Bacteriophages are a promising tool for microbiome editing, yet their development has been constrained by limited insights into bacteriophage-host interactions within their shared mammalian body habitat. We isolated a lytic phage ΦKL11 that efficiently targets a disease-associated member of the human gut microbiota, Eggerthella lenta, during in vitro growth. However, ΦKL11 selects for a pre-existing and reversible bacteriophage-resistant sub-population in mice. Long-read sequencing revealed a massive genomic inversion event, representing >50% of the E. lenta genome, enriched in response to bacteriophage infection. Transcriptomics linked this inversion to the altered expression of three capsular polysaccharide synthesis (CPS) gene clusters and transmission electron microscopy confirmed differential capsule production. Finally, we show that ΦKL11 has a broad host range attributable to CPS and other strain-variable genes. These findings suggest a previously unrecognized strategy for phage evasion in the gut, involving megabase-scale genomic inversions and reversible capsule variation driving phage resistance.

A summary of the data from saturation binding assays used to calculate Bmax and receptor number per cell for bladder cancer cell lines. The data are representative of two independent assays. Data are reported as mean ± standard deviation.

Supplemental Movies 2a-2c: 22Rv1-NuclightGreen and SHP-77-NuclightRed mixed population co-cultures. Time-lapse movies of 22Rv1 cells (labeled with NuclightGreen) and SHP-77 cells (labeled with NuclightRed) co-cultured with T cells. Both the 22Rv1 and SHP-77 cells were plated at equal densities and treated with either NT control BiTE®, AMG 757, or the PSMA-targeting HLE BiTE, AMG 160. T cells were added to each well at an E:T ratio of 10:1. Images were captured every 4 hours at 4x magnification for 5 days. a. 22Rv1, SHP-77, and T cells co-cultured with NT control BiTE®. b. 22Rv1, SHP-77, and T cells co-cultured with AMG 757. c. 22Rv1, SHP-77, and T cells co-cultured with AMG 160.

CDCP1 mRNA expression by pathologic stage in the TCGA dataset. p=0.72 by Kruskal-Wallis test.

Supplementary Data: Figures and Legends combined

CDCP1 protein expression by clinical stage using the tissue microarray. p=0.32 by Kruskal-Wallis test.

A plot showing the mean tumor volume for the vehicle and treated arms from the antitumor assessment study in mice bearing UMUC3 xenografts. The differences in mean between treatment arms was found to be significant (P < 0.001) using an unpaired, two tailed t test by day 20. Data are reported as mean ± standard deviation.

CDCP1 mRNA expression in locally-advanced vs metastatic patients, from the IMvigor210 dataset. p=0.55 by Wilcoxon test, n=316 patients.

CDCP1 mRNA is overexpressed in basal/squamous subtypes of bladder cancer, related to Figure 1A-D. Pairwise comparisons of CDCP1 expression between bladder cancer consensus subtypes using the Wilcoxon rank-sum test across four different datasets are shown.

CDCP1 mRNA expression is not associated with overall survival. Kaplan–Meier survival analyses were performed on quartiles of CDCP1 mRNA expression with overall survival as the endpoint. p > 0.05 in both public datasets for which overall survival data were available.

A plot showing the mean tumor volume for the vehicle and treated arms from the antitumor assessment study in mice bearing UMUC3 xenografts. The differences in mean between treatment arms was found to be significant (P < 0.001) using an unpaired, two tailed t test by day 20. Data are reported as mean ± standard deviation.