Bad bugs and beleaguered bladders: interplay between uropathogenic Escherichia coli and innate host defenses

Abstract

Strains of uropathogenic Escherichia coli (UPEC) are the causative agents in the vast majority of all urinary tract infections. Upon entering the urinary tract, UPEC strains face a formidable array of host defenses, including the flow of urine and a panoply of antimicrobial factors. To gain an initial foothold within the bladder, most UPEC strains encode filamentous surface adhesive organelles called type 1 pili that can mediate bacterial attachment to, and invasion of, bladder epithelial cells. Invasion provides UPEC with a protective environment in which bacteria can either replicate or persist in a quiescent state. Infection with type 1-piliated E. coli can trigger a number of host responses, including cytokine production, inflammation, and the exfoliation of infected bladder epithelial cells. Despite numerous host defenses and even antibiotic treatments that can effectively sterilize the urine, recent studies demonstrate that uropathogens can persist within the bladder tissue. These bacteria may serve as a reservoir for recurrent infections, a common problem affecting millions each year.

Figures

Figure 1

Type 1 pilus architecture and crystal structure of the FimH adhesin. (Left) High-resolution EM reveals the composite structure of the type 1 pilus. The pilus tip containing the FimH adhesin is indicated. (Right) FimH has two domains, each with Ig-like folds. The adhesin domain has a binding pocket (arrow) that can accommodate d-mannose.

Figure 2

Type 1 pilus-mediated bacterial attachment to the bladder epithelium. After inoculation of C57BL/6 mice with type 1-piliated UPEC, numerous bacteria (yellow) can be found attached to the luminal surface of the bladder (blue) as detected by scanning EM (A) and high-resolution freeze-dry/deep-etch EM (B). Type 1 pili mediating bacterial attachment were resolved with the high-resolution technique. The scalloped appearance of the bladder surface is attributable to the presence of the uroplakin plaques (≈0.5 μm in diameter). [Bars = 3 μm (A) and 0.5 μm (B).]

Figure 3

Internalization of type 1-piliated UPEC by superficial facet cells. (A–F) Transmission EM shows the AUM of mouse superficial facet cells in various stages of enveloping adherent type 1-piliated UPEC. (F and G) High-resolution freeze-fracture/deep-etch EM also reveals type 1-piliated E. coli (yellow) seemingly being enveloped by facet cells (brown; the AUM and extracellular milieu are colored blue). The cytoplasmic face of three uroplakin plaques, separated by smooth interplaque regions, can be seen in the host membrane that is partially enveloping the bacterium in G. All images were obtained by using C57BL/6 mouse bladders recovered about 1 h after infection with UPEC. [Bars = 1 μm (A and B) and 0.5 μm (C–G).]

Figure 4

FimH-mediated internalization of latex beads. Native FimH can be purified from bacteria in complex with the type 1 pilus chaperone FimC (14). Transmission EM (and other data not shown) demonstrate that latex beads coated with FimC-FimH complexes are readily internalized by bladder epithelial cells (5637 cells) grown in culture. Beads coated with FimC alone or with BSA are not internalized (23). The envelopment of FimC-FimH-coated beads by 5637 cells is similar in appearance to the envelopment of type 1-piliated E. coli by bladder epithelial cells. (Bar = 1 μm.)

Figure 5

Exfoliation of bladder epithelial cells. (A) The luminal surface of the bladder is normally covered by extremely long-lived, highly differentiated superficial facet cells with distinctive pentagonal or hexagonal outlines. These large cells (ranging from 20 to 150 μm in diameter) are often bi- or multinucleate and can be easily identified on the surface of whole-mount bladders stained with Hoechst dye. (B) Within 6 h after inoculation with type 1-piliated E. coli, many infected facet cells in C57BL/6 mice exfoliate and are rinsed away, revealing the smaller, mono-nucleate underlying epithelial cells. (C) A stained paraffin section from an infected mouse bladder shows a facet cell in the process of exfoliating and taking a large mass of adherent E. coli with it. Host cell nuclei were stained with Hoechst dye whereas bacteria were stained red by using anti-E. coli primary and Cy3-labeled secondary antibodies. [Bars = 100 μm (A and B) and 10 μm (C).]

Figure 6

Neutrophil influx into the urothelium in response to infection. (A) Paraffin sections of C57BL/6 mouse bladders recovered 6 h after infection with type 1-piliated UPEC and stained with hematoxylin and eosin show PMNs (small, darkly stained cells) migrating from blood vessels within the lamina propria and into the urothelium. (B) PMNs appeared to aggregate beneath the luminal surface of the bladder and could occasionally be seen, by scanning EM, emerging in the vicinity of adherent bacteria on the surface of newly exposed immature urothelial cells (C). PMNs were also found associated with infected facet cells in the process of exfoliating (D). [Bars = 20 μm (A), 10 μm (B), 50 μm (C), and 30 μm (D).]

Figure 7

Resistance of intracellular UPEC to antibiotic treatment. C57BL/6 mice were infected with 1 × 108 colony-forming units of the clinical cystitis isolate NU14. Six hours after infection, mice were given trimethoprim-sulfamethoxazole (54 μg/ml and 270 μg/ml, respectively) in their drinking water or were left untreated. Mice were provided fresh water with or without antibiotics daily, and at 3 days after inoculation bladders were recovered and bisected. Bladder halves were treated with either gentamicin (to kill any extracellular bacteria) or were left untreated before homogenization and determination of bacterial titers. Control experiments (not shown) demonstrated that trimethoprim-sulfamethoxazole completely inhibited the growth of NU14 in vitro.

Figure 8

Interplay between innate host defenses and UPEC within the bladder. The flow of urine in addition to a variety of host factors that can act as soluble receptor analogues for type 1 pili can impede contact between type 1-piliated UPEC and host superficial facet cells. If contact is established, FimH-receptor interactions can trigger the internalization of adherent bacteria into facet cells, in which UPEC can replicate to high levels. However, attachment and/or invasion can result in the activation of apoptotic pathways within facet cells, leading to the eventual exfoliation and clearance of infected host cells. The release of infected bladder cells in urine may facilitate the spread of UPEC strains in the environment. Initial interactions between type 1-piliated E. coli and urothelial cells can also result in the induction of pro-inflammatory cytokines, leading to the influx of PMNs into the urothelium. To avoid clearance by exfoliation, UPEC is able to escape from dying facet cells and can go on to infect surrounding and underlying epithelial cells. These bacteria may eventually be able to enter a niche within the urothelium in which they can persist (at subclinical levels) undetected by immunosurveillance mechanisms.