Fig 1.
Type 1 pili expression promotes optimal adhesion of ETEC to intestinal epithelia.
a. Transmission electron micrograph of ETEC H10407 expressing type 1 pili. The FimH tip adhesin was detected using α-FimH antibody and gold secondary antibody conjugate. b. Flow cytometric analysis of type 1 pili expression by ETEC H10407 and fimH mutants. c. Assessment of type 1 pili function using yeast agglutination assays. Negative yeast agglutination reflected the loss of type 1 pili activity. d. FimA immunoblot of type 1 pili extracts from static culture of WT ETEC, fimH mutants and mutants complemented with wild type fimH gene (pfimH). The fimH mutant complemented with a plasmid encoding a Q133K substitution in FimH is included as a negative control. e. Confocal microscopic images showing adhesion of WT ETEC, fimH mutants or complemented mutants to polarized cultured intestinal epithelia. Bacteria (anti-O78, green), cell membrane (CellMask, red), nuclei (DAPI, blue). f. Quantitative analysis was done by counting number of bacteria per focus area. Horizontal dashed lines represent geometric means of 3 combined individual experiments. P values were calculated by nonparametric Mann-Whitney test. *** indicates p<0.0001.
Fig 2.
Inhibition of type 1 pili mediated interaction impairs ETEC adhesion.
Yeast agglutination assay of WT ETEC (a) is inhibited by methyl-α-D-mannose (man) but not equimolar concentrations of methyl-α-D-galactose (gal) control sugar. b. Anti-FimH antibodies, generated against the lectin domain of FimH, but not pre-immune sera inhibit yeast agglutination. c. In vitro adhesion of WT ETEC or fimH mutants in the absence or presence of methyl-α-D-galactose or methyl-α-D-mannose. d. WT ETEC adhesion is inhibited by anti-FimH antibodies. The percentage of cell associated bacteria represents the proportion of bacteria associated with the CIE at the end of 1 h relative to the inoculum. Bars represent mean values + SEM (n = 5). P values were calculated by nonparametric Mann-Whitney test. * indicates p<0.05.
Fig 3.
FimH adhesin of ETEC interacts with intestinal epithelial cells.
a. Confocal microscopy images show binding of the biotinylated FimH lectin domain (FimHLD) or FimHLD:Q133K to the apical surface cultured intestinal epithelium (CIE). Biotinylated FimHLD was detected with streptavidin-conjugated fluorescent nanocrystals (Qdot, green); plasma membranes were stained with CellMask (red) and nuclei with DAPI (blue). Image at right shows three dimensional reconstruction of z stacks of CIE following interaction with FimHLD or the mutant protein. b. Quantitative analysis of FimHLD binding to CIE represented in panels a using Volocity three-dimensional (3D) image analysis software (version 6.2; PerkinElmer, Inc.). P value was calculated using nonparametric Mann-Whitney testing. c. Immunoelectron microscopy images of CIE infected with ETEC H10407. Left panel, microvilli structure at the apical surface of the CIE; right panels show immunogold labeling of FimH localized to the ETEC-host interacting surface.
Fig 4.
Enhanced presentation of mannosylated glycoproteins increases FimH binding and ETEC adhesion.
a. Confocal microscopy images detecting FimHLD binding to the kifunensine treated CIE. Biotinylated FimHLD was detected with streptavidin-conjugated fluorescent nanocrystals (Qdot, green) and nuclei with DAPI (blue). b. Quantitative analysis of FimHLD binding to CIE represented in panels a using Volocity three-dimensional (3D) image analysis software (version 6.2; PerkinElmer, Inc.). Data represent mean ± standard deviation of results of 3 independent experiments each with triplicate wells per concentration tested (n = 9). c. Confocal microscopic images showing ETEC adhesion to kifunensine treated CIE. The CIE grown on trans-well filters were treated with kifunensine and infected with WT ETEC or fimH mutants. One hour post infection wells were processed for microscopic examination. Bacteria (green), cell membrane (red), DAPI (blue). d. Quantitative analysis was done by counting number of bacteria present per focus area. Horizontal dashed lines represent geometric mean of 25 total data points combined from 2 replicate experiments. P values were calculated by nonparametric Mann-Whitney test. *** indicates p<0.0001.
Fig 5.
Properties of human enteroid-derived small intestinal monolayers.
a. Hematoxylin and eosin staining, DIC, and UEA1 lectin immunofluorescence confocal microscopic images showing formation of continuous monolayers. The apical surface is detected in the bottom image with FITC conjugated UEA1 lectin. b. UEA1 and anti-villin 1 antibody immunofluorescence in Laser Scanning Confocal Microscopy (LSCM) images from sections of polarized small intestinal enteroid monolayers. Nuclei (DAPI) are shown in blue. c. LSCM of sections showing MUC2 immunofluorescence (red) in probable goblet cell and co-localization with UEA1 in the merged image. d. Chromogranin A positive cells (red). e. Transmission electron microscopy (3000x) of polarized small intestinal enteroid monolayer sections showing a goblet cell flanked by enterocytes with distinct microvilli on the apical surface. Transwell filters in sections are indicated by *.
Fig 6.
Type 1 pili are required for optimal adhesion to small intestinal epithelia.
a. Representative laser scanning confocal microscopy (LSCM) images of sections prepared from human small intestinal enteroid-derived polarized monolayers infected with WT H10407 (red, anti-O78). Surface staining with the UEA1 lectin is shown in green. Nuclei are stained in blue (DAPI). b. Three dimensional reconstruction of LSCM z stacks of ETEC H10407 infected polarized monolayers showing the distribution of zonula occludens-1 (ZO-1, green) at the apical surface. Bacteria were visualized with anti-O78 (red) and nuclei are stained with DAPI (blue). c. Transmission electron microscopy images of ETEC H10407 adhering to microvilli on the surface of small intestinal monolayers (magnification 7500x and 15000x for left and right images respectively). d. LCSM images of wild type (wt) versus fimH mutant bacteria adherent to the surface of small intestinal enteroids Bacteria were visualized with anti-O78 (green) and cell membranes with CellMask (red), nuclei (DAPI, blue). e. Quantitative analysis of ETEC adhesion to enteroid-derived monolayers represented in panel d. Each dot plot represents adhesion data obtained using ileal cell derived from two individual subjects. Horizontal lines represent geometric means of data combined from 2 independent experiments. P values were calculated by nonparametric Mann-Whitney testing.
Fig 7.
Type 1 pili mediated interactions enhance toxin delivery.
a. Quantification of intracellular cGMP in infected cells. Cells were infected with WT ETEC in the absence or presence mannose sugar or with fimH or fimA mutants. Cells infected with estH/estP mutants which lack production of both heat stable toxins ST-H (ST-1b) and ST-P (ST-1a), represent basal level of cGMP in cells. b. Quantification of the amount of intracellular cAMP in infected cells. Cells were infected with WT ETEC in the absence or presence mannose sugar or with fimH or fimA mutants. Cells infected with eltAB, LT mutants, represent basal level of cAMP in cells. c. LT secretion by different mutants. Each bar represent mean with SEM (error bar) of 2 experiments consisting of 5 replicates per experiment for each strain. All P values were calculated by nonparametric Mann-Whitney test. *** p<0.0001.
Fig 8.
Type 1 pili are required for virulence in the rabbit ileal loop assay.
Type 1 pili are required for optimal bacterial engagement of rabbit intestinal epithelia. a. sections of rabbit ileum in which attached bacteria (green) are identified with anti-O78 (top panel) or anti-FimH (bottom panel). Nuclei are stained with DAPI (blue) and membranes are stained with CellMask (red). b. bacteria adherent to the ileal mucosal surface following infection with wild type ETEC H10407 or fimH and fimA mutants. c. Type 1 pili are required for toxicity in the rabbit ileal loop assay. Shown in the graph is the amount of fluid accumulation in each loop infected with WT or fimH or fimA mutants 18 h post inoculation. Loops infected with eltAB mutants or mock infected (PBS) were used as controls. Data represent the summary of experiments from 7 different rabbits (n = 7). Inset image shows infected ileal loops from one representative experiment. Each loop in the inset image is labeled with the infecting bacterial strain or with the negative control (PBS).
Table 1.
Relationship of functional type 1 pili expresion to colonization factors.