- Email: [email protected]
Leptin mediates Clostridium difficile toxin A–induced enteritis in mice
GASTROENTEROLOGY 2003;124:683– 691
Leptin Mediates Clostridium difficile Toxin A–Induced Enteritis in Mice ANDREAS MYKONIATIS,* PAULINE M. ANTON,* MICHAEL WLK,* CHI CHUNG WANG,* ¨ HER,‡ MARIA VENIHAKI,§ SIMOS SIMEONIDIS,* JEFF ZACKS,㛳 LINDA UNGSUNAN,‡ SUSANN BLU DEZHENG ZHAO,* STAVROS SOUGIOULTZIS,* KATIA KARALIS,§ CHRISTOS MANTZOROS,‡ and CHARALABOS POTHOULAKIS* *Divisions of Gastroenterology and ‡Endocrinology, Beth Israel Deaconess Medical Center; §Division of Endocrinology, Children’s Hospital, Harvard Medical School; and 㛳Mallory Institute, Department of Pathology, Boston University School of Medicine, Boston, Massachusetts
Background & Aims: Leptin regulates energy homeostasis and participates in the regulation of the hypothalamicpituitary-adrenal axis. Although hyperleptinemia is described in experimental colitis, its role in the pathophysiology of enterotoxin-mediated diarrhea and inflammation remains unclear. We examined the role of leptin in the inflammatory diarrhea induced by toxin A from Clostridium difficile, the causative agent of antibioticrelated colitis. Methods: Toxin A (10 g) or buffer were administered in ileal loops of leptin-deficient (ob/ob), leptin-resistant (db/db), or wild-type mice and enterotoxic responses were measured. Results: In toxin A–treated wild-type mice, circulating leptin and corticosterone levels were increased compared with bufferinjected animals. Toxin A also stimulated increased mucosal expression of the Ob-Rb at the messenger RNA (mRNA) and protein level. Ob/ob and db/db mice were partially protected against toxin A–induced intestinal secretion and inflammation, and this effect was reversed by leptin administration in ob/ob, but not db/db, mice. Basal- and toxin A–stimulated plasma corticosterone levels in ob/ob and db/db mice were higher compared with toxin A–treated wild-type mice. To assess whether the effect of leptin in intestinal inflammation is mediated by corticosteroids we performed adrenalectomy experiments in db/db and wild-type mice. Our results suggested that the diminished intestinal response to toxin A in db/db mice was related only in part to increased levels of corticosteroids. Conclusions: Leptin plays an important role in regulating the severity of enterotoxin-mediated intestinal secretion and inflammation by activating both corticosteroid-dependent and -independent mechanisms.
lostridium difficile, the primary cause of antibioticassociated colitis, is a major pathogen in hospitals and nursing homes.1 This anaerobic bacterium mediates diarrhea and colitis by releasing 2 exotoxins, toxin A and toxin B.1 The cellular proinflammatory mechanism of these toxins involves binding to their receptors,2 effects
C
on cellular mitochondria,3 generation of reactive oxygen species,3 and activation of the transcription nuclear factor B.4 Animal experiments suggest that toxin A causes intestinal secretion and inflammation by stimulating a complex cascade involving neuropeptides, such as substance P and neurotensin, and activating lamina propria macrophages, mast cells, and neutrophils.5 Recent results also showed that endogenous glucocorticoids modulate the secretory and inflammatory effects of C. difficile toxin A in rat ileum by inhibiting toxin A–induced neutrophil infiltration and expression of proinflammatory cytokines.6 Leptin, the product of the ob gene,7 is an adiposesecreted tissue peptide that signals to the brain the magnitude of fat stores8 and regulates energy homeostasis.9 Leptin mediates its effects by binding to a specific receptor belonging to the cytokine receptor superfamily.10 Several forms of the leptin receptor (Ob-R) exist, but the signaling Ob-Rb isoform, which contains a long intracytoplasmic domain, has been shown to be important for mediation of leptin’s effects.11 Several studies suggest that leptin and the Ob-Rb may be associated with inflammatory conditions. Leptin acting on the hypothalamus regulates neuroendocrine systems important in immune function such as the adrenocorticotropic hormone– corticosterone axis.12,13 More specifically, a leptin-deficient or leptin-resistant state is associated with increased corticosterone levels whereas leptin replacement results in partial normalization of corticosterone levels.12 Leptin-deficient (ob/ob) mice and leptin-resistant (db/db) mice have impaired T-cell imAbbreviations used in this paper: db/db, leptin resistant; IL, interleukin; MPO, myeloperoxidase; ob/ob, leptin deficient; Ob-R, leptin receptor; Ob-Rb, long form of leptin receptor. © 2003 by the American Gastroenterological Association 0016-5085/03/$30.00 doi:10.1053/gast.2003.50101
684
MYKONIATIS ET AL.
munity associated with marked reduction in size and cellularity of the thymus.14,15 In addition, peritoneal macrophages isolated from leptin-deficient and leptinresistant mice have decreased phagocytic activity, and expression of proinflammatory cytokines including interleukins (ILs) and tumor necrosis factor ␣.16 Leptin administration to ob/ob mice normalizes these responses, reduces thymocyte apoptosis, and substantially increases thymic cellularity.17 Leptin administration to rats results in increased levels of the proinflammatory cytokine IL-1 in the rat hypothalamus.18 Further, administration of leptin increases Th1 and suppresses Th2 cytokine production by T cells in culture.19 Thus, available evidence indicates that leptin may have both direct and indirect action on the immune system. Moreover, factors that activate the immune system such as injection of lipopolysaccharide increases leptin messenger RNA (mRNA) in adipose tissues and leptin levels in serum.20,21 The signaling Ob-Rb isoform of the leptin receptor is expressed in the gastrointestinal tract of animals and humans and on colonic epithelial cells,22–25 and leptin binding to this receptor inhibits absorption of sugars23 and stimulates growth of colonocytes.24 Further, leptin blood levels are elevated transiently at the early stages of trinitrobenzene sulfonic acid–mediated colitis in rats, and this increase correlates with the degree of inflammation and anorexia.26 Decreasing leptin levels by a cholecystokinin B antagonist and a 3 agonist treatment are associated with an improvement in the severity of colitis.27 Siegmund et al.28 showed that the inflammatory responses in sodium dextran sulfate– and trinitrobenzene sulfonic acid–induced colitis were significantly lower in leptin-deficient ob/ob mice compared with wild-type mice and injection of leptin restored the colonic inflammatory response. However, whether leptin and Ob-Rb participate in the pathogenesis of enterotoxin-mediated intestinal secretion and inflammation is not known. As well, the role of the hypothalamo-pituitary-adrenal A axis in the development of intestinal inflammation as it relates to leptin deficiency has not been examined. In this study, we examined the role of leptin and its long receptor isoform in fluid secretion and inflammation induced by C. difficile toxin A. We used the ileal loop model of toxin A–induced ileitis in wild-type as well as leptin-deficient ob/ob mice and leptin-resistant db/db mice, which have a truncated and inactive form of the leptin receptor,29 to investigate the possibility that leptin participates in intestinal inflammation. We also correlated the levels of leptin and ileal Ob-Rb expression with the time-course appearance of intestinal secretion,
GASTROENTEROLOGY Vol. 124, No. 3
epithelial cell damage, and inflammation in response to C. difficile toxin A. Finally, we examined whether this role of leptin is corticosterone independent or not by studying adrenalectomized mice. Our results point to a novel proinflammatory role of leptin in the pathophysiology of enterotoxin-mediated enteritis that is in part corticosterone independent.
Materials and Methods Toxin A was purified from culture supernatants of C. difficile strain 10,463 as previously described.30 Protein concentrations were determined with the DC protein assay (BioRad Laboratories, Hercules, CA).
Mouse Ileal Loops Twelve-week-old male C57BL/6J (20 –25 g), ob/ob (B6.V-Lepob, 45–55 g) and db/db (BKS.Cg-m⫹/⫹Leprdb, 45–55 g) mice (Jackson Laboratory, Bar Harbor, ME) were housed on a 12-hour light-dark cycle for 3 days before surgery. Experiments were performed between 9:30 AM–11:30 AM to minimize influence of the circadian rhythm. Mice were fasted (16 h) and then anesthetized with a mixture of ketamine and xylazine,31 and after laparotomy, one 3- to 5-cm ileal loops were formed32 and injected with 0.15 mL of phosphate buffer saline (pH 7.4) containing 10 g of purified toxin A or buffer alone. The abdomen was then sutured, animals were placed on a heating pad at 37°C, and after 4 hours were killed with CO2 inhalation. Ob/ob and db/db mice (n ⫽ 5) were injected intraperitoneally with 1 g/g of body weight12,33,34 mouse recombinant leptin (Sigma, St. Louis, MO), 20, 14, and 0.5 hours before toxin A administration. Previous studies showed that this dose, in fasted mice, reestablishes the circulating levels of leptin observed in nonfasted animals.12 At death, fluid secretion was estimated as the loop weightto-length ratio (mg/cm) as previously described.31,32 Pieces of tissues were removed for myeloperoxidase (MPO) activity measurement, and full-thickness loop sections were fixed in formalin, paraffin-embedded, stained with H&E, and graded histopathologically by using parameters associated with toxin A–induced enterotoxicity.32 Moreover, ileal sections from C57BL/6J mice (n ⫽ 4) were snap frozen for immunohistochemistry and mRNA amplification for OB-Rb. Tissues were stored at – 80°C until determination.
Adrenalectomy C57BL/6J and db/db mice (n ⫽ 6) were adrenalectomized bilaterally under general anesthesia by the retroperitoneal route,35 whereas sham-operated C57BL/6J and db/db mice (n ⫽ 4) underwent the same procedure without removing the adrenals. Adrenalectomized animals were given 0.9% NaCl to compensate for the salt loss. Six days after adrenalectomy ileal loops were formed, exposed to toxin A or buffer, and fluid secretion was measured at 4 hours.36 The institutional Animal Care and Use Committee approved the animal studies.
March 2003
Leptin and Corticosterone Measurements Ileal loops of male wild-type, ob/ob, and db/db mice were injected with 10 g of purified toxin A or buffer (control). Blood samples (0.2 mL) were collected from the retroorbital plexus before buffer or toxin A injection and after 30 minutes, 2 hours, and 4 hours for leptin and corticosterone assays. They were centrifuged (2000g for 10 min at 4°C), and aliquots of plasma were stored at ⫺80°C. Leptin and corticosterone levels were measured using commercially available radioimmunoassay kits from Linco Research Institute (St. Louis, MO), and ICN Biomedicals, Inc. (Costa Mesa, CA), respectively.
Myeloperoxidase Measurements Tissue MPO activity was determined by a modified method of Bradley et al.37 Briefly, ileal loop samples were homogenized (Ultra Thurrax; Tekmar Co., Cincinnati, OH) and frozen and thawed 3 times. Samples were sonicated (Heat Systems; Ultrasonics, Plain View, NY) in 1.5 mL of 50 mmol/L phosphate buffer containing 0.5% of hexadecyl-trimethyl ammonium bromide. Samples were then centrifuged at 10,000g (15 min at 4°C) and supernatants were collected. Supernatants were then diluted into the same phosphate buffer containing 0.167 mg/mL of O-dianisidine dihydrochloride and 5.10⫺4% of hydrogen peroxide. MPO activity was measured spectrophotometrically (Lambda 20, UV/VIS spectrophotometer; Perkin Elmer, Norwalk, CT) at 450 nm using human MPO (0.1 U/10 L; Sigma, St. Louis, MO) as a standard. Proteins were evaluated using the DC protein assay (Bio-Rad Laboratories, Hercules, CA).
Total Mucosal RNA Extraction and ReverseTranscription Polymerase Chain Reaction Amplification for Ob-Rb Messenger RNA Ileal loops were injected with either phosphate-buffered saline or toxin A, and after 1 or 2 hours loops were removed, opened, washed in ice-cold phosphate-buffered saline, and the mucosa was scraped with RNAse-free glass slides (Fisher, Pittsburgh, PA). Total RNA was isolated as previously described.31 The method described by Bjorbaek et al.38 was applied for reverse-transcription polymerase chain reaction. Briefly, a final volume of 100 L complementary DNA (cDNA) was synthesized from 1 g of total mRNA using the Advantage reverse-transcription polymerase chain reaction kit from Clontech (Palo Alto, CA). Ob-Rb cDNA was amplified using the following primers: upstream: 5⬘-ACA GCG TGC TTC CTG GGT CTT C-3⬘, downstream: 5⬘-TGG ATA AAC CCT TGC TCT TCA-3⬘. The 201-bp -actin cDNA was amplified using the following primers: upstream: 5⬘-CGT ACC ACG GGC ATT GTG ATG G-3⬘, and downstream: 5⬘-TTT GAT GTC ACG CAC GAT TTC CC-3⬘. Preliminary polymerase chain reaction experiments showed that the rate of amplification was linear for -actin when applied for fewer than 20 cycles and for Ob-Rb when applied for fewer than 35 cycles. Each 50-L polymerase chain reaction was performed
LEPTIN AND INFLAMMATION
685
with 5.0 L of template cDNA. Conditions were: 10 mmol/L Tris-HCl (pH 8.8), 50 mmol/L KCl, 1.5 mmol/L MgCl2 , 0.01% gelatin, 0.2 mmol/L deoxynucleoside triphosphate, 20 pmol of each primer, 2.5 U of Taq polymerase (Stratagene, La Jolla, CA), and 0.5 L of 32P-dGTP (29.6 Tbq/mmol, 370 Mbq/mL) (NEN, Boston, MA). The mixture was overlaid with 50 L of mineral oil and, after initial denaturation at 96°C for 4 minutes, samples were subjected to 18 cycles of amplification for -actin and to 30 cycles for Ob-Rb (denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and extension at 72°C for 45 sec). Ten microliters of the reaction were then combined with 5 L of sequencing stop solution (Amersham International, Buckinghamshire, UK) and heated to 85°C for 5 minutes, before loading 4 L onto a 4% urea-acrylamide gel (38 ⫻ 31 ⫻ 0.03 cm). The electrophoresis conditions were 60 W of constant power for 1.45 hours. After electrophoresis, gels were transferred to filter paper, dried, and subjected to 32P quantitation by PhosphoImager Analysis (Molecular Dynamics, Sunnyvale, CA).
Immunohistochemistry Toxin A or buffer were injected into ileal loops of wild-type mice (n ⫽ 3 per group), and after 15, 30, and 60 minutes after toxin A exposure or 60 minutes after buffer exposure, mice were killed and freshly frozen sections were prepared. Frozen ileal sections were cut (5 m) and fixed in Teck tissue fixative (Teck, Redding, CA) for 10 minutes. Sections were then washed in Tris buffer saline (pH 7.5) containing 0.1% Tween 20 (TBST; Dako Corporation, Carpenteria, CA), and incubated with 2% normal donkey serum in TBST for 30 minutes at 22°C. Sections were incubated for 2 hours with 1:20 dilution of rabbit anti-human Ob-Rb antibody or control rabbit immunoglobulin G (Linco Research, Inc), washed in 1 ⫻ TBST and incubated for 30 minutes with 1:50 dilution of a fluorescein isothiocyanate– conjugated donkey anti-rabbit immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Sections were mounted with antibleaching solution and images were viewed under a confocal microscope (model MRC 1024; Bio-Rad Microsciences, Cambridge, MA) using Plan-Neofluar objectives (20⫻) and were stored digitally in Bio-Rad COMOS software. Specificity of the antibody was assessed in separate experiments in which slides were exposed to the anti-human Ob-Rb antibody, which was preincubated with an excess of leptin protein.
Statistical Analyses Results are presented as mean ⫾ SEM and, unless otherwise stated, data were analyzed using the SIGMA-STAT statistics software program (Jandel Scientific Software, San Rafael, CA). Analysis of variance followed by a Tukey post hoc test was used for intergroup comparisons. A value of P less than 0.05 was considered significant.
686
MYKONIATIS ET AL.
GASTROENTEROLOGY Vol. 124, No. 3
toxin A stimulated increased fluid secretion, MPO activity, and caused severe histologic changes in the ileum of wild-type mice compared with buffer exposure (Figure 1A), consistent with prior observations.31,32 In contrast, ob/ob and db/db mice had reduced fluid secretion, attenuated epithelial damage, mucosal congestion and edema, and neutrophil (polymorphonuclear leukocyte) infiltration in response to toxin A compared with wild-type mice (Figure 1A, Table 1). Leptin replacement before toxin A injection in ob/ob mice resulted in increased susceptibility to toxin A–induced fluid secretion whereas leptin administration did not alter toxin A–induced ileal secretion in the leptin receptor– deficient db/db mice (Figure 1A). Histologic quantification of buffer-exposed ileal tissues of ob/ob and db/db mice was no different than buffer-injected ileal tissues of wild-type mice (Table 1). Together, these results provide direct evidence for the importance of leptin in inflammatory diarrhea caused by C. difficile toxin A. Increased Plasma Leptin During Toxin A–Mediated Enteritis
Figure 1. (A) Reduced toxin A–induced ileal fluid secretion in db/db and ob/ob mice. Ileal loops of anesthetized mice were injected with buffer or toxin A and, after 4 hours, mice were killed and fluid secretion was measured by loop weight-to-length ratio (mg/cm). Some ob/ob and some db/db mice were injected with 3 single intraperitoneal leptin injections (1 g/g of body weight), 20, 14, and 0.5 hours before toxin A administration. Data are means ⫾ SEM of 5 animals for each group. **P ⬍ 0.01 vs. the respective buffer; ⫹P ⬍ 0.05 vs. toxin A of C57BL/6J mice; #P ⬍ 0.05 vs. toxin A of ob/ob mice. 䊐, buffer; ■, toxin A; , leptin ⫹ toxin A. (B) Increased leptin plasma levels during toxin A–induced enteritis. Ileal loops of C57/BL6J anesthetized mice were injected with buffer or toxin A and, at the indicated time points, blood samples were collected and leptin concentration was measured in the plasma by radioimmunoassay. Data are means ⫾ SEM of 5 animals for each group. *P ⬍ 0.05 vs. the respective buffer or basal leptin levels (after anesthesia, before injection of loops). 䊐, buffer; ■, toxin A.
Results Reduced Toxin A–Induced Fluid Secretion and Intestinal Inflammation in LeptinDeficient and Leptin-Resistant Mice To assess directly the role of leptin in the toxin A model of intestinal inflammation, we injected ileal loops of anesthetized wild-type, leptin-deficient (ob/ob), and leptin-resistant (db/db) mice, with either buffer or purified toxin A and, after 4 hours, we measured toxin A–associated ileal responses. Basal fluid secretion in response to buffer injection was comparable between wildtype and ob/ob and db/db mice (Figure 1A). Moreover,
Because these results (Figure 1A, Table 1) suggested participation of leptin in toxin A–mediated enteritis, we next examined the plasma levels of leptin after toxin A administration. Thirty minutes after ileal toxin A injection in wild-type mice plasma leptin levels were elevated significantly, and further increased after 4 hours (Figure 1B). We also measured circulating leptin levels in db/db mice before and after toxin A exposure. Db/db mice were hyperleptinemic at baseline (17.01 ⫾ 0.43 in buffer-treated db/db mice vs. 1.32 ⫾ 0.16 ng/dL in buffer-treated wild-type mice, n ⫽ 5 per group, P ⱕ 0.05), consistent with prior observations of high leptin levels in these mice.8,25 Moreover, plasma leptin levels were not altered significantly in response to toxin A treatment at any of the time points studied (30 min– 4 h, n ⫽ 5 per group). We suspect that the reason for the inability of toxin A to increase leptin levels in db/db mice, in contrast to wild-type mice, is related to the already established hyperleptinemic state of these animals. Role of Corticosteroids in Toxin A–Induced Secretion in Leptin Receptor–Deficient Mice Because the inflammatory and secretory responses to toxin A are modulated by endogenous corticosteroids,6 we examined corticosteroid plasma levels in response to toxin A in wild-type, ob/ob, and db/db mice 2 hours after toxin A administration. Plasma corticosterone in buffer-injected ob/ob and db/db mice was higher compared with buffer-injected wild-type mice (Figure 2A),
March 2003
LEPTIN AND INFLAMMATION
687
Table 1. Reduced Toxin A–Mediated Histologic Responses in ob/ob and db/db Mice Treatment
Epithelial damage
Congestion and edema
Neutrophil infiltration
MPO U/g protein
Buffer C57BL/6J Toxin A C57BL/6J Buffer db/db Toxin A db/db Buffer ob/ob Toxin A ob/ob
0.4 ⫾ 0.24 2.8 ⫾ 0.16a 0.3 ⫾ 0.2 1.16 ⫾ 0.16b,c 0.46 ⫾ 0.18 1.31 ⫾ 0.11a,c
0.5 ⫾ 0.28 2.5 ⫾ 0.2a 0.1 ⫾ 0.1 1.14 ⫾ 0.09a,c 0.2 ⫾ 0.13 1.37 ⫾ 0.18a,c
0.0 ⫾ 0.0 2.6 ⫾ 0.2a 0.1 ⫾ 0.1 1.0 ⫾ 0.0b,c 0.4 ⫾ 0.16 1.28 ⫾ 0.18b,c
29.16 ⫾ 5.84 108.2 ⫾ 16.5b 30 ⫾ 4.61 51.3 ⫾ 2.07b,d 32.5 ⫾ 3.57 75.56 ⫾ 3.4b,d
NOTE. Ileal loops were injected with 10 g of toxin A or phosphate-buffered saline. After 4 hours, animals were killed, ileal loops were removed, and histologic severity of enteritis was graded by a score of 0 –3 for epithelial cell damage, congestion and edema of the mucosa, and mucosal neutrophil infiltration. For MPO, ileal loops were removed and homogenized in ice-cold phosphate-buffered saline containing hexadecyl-trimethyl ammonium bromide (0.5%) (see Methods section). Homogenates were then centrifuged and aliquots from the supernatants were taken for MPO and protein measurements. Data are means ⫾ SEM per group; n ⫽ 8 –10 for histologic quantitation and 4 for MPO measurements, each with duplicate determinations. The nonparametric Mann-Whitney test was used to calculate the histologic differences. aP ⬍ 0.01 vs. control. bP ⬍ 0.05 vs. control. cP ⬍ 0.01 vs. toxin A of C57BL/6J mice. dP ⬍ 0.05 vs. toxin A of C57BL/6J mice.
consistent with the high basal corticosterone level of ob/ob and db/db mice.39,40 Moreover, exposure of ileal loops from ob/ob and db/db mice to toxin A stimulated a further increase in circulating corticosterone levels (Figure 2A). Because the higher corticosterone concentrations in ob/ob and db/db mice could raise the possibility that the relative protection of this strain could be caused by their increased corticosterone and not leptin levels per se, we sought to address this possibility directly using adrenalectomized db/db and wild-type mice. Plasma corticosterone level before toxin A and buffer injection was 1.30 ⫾ 0.16 and 1.60 ⫾ 0.25 g/dL for adrenalectomized C57BL/6J and db/db mice, respectively, compared with 22.0 ⫾ 3.8, and 85.0 ⫾ 5.4 g/dL for nonadrenalectomized C57BL/6J, and db/db mice, respectively. Intestinal secretion was similar in sham-operated or adrenalectomized mice injected with buffer (Figure 2B). Adrenalectomized mice of both genotypes had increased toxin A–induced fluid secretion, compared with the respective sham-operated animals, indicating that glucocorticoids contribute to the reduced secretory responses to toxin A in db/db mice. However, toxin A–mediated secretion in db/db mice was significantly lower compared with fluid secretion in toxin A– exposed wild-type mice (Figure 2B). This suggests that an additional glucocorticoid-independent mechanism may be responsible for the reduced toxin A–associated responses in leptin-resistant db/db mice. Increased Leptin Receptor Expression During Toxin A–Induced Intestinal Inflammation Because the leptin receptor Ob-Rb was shown to be involved in toxin A–induced secretion and inflamma-
tion, we examined the expression and localization of the Ob-Rb in the ileal mucosa of wild-type mice after toxin A exposure. Reverse-transcription polymerase chain reaction showed a gradual increase of Ob-Rb mucosal mRNA levels in toxin A– exposed ileum that became statistically significant (by 42.2% in wild-type mice vs. controls, P ⬍ 0.001) after 2 hours (Table 2). Immunohistochemical studies showed Ob-Rb–positive cells in buffer-exposed ileum and increased expression of these receptors (localized on epithelial as well as on subepithelial cells) after 15 and 30 minutes of toxin A exposure. However, Ob-Rb expression was less evident 60 minutes after toxin A exposure, probably owing to epithelial cell damage in response to the toxin.
Discussion We have reported previously that C. difficile toxin A stimulates fluid secretion and elicits an acute inflammatory response in animal intestine characterized by neutrophil infiltration and epithelial cell destruction.5 The results presented here show that animals either genetically lacking leptin itself or resistant to leptin’s effects have substantially reduced responses to toxin A, which is normalized in response to the administration of leptin in leptin-deficient ob/ob mice but not in leptinresistant db/db mice. This indicates that leptin mediates to a large extent ileal fluid secretion and histologic changes in response to this toxin. This study showed a proinflammatory role for leptin in the pathophysiology of intestinal secretion and inflammation in response to a bacterial enterotoxin. Our results are consistent with the findings of Barbier et al.26,27 and Siegmund et al.,28 suggesting participation of leptin in the pathogenesis of experimental colitis.
688
MYKONIATIS ET AL.
Figure 2. (A) Plasma corticosterone levels in ob/ob and db/db mice in response to toxin A. One closed ileal loop was prepared in the distal ileum of anesthetized mice and injected with either buffer or toxin A (10 g). After 2 hours mice were killed and blood samples were collected by heart puncture. Data are means ⫾ SEM of 5 loops per experimental condition. ***P ⬍ 0.001 vs. buffer for C57BL/6J, ob/ob, and db/db mice; ⫹⫹⫹P ⬍ 0.001 vs. buffer-injected C57BL/6J; †††P ⬍ 0.001 vs. toxin A–injected C57BL/6J mice. 䊐, buffer; ■, toxin A. (B) Adrenalectomy in db/db mice reverses reduced toxin A–mediated fluid secretion. Wild-type or db/db mice were subjected to adrenalectomy whereas another group was sham-operated without removing the adrenals. Mice were left to recover for 6 days and then fluid secretion (weight-to-length ratio) was measured 4 hours after toxin A (10 g, n ⫽ 7– 8 per group) or buffer (n ⫽ 3 per group) injection into ileal loops. Data are expressed as mean ⫾ SEM. ††P ⬍ 0.01 vs. buffer-infected loops of sham adrenalectomized C57BL/6J; **P ⬍ 0.01 vs. toxin A–injected loops of sham adrenalectomized of both genotypes; ⫹⫹P ⬍ 0.01 vs. toxin A–injected adrenalectomized C57BL/6J mice, ##P ⬍ 0.01 vs. toxin A–injected sham C57BL/6J mice. 䊐, sham buffer; , sham toxin A; , adrenalectomized buffer; , adrenalectomized toxin A.
We observed increased basal and C. difficile toxin A–stimulated corticosterone levels in both ob/ob and db/db mice compared with wild-type mice (Figure 2B). This association between the hypothalamo-pituitary-adrenal axis and circulating leptin has been established previously. Thus, hypercorticosteronemia observed in ob/ob and fasting mice can be reduced significantly by
GASTROENTEROLOGY Vol. 124, No. 3
chronic leptin administration,12,40 whereas leptin-resistant Zucker rats show an excessive hypothalamo-pituitary-adrenal axis activation, along with hypercorticosteronemia that is not corrected by leptin administration.41 Although high basal corticosterone levels in ob/ob and db/db mice were expected,39,40 the increase of glucocorticoids in the absence of functional leptin or functional leptin receptors suggest that other mediators enhance the secretion of glucocorticoids in C. difficile–induced inflammation. Indeed, we have shown recently that increased peripheral corticotropin-releasing hormone expression was associated with increased corticosteroid levels in the toxin A model of enteritis.42 We also have shown that endogenous glucocorticoids diminish the secretory and inflammatory effect of toxin A.6 This protective effect of glucocorticoids also has been described in other models of inflammation, including animals injected with RU 48643,44 or mice deficient in corticotropin-releasing hormone or corticosterone,45,46 which develop an increased innate inflammatory response. To determine whether a glucocorticoid-independent component is involved in the effects of leptin in toxin A–mediated intestinal secretion, we used the adrenalectomized mouse model. Our results using adrenalectomized wild-type and db/db mice point out that only part of toxin A–induced fluid secretion is linked to release of glucocorticoids, whereas a glucocorticoid-independent, leptin-associated mechanism appears also to be involved. The nature of this mechanism remains to be identified, but several studies have suggested that leptin–Ob-Rb interactions may affect directly or alter inflammatory responses. For example, increased IL-1 mRNA expression has been shown in glial cells in response to leptin.47 Leptin also can enhance the effect of LPS on the production of different cytokines by murine peritoneal macroTable 2. Increased Toxin A–Mediated Ob-Rb mRNA Accumulation in Mouse Ileum Time points 1h 2h
Treatment Buffer Toxin A Buffer Toxin A
Leptin mRNA 99.7 ⫾ 6.6 152.4 ⫾ 20.4 100.0 ⫾ 6.8 173.4 ⫾ 15.8a
NOTE. Ileal loops were injected with phosphate-buffered saline or toxin A and, after 1 and 2 hours, mRNA was extracted and reverse transcribed to obtain cDNA. Ob-Rb mRNA was quantified by reversetranscription polymerase chain reaction as described in Materials and Methods. Data represent the mean values (from arbitrary Phospholmager units) of each group corrected for the -actin values and then normalized to the mRNA expression of the buffer-treated group, which is considered to be 100. Data are mean ⫾ SEM, n ⫽ 5 per group. Statistical significance was assessed by unpaired, 2-tailed Student t test using the Statview program (Abacus, CA). aP ⬍ 0.001 vs. buffer C57BL/6J mice.
March 2003
LEPTIN AND INFLAMMATION
689
Figure 3. Immunohistochemistry of Ob-Rb after toxin A administration. (A) Buffer or (B–D) toxin A were injected into ileal loops of wild-type mice. After (B) 15, (C) 30, and (D) 60 minutes the animals were killed and sections were processed for immunohistochemistry using a rabbit anti-human leptin receptor antibody. Leptin receptor staining is present in control (buffer-exposed) mouse ileum. However, 15 and 30 minutes after toxin A injection, a substantial increase in the expression of leptin receptor compared with control was noted. Ob-Rb expression was less evident 60 minutes after toxin A exposure. Leptin receptor immunoreactivity is present on intestinal epithelial cells as well as in cells of the lamina propria. Results are representative of 3 separate experiments. (A) Cross-section. (B–D) Longitudinal sections. (Magnification 200⫻).
phages, including IL-1Ra, IL-6, IL-12, and tumor necrosis factor ␣.16 Moreover, exposure of the macrophage cell line RAW264.7, or monocytic THP-1 cells, or peripheral blood mononuclear cells to leptin causes secretion of IL-1Ra.16,48 In human monocytes, leptin can induce the expression of cell-surface markers and cytokines, including IL-6 and tumor necrosis factor ␣49 and, in human lymphocytes, exogenous leptin increases Th1 and suppresses Th2 cytokine production.19 Further, administration of leptin in ob/ob mice reduced apoptosis of thymocytes and increased the CD4⫾CD8⫾/CD4⫺CD8⫺ ratio and thymic cellularity.17 Finally, our immunohistochemical experiments (Figure 3) show that expression of Ob-Rb is increased in the intestinal epithelium 15 minutes after toxin A exposure, whereas increases in Ob-Rb mRNA expression are evident at 1 and 2 hours (Table 2). These results suggest that expression of preformed Ob-Rb receptors may be stimulated at the early stages of toxin A–mediated intestinal inflammation. Our observation that both leptin levels and immunoreactive Ob-Rb are increased significantly within the first 30 minutes of toxin A exposure suggests a direct proinflammatory effect of leptin on intestinal epithelial cells via binding to this receptor.24,25 Indeed, in intestinal cells, leptin administration activates the JAK/STAT pathway that only can be stimulated by the long form of the leptin receptor, Ob-Rb.25 This study showed altered expression of the Ob-Rb during inflammation. However, the mechanism underlying this response remains to be elucidated. The present results may be relevant to the pathophysiology of human C. difficile colitis. Our observations, together with recent studies showing participation of leptin in animal models of inflammatory bowel disease,26 –28 suggest that use of specific leptin or leptin receptor antagonists may represent a novel therapeutic
approach for the treatment of intestinal inflammatory conditions.
References 1. Kelly CP, Pothoulakis C, LaMont JT. Clostridium difficile colitis. N Engl J Med 1994;330:257–262. 2. Pothoulakis C, Gilbert RJ, Cladaras C, Castagliuolo I, Semenza G, Hitti Y, Montcrief JS, Linevsky J, Kelly CP, Nikulasson S, Desai H, Wilkins TD, LaMont JT. Rabbit sucrase-isomaltase contains a functional intestinal receptor for Clostridium difficile toxin A. J Clin Invest 1996;98:641– 649. 3. He D, Hagen SJ, Pothoulakis C, Chen M, Medina ND, Warny M, LaMont JT. Clostridium difficile toxin A causes early damage to mitochondria in cultured cells. Gastroenterology 2000;119:139 – 150. 4. Warny M, Keates A, Keates S, Castagliuolo I, Zacks JK, Aboudola S, Qamar A, Pothoulakis C, LaMont T, Kelly CP. p38 MAP kinase activation by Clostridium difficile toxin A mediates monocyte necrosis, Il-8 production, and enteritis. J Clin Invest 2000;105: 1147–1156. 5. Pothoulakis C, Lamont JT. Microbes and microbial toxins: paradigms for microbial-mucosal interactions II. The integrated response of the intestine to Clostridium difficile toxins. Am J Physiol 2001;280:178 –183. 6. Castagliuolo I, Karalis K, Valenick L, Pasha A, Nikulasson S, Wlk M, Pothoulakis C. Endogenous corticosteroids modulate Clostridium difficile toxin A-induced enteritis in rats. Am J Physiol 2001; 280:G539 –G545. 7. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425– 432. 8. Frederich RC, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat Med 1995;1: 1311–1314. 9. Friedman JM. Leptin, leptin receptors, and the control of body weight. Nutr Rev 1998;53:S38 –S46. 10. Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J. Identification and expression cloning of a leptin receptor, OB-R. Cell 1995;83: 1263–1271. 11. Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, Lee JI, Friedman JM. Abnormal splicing of the leptin receptor in diabetic mice. Nature 1996;379:632– 635. 12. Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-
690
13.
14. 15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
MYKONIATIS ET AL.
Flier E, Flier JS. Role of leptin in the neuroendocrine response to fasting. Nature 1996;382:250 –252. Heiman ML, Ahima RS, Craft LS, Schoner B, Stephens TW, Flier JS. Leptin inhibition of the hypothalamic-pituitary-adrenal axis in response to stress. Endocrinology 1997;138:3859 –3863. Chandra RK. Cell-mediated immunity in genetically obese C57BL/6J ob/ob mice. Am J Clin Nutr 1980;33:13–16. Fernandes G, Handwerger BS, Yunis EJ, Brown DM. Immune response in the mutant diabetic C57BL/Ks-dt⫹ mouse. Discrepancies between in vitro and in vivo immunological assays. J Clin Invest 1978;61:243–250. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang DJ, Klein AS, Bulkley GB, Bao C, Noble PW, Lane MD, Diehl AM. Leptin regulates proinflammatory immune responses. FASEB J 1998;12:57– 65. Howard JK, Lord GM, Matarese G, Vendetti S, Ghatei MA, Ritter MA, Lechler RI, Bloom SR. Leptin protects mice from starvationinduced lymphoid atrophy and increases thymic cellularity in ob/ob mice. J Clin Invest 1999;104:1051–1059. Luheshi GN, Gardner JD, Rushforth DA, Loudon AS, Rothwell NJ. Leptin actions on food intake and body temperature are mediated by IL-1. Proc Natl Acad Sci U S A 1999;96:7047–7052. Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, Lechler RI. Leptin modulates the T-cell immune response and reverses starvation-induced immunosuppression. Nature 1998;394:897– 901. Grunfeld C, Zhao C, Fuller J, Pollack A, Moser A, Friedman J, Feingold KR. Endotoxin and cytokines induce expression of leptin, the ob gene product, in hamsters. J Clin Invest 1996;97: 2152–2157. Sarraf P, Frederich RC, Turner EM, Ma G, Jaskowiak NT, Rivet DJ III, Flier JS, Lowell BB, Fraker DL, Alexander HR. Multiple cytokines and acute inflammation raise mouse leptin levels: potential role in inflammatory anorexia. J Exp Med 1997;185:171–175. Cioffi JA, Shafer AW, Zupancic TJ, Smith-Gbur J, Mikhail A, Platika D, Snodgrass HR. Novel B219/OB receptor isoforms: possible role of leptin in hematopoiesis and reproduction. Nat Med 1996; 2:585–589. Lostao MP, Urdaneta E, Martinez-Anso E, Barber A, Martinez JA. Presence of leptin receptors in rat small intestine and leptin effect on sugar absorption. FEBS Lett 1998;423:302–306. Hardwick JC, Van Den Brink GR, Offerhaus GJ, Van Deventer SJ, Peppelenbosch MP. Leptin is a growth factor for colonic epithelial cells. Gastroenterology 2001;121:79 –90. Morton NM, Emilsson V, Liu YL, Cawthorne MA. Leptin action in intestinal cells. J Biol Chem 1998;273:26194 –26201. Barbier M, Cherbut C, Aube AC, Blottiere HM, Galmiche JP. Elevated plasma leptin concentrations in early stages of experimental intestinal inflammation in rats. Gut 1998;43:783–790. Barbier M, Attoub S, Joubert M, Bado A, Laboisse C, Cherbut C, Galmiche JP. Proinflammatory role of leptin in experimental colitis in rats. Benefit of cholecystokinin-B antagonist and beta 3-agonist. Life Sci 2001;69:567–580. Siegmund B, Lehr HA, Fantuzzi G. Leptin: a pivotal mediator of intestinal inflammation in mice. Gastroenterology 2002;122: 2011–2015. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RE, Duyk GM, Tepper RI, Morgenstern JP. Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell 1996;84:491– 495. Pothoulakis C, LaMont JT, Eglow R, Gao N, Rubins JB, Theoharides TC, Dickey BF. Characterization of rabbit ileal receptors for Clostridium difficile toxin A. Evidence for a receptor-coupled G protein. J Clin Invest 1991;88:119 –125. Castagliuolo I, Riegler M, Pasha A, Nikulasson S, Lu B, Gerard C, Gerard N, Pothoulakis C. Neurokinin 1 receptor is required in
GASTROENTEROLOGY Vol. 124, No. 3
32.
33.
34.
35.
36. 37.
38.
39. 40.
41.
42.
43.
44.
45.
46.
47.
48.
Clostridium difficile-induced enteritis. J Clin Invest 1998;101: 1547–1550. Pothoulakis C, Castagliuolo I, LaMont T, Jaffer A, O’Keane JC, Snider RM, Leeman SE. CP-96,345, a substance P antagonist, inhibits rat intestinal responses to Clostridium difficile toxin A but not cholera toxin. Proc Natl Acad Sci U S A 1994;91:947–951. Chaudhary M, Mandir N, FitzGerald AJ, Howard JK, Lord GM, Ghatei MA, Bloom SR, Goodlad RA. Starvation, leptin and epithelial cell proliferation in the gastrointestinal tract of the mouse. Digestion 2000;61:223–229. Ikejima K, Honda H, Yoshikawa M, Hirose M, Kitamura T, Takei Y, Sato N. Leptin augments inflammatory and profibrogenic responses in the murine liver induced by hepatotoxic chemicals. Hepatology 2001;34:288 –297. Jacobson L, Sapolsky R. Augmented ACTH responses to stress in adrenalectomized rats replaced with constant, physiological levels of corticosterone are partially normalized by acute increases in corticosterone. Neuroendocrinology 1993;58:420 – 429. Pothoulakis C. Effects of Clostridium difficile toxins on epithelial cell barrier. Ann N Y Acad Sci 2000;915:347–356. Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 1982;78:206 – 209. Bjorbaek C, Elmquist JK, El-Haschimi K, Kelly J, Ahima RS, Hileman S, Flier JS. Activation of SOCS-3 messenger ribonucleic acid in the hypothalamus by ciliary neurotrophic factor. Endocrinology 1999;140:2035–2043. Coleman DL, Burkart DL. Plasma corticosterone concentrations in diabetic (db) mice. Diabetologia 1977;13:25–26. Stephens TW, Basinski M, Bristow PK, Bue-Valleskey JM, Burgett SG, Craft L, Hale J, Hoffmann J, Hsiung HM, Kriauciunas A, MacKellar A, Rosteck W, Schoner PRJ, Smith B, Tinsley D, Zhang Xing-Yue FC, Heiman M. The role of neuropeptide Y in the antiobesity action of the obese gene product. Nature 1995;377: 530 –532. Pacak K, McCarty R, Palkovits M, Cizza G, Kopin IJ, Goldstein DS, Chrousos GP. Decreased central and peripheral catecholaminergic activation in obese Zucker rats. Endocrinology 1995;136: 4360 – 4367. Wlk M, Wang CC, Venihaki M, Liu J, Zhao D, Anton PM, Mykoniatis A, Pan A, Zacks J, Karalis K, Pothoulakis C. Corticotropinreleasing hormone antagonists possess anti-inflammatory effects in the mouse ileum. Gastroenterology 2002;123:505–515. Laue L, Kawai S, Brandon DD, Brightwell D, Barnes K, Knazek RA, Loriaux DL, Chrousos GP. Receptor-mediated effects of glucocorticoids on inflammation: enhancement of the inflammatory response with a glucocorticoid antagonist. J Steroid Biochem 1988;29:591–598. Laue L, Loriaux DL, Chrousos GP. Glucocorticoid antagonists and the role of glucocorticoids at the resting and stress state. Adv Exp Med Biol 1988;245:225–235. Sternberg EM, Young W Sr, Bernardini R, Calogero AE, Chrousos GP, Gold PW, Wilder RL. A central nervous system defect in biosynthesis of corticotropin-releasing hormone is associated with susceptibility to streptococcal cell wall-induced arthritis in Lewis rats. Proc Natl Acad Sci U S A 1989;86:4771– 4775. Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ, Gold PW, Wilder RL. Inflammatory mediator-induced hypothalamic-pituitary-adrenal axis activation is defective in streptococcal cell wall arthritis-susceptible Lewis rats. Proc Natl Acad Sci U S A 1989;86:2374 –2378. Hosoi T, Okuma Y, Nomura Y. Expression of leptin receptors and induction of IL-1beta transcript in glial cells. Biochem Biophys Res Commun 2000;273:312–315. Gabay C, Dreyer M, Pellegrinelli N, Chicheportiche R, Meier CA.
March 2003
Leptin directly induces the secretion of interleukin 1 receptor antagonist in human monocytes. J Clin Endocrinol Metab 2001; 86:783–791. 49. Santos-Alvarez J, Goberna R, Sanchez-Margalet V. Human leptin stimulates proliferation and activation of human circulating monocytes. Cell Immunol 1999;194:6 –11.
Received March 7, 2002. Accepted December 2, 2002.
LEPTIN AND INFLAMMATION
691
Address requests for reprints to: Charalabos Pothoulakis, M.D., Beth Israel Deaconess Medical Center, Division of Gastroenterology, Dana 501, 330 Brookline Avenue, Boston, Massachusetts 02215. e-mail: [email protected]; fax: (617) 667-2767. Supported by research grants PO DK33506 (to C.P.), PO 1 DK 56116 (to C.M.), RO1 DK 58785 (to C.M.), MO RR 01032 (to C.M.), and a Pilot Feasibility Study from P30 DK 40561 from the National Institutes of Health. This study also was supported by a research grant from the Crohn’s and Colitis Foundation.
Leptin Mediates Clostridium difficile Toxin A–Induced Enteritis in Mice ANDREAS MYKONIATIS,* PAULINE M. ANTON,* MICHAEL WLK,* CHI CHUNG WANG,* ¨ HER,‡ MARIA VENIHAKI,§ SIMOS SIMEONIDIS,* JEFF ZACKS,㛳 LINDA UNGSUNAN,‡ SUSANN BLU DEZHENG ZHAO,* STAVROS SOUGIOULTZIS,* KATIA KARALIS,§ CHRISTOS MANTZOROS,‡ and CHARALABOS POTHOULAKIS* *Divisions of Gastroenterology and ‡Endocrinology, Beth Israel Deaconess Medical Center; §Division of Endocrinology, Children’s Hospital, Harvard Medical School; and 㛳Mallory Institute, Department of Pathology, Boston University School of Medicine, Boston, Massachusetts
Background & Aims: Leptin regulates energy homeostasis and participates in the regulation of the hypothalamicpituitary-adrenal axis. Although hyperleptinemia is described in experimental colitis, its role in the pathophysiology of enterotoxin-mediated diarrhea and inflammation remains unclear. We examined the role of leptin in the inflammatory diarrhea induced by toxin A from Clostridium difficile, the causative agent of antibioticrelated colitis. Methods: Toxin A (10 g) or buffer were administered in ileal loops of leptin-deficient (ob/ob), leptin-resistant (db/db), or wild-type mice and enterotoxic responses were measured. Results: In toxin A–treated wild-type mice, circulating leptin and corticosterone levels were increased compared with bufferinjected animals. Toxin A also stimulated increased mucosal expression of the Ob-Rb at the messenger RNA (mRNA) and protein level. Ob/ob and db/db mice were partially protected against toxin A–induced intestinal secretion and inflammation, and this effect was reversed by leptin administration in ob/ob, but not db/db, mice. Basal- and toxin A–stimulated plasma corticosterone levels in ob/ob and db/db mice were higher compared with toxin A–treated wild-type mice. To assess whether the effect of leptin in intestinal inflammation is mediated by corticosteroids we performed adrenalectomy experiments in db/db and wild-type mice. Our results suggested that the diminished intestinal response to toxin A in db/db mice was related only in part to increased levels of corticosteroids. Conclusions: Leptin plays an important role in regulating the severity of enterotoxin-mediated intestinal secretion and inflammation by activating both corticosteroid-dependent and -independent mechanisms.
lostridium difficile, the primary cause of antibioticassociated colitis, is a major pathogen in hospitals and nursing homes.1 This anaerobic bacterium mediates diarrhea and colitis by releasing 2 exotoxins, toxin A and toxin B.1 The cellular proinflammatory mechanism of these toxins involves binding to their receptors,2 effects
C
on cellular mitochondria,3 generation of reactive oxygen species,3 and activation of the transcription nuclear factor B.4 Animal experiments suggest that toxin A causes intestinal secretion and inflammation by stimulating a complex cascade involving neuropeptides, such as substance P and neurotensin, and activating lamina propria macrophages, mast cells, and neutrophils.5 Recent results also showed that endogenous glucocorticoids modulate the secretory and inflammatory effects of C. difficile toxin A in rat ileum by inhibiting toxin A–induced neutrophil infiltration and expression of proinflammatory cytokines.6 Leptin, the product of the ob gene,7 is an adiposesecreted tissue peptide that signals to the brain the magnitude of fat stores8 and regulates energy homeostasis.9 Leptin mediates its effects by binding to a specific receptor belonging to the cytokine receptor superfamily.10 Several forms of the leptin receptor (Ob-R) exist, but the signaling Ob-Rb isoform, which contains a long intracytoplasmic domain, has been shown to be important for mediation of leptin’s effects.11 Several studies suggest that leptin and the Ob-Rb may be associated with inflammatory conditions. Leptin acting on the hypothalamus regulates neuroendocrine systems important in immune function such as the adrenocorticotropic hormone– corticosterone axis.12,13 More specifically, a leptin-deficient or leptin-resistant state is associated with increased corticosterone levels whereas leptin replacement results in partial normalization of corticosterone levels.12 Leptin-deficient (ob/ob) mice and leptin-resistant (db/db) mice have impaired T-cell imAbbreviations used in this paper: db/db, leptin resistant; IL, interleukin; MPO, myeloperoxidase; ob/ob, leptin deficient; Ob-R, leptin receptor; Ob-Rb, long form of leptin receptor. © 2003 by the American Gastroenterological Association 0016-5085/03/$30.00 doi:10.1053/gast.2003.50101
684
MYKONIATIS ET AL.
munity associated with marked reduction in size and cellularity of the thymus.14,15 In addition, peritoneal macrophages isolated from leptin-deficient and leptinresistant mice have decreased phagocytic activity, and expression of proinflammatory cytokines including interleukins (ILs) and tumor necrosis factor ␣.16 Leptin administration to ob/ob mice normalizes these responses, reduces thymocyte apoptosis, and substantially increases thymic cellularity.17 Leptin administration to rats results in increased levels of the proinflammatory cytokine IL-1 in the rat hypothalamus.18 Further, administration of leptin increases Th1 and suppresses Th2 cytokine production by T cells in culture.19 Thus, available evidence indicates that leptin may have both direct and indirect action on the immune system. Moreover, factors that activate the immune system such as injection of lipopolysaccharide increases leptin messenger RNA (mRNA) in adipose tissues and leptin levels in serum.20,21 The signaling Ob-Rb isoform of the leptin receptor is expressed in the gastrointestinal tract of animals and humans and on colonic epithelial cells,22–25 and leptin binding to this receptor inhibits absorption of sugars23 and stimulates growth of colonocytes.24 Further, leptin blood levels are elevated transiently at the early stages of trinitrobenzene sulfonic acid–mediated colitis in rats, and this increase correlates with the degree of inflammation and anorexia.26 Decreasing leptin levels by a cholecystokinin B antagonist and a 3 agonist treatment are associated with an improvement in the severity of colitis.27 Siegmund et al.28 showed that the inflammatory responses in sodium dextran sulfate– and trinitrobenzene sulfonic acid–induced colitis were significantly lower in leptin-deficient ob/ob mice compared with wild-type mice and injection of leptin restored the colonic inflammatory response. However, whether leptin and Ob-Rb participate in the pathogenesis of enterotoxin-mediated intestinal secretion and inflammation is not known. As well, the role of the hypothalamo-pituitary-adrenal A axis in the development of intestinal inflammation as it relates to leptin deficiency has not been examined. In this study, we examined the role of leptin and its long receptor isoform in fluid secretion and inflammation induced by C. difficile toxin A. We used the ileal loop model of toxin A–induced ileitis in wild-type as well as leptin-deficient ob/ob mice and leptin-resistant db/db mice, which have a truncated and inactive form of the leptin receptor,29 to investigate the possibility that leptin participates in intestinal inflammation. We also correlated the levels of leptin and ileal Ob-Rb expression with the time-course appearance of intestinal secretion,
GASTROENTEROLOGY Vol. 124, No. 3
epithelial cell damage, and inflammation in response to C. difficile toxin A. Finally, we examined whether this role of leptin is corticosterone independent or not by studying adrenalectomized mice. Our results point to a novel proinflammatory role of leptin in the pathophysiology of enterotoxin-mediated enteritis that is in part corticosterone independent.
Materials and Methods Toxin A was purified from culture supernatants of C. difficile strain 10,463 as previously described.30 Protein concentrations were determined with the DC protein assay (BioRad Laboratories, Hercules, CA).
Mouse Ileal Loops Twelve-week-old male C57BL/6J (20 –25 g), ob/ob (B6.V-Lepob, 45–55 g) and db/db (BKS.Cg-m⫹/⫹Leprdb, 45–55 g) mice (Jackson Laboratory, Bar Harbor, ME) were housed on a 12-hour light-dark cycle for 3 days before surgery. Experiments were performed between 9:30 AM–11:30 AM to minimize influence of the circadian rhythm. Mice were fasted (16 h) and then anesthetized with a mixture of ketamine and xylazine,31 and after laparotomy, one 3- to 5-cm ileal loops were formed32 and injected with 0.15 mL of phosphate buffer saline (pH 7.4) containing 10 g of purified toxin A or buffer alone. The abdomen was then sutured, animals were placed on a heating pad at 37°C, and after 4 hours were killed with CO2 inhalation. Ob/ob and db/db mice (n ⫽ 5) were injected intraperitoneally with 1 g/g of body weight12,33,34 mouse recombinant leptin (Sigma, St. Louis, MO), 20, 14, and 0.5 hours before toxin A administration. Previous studies showed that this dose, in fasted mice, reestablishes the circulating levels of leptin observed in nonfasted animals.12 At death, fluid secretion was estimated as the loop weightto-length ratio (mg/cm) as previously described.31,32 Pieces of tissues were removed for myeloperoxidase (MPO) activity measurement, and full-thickness loop sections were fixed in formalin, paraffin-embedded, stained with H&E, and graded histopathologically by using parameters associated with toxin A–induced enterotoxicity.32 Moreover, ileal sections from C57BL/6J mice (n ⫽ 4) were snap frozen for immunohistochemistry and mRNA amplification for OB-Rb. Tissues were stored at – 80°C until determination.
Adrenalectomy C57BL/6J and db/db mice (n ⫽ 6) were adrenalectomized bilaterally under general anesthesia by the retroperitoneal route,35 whereas sham-operated C57BL/6J and db/db mice (n ⫽ 4) underwent the same procedure without removing the adrenals. Adrenalectomized animals were given 0.9% NaCl to compensate for the salt loss. Six days after adrenalectomy ileal loops were formed, exposed to toxin A or buffer, and fluid secretion was measured at 4 hours.36 The institutional Animal Care and Use Committee approved the animal studies.
March 2003
Leptin and Corticosterone Measurements Ileal loops of male wild-type, ob/ob, and db/db mice were injected with 10 g of purified toxin A or buffer (control). Blood samples (0.2 mL) were collected from the retroorbital plexus before buffer or toxin A injection and after 30 minutes, 2 hours, and 4 hours for leptin and corticosterone assays. They were centrifuged (2000g for 10 min at 4°C), and aliquots of plasma were stored at ⫺80°C. Leptin and corticosterone levels were measured using commercially available radioimmunoassay kits from Linco Research Institute (St. Louis, MO), and ICN Biomedicals, Inc. (Costa Mesa, CA), respectively.
Myeloperoxidase Measurements Tissue MPO activity was determined by a modified method of Bradley et al.37 Briefly, ileal loop samples were homogenized (Ultra Thurrax; Tekmar Co., Cincinnati, OH) and frozen and thawed 3 times. Samples were sonicated (Heat Systems; Ultrasonics, Plain View, NY) in 1.5 mL of 50 mmol/L phosphate buffer containing 0.5% of hexadecyl-trimethyl ammonium bromide. Samples were then centrifuged at 10,000g (15 min at 4°C) and supernatants were collected. Supernatants were then diluted into the same phosphate buffer containing 0.167 mg/mL of O-dianisidine dihydrochloride and 5.10⫺4% of hydrogen peroxide. MPO activity was measured spectrophotometrically (Lambda 20, UV/VIS spectrophotometer; Perkin Elmer, Norwalk, CT) at 450 nm using human MPO (0.1 U/10 L; Sigma, St. Louis, MO) as a standard. Proteins were evaluated using the DC protein assay (Bio-Rad Laboratories, Hercules, CA).
Total Mucosal RNA Extraction and ReverseTranscription Polymerase Chain Reaction Amplification for Ob-Rb Messenger RNA Ileal loops were injected with either phosphate-buffered saline or toxin A, and after 1 or 2 hours loops were removed, opened, washed in ice-cold phosphate-buffered saline, and the mucosa was scraped with RNAse-free glass slides (Fisher, Pittsburgh, PA). Total RNA was isolated as previously described.31 The method described by Bjorbaek et al.38 was applied for reverse-transcription polymerase chain reaction. Briefly, a final volume of 100 L complementary DNA (cDNA) was synthesized from 1 g of total mRNA using the Advantage reverse-transcription polymerase chain reaction kit from Clontech (Palo Alto, CA). Ob-Rb cDNA was amplified using the following primers: upstream: 5⬘-ACA GCG TGC TTC CTG GGT CTT C-3⬘, downstream: 5⬘-TGG ATA AAC CCT TGC TCT TCA-3⬘. The 201-bp -actin cDNA was amplified using the following primers: upstream: 5⬘-CGT ACC ACG GGC ATT GTG ATG G-3⬘, and downstream: 5⬘-TTT GAT GTC ACG CAC GAT TTC CC-3⬘. Preliminary polymerase chain reaction experiments showed that the rate of amplification was linear for -actin when applied for fewer than 20 cycles and for Ob-Rb when applied for fewer than 35 cycles. Each 50-L polymerase chain reaction was performed
LEPTIN AND INFLAMMATION
685
with 5.0 L of template cDNA. Conditions were: 10 mmol/L Tris-HCl (pH 8.8), 50 mmol/L KCl, 1.5 mmol/L MgCl2 , 0.01% gelatin, 0.2 mmol/L deoxynucleoside triphosphate, 20 pmol of each primer, 2.5 U of Taq polymerase (Stratagene, La Jolla, CA), and 0.5 L of 32P-dGTP (29.6 Tbq/mmol, 370 Mbq/mL) (NEN, Boston, MA). The mixture was overlaid with 50 L of mineral oil and, after initial denaturation at 96°C for 4 minutes, samples were subjected to 18 cycles of amplification for -actin and to 30 cycles for Ob-Rb (denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and extension at 72°C for 45 sec). Ten microliters of the reaction were then combined with 5 L of sequencing stop solution (Amersham International, Buckinghamshire, UK) and heated to 85°C for 5 minutes, before loading 4 L onto a 4% urea-acrylamide gel (38 ⫻ 31 ⫻ 0.03 cm). The electrophoresis conditions were 60 W of constant power for 1.45 hours. After electrophoresis, gels were transferred to filter paper, dried, and subjected to 32P quantitation by PhosphoImager Analysis (Molecular Dynamics, Sunnyvale, CA).
Immunohistochemistry Toxin A or buffer were injected into ileal loops of wild-type mice (n ⫽ 3 per group), and after 15, 30, and 60 minutes after toxin A exposure or 60 minutes after buffer exposure, mice were killed and freshly frozen sections were prepared. Frozen ileal sections were cut (5 m) and fixed in Teck tissue fixative (Teck, Redding, CA) for 10 minutes. Sections were then washed in Tris buffer saline (pH 7.5) containing 0.1% Tween 20 (TBST; Dako Corporation, Carpenteria, CA), and incubated with 2% normal donkey serum in TBST for 30 minutes at 22°C. Sections were incubated for 2 hours with 1:20 dilution of rabbit anti-human Ob-Rb antibody or control rabbit immunoglobulin G (Linco Research, Inc), washed in 1 ⫻ TBST and incubated for 30 minutes with 1:50 dilution of a fluorescein isothiocyanate– conjugated donkey anti-rabbit immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Sections were mounted with antibleaching solution and images were viewed under a confocal microscope (model MRC 1024; Bio-Rad Microsciences, Cambridge, MA) using Plan-Neofluar objectives (20⫻) and were stored digitally in Bio-Rad COMOS software. Specificity of the antibody was assessed in separate experiments in which slides were exposed to the anti-human Ob-Rb antibody, which was preincubated with an excess of leptin protein.
Statistical Analyses Results are presented as mean ⫾ SEM and, unless otherwise stated, data were analyzed using the SIGMA-STAT statistics software program (Jandel Scientific Software, San Rafael, CA). Analysis of variance followed by a Tukey post hoc test was used for intergroup comparisons. A value of P less than 0.05 was considered significant.
686
MYKONIATIS ET AL.
GASTROENTEROLOGY Vol. 124, No. 3
toxin A stimulated increased fluid secretion, MPO activity, and caused severe histologic changes in the ileum of wild-type mice compared with buffer exposure (Figure 1A), consistent with prior observations.31,32 In contrast, ob/ob and db/db mice had reduced fluid secretion, attenuated epithelial damage, mucosal congestion and edema, and neutrophil (polymorphonuclear leukocyte) infiltration in response to toxin A compared with wild-type mice (Figure 1A, Table 1). Leptin replacement before toxin A injection in ob/ob mice resulted in increased susceptibility to toxin A–induced fluid secretion whereas leptin administration did not alter toxin A–induced ileal secretion in the leptin receptor– deficient db/db mice (Figure 1A). Histologic quantification of buffer-exposed ileal tissues of ob/ob and db/db mice was no different than buffer-injected ileal tissues of wild-type mice (Table 1). Together, these results provide direct evidence for the importance of leptin in inflammatory diarrhea caused by C. difficile toxin A. Increased Plasma Leptin During Toxin A–Mediated Enteritis
Figure 1. (A) Reduced toxin A–induced ileal fluid secretion in db/db and ob/ob mice. Ileal loops of anesthetized mice were injected with buffer or toxin A and, after 4 hours, mice were killed and fluid secretion was measured by loop weight-to-length ratio (mg/cm). Some ob/ob and some db/db mice were injected with 3 single intraperitoneal leptin injections (1 g/g of body weight), 20, 14, and 0.5 hours before toxin A administration. Data are means ⫾ SEM of 5 animals for each group. **P ⬍ 0.01 vs. the respective buffer; ⫹P ⬍ 0.05 vs. toxin A of C57BL/6J mice; #P ⬍ 0.05 vs. toxin A of ob/ob mice. 䊐, buffer; ■, toxin A; , leptin ⫹ toxin A. (B) Increased leptin plasma levels during toxin A–induced enteritis. Ileal loops of C57/BL6J anesthetized mice were injected with buffer or toxin A and, at the indicated time points, blood samples were collected and leptin concentration was measured in the plasma by radioimmunoassay. Data are means ⫾ SEM of 5 animals for each group. *P ⬍ 0.05 vs. the respective buffer or basal leptin levels (after anesthesia, before injection of loops). 䊐, buffer; ■, toxin A.
Results Reduced Toxin A–Induced Fluid Secretion and Intestinal Inflammation in LeptinDeficient and Leptin-Resistant Mice To assess directly the role of leptin in the toxin A model of intestinal inflammation, we injected ileal loops of anesthetized wild-type, leptin-deficient (ob/ob), and leptin-resistant (db/db) mice, with either buffer or purified toxin A and, after 4 hours, we measured toxin A–associated ileal responses. Basal fluid secretion in response to buffer injection was comparable between wildtype and ob/ob and db/db mice (Figure 1A). Moreover,
Because these results (Figure 1A, Table 1) suggested participation of leptin in toxin A–mediated enteritis, we next examined the plasma levels of leptin after toxin A administration. Thirty minutes after ileal toxin A injection in wild-type mice plasma leptin levels were elevated significantly, and further increased after 4 hours (Figure 1B). We also measured circulating leptin levels in db/db mice before and after toxin A exposure. Db/db mice were hyperleptinemic at baseline (17.01 ⫾ 0.43 in buffer-treated db/db mice vs. 1.32 ⫾ 0.16 ng/dL in buffer-treated wild-type mice, n ⫽ 5 per group, P ⱕ 0.05), consistent with prior observations of high leptin levels in these mice.8,25 Moreover, plasma leptin levels were not altered significantly in response to toxin A treatment at any of the time points studied (30 min– 4 h, n ⫽ 5 per group). We suspect that the reason for the inability of toxin A to increase leptin levels in db/db mice, in contrast to wild-type mice, is related to the already established hyperleptinemic state of these animals. Role of Corticosteroids in Toxin A–Induced Secretion in Leptin Receptor–Deficient Mice Because the inflammatory and secretory responses to toxin A are modulated by endogenous corticosteroids,6 we examined corticosteroid plasma levels in response to toxin A in wild-type, ob/ob, and db/db mice 2 hours after toxin A administration. Plasma corticosterone in buffer-injected ob/ob and db/db mice was higher compared with buffer-injected wild-type mice (Figure 2A),
March 2003
LEPTIN AND INFLAMMATION
687
Table 1. Reduced Toxin A–Mediated Histologic Responses in ob/ob and db/db Mice Treatment
Epithelial damage
Congestion and edema
Neutrophil infiltration
MPO U/g protein
Buffer C57BL/6J Toxin A C57BL/6J Buffer db/db Toxin A db/db Buffer ob/ob Toxin A ob/ob
0.4 ⫾ 0.24 2.8 ⫾ 0.16a 0.3 ⫾ 0.2 1.16 ⫾ 0.16b,c 0.46 ⫾ 0.18 1.31 ⫾ 0.11a,c
0.5 ⫾ 0.28 2.5 ⫾ 0.2a 0.1 ⫾ 0.1 1.14 ⫾ 0.09a,c 0.2 ⫾ 0.13 1.37 ⫾ 0.18a,c
0.0 ⫾ 0.0 2.6 ⫾ 0.2a 0.1 ⫾ 0.1 1.0 ⫾ 0.0b,c 0.4 ⫾ 0.16 1.28 ⫾ 0.18b,c
29.16 ⫾ 5.84 108.2 ⫾ 16.5b 30 ⫾ 4.61 51.3 ⫾ 2.07b,d 32.5 ⫾ 3.57 75.56 ⫾ 3.4b,d
NOTE. Ileal loops were injected with 10 g of toxin A or phosphate-buffered saline. After 4 hours, animals were killed, ileal loops were removed, and histologic severity of enteritis was graded by a score of 0 –3 for epithelial cell damage, congestion and edema of the mucosa, and mucosal neutrophil infiltration. For MPO, ileal loops were removed and homogenized in ice-cold phosphate-buffered saline containing hexadecyl-trimethyl ammonium bromide (0.5%) (see Methods section). Homogenates were then centrifuged and aliquots from the supernatants were taken for MPO and protein measurements. Data are means ⫾ SEM per group; n ⫽ 8 –10 for histologic quantitation and 4 for MPO measurements, each with duplicate determinations. The nonparametric Mann-Whitney test was used to calculate the histologic differences. aP ⬍ 0.01 vs. control. bP ⬍ 0.05 vs. control. cP ⬍ 0.01 vs. toxin A of C57BL/6J mice. dP ⬍ 0.05 vs. toxin A of C57BL/6J mice.
consistent with the high basal corticosterone level of ob/ob and db/db mice.39,40 Moreover, exposure of ileal loops from ob/ob and db/db mice to toxin A stimulated a further increase in circulating corticosterone levels (Figure 2A). Because the higher corticosterone concentrations in ob/ob and db/db mice could raise the possibility that the relative protection of this strain could be caused by their increased corticosterone and not leptin levels per se, we sought to address this possibility directly using adrenalectomized db/db and wild-type mice. Plasma corticosterone level before toxin A and buffer injection was 1.30 ⫾ 0.16 and 1.60 ⫾ 0.25 g/dL for adrenalectomized C57BL/6J and db/db mice, respectively, compared with 22.0 ⫾ 3.8, and 85.0 ⫾ 5.4 g/dL for nonadrenalectomized C57BL/6J, and db/db mice, respectively. Intestinal secretion was similar in sham-operated or adrenalectomized mice injected with buffer (Figure 2B). Adrenalectomized mice of both genotypes had increased toxin A–induced fluid secretion, compared with the respective sham-operated animals, indicating that glucocorticoids contribute to the reduced secretory responses to toxin A in db/db mice. However, toxin A–mediated secretion in db/db mice was significantly lower compared with fluid secretion in toxin A– exposed wild-type mice (Figure 2B). This suggests that an additional glucocorticoid-independent mechanism may be responsible for the reduced toxin A–associated responses in leptin-resistant db/db mice. Increased Leptin Receptor Expression During Toxin A–Induced Intestinal Inflammation Because the leptin receptor Ob-Rb was shown to be involved in toxin A–induced secretion and inflamma-
tion, we examined the expression and localization of the Ob-Rb in the ileal mucosa of wild-type mice after toxin A exposure. Reverse-transcription polymerase chain reaction showed a gradual increase of Ob-Rb mucosal mRNA levels in toxin A– exposed ileum that became statistically significant (by 42.2% in wild-type mice vs. controls, P ⬍ 0.001) after 2 hours (Table 2). Immunohistochemical studies showed Ob-Rb–positive cells in buffer-exposed ileum and increased expression of these receptors (localized on epithelial as well as on subepithelial cells) after 15 and 30 minutes of toxin A exposure. However, Ob-Rb expression was less evident 60 minutes after toxin A exposure, probably owing to epithelial cell damage in response to the toxin.
Discussion We have reported previously that C. difficile toxin A stimulates fluid secretion and elicits an acute inflammatory response in animal intestine characterized by neutrophil infiltration and epithelial cell destruction.5 The results presented here show that animals either genetically lacking leptin itself or resistant to leptin’s effects have substantially reduced responses to toxin A, which is normalized in response to the administration of leptin in leptin-deficient ob/ob mice but not in leptinresistant db/db mice. This indicates that leptin mediates to a large extent ileal fluid secretion and histologic changes in response to this toxin. This study showed a proinflammatory role for leptin in the pathophysiology of intestinal secretion and inflammation in response to a bacterial enterotoxin. Our results are consistent with the findings of Barbier et al.26,27 and Siegmund et al.,28 suggesting participation of leptin in the pathogenesis of experimental colitis.
688
MYKONIATIS ET AL.
Figure 2. (A) Plasma corticosterone levels in ob/ob and db/db mice in response to toxin A. One closed ileal loop was prepared in the distal ileum of anesthetized mice and injected with either buffer or toxin A (10 g). After 2 hours mice were killed and blood samples were collected by heart puncture. Data are means ⫾ SEM of 5 loops per experimental condition. ***P ⬍ 0.001 vs. buffer for C57BL/6J, ob/ob, and db/db mice; ⫹⫹⫹P ⬍ 0.001 vs. buffer-injected C57BL/6J; †††P ⬍ 0.001 vs. toxin A–injected C57BL/6J mice. 䊐, buffer; ■, toxin A. (B) Adrenalectomy in db/db mice reverses reduced toxin A–mediated fluid secretion. Wild-type or db/db mice were subjected to adrenalectomy whereas another group was sham-operated without removing the adrenals. Mice were left to recover for 6 days and then fluid secretion (weight-to-length ratio) was measured 4 hours after toxin A (10 g, n ⫽ 7– 8 per group) or buffer (n ⫽ 3 per group) injection into ileal loops. Data are expressed as mean ⫾ SEM. ††P ⬍ 0.01 vs. buffer-infected loops of sham adrenalectomized C57BL/6J; **P ⬍ 0.01 vs. toxin A–injected loops of sham adrenalectomized of both genotypes; ⫹⫹P ⬍ 0.01 vs. toxin A–injected adrenalectomized C57BL/6J mice, ##P ⬍ 0.01 vs. toxin A–injected sham C57BL/6J mice. 䊐, sham buffer; , sham toxin A; , adrenalectomized buffer; , adrenalectomized toxin A.
We observed increased basal and C. difficile toxin A–stimulated corticosterone levels in both ob/ob and db/db mice compared with wild-type mice (Figure 2B). This association between the hypothalamo-pituitary-adrenal axis and circulating leptin has been established previously. Thus, hypercorticosteronemia observed in ob/ob and fasting mice can be reduced significantly by
GASTROENTEROLOGY Vol. 124, No. 3
chronic leptin administration,12,40 whereas leptin-resistant Zucker rats show an excessive hypothalamo-pituitary-adrenal axis activation, along with hypercorticosteronemia that is not corrected by leptin administration.41 Although high basal corticosterone levels in ob/ob and db/db mice were expected,39,40 the increase of glucocorticoids in the absence of functional leptin or functional leptin receptors suggest that other mediators enhance the secretion of glucocorticoids in C. difficile–induced inflammation. Indeed, we have shown recently that increased peripheral corticotropin-releasing hormone expression was associated with increased corticosteroid levels in the toxin A model of enteritis.42 We also have shown that endogenous glucocorticoids diminish the secretory and inflammatory effect of toxin A.6 This protective effect of glucocorticoids also has been described in other models of inflammation, including animals injected with RU 48643,44 or mice deficient in corticotropin-releasing hormone or corticosterone,45,46 which develop an increased innate inflammatory response. To determine whether a glucocorticoid-independent component is involved in the effects of leptin in toxin A–mediated intestinal secretion, we used the adrenalectomized mouse model. Our results using adrenalectomized wild-type and db/db mice point out that only part of toxin A–induced fluid secretion is linked to release of glucocorticoids, whereas a glucocorticoid-independent, leptin-associated mechanism appears also to be involved. The nature of this mechanism remains to be identified, but several studies have suggested that leptin–Ob-Rb interactions may affect directly or alter inflammatory responses. For example, increased IL-1 mRNA expression has been shown in glial cells in response to leptin.47 Leptin also can enhance the effect of LPS on the production of different cytokines by murine peritoneal macroTable 2. Increased Toxin A–Mediated Ob-Rb mRNA Accumulation in Mouse Ileum Time points 1h 2h
Treatment Buffer Toxin A Buffer Toxin A
Leptin mRNA 99.7 ⫾ 6.6 152.4 ⫾ 20.4 100.0 ⫾ 6.8 173.4 ⫾ 15.8a
NOTE. Ileal loops were injected with phosphate-buffered saline or toxin A and, after 1 and 2 hours, mRNA was extracted and reverse transcribed to obtain cDNA. Ob-Rb mRNA was quantified by reversetranscription polymerase chain reaction as described in Materials and Methods. Data represent the mean values (from arbitrary Phospholmager units) of each group corrected for the -actin values and then normalized to the mRNA expression of the buffer-treated group, which is considered to be 100. Data are mean ⫾ SEM, n ⫽ 5 per group. Statistical significance was assessed by unpaired, 2-tailed Student t test using the Statview program (Abacus, CA). aP ⬍ 0.001 vs. buffer C57BL/6J mice.
March 2003
LEPTIN AND INFLAMMATION
689
Figure 3. Immunohistochemistry of Ob-Rb after toxin A administration. (A) Buffer or (B–D) toxin A were injected into ileal loops of wild-type mice. After (B) 15, (C) 30, and (D) 60 minutes the animals were killed and sections were processed for immunohistochemistry using a rabbit anti-human leptin receptor antibody. Leptin receptor staining is present in control (buffer-exposed) mouse ileum. However, 15 and 30 minutes after toxin A injection, a substantial increase in the expression of leptin receptor compared with control was noted. Ob-Rb expression was less evident 60 minutes after toxin A exposure. Leptin receptor immunoreactivity is present on intestinal epithelial cells as well as in cells of the lamina propria. Results are representative of 3 separate experiments. (A) Cross-section. (B–D) Longitudinal sections. (Magnification 200⫻).
phages, including IL-1Ra, IL-6, IL-12, and tumor necrosis factor ␣.16 Moreover, exposure of the macrophage cell line RAW264.7, or monocytic THP-1 cells, or peripheral blood mononuclear cells to leptin causes secretion of IL-1Ra.16,48 In human monocytes, leptin can induce the expression of cell-surface markers and cytokines, including IL-6 and tumor necrosis factor ␣49 and, in human lymphocytes, exogenous leptin increases Th1 and suppresses Th2 cytokine production.19 Further, administration of leptin in ob/ob mice reduced apoptosis of thymocytes and increased the CD4⫾CD8⫾/CD4⫺CD8⫺ ratio and thymic cellularity.17 Finally, our immunohistochemical experiments (Figure 3) show that expression of Ob-Rb is increased in the intestinal epithelium 15 minutes after toxin A exposure, whereas increases in Ob-Rb mRNA expression are evident at 1 and 2 hours (Table 2). These results suggest that expression of preformed Ob-Rb receptors may be stimulated at the early stages of toxin A–mediated intestinal inflammation. Our observation that both leptin levels and immunoreactive Ob-Rb are increased significantly within the first 30 minutes of toxin A exposure suggests a direct proinflammatory effect of leptin on intestinal epithelial cells via binding to this receptor.24,25 Indeed, in intestinal cells, leptin administration activates the JAK/STAT pathway that only can be stimulated by the long form of the leptin receptor, Ob-Rb.25 This study showed altered expression of the Ob-Rb during inflammation. However, the mechanism underlying this response remains to be elucidated. The present results may be relevant to the pathophysiology of human C. difficile colitis. Our observations, together with recent studies showing participation of leptin in animal models of inflammatory bowel disease,26 –28 suggest that use of specific leptin or leptin receptor antagonists may represent a novel therapeutic
approach for the treatment of intestinal inflammatory conditions.
References 1. Kelly CP, Pothoulakis C, LaMont JT. Clostridium difficile colitis. N Engl J Med 1994;330:257–262. 2. Pothoulakis C, Gilbert RJ, Cladaras C, Castagliuolo I, Semenza G, Hitti Y, Montcrief JS, Linevsky J, Kelly CP, Nikulasson S, Desai H, Wilkins TD, LaMont JT. Rabbit sucrase-isomaltase contains a functional intestinal receptor for Clostridium difficile toxin A. J Clin Invest 1996;98:641– 649. 3. He D, Hagen SJ, Pothoulakis C, Chen M, Medina ND, Warny M, LaMont JT. Clostridium difficile toxin A causes early damage to mitochondria in cultured cells. Gastroenterology 2000;119:139 – 150. 4. Warny M, Keates A, Keates S, Castagliuolo I, Zacks JK, Aboudola S, Qamar A, Pothoulakis C, LaMont T, Kelly CP. p38 MAP kinase activation by Clostridium difficile toxin A mediates monocyte necrosis, Il-8 production, and enteritis. J Clin Invest 2000;105: 1147–1156. 5. Pothoulakis C, Lamont JT. Microbes and microbial toxins: paradigms for microbial-mucosal interactions II. The integrated response of the intestine to Clostridium difficile toxins. Am J Physiol 2001;280:178 –183. 6. Castagliuolo I, Karalis K, Valenick L, Pasha A, Nikulasson S, Wlk M, Pothoulakis C. Endogenous corticosteroids modulate Clostridium difficile toxin A-induced enteritis in rats. Am J Physiol 2001; 280:G539 –G545. 7. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425– 432. 8. Frederich RC, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS. Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat Med 1995;1: 1311–1314. 9. Friedman JM. Leptin, leptin receptors, and the control of body weight. Nutr Rev 1998;53:S38 –S46. 10. Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J. Identification and expression cloning of a leptin receptor, OB-R. Cell 1995;83: 1263–1271. 11. Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, Lee JI, Friedman JM. Abnormal splicing of the leptin receptor in diabetic mice. Nature 1996;379:632– 635. 12. Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-
690
13.
14. 15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
MYKONIATIS ET AL.
Flier E, Flier JS. Role of leptin in the neuroendocrine response to fasting. Nature 1996;382:250 –252. Heiman ML, Ahima RS, Craft LS, Schoner B, Stephens TW, Flier JS. Leptin inhibition of the hypothalamic-pituitary-adrenal axis in response to stress. Endocrinology 1997;138:3859 –3863. Chandra RK. Cell-mediated immunity in genetically obese C57BL/6J ob/ob mice. Am J Clin Nutr 1980;33:13–16. Fernandes G, Handwerger BS, Yunis EJ, Brown DM. Immune response in the mutant diabetic C57BL/Ks-dt⫹ mouse. Discrepancies between in vitro and in vivo immunological assays. J Clin Invest 1978;61:243–250. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang DJ, Klein AS, Bulkley GB, Bao C, Noble PW, Lane MD, Diehl AM. Leptin regulates proinflammatory immune responses. FASEB J 1998;12:57– 65. Howard JK, Lord GM, Matarese G, Vendetti S, Ghatei MA, Ritter MA, Lechler RI, Bloom SR. Leptin protects mice from starvationinduced lymphoid atrophy and increases thymic cellularity in ob/ob mice. J Clin Invest 1999;104:1051–1059. Luheshi GN, Gardner JD, Rushforth DA, Loudon AS, Rothwell NJ. Leptin actions on food intake and body temperature are mediated by IL-1. Proc Natl Acad Sci U S A 1999;96:7047–7052. Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, Lechler RI. Leptin modulates the T-cell immune response and reverses starvation-induced immunosuppression. Nature 1998;394:897– 901. Grunfeld C, Zhao C, Fuller J, Pollack A, Moser A, Friedman J, Feingold KR. Endotoxin and cytokines induce expression of leptin, the ob gene product, in hamsters. J Clin Invest 1996;97: 2152–2157. Sarraf P, Frederich RC, Turner EM, Ma G, Jaskowiak NT, Rivet DJ III, Flier JS, Lowell BB, Fraker DL, Alexander HR. Multiple cytokines and acute inflammation raise mouse leptin levels: potential role in inflammatory anorexia. J Exp Med 1997;185:171–175. Cioffi JA, Shafer AW, Zupancic TJ, Smith-Gbur J, Mikhail A, Platika D, Snodgrass HR. Novel B219/OB receptor isoforms: possible role of leptin in hematopoiesis and reproduction. Nat Med 1996; 2:585–589. Lostao MP, Urdaneta E, Martinez-Anso E, Barber A, Martinez JA. Presence of leptin receptors in rat small intestine and leptin effect on sugar absorption. FEBS Lett 1998;423:302–306. Hardwick JC, Van Den Brink GR, Offerhaus GJ, Van Deventer SJ, Peppelenbosch MP. Leptin is a growth factor for colonic epithelial cells. Gastroenterology 2001;121:79 –90. Morton NM, Emilsson V, Liu YL, Cawthorne MA. Leptin action in intestinal cells. J Biol Chem 1998;273:26194 –26201. Barbier M, Cherbut C, Aube AC, Blottiere HM, Galmiche JP. Elevated plasma leptin concentrations in early stages of experimental intestinal inflammation in rats. Gut 1998;43:783–790. Barbier M, Attoub S, Joubert M, Bado A, Laboisse C, Cherbut C, Galmiche JP. Proinflammatory role of leptin in experimental colitis in rats. Benefit of cholecystokinin-B antagonist and beta 3-agonist. Life Sci 2001;69:567–580. Siegmund B, Lehr HA, Fantuzzi G. Leptin: a pivotal mediator of intestinal inflammation in mice. Gastroenterology 2002;122: 2011–2015. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, Lakey ND, Culpepper J, Moore KJ, Breitbart RE, Duyk GM, Tepper RI, Morgenstern JP. Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell 1996;84:491– 495. Pothoulakis C, LaMont JT, Eglow R, Gao N, Rubins JB, Theoharides TC, Dickey BF. Characterization of rabbit ileal receptors for Clostridium difficile toxin A. Evidence for a receptor-coupled G protein. J Clin Invest 1991;88:119 –125. Castagliuolo I, Riegler M, Pasha A, Nikulasson S, Lu B, Gerard C, Gerard N, Pothoulakis C. Neurokinin 1 receptor is required in
GASTROENTEROLOGY Vol. 124, No. 3
32.
33.
34.
35.
36. 37.
38.
39. 40.
41.
42.
43.
44.
45.
46.
47.
48.
Clostridium difficile-induced enteritis. J Clin Invest 1998;101: 1547–1550. Pothoulakis C, Castagliuolo I, LaMont T, Jaffer A, O’Keane JC, Snider RM, Leeman SE. CP-96,345, a substance P antagonist, inhibits rat intestinal responses to Clostridium difficile toxin A but not cholera toxin. Proc Natl Acad Sci U S A 1994;91:947–951. Chaudhary M, Mandir N, FitzGerald AJ, Howard JK, Lord GM, Ghatei MA, Bloom SR, Goodlad RA. Starvation, leptin and epithelial cell proliferation in the gastrointestinal tract of the mouse. Digestion 2000;61:223–229. Ikejima K, Honda H, Yoshikawa M, Hirose M, Kitamura T, Takei Y, Sato N. Leptin augments inflammatory and profibrogenic responses in the murine liver induced by hepatotoxic chemicals. Hepatology 2001;34:288 –297. Jacobson L, Sapolsky R. Augmented ACTH responses to stress in adrenalectomized rats replaced with constant, physiological levels of corticosterone are partially normalized by acute increases in corticosterone. Neuroendocrinology 1993;58:420 – 429. Pothoulakis C. Effects of Clostridium difficile toxins on epithelial cell barrier. Ann N Y Acad Sci 2000;915:347–356. Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 1982;78:206 – 209. Bjorbaek C, Elmquist JK, El-Haschimi K, Kelly J, Ahima RS, Hileman S, Flier JS. Activation of SOCS-3 messenger ribonucleic acid in the hypothalamus by ciliary neurotrophic factor. Endocrinology 1999;140:2035–2043. Coleman DL, Burkart DL. Plasma corticosterone concentrations in diabetic (db) mice. Diabetologia 1977;13:25–26. Stephens TW, Basinski M, Bristow PK, Bue-Valleskey JM, Burgett SG, Craft L, Hale J, Hoffmann J, Hsiung HM, Kriauciunas A, MacKellar A, Rosteck W, Schoner PRJ, Smith B, Tinsley D, Zhang Xing-Yue FC, Heiman M. The role of neuropeptide Y in the antiobesity action of the obese gene product. Nature 1995;377: 530 –532. Pacak K, McCarty R, Palkovits M, Cizza G, Kopin IJ, Goldstein DS, Chrousos GP. Decreased central and peripheral catecholaminergic activation in obese Zucker rats. Endocrinology 1995;136: 4360 – 4367. Wlk M, Wang CC, Venihaki M, Liu J, Zhao D, Anton PM, Mykoniatis A, Pan A, Zacks J, Karalis K, Pothoulakis C. Corticotropinreleasing hormone antagonists possess anti-inflammatory effects in the mouse ileum. Gastroenterology 2002;123:505–515. Laue L, Kawai S, Brandon DD, Brightwell D, Barnes K, Knazek RA, Loriaux DL, Chrousos GP. Receptor-mediated effects of glucocorticoids on inflammation: enhancement of the inflammatory response with a glucocorticoid antagonist. J Steroid Biochem 1988;29:591–598. Laue L, Loriaux DL, Chrousos GP. Glucocorticoid antagonists and the role of glucocorticoids at the resting and stress state. Adv Exp Med Biol 1988;245:225–235. Sternberg EM, Young W Sr, Bernardini R, Calogero AE, Chrousos GP, Gold PW, Wilder RL. A central nervous system defect in biosynthesis of corticotropin-releasing hormone is associated with susceptibility to streptococcal cell wall-induced arthritis in Lewis rats. Proc Natl Acad Sci U S A 1989;86:4771– 4775. Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ, Gold PW, Wilder RL. Inflammatory mediator-induced hypothalamic-pituitary-adrenal axis activation is defective in streptococcal cell wall arthritis-susceptible Lewis rats. Proc Natl Acad Sci U S A 1989;86:2374 –2378. Hosoi T, Okuma Y, Nomura Y. Expression of leptin receptors and induction of IL-1beta transcript in glial cells. Biochem Biophys Res Commun 2000;273:312–315. Gabay C, Dreyer M, Pellegrinelli N, Chicheportiche R, Meier CA.
March 2003
Leptin directly induces the secretion of interleukin 1 receptor antagonist in human monocytes. J Clin Endocrinol Metab 2001; 86:783–791. 49. Santos-Alvarez J, Goberna R, Sanchez-Margalet V. Human leptin stimulates proliferation and activation of human circulating monocytes. Cell Immunol 1999;194:6 –11.
Received March 7, 2002. Accepted December 2, 2002.
LEPTIN AND INFLAMMATION
691
Address requests for reprints to: Charalabos Pothoulakis, M.D., Beth Israel Deaconess Medical Center, Division of Gastroenterology, Dana 501, 330 Brookline Avenue, Boston, Massachusetts 02215. e-mail: [email protected]; fax: (617) 667-2767. Supported by research grants PO DK33506 (to C.P.), PO 1 DK 56116 (to C.M.), RO1 DK 58785 (to C.M.), MO RR 01032 (to C.M.), and a Pilot Feasibility Study from P30 DK 40561 from the National Institutes of Health. This study also was supported by a research grant from the Crohn’s and Colitis Foundation.