Defensins in the Urinary Tract and Other Tissues

The epithelial surfaces of the urinary tract, like those of other mucosal surfaces, are covered by a thin layer of fluid secreted by the epithelial cells and any inflammatory cells that may have been recruited in response to infection or inflammatory stimuli. As first described for nasal secretions by Alexander Fleming in 1922, epithelial secretions possess antimicrobial properties that are due mostly to their content of various antimicrobial proteins and peptides. These substances are present constitutively, but their activity can be augmented by the release of additional substances from epithelial cells and phagocytes stimulated by infection or inflammation. Among the antimicrobial substances identified in various epithelial secretions, lysozyme, lactoferrin, secretory leukoprotease inhibitor, peroxidases, and defensins are the most abundant. It is noteworthy that many of these substances are also prominent components of the microbicidal granules of polymorphonuclear leukocytes [1]. The various antimicrobial substances act by distinct mechanisms, and there is increasing evidence that they often act synergistically Defensins are 3- to 5-kDa, three-disulfide cationic peptides abundantly expressed in vertebrate phagocytes and epithelia. Similar but not necessarily genetically related peptides are also found in host defense settings in plants and invertebrates [2]. Most defensins manifest antimicrobial activity against a broad range of microbes, including gram-positive and gram-negative bacteria, fungi, and some enveloped viruses. In addition, defensins display other activities that implicate them as regulators of inflammation and of the adaptive immune response Three defensins, human neutrophil peptides (HNP)-1–3 are among the most abundant constituents of the azurophil granules of neutrophils, while HNP-4, also in neutrophil granules, is ∼100-fold less abundant [3]. Neutrophil defensins form pores in artificial membranes that contain anionic phospholipids, and they permeabilize the membranes of bacteria. Their antimicrobial mechanism of action depends on membrane permeabilization but may also involve subsequent and as yet poorly understood intracellular events. The initial interaction of cationic defensins with anionic microbial surfaces is likely to involve electrostatic forces, accounting for the known inhibitory effects of increasing salt and divalent cation concentrations on defensin activity. However, the fusion of a defensin-containing granule to a phagocytic vacuole (phagosome) is expected to generate defensin concentrations in the milligram per milliliter range, a concentration likely sufficient to overcome the known inhibitory effects of salt and other serum components. Phagosomal contents in human neutrophils can be assayed by phagocytosis of bacteria in the presence of radioactive iodide, a procedure that leads to radioiodination of phagosomal contents. Among the radioiodinated phagosomal polypeptides, defensins are predominant, confirming their expected abundance within this compartment In humans, neutrophil defensins and the other known human defensins expressed in various epithelia are encoded by a cluster of genes on chromosome 8 band p23. In vivo, human neutrophil defensins are synthesized in late promyelocytes, the bone marrow precursors of neutrophils. They are encoded as 94–amino acid precursors and are processed posttranslationally to mature 29– to 30–amino acid peptides prior to storage in granules. The specific processing enzymes have not yet been identified. Among animals used for models of human pathology, rabbits, guinea pigs, rats, and hamsters have neutrophil defensins similar to those of humans; mice and pigs do not. Mice and pig neutrophils are not defenseless since they have antimicrobial peptides that belong to the cathelicidin family, and they have other antimicrobial proteins and mechanisms. It is likely that each animal species evolves its antimicrobial arsenal under pressure from distinct pathogens, accounting for the observed interspecies differences in antimicrobial molecules. Remarkably, mice and pigs do express homologs of human epithelial defensins The first group of epithelial defensins was described as an abundant class of mRNAs in mouse intestinal Paneth cells [4]. Subsequently, the human intestinal defensins, HD-5 and HD-6, were cloned by homology to neutrophil defensins and localized by in situ hybridization and immunostaining to the granules of Paneth cells [5]. Their spectrum of activity is similar to that of neutrophil defensins. Like neutrophil defensins, they are synthesized as 90– to 100–amino acid preprodefensins. In the mouse model, Paneth cell granules contain prodefensins (procryptdins) that are activated to mature defensins by matrilysin. Mice with disrupted matrilysin genes are rendered incapable of processing (most) cryptdins and have a defect in small intestinal killing of ingested microbes. There is evidence that human intestinal defensins are not processed in the same manner as mouse intestinal defensins but may instead be stored in Paneth cell granules largely in their mature form The known human neutrophil and Paneth cell defensins belong to the structural subclass of α-defensins, which are characterized by specific spacing and disulfide linkage of their six cysteines. Another class of defensins, β-defensins, has different cysteine spacing and disulfide linkage but assumes a very similar three-dimensional conformation and is genetically related to α-defensins. β-defensins were initially found as lipopolysaccharide-inducible antimicrobial peptides of the bovine tracheal epithelium [6] and abundant antimicrobial peptides of bovine polymorphonuclear leukocytes. Since these initial studies, β-defensins have also been found in avian leukocytes. Peptides with the structure of β-defensins are also components of snake venoms, in which they are classified as myotoxins. It is possible that these myotoxic peptides evolved by modification of the ancient antimicrobial peptides of lower vertebrates Two human β-defensins, HBD-1 and HBD-2, were discovered respectively by screening plasma dialysate and psoriatic skin for novel peptides. HBD-1 is expressed predominantly in the kidney and the female genital tract, with lesser expression in the pancreas, liver, and other epithelia. Within the kidney, in situ hybridization indicates that HBD-1 is produced in distal tubules, loops of Henle, and collecting ducts [7]. Human urine contains 10–100 μg/L HBD-1, with variable N-terminal processing ranging from 36–47 amino acids. The release of HBD-1 rises about 3-fold during pyelonephritis [8] and is accompanied by increased blood HBD-1 concentrations. It is not clear whether this represents increased release from damaged kidney cells or increased synthesis. In the female genital tract, HBD-1 mRNA is found in the epithelia of uterine glands and the endocervix and in the parabasal cells of the vaginal lining. Although several of the HBD-1 forms show activity against Escherichia coli ML-35, the required concentrations are at least 10-fold higher than those detected in urine. It is possible that higher concentrations of HBD-1 exist in or on epithelial cells, where they could form an antimicrobial barrier. Alternatively, HBD-1 could have another role in the kidney, unrelated to its antimicrobial properties, such as the recently reported chemoattractant activity of HBD-1 and HBD-2 toward immature dendritic cells and memory T cells The second known human β-defensin, HBD-2, was isolated from the shed cornified skin of patients with psoriasis [9]. On the basis of the yield of HBD-2 from the purification, 200–400 μg/50 g of psoriatic skin, we estimate a minimum concentration in inflamed skin of 6 μg/mL (i.e., ∼1–2 μM). Our immunostaining studies indicate that this material is not uniformly distributed, so the local concentrations in inflamed skin are likely to be even higher. At these concentrations, HBD-2 is broadly microbicidal toward gram-negative and gram-positive bacteria at both low and normal salt concentrations. An interesting exception is the impressive resistance to HBD-2 shown by the skin pathogen Staphylococcus aureus. Lower concentrations of HBD-2 are also present in other inflamed epithelia and epithelial surfaces exposed to bacteria. Immunostaining experiments show that the HBD-2 induction in response to inflammation or infection is a local phenomenon, restricted to the area involved. The known consensus NF-κB–binding motifs in the HBD-2 gene promoter [10] may mediate the transcriptional component of the response to inflammatory signals Although the role of defensins in urinary tract infections has not been systematically studied, the tissue distribution and properties of human defensins allow the construction of a speculative scenario. HBD-1 may work together with other constitutive components of the mucosal lining of the distal nephron, ureters, bladder, and urethra to form a barrier to tissue penetration by ascending bacteria. The neutrophil defensins HNP-1–3 may contribute to clearance of bacteria phagocytized by exudate neutrophils, and the inducible HBD-2 could fortify epithelial barriers that have encountered bacteria or other inflammatory signals. In turn, the diffusion of these defensins from the site of infection may contribute to the chemotactic signal that attracts the cells necessary for the initiation of an adaptive immune response (dendritic cells and T cells). A similar scenario may also apply to the lining of the female genital tract. Inherited or acquired disorders of defensin production could contribute to susceptibility to urogenital infections. Given the demonstrated importance of innate immunity in the host defense against urinary tract infection, we can look forward to exploring these scenarios in suitable animal models