Figure 1. Experimental design and metrics and CFU dynamics.

(A) The experimental schematic depicts the extraintestinal pathogenic Escherichia coli (ExPEC) barcoded library as colored bacteria, where colors indicate unique barcodes; the library was injected intravenously (i.v.) into mice, after which organs were harvested, homogenized, and plated for enumeration of CFU. Bacterial cells were scraped, pooled, and the barcode abundances in the output populations were quantified. The distributions of the barcodes (far-right graph) define the population structure of the organ, schematized in (B–E). (B–E) Graphical depiction of populations with different combinations of N_b and N_r values. These populations arise from (B) a tight bottleneck and subsequent expansion of purple cells, (C, D) a wider bottleneck and expansion of purple (C) or red (D) cells and (E) a wide bottleneck and even growth of all cells. Metrics used for comparisons between samples (genetic distance [GD], “resilient” genetic distance [RD], and fractional RD [FRD]) are indicated. (F) CFU recovered from select organs on days 0–5; the full CFU data set is provided in Figure 1—figure supplements 2–4. Points with the same fill and border color were obtained from the same animal. Arrows pointing to G–J correspond to the points where complete barcode frequency distributions are shown below (in G–J). (G–J) Barcode distributions and N_b and N_r values are shown. Red lines separate populations that were distinguished by the N_r algorithm (Hullahalli et al., 2021). Blue line indicates the algorithm-defined threshold for noise. Percentages represent the relative abundance of barcodes within each region. G and H represent examples with highly abundant barcodes that skew N_b values lower. (G) reproduced from Figure S5B (Hullahalli et al., 2021).

Figure 1—figure supplement 1. Barcode stability and creation of the STAMPR standard curve.

(A) The barcoded library was serially passed in LB lacking antibiotic. Each day, colonies containing barcodes were enumerated as a fraction of the kanamycin-resistant cells. In addition, every day, 30 single colonies derived from the serially passaged culture were grown on plates lacking kanamycin and then patched onto media containing kanamycin. All colonies retained the antibiotic resistance marker, suggesting that the barcode is stable for at least ~50 generations. (B) Growth curves of nine individual barcoded clones (STAMP) and the WT untagged strain (CHS7) were indistinguishable, indicating that the tags do not confer a growth defect. (C) Sequencing of the barcodes at different known bottleneck sizes (plated serial dilutions of the barcoded library) showed that both N_b and N_r accurately reflect founding population (FP) values up to 10⁵ – corresponding to the approximate sequencing depth of the sample. The underlying barcode distributions that result from the serial dilutions are shown in (D).

Figure 1—figure supplement 2. Longitudinal CFU, N_b, and N_r measurements from the blood, bile, and spleen.

Points with the same fill and border color are from the same animal; the same fill and border coloring scheme is used for the animals shown in Figure 1—figure supplement 3 and Figure 1—figure supplement 4.

Figure 1—figure supplement 3. Longitudinal CFU, N_b, and N_r measurements from the lungs and kidneys.

Points with the same fill and border color are from the same animal and correspond to animals shown in Figure 1—figure supplement 2 and Figure 1—figure supplement 4.

Figure 1—figure supplement 4. Longitudinal CFU, N_b, and N_r measurements from the liver.

Points with the same fill and border color are from the same animal and correspond to animals shown in Figure 1—figure supplement 2 and Figure 1—figure supplement 3.