 1; n = 9 for day 2 and cecal content) and 1:1000 ratios (n = 9). It also includes data in the absence of ATCC14028S (day 1: n = 12; day 2 and cecal content: n = 11). Box-and-whisker plots (a–f, h, k) display the median, interquartile range (25th to 75th percentiles), minimum and maximum values, and individual data points. Bar plots (i, j) show medians with individual points. P values (Mann–Whitney U-test) are: **** ≙ P < 0.0001; *** ≙ P < 0.0005; ** ≙ P < 0.005; * ≙ P < 0.05; ns ≙ P > 0.05. The source data for panels (a–i) are provided in the Source Data file.