Skip to main content
. 2017 Nov 7;6:e29132. doi: 10.7554/eLife.29132

Figure 1. HIOs can be stably associated with non-pathogenic E. coli.

(A) Mean fluorescent intensity of a human intestinal organoid (HIO) containing live GFP+E. coli str. ECOR2. The lower panels show representative images from the time series. Representative of three independent experiments. Video 1 is an animation corresponding to this dataset. (B) Confocal micrograph of the HIO epithelium (E-cadherin) in direct association with GFP+ E. coli at 48 hr post-microinjection with 104 live E. coli. 60X magnification. (C) Luminal CFU per HIO E. coli at 24 hr post-microinjection relative to the injected concentration of 5 × 10-1 to 5 × 105 CFU per HIO at the start of the experiment. N = 10 biological replicates per E. coli dose. The r2 and P value shown in the figure represent the results of a linear regression analysis of the relationship between the 24 hr change in CFU/HIO and the initial number of CFU injected. (D) Luminal CFU per HIO at 0–72 hr following microinjection with 10 CFU E. coli per HIO. N = 13–17 replicate HIOs per time point. The p-value represents the results of a two-tailed Student’s t-test comparing the two conditions indicated. (E) Daily proportion of HIO cultures with no culturable E. coli in the external media following E. coli microinjection (N = 48) or PBS microinjection (N = 8).

Figure 1.

Figure 1—figure supplement 1. Pluripotent stem-cell-derived intestinal epithelium transcriptionally resembles the immature human intestinal epithelium.

Figure 1—figure supplement 1.

(A) Pearson’s correlation matrix with heirarchical clustering forwhole-transcriptome normalized RNA-seq gene counts from epithelium isolated from the tissuesindicated on the axes. (B) Pearson’s correlation coefficient for the comparison of whole-transcriptome normalized RNA-seq gene counts between each of the sample types listed onthe x-axis and adult small intestinal epithelium. p-Value indicates the results of an unpaired two-sided Student’s t-test. (C) Principle component analysis of whole-transcriptome RNA-seq normalized gene counts. Cumulative explained variance for PC1 and PC2 is indicated as a percentage on the x- and y-axes, respectively. (D) Density plot of the Log2 -transformed Fold change in gene expression in epithelium from transplanted HIOs over epithelium from HIOs cultured in vitro plotted against the Log 2 -transformed Fold change in gene expression in adult small intestinal epithelium over fetal small intestinal epithelium. The intensity of the blue color indicates the density of points in 2-dimensional space. SI, small intestine; tx, transplanted tissue; hPSC, human pluripotent stem cell; HIO, human intestinal organoid.
Figure 1—figure supplement 2. Phylogenetic tree based on maximum liklihood genomic distance among E.coli str.

Figure 1—figure supplement 2.

ECOR2 (Ochman and Selander, 1984), the strain used in the HIO colonization experiments, closely related E. coli isolates available on the PATRIC (Wattam et al., 2017) database, and pathogenic type strains from the genera Esherichia, Shigella, and Salmonella. A PATRIC genome reference number follows name of each taxa.
Figure 1—figure supplement 3. HIO colonization protocol.

Figure 1—figure supplement 3.

(A) Schematic representation of the microinjection of HIOs with live E. coli. See Materials and Methods for additional details. (B) A comparison of CFU/HIO at 24 hpost-microinjection of HIOs microinjected with 10 × 103 CFU live E. coli diluted in sterile PBS or fresh LB broth. N = 5 HIOs per condition. All experiments presented in the main paper represent E. coli diluted in PBS. (C) To test for the effect of antibiotic carryover of E. coli growth in the HIO lumen, we compared CFU/HIO in HIOs cultured in antibiotic-free media ('Remove abx') or media containing penicillin and streptomycin at 24 hr post-microinjection with 10 3 CFU (colony forming units) live E. coli. All experiments presented in the main paper represent HIOs cultured in antibiotic-free media due to the apparent effect of antibiotics in suppressing E. coli growth within the HIO lumen. N = 5 HIOs per condition. (D) In several experiments, bacterial translocation was measured by samplign the external HIO culture media (Figure 1 and 8). To evaluate the potential influence of antibiotic carryover in our estimates of bacterial translocation, we measured growth inhibition in E. coli cultures plated as a lawn of LB agar and treated with 1 μl samples of HIO media collected during the 1 hr antibiotic wash step or after HIO culture washout with PBS and replacement with fresh antibiotic-free media (see panel A). N = 6 replicates per culture condition. All p-values represent the results of unpaired one-tailed Student’s t-test comparisons.