Distinct small intestinal modifications such as presence. Interleukin 15 Mediates Epithelial Changes in Celiac Disease

Transcription

1 GASTROENTEROLOGY 2000;119: Interleukin 15 Mediates Epithelial Changes in Celiac Disease LUIGI MAIURI,*, CAROLINA CIACCI, SALVATORE AURICCHIO,*, VIRGINIA BROWN, SONIA QUARATINO, and MARCO LONDEI, Departments of *Pediatrics and Gastroenterology, University Federico II, Naples, Italy; European Laboratory for the Investigation of Food Induced Diseases; and Imperial College School of Medicine, Kennedy Institute of Rheumatology Division, London, England Background & Aims: Villous atrophy and crypt proliferation are key epithelial features of untreated celiac disease. We tried to define whether cytokines such as interleukin (IL)-15, IL-2, IL-4, and IL-7, which share chains of their receptors, could influence the epithelial modifications. Methods: Duodenal biopsy specimens (14 treated and 13 untreated celiac patients, 7 controls) were cultured in vitro for 24 hours with or without gliadin (1 mg/ml), IL-15, IL-7, IL-4, or IL-2 (10 ng/ml). Tumor necrosis factor (TNF)- and interferon (IFN)- were also used in some specimens of untreated celiacs. Epithelial expression of Ki67, FAS, and transferrin receptor (TFR) was detected by immunohistochemistry, and apoptosis by TUNEL technique (percentage of positive enterocytes). IL-15 positive cells were detected by immunohistochemistry in celiac disease and control biopsy specimens; presence of IL-15 was also determined by semiquantitative polymerase chain reaction. Results: Only IL-15 induced enterocyte expression of Ki67, TFR, and FAS in treated celiac (P < 0.01 vs. medium) and enterocyte apoptosis in untreated celiac disease specimens. Anti IL-15 monoclonal antibodies neutralized gliadin-induced enterocyte TFR and FAS expression in treated celiac and enterocyte apoptosis in untreated celiac disease specimens (P < 0.05 vs. gliadin). IL-15 positive cells were increased in untreated celiacs (P < vs. treated celiacs and controls). Conclusions: IL-15 is involved in the modulation of epithelial changes in celiac disease, indicating that this cytokine has an unforeseen role in the pathologic manifestations of celiac disease. Distinct small intestinal modifications such as presence of intraepithelial leukocyte infiltration 1 and local activation of mucosal T cells with production of T-cell cytokines 2,3 characterize celiac disease. The main feature of the small intestine of celiac patients is, however, the classic appearance of flat mucosa, 1 which is still the gold standard for diagnosis. 4 All these modifications are widely considered to be driven by the specific immunologic recognition of gliadin in genetically predisposed individuals. 1,5 It has to be stressed that the epithelial phase is not simply characterized by mucosal atrophy, 1 but clear evidence of epithelial proliferation is also observed in the crypts of patients with active celiac disease. 6 8 Thus, in celiac disease, 2 apparent opposite forces seem to be involved in the pure epithelial stage of the disease, one leading to destruction and the other leading to proliferation. The overwhelming body of studies, performed in the last few years, indicate that the adaptive branch of the immune system holds the key to the pathogenic cascade leading to celiac disease. 2,3,9 12 Some of this evidence has been generated using the organ culture model of small intestine of patients with treated as well as untreated celiac disease, 3,13,14 which allows some dynamic studies that, although possible in vivo in patients, 15 are difficult to execute and analyze. This model has provided many important clues to the understanding of the pathogenesis of celiac disease. To further our studies on the organ culture of celiac disease biopsy specimens, we tried to evaluate whether cytokines produced by T lymphocytes, the archetype of cells of the adaptive branch of the immune system involved in the pathogenesis of celiac disease, could modulate both epithelial modifications observed in celiac disease. Different hypotheses have been proposed; it has been suggested that cytokines, e.g., interferon (IFN)-, 16 released at the mucosal level of celiac disease upon gliadin challenge could be involved in producing such modifications, in particular the destructive ones. In other organ culture models, such as the organ culture of human fetal gut, other cytokines (e.g., tumor necrosis factor [TNF]- ) have been shown to induce a modification reminiscent of the one observed in celiac disease. 17 However, no definitive evidence has been provided on the mechanisms controlling these modifications in celiac dis- Abbreviations used in this paper: EMA, antiendomysial antibodies; IFN, interferon; IL, interleukin; LPMNC, lamina propria mononuclear cell; mab, monoclonal antibody; PT, peptic-tryptic; RT-PCR, reverse-transcription polymerase chain reaction; TFR, transferrin receptor; TNF, tumor necrosis factor; TUNEL, terminal deoxynucleotidyl transferase mediated deoxyuridine triphosphate/digoxigenin nick-end labeling by the American Gastroenterological Association /00/$10.00 doi: /gast

2 October 2000 INTERLEUKIN 15 AND CELIAC DISEASE 997 Figure 1. Effect of IL-15 on TFR antigen expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. Expression of TFR in enterocytes after a 24-hour in vitro challenge of treated celiac disease specimens with or without IL-15 (n 14), IL-7 (n 7), IL-4 (n 7), IL-2 (n 7), or gliadin (n 14). Open bars show low expression (0/1 ) in 70% enterocytes; filled bars show high expression (2 ) in 70% enterocytes. *P 0.01 vs. cultures with medium alone. ease, nor a clear explanation of the connection between the destructive and proliferating phases of the epithelial manifestations of celiac disease. We considered that because the adaptive branch of the immune system clearly controls the evolution of celiac disease, we should have focused our attention on a cytokine mainly produced by activated T cells, which should be present in the small intestine of celiac patients and be able to directly act on epithelial cells. A cytokine that possesses all these characteristics is interleukin (IL)- 2. 2,18 22 As internal controls, 3 other cytokines, IL-4, IL-7, and IL-15, which share 1 of 3 of the receptor chains with IL-2, 18 were used. Some of the latter cytokines have also been reported to influence small intestinal epithelial cell functions. 21,22,23,24 We also included, in a selected group of biopsy specimens, IFN- and TNF-, because they have been reported to modulate epithelial damage. 16,17 Unexpectedly, only IL-15 proved to induce epithelial modifications in patients with celiac disease, but not in controls, with a pattern highly comparable with the one observed on gliadin challenge. 3,13,14,25,26 More importantly, the gliadin-induced epithelial changes were inhibited by neutralizing anti IL-15, but not isotypematched monoclonal antibodies (mabs). These results thus suggest that IL-15, induced by gliadin challenge, is a key factor, but not the only one because others are likely to be involved, in producing the disease-specific epithelial modifications characteristic of celiac disease. This study provides a new focus on the potential role of IL-15 in the pathogenesis of celiac disease. Materials and Methods Patients Duodenal endoscopy was performed, and multiple biopsy specimens were obtained from the duodenal-jejunal flexure for diagnostic purposes in 18 patients with untreated celiac disease (mean age, 41.5 years; range, years; 6 men), who had villous atrophy and serum antiendomysial antibodies (EMA), and from 14 patients with treated celiac disease (10 with follow-up of treated celiac disease). The latter group had been on a gluten-free diet for at least 12 months, and had normal villous architecture 27 and absence of EMA in sera. Ten non celiac disease patients (mean age, 40.5 years; range, years; 4 men) with esophagitis (3 of 10), gastritis (3 of 10), and chronic nonspecific diarrhea (4 of 10) were used as controls. All had normal duodenal morphology and were EMA negative. Fully informed consent was obtained from all individuals. All specimens were washed in 0.15 mol/l sodium chloride and examined with a dissecting microscope. One specimen from each patient was oriented and embedded in O.C.T. compound (Tissue Tek, Miles Laboratories, Elkhart, IN), snap-frozen in isopenthane, cooled in liquid nitrogen, and stored at 70 C until cryosectioning. Sections of 5 m were stained with hematoxylin and used for diagnosis. The other samples were cultured in vitro as described later. Preparation of the Culture Medium and In Vitro Culture In vitro organ culture of biopsy specimens from treated and untreated celiac disease and controls. Duodenal explants from all patients with treated celiac disease were cultured in vitro for 24 hours in the presence of medium alone as well as medium added with peptic-tryptic (PT) digest of gliadin from bread wheat (1 mg/ml), as reported previously. 3,13,14 IL-15 (Immunex, Seattle, WA), IL-7 (Immunex), IL-4 (Roche, Basel, Switzerland), and IL-2 (Sandoz, Basel, Figure 3. Effect of IL-15 on Ki67 antigen expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. Expression of enterocyte Ki67 after a 24-hour in vitro challenge of treated celiac disease specimens with or without IL-15 (n 10), IL-7 (n 7), IL-4 (n 7), IL-2 (n 7), or gliadin (n 10). *P ; P 0.05 vs. cultures with medium alone.

3 998 MAIURI ET AL. GASTROENTEROLOGY Vol. 119, No. 4 Figure 9. Figure 2. Effect of IL-15 on TFR antigen expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. (A) Expression of TFR after a 24-hour challenge of treated celiac disease biopsy specimens with medium alone. Weak staining is observed in some crypt enterocytes. (B) Expression of TFR after a 24-hour challenge with IL-15. Intense staining is observed in the cytoplasm of crypt enterocytes. (Peroxidase staining technique; original magnification 180.) Figure 12. Figure 4. Effect of IL-15 on Ki67 antigen expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. (A) Expression of Ki67 after a 24-hour challenge of treated celiac disease specimens with medium alone. Expression of Ki67 is observed in a few crypt enterocytes. (B) Expression of Ki67 after a 24-hour challenge of treated celiac disease specimens with IL-15. Ki67 nuclear expression is evident in many crypt enterocytes. (Peroxidase staining technique; original magnification 180.) Figure 6. Figure 14.

4 October 2000 INTERLEUKIN 15 AND CELIAC DISEASE 999 Switzerland) were also used in some biopsy specimens (Figures 1 3) at a final concentration of 10 ng/ml, which was chosen after titration experiments. Neutralizing anti IL-15 mabs M110 or M111 IgG1 (Immunex; 5 g/ml) or isotype control mabs, anti B7-1 clone BB7.1, IgG1 (5 g/ml; CAMFOLIO; Becton Dickinson, San Jose, CA), or anti-human lactase mlac1, IgG1 (5 g/ml), 28 were added to PT gliadin digest of some treated celiac disease samples (Figure 4). Duodenal explants from 5 patients with treated celiac disease were also cultured for 3 hours with or without PT gliadin digest or maize digest (1 mg/ml). Duodenal explants (9) from untreated celiac disease patients with villous atrophy were cultured for 24 hours in the presence or absence of IL-15 (10 ng/ml) or PT gliadin digest (1 mg/ml) with or without neutralizing anti IL-15 mabs M110 and M111 IgG1 (5 g/ml) or control antibodies, as described earlier. In 4 cases, in vitro cultures were performed with or without PT gliadin digest, IL-15, TNF- (BASF, Ludwigshafen, Germany), and IFN- (20 ng/ml; Roche, Basel, Switzerland). Duodenal biopsy specimens from 7 non celiac disease controls were cultured in vitro for 24 hours in the presence of the sole medium and medium added with PT gliadin digest (1 mg/ml) or IL-15 (10 ng/ml). Some of these specimens were cultured for 3 hours with or without PT gliadin digest (1 mg/ml). Markers and Cell Detection The following mabs were used on frozen tissue sections: Ki67 (Dako, Copenhagen, Denmark; 1:25), transferrin receptor (TFR; Dako; 1:30), IL-15 (M110, M111, or M112 mouse IgG1 mabs; 5 g/ml), and FAS (M3 and M38, mouse Ig mabs; Immunex; 1:30), and immunodetection was performed by alkaline phosphatase or indirect immunofluorescence (IL-15), peroxidase (TFR, Ki67, FAS) staining techniques. For detection of IL-15, serial sections of the same duodenal explants were stained with the 3 different mabs to validate the IL-15 staining. In some cases, to amplify the signal, the procedure was repeated twice. Immunofluorescence was used because it allows a better resolution of 2-color procedures, as described below. Two-color immunofluorescence was used to characterize the type of cells producing IL-15 by anti IL-15 mab, immunodetection by biotinylated goat anti-mouse Ig and R-phycoerythrin conjugated streptavidin (red), and fluorescein isothiocyanate conjugated anti- CD68 mab, clone KP1, IgG (Pharmingen, San Diego, CA; green). Double staining is yellow, resulting from the overlap of green (CD68 ) and red (IL-15 positive). The morphometric assessment of stained lamina propria mononuclear cells (LPMNCs) was done as previously described. 3,13 The number of crypt enterocytes, which express Ki67 antigen, was calculated as percentage of total crypt enterocytes. Staining of epithelial cells by anti-fas and anti-tfr mabs was arbitrarily graded from 0 to 2, based on the intensity of staining in 70% of the cells (undetectable, 0; low, 1 ; intense, 2 ). Guidelines for this scoring system were established at the start of the study, and the samples were independently analyzed by 2 observers; the results were compared afterwards and found to be equivalent. At least 5 slides for each sample were blindly evaluated. Specificity control experiments were performed using mouse IgG or IgM against inappropriate blood group antigens or other unrelated antigens, 28 and simultaneously analyzing different cultured samples belonging to the same individual. IL-15 Quantitative Competitive Reverse- Transcription Polymerase Chain Reaction in Whole Biopsy Specimens Competitive reverse-transcription polymerase chain reaction (RT-PCR) was used to quantitate IL-15 transcripts from Š Figure 6. Effect of IL-15 on FAS expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. (A) FAS expression after a 24-hour challenge of treated celiac disease specimens with medium alone. Very faint staining is observed on basolateral cell membranes of a few enterocytes. (B) FAS expression after 24 hours of challenge of treated celiac disease biopsy specimens with IL-15. Intense FAS expression is evident on basolateral cell membranes of most enterocytes and in some LPMNCs. (Peroxidase staining technique; original magnification 160.) Figure 9. (A) Pattern of enterocyte DNA fragmentation after a 24-hour gliadin challenge of untreated celiac disease specimens with villous atrophy. Many enterocytes are TUNEL positive; the staining is also evident in LPMNCs. (B) Pattern of DNA fragmentation after a 24-hour in vitro challenge of untreated celiac disease specimens with gliadin supplemented with neutralizing anti IL-15 M110 mab. A marked reduction in the number of TUNEL-positive enterocytes is observed with respect to the pattern observed after gliadin challenge. (TUNEL technique; original magnification 160.) Figure 12. IL-15 protein expression in celiac disease and control small intestine. (A) Expression of IL-15 in untreated celiac disease intestine. Some LPMNCs are clearly stained (red) in untreated celiac disease duodenum. (B) Expression of IL-15 in treated celiac disease intestine. No detectable staining is found in the lamina propria region shown. Only a very few LPMNCs are IL-15 positive when 1 mm 2 of lamina propria is analyzed. (Alkaline phosphatase staining technique; original magnification 160.) Figure 14. IL-15 protein expression after 3 hours of in vitro challenge of treated celiac disease small intestinal biopsy specimens with PT gliadin digest. (A) Three-hour in vitro organ culture of treated celiac disease specimens with medium alone. Only a very few IL-15 positive cells are detected (arrow); this pattern is similar to that observed before in vitro manipulation. (B) Three-hour in vitro challenge of treated celiac disease specimens with PT gliadin digest. Numerous IL-15 positive LPMNCs can be seen (arrow). (C) Three-hour in vitro challenge of treated celiac disease specimens with PT gliadin digest: distribution of IL-15 protein expression. IL-15 positive cells detected have the phenotype of resident macrophage/monocytes. Two-color immunofluorescence with anti IL-15 (red) and CD68 (green) mabs (see Materials and Methods). Note that some cells (5 in the present field) show colocalization of both markers (orange/yellow). Some CD68 cells do not express IL-15 (green). (Immunofluorescence; original magnification 150 [A and B] and 260 [C].)

5 1000 MAIURI ET AL. GASTROENTEROLOGY Vol. 119, No. 4 intestinal specimens. The RT reaction was performed as described previously 29 from 2 g total RNA. To normalize each RT reaction, 1 L of each complementary DNA (cdna) was amplified using the following human -actin primers: CAC- CTTCTACAATGAGCTGCG (forward) and CAGCACTGT- GTTGGCGTACAGG (reverse). The competitor template for IL-15 (kindly provided by Dr. N. Azimi, National Institutes of Health, Bethesda, MD) was generated by inserting a 60 base pair fragment in IL-15 cdna. 30 The endogenous IL-15 and the competitor template were amplified using the primers GAAACCACATTT- GAGAAGTATTTC (forward) and CCATTAGAAGACAAA- CTGTTG (reverse). In each reaction, a known concentration of competitor was added to 2 L of endogenous IL-15 cdna. The amount of the competitor used in each reaction is indicated in the Figure 10. Endogenously amplified IL-15 cdna migrated below the competitor IL-15 DNA. The conditions for the PCR reaction were 92 C for 4 minutes (one cycle), followed by 30 cycles at 92 C for 10 seconds, 55 C for 30 seconds, and 72 C for 1 minute; and 72 C for 4 minutes (one cycle) on an MJ Research PTC-200 DNA Engine (Watertown, MA). Ten microliters of the amplified PCR products was run on a 2% agarose gel. Detection of DNA Fragmentation DNA fragmentation on tissue section was assayed by terminal deoxynucleotidyl transferase mediated deoxyuridine triphosphate/digoxigenin nick-end labeling (TUNEL), as described previously. 31 The number of TUNEL-positive enterocytes was calculated as percentage of total surface and crypt enterocytes. Statistical Analysis The samples belonging to every category were compared with each other. The Student 2-tailed t test for paired samples was applied for the analysis of the number of mucosal cells that express Ki67 or IL-15, as well as for the analysis of TUNEL-positive enterocytes. Nonparametric tests (Wilcoxon test) were also applied, and the results were concordant with those obtained using parametric tests. The Fisher test was applied to compare tissues with FAS or TFR expression that was undetectable or low (0 to 1 ) with those showing intense (2 ) expression. Tissue specimens, after challenge with the tested cytokines or gliadin, were compared with those after challenge with medium alone or with gliadin added with neutralizing anti IL-15 mabs. Results IL-15 Challenge Mimics Most of the Epithelial Modifications Induced by Gliadin Challenge in Cultures of Treated Celiac Disease Duodenum We investigated whether IL-15 could induce increase in TFR expression, because this cell proliferation marker is overexpressed in crypt enterocytes of untreated celiac disease intestine. 32 IL-15, as well as gliadin, induced TFR expression (intense TFR expression in 70% enterocytes) in crypt and villous enterocytes (Figure 1). IL-4 was also able to increase the expression of TFR in enterocytes, but to a reduced degree compared with IL-15, and higher doses did not significantly modify the pattern (data not shown; and Figures 1 and 2A and B), whereas IL-7 and IL-2 did not generate any TFR up-regulation (Figure 1). In all experiments scored as negative, increased expression of TFR was infrequently observed in very few villous or crypt enterocytes. No effect of IL-15 or gliadin was observed in control biopsy specimens after 24 hours of in vitro challenge (7 cultured specimens with IL-15 or with gliadin, intense TFR expression in 70% of enterocytes in 0 of 7 cases vs. 0 of 7 cases after challenge with medium alone; P 0.5). We also analyzed whether IL-15 could up-regulate the expression of Ki67, a widely used marker to study cell proliferation 33 also in intestinal inflammatory pathologies. 34,35 Ki67 expression is indeed increased in untreated celiac disease intestine. 8 In treated celiac disease specimens, IL-15 drives the expression of Ki67 in crypt epithelial cells (Figures 3 and 4A and B), whereas no effect on Ki67 expression was observed after 24-hour challenge with gliadin (Figure 3). In the 7 cultured specimens, IL-7 and IL-4 were devoid of such in vitro effect, whereas IL-2 was effective, although to a much lesser degree than IL-15, in inducing Ki67 expression in crypt epithelial cells (Figure 3). No effect of IL-15 was observed in control biopsy specimens after 24 hours of in vitro challenge (7 cultured specimens with IL-15; mean [SD] percentage of Ki67 crypt enterocytes, 2.28 [0.9] vs. 2.4 [0.9] after culture with medium alone; P 0.5). IL-15 was also effective in inducing FAS expression (intense FAS expression in 70% enterocytes) in crypt and villous enterocytes (Figures 5 and 6A and B) of treated celiac disease specimens. Manifestation of FAS was prevalent on basolateral cell membranes of enterocytes (Figure 6B), as previously described in other studies. 36 Gliadin challenge was similarly able to significantly enhance, in 24 hours of challenge, enterocyte FAS expression (Figure 5). IL-7, IL-4, and IL-2 were unable to increase the expression of FAS in enterocytes (Figure 5). In all negative samples, intense FAS expression was rarely observed in a few villous or crypt enterocytes. No effect of IL-15 or gliadin was observed in control specimens after 24 hours of in vitro challenge (7 cultured biopsies with IL-15 or gliadin; intense FAS expression in 70% of enterocytes

6 October 2000 INTERLEUKIN 15 AND CELIAC DISEASE 1001 Figure 5. Effect of IL-15 on FAS expression by enterocytes of treated celiac disease small intestinal biopsy specimens after 24 hours of in vitro challenge. Expression of FAS in enterocytes after a 24-hour in vitro challenge of treated celiac disease specimens with or without IL-15 (n 14), IL-7 (n 7), IL-4 (n 7), IL-2 (n 7), or gliadin (n 14). Open bars show low expression (0/1 ) in 70% enterocytes; filled bars show high expression (2 ) in 70% enterocytes. *P 0.01 vs. cultures with medium alone. in 0 of 7 cases vs. 0 of 7 cases after challenge with medium alone; P 0.5). Neutralizing Anti IL-15 mabs Control Gliadin-Induced Enterocyte Modifications in Treated Celiac Disease Biopsy Specimens Because gliadin and IL-15 had a very similar spectrum of effects on enterocytes of cultured treated celiac disease explants, we tested whether neutralization of IL-15 could eliminate the gliadin-induced enterocyte modifications. The neutralizing anti IL-15 mabs (M110 and M111) were highly effective and consistently blocked gliadin-induced expression of TFR and FAS on enterocytes (Figure 7A and B). Isotype-matched control mabs did not interfere with gliadin-induced up-regulation of TFR and FAS (Figure 7A and B). To validate the efficacy of neutralizing anti IL-15 M110 and M111 mabs in the organ culture system of duodenal explants, 2 further treated celiac disease specimens were challenged with IL-15 and IL-15 supplemented with anti IL-15 M110 or M111 mabs, respectively. In both cases, all the epithelial modifications induced by IL-15 on celiac intestine (increased expression of Ki67, TFR, and FAS) were controlled by neutralizing anti IL-15 mabs, but not by isotype-matched control antibodies (data not shown). IL-15 and Gliadin Induce Epithelial Apoptosis in Biopsy Specimens of Untreated Celiac Patients That Is Inhibited by Neutralizing Anti IL-15 mab To expand our study on IL-15, gliadin epithelial modifications, and celiac disease, we studied the effects of IL-15, gliadin, and control cytokines in organ culture of untreated celiac disease. In cultures of untreated celiac disease biopsy specimens with villous atrophy, IL-15 increased the number of TUNEL-positive enterocytes (Figure 8A). The same effect was observed after gliadin challenge (Figure 8A). None of the other cytokines (IL-2, IL-4, and IL-7) (data not shown) induced any increase of TUNEL-positive epithelial cells compared with medium alone. Conversely, TNF- induced mucosal damage, but the epithelial pattern was different from the one induced by IL-15; similar results were obtained with IFN- (Table 1). To define whether a chronic inflammatory milieu could promote the epithelial apoptotic effects of gliadin, Figure 7. Effect of neutralizing anti IL-15 M110 mab on epithelial modifications induced by gliadin challenge after 24 hours of in vitro culture of treated celiac disease small intestinal biopsy specimens. (A) Effect of M110 mab and of isotype-matched control mabs (antihuman lactase mlac1 and anti-cd80 BB7.1 mabs; n 5) on gliadininduced TFR enterocyte expression. Open bars show low expression (0/1 )in 70% enterocytes; filled bars show high expression (2 )in 70% enterocytes. P vs. cultures with medium alone; *P 0.02 vs. cultures with PT gliadin digest. (B) Effect of M110 mab and of isotype-matched control mabs (antihuman lactase mlac1 and anti- CD80 BB7.1 mabs; n 5) on gliadin-induced FAS enterocyte expression. Open bars show low expression (0/1 ) in 70% enterocytes; filled bars show high expression (2 ) in 70% enterocytes. P vs. cultures with medium alone; *P 0.02 vs. cultures with PT gliadin digest.

7 1002 MAIURI ET AL. GASTROENTEROLOGY Vol. 119, No. 4 Figure 8. Cultured untreated celiac disease biopsy specimens: effect of IL-15 on enterocyte DNA fragmentation and neutralizing effect of anti IL-15 M110 mab on gliadin-induced enterocyte DNA fragmentation. (A) Effect of IL-15 and gliadin on enterocyte DNA fragmentation after a 24-hour challenge of untreated celiac disease specimens with villous atrophy (n 4). *P 0.05 vs. cultures with medium alone. (B) Effect of M110 mab and of isotype-matched control mabs (anti-human lactase mlac1 and anti-cd80 BB7.1 mabs) on gliadin-induced enterocyte DNA fragmentation after 24-hour challenge of untreated celiac disease biopsy specimens with villous atrophy (n 5). P 0.01 vs. cultures with medium alone; *P 0.01 vs. cultures with PT gliadin digest. organ cultures of 2 patients with Crohn s disease were also studied with no case apoptosis observed (data not shown). To define the potential relevance of IL-15 in the gliadin-induced epithelial apoptosis, we used neutralizing anti IL-15 mabs and isotype-matched control mabs in combination with gliadin challenge. In these experimental conditions, the proapoptotic action of gliadin was severely curbed by the use of neutralizing anti IL-15 mabs, but not controls (Figures 8A and B and 9A and B). Expression of IL-15 in Celiac and Control Intestine The results so far obtained indicated that IL-15 might have a role in the progression of celiac disease. We therefore analyzed the messenger RNA (mrna) and protein expression in biopsy specimens from patients with untreated celiac disease and controls. To monitor mrna levels, we implemented a competitive PCR method 30 ; we detected mrna in controls as well as Table 1. Comparison of the Effects of Gliadin, IL-15, TNF-, and IFN- on Biopsy of Specimens From Patients With Untreated Celiac Disease Gliadin IL-15 TNF- IFN- Increase of enterocyte apoptosis a 4/4 4/4 2/4 2/4 Increase of enterocyte Ki67 expression a 4/4 4/4 0/4 0/4 Mucosal swelling b 0/4 0/4 3/4 2/4 Presence of large necrotic areas b 0/4 0/4 4/4 2/4 a With respect to the pattern observed after incubation with medium alone. b Histologic assessment of the specimens after culture was blindly performed to detect mucosal swelling and necrotic areas. diseased samples, and no difference was found (Figure 10). By immunohistochemistry, however, we detected a significantly increased number of lamina propria cells that stained positive for IL-15 in untreated celiac disease patients, but not in treated patients or controls, although the absolute number of IL-15 positive cells was relatively low (Figures 11 and 12A and B). Gliadin Induces IL-15 Positive Cells After 3 Hours of In Vitro Challenge of Treated Celiac Disease Biopsy Specimens As soon as 3 hours after challenge with gliadin, several LPMNCs became IL-15 positive (Figures 13 and 14A and B). The stained cells showed the phenotype of resident macrophages/monocytes, as revealed by double immunofluorescence (Figure 14C). No increase in the number of IL-15 positive cells was observed upon challenge with other alimentary antigens such as maize (Figure 13), nor was an increase of IL-15 positive cells observed in controls challenged with gliadin (5 cultured specimens; mean [SD] number of IL15 positive cells/ mm 2 lamina propria, 5.4 [2.7] vs. 4.8 [1.9] after challenge with medium alone; P 0.5). We choose to identify the presence of IL-15 positive cells by immunohistochemistry or immunofluorescence (both techniques gave overlapping results) rather than in situ hybridization, because mrna expression does not correlate with protein production caused by posttranscriptional regulation of IL-15 mrna. 37 Also, to validate our findings, we used 3 different mabs specific for IL-15, recognizing different epitopes on IL-15, which provided totally overlapping results.

8 October 2000 INTERLEUKIN 15 AND CELIAC DISEASE 1003 Figure 10. IL-15 competitive PCR in celiac disease and control small intestine. Competitive RT-PCR using RNA isolated from intestinal biopsy specimens from 3 patients and 3 controls. Total RNA (2 g) was used to generate cdna in an RT reaction. Next, 2 L of cdna was amplified in a competitive PCR using decreasing concentrations of the competitor template. Endogenously amplified IL-15 migrates below the competitor in the agarose gel. The amounts of competitor DNA added to each PCR reaction were: ,2 10 3,2 10 4,2 10 5, and ng. Amplification of -actin served as control to normalize the amount of cdna input in the competitive PCR reaction. Discussion The mucosa of patients with untreated celiac disease is characterized by modifications that are epitomized by the appearance of villous atrophy, 1 which is the trademark of celiac disease and still used as the gold standard for diagnostic purposes. 4 The presence of villous atrophy is clearly suggestive of a destructive event that has been widely studied and that has conditioned the thinking and strategy of research for many scientists. Different hypotheses have been proposed to explain the origin of the mucosal damage: from the direct toxic action of gliadin to a secondary killing of epithelial cells by the adaptive branch of the immune system, in particular T cells and cytokines released upon gliadin challenge. 1,16 In the mucosa of celiac patients, there is not only a destructive event, leading to the flat mucosa, but also a preponderant proliferative response, 1,6,8 which leads to an overall hypertrophy of the mucosa, as described by Marsh. 1 Thus, in active celiac disease, 2 opposite forces are acting in the epithelial compartment, and so far no clear explanation of the mechanisms controlling these 2 phases has been satisfactorily proposed. In celiac disease, it is widely accepted that the immunologic recognition and subsequent activation of the adaptive branch of the immune system modulate the pathognomonic pattern of celiac disease. 5,9 It is conceptually easy to accept that activation of the immune system leads to epithelial damage directly or indirectly via the release of proinflammatory cytokines, 2,3,9 12,16 Figure 11. IL-15 protein expression in celiac disease and control small intestine. Expression of IL-15 by LPMNCs of untreated celiac (n 14), treated celiac (n 14), and control (n 10) intestine (mean SD). *P vs. treated celiac disease and control samples. Figure 13. IL-15 protein expression after 3 hours of in vitro challenge of treated celiac disease small intestinal biopsy specimens with PT gliadin digest. Expression of IL-15 by LPMNCs of treated celiac disease intestine after a 3-hour in vitro culture with or without PT gliadin digest (n 5; mean SD). *P vs. samples cultured with medium alone.

9 1004 MAIURI ET AL. GASTROENTEROLOGY Vol. 119, No. 4 but less clear is the mechanism that modulates the repairing (proliferating) phase. 6,35,42 A repairing phase must be present and logically follow the destructive events leading to mucosal atrophy. However, whether the proliferating phase precedes or follows the destructive phase is debated, and it has also been hypothesized that the proliferating phase may precede the destructive one. 1,8 Previous reports have shown in different experimental models, such as human fetal organ culture, that keratinocyte growth factor and transforming growth factor might have a central function in up-regulating proliferation of mucosal epithelial cells, thus being involved in this remodeling/proliferating phase often observed in human gastrointestinal pathologies. 35 Interestingly, keratinocyte growth factor in particular is produced by mesenchymal cells in the lamina propria, and, although not yet confirmed, it could play a role in the regenerative phase even in celiac disease. In this report, we show that a new player, IL-15, may be involved in both epithelial phases of celiac disease. IL-15 is a pleiotropic cytokine that has been considered to be closely related to IL-2 because it shares many of the biological actions and also most of the receptor chains. 37 However, IL-15 has many unique features that clearly differentiate it from IL-2: the producing cell types are lymphoid for IL-2 (mainly T cells), 18 whereas for IL-15 they are myeloid or mesenchymal cells such as monocytes and epithelial cells. 37 The most intriguing difference is the way in which the production of these cytokines is controlled. For IL-2, there is strict control at the level of mrna transcription and stabilization, thus the generation and stability of mrna that tunes the production of this cytokine. 18 Posttranscriptional regulatory mechanisms, on the contrary, dominate the control of IL-15 secretion. 37 Thus, the detection of IL-15 specific mrna does not have the same meaning as mrna detection of IL-2, as studied, for instance, in celiac disease. 2,43 Often, the unequivocal presence and effects of IL-15 are proved by immunostaining and the use of neutralizing antibodies to block the biological actions ascribed to IL-15, as we do in this study, even when no protein is detected by enzyme-linked immunosorbent assay or bioassay. 44 We have experienced the same problem, because competitive PCR did not differentiate biopsy specimens in which IL-15 was detected by immunohistochemistry from specimens in which IL-15 was not detected. These results have been supported by other groups (R. Troncone and V. Salvati, personal communication, March 2000), and the pattern of expression, although significantly increased in untreated celiac disease, is not comparable with the one observed in other pathologies such as cutaneous T-cell lymphoma. 45 Some reports indicate that IL-15 is involved in other inflammatory pathologies such as rheumatoid arthritis and, more pertinent to this report, in inflammatory bowel disease The latter evidence on IL-15 and inflammatory bowel disease has been challenged, however, by studies from other groups. 52 In any case, IL-15, a typical cytokine of the innate immune system, seems to be involved in several immunomediated pathologies 37 and, as we report now, in celiac disease. Clearly, IL-15 does not induce identical modifications in these different settings. Our present results do not exclude the involvement of other factors and/or cytokines such as IFN- and TNF-, although in our study they did not induce the same modifications as IL-15. Although celiac disease is clearly associated with an involvement of the adaptive branch of the immune system, 37 evidence indicates that a more complex pathogenic pattern controls the disease evolution. Previous studies by our groups have suggested a dual involvement of the innate and adaptive branch of the immune system. 3,13 The present study continues in this direction pointing to IL-15, a product of monocytes, mesenchymal or epithelial cells, 37 as a potential link between innate and adaptive sections of the immune systems in celiac disease. Importantly, this cytokine also acts on small intestinal epithelial cells. 24 Our study provides some results that are apparently in disagreement with the literature; in particular, it has been reported that IL-15 protects from the induction of apoptosis, 53 whereas in our study it seems to drive it. These differences can be explained by the differential experimental settings (mouse and human) and different cell types in which the antiapoptotic action of IL-15 was observed, because lymphoid cells and hepatocytes were the cell types mainly protected. 53 Further studies are required to reconcile these apparent differences. However, the genetic makeup of patients with celiac disease as well as a previous activation or priming, even in perfectly treated celiacs, of the mucosal environment are likely to contribute to the IL-15 sensitivity in these patients. Our report indicates that IL-15 has a part in the whole progression of the disease, a finding that represents a novel view in the pathogenesis of celiac disease. References 1. Marsh MN. Gluten, major histocompatibility complex, and the small intestine. A molecular and immunobiologic approach to the spectrum of gluten sensitivity ( celiac sprue ). Gastroenterology 1992;102: Nilsen EM, Jahnsen FL, Lundin KE, Johansen FE, Fausa O, Sollid LM, Jahnsen J, Scott H, Brandtzaeg P. Gluten induces an intestinal cytokine response strongly dominated by interferon

10 October 2000 INTERLEUKIN 15 AND CELIAC DISEASE 1005 gamma in patients with celiac disease. Gastroenterology 1998; 115: Maiuri L, Auricchio S, Coletta S, De Marco G, Picarelli A, Di Tola M, Quaratino S, Londei M. Blockage of T-cell costimulation inhibits T-cell action in celiac disease. Gastroenterology 1998;115: Maki M, Collin P. Coeliac disease. Lancet 1997;349: Sollid LM, Thorsby E. HLA susceptibility genes in celiac disease: genetic mapping and role in pathogenesis. Gastroenterology 1993;105: Savidge TC, Shmakov AN, Walker-Smith JA, Phillips AD. Epithelial cell proliferation in childhood enteropathies. Gut 1996;39: Przemioslo R, Wright NA, Elia G, Ciclitira PJ. Analysis of crypt cell proliferation in coeliac disease using MI-B1 antibody shows an increase in growth fraction. Gut 1995;36: Moss SF, Attia L, Scholes JV, Walters JR, Holt PR. Increased small intestinal apoptosis in coeliac disease. Gut 1996;39: Scott H, Nilsen E, Sollid LM, Lundin KE, Rugtveit J, Molberg O, Thorsby E, Brandtzaeg P. Immunopathology of gluten-sensitive enteropathy. Springer Semin Immunopathol 1997;18: Lundin KE, Scott H, Hansen T, Paulsen G, Halstensen TS, Fausa O, Thorsby E, Sollid LM. Gliadin-specific, HLA-DQ(alpha 1*0501,beta 1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients. J Exp Med 1993; 178: Molberg O, McAdam SN, Korner R, Quarsten H, Kristiansen C, Madsen L, Fugger L, Scott H, Noren O, Roepstorff P, Lundin KE, Sjostrom H, Sollid LM. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998;4: Schuppan D, Dieterich W, Riecken EO. Exposing gliadin as a tasty food for lymphocytes. Nat Med 1998;4: Maiuri L, Picarelli A, Boirivant M, Coletta S, Mazzilli MC, De Vincenzi M, Londei M, Auricchio S. Definition of the initial immunologic modifications upon in vitro gliadin challenge in the small intestine of celiac patients. Gastroenterology 1996;110: Picarelli A, Maiuri L, Frate A, Greco M, Auricchio S, Londei M. Production of antiendomysial antibodies after in-vitro gliadin challenge of small intestine biopsy samples from patients with coeliac disease. Lancet 1996;348: Sturgess R, Day P, Ellis HJ, Lundin KE, Gjertsen HA, Kontakou M, Ciclitira PJ. Wheat peptide challenge in coeliac disease. Lancet 1994;343: Przemioslo RT, Lundin KE, Sollid LM, Nelufer J, Ciclitira PJ. Histological changes in small bowel mucosa induced by gliadin sensitive T lymphocytes can be blocked by anti-interferon gamma antibody. Gut 1995;36: Pender SL, Tickle SP, Docherty AJ, Howie D, Wathen NC, Mac- Donald TT. A major role for matrix metalloproteinases in T cell injury in the gut. J Immunol 1997;158: Waldmann T, Tagaya Y, Bamford R. Interleukin-2, interleukin-15, and their receptors. Int Rev Immunol 1998;16: Kontakou M, Przemioslo RT, Sturgess RP, Limb GA, Ellis HJ, Day P, Ciclitira PJ. Cytokine mrna expression in the mucosa of treated coeliac patients after wheat peptide challenge. Gut 1995;37: Ciacci C, Mahida YR, Dignass A, Koizumi M, Podolsky DK. Functional interleukin-2 receptors on intestinal epithelial cells. J Clin Invest 1993;92: Dignass AU, Podolsky DK. Interleukin 2 modulates intestinal epithelial cell function in vitro. Exp Cell Res 1996;225: Reinecker HC, Podolsky DK. Human intestinal epithelial cells express functional cytokine receptors sharing the common gamma c chain of the interleukin 2 receptor. Proc Natl Acad Sci U S A 1995;92: Reinecker HC, MacDermott RP, Mirau S, Dignass A, Podolsky DK. Intestinal epithelial cells both express and respond to interleukin 15. Gastroenterology 1996;111: Colgan SP, Resnick MB, Parkos CA, Delp-Archer C, McGuirk D, Bacarra AE, Weller PF, Madara JL. IL-4 directly modulates function of a model human intestinal epithelium. J Immunol 1994; 153: Halstensen TS, Scott H, Fausa O, Brandtzaeg P. Gluten stimulation of coeliac mucosa in vitro induces activation (CD25) of lamina propria CD4 T cells and macrophages but no crypt-cell hyperplasia. Scand J Immunol 1993;38: Fais S, Maiuri L, Pallone F, De Vincenzi M, De Ritis G, Troncone R, Auricchio S. Gliadin induced changes in the expression of MHC-class II antigens by human small intestinal epithelium. Organ culture studies with coeliac disease mucosa. Gut 1992; 33: Kuitunen P, Kosnai I, Savilahti E. Morphometric study of the jejunal mucosa in various childhood enteropathies with special reference to intraepithelial lymphocytes. J Pediatr Gastroenterol Nutr 1982;1: Maiuri L, Rossi M, Raia V, Garipoli V, Hughes LA, Swallow D, Noren O, Sjostrom H, Auricchio S. Mosaic regulation of lactase in human adult-type hypolactasia. Gastroenterology 1994;107: Quaratino S, Feldmann M, Dayan CM, Acuto O, Londei M. Human self-reactive T cell clones expressing identical T cell receptor beta chains differ in their ability to recognize a cryptic self-epitope. J Exp Med 1996;183: Azimi N, Jacobson S, Leist T, Waldmann TA. Involvement of IL-15 in the pathogenesis of human T lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis: implications for therapy with a monoclonal antibody directed to the IL-2/15R beta receptor. J Immunol 1999;163: Maiuri L, Raia V, De Marco G, Coletta S, de Ritis G, Londei M, Auricchio S. DNA fragmentation is a feature of cystic fibrosis epithelial cells: a disease with inappropriate apoptosis. FEBS Lett 1997;408: Maiuri L, Troncone R, Fais S, Coletta S, Picarelli A, Pallone F. Crypt epithelial cells express the 4F2 antigen in untreated coeliac mucosa. Adv Exp Med Biol 1995;371B: Brown DC, Gatter KC. Monoclonal antibody Ki67: its use in histopathology. Histopathology 1990;17: MacDonald TT. Epithelial proliferation in response to gastrointestinal inflammation. Ann N Y Acad Sci 1992;664: Bajaj-Elliott M, Poulsom R, Pender SL, Wathen NC, MacDonald TT. Interactions between stromal cell derived keratinocyte growth factor and epithelial transforming growth factor in immune-mediated crypt cell hyperplasia. J Clin Invest 1998;102: Leithauser F, Dhein J, Mechtersheimer G, Koretz K, Bruderlein S, Henne C, Schmidt A, Debatin KM, Krammer PH, Moller P. Constitutive and induced expression of APO-1, a new member of the nerve growth factor/tumor necrosis factor receptor superfamily, in normal and neoplastic cells. Lab Invest 1993;69: Waldmann TA, Tagaya Y. The multifaceted regulation of interleukin-15 expression and the role of this cytokine in NK cell differentiation and host response to intracellular pathogens. Annu Rev Immunol 1999;17: Auricchio S, De Ritis G, De Vincenzi M, Magazzu G, Maiuri L, Mancini E, Minetti M, Sapora O, Silano V. Mannan and oligomers of N-acetylglucosamine protect intestinal mucosa of celiac patients with active disease from in vitro toxicity of gliadin peptides. Gastroenterology 1990;99: Auricchio S, De Ritis G, De Vincenzi M, Gentile V, Maiuri L, Mancini E, Porta R, Raia V. Amines protect in vitro the celiac

11 1006 MAIURI ET AL. GASTROENTEROLOGY Vol. 119, No. 4 small intestine from the damaging activity of gliadin peptides. Gastroenterology 1990;99: Auricchio S, De Ritis G, De Vincenzi M, Silano V. Toxicity mechanisms of wheat and other cereals in celiac disease and related enteropathies. J Pediatr Gastroenterol Nutr 1985;4: Weiser MM, Douglas AP. An alternative mechanism for gluten toxicity in coeliac disease. Lancet 1976;1: Podolsky DK. Healing the epithelium: solving the problem from two sides. J Gastroenterol 1997;32: Kontakou M, Przemioslo RT, Sturgess RP, Limb AG, Ciclitira PJ. Expression of tumour necrosis factor-alpha, interleukin-6, and interleukin-2 mrna in the jejunum of patients with coeliac disease. Scand J Gastroenterol 1995;30: Barzegar C, Meazza R, Pereno R, Pottin-Clemenceau C, Scudeletti M, Brouty-Boye D, Doucet C, Taoufik Y, Ritz J, Musselli C, Mishal Z, Jasmin C, Indiveri F, Ferrini S, Azzarone B. IL-15 is produced by a subset of human melanomas, and is involved in the regulation of markers of melanoma progression through juxtacrine loops. Oncogene 1998;16: Dobbeling U, Dummer R, Laine E, Potoczna N, Qin JZ, Burg G. Interleukin-15 is an autocrine/paracrine viability factor for cutaneous T-cell lymphoma cells. Blood 1998;92: McInnes IB, al-mughales J, Field M, Leung BP, Huang FP, Dixon R, Sturrock RD, Wilkinson PC, Liew FY. The role of interleukin-15 in T-cell migration and activation in rheumatoid arthritis. Nat Med 1996;2: McInnes IB, Leung BP, Sturrock RD, Field M, Liew FY. Interleukin-15 mediates T cell dependent regulation of tumor necrosis factor alpha production in rheumatoid arthritis. Nat Med 1997; 3: McInnes IB, Liew FY. Interleukin 15: a proinflammatory role in rheumatoid arthritis synovitis. Immunol Today 1998;19: Kirman I, Nielsen OH. Increased numbers of interleukin-15 expressing cells in active ulcerative colitis. Am J Gastroenterol 1996;91: Sakai T, Kusugami K, Nishimura H, Ando T, Yamaguchi T, Ohsuga M, Ina K, Enomoto A, Kimura Y, Yoshikai Y. Interleukin 15 activity in the rectal mucosa of inflammatory bowel disease. Gastroenterology 1998;114: Liu Z, Geboes K, Colpaert S, D Haens GR, Rutgeerts P, Ceuppens JL. IL-15 is highly expressed in inflammatory bowel disease and regulates local T cell dependent cytokine production. J Immunol 2000;164: Inoue S, Matsumoto T, Iida M, Mizuno M, Kuroki F, Hoshika K, Shimizu M. Characterization of cytokine expression in the rectal mucosa of ulcerative colitis: correlation with disease activity. Am J Gastroenterol 1999;94: Bulfone-Paus S, Ungureanu D, Pohl T, Lindner G, Paus R, Ruckert R, Krause H, Kunzenforf U. Interleukin-15 protects from lethal apoptosis in vivo. Nat Med 1997;3: Received September 24, Accepted May 16, Address requests for reprints to: Marco Londei, M.D., Kennedy Institute of Rheumatology, 1 Aspenlea Road, Hammersmith, London W6 8LH. m.londei@ic.ac.uk; fax: (39) Supported by Telethon-Italy (Grant no E762), The Wellcome Trust (United Kingdom), Associazione Italiana Celiachia (Italy), and European Community grants QLK1-CT and BMH The Kennedy Institute of Rheumatology is supported by the Arthritis Research Campaign (ARC). The authors thank Dr. A. Picarelli (Cattedra di Gastroenterologia, Clinica Medica II, Universitá La Sapienza, Rome, Italy) for providing some in vitro cultured bioptic samples; and Dr. G. De Marco, Dr. S. Coletta, and L. Vacca (Department of Pediatrics, University Federico II of Naples) for technical support.