Structure and Function of Escherichia coli Type 1 Pili: New Insight into the Pathogenesis of Urinary Tract Infections

Transcription

1 S36 Structure and Function of Escherichia coli Type 1 Pili: New Insight into the Pathogenesis of Urinary Tract Infections Joel D. Schilling, Matthew A. Mulvey, and Scott J. Hultgren Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri The initial step in the establishment of a mucosal bacterial infection is the interaction of bacterial adhesive proteins with epithelial cells, an event that is often followed by invasion of the epithelial cell. Invasion of host cells has been proposed to provide bacterial pathogens with a means of escaping the harsh extracellular environment, where antibodies, complement, defensins, and other antibacterial molecules are abundant. Bacterial invasion can also facilitate the spread of microbes across and within tissue barriers, allowing for the dissemination of the infection. Uropathogenic Escherichia coli (UPEC), the primary causative agent of urinary tract infections (UTIs), has been presumed to be a predominantly extracellular pathogen. This concept has been challenged by recent studies demonstrating the ability of UPEC to invade bladder epithelial cells [1 3]. Two adhesive organelles associated with UPEC, the Dr adhesins and type 1 pili, have been linked to epithelial invasion by UPEC. This review will focus on the structure and assembly of type 1 pili and the mechanisms and consequences of type 1 pilus mediated invasion of bladder epithelial cells. Current Knowledge Type 1 pili. Type 1 pili are peritrichously expressed filamentous surface structures ranging from a few fractions of a micron to greater than 3 microns in length [4, 5]. These organelles are composite structures consisting of a 7-nm thick righthanded helical rod made up of repeating FimA subunits joined to a 3-nm thick distal tip fibrillum containing two adaptor proteins, FimG and FimF, and the adhesin, FimH (figure 1A) [5, 6]. The expression and assembly of type 1 pili require at least eight genes that are present in the type 1 pilus gene cluster (figure 1D). Donor strand complementation mechanism of pilus assembly. The assembly of type 1 pili proceeds by the highly conserved chaperone-usher pathway, which participates in the biogenesis of at least 30 diverse surface organelles in gram-negative bacterial pathogens [7 9]. Although the following discussion focuses on type 1 pili, significant insights into pilus biogenesis in Grant support: NIH (DK-51406, AI-29549). Reprints or correspondence: Dr. Scott Hultgren, Dept. of Molecular Microbiology, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, MO (Hultgren@borcim.wustl.edu). The Journal of Infectious Diseases 2001;183(Suppl 1):S by the Infectious Diseases Society of America. All rights reserved /2001/18305S-0010$02.00 general have also come from studies investigating the assembly of P pili, which are produced by most pyelonephritic strains of E. coli [10]. The FimC chaperone is part of the PapD superfamily of chaperones and consists of two immunoglobulin-like domains oriented in an L shape [8, 9]. The FimH adhesin also consists of two domains, a receptor-binding domain and a pilin domain (figure 1B). The receptor-binding domain, which does not interact significantly with the chaperone, is an elongated 11- stranded b-barrel with a putative mannose-binding pocket at its distal end. The pilin domain of FimH, like the pilin subunits, has an immunoglobulin fold that lacks the seventh (C-terminal) b-strand present in canonical immunoglobulin folds [11]. The absence of this strand produces a deep groove along the surface of the pilin domain and exposes its hydrophobic core. FimC and other PapD-like chaperones have three functions: They facilitate the folding of subunits as they emerge from the cytoplasmic membrane in semi-unfolded conformations, they stabilize subunits in the periplasm as soluble chaperone-subunit complexes, and they cap interactive surfaces on subunits to prevent their premature and nonproductive interaction in the periplasm. These three functions are carried out as part of a single mechanism called donor strand complementation. In the chaperone-subunit complexes, the G 1 strand of the chaperone and a portion of the F 1 -G 1 loop, which extends to lengthen the G 1 strand, complete the immunoglobulin fold of the subunit by occupying the groove (figure 1B) [12]. This interaction is thought to nucleate folding directly on the chaperone template. The folded subunit remains bound to the chaperone, stabilizing the subunit by shielding its hydrophobic core. The groove of the subunit is also directly involved in subunit-subunit interactions; thus, as long as the chaperone remains bound to the subunit, its interactive surface is capped. Once formed, chaperone-subunit complexes are specifically targeted to the outer-membrane usher, FimD, which has been shown to form a channel 2 3 nm in diameter, large enough to allow the passage of pilus subunits. The FimC-FimH complex binds more rapidly and tightly to the usher than do other chaperone-subunit complexes, presumably initiating pilus assembly. Formation of a chaperone-adhesin-usher ternary complex induces a conformational change in the usher to an assembly-competent form that is maintained throughout pilus assembly. As the pilus is assembled, the amino terminal strand of a neighboring subunit replaces the G 1 strand of the chaperone in a process called donor strand exchange [13], which

2 JID 2001;183 (Suppl 1) Structure and Function of Type 1 Pili S37 Figure 1. Structure and genetics of type 1 pili. A, High-resolution electron micrograph of a type 1 pilus, revealing the composite structure of this adhesive organelle. Arrowhead indicates the pilus tip containing the FimH adhesin. B, Crystal structure of the FimC chaperone in complex with the FimH adhesin. The receptor-binding domain of FimH is an 11-stranded elongated b-barrel with an overall jelly-roll topology, while the pilin domain of FimH consists of an immunoglobulin-like fold. The * indicates the mannose-binding pocket of FimH. Also shown is the insertion of the G 1 b-strand of the FimC chaperone into the hydrophobic groove formed between the A and F strands of the FimH pilin domain, an interaction known as donor-strand complementation. C, A model of FimH-subunit interactions demonstrating the insertion of the N-terminal extension of a subunit into the hydrophobic core of the FimH pilin domain. The mechanism by which a subunit replaces the chaperone to interact with this surface of the adhesin is called donor strand exchange. D, Genetic organization of the type 1 pilus gene cluster. appears to be facilitated by the usher. Thus, in the mature pilus, every subunit completes the immunoglobulin fold of its neighbor (figure 1C). The diameter of the usher pore suggests that the pilus grows through the usher as a linear fiber and only packages into its final quaternary structure upon reaching the bacterial surface. Subunits assembled into pili by FimC-like chaperones share critical features, indicating that donor strand complementation represents a general mechanism of protein interactions in pilus biogenesis [14]. FimH-mediated invasion into bladder epithelial cells. Until recently, the only identified function of type 1 pili has been to mediate adherence to host cells. However, recent studies have implicated type 1 pili in the invasion of bladder epithelial cells [1, 2]. Type 1 piliated strains of E. coli, such as the clinical stain NU14 and the K12 strain AAEC185/pSH2, can efficiently invade bladder epithelial cells in culture [1]. In contrast, type 1 piliated, FimH isogenic mutants of these strains, NU14-1 and AAEC185/pUT2002, cannot enter bladder epithelial cells. These results indicate a strong link between FimH and invasion of bladder epithelial cells. A more definitive study utilized latex beads coated with FimH in complex with the FimC chaperone. These beads are can efficiently enter bladder epithelial cells, whereas beads coated with bovine serum albumin or FimC alone are not efficiently internalized. Furthermore, bacterial adherence to bladder epithelial cells mediated by P pili does not induce bacterial uptake [1]. These data demonstrate that FimH is sufficient to mediate internalization in the absence of other bacterial factors and show that bacterial invasion mediated by type 1 pili is not simply the result of the nonspecific phagocytosis of adherent particles. The process of type 1 pilus mediated invasion occurs via a membrane zippering mechanism that involves host cell actin cytoskeletal rearrangements [1]. These rearrangements require host protein tyrosine phosphorylation and the activation of phosphoinositide 3-kinase (PI 3-kinase). Type 1 pilus mediated invasion also appears to involve the transient formation of

3 S38 Schilling et al. JID 2001;183 (Suppl 1) protein complexes between focal adhesin kinase and PI 3-kinase and between a-actinin and vinculin. These interactions likely facilitate and modulate the cytoskeletal rearrangements that lead to the envelopment and eventual internalization of adherent type 1 piliated bacteria. Invasion and innate host defenses. While the molecular players involved in type 1 pilus mediated bacterial invasion are being defined, the role of invasion in UTI pathogenesis is also beginning to crystallize. Shortly after transurethral inoculation of mouse or rat bladders with UPEC, bacteria can be found both entering and within the large superficial epithelial cells that line the lumenal surface of the bladder [2, 15, 16] (unpublished observations). Accumulations of intracellular bacteria have also been observed within bladder epithelial cells that were released into the urine in response to infection. On the basis of such observations, it was proposed that bladder epithelial cells could phagocytose UPEC and subsequently exfoliate as part of an innate host defense mechanism. Consistent with the hypothesis of exfoliation as an antibacterial host defense, the instillation of lipopolysaccharide (LPS) into the bladder has also been shown to induce the release of bladder epithelial cells in a mouse model [17, 18]. More recently, we have shown that type 1 piliated K12 and UPEC strains can trigger bladder cell exfoliation in C57BL/6 mice, whereas type 1 piliated, FimH strains do not elicit this response [2]. The release of bladder epithelial cells following infection with type 1 piliated E. coli occurs via an apoptosis-like mechanism [2]. Prior to the onset of exfoliation, DNA fragmentation, one of the hallmarks of apoptosis, can be detected within superficial epithelial cells in bladders infected with type 1 piliated E. coli. Furthermore, the treatment of infected mouse bladders with an inhibitor of caspases (cysteine proteases that mediate apoptotic programs) [19] impedes bladder cell exfoliation and significantly hinders bacterial clearance from the bladder [2]. Within the bladder, the precise role of type 1 pili, and specifically FimH, in triggering exfoliation of bladder epithelial cells is unclear. One scenario consistent with available data is that type 1 pilus mediated attachment and invasion facilitates the efficient delivery of bacteria-associated LPS to host epithelial cells. Subsequently, LPS interactions with host receptors may activate signaling pathways leading to the induction of apoptosis and exfoliation. However, bladder epithelial cells in LPS-hyporesponsive mice (C3H/HeJ mice) exfoliate in response to infection with type 1 piliated UPEC, although with significantly slower kinetics than in C57BL/6 mice, suggesting that other factors can also induce the release of infected bladder cells independent of LPS (unpublished data). Considering the invasive capacity of type 1 piliated UPEC and the involvement of an apoptosis-like mechanism in mediating bladder cell exfoliation, it is interesting to note that in other infectious disease systems, bacterial invasion has been associated with epithelial cell apoptosis [20]. Figure 2. Mechanisms and consequences of type 1 pilus mediated bacterial invasion. Type 1 pili facilitate adherence of uropathogenic Escherichia coli to superficial bladder epithelial cells. Interactions between FimH and bladder epithelial cells induce host signaling cascades, including the formation of protein complexes between focal adhesion kinase (FAK) and phosphatidylinositol 3-kinase (PI 3-kinase) and a-actinin and vinculin, that lead to cytoskeletal rearrangements and the envelopment of the attached bacterium. Invasion of bladder epithelial cells may trigger the production of proinflammatory cytokines/chemokines and epithelial cell apoptosis, leading to the recruitment of inflammatory cells and exfoliation, respectively. To prevent bacterial clearance by exfoliation of superficial bladder epithelial cells, and other innate host defenses, uropathogenic E. coli may invade the underlying bladder epithelial cells, potentially facilitating long-term persistence of bacteria within the bladder mucosa.

4 JID 2001;183 (Suppl 1) Structure and Function of Type 1 Pili S39 Invasion and bacterial persistence in the bladder. Interactions with superficial bladder epithelial cells thus present UPEC with a challenging situation. While providing UPEC with a temporary reprieve from the harsh environment of the urine and numerous extracellular antimicrobial factors [21], the uptake of UPEC by superficial bladder cells may also serve to entrap and contain invading bacteria. From this perspective, the superficial bladder cells could be viewed as a sink, collecting and storing UPEC for later disposal via exfoliation followed by micturition. In addition, it has been demonstrated that bacterial invasion of bladder cells can stimulate the induction of epithelial cell cytokine and chemokine production, an important component of mucosal host defenses [22]. Thus, type 1 pilus mediated bacterial invasion could potentially signal the beginning of the end for UPEC within the urinary tract. However, this does not appear to be the case. Between 24 and 48 h after inoculation of mouse bladders with type 1 piliated UPEC, the majority of bacteria that continue to persist within the bladder are intracellular, as determined by ex vivo gentamicin protection assays in which extracellular bacteria are specifically eliminated [2]. This observation indicates that bladder cell invasion by UPEC provides the bacteria with a survival advantage within the urinary tract. Since infected superficial bladder epithelial cells eventually slough off into the urine, carrying with them any internalized bacteria, UPEC must have means of escaping from within dying superficial cells before completion of exfoliation in order to persist within the bladder. Prior to exfoliation, intracellular bacterial replication and efflux from infected superficial bladder cells are pathogenic events that could increase the likelihood that UPEC successfully evades the exfoliation response and colonizes other bladder epithelial cells. In essence, from the perspective of incoming UPEC, the superficial bladder cells may serve as a temporary beachhead, providing a passageway to surrounding superficial cells and to the less accessible underlying bladder epithelium. The underlying epithelial cells may provide a more stable environment for the long-term persistence of UPEC within the bladder as well as protect UPEC from potentially detrimental interactions with soluble antimicrobial products within the urine, including antibiotics. Indeed, recent studies have demonstrated that UPEC can persist within the bladder tissue even in the face of antibiotic treatments that effectively sterilize the urine [23, 24]. Commentary These observations indicating that UPEC is not strictly an extracellular pathogen may directly impact the ways in which UTIs are detected and treated. The ability of type 1 piliated UPEC to invade the bladder epithelium and persist, despite the presence of sterile urine, may also provide an explanation for the recurrent nature of UTIs. A significant number of recurrent UTIs are caused by the same bacterial strain isolated from the original infection [25], suggesting that the reemergence of intracellular UPEC from within the underlying bladder epithelium is a potential etiology for some cases of recurrent UTI. Studies are currently underway to address these issues. Figure 2 depicts a model of type 1 pilus mediated bacterial invasion of bladder epithelial cells and its consequences in UTI pathogenesis. It is hoped that the elucidation of the signaling events leading to type 1 pilus mediated invasion and the determination of the consequences of bacterial internalization, in concert with detailed studies of pilus assembly, will eventually lead to new and better approaches for treating UTIs. Acknowledgments We thank F. Sauer, C. Hung, M. Chapman, J. Martinez, and C. Vincent for their assistance in assembling this manuscript. References 1. Martinez JJ, Mulvey MA, Schilling JD, Pinkner JS, Hultgren SJ. Type 1 pilus mediated bacterial invasion of bladder epithelial cells. EMBO J 2000;19: Mulvey MA, Lopez-Boado YS, Wilson CL, et al. Induction and evasion of host defenses by type 1 piliated uropathogenic Escherichia coli. Science 1998;282: (published erratum appears in Science 1999;283:795). 3. Goluszko P, Selvarangan R, Popov V, Pham T, Wen JW, Singhal J. Decayaccelerating factor and cytoskeleton redistribution pattern in HeLa cells infected with recombinant Escherichia coli strains expressing Dr family of adhesins. Infect Immun 1999;67: Brinton CC Jr. The structure, function, synthesis and genetic control of bacterial pili and a molecular model for DNA and RNA transport in gram negative bacteria. Trans N Y Acad Sci 1965;27: Jones CH, Pinkner JS, Roth R, et al. FimH adhesin of type 1 pili is assembled into a fibrillar tip structure in the Enterobacteriaceae. Proc Natl Acad Sci USA 1995;92: Russell PW, Orndorff PE. Lesions in two Escherichia coli type 1 pilus genes alter pilus number and length without affecting receptor binding. J Bacteriol 1992;174: Thanassi DG, Saulino ET, Hultgren SJ. The chaperone/usher pathway: a major terminal branch of the general secretory pathway. Curr Opin Microbiol 1998;1: Sauer FG, Knight SD, Waksman GJ, Hultgren SJ. PapD-like chaperones and pilus biogenesis. Semin Cell Dev Biol 2000;11: Hung DL, Hultgren SJ. Pilus biogenesis via the chaperone/usher pathway: an integration of structure and function. J Struct Biol 1998;124: Roberts JA, Marklund BI, Ilver D, et al. The Gal(alpha 1-4)Gal-specific tip adhesin of Escherichia coli P fimbriae is needed for pyelonephritis to occur in the normal urinary tract. Proc Natl Acad Sci USA 1994;91: Choudhury D, Thompson A, Stojanoff V, et al. x-ray structure of the FimC- FimH chaperone-adhesin complex from uropathogenic Escherichia coli [see comments]. Science 1999;285: Sauer FG, Futterer K, Pinkner JS, Dodson KW, Hultgren SJ, Waksman G. Structural basis of chaperone function and pilus biogenesis [see comments]. Science 1999;285: Barnhart MM, Pinkner JS, Soto GE, et al. From the cover: PapD-like chaperones provide the missing information for folding of pilin proteins [see comments]. Proc Natl Acad Sci USA 2000;97: Soto GE, Hultgren SJ. Bacterial adhesins: common themes and variations in architecture and assembly. J Bacteriol 1999;181: Fukushi Y, Orikasa S, Kagayama M. An electron microscopic study of the interaction between vesical epithelium and E. coli. Invest Urol 1979;17: McTaggart LA, Rigby RC, Elliott TS. The pathogenesis of urinary tract

5 S40 Schilling et al. JID 2001;183 (Suppl 1) infections associated with Escherichia coli, Staphylococcus saprophyticus and S. epidermidis. J Med Microbiol 1990;32: Aronson M, Medalia O, Amichay D, Nativ O. Endotoxin-induced shedding of viable uroepithelial cells is an antimicrobial defense mechanism. Infect Immun 1988;56: Jezernik K, Medalia O, Aronson M. A comparative study of the desquamation of urothelial cells during gestation and in adult mice following moderate stress or endotoxin treatment. Cell Biol Int 1995;19: Cohen GM. Caspases: the executioners of apoptosis. Biochem J 1997;326: Kim JM, Eckmann L, Savidge TC, Lowe DC, Witthoft T, Kagnoff MF. Apoptosis of human intestinal epithelial cells after bacterial invasion. J Clin Invest 1998;102: Sobel JD. Pathogenesis of urinary tract infection. Role of host defenses. Infect Dis Clin North Am 1997;11: Schilling JD, Mulvey MA, Vincent CV, Lorenz RG, Hultgren SJ. Bacterial invasion augments epithelial responsiveness to Escherichia coli through an LPS-dependent mechanism. J Immunol (in press). 23. Hvidberg H, Struve C, Krogfelt KA, Christensen N, Rasmussen SN, Frimodt-Moller N. Development of a long-term ascending urinary tract infection mouse model for antibiotic treatment studies. Antimicrob Agents Chemother 2000;44: Mulvey MA, Schilling JD, Martinez JJ, Hultgren SJ. From the cover: bad bugs and beleaguered bladders: interplay between uropathogenic Escherichia coli and innate host defenses [in process citation]. Proc Natl Acad Sci USA 2000;97: Stapleton A, Stamm WE. Prevention of urinary tract infection. Infect Dis Clin North Am 1997;11: