Göran Bergsten

Neutrophil recruitment, chemokine receptors, and resistance to mucosal infection

Neutrophil migration to infected mucosal sites involves a series of complex interactions with molecules in the lamina propria and at the epithelial barrier. Much attention has focussed on the vascular compartment and endothelial cells, but less is known about the molecular determinants of neutrophil behavior in the periphery. We have studied urinary tract infections (UTIs) to determine the events that initiate neutrophil recruitment and interactions of the recruited neutrophils with the mucosal barrier. Bacteria activate a chemokine response in uroepithelial cells, and the chemokine repertoire depends on the bacterial virulence factors and on the specific signaling pathways that they activate. In addition, epithelial chemokine receptor expression is enhanced. Interleukin (IL)‐8 and CXCR1 direct neutrophil migration across the epithelial barrier into the lumen. Indeed, mIL‐8Rh knockout mice showed impaired transepithelial neutrophil migration, with tissue accumulation of neutrophils, and these mice developed renal scarring. They had a defective antibacterial defense and developed acute pyelonephritis with bacteremia. Low CXCR1 expression was also detected in children with acute pyelonephritis. These results demonstrate that chemokines and chemokine receptors are essential to orchestrate a functional antimicrobial defense of the urinary tract mucosa. Mutational inactivation of the IL‐8R caused both acute disease and chronic tissue damage.

P fimbriae enhance the early establishment of Escherichia coli in the human urinary tract.

This study examined the role of P fimbriae in the establishment of bacteriuria. Patients (n = 17) were subjected to intravesical inoculation with an asymptomatic bacteriuria strain, Escherichia coli 83972, or its P‐fimbriated (pap+/prs+) transformants. As shown by groupwise analysis, the pap+/prs+ transformants established bacteriuria more rapidly than E. coli 83972 (P = 0.021) and required a lower number of inoculations to reach 105 cfu ml−1 (P = 0.018). Intraindividual analysis showed that the pap+/prs+ transformants established bacteriuria more rapidly than E. coli 83972 in the patients who subsequently became carriers of both strains. Finally, bacterial establishment was shown to vary with the in vivo expression of P fimbriae. Bacterial counts were higher when P‐fimbrial expression was detected than when the pap+/prs+ strain showed a negative phenotype. The results suggested that P fimbriae enhance the establishment of bacteriuria and fulfil the molecular Koch postulates as a colonization factor in the human urinary tract.

P-fimbriae trigger mucosal responses to Escherichia coli in the human urinary tract.

Uropathogenic Escherichia coli elicit a host response that determines the severity of urinary tract infection (UTI). Specific adherence mechanisms allow the bacteria to initiate this process by targeting epithelial cells in the urinary tract mucosa. Epidemiological studies show a strong association of P‐fimbriae with disease severity, suggesting that adherence mediated by these organelles has a direct effect on mucosal inflammation in vivo. The present study examined the ability of P‐fimbriae to induce inflammation in the human urinary tract. Patients were subjected to intravesical inoculation with a non‐fimbriated E. coli strain or transformants of this strain expressing P‐fimbriae. The inflammatory response was analysed as a function of P‐fimbrial expression. The P‐fimbriated transformants invariably caused higher interleukin (IL)‐8, IL‐6 and neutrophil responses in the urinary tract than the ABU strain. Furthermore, loss of P‐fimbrial expression in vivo was accompanied by a return to background levels of neutrophils, IL‐6 and IL‐8 in individual patients. The results demonstrate that the pap sequences confer on a non‐fimbriated, avirulent strain the ability to induce a host response in the human urinary tract. P‐fimbriae thus fulfil the ‘molecular Koch–Henle postulates’ linking a single virulence factor to host response induction.

PapG-Dependent Adherence Breaks Mucosal Inertia and Triggers the Innate Host Response

Mucosal pathogens differ from normal flora constituents in that they provoke a host response that upsets mucosal integrity. We investigated whether the elaboration of discrete adherence factors is sufficient to break the inertia of the mucosal barrier. PapG-mediated adherence was selected as an example, because P fimbrial expression characterizes uropathogenic Escherichia coli and because adherence starts the attack on the mucosal barrier. Patients were inoculated intravesically with transformed nonvirulent E. coli strains expressing functional P fimbriae (E. coli pap(+)) or mutant fimbriae lacking the adhesin (E. coli Delta papG). E. coli pap(+) was shown to activate the innate host response, and adherent gfp(+) bacteria were observed on excreted uroepithelial cells. E. coli Delta papG failed to trigger a response and was nonadhesive. We conclude that PapG-mediated adherence breaks mucosal inertia in the human urinary tract by triggering innate immunity and propose that this activation step differentiates asymptomatic carriage from infection.

The role of P fimbriae for Escherichia coli establishment and mucosal inflammation in the human urinary tract.

Bacterial adhesion to the bladder mucosa is a critical step for the establishment of Escherichia coli bacteriuria. The P-fimbriae, encoded by the pap gene cluster, are considered as virulence factors but the mechanisms have been debated. This study defined the roles for P fimbriation during the early colonization of the human urinary tract. Patients with recurrent UTI were first subjected to deliberate colonization with the non-fimbriated ABU strain E. coli 83972. Bacteriuria was established long term (1-4 years) in patients with dysfunctional bladders, but not in the patients with normal bladder function. Super-infections were transient and asymptomatic. P fimbriated transformants of the ABU strain (E. coli 83972pap+/prs+) reached 105 CFU/ml more rapidly than E. coli 83972 and the vector control. This was demonstrated by group wise and intra-individual analysis in patients colonized on different occasions with E. coli 83972 or the P fimbriated transformants. Higher neutrophil numbers and IL-8 and IL-6 concentrations in urine were obtained after colonization with the P fimbriated transformants. These results demonstrated that transformation of E. coli 83972 with the pap sequences is sufficient to convert it to a more potent host response inducer. The P fimbriae were shown to lower the significant bacteriuria threshold. The P fimbriated transformants needed lower bacterial numbers (103-4 CFU/ml) to predict a positive second urine culture with a >80% accuracy and to trigger a significant host response. These studies show that P fimbriae fulfil the Koch Henles molecular postulates for bacterial establishment and host response induction in the human urinary tract.

The host response to urinary tract infection.

The authors use the UTI model to identify basic mechanisms of disease pathogenesis, host response induction, and defense. Their studies hold the promise to provide a molecular and genetic explanation for susceptibility to UTI, and to offer more precise tools for diagnosis and therapy of these infections. There are few infections where the host response is understood in such detail and where pathologic host responses can be linked to distinct disease states. The susceptibility to UTI varies greatly in the population. The studies suggest that distinct molecular defects can cause the clinical entity of acute pyelonephritis with renal scarring, and suggest that the susceptibility to UTI in certain patient groups may have a genetic basis. In addition, the distinct signal transduction pathways explain the development of symptoms, and propose that defects in those signaling mechanisms may occur in patients with ABU. In the future, it may be useful to include these host response parameters in the diagnostic arsenal, to help in early detection of patients susceptible to recurrent UTI and renal scarring. These patients may then be offered therapies that strengthen their defense, and be offered close surveillance for recurrences and other complications.

The ‘innate’ host response protects and damages the infected urinary tract

Symptoms of infection and tissue pathology are caused by the host response: not by the microbe per se. The same response is also critical for the defence and is needed to clear infection. It is therefore essential to understand how the host response is activated and to identify the critical effector mechanisms of the defence. We have studied these issues in the urinary tract infection (UTI) model. The symptoms of UTI and the host defence both rely on the so-called ‘innate’ immune system, making this one of the best characterized human disease models of ‘innate immunity’. We discuss the critical molecular events that determine whether the host response will be activated by P-fimbriated uropathogenic Escherichio coli as well as factors determining whether the patient develops acute pyelonephritis or asymptomatic bacteriuria. We will describe the glycoconjugate receptors used by the P-fimbriated bacteria adhering to host tissues, the recruitment of TLR4 co-receptors and the signalling pathways that allow progression to symptomatic disease, and discuss how these mechanisms are altered in asymptomatic carriers, presenting the possible genetic basis for unresponsiveness. We have shown that neutrophils are the critical effectors of the host defence and that neutrophil dysfunctions lead to acute pyelonephritis and renal scarring. Here we discuss the mechanisms of neutrophil-mediated, chemokine receptor (CXCR1)-dependent clearance, and the defect in inter-leukin-8 receptor homolog knock-out (IL-8Rh KO) mice and describe the data linking low CXCRI expression to recurrent pyelonephritis in man, as well as the information on the genetic basis for low CXCR1 expression in affected patients. Finally, the mechanisms of renal scarring in IL8Rh KO mice will be discussed in relation to human disease. Our studies hold the promise to provide a molecular and genetic explanation for disease susceptibility in some patients with UTI and to offer more precise tools for the diagnosis and therapy of these infections.

Do type 1 fimbriae promote inflammation in the human urinary tract

Type 1 fimbriae have been implicated as virulence factors in animal models of urinary tract infection (UTI), but the function in human disease remains unclear. This study used a human challenge model to examine if type 1 fimbriae trigger inflammation in the urinary tract. The asymptomatic bacteriuria strain Escherichia coli 83972, which fails to express type 1 fimbriae, due to a 4.25 kb fimB–fimD deletion, was reconstituted with a functional fim gene cluster and fimbrial expression was monitored through a gfp reporter. Each patient was inoculated with the fim+ or fim− variants on separate occasions, and the host response to type 1 fimbriae was quantified by intraindividual comparisons of the responses to the fim+ or fim− isogens, using cytokines and neutrophils as end‐points. Type 1 fimbriae did not promote inflammation and adherence was poor, as examined on exfoliated cells in urine. This was unexpected, as type 1 fimbriae enhanced the inflammatory response to the same strain in the murine urinary tract and as P fimbrial expression by E. coli 83972 enhances adherence and inflammation in challenged patients. We conclude that type 1 fimbriae do not contribute to the mucosal inflammatory response in the human urinary tract.

The Role of P Fimbriae for Colonization and Host Response Induction in the Human Urinary Tract

Bacterial attachment is thought to enhance virulence by promoting colonization of the urinary tract and by attacking tissue [1]. Uropathogenic Escherichia coli express several classes of fimbriae-associated adhesins that mediate attachment through specific binding to different glycoconjugate receptors [2] (for review see Mirelman [3]). P fimbriae, encoded by the pap gene cluster, show the most clear-cut association with virulence as defined by acute disease severity [4]. It has been proposed that P fimbriae augment the virulence of uropathogenic E. coli at different stages during the pathogenesis of urinary tract infection (UTI). pap+ strains remain longer in the intestinal flora and spread more efficiently to the urinary tract than do papstrains [5]. Once in the urinary tract, P-fimbriated strains establish bacteriuria and may cross the epithelial barrier into the bloodstream [6]. P fimbriae enhance epithelial cytokine responses in vitro [7] and in the mouse UTI model [8]. The role of P fimbriae per se for colonization and host response induction is not clear and has been the subject of some discussion. While fimbriae-mediated attachment enhances bac-

Innate defences and resistance to gram negative mucosal infection

Mucosal surfaces are continuously exposed to a diverse environmental microflora. Yet, infections are fairly rare and health is maintained. So, how is this achieved? Many and diverse effectors of the host defense have converged at mucosal sites and cooperate to resist the onslaught of pathogens. These are commonly assigned to two categories; the specific or the ‘’innate‘’ defenses. Specific immunity is executed by different lymphocyte populations in response to specific microbial antigens. The term ‘’innate’’ immunity is used to denote the defenses, which lack antigen specificity, but execute their defense functions through effector molecules with highly specialized anti-microbial activity. While many aspects of specific and ‘’innate’’ defense mechanisms have been worked out in vitro , we understand relatively little about their contribution to host resistance in vivo at a given infection site.