Functional heterogeneity of the UpaH autotransporter protein from uropathogenic Escherichia coli

Abstract

Uropathogenic Escherichia coli (UPEC) is responsible for the majority of urinary tract infections (UTI). To cause a UTI, UPEC must adhere to the epithelial cells of the urinary tract and overcome the shear flow forces of urine. This function is mediated primarily by fimbrial adhesins, which mediate specific attachment to host cell receptors. Another group of adhesins that contributes to UPEC-mediated UTI is autotransporter (AT) proteins. AT proteins possess a range of virulence properties, such as adherence, aggregation, invasion, and biofilm formation. One recently characterized AT protein of UPEC is UpaH, a large adhesin-involved-in-diffuse-adherence (AIDA-I)-type AT protein that contributes to biofilm formation and bladder colonization. In this study we characterized a series of naturally occurring variants of UpaH. We demonstrate that extensive sequence variation exists within the passenger-encoding domain of UpaH variants from different UPEC strains. This sequence variation is associated with functional heterogeneity with respect to the ability of UpaH to mediate biofilm formation. In contrast, all of the UpaH variants examined retained a conserved ability to mediate binding to extracellular matrix (ECM) proteins. Bioinformatic analysis of the UpaH passenger domain identified a conserved region (UpaH(CR)) and a hydrophobic region (UpaH(HR)). Deletion of these domains reduced biofilm formation but not the binding to ECM proteins. Despite variation in the upaH sequence, the transcription of upaH was repressed by a conserved mechanism involving the global regulator H-NS, and mutation of the hns gene relieved this repression. Overall, our findings shed new light on the regulation and functions of the UpaH AT protein.

Fig 1

Evolutionary relationships and genome context of UpaH. The evolutionary history of UpaH (red) was inferred using the neighbor-joining method, and its genome context was compared in 19 E. coli strains. The bootstrap consensus tree was drawn to scale, and the evolutionary distances were computed using the p-distance method in MEGA5 (70). The genome context and gene variability were evaluated using complete and annotated E. coli genome sequences in Easyfig (67). Core (black) and variable (gray) genes are compared to those in the upaH-negative E. coli strain IAI1.

Fig 2

(A) Western blot analysis of UpaH production in whole-cell lysates prepared from E. coli strains CFT073, M161, M357, M369, and IA2 and their respective upaH or hns mutants. (B) Schematic illustration of the domain organization of UpaH from E. coli. Indicated are the signal peptide, the small repeat region, the longer repeat region, and the transmembrane β-domain. The repeat regions contain the majority of the sequence differences identified in the UpaH variants.

Fig 3

UpaH variants mediate different levels of biofilm formation in E. coli. (A) Overhead-view and (B) 3D-reconstructed fluorescence micrographs of biofilms formed under continuous flow in the dynamic-flow-chamber assay by E. coli OS56 strains expressing different UpaH variants (pUpaH^CFT073, pUpaH^M161, pUpaH^M357, pUpaH^M369, pUpaH^IA2, or pBAD). Biofilm development was monitored via scanning confocal laser microscopy of the GFP-tagged E. coli strain OS56 cells induced with arabinose. The images in panel A are representative horizontal sections collected within each biofilm and vertical sections (the right panel and above each larger panel, representing the yz plane and the xz plane, respectively). Under 24 h of continuous flow, UpaH variants promoted biofilm growth with significant differences in level of biovolume (P < 0.001), substratum coverage (P = 0.002), vertical spreading (P = 0.001), surface area-to-biovolume ratio (P < 0.001), mean thickness (P = 0.005), and biofilm roughness (P = 0.001). Measurements were analyzed using COMSTAT and the Kruskal-Wallis test. These results demonstrate that UpaH variants resulted in different levels of biofilm formation when expressed from the same plasmid and in the same host background. (C) UpaH variants that mediate E. coli binding to MaxGel, collagen V, fibronectin, and laminin in ELISA-based binding to the same levels. The bar charts represent the average absorbance measurements at 405 nm of 4 independent experiments plus the standard errors of the mean (SEM). The mean absorbance readings were compared with negative control readings for OS56(pBAD).

Fig 4

(A) Schematic illustration of the domain organization of UpaH, UpaH^HR, and UpaH^CR and a summary of phenotypic results. Indicated are the signal peptide, the hydrophobic region, the conserved region, the small repeat region, the longer repeat region, and the transmembrane β-domain. (B) Immunofluorescence (top) or phase-contrast (bottom) microscopy employing UpaH-specific antiserum against the cells of the specified E. coli strains. Strains were grown in the presence of 0.2% arabinose. The overnight cultures were fixed and incubated with anti-UpaH serum followed by incubation with goat anti-rabbit IgG coupled to Alexa Fluor 350. Positive reactions indicating surface localization of UpaH, UpaH^HR, and UpaH^CR were detected. (C) UpaH, UpaH^HR, and UpaH^CR mediate E. coli binding to MaxGel, collagen V, fibronectin, and laminin in ELISA-based binding to the same levels. The bar charts represent average absorbance measurements at 405 nm of 3 independent experiments plus the SEM. The mean absorbance readings were compared with negative control readings from strain OS56(pBAD). (D) UpaH mutants deleted for the hydrophobic or conserved region that mediates different levels of biofilm formation in E. coli. Shown are the overhead views of biofilms formed under continuous flow by E. coli OS56 strains expressing pUpaH^CFT073, pUpaH^HR, pUpaH^CR, or pBAD. Biofilm development was monitored via scanning confocal laser microscopy of the GFP-tagged E. coli strain OS56 cells. The images are representative horizontal sections collected within each biofilm and vertical sections (the right panel and above each larger panel, representing the yz plane and the xz plane, respectively). Under 24 h of continuous flow, the UpaH mutants affected biofilm growth with significant differences in level of biovolume (P = 0.001), substratum coverage (P = 0.003), vertical spreading (P = 0.001), mean thickness (P = 0.002), and biofilm roughness (P = 0.001). Measurements were analyzed using COMSTAT and the Kruskal-Wallis test. These results demonstrate that expression of pUpaH^CFT073, pUpaH^HR, and pUpaH^CR resulted in different levels of biofilm formation when expressed from the same plasmid and in the same host background.

Fig 5

Evolutionary relationships of UpaH. The evolutionary relationship of UpaH was inferred using the N-terminal 300 amino acids of 143 upaH-positive E. coli strains. The neighbor-joining method was used, and the bootstrap consensus tree was taken to represent the evolutionary history of the taxa analyzed in MEGA5 (70). The branches corresponding to partitions reproduced in less than 50% of the bootstrap replicates are collapsed. The tree is drawn to scale, and the distances were computed using the p-distance method. The E. coli strains are color coded according to their pathotypes, and the UPEC strains used in this study are indicated by an asterisk.

Fig 6

H-NS represses upaH expression. (A) β-Galactosidase assay of upaH::lacZ-zeo reporter fusions in strain CFT073, isogenic hns deletion mutant, complemented mutant, and complement vector control. (B) Electrophoretic band shift of the amplified 250-bp upaH promoter and the bla promoter from TaqI-SspI-digested pBR322 DNA in the presence of increasing concentrations of H-NS (0.05, 0.1, 0.2, 0.5, 1, and 2 μM). The pBR322 fragments not containing the bla-promoter (313/315 bp and 475 bp) were not shifted by H-NS. (C) Curvature-propensity plot of the sequence (250 bp) upstream of the upaH gene from CFT073(PupaH^CFT073) and M369(PupaH^M369) showing the predicted regions of curved DNA. (D) Curved DNA PAGE gel of the amplified 250-bp promoters PupaH^CFT073 and PupaH^M369.