点击显示 收起
Center for Global Health, Division of Infectious Disease and International Health, University of Virginia, Charlottesville
National Institutes of Health Philippines, University of the Philippines Manila, Philippines
Federal University of Ceará, Fortaleza Ceara, Brazil
Angiotensin II (ANG II) has been described in the regulation of intestinal secretion and absorption via angiotensin subtype 1 (AT1) and AT2 receptors, respectively, in rats. We investigated the role that ANG II plays in the rabbit ileal-loop model of Clostridium difficile infection. Expression of AT1, the more abundant ANG II receptor, was demonstrated in ileal loops, and an AT1 receptor blocker, losartan, inhibited hypersecretion induced by C. difficile toxin A (mean volume : length ratio, 0.27 ± 0.06 vs. 0.60 ± 0.06 mL/cm in controls). Losartan also decreased production of ANG II in the ileum (0.48 ± 0.06 vs. 0.87 ± 0.12 pg/mg in controls), raising the possibility that ANG II may participate in a positive feedback loop involving the hypersecretory response. Our findings suggest that ANG II plays a significant role in the pathogenesis of C. difficile toxininduced diarrhea.
Toxigenic Clostridium difficile is the most important known cause of both antibiotic-associated and nosocomial diarrhea. The clinical manifestations of C. difficile colitis are caused by either one or both of the large exotoxinstoxins A and Bsecreted by C. difficile. Most of the inflammatory and secretory effects of C. difficile are due to toxin A, an enterotoxin. Toxin A has been shown to cause the release of inflammatory cytokines, the recruitment of polymorphonuclear cells, and the stimulation of prostaglandin synthesis [13]. Toxin B, however, is a cytotoxin that causes cytopathic changes in cell culture. Recent studies have implicated toxin Anegative/toxin Bpositive strains of C. difficile as a cause of disease outbreaks [4]. Both toxin A and toxin B exert at least some of their effects via the monoglycosylation and inactivation of an intracellular signaling target, rho, which leads to cytoskeletal disruption [57].
Angiotensin II (ANG II) receptors exist as subtype 1 (AT1) and subtype 2 (AT2) and are known to play a major role in the cardiovascular and renovascular systems that mediate inflammation, cell growth, fibrosis, and vascular tone. There has been considerable interest in the presence of a local angiotensin system in the gastrointestinal tract, including the demonstration of ANG II binding sites, angiotensin-converting enzyme, and AT1 and AT2 receptors in the rat jejunum [811] and, more recently, in a human intestinal cell line (B.A.C.-F., G.A.C.B., and R.L.G., personal communication). The AT1 receptor is present throughout the body in most species that have been studied to date [12]. Autoradiography performed on rat intestines detected mostly AT1 receptors in the mucosa and muscularis of the jejunum, ileum, and colon, whereas only a small population of AT2 receptors was observed [10]. The AT2 receptor is more localized in the brain and adrenal glands than in other tissues, and its location may be more variable by species than is that of the AT1 receptor [12]. Activation of the AT1 receptor inhibits absorption or stimulates secretion, whereas activation of the AT2 receptor causes net absorption of ions and water, although these effects may be species and/or site specific [1315]. Losartan, a nonpeptide competitive antagonist, selectively blocks AT1 receptors. In the present study, we investigated whether the same local angiotensin system, through AT1 receptors, operates in the rabbit model of C. difficileassociated diarrhea (CDAD). Furthermore, we investigated the association between ANG II and C. difficile toxin Ainduced inflammatory and secretory responses with and without AT1 receptor blockade.
MATERIALS AND METHODS
Rabbit ileal loops.
The animal experiment protocol was reviewed and approved by the University of Virginia Center for Comparative Medicine. Twenty-two New Zealand white rabbits weighing 2 kg were fasted overnight. The rabbits were anesthetized with ketamine (6080 mg/kg) and xylazine (510 mg/kg), administered intramuscularly, and a midline abdominal incision was made to expose the small bowel. After the ileum was flushed with 510 mL of PBS, 814 loops of 4 cm each were ligated using double ties, and a 1-cm interval was left between loops.
We studied 121 ligated ileal loops, of which 68 loops from 13 rabbits served as controls. Each of the control loops was injected intraluminally with a 1-mL solution of PBS with either 10 g/mL C. difficile toxin A (Techlab), 2 g/mL cholera toxin (CT; ICN Biomedicals), or 20 g/mL E. coli heat-stable toxin (ST; Sigma-Aldrich). Losartan powder (DuPont-Merck Pharmaceutical) was diluted in PBS to concentrations of 10, 3, 1, 0.3, or 0.1 mg/mL. Immediately before the enterotoxins were administered, 53 loops from another 9 rabbits were treated with losartan. The ileal loops were replaced intraperitoneally, and the abdominal incision was sutured closed. The rabbits were maintained under anesthesia until they were euthanized 5 h later.
Measurement of intestinal secretion.
After 5 h of incubation, the ligated ileal loops were removed. The length of each ligated ileal segment was measured. Intraluminal fluid was extracted from each ileal loop and quantified. The volume : length ratio (V : L) was calculated in milliliters per centimeter per loop. The gross description (clear, serous, purulent, or bloody) of the collected fluid was also noted.
Histopathologic assessment.
Samples of intestinal tissue from control and treated ileal loops were fixed with 10% formalin and stained with hematoxylin-eosin. Three of the investigators (C.S.A., R.L.G., and G.A.C.B.) randomly read each slide and were blinded to its source. A grading scale was formulated, and each slide was graded from 0 (none) to 4 (worst) on the basis of the degree of mucosal disruption, the increase in cellularity, and the intensity of vascular congestion (table 1). The final histologic score of each slide was the mean of the scores given by the 3 investigators. This grading scale was used in a study published elsewhere on enterotoxin-induced mucosal injury in the same rabbit ileal-loop model [3].
ANG II EIA.
Ileal tissues were immediately stored in liquid nitrogen until the EIA was performed. For the ANG II assay, 1 mL of PBS/g of tissue was added. The tissue was homogenized and sonicated. The homogenate was centrifuged at 1500 g for 10 min. The supernatant was tested for ANG II by specific EIA (SPI-BIO), in accordance with the manufacturer's instructions. Briefly, the supernatant was dried by vacuum centrifugation. The sample was resuspended in EIA buffer, vortexed, and centrifuged at 3000 g at 4°C for 10 min. An ANG II standard and samples were dispensed in duplicate onto plates coated with monoclonal antiANG II. After 1 h of incubation at room temperature, glyceraldehydes and borane trimethylamine were successively dispensed onto the plates. The plates were washed with buffer, and an antiANG II IgG tracer was dispensed into each well. After an overnight incubation at 4°C, the plates were again washed with buffer, incubated with Ellman's reagent, and read at 405 nm with an ELISA reader (Titertek Multiskan Plus).
Western blot analysis.
Ileal tissues were harvested from 2 control rabbits treated with either PBS or toxin A only. The tissue samples were immediately frozen and kept in liquid nitrogen until protein extraction was performed. The samples were homogenized in 1 mL of cold lysis buffer that contained 50 mmol/L HEPES (pH 7.48.9), 1% Triton X, and a 10% (wt/vol) protease inhibitor mixture (Sigma). The homogenate was sonicated for 10 s, incubated in ice for 30 min, and centrifuged at 20,000 g at 4°C for 10 min. The supernatant was then aliquoted and stored at -70°C until the protein assay was performed. Concentrations of proteins were determined by bioinchoninic acid protein assay (Pierce), in accordance with the manufacturer's instructions.
The samples were diluted in SDS buffer (Sigma-Aldrich) and heated at 100°C for 4 min before being loaded on a 10% Tris-HCL SDS-polyacrylamide gel (Bio-Rad Laboratories). Each lane was loaded with 20-g samples of protein, and 1 lane of each gel contained prestained molecular weight standards (Bio-Rad Laboratories). Vascular smooth muscle protein extract and PC-12 whole-cell lysate (Santa Cruz Biotechnology) served as positive controls. After electrophoresis, the proteins were transferred to nitrocellulose membranes, which were then placed in Tris-buffered saline with 5% nonfat milk and 0.05% Tween 20. Conjugation of the primary AT1 (N-10) affinity purified rabbit polyclonal antibody (Santa Cruz Biotechnology) was performed using a Zenon horseradish peroxidase (HRP)conjugated rabbit IgG labeling kit (Molecular Probes). After an overnight incubation at 4°C, 1 nitrocellulose membrane was incubated with the HRP-conjugated AT1 antibody (1 : 500 dilution). For the preadsorption study, another membrane was incubated with an HRP-conjugated primary antibody that was initially mixed for 1 h with a 10-fold molar excess of AT1 (N-10) peptide (Santa Cruz Biotechnology), a blocking peptide. Immunoreactivity was visualized by enhanced chemiluminescence (ECL Plus; Amersham Biosciences). Quantitative assessment of band densities was performed by scanning densitometry (ImageQuant; Molecular Dynamics).
Statistical analysis.
Continuous variables, such as V : L, histologic score, Western blot band densities, and levels of ANG II, are expressed as means ± SE. For statistical comparison of the magnitude of secretion, the amount of inflammation, and protein levels, a 2-tailed Student's t test was applied. Analysis of variance (ANOVA) was performed to compare ileal loops treated with varying doses of losartan. P < .05 was considered to be statistically significant.
RESULTS
Toxin Ainduced hemorrhagic secretion and mucosal injury.
Toxin Atreated loops had significantly elevated levels of intestinal secretions (mean V : L, 0.60 ± 0.06 mL/cm), compared with PBS-treated loops (mean V : L, 0.02 ± 0.01 mL/cm; P < .0001). Most (89%) of the toxin Atreated loops had blood-tinged or grossly hemorrhagic secretions. Levels of intestinal secretions were also significantly elevated in CT-treated loops (mean V : L, 0.68 ± 0.12 mL/cm; P = .0007) and in ST-treated loops (mean V : L, 0.28 ± 0.06 mL/cm; P = .008), compared with those in PBS-treated loops (mean V : L, 0.01 ± 0.01 mL/cm). In contrast to the hemorrhagic secretions in toxin Atreated loops, most (86%) CT-treated loops had clear serous fluid, whereas ST-treated loops had mostly cloudy (16%) to blood-tinged (66%) secretions.
Microscopic studies of the intestinal mucosa exposed to toxin A (histologic score, 3.28 ± 0.24 vs. 0.25 ± 0.13 in PBS-treated loops; P < .0001) revealed severe epithelial disruption, inflammation, and vascular congestion (figure 1). In contrast, in PBS-, CT-, or ST-treated loops, no gross changes in the epithelial mucosa or differences in the histologic scores were observed.
AT1 receptors expressed in toxin Atreated loops.
Western blot analysis using antibody specific for AT1 receptor proteins recognized 2 bands within the expected molecular weight range of 3750 in the lanes with rabbit ileal proteins. The protein bands from toxin Atreated loops appeared darker than those from PBS-treated loops (figure 2A and 2B). Densitometric analysis of bands from toxin Atreated loops showed a mean increase in density volume of 107%, compared with those from PBS-treated loops. The lower band in the lanes with rabbit ileal proteins corresponded to the level of AT1 receptor expression seen in the positive control, vascular smooth muscle (figure 2A). Preadsorption studies demonstrated complete disappearance of the lower bands in rabbit ileal proteins, as was observed in the AT1 receptor protein band in the positive control, PC-12 whole-cell lysate (figure 2B and 2C). In the neutralization study, the upper bands demonstrated partial fading. Densitometric analysis of the upper bands showed a mean decrease in density volume of 35% after preadsorption.
AT1 receptor blockade inhibition of toxin Ainduced secretion.
All ileal loops treated with both enterotoxin and losartan showed a statistically significant reduction in mean V : L. Toxin Atreated loops had a 57% reduction in their mean secretory response when they were also treated with losartan (mean V : L, 0.27 ± 0.06 vs. 0.60 ± 0.06 mL/cm in loops treated with toxin A alone; P < .001) (figure 3A). Furthermore, a dose-dependent reduction in secretion occurred as the dose of losartan was increased from 0.1 to 10 mg/mL (P < .0005, ANOVA) (figure 3B). When each treatment group was compared with the group that was treated with toxin A only, a statistically significant difference was found in all groups (10-mg dose group, n = 3 [P < .0001]; 3-mg dose group, n = 4 [P = .002]; 1-mg dose group, n = 5 [P < .0001]; 0.1-mg dose group, n = 4 [P = .008]), except in the 0.3-mg dose group. CT- and ST-treated loops showed a reduction in the levels of secretions of 59% (P = .009) and 64% (P = .04), respectively, when they were also treated with losartan.
Loops treated with losartan at all doses showed a dose-related trend toward inhibition of inflammation and tissue damage (P = .053, ANOVA), but a statistically significant difference was observed only at the 10-mg dose (n = 3; mean histologic score, 1.67 ± 0.33; P = .02). At the 1-mg dose, some loops already showed almost complete inhibition of mucosal injury.
AT1 receptor blockade inhibition of toxin Ainduced secretion of ANG II.
ANG II levels were higher in toxin Atreated loops than in PBS-treated loops (0.87 ± 0.12 vs. 0.50 ± 0.09 pg/mg; P = .03). Loops treated with both losartan and toxin A showed significantly decreased ANG II levels (0.48 ± 0.06 pg/mg), compared with loops treated with toxin A only (0.87 ± 0.12 pg/mg; P = .02) (figure 4).
DISCUSSION
In the rabbit ileum, toxin A induced severe mucosal injury, invoked an intense inflammatory reaction, and augmented intestinal secretion. These destructive changes brought about by toxin A have been well documented in previous studies and may even be consistent with what is observed in severe clinical cases of CDAD [16]. The present study has demonstrated a possible role for the local angiotensin system in CDAD. The rabbit ileal tissue showed increased levels of ANG II in the presence of toxin A. This effect corroborates what has been reported for other inflammatory processes, such as Crohn disease, in which ANG I and ANG II were noted to be elevated [17]. Until now, there had been no information available that implicated ANG II in infectious inflammatory enteritides such as C. difficile colitis. Of interest, human neutrophils and mast cells, which are observed to be recruited in toxin Ainduced colitis, contain cathepsin G and chymase, respectively, which are potent enzymes that convert ANG I to ANG II [1820]. Whether the synthesis of these enzymes is stimulated in CDAD remains to be explored. Elevated levels of ANG II may also be part of a systemic activation of the renin-angiotensin-aldosterone system. Enhanced vascularity secondary to treatment with toxin A may have enhanced the shift of circulating ANG II in the intestinal milieu.
The demonstration of AT1 receptors in the rabbit ileum further implicates the tissue angiotensin system in the modulation or mediation of inflammatory and secretory processes in the intestinal mucosa. The AT1 receptor protein bands, which were within the expected molecular weight range, were confirmed by the complete neutralization (lower bands in the Western blot analysis) or decrease in intensity (upper bands in the Western blot analysis) of the bands in the presence of the specific blocking peptide (figure 2). The upper protein bands that were recognized by the specific AT1 receptor IgG but were partially neutralized by the peptide might represent heavily glycosylated epitopes. Interestingly, toxin Atreated tissues had darker bands, as was confirmed through both visual inspection and densitometry. This difference may indicate that the expression of AT1 receptors may be up-regulated in inflammatory enteritides, as has been reported in animal studies of other pathological states, such as renal disease and insulin resistance [21, 22].
ANG II levels were elevated in toxin Atreated rabbit ileum, which suggests that the intense mucosal inflammation and accumulation of intraluminal fluid were mediated by ANG II. Blocking the AT1 receptor with losartan inhibited the secretory response to toxin A. This is consistent with what has been reported in the rat jejunum, where the administration of a high dose of ANG II resulted in intestinal secretion through AT1 receptors [1315, 23]. Losartan also caused a concomitant decrease in ANG II levels, and this indicates that ANG II synthesis is regulated by AT1 receptors. Both CT- and ST-induced secretion were also inhibited by losartan, possibly through a distal common pathway, which suggests that ANG II receptor activation may not be specific to toxin A but can be accomplished by various enterotoxins.
In a previous study, rabbit ileum stimulated with toxin A yielded elevated levels of prostaglandin E2 (PGE2) because of the induction of cyclooxygenase 2 expression [3]. AT1 is also known to stimulate phospholipase A2, which results in increased levels of PGE2. Although it was not statistically significant (P = .053), a trend toward an effect on inflammation, vascular congestion, and epithelial disruption caused by toxin A was observed in the intestinal mucosa. Other factors in the local tissue angiotensin system may be involved in mucosal integrity. Although it was not demonstrated in the present study, AT2 receptors have been implicated in the maintenance of mucosal barrier functions. Stimulation of AT2 receptors has been shown to stimulate the production of nitric oxide, an important mediator of mucosal defense [24, 25]. Whether the ANG II receptors play a role in the alteration of the monoglucosylation of rho substrates in enterocytes in CDAD remains to be proven.
In summary, the findings in the present study suggest that toxin A enhances the expression of ANG II, which, in turn, mediates mucosal injury and intestinal secretion through AT1 receptors. Blockade of the inflammatory and secretory cascade by use of a pharmacologic approach, such as the inhibition of AT1 receptors, provides a potentially novel approach to control CDAD and C. difficile colitis.
Acknowledgments
We thank Helen McGrath, John Gildea, Charlotte Martin, Nancy Howell, and Ruth Aldridge for their valuable technical support.
References
1. Flegel WA, Muller F, Daubener W, Fischer HG, Hadding U, Northoff H. Cytokine response by human monocytes to Clostridium difficile toxin A and toxin B. Infect Immun 1991; 59:365966. First citation in article
2. Pothoulakis C, Sullivan R, Melnick DA, et al. Clostridium difficile toxin A stimulates intracellular calcium release and chemotactic response in human granulocytes. J Clin Invest 1988; 81:17415. First citation in article
3. Alcantara C, Stenson WF, Steiner TS, Guerrant RL. Role of inducible cyclooxygenase and prostaglandins in Clostridium difficile toxin Ainduced secretion and inflammation in an animal model. J Infect Dis 2001; 184:64852. First citation in article
4. Alfa MJ, Kabani A, Lyerly D, et al. Characterization of a toxin A-negative, toxin B-positive strain of Clostridium difficile responsible for a nosocomial outbreak of Clostridium difficile-associated diarrhea. J Clin Microbiol 2000; 38:270614. First citation in article
5. Just I, Wilm M, Selzer J, et al. The enterotoxin from Clostridium difficile (ToxA) monoglucosylates the rho proteins. J Biol Chem 1995; 270:139326. First citation in article
6. Just I, Selzer J, Wilm M, von Eichel-Streiber C, Mann M, Aktories K. Glucosylation of rho proteins by Clostridium difficile toxin B. Nature 1995; 375:5003. First citation in article
7. Brito GAC, Sullivan GW, Ciesla WP Jr, Carper HT, Mandell GL, Guerrant RL. Clostridium difficile toxin A alters in vitroadherent neutrophil morphology and function. J Infect Dis 2002; 185:1297306. First citation in article
8. Cox HM, Munday KA, Poat JA. Identification of selective, high affinity [125I]-angiotensin and [125I]-bradykinin binding sites in rat intestinal epithelia. Br J Pharmacol 1986; 87:2019. First citation in article
9. Duggan KA, Mendelsohn FA, Levens NR. Angiotensin receptors and angiotensin I-converting enzyme in rat intestine. Am J Physiol 1989; 257:G50410. First citation in article
10. Sechi LA, Valentin JP, Griffin CA, Schambelan M. Autoradiographic characterization of angiotensin II receptor subtypes in rat intestine. Am J Physiol 1993; 265:G217. First citation in article
11. Schinke M, Doods HN, Ganten D, Wienen W, Entzeroth M. Characterization of rat intestinal angiotensin II receptors. Eur J Pharmacol 1991; 204:16570. First citation in article
12. Timmermans PB, Wong PC, Chiu AT, et al. Angiotensin II receptors and angiotensin II receptor antagonists. Pharmacol Rev 1993; 45:20551. First citation in article
13. Jin X-H, Wang Z-Q, Siragy HM, Guerrant RL, Carey RM. Regulation of jejunal sodium and water absorption by angiotensin subtype receptors. Am J Physiol 1998; 275:R51523. First citation in article
14. Levens NR, Peach MJ, Carey RM, Poat JA, Munday KA. Response of rat jejunum to angiotensin II: role of norepinephrine and prostaglandins. Am J Physiol 1981; 240:G1724. First citation in article
15. Levens NR, Peach MJ, Carey RM. Interactions between angiotensin peptides and the sympathetic nervous system mediating intestinal sodium and water absorption in the rat. J Clin Invest 1981; 67:1197207. First citation in article
16. Lima AAM, Lyerly DM, Wilkins TD, Innes DJ, Guerrant RL. Effects of Clostridium difficile toxins A and B in rabbit small and large intestine in vivo and on cultured cells in vitro. Infect Immun 1988; 56:5828. First citation in article
17. Jaszewski R, Tolia V, Ehrinpreis MN, et al. Increased colonic mucosal angiotensin I and II concentrations in Crohn's colitis. Gastroenterology 1990; 98:15438. First citation in article
18. Wershil BK, Castagliuolo I, Pothoulakis C. Direct evidence of mast cell involvement in Clostridium difficile toxin A-induced enteritis in mice. Gastroenterology 1998; 114:95664. First citation in article
19. Tonnesen MG, Klempner MS, Austen KF, Wintroub BU. Identification of a human neutrophil angiotension II-generating protease as cathepsin G. J Clin Invest 1982; 69:2530. First citation in article
20. Fukami H, Okunishi H, Miyazaki M. Chymase: its pathophysiological roles and inhibitors. Curr Pharm Des 1998; 4:43953. First citation in article
21. Graciano ML, Cavaglieri R de C, Delle H, et al. Intrarenal rennin-angiotensin system is upregulated in experimental model of progressive renal disease induced by chronic inhibition of nitric oxide synthesis. J Am Soc Nephrol 2004; 15:180515. First citation in article
22. Shinozaki K, Ayajiki K, Nishio Y, Sugaya T, Kashiwagi A, Okamura T. Evidence for a causal role of the rennin-angiotensin system in vascular dysfunction associated with insulin resistance. Hypertension 2004; 43:25562. First citation in article
23. Levens NR, Peach MJ, Carey RM, Poat JA, Munday KA. Stimulation of intestinal sodium and water transport in vivo by angiotensin II and analogs. Endocrinology 1980; 107:194653. First citation in article
24. Ewert S, Laesser M, Johansson B, Holm M, Aneman A, Fandriks L. The angiotensin II receptor type 2 agonist CGP 42112A stimulates NO production in the porcine jejunal mucosa. BMC Pharmacol 2003; 3:18. First citation in article
25. Wallace JL, Miller MJ. Nitric oxide in mucosal defense: a little goes a long way. Gastroenterology 2000; 119:51220. First citation in article