 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Basic Immunology and Pathology |
Departments of * Internal Medicine, Research Histopathology, Pediatrics, || Obstetrics and Gynecology and Microbiology and Immunology, The University of Texas Medical Branch at Galveston, Texas; and Division of Immunology and Allergy, Centre Hospitalier Universitaire Vaudois at Lausanne, Lausanne, Switzerland
Address correspondence to: Dr. James C. Rice, JSA 4.200, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0562. Phone: 409-772-1811; Fax: 409-772-5451; E-mail: jrice{at}utmb.edu
Received for publication March 3, 2005. Accepted for publication August 26, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
48 h. The decrease in pIgR was associated with decreased urinary IgA levels in the P(+)-infected group at 48 h. pIgR mRNA and protein also decline in P(+) E. coli–infected LPS-hyporesponsive mice. These studies identify a novel virulence mechanism of E. coli that express P fimbriae. It is proposed that P fimbriae decrease pIgR expression in the kidney and consequently decrease IgA transport into the urinary space. This may explain, in part, how E. coli that bear P fimbriae exploit the immune system of human hosts to establish ascending pyelonephritis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(1 to 4)Gal disaccharide on the apical surface of renal epithelial cells, glomeruli, and endothelia (2). The expression of P fimbriae on E. coli is important in establishing pyelonephritis, as P fimbriae–specific antibodies prevent the adherence of bacteria to uroepithelial cells in vitro (3) and protect animals from ascending E. coli pyelonephritis in vivo (4). Although P fimbriae are not the sole virulence factor on uropathic E. coli, pyelonephritis with P(+) E. coli strains are more often associated with kidney histopathology than P(–) E. coli (5). In addition to its role as an adhesin, E. coli that express P fimbriae may determine the pattern of the local immune response via interaction with its P-specific glycosphingolipid receptors present on uroepithelial cells (6) and/or by signaling via inflammatory cytokines (7). The secretions that protect the mucosal surface of renal epithelia contain an array of host defense factors, including secretory immunoglobulins, of which IgA is the major class. The polymeric Ig receptor (pIgR) is responsible for transporting these secretory immunoglobulins (S-Ig) in vesicles from the basolateral to the apical surface of epithelial cells (8–10), including renal tubule cells. The extracellular portion of the pIgR, termed secretory component (SC), is cleaved at the apical surface and excreted into the urinary space either bound to IgA as secretory IgA (S-IgA) or unbound (free SC) (11). Our goal was to determine whether P fimbriae on E. coli affect the expression of the pIgR and the transport of IgA into the urine during ascending pyelonephritis and to determine whether pIgR expression was affected by the innate immune response to LPS in vivo.
We have demonstrated that pIgR mRNA levels and pIgR protein levels are decreased in the rat kidney by renal ischemia (12). The significance of decreased pIgR protein levels induced by ischemia in the mucosal defense of the kidney is suggested by the frequent association of UTI with acute renal failure (13) and with the early (postischemic) renal transplant period (14). Although the exact roles of S-IgA antibodies and free SC (FSC) in pyelonephritis have not been fully elucidated, several studies suggest that S-Ig are important in protecting the urinary tract mucosa. S-IgA excretion rates are elevated during acute pyelonephritis in normal patients (15), depressed in girls with recurrent UTI compared with normal girls (16), and associated with a protective immune response to P-fimbriated E. coli UTI in a primate model (17). However, the question of whether E. coli that express P fimbriae interact with renal epithelial cells and affect the local secretory antibody immune response have not been elucidated.
The past three to four decades of studies on the mechanisms of ascending UTI have established a concept of bacterial adhesin-to-host receptor-mediated infection. In this process, E. coli adherence factors, such as P fimbriae, allow pathogens to anchor to the uroepithelium via P blood group receptor(s) (18). This was originally proposed to allow pathogen to evade shearing forces of urine flow and persist attached to the renal epithelial cells. Only relatively recent studies show that binding mediated by the P fimbriae not only anchor bacterium but also signal and cross-talk to the host and may involve the LPS receptor, toll-like receptor-4 (TLR-4) (19).
In this report, we describe the results of an experimental ascending pyelonephritis model that mimics the pathogenesis in the anatomically normal urinary tract (20) on the expression of the pIgR and urinary transport of IgA. We compared pIgR levels in the kidneys of mice that were infected with E. coli that express P fimbriae with a sister strain of E. coli that does not express P fimbriae. To define the roll of LPS in this component of the host response, we used both an LPS responsive strain (C3H/HeN) and in some experiments an LPS nonresponsive strain (C3H/HeJ) of mice, as renal tubule cells have been reported to respond to LPS in vitro (21). The C3H/HeJ mice are resistant to the LPS present in E. coli as a result of a naturally occurring mutation in the LPS receptor, TLR-4 (22). We found that E. coli that express P fimbriae decrease pIgR mRNA and protein expression in the kidney and decrease IgA transport into the urinary space during ascending pyelonephritis in LPS-sensitive mice. The parallel decrease in pIgR mRNA and protein levels in P(+) E. coli–infected LPS-insensitive mice indicates that these changes are independent of the LPS response. Hence, these results suggest that P fimbriae on E. coli may facilitate the establishment of acute pyelonephritis, in part, by impairing the local mucosal host defense by depressing S-IgA secretion into the urinary space.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
were performed using 7- to 9-wk-old female C3H/HeN (LPS-sensitive; Harlan Sprague-Dawley, Indianapolis, IN) mice; in some cases (Figure 6), parallel experiments were performed in age- and gender-matched C3H/HeJ (LPS-insensitive; Jackson Laboratory, Bar Harbor, MA) mice. All animal experimentation described in this article was conducted in accord with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Animals were housed together for 4 to 6 d before the study was started and received standard mouse diet (Diet #7001; Harlan Teklad Laboratory, Madison, WI) and water ad libitum during the study period.
|
|
|
|
|
|
All mice received ciprofloxacin (10 µg/g) by intraperitoneal injection 48 to 60 h before bladder inoculation to ensure the sterilization of the urinary tract. Before infusion of the inoculum into the bladder, anesthesia was induced by either halothane (2-bromo-2-chloro-1,1,1-triflaorethane, Sigma) or intraperitoneal injection of sodium pentobarbital (35 mg/kg). Animals were grouped in cages according to their experimental treatment in a biologic hazard isolation room for the duration of the experiment. After the completion of the study, anesthesia was induced by intraperitoneal injection of sodium pentobarbital (50 mg/kg), the peritoneum was opened, and both kidneys were rapidly removed. Transverse sections were obtained from one kidney for immunohistochemistry, protein isolation, and quantification of bacteria; the opposite kidney was rapidly frozen in liquid nitrogen for subsequent RNA isolation.
Bacterial Culture and Induction of Ascending Pyelonephritis
To clarify the roles of E. coli that express P fimbriae in modulating the epithelial cell’s response to ascending pyelonephritis, we used sister strains of E. coli that differ only in their expression of the P fimbriae adhesin. All of the bacterial strains are derivatives of E. coli FN506, a nonadherent fecal strain (24). This strain of E. coli was transformed with either high copy number of the pBR322 recombinant plasmid encoding pap-1 (P+) or the same plasmid without the pap-1 fimbriae (P–) (23). E. coli bacteria were grown overnight at 37°C on Luria agar, and the presence or absence of the P fimbriae expression on the strains was confirmed before bladder inoculation by the hemagglutination of 3% human blood cells that express the P antigen; E. coli express the P antigen P(+) agglutinate human blood cells, whereas E. coli that lack the P antigen P(–) do not agglutinate human blood cells.
Quantification of Bacteria in Kidney Tissue Homogenate
Bacterial infection of the upper urinary tract (pyelonephritis) was confirmed by culturing kidney homogenate and quantified by determining the number of CFU of bacteria per gram of whole kidney tissue as described (23). Results are reported as the mean CFU per gram of kidney tissue homogenate on two to three separate cultures. Kidneys with >103 CFU/g tissue were considered infected.
Total RNA Isolation and Northern Hybridization Analysis
Total RNA was prepared from whole kidney tissue that was rapidly removed at the time of tissue harvest, placed under liquid nitrogen, stored at –70°C, and isolated using Trireagent (Sigma, St. Louis, MO). Northern blot of total RNA was performed using a formaldehyde 1% agarose gel and transferred to a nylon membrane (Schleicher & Schuell, Keene, NH) as described (11). We used the 2.2-kb SacI fragment of the full-length cDNA probe for mouse pIgR mRNA provided by Charlotte Kaetzel (25) for Northern hybridization analysis of pIgR. Variability in loading of intact mRNA was assessed by reprobing the same blot for glyceraldehyde-3-phosphate-dehydrogenase RNA. Quantification of RNA was determined using NIH Image analysis performed on a Macintosh computer using the public domain NIH Image program (http://rsb.info.nih.gov/nih-image/), and pIgR was expressed as a ratio to that of glyceraldehyde-3-phosphate-dehydrogenase in the same sample. An equal area adjacent to the band of the interest was also quantified as background, allowing us to subtract background values for each band.
Antibodies
Polyclonal rabbit anti-mouse recombinant SC antiserum (26) was used as primary reagent with subsequent horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Amersham Biosciences, Piscataway, NJ) as detecting antibodies for Western blots and immunohistochemistry. We used the purified goat anti-mouse IgA for coating (Southern Biotechnology, Birmingham, AL), with the HRP-conjugated goat anti-mouse IgA (Southern Biotechnology) for detecting antibody and the purified mouse IgA TEPC 15 (Sigma) for our IgA standard for ELISA. For Western analysis of urine samples, we used goat anti-mouse 
light chain (Southern Biotechnology) to determine the presence of light chains and sheep anti–Tamm-Horsfall glycoprotein (uromucoid) antibody (Biogenesis Inc., Brentwood, NH) to confirm that equal amount of urinary protein was loaded for each urine sample and to facilitate comparison between groups.
ELISA
IgA concentrations in urine and kidney homogenates were quantified in a sandwich ELISA modified from methods previously described (11). Briefly, 96-well polystyrene microtiter plates (Dynatech Laboratories, Chantilly, VA) were coated with 3 µg/ml goat anti-mouse IgA in borate-buffered saline, samples were added in duplicate using serial dilutions and detected using polyclonal goat anti-mouse IgA-HRP and monoclonal IgA as a standard as described (27). Urine samples were tested at multiple dilutions, and the concentration of IgA was determined by linear regression, using at least two different dilution points for each sample.
Immunohistochemistry
Tissue sections were prepared from whole kidney fixed in 4% neutral-buffered formaldehyde before embedding in paraffin modified from methods previously described (28). Cross-sections of the renal hilum were labeled with a 1:300 dilution rabbit anti-recombinant mouse SC antibody. All immunohistochemistry sections were reviewed in a blinded manner (by H.-Q.W.) to eliminate any interpretation bias.
Western Blots
Equal amounts of protein from the whole kidney homogenate or from urine (by protein quantification) with radioimmunoprecipitation buffer was run on an 8.0% polyacrylamide gel (except urine Western analysis run on 4 to 20% nonreducing gels), transferred to polyvinylidene difluoride membrane (Schleicher & Schuell) in Tris-glycine electrophoresis buffer (pH 8.3) and blocked with 5% dried milk in 20 mM Tris-buffered saline-0.1% Tween-20 as per standard methods. The polyvinylidene difluoride membrane was incubated with rabbit anti-recombinant mouse SC at 1:2000 dilution, and the primary antibody was detected using goat anti-rabbit IgG:HRP conjugate (Bio-Rad, Hercules, CA) at 1:5000 dilution and oxidation of luminol in an enhanced chemiluminescence assay (Amersham Life Sciences, Arlington Heights, IL). The intensity of pIgR and 
-actin bands on the film was quantified using NIH Image 1.61 Software; pIgR levels were normalized to -actin levels in the same sample.
Urine Collection
Urine was collected from mice that spontaneously voided urine during handling or by direct aspiration of the bladder at the time of tissue harvest and immediately placed on ice. Protease inhibitor cocktail (Sigma) in sodium borate–buffered saline was added to each urine collection tube (0.15 µl/tube) to prevent proteolysis and microbial growth. Urinary levels of IgA were compared with urine creatinine levels (Sigma Kit, 555A) in the same specimen to normalize for urine concentration.
Statistical Analyses
The original bacteria colony count data were transformed to the log-log scale to stabilize the variance and induce symmetry. Statistical analyses were obtained on the transformed data using the general linear model procedure in SAS (SAS Institute Inc., Cary, NC). Pairwise mean comparisons are displayed with least significant interval plots in which nonoverlapping intervals indicate significantly different means at a 95% confidence level. When significance is obtained (P < 0.05) in the ANOVA, pairwise means of a priori interest are assessed using Fisher protected least significant difference; otherwise, a Bonferroni correction is applied to the intervals. P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
E. coli Expressing P Fimbriae Decrease pIgR RNA Expression in the Kidney during Ascending Pyelonephritis
We compared pIgR mRNA levels in kidneys of C3H/HeN (LPS sensitive) mice with documented P(+) and P(–) E. coli pyelonephritis, matched for the level of infection (CFU/g tissue), to isolate the effects of P fimbriae. We found significantly decreased pIgR mRNA expression in mice that were infected with E. coli that express P fimbriae (P+) than in mice that were infected with E. coli that did not express P fimbriae (P–) or with PBS-infused controls at 48 h after infusion by Northern hybridization (Figure 2). There was no statistically significant difference in RNA levels between P(+)- and P(–)-infected mice at 6 h (data not shown) or 24 h (Figure 2) by Northern hybridization. However, we detected an earlier decrease in pIgR mRNA levels with P(+)-infected versus P(–) E. coli–or PBS-infused controls by real-time reverse transcription–PCR that was significant at 24 h (ANOVA, P = 0.025), with the P(+) group being 64% of the P(–) group at 24 h (1.06 versus 1.66; P < 0.05; n = 3 to 5/group; data not shown).
pIgR Protein Levels Decreased during Infection with P(+) E. coli
We next compared pIgR protein expression at 6, 24, and 48 h in whole kidney homogenate from C3H/HeN mice that were infected with P(+) E. coli and P(–) E. coli that were matched for the bacterial colony counts (CFU/g). We found that pIgR protein is decreased during infection with E. coli that express P fimbriae (P+) but not during infection with E. coli that do not express P fimbriae (P–) or PBS infusion at 48 h (n = 8 to 15/group; P < 0.05; Figure 3). The pIgR protein was intact (approximately 116 to 120 kD; Figure 3). There was no difference in protein levels between P(+)- and P(–)-infected mice at 6 h (data not shown) or 24 h (Figure 3).
E. coli Expressing P Fimbriae Decreases pIgR Protein in the Kidney Cortex
We investigated the location and amount of pIgR protein expressed in the mouse kidney, using immunohistochemistry. We found that pIgR protein was expressed predominantly in the kidney cortex epithelia in the proximal and distal nephron tubule segments (cortical thick ascending limb and distal convoluted tubules; Figure 4). There was a significant decrease in pIgR staining in the tubule epithelia of mice that were infected with P(+) E. coli compared with mice that were infected with P(–) E. coli or PBS-infused controls at 48 h (Figure 4). The distribution and the intensity of staining for pIgR protein was similar in control (PBS-infused), P(–) E. coli–infected, and P(+) E. coli–infected mice at 6 and 24 h (data not shown), as determined in blinded analysis by our pathologist (H.-Q.W.).
Urinary IgA Levels Are Decreased in P(+) E. coli–Infected Mice
We measured the concentration of IgA in the urine collected from LPS-sensitive mice that were infected with P(+) E. coli or P(–) E. coli and compared these values with the preinfusion levels of IgA. We found that IgA levels in the urine are decreased in the P(+) E. coli–infected group, compared with baseline (preinfusion) levels at 48 and 72 h (Figure 5). There were no significant differences between the infected groups and baseline levels at 24 h (data not shown). The decrease in IgA levels in the P(+) group at 48 and 72 h compared with baseline levels persists even when urine IgA values are normalized for urinary creatinine (data not shown). We also compared the time-averaged values for urine IgA levels between groups (PBS versus P– versus P+) and found a significant difference between the PBS-infused and P(+)-infected groups (P = 0.049) but no difference between PBS-infused versus P(–)-infected groups (P = 0.172; data not shown).
Further analysis of the urine IgA fractions by Western blot analysis from preinoculation and P(+)-infected, P(–)-infected, and PBS-infused C3H/HeN (LPS-sensitive) mice at 48 h revealed that all forms of urinary IgA were decreased in the P(+)-infected group, compared with the P(–)-infected or PBS-infused groups, at 48 h. Urinary S-IgA levels were decreased to a greater extent than the monomeric IgA fraction in the P(+)-infected group. Urinary FSC levels were also decreased in the P(+)-infected group compared with the P(–)-infected or PBS-infused mice at 48 h (Figure 5). In addition, we measured IgA levels in whole kidney homogenate and found that IgA levels (µg IgA/mg protein) in the kidney were approximately two-fold higher in the P(+) E. coli–infected versus P(–) E. coli–infected mice at 48 h (0.192 versus 0.104 µg/mg protein), which although did not reach statistical significance (P = 0.067) may be physiologically relevant.
PIgR mRNA Levels and pIgR Protein Expression Decreased in P(+) E. coli–Infected, LPS-Insensitive (C3H/HeJ) Mice
We measured pIgR mRNA levels in an isogeneic strain of LPS-insensitive mice (C3H/HeJ) to determine whether the changes in renal pIgR mRNA and protein levels in the ascending pyelonephritis were mediated through an LPS response. Unlike the LPS-responsive C3H/HeN strain, C3H/HeJ mice are resistant to LPS present in E. coli as a result of a naturally occurring mutation in the LPS receptor TLR-4 (22). We found that pIgR mRNA levels were decreased in P(+) E. coli–infected mice by 40%, compared with P(–) E. coli–infected or PBS-infused controls in the LPS-insensitive (C3H/HeJ) mice at 24 h (Figure 6). In addition, we found that P(+) E. coli–infected C3H/HeJ mice had decreased immunostaining for pIgR in the cortex compared with P(–) E. coli–infected or control mice at 48 h (Figure 6) and confirmed that the decrease in pIgR staining was significant in P(+) E. coli–infected mice compared with P(–) mice by quantitative image analysis using NIH Quant (P = 0.025; data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
P(+) Fimbriae Facilitate Persistent Infection of the Kidney
As suggested by human and primate studies (30,31), our results indicate that E. coli that express P fimbriae cause more extensive and persistent infection in LPS-sensitive mice (Figure 1). These results are consistent with previous reports that transurethral inoculation of E. coli that express P fimbriae (P+) results in increased amount of bacterial infection and a persistence to a greater amount than inoculation with equal amounts of the isogenic strain of E. coli without P fimbriae (P–) in the mouse UTI model in LPS-sensitive (C3H/HeN) mice as reported for LPS-insensitive (C3H/HeJ) mice (23,24,29). Hence, the persistence of P(+) E. coli infection of the kidney may be due, in part, to the combined effects of the adhesion properties of P fimbriae and the ability of E. coli that express P fimbriae to impair IgA transport into the urinary space.
P(+) E. coli Decrease pIgR RNA Expression in Kidneys with Pyelonephritis
As the amount of infection is greater in P(+)-infected mice, we chose to compare RNA and protein levels from P(+) E. coli–infected and P(–) E. coli–infected mice that were matched for the same level of infection in the kidney (CFU/g). Our goal was to determine whether the presence of P fimbriae on E. coli affected the IgA transporter pIgR and to ensure that any changes were not due to differences in the number of bacteria colonizing the kidney. The significant decrease in pIgR RNA and protein levels at 48 h in mice that were infected with P(+) E. coli, compared with those that were infected with P(–) E. coli and PBS-infused mice, supports a dominant role for P fimbriae in the regulation of pIgR RNA and protein expression in the kidney.
E. coli Expressing P Fimbriae Decrease pIgR Protein Levels in the Cortex
We and others have localized pIgR protein in the mouse kidney (32,33). We are the first to demonstrate that infection caused by P(+) E. coli decreases pIgR expression in the cortical nephron segments. The distribution of pIgR protein seems to be similar to that of the normal rat kidney, in which we have demonstrated that pIgR protein is preferentially expressed near the luminal surface of the thick ascending limb and throughout some proximal tubules (11). The decreased pIgR staining in the cortical tubules of the P(+) E. coli–infected mice suggests that pIgR protein is depleted from the apical storage vesicles of the renal epithelia, where it has been shown clearly to be stored by electron microscopy in normal mice (33), and is consistent with our findings of decreased pIgR protein levels by Western blot in the P(+) E. coli–infected group at 48 h. Hence, the depletion of pIgR protein in the kidney cortex of pyelonephritis animals at 48 h (Figure 4) results from the combined effects of decreased pIgR RNA levels (Figure 2) and the continued excretion of SC (cleaved fragment of pIgR) either attached to IgA (Figure 5) or as FSC.
P(+) E. coli Decrease Urinary IgA
Western analysis of the urine samples from preinoculation and P(+)-infected, P(–)-infected and PBS-infused mice at 48 h revealed decreased levels of all forms of urinary IgA in the P(+)-infected group (Figure 5). Our results demonstrate that S-IgA is decreased to a greater extent than the monomeric IgA fraction in the P(+)-infected group (Figure 5). As these studies were performed on urine samples that previously were used for ELISA, we acknowledge that there may be some S-IgA breakdown in the urine during storage (34). However, as urinary S-IgA is decreased to a greater extent than the monomeric IgA fraction and FSC levels are lower in the P(+)-infected mice, these results support our data that decreased pIgR/SC levels in the kidney result in decreased transport of IgA into the urine in the P(+)-infected group and emphasize the importance of pIgR/SC in the humoral immune response. In addition, the trend toward higher kidney IgA levels in the kidney of the P(+)-infected group also supports the concept that decreased pIgR/SC levels in the kidney result in impaired IgA transport into the urine. Hence, only the P(+) E. coli–infected group has reduced pIgR mRNA and pIgR protein in the kidney, resulting in decreased S-IgA and decreased FSC (secreted cleaved component of pIgR) transport into the urinary space. These results support our hypothesis that uropathic P(+) E. coli may facilitate the persistence of infection in the kidney by reducing pIgR mRNA and pIgR protein levels in the kidney, resulting in decreased transport of S-IgA and FSC into the urine.
Regulation of pIgR RNA and Protein Expression in the Kidney Is Independent of LPS Sensitivity
Our results demonstrate decreased pIgR mRNA levels in the kidney in P(+) E. coli–infected compared with P(–) E. coli–infected (matched for CFU/g) or PBS-infused controls and indicate that P fimbriae decrease the constitutive expression of pIgR mRNA (35,36). We reported decreased pIgR mRNA and protein levels in the kidney in other models of renal stress compared with control, including water-loading (11) and ischemia-reperfusion injury (12). As several factors have been reported to increase transcription of the pIgR, including glucocorticoids (37) and proinflammatory cytokines (38), and E. coli bacterial infection increases cytokines (39) that are known to increase pIgR expression (8,40), we expected to see an increase in pIgR expression in the kidney in our ascending pyelonephritis model. The increase in proinflammatory cytokines/chemokines during renal infection or injury is believed to be due, in part, to a response to LPS (21,41). Intestinal epithelial cells have recently been shown to increase pIgR expression in response to LPS in vitro (42). However, the ability of renal epithelial cells to react to LPS directly is unclear, as some report that renal tubule cells are poorly responsive to LPS as a result of lack of CD14 (19), whereas others report that renal epithelia respond to LPS in vitro (21) or that the LPS receptor TLR-4 (43) and the CD14 receptor (44) are induced during kidney disease. The mutations in the LPS receptor that result in LPS insensitivity in the C3H/HeJ mice are applicable and translatable to humans, as mutations in the LPS receptor underlie some of the variability in responsiveness to LPS in humans (45). Furthermore, recent studies suggest that the LPS receptor (TLR-4) is not restricted to the endotoxin/LPS ligand and can signal in response to other microbial products (46), including P fimbriae (19). Hedlund et al. (7) reported that the renal epithelial cell cytokine response to P fimbriae in vitro is independent of LPS but requires a functional LPS receptor (TLR-4) (19).
To address the role of LPS and its receptor TLR-4 in the regulation of pIgR expression during infection, we used LPS-resistant (C3H/HeJ) mice. The similar decrease in pIgR mRNA and protein expression in LPS-sensitive mice (Figures 2 and 3) and LPS-resistant mice (Figure 6) indicates that the decrease in pIgR in the kidneys of P(+) E. coli–infected mice does not depend on an intact LPS response.
Our results support recent studies showing that P fimbriae not only anchor bacterium but also signal and cross-talk to the host. In contrast to most reports that P fimbriae signaling results in increased expression of proinflammatory molecules and involves an intact LPS receptor, TLR-4, in our report we unexpectedly found that this cross-talk results in decreased expression of pIgR and, respectively, decreased levels of urine IgA. In addition, the decrease in pIgR mRNA and protein did not depend on an intact TLR-4 signaling. One of the key roles of IgA previously described is the blocking of bacterial binding to epithelia, thereby interfering with the receptor:adhesin interaction (17). We propose a working hypothesis that P fimbriated E. coli associated with acute pyelonephritis evolved to exploit this system after binding and signal the host to decrease pIgR-mediated transport of IgA into the urine, therefore evading one of the key protective systems involved in the anti-adherence response.
Taken together, our results indicate that P fimbriae on E. coli regulate the mucosal immune response by decreasing pIgR expression in the kidneys in a mouse model of ascending pyelonephritis. The regulation of pIgR-mediated transport of IgA is in addition to its role as an adhesin and does not require an intact LPS response, as infection with P(+) E. coli decreases pIgR mRNA and protein expression in the kidneys in both LPS-sensitive and LPS-resistant mice. The decrease in pIgR levels in the kidney correlate with the decrease in urinary IgA levels in P(+) E. coli–infected mice. Hence, we present the novel findings that the presence of P fimbriae on E. coli may facilitate pyelonephritis, in part, by impairing the kidney’s ability to transport S-IgA in response to ascending infection. A better understanding of the mechanisms that regulate the mucosal immune response of the kidney to E. coli and P fimbriae in acute ascending pyelonephritis may lead to treatment modalities that augment the mucosal defense of the kidney and potentially reduce the renal disease as a result of ascending pyelonephritis.
|
Acknowledgments |
|---|
Portions of this work have previously been published in abstract form (Imunol Lett 69: 22, 2000).
We acknowledge the technical assistance of Jyotsana Singhal, Deborah Yetman, and Ning-ping Yang and the manuscript review and the research guidance of Drs. David Good and William Mitch.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Stapleton Novel Mechanism of P-Fimbriated Escherichia coli Virulence in Pyelonephritis J. Am. Soc. Nephrol., December 1, 2005; 16(12): 3458 - 3460. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
) compared with P(–) E. coli–infected (
) and PBS-infused control (hatched gray bars) C3H/HeN (LPS-sensitive) mice at 48 h. Total pIgR mRNA levels in the P(+)-infected mice are 75% of P(–) E. coli–infected and 63% of PBS-infused control mice at 48 h, whereas there is no difference in pIgR levels between groups at 6 h (data not shown) or 24 h. RNA levels for pIgR were normalized to glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) RNA in the same sample and expressed as a ratio to pIgR/GAPDH RNA levels for PBS-infused controls on the same blot. *P < 0.05, P(+) versus P(–) E. coli–infected and PBS-infused controls; n = 12 to 15 mice per group.
) group at 24 h. There are no differences between groups at 6 h. pIgR mRNA levels are normalized to GAPDH in the same specimen and expressed as a percentage of the pIgR/GAPDH ratio of PBS-infused controls at the same time point; n = 12/group; *P < 0.05. (Right) Representative immunohistochemistry section demonstrates decreased staining for pIgR in the cortical tubule segments of P(+) E. coli–infected (bottom right) versus P(–) E. coli–infected kidneys (top right). There is no staining of glomeruli (G). Magnification, x200.