Advertisement

Gluten induces an intestinal cytokine response strongly dominated by interferon gamma in patients with celiac disease

      Abstract

      Background & Aims: Celiac disease appears to be a T cell–mediated enteropathy induced by gluten in genetically predisposed individuals. Duodenal biopsy specimens from patients with celiac disease and histologically normal controls were investigated to see if cytokine expression is related to disease activity. Methods: Cytokine messenger RNA (mRNA) expression was determined by quantitative reverse-transcription polymerase chain reaction and in situ expression by immunohistochemistry. Results: In normal controls, mRNA levels were usually below the quantitative limit, even after in vitro gluten stimulation. By contrast, interferon (IFN)-γ mRNA was increased more than 1000-fold in untreated disease. In vitro gluten stimulation of specimens from treated patients (gluten-free diet) increased IFN-γ mRNA to the levels of untreated patients. In addition, increased mRNA levels for interleukin (IL)-2, IL-4, IL-6, and tumor necrosis factor α were found after such stimulation, whereas mRNA for IL-5, IL-10, and IL-12p40 was usually below the quantitative level. Biopsy specimens from untreated patients contained on average 10-fold more lamina propria cells positive for IFN-γ than normal controls, whereas cells containing IL-4 were rare in both subject groups. Conclusions: The results show that mucosal gluten exposure in patients with celiac disease rapidly elicits high levels of IFN-γ expression and lower levels of IL-2, IL-4, IL-6, and tumor necrosis factor α even in the virtual absence of IL-12.
      GASTROENTEROLOGY 1998;115:551-563

      Abbreviations:

      ELISA (enzyme-linked immunosorbent assay), IEL (intestinal epithelial cell), IFN (interferon), IL (interleukin), MAb (monoclonal antibody), RT-PCR (reverse-transcription polymerase chain reaction), TCC (T-cell clone), TCR (T-cell receptor), TGF (transforming growth factor), Th (T-helper), TNF (tumor necrosis factor)
      Celiac disease, or gluten-sensitive enteropathy, appears to be immune mediated and is precipitated by wheat gluten and related cereal proteins in the proximal small intestine of genetically susceptible individuals. The result is malabsorption and often diarrhea. The disease is characterized by crypt cell hyperplasia and villous atrophy,
      • Trier JS
      Celiac sprue.
      • Marsh MN
      Gluten, major histocompatibility complex, and the small intestine.
      in addition to increased epithelial expression of HLA class II molecules
      • Scott H
      • Sollid LM
      • Fausa O
      • Brandtzaeg P
      • Thorsby E
      Expression of major histocompatibility complex class II subregion products by jejunal epithelium in patients with coeliac disease.
      and transmembrane secretory component, the polymeric immunoglobulin receptor.
      • Scott H
      • Brandtzaeg P
      • Solheim BG
      • Thorsby E
      Relation between HLA-DR–like antigens and secretory component (SC) in jejunal epithelium of patients with coeliac disease or dermatitis herpetiformis.
      There is also markedly increased density of intraepithelial lymphocytes (IELs).
      • Ferguson A
      • Murray D
      Quantitation of intraepithelial lymphocytes in human jejunum.
      • Spencer J
      • MacDonald T
      • Walker-Smith J
      • Ciclitira P
      • Isaacson P
      Changes in intra-epithelial lymphocyte subpopulations in coeliac disease and enteropathy associated T-cell lymphoma (malignant histiocytosis of the intestine).
      • Brandtzaeg P
      • Halstensen TS
      • Kett K
      • Krajci P
      • Kvale D
      • Rognum TO
      • Scott H
      • Sollid LM
      Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes.
      • Halstensen TS
      • Scott H
      • Brandtzaeg P
      Intraepithelial T cells of the TcRγ/δ+CD8− and Vδ1/Jδ1+ phenotypes are increased in coeliac disease.
      Withdrawal of gluten from the diet usually results in normalization of the intestinal morphology.
      Gluten-specific mucosal CD4+ T cells employing the α/β T-cell receptor (TCR) seem to be central in the immunopathology of celiac disease.
      • Halstensen TS
      • Brandtzaeg P
      Activated T lymphocytes in the celiac lesion: non-proliferative activation (CD25) of CD4+ α/β cells in the lamina propria but proliferation (Ki-67) of α/β and γ/δ cells in the epithelium.
      • 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.
      • Lundin KEA
      • Scott H
      • Hansen T
      • Paulsen G
      • Halstensen TS
      • Fausa O
      • Thorsby E
      • Sollid LM
      Gliadin-specific, HLA-DQ (α1*0501,β1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients.
      • Lundin KEA
      • Scott H
      • Fausa O
      • Thorsby E
      • Sollid LM
      T cells from the small intestinal mucosa of a DR4, DQ7/DR4, DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8.
      It has been suggested that the mucosal changes found in untreated patients or in treated patients after gluten challenge are caused by increased local production of certain cytokines,
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      namely, interferon (IFN)-γ, that alone or in combination with tumor necrosis factor (TNF)-α are cytotoxic to epithelial cells.
      • Deem RL
      • Shanahan F
      • Targan SR
      Triggered human mucosal T cells release tumour necrosis factor-alpha and interferon-gamma which kill human colonic epithelial cells.
      We have previously established gluten-specific HLA-DQ–restricted T-cell clones (TCCs) from the celiac lesion
      • Lundin KEA
      • Scott H
      • Hansen T
      • Paulsen G
      • Halstensen TS
      • Fausa O
      • Thorsby E
      • Sollid LM
      Gliadin-specific, HLA-DQ (α1*0501,β1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients.
      • Lundin KEA
      • Scott H
      • Fausa O
      • Thorsby E
      • Sollid LM
      T cells from the small intestinal mucosa of a DR4, DQ7/DR4, DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8.
      that show a T-helper (Th)1 or Th0 profile, IFN-γ being the predominant cytokine.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      In addition, most of these mucosal TCCs produce variable levels of TNF, transforming growth factor (TGF)-β, and interleukin (IL)-6, and some secrete IL-4, IL-5 and IL-10 as well.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      The possibility exists that isolation of TCC is biased, not truly reflecting the in situ T-cell profile.
      • Scott P
      Selective differentiation of CD4+ T helper cell subsets.
      • Kelso A
      Th1 and Th2 subsets: paradigms lost?.
      • Romagnani S
      Development of Th1- or Th2-dominated immune responses: what about the polarizing signals?.
      • Mosmann TR
      • Sad S
      The expanding universe of T-cell subsets: Th1, Th2 and more.
      Therefore, in the present study we used a highly sensitive quantitative reverse-transcription polymerase chain reaction (RT-PCR) method to evaluate cytokine expression patterns in intestinal biopsy specimens from patients with active (untreated) or inactive (treated with gluten-free diet) celiac disease. Mucosal gluten stimulation in vitro of specimens obtained from treated celiac lesions was also performed, and the expression of IFN-γ messenger RNA (mRNA) was shown to increase after 30 minutes, usually peaking after 2 hours at a level comparable to that obtained for untreated disease. In addition, low levels of IL-2, IL-4, IL-5, IL-6, and TNF-α mRNA were detected, whereas the macrophage cytokines IL-10 and IL-12p40 usually remained below the detection limit. These results were supported by immunohistochemistry showing a significantly increased number of IFN-γ–positive mucosal lamina propria cells compared with histologically normal controls.
      Our findings substantiate the notion that gluten induces rapid activation of mucosal Th1 or Th0 T cells in patients with susceptibility for celiac disease, thereby leading to local secretion of high levels of IFN-γ. This cytokine might directly or indirectly (e.g., via activation of macrophages) cause the mucosal alterations characteristic of untreated celiac disease.

      Materials and methods

       Patients

      The patients enrolled in the cytokine mRNA study were evaluated according to the European Society for Pediatric Gastroenterology and Nutrition criteria for celiac disease of 1990.
      • Walker-Smith JA
      • Guandalini S
      • Schmitz J
      • Shmerling DH
      • Visakorpi JK
      Revised criteria for diagnosis of coeliac disease.
      Biopsy specimens from the distal duodenum (n = 1–5 per subject) were obtained during upper gastrointestinal endoscopy (Olympus GIF IT20 endoscope; biopsy forceps, Olympus FB 13K; Tokyo, Japan) from 20 patients with celiac disease (median age, 38 years; range, 6–68 years), 7 with untreated disease and 13 on a gluten-free diet (Table 1). The histopathologic diagnosis was based on typical mucosal lesions with crypt hyperplasia, villous atrophy, and an increased number of IELs. H&E-stained sections from parallel formalin-fixed specimens were evaluated by light microscopy, and the ratio of TCRγ/δ+ IELs was determined in frozen duodenal tissue sections as previously described.
      • Halstensen TS
      • Scott H
      • Brandtzaeg P
      Intraepithelial T cells of the TcRγ/δ+CD8− and Vδ1/Jδ1+ phenotypes are increased in coeliac disease.
      An increased γ/δ cell percentage supported the celiac diagnosis. Some, but not all, patients showed normalization of the small intestinal morphology after a presumptive gluten-free diet (Table 1). Patients enrolled in the control group (n = 12; median age, 39 years; range, 14–74 years) were under investigation for gastrointestinal symptoms, but none had abnormal endoscopic or histological findings and had a much lower density of IELs than that found in patients with untreated celiac disease (Table 1). Moreover, all controls except 1 (control 9) had a normal ratio (≤10%) of intraepithelial γ/δ cells (Table 1).
      Table 1Clinicopathologic information on patients with celiac disease and histologically normal controls providing duodenal specimens for cytokine mRNA analyses
      Subject
      No.Age (y)Small intestinal morphologyGluten-free diet (time)IELs/100 epithelial cells% of TCRγ/δ+ IELsa
      Pt. 1937Subtotal villous atrophy04136
      Pt. 2227Subtotal villous atrophy3 mo4535
      Pt. 2345Subtotal villous atrophy08223
      Pt. 2848Subtotal villous atrophy03934
      Pt. 3329Total villous atrophy04711
      Pt. 346Subtotal villous atrophy05325
      Pt. 389Subtotal villous atrophy04643
      Pt. 1355Partial villous atrophy18 mo2630
      Pt. 1450Partial villous atrophy12 yr2652
      Pt. 1550Partial villous atrophy8 mo21ND
      Pt. 2114Minimal villous changes22 mo1428
      Pt. 2440Subtotal villous atrophy6 mo1718
      Pt. 2945Minimal villous changes4 yr100
      Pt. 3165Minimal villous changes6 mo1946
      Pt. 3518Minimal villous changes6 yr1051
      Pt. 3768Partial villous atrophy5 mo847
      Pt. 4043Normal14 mo102
      Pt. 4141Partial villous atrophy3 yr3839
      Pt. 4240Minimal villous changes3 yr2260
      Pt. 4329Partial villous atrophy12 mo4651
      Ctr. 137Normal16<1
      Ctr. 234Normal90.5
      Ctr. 342Normal81
      Ctr. 414Normal72.5
      Ctr. 530Normal80.5
      Ctr. 626NormalND1
      Ctr. 747Normal64
      Ctr. 836Normal72
      Ctr. 931Normal1228
      Ctr. 1074Normal50.5
      Ctr. 1163Normal810
      Ctr. 1218Normal131
      aThe percentage of γ/δ IELs in relation to total CD3+ IELs was determined by paired immunofluorescence staining in situ.
      Pt., patient; Ctr., control; ND, not determined.
      Immunohistochemical detection of cytokines to support the mRNA results was performed on biopsy specimens obtained from 6 patients with untreated celiac disease and 5 histologically normal controls. The criteria used for enrollment of these additional subjects were as described above.

       In vitro gluten stimulation of biopsy specimens

      One biopsy specimen from each subject was immediately snap-frozen in liquid nitrogen and stored at −70°C until use to reflect the in vivo situation. When more than 1 specimen was obtained, the remaining ones were placed on iron grids and challenged for different time intervals with 5 mg/mL of a peptic/tryptic
      • Frazer AC
      • Fletcher RF
      • Ross CAC
      • Shaw B
      • Sammons HG
      • Schneider R
      Gluten-induced enteropathy. The effects of partially digested gluten.
      digest of commercial gluten (Sigma Chemical Co., St. Louis, MO) in an organ culture chamber at 37°C with 5% CO2 and 95% O2 at 1-bar pressure in RPMI 1640 (GIBCO, Paisley, Scotland) containing 15% inactivated pooled normal human serum and antibiotics. The celiac specimens were usually stimulated for 2 or 6 hours with gluten. An additional specimen from 3 individuals (1 untreated and 2 treated) was stimulated in parallel with medium containing 15% normal human serum (without gluten) as control. Early and late cytokine expression was evaluated in parallel specimens from 3 patients after 30 minutes and 1 and 2 hours, and from 1 patient after 2, 6, 8, and 10 hours. Normal control specimens were stimulated for 2–10 hours. After such in vitro stimulation, the specimens were frozen in liquid nitrogen and stored at −70°C.
      Culture supernatants were collected after in vitro gluten stimulation of 4 celiac specimens for 18 hours as described above. The supernatants (2 mL) were stored at −70°C until used.

       RNA isolation and RT-PCR

      Biopsy specimens were homogenized with a glass Dounce homogenizer (Kontes, Vineland, NJ), and total RNA was isolated by the guanidium thiocyanate method
      • MacDonald RJ
      • Swift GH
      • Przybyla AE
      • Rutters WJ
      • Chirgwin JM
      Isolation of RNA using guanidium salts.
      and stored at −70°C in diethyl pyrocarbonate–treated water. The RNA was quantified by absorbance at 260 nm. Total RNA (500 ng) was reverse transcribed into complementary DNA (cDNA) according to a standard protocol (Perkin–Elmer, Branchburg, NJ) in a 20-μL reaction mixture containing 5 mmol/L MgCl2, 100 mmol/L Tris-HCl (pH 8.3), 500 mmol/L KCl, 1 mmol/L deoxynucleotide triphosphates (Perkin–Elmer), 20 U RNAsin (Promega Corp., Madison, WI), 50 U reverse transcriptase (Perkin–Elmer), and 20 pmol oligo(dT)16 primer. The reverse transcription reactions were performed for 60 minutes (42°C), followed by heat inactivation for 10 minutes (99°C), and the cDNA was stored at −20°C until use.
      PCR was performed on 2 μL cDNA and amplified by gene-specific primers as described previously.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      Amplification was performed for 25 cycles for β-actin and 35 cycles for the cytokines. The size of the PCR products, the primer set sequences, and the annealing temperatures for most primer sets have been reported previously.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      • Jung HC
      • Eckmann L
      • Yang SK
      • Panja A
      • Fierer J
      • Morzycka-Wroblewska E
      • Kagnoff MF
      A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion.
      In addition, primers for the TCRα chain were used (courtesy of Dr. M. F. Kagnoff, University of California, San Diego, CA). Because multiple bands were obtained with the IL-12p40 primers described by Jung et al.,
      • Jung HC
      • Eckmann L
      • Yang SK
      • Panja A
      • Fierer J
      • Morzycka-Wroblewska E
      • Kagnoff MF
      A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion.
      we included two additional primer sets for this cytokine; one of these (forward primer 5'-AGTGTCAAAAGCAGCAGAGG-3', reverse primer 5'-AACGCAGAATGTCAGGGAG-3') yielded a 363–base pair (bp) product, and the other set, which was in a nested position to the first one (forward primer 5'-TCAGAGGGGACAACA AGGAG-3', reverse primer 5'-AGGGAGAAGTAGGAATGTGGAG-3'), yielded a 274-bp product. Both primer sets had an annealing temperature at 60°C. All primer sequences were derived from separate exons, thus spanning one or more introns; amplification products of genomic DNA could therefore easily be distinguished from those of reverse-transcribed mRNA.

       RNA standards for quantitative PCR

      To quantify the amounts of different gene transcripts, we used standard mRNA generated from the plasmid pHCQ1 (courtesy of Dr. M. F. Kagnoff). The construction of this plasmid has been described.
      • Jung HC
      • Eckmann L
      • Yang SK
      • Panja A
      • Fierer J
      • Morzycka-Wroblewska E
      • Kagnoff MF
      A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion.
      To generate standard mRNA, the plasmid was linearized with HindIII and transcribed in vitro by use of the MEGAscript T7 kit under conditions recommended by the supplier (Ambion Inc., Intermedica, Stockholm, Sweden). The distance between specific 5' and 3' primer sequences and, therefore, the size of the PCR amplification products differed for standard and target RNAs.
      • Jung HC
      • Eckmann L
      • Yang SK
      • Panja A
      • Fierer J
      • Morzycka-Wroblewska E
      • Kagnoff MF
      A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion.

       Quantification of mRNA expression by competitive PCR

      For quantification of mRNA expression, a constant amount of total RNA (500 ng) from each mucosal specimen was reverse-transcribed into cDNA together with serial dilutions of standard RNA transcripts in the same reaction volume and coamplified by PCR in a total volume of 25 μL. After amplification, the PCR products generated by standard and target RNA were electrophoresed in 1.6% agarose gels and the bands were visualized by ethidium bromide staining. Photographs were taken on Polaroid 665 film, and the negatives were used to quantify band intensities by means of an image analyzer (Joyce-Loebl Ltd., Gateshead, England). With this approach, a point could be determined at which the starting number of standard RNA transcripts was equal to that of target RNA transcripts; the ratios of the band intensities of the standard RNA to target RNA were plotted against the starting number of standard RNA molecules on a double logarithmic scale. Although all transcripts could be detected and quantified below 103 transcripts per microgram of total RNA, this value was selected as a lower limit for the quantitative PCR analysis; biologically meaningful cytokine production is unlikely to take place below this limit.

       Immunoassays for secreted cytokines

      Commercial enzyme-linked immunosorbent assay (ELISA) systems were used according to the manufacturers' recommendations for quantification of human IFN-γ (Medgenix Diagnostics, Brussels, Belgium) and human IL-4 (Genzyme, Cambridge, MA) in culture supernatants.

       Bioassays for secreted cytokines

      IL-2 and IL-6 activities in culture supernatants were determined in proliferative bioassays by applying the murine cell line HT2 or B9 (clone B-92504), respectively. TNF (α and β) activities were measured in a cytotoxic bioassay with the murine cell line WEHI 164 clone 13. These methods and their specificities have been detailed previously.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.

       Cellular cytokine detection by immunohistochemistry

      The method was performed as described by Andersson et al.
      • Andersson J
      • Abrams J
      • Björk L
      • Funa K
      • Litton M
      • Ågren K
      Concomitant in vivo production of 19 different cytokines in human tonsils.
      Briefly, the biopsy specimens were immediately placed on a thin slice of carrot for appropriate orientation and handling, embedded in OCT (Tissue-Tek, Miles Laboratories, Elkhart, IN), snap-frozen in liquid nitrogen, and stored at −70°C. Cryosections cut serially at 8 μm were fixed in 4% paraformaldehyde (pH 7.4) for 15 minutes. Endogenous peroxidase was blocked by 1% H2O2/0.02% NaN3 in phosphate-buffered saline (PBS), supplemented with 0.1% saponin for 30 minutes. Thereafter the sections were incubated with monoclonal antibody (MAb) specific for human IFN-γ (7B6; purified mouse immunoglobulin G [IgG]1, 5 μg/mL; Mabtech AB, Nacka, Sweden) or MAb specific for human IL-4 (1-41-1; purified mouse IgG1, 5 μg/mL; courtesy Dr. F. Kalthoff, Sandoz Forschungsinstitut, Vienna, Austria) for 20 hours at 4°C. This incubation was followed by biotinylated horse anti-mouse IgG (5 μg/mL; Vector Laboratories, Burlingame, CA) for 3 hours at room temperature. The sections were finally incubated with horseradish peroxidase–conjugated streptavidin (Ventana DAB detection kit; Ventana Medical Systems, Inc., Tucson, AZ) followed by diaminobenzidine (Ventana) and counterstained with hematoxylin. All antibody reagents were appropriately diluted in PBS containing 0.1% saponin, and the incubations took place with intervening thorough rinses in Tris-buffered saline with 0.1% saponin (except the steps following horseradish peroxidase–conjugated streptavidin, from which saponin was omitted).
      Gluten-specific TCCs previously shown to produce high levels of both IFN-γ and IL-4
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      were used as positive controls. After stimulation with gluten peptides, as described previously,
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      and blockade of the secretory pathway by monensin,
      • Jung T
      • Schauer U
      • Heusser C
      • Neumann C
      • Rieger C
      Detection of intracellular cytokines by flow cytometry.
      paraformaldehyde-fixed cytospins showed characteristic perinuclear cytoplasmic staining
      • Andersson J
      • Abrams J
      • Björk L
      • Funa K
      • Litton M
      • Ågren K
      Concomitant in vivo production of 19 different cytokines in human tonsils.
      for both IFN-γ and IL-4 with the staining protocol described above.
      To control for specificity, immunohistochemistry was performed with two other MAbs, clone 1-D1K (purified mouse IgG1; Mabtech AB) and clone MP4-25D2 (purified rat IgG1; Pharmingen, San Diego, CA) directed against IFN-γ and IL-4, respectively, and a similar staining pattern was obtained. Furthermore, preincubation of 7B6 (5 μg/mL) with recombinant human IFN-γ (1000 U/mL; R&D Systems, Abingdon, England) for 30 minutes blocked completely the cellular reactivity of this MAb, except for tissue elements considered to be nonspecifically stained (see later). A murine MAb directed towards keyhole limpet hemocyanin (IgG1; Becton Dickinson, Mountain View, CA) applied in the same staining protocol at the usual working concentration also provided a negative control reference.

       Immunohistochemical evaluation

      IFN-γ– and IL-4–positive cells were counted systematically throughout one complete tissue section per subject with an ordinary light microscope at 600× magnification and expressed as cell number per mucosal unit area (square millimeters). All tissue sections were examined blindly by two investigators (E.M.N. and F.L.J.), and the coefficient of variation (16%) indicated satisfactory reproducibility. The Mann–Whitney test was performed to compare the results for biopsy specimens from active celiac disease with histologically normal controls. Wilcoxon matched pairs signed-rank sum test was used to compare IFN-γ– and IL-4–positive cells within the two subject groups. A difference was considered to be significant with P values of <0.05.

      Results

       Estimation of cytokine mRNA expression by RT-PCR

      Intestinal specimens from histologically normal controls, both unstimulated and stimulated in vitro with gluten, were either negative or showed only low expression of IFN-γ, IL-2, IL-4, IL-5, IL-6, and IL-12p40 mRNA. Moreover, we were unable to detect transcripts for IL-10 and TNF-α (Figure 1).
      Figure thumbnail gr1
      Fig. 1Agarose gel electrophoresis for semiquantitative determination of PCR-amplified cytokine mRNA in duodenal specimens from 8 control subjects with normal histology. Specimens from 2 subjects (controls [Ctr.] 1 and 2) were stimulated for 2, 4, and 6 hours with gluten or for 6 hours with medium alone (*) (control 1), and were compared with unstimulated counterparts (0). For the other subjects (controls 3–8), results are shown only for unstimulated specimens. Total RNA (500 ng) was reverse-transcribed to cDNA. Amplification was performed by gene-specific primers with 2 μL cDNA for 25 cycles (β-actin) or 35 cycles (cytokines). The indicated size (base pairs) of the amplification product matched that predicted from the position of the primer pairs.
      Conversely, all celiac disease specimens from both untreated and treated patients, as well as those stimulated in vitro with gluten, expressed mRNA for IFN-γ. In addition, several such specimens contained mRNA for IL-2, IL-4, IL-5, IL-6, and TNF-α after in vitro gluten stimulation (Figure 2).
      Figure thumbnail gr2
      Fig. 2Agarose gel electrophoresis for semiquantitative determination of PCR-amplified cytokine mRNA in duodenal specimens from 4 patients with celiac disease. Specimens taken from 1 untreated patient (Pt. 23) and 3 treated patients (Pt. 37, Pt. 40, and Pt. 41) were gluten-stimulated for different times (0.5–10 hours) and compared with unstimulated counterparts (0). Amplification and display as described in the legend to .
      All specimens (histologically normal controls as well as untreated and treated celiac disease) expressed mRNA for TGF-β both in the unstimulated and stimulated state (Fig. 1, Fig. 2).
      The most optimal time point for IFN-γ mRNA expression after in vitro gluten stimulation was found to be 2–6 hours. However, to control for early or late cytokine expression, specimens from 3 treated patients (patients 40, 42, and 43; Table 1) were stimulated with gluten for 30 minutes and for 1 and/or 2 hours, and from 1 patient (patient 41) for 2, 6, 8, and 10 hours. Weak expression of IFN-γ mRNA was observed as early as after 30 minutes, whereas the other cytokines remained at the basal level at this time point (Figure 2, and data not shown). The highest expression of IFN-γ was obtained after gluten stimulation for 2–6 hours for all patients tested, reaching a level comparable to that obtained for untreated lesions. In specimens stimulated for 2–10 hours, IFN-γ mRNA and IL-2 mRNA were elevated after 2 hours, whereas the TNF-α message did not increase until after 6 hours (Figure 2).

       Quantification of β-actin mRNA

      To evaluate differences in intestinal cytokine expression between untreated celiac disease, treated celiac disease (without and with in vitro gluten stimulation), and histologically normal controls, mRNA levels were assessed by quantitative RT-PCR with internal standards. Differences caused by varying amounts of total RNA in specimens of histologically normal or diseased intestinal mucosa were accounted for by quantification of β-actin mRNA. Samples taken from the same patient, representing different stimulation time points, contained approximately equal amounts of β-actin–specific transcripts (Tables 2 and 3). Thus, the marked differences observed for cytokine mRNA concentrations could not be ascribed to altered amounts of total RNA. However, a marked difference was observed for β-actin mRNA concentrations in specimens taken from different patients (range, 2 × 107 to 1.8 × 108 transcripts/μg total RNA).
      Table 2Cytokine, TCRα, and β-actin mRNA levels in duodenal specimens from patients with untreated celiac disease as assessed by quantitative PCR
      Patient
      No.Culture timeIFN-γ transcripts/μgIL-2 transcripts/μgIL-4 transcripts/μgTCRα transcripts/μgβ-Actin transcripts/μg
      190 h1.6 × 105<1031.8 × 1036.8 × 1069.6 × 107
      220 h2.4 × 105<103<1031.2 × 1061 × 108
      230 h1.2 × 106NegativeNegative4 × 107NDa
      232 hND<103<1036 × 1061.8 × 108
      236 h1.2 × 106<103<1035.2 × 1061.48 × 108
      236 ha1.2 × 106NegativeNegativeNDND
      280 h1.2 × 105<103<1032.8 × 1063.6 × 107
      282 h1 × 105NegativeNegative1.4 × 1065.6 × 107
      286 h1.2 × 105NDNDND5.6 × 107
      330 h4.8 × 105NegativeNegative2 × 1061.12 × 108
      332 h1 × 105<103<1031.8 × 1061 × 108
      340 h6 × 104<103<1031.2 × 1063.2 × 107
      380 h1.6 × 105NegativeNegative1.4 × 1068.8 × 107
      aThis biopsy specimen was stimulated with medium containing 15% normal human serum (without gluten) as control.
      ND, not determined.
      Table 3Cytokine, TCRα, and β-actin mRNA levels in duodenal specimens from patients with treated celiac disease as assessed by quantitative PCR
      Patient
      No.Culture timeIFN-γ transcripts/μgIL-2 transcripts/μgIL-4 transcripts/μgTCRα transcripts/μgβ-Actin transcripts/μg
      150 h1.2 × 104NegativeNDND9.2 × 107
      152 hNDNDNDNDND
      156 h1.6 × 105<103NDNDND
      210 h6 × 104NegativeNegative1 × 106ND
      212 h3.2 × 105<103<1031 × 106ND
      216 h6 × 104<103<103NDND
      216 ha6 × 104NegativeNegativeNDND
      290 h1.8 × 105Negative1 × 1077.2 × 107
      292 h3.6 × 1051.4 × 104<1032 × 1064 × 107
      296 h1.4 × 1053.6 × 103<1036 × 1056.8 × 107
      310 h1 × 105NegativeNegative2.04 × 1071.12 × 108
      312 h1 × 1062.8 × 1051.8 × 1032.48 × 1071.4 × 108
      370 h8 × 103<103<1036 × 1052 × 107
      372 h3 × 1047.2 × 103<1032.6 × 1052.4 × 107
      374 hNDND<103NDND
      376 h6 × 1041.2 × 104<1033.2 × 1056 × 107
      410 h1.5 × 104<103<1034 × 1053.2 × 107
      412 h2 × 1047.4 × 103<1033.2 × 1056.4 × 107
      416 h7.6 × 1054 × 1057.4 × 1032 × 1057.2 × 107
      aThis biopsy specimen was stimulated with medium containing 15% normal human serum (without gluten) as control.
      ND, not determined.

       Quantification of TCRα chain mRNA

      To account for cytokine differences caused by a varying number of T cells in the specimens derived from histologically normal and diseased intestinal mucosa, T cell–derived mRNA was quantified by the use of TCRα chain–specific primers. Samples taken from the same patient, representing different stimulation time points, contained approximately equal amounts of such transcripts (Table 2, Table 3). Thus, the marked differences observed for cytokine mRNA concentrations could not be ascribed to an altered T-cell number. A marked difference was observed for TCRα chain–specific transcripts in specimens taken from different patients (range, 2.6 × 105 to 4 × 107 transcripts/μg total RNA). However, this variation could to some extent be explained by the fact that each T cell can functionally rearrange two TCRα genes.

       Quantification of IFN-γ mRNA

      Histologically normal intestinal controls showed little expression of IFN-γ mRNA. When such specimens were stimulated in vitro with gluten for 2–10 hours, only a minimal increase of transcripts was observed (Figure 1). Twelve histologically normal controls (both unstimulated and stimulated) had transcript levels below the limit defined for quantitative evaluation (≤103 transcripts/μg total RNA).
      By contrast, the expression of IFN-γ mRNA was remarkably increased in 7 untreated celiac disease specimens subjected to quantification, containing between 6 × 104 and 1.2 × 106 transcripts/μg total RNA (Table 2). Specimens from 2 untreated patients (patients 23 and 28; Table 1) stimulated in vitro with gluten had unchanged IFN-γ mRNA levels; this was also the case for 3 specimens (1 untreated and 2 treated) incubated for 6 hours in medium containing 15% normal human serum alone (Table 2, Table 3, and data not shown).
      Gluten stimulation in vitro of mucosal specimens from 12 of 13 treated celiac patients produced increased IFN-γ mRNA expression, approaching the level for untreated cases; 1 specimen (patient 35) remained negative for all cytokines when tested after 2 and 6 hours with gluten stimulation (data not shown). This patient had been on a gluten-free diet for several years, which might account for the lacking response (absence of gluten-specific mucosal T cells?). Specimens from 6 representative treated patients were subjected to IFN-γ mRNA quantification before and after in vitro gluten stimulation; they showed basal levels of 8 × 103 to 1 × 105 transcripts/μg total RNA that increased to 6 × 104 to 1 × 106 transcripts/μg total RNA after 2-hour exposure (Table 3). An increased level was obtained as early as after 30 minutes (Figure 2), whereas the most optimal time point of IFN-γ mRNA expression was reached after 2–6 hours with gluten stimulation.

       Quantification of IL-2 and IL-4 mRNA

      All histologically normal intestinal controls (both unstimulated and gluten-stimulated) expressed IL-2 and IL-4 mRNA below the limit set for quantitation (data not shown). In specimens from untreated celiac patients, it was not possible to quantify the expression of these two cytokines (Table 2), as was also the case in specimens from treated patients (Table 3). However, in the latter category, IL-2 mRNA increased to 7.2 × 103 to 2.8 × 105 transcripts/μg total RNA after 2 hours, and to 3.6 × 103 to 5 × 105 transcripts/μg total RNA after 6 hours with gluten stimulation in vitro (Table 3). Specimens from treated as well as untreated patients showed slightly increased IL-4 mRNA expression after in vitro gluten stimulation for more than 2 hours (Table 2, Table 3), the levels corresponding to 1.8 × 103 to 8.4 × 103 transcripts/μg total RNA.

       Quantification of IL-6, TNF-α, and TGF-β mRNA

      All histologically normal intestinal controls (both unstimulated and gluten-stimulated) expressed IL-6 and TNF-α mRNA below the level defined for quantitative evaluation (data not shown). In specimens from untreated celiac patients, it was not possible to quantify the expression of the two cytokines, as was also the case in specimens from treated patients (Table 4). However, in the latter group, IL-6 mRNA increased to 2.7 × 103 to 1.24 × 105 transcripts/μg total RNA after 6 hours and TNF-α mRNA increased to 3.6 × 103 to 7.2 × 104 transcripts/μg total RNA after 2 and 6 hours of gluten stimulation in vitro (Table 4). Expression of TGF-β was almost unchanged in all specimens tested (without and with in vitro gluten stimulation), although some variations were obtained between different patients. The specimens contained 8 × 104 to 2 × 106 TGF-β transcripts/μg total RNA (Table 4), as was also the case for specimens obtained from histologically normal controls (data not shown).
      Table 4Cytokine mRNA levels in duodenal specimens from patients with celiac disease as assessed by quantitative PCR
      Patient
      No.Culture timeIL-6 transcripts/μgTNF-α transcripts/μgTGF-β transcripts/μg
      290 hNegativeNegative4 × 105
      292 hNegative3.6 × 1038 × 104
      296 h6.2 × 104<1035 × 105
      310 hNegativeNegative8 × 105
      312 h1.24 × 1051.5 × 1042 × 106
      370 hNegativeNegative1 × 105
      372 h<103Negative1.24 × 105
      376 h<103Negative2.72 × 105
      410 hNegativeNegative1.24 × 105
      412 h<103Negative8.8 × 104
      416 h2.7 × 1037.2 × 1043.12 × 105
      190 hNegativeNegative1.28 × 105
      340 hNegativeNegative1.2 × 105

       Quantification of IL-5 and IL-10 mRNA

      Intestinal expression of both IL-5 and IL-10 mRNA was either negative or below the level defined for quantitative evaluation in all specimens (histologically normal controls as well as untreated and treated celiac disease), both with and without in vitro gluten stimulation (data not shown).

       Detection of IL-12p40 mRNA

      Intestinal expression of IL-12p40 mRNA was either negative or gave multiple bands in all specimens (histologically normal controls as well as treated and untreated celiac disease), both with and without in vitro gluten stimulation (data not shown). We were unable to detect a single band with correct base pair size even with the use of three different primer sets, in contrast to the positive control result obtained for peripheral blood mononuclear cells stimulated with phytohemagglutinin and pokeweed mitogen (data not shown). However, nested PCR performed on the 363-bp PCR product (1 μL, diluted 1:10, 20 cycles) gave a single band of 274 bp; sequencing of this product confirmed that it was IL-12p40 (data not shown). Both histologically normal controls and celiac disease specimens expressed mRNA for IL-12p40 detectable by this method (Figure 3).
      Figure thumbnail gr3
      Fig. 3Agarose gel electrophoresis for semiquantitative determination of PCR-amplified IL-12p40 mRNA in duodenal specimens taken from both histologically normal controls and patients with celiac disease. Specimens were stimulated with gluten for different times as indicated and compared with unstimulated counterparts (0). In addition, 2 specimens were stimulated for 6 hours with medium alone (6*). Total RNA (500 ng) was reverse-transcribed to cDNA. Amplification was performed on 2 μL cDNA for 35 cycles followed by nested PCR on 1 μL of the product (diluted 1:10, 20 cycles). The 274-bp amplification product matched that predicted from the position of the primer pairs. A 123-bp ladder was used as molecular-weight marker, and β-actin served as sample control.

       Mucosal secretion of cytokines

      ELISA for IFN-γ and IL-4, in addition to bioassays for IL-2, IL-6, and TNF, were used to quantify cytokines in supernatants from gluten-stimulated intestinal biopsy specimens from 23 patients with celiac disease (both untreated and treated). Except for IL-6, only low levels were detected (data not shown).

       Immunohistochemistry for cellular IFN-γ and IL-4

      Cells deemed by control stainings to be truly cytokine expressing were exclusively located in the lamina propria (Figure 4).
      Figure thumbnail gr4
      Fig. 4Immunohistochemical cytokine detection in duodenal specimen from a patient with untreated celiac disease and in cytospins from gluten-specific T-cell clones. (A and B) Perinuclear staining of IFN-γ in lamina propria cells (closed arrows) identifies mucosal expression of this cytokine, whereas a few other cells with a diffuse brown cytoplasm (open arrow) were deemed to be nonspecifically decorated. Note that cytokine-expressing cells were absent from the epithelium (EP) shown at the top. (C and D) Gluten-specific T-cell clones used as positive controls. After stimulation with gluten peptides and blockade of the secretory pathway by monensin, cytospins showed characteristic perinuclear staining both for (C) IFN-γ and (D) IL-4 (immunoperoxidase staining lightly counterstained with hematoxylin; original magnification 1000×).
      Only cells showing a characteristic paranuclear expression pattern were included in the counts (Figure 5).
      Figure thumbnail gr5
      Fig. 5Scatter diagram comparing the density of lamina propria cells positive for IFN-γ or IL-4 in duodenal specimens from histologically normal controls (Ctr.) and patients with untreated celiac disease (Pt.).
      Biopsy specimens from untreated celiac disease patients contained significantly more IFN-γ– than IL-4–positive cells (P = 0.03), and specimens from untreated patients contained a significantly larger (P = 0.006) number of IFN-γ–positive cells than histologically normal controls (20–84 vs. 2–11/mm2). Conversely, there was no difference (P = 0.08) in the number of IL-4–positive cells (1–10 vs. 0–4/mm2). Scattered undefined cells with diffuse cytoplasmic staining were excluded because they were seen both in diseased specimens and controls and occurred also after incubation with irrelevant MAb as well as after absorption control with IFN-γ. The same was true for weak apical epithelial staining seen in some tissue specimens.

      Discussion

      Quantification of mRNA levels by RT-PCR showed that the expression of IFN-γ was remarkably increased in the duodenal mucosa of patients with celiac disease after gluten exposure in vivo or in vitro compared with similarly exposed histologically normal control mucosa. Conversely, only low mucosal levels of IL-2, IL-4, IL-5, IL-6, IL-12p40, and TNF-α mRNA were detected even after gluten stimulation, whereas all specimens showed constitutive expression of TGF-β but were negative for IL-10. In addition, biopsy specimens taken from patients with untreated celiac disease contained a significantly increased number of IFN-γ–positive lamina propria cells compared with histologically normal controls, whereas IL-4–positive cells were rare in both categories of mucosal specimens. These results strongly supported the disease relevance of our previous data obtained with gluten-specific HLA-DQ–restricted TCCs isolated from celiac disease mucosa after in vitro gluten stimulation.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      Our findings also agreed well with an in situ hybridization study of IFN-γ mRNA performed on duodenal mucosa; numerous positive lamina propria cells (but no IELs) were seen in active celiac lesions but only a few scattered ones in normal mucosa.
      • Kontakou M
      • Sturgess RP
      • Przemioslo RT
      • Limb GA
      • Nelufer JM
      • Ciclitira PJ
      Detection of interferon gamma mRNA in the mucosa of patients with coeliac disease by in situ hybridisation.
      Human intestinal mucosa normally contains a large population of T lymphocytes
      • Brandtzaeg P
      • Halstensen TS
      • Kett K
      • Krajci P
      • Kvale D
      • Rognum TO
      • Scott H
      • Sollid LM
      Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes.
      ; CD4+ T cells dominate in the lamina propria, whereas CD8+ T cells localize preferentially (80%–90%) in the villous epithelium.
      • Brandtzaeg P
      • Halstensen TS
      • Kett K
      • Krajci P
      • Kvale D
      • Rognum TO
      • Scott H
      • Sollid LM
      Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes.
      When duodenal mucosa from treated celiac patients is stimulated with gluten in vitro, lamina propria CD4+TCRα/β+ T cells become preferentially activated and express CD25; also a dominating population of subepithelial pan–HLA class II+ macrophages (CD4+CD68+) with variable expression of CD25 often appears to be induced.
      • 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.
      IFN-γ, IL-2, IL-4, and IL-5 are known to be produced mainly by activated T cells, and the lamina propria CD4+ subset is the most likely source of these cytokines in celiac disease. However, the additional evidence for activated mucosal macrophages after gluten stimulation prompted us to test for cytokines such as IL-6, IL-10, IL-12, and TNF-α as well.
      A strict gluten-free diet usually reverses the histopathologic and clinical features of celiac disease. Therefore, inclusion of biopsy specimens from treated patients allowed us to study early immunopathologic events induced in the mucosa by exposure to gluten in vitro. Such stimulation for only 30 minutes produced increased expression of IFN-γ mRNA that persisted for up to 2–6 hours and then usually declined. IL-2 and IL-4 mRNA expression was also observed relatively early, whereas TNF-α mRNA did not appear until after 6–8 hours of gluten stimulation. Conversely, IL-10 mRNA remained negative for up to 10 hours of stimulation, which was the latest time point tested. These results supported the notion that the immunopathogenesis of celiac disease largely develops in the lamina propria
      • Halstensen TS
      • Scott H
      • Brandtzaeg P
      Intraepithelial T cells of the TcRγ/δ+CD8− and Vδ1/Jδ1+ phenotypes are increased in coeliac disease.
      • Halstensen TS
      • Brandtzaeg P
      Activated T lymphocytes in the celiac lesion: non-proliferative activation (CD25) of CD4+ α/β cells in the lamina propria but proliferation (Ki-67) of α/β and γ/δ cells in the epithelium.
      and suggested that gluten specifically induces a rapid increase in IFN-γ in addition to inducing low levels of IL-2 and IL-4 mRNA. Secondary to this event, macrophages might become activated and produce TNF-α and probably other factors.
      It cannot be excluded that additional immune cells besides T lymphocytes and macrophages might contribute to the Th1-like cytokine mRNA expression profile observed in celiac disease mucosa. In a study based on in vitro gluten stimulation of biopsy specimens from treated patients, Maiuri et al.
      • 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.
      proposed two parallel initiating pathways, one leading to T-cell and macrophage activation and an undefined preceding one reflected by increased epithelial HLA-DR expression observed as early as after 1–2 hours. This latter pathway was suggested to be a direct stimulatory effect of gluten or, alternatively, an effect of preformed cytokines released from cellular stores such as mast cells. However, our results regarding increased IFN-γ mRNA expression after gluten stimulation suggested that the HLA-DR expression might be explained by gluten-induced reactivation of Th1/Th0 memory cells. Another known source of IFN-γ is natural killer cells, but it is unlikely that gluten exerts a disease-specific effect on them. We found no IFN-γ or IL-4 expression in IELs by immunohistochemistry, and gluten-specific IEL responses, to our knowledge, have not been demonstrated. Nevertheless, it is quite possible that both lamina propria T cells and IELs can act as cytolytic effector cells when activated by proinflammatory cytokines and thereby contribute to tissue disruption in celiac disease.
      • Oberhuber G
      • Vogelsang H
      • Stolte M
      • Muthenhaler S
      • Kummer AJ
      • Radaszkiewicz T
      Evidence that intestinal intraepithelial lymphocytes are activated cytotoxic T cells in celiac disease but not in giardiasis.
      We found significant quantitative differences in mucosal IFN-γ mRNA levels between specimens derived from patients with treated and untreated celiac disease as well as between treated disease and histologically normal controls. The transcript levels were 10–100 times higher in untreated than in treated mucosal lesions, and more than 1000 times higher in untreated lesions than in histologically normal controls. Gluten stimulation in vitro of celiac specimens from patients treated with a gluten-free diet increased the expression of IFN-γ mRNA to the level in untreated lesions that contained 2.9 × 105 to 1.2 × 106 IFN-γ transcripts/μg total RNA (corresponding to 0.14–0.58 pg IFN-γ/μg total RNA). This level was comparable to, or even higher than, that reported for active Crohn's disease lesions.
      • Niessner M
      • Volk BA
      Altered Th1/Th2 cytokine profiles in the intestinal mucosa of patients with inflammatory bowel disease as assessed by quantitative reversed transcribed polymerase chain reaction (RT-PCR).
      Our results were supported by immunohistochemistry that showed significantly more IFN-γ–positive lamina propria cells in untreated celiac disease (20–84/mm2) than in histologically normal controls (2–11/mm2).
      Low levels of TNF-α mRNA were found in duodenal mucosa of some celiac patients. In combination with IFN-γ, this cytokine has been reported to be particularly cytotoxic to epithelial cells,
      • Deem RL
      • Shanahan F
      • Targan SR
      Triggered human mucosal T cells release tumour necrosis factor-alpha and interferon-gamma which kill human colonic epithelial cells.
      which might contribute to the immunopathology in celiac disease.
      • Lionetti P
      • Breese E
      • Braegger CP
      • Murch SH
      • Taylor J
      • MacDonald TT
      T-cell activation can induce either mucosal destruction or adaptation in cultured human fetal small intestine.
      • Przemioslo RT
      • Lundin KEA
      • Sollid LM
      • Nelufer J
      • Ciclitira PJ
      Histological changes in small bowel mucosa induced by gliadin sensitive T lymphocytes can be blocked by anti-interferon γ antibody.
      Interestingly, normal small intestinal biopsy specimens showed a significant reduction in enterocyte height when cultured for 24 hours in vitro with supernatants from our stimulated gluten-reactive TCCs, and antibodies to IFN-γ blocked this toxic effect.
      • Przemioslo RT
      • Lundin KEA
      • Sollid LM
      • Nelufer J
      • Ciclitira PJ
      Histological changes in small bowel mucosa induced by gliadin sensitive T lymphocytes can be blocked by anti-interferon γ antibody.
      IFN-γ is also an important macrophage activator.
      • Doherty TM
      T-cell regulation of macrophage function.
      One hypothesis is that activated macrophages secrete metalloproteinases that disintegrate the mucosal matrix, thereby causing crypt hyperplasia in celiac disease.
      • Pender SLF
      • Lionetti P
      • Murch SH
      • Wathan N
      • MacDonald TT
      Proteolytic degradation of intestinal mucosal extracellular matrix after lamina propria T cell activation.
      All the histologically normal control specimens contained IFN-γ mRNA below the level defined for quantitative evaluation (≤103 transcripts/μg total RNA), whether unstimulated or stimulated with gluten for up to 10 hours.
      Increased expression of IL-2 mRNA was detected after gluten stimulation of specimens taken from treated celiac patients, compared with very low transcript levels found in unstimulated counterparts and histologically normal controls (unstimulated or gluten-stimulated). These findings agreed with our data from gluten-reactive TCCs, which were found to express IL-2 mRNA and produced small amounts of IL-2 protein only after specific stimulation.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      Some expression of IL-4 mRNA was likewise detected in celiac specimens after gluten stimulation for more than 2 hours. In addition, specimens from patients with untreated celiac disease tended to contain somewhat more IL-4–positive cells than histologically normal controls, but the number was significantly smaller than that for IFN-γ–positive cells (1–10 vs. 20–84/mm2). Altogether, these findings were again in keeping with our TCC results.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      The reason why IL-2 and IL-4 mRNA expression was seen only after in vitro stimulation remains unclear. However, in vivo gluten peptides are likely to penetrate the mucosa in waves (after meals) providing temporary T-cell activation, whereas the in vitro stimulation was continuous at a high level. Notably, most duodenal biopsy specimens were obtained in the morning after a fast of at least 8 hours. Thus, the results induced in vivo and in vitro might not be contradictory but rather reflect different phases of mucosal immune responsiveness to gluten. Also, others have found only minute amounts of IL-2 and IL-4 in human intestinal mucosa.
      • Breese E
      • Braegger CP
      • Corrigan CJ
      • Walker-Smith JA
      • MacDonald TT
      Interleukin-2 and interferon-γ secreting T cells in normal and diseased human intestinal mucosa.
      • McCabe RP
      • Secrist H
      • Botney M
      • Egan M
      • Peters M
      Cytokine mRNA expression in intestine from normal and inflammatory bowel disease patients.
      • Fuss IJ
      • Neurath M
      • Boirivant M
      • Klein JS
      • Motte Cdl
      • Strong SA
      • Fiocchi C
      • Strober W
      Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease.
      Although the increased epithelial expression of the secretory component found in the untreated celiac lesion appears to be explained mainly by the high levels of IFN-γ, an additional role of IL-4 and TNF-α is likely.
      • Sollid LM
      • Kvale D
      • Brandtzaeg P
      • Markussen G
      • Thorsby E
      Interferon-γ enhances expression of secretory component, the epithelial receptor for polymeric immunoglobulins.
      • Kvale D
      • Brandtzaeg P
      • Løvhaug D
      Up-regulation of the expression of secretory component and HLA molecules in a human colonic cell line by tumour necrosis factor–α and gamma interferon.
      • Phillips JO
      • Everson MP
      • Moldoveanu Z
      • Lue C
      • Mestecky J
      Synergistic effect of IL-4 and IFN-γ on the expression of polymeric Ig receptor (secretory component) and IgA binding by human epithelial cells.
      • Krajci P
      • Taskén K
      • Kvale D
      • Brandtzaeg P
      Interferon-γ stimulation of messenger RNA for human secretory component (poly-Ig receptor) depends on continuous intermediate protein synthesis.
      Because the transmembrane secretory component functions as polymeric Ig receptor, its enhanced expression promotes the external transport of dimeric IgA and pentameric IgM antibodies, and this is known to be the case in active celiac disease.
      • Brandtzaeg P
      • Halstensen TS
      • Kett K
      • Krajci P
      • Kvale D
      • Rognum TO
      • Scott H
      • Sollid LM
      Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes.
      IL-10 is a cytokine with anti-inflammatory properties, mainly functioning to inhibit Th1-induced macrophage production of proinflammatory cytokines such as IL-1β, IL-6, and TNF-α.
      • Fiorentino DF
      • Zlotnik A
      • Vieira P
      • Mosmann TR
      • Howard M
      • Moore KW
      • O'Garra A
      IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Th1 cells.
      Because only low expression of these cytokines were found in untreated celiac lesions, we had expected to find increased IL-10 levels as an important down-regulatory factor in the mucosa. However, no IL-10 mRNA expression was observed after gluten challenge of celiac specimens for up to 10 hours, although we could not exclude a later appearance.
      In agreement with our data for gluten-reactive TCCs,
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      constitutive expression of TGF-β was found in specimens from all celiac patients, but it was also present in histologically normal controls. Particularly in cooperation with IL-10, this cytokine has been identified in vitro as important for B-cell expansion and preferential IgA differentiation, perhaps as a consequence of class switching.
      • Defrance T
      • Vanbervliet B
      • Briére F
      • Durand I
      • Rousset F
      • Banchereau J
      Interleukin 10 and transforming growth factor β cooperate to induce anti-CD40–activated naive human B cells to secrete immunoglobulin A.
      At the same time, it is believed that T cells producing TGF-β are involved in oral tolerance.
      • Weiner HL
      Oral tolerance.
      Polarized differentiation of Th cells toward a Th1 or Th2 cytokine profile is influenced by many factors, including the nature and concentration of stimulatory antigen, the type of the antigen-presenting cells, and the local cytokine milieu.
      • Romagnani S
      Development of Th1- or Th2-dominated immune responses: what about the polarizing signals?.
      • Mosmann TR
      • Sad S
      The expanding universe of T-cell subsets: Th1, Th2 and more.
      • Seder RA
      • Paul WE
      Acquisition of lymphokine-producing phenotype by CD4+ T cells.
      IL-12 is believed to promote differentiation of Th1 cells, mainly through its ability to prime T cells for high IFN-γ production.
      • Manetti R
      • Parronchi P
      • Giudizi MG
      • Piccinni MP
      • Maggi E
      • Trinchieri G
      • Romagnani S
      Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper type 1 (Th1)-specific immune responses and inhibits the development of IL-4–producing Th cells.
      • Macatonia SE
      • Hosken NA
      • Litton M
      • Vieira P
      • Hsieh CS
      • Culpepper JA
      • Wysocka M
      • Trinchieri G
      • Murphy KM
      • O'Garra A
      Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells.
      • Kato T
      • Hakamada R
      • Yamane H
      • Nariuchi H
      Induction of IL-12 p40 messenger RNA expression and IL-12 production of macrophages via CD40-CD40 ligand interaction.
      • Gerosa F
      • Paganin C
      • Peritt D
      • Paiola F
      • Scupoli MT
      • Aste-Amezaga M
      • Frank I
      • Trinchieri G
      Interleukin-12 primes human CD4 and CD8 T cell clones for high production of both interferon-γ and interleukin-10.
      IL-12 is produced by macrophages and professional antigen-presenting cells such as interdigitating dendritic cells and Langerhans' cells.
      • Macatonia SE
      • Hosken NA
      • Litton M
      • Vieira P
      • Hsieh CS
      • Culpepper JA
      • Wysocka M
      • Trinchieri G
      • Murphy KM
      • O'Garra A
      Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells.
      • Kobayashi M
      • Fitz L
      • Ryan M
      • Hewick RM
      • Clark SC
      • Chan S
      • Loudon R
      • Sherman F
      • Perussia B
      • Trinchieri G
      Identification and purification of natural killer cell stimulatory factor (NKSF), a cytokine with multiple biologic effects on human lymphocytes.
      • Trinchieri G
      Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive immunity.
      Accumulating evidence from in vivo and in vitro experimental test systems supports the notion that IL-12 in fact is required for optimal in vivo Th1 responses.
      • Manetti R
      • Parronchi P
      • Giudizi MG
      • Piccinni MP
      • Maggi E
      • Trinchieri G
      • Romagnani S
      Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper type 1 (Th1)-specific immune responses and inhibits the development of IL-4–producing Th cells.
      • Trinchieri G
      Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive immunity.
      • Hsieh CS
      • Macatonia SE
      • Tripp CS
      • Wolf SF
      • O'Garra A
      • Murphy KM
      Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages.
      Interestingly, we detected only minute amounts of IL-12p40 mRNA expression, and no difference was seen between specimens from celiac disease patients (untreated or treated) and normal controls. Therefore, our data might reflect the memory nature of the studied mucosal response to gluten peptides. Animal experiments have suggested that priming of T cells with a soluble protein in the presence of IL-12 renders their capacity to secrete IFN-γ independently of IL-12 on subsequent antigen stimulation.
      • Seder RA
      • Kelsall BL
      • Jankovic D
      Differential roles for IL-12 in the maintenance of immune responses in infectious versus autoimmune disease.
      This could indeed be the case in celiac disease in which the initiating event likely is associated with some sort of intestinal infection inducing bystander release of IL-12.
      • Scott H
      • Nilsen EM
      • Sollid LM
      • Lundin KEA
      • Rugtveit J
      • Molberg Ø
      • Thorsby E
      • Brandtzaeg P
      Immunopathology of gluten-sensitive enteropathy.
      We were unable to detect cytokines in culture supernatants by ELISA and bioassay after gluten stimulation of celiac mucosal specimens in vitro. Four specimens with a size corresponding to 25–35 mg wet wt each were stimulated in 2 mL of medium, which probably resulted in a too-high dilution of tissue-derived mediators for our detection methods. Moreover, cytokines exert autocrine and paracrine effects and may hence become bound to receptors near the production site. Although mRNA levels might be misleading in terms of actual cytokine production because some cytokines are posttranscriptionally regulated, our immunohistochemical findings did show increased mucosal production of IFN-γ, particularly in untreated celiac disease, thus supporting our RT-PCR data. Collectively, our findings accorded well with what we have previously reported for gluten-specific HLA-DQ–restricted TCCs isolated from celiac mucosa. Those TCCs were shown to produce cytokines with a Th1-like or Th0-like profile, IFN-γ being the predominant product.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      In addition, most of the clones produced variable levels of TNF, TGF-β, and IL-6, and some secreted IL-4, IL-5, and IL-10 as well.
      Together our results strongly support the notion that gluten-induced activation of lamina propria CD4+ T cells, followed by secretion of high levels of IFN-γ, is an early pathogenic event in celiac disease.
      • Nilsen EM
      • Lundin KEA
      • Krajci P
      • Scott H
      • Sollid LM
      • Brandtzaeg P
      Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
      This cytokine alone or together with other mediators may directly or indirectly damage enterocytes or alter their maturation.
      • Scott H
      • Nilsen EM
      • Sollid LM
      • Lundin KEA
      • Rugtveit J
      • Molberg Ø
      • Thorsby E
      • Brandtzaeg P
      Immunopathology of gluten-sensitive enteropathy.
      In addition, IFN-γ may cause activation of macrophages that produce cytokines such as IL-10, IL-12, and TNF-α, as well as metalloproteinases that can disintegrate the mucosal matrix.
      • Pender SLF
      • Lionetti P
      • Murch SH
      • Wathan N
      • MacDonald TT
      Proteolytic degradation of intestinal mucosal extracellular matrix after lamina propria T cell activation.
      An unanswered crucial question that remains to be solved, however, is whether the human gut mucosa is at all able to respond with a Th2-like profile, e.g., in individuals with atopic allergy. Perhaps a Th1-like profile with predominantly high levels of IFN-γ results from all types of intestinal immune responses, as recently observed by us after intestinal astrovirus infection
      • Molberg Ø
      • Nilsen EM
      • Sollid LM
      • Scott H
      • Brandtzaeg P
      • Thorsby E
      • Lundin KEA
      CD4+ T-cells with specific reactivity against astrovirus isolated from normal human small intestine.
      and by others in cow's milk–sensitive enteropathy.
      • Hauer AC
      • Breese EJ
      • Walker-Smith JA
      • MacDonald TT
      The frequency of cells secreting interferon-γ and interleukin-4, -5, and -10 in the blood and duodenal mucosa of children with cow's milk hypersensitivity.

      Acknowledgements

      The authors thank Aaste Aursjø, Drude Hansson, and Tone Narvesen for excellent technical assistance; Dr. Ø. Molberg for providing T-cell clones; and Dr. M.F. Kagnoff for the generous gift of plasmids to generate standard mRNA. Mabtech AB is acknowledged for covering the costs of the color plate.

      References

        • Trier JS
        Celiac sprue.
        N Engl J Med. 1991; 325: 1709-1719
        • Marsh MN
        Gluten, major histocompatibility complex, and the small intestine.
        Gastroenterology. 1992; 102: 330-354
        • Scott H
        • Sollid LM
        • Fausa O
        • Brandtzaeg P
        • Thorsby E
        Expression of major histocompatibility complex class II subregion products by jejunal epithelium in patients with coeliac disease.
        Scand J Immunol. 1987; 26: 563-571
        • Scott H
        • Brandtzaeg P
        • Solheim BG
        • Thorsby E
        Relation between HLA-DR–like antigens and secretory component (SC) in jejunal epithelium of patients with coeliac disease or dermatitis herpetiformis.
        Clin Exp Immunol. 1981; 44: 233-238
        • Ferguson A
        • Murray D
        Quantitation of intraepithelial lymphocytes in human jejunum.
        Gut. 1971; 12: 988-994
        • Spencer J
        • MacDonald T
        • Walker-Smith J
        • Ciclitira P
        • Isaacson P
        Changes in intra-epithelial lymphocyte subpopulations in coeliac disease and enteropathy associated T-cell lymphoma (malignant histiocytosis of the intestine).
        Gut. 1989; 30: 339-346
        • Brandtzaeg P
        • Halstensen TS
        • Kett K
        • Krajci P
        • Kvale D
        • Rognum TO
        • Scott H
        • Sollid LM
        Immunobiology and immunopathology of human gut mucosa: humoral immunity and intraepithelial lymphocytes.
        Gastroenterology. 1989; 97: 1562-1584
        • Halstensen TS
        • Scott H
        • Brandtzaeg P
        Intraepithelial T cells of the TcRγ/δ+CD8− and Vδ1/Jδ1+ phenotypes are increased in coeliac disease.
        Scand J Immunol. 1989; 30: 665-672
        • Halstensen TS
        • Brandtzaeg P
        Activated T lymphocytes in the celiac lesion: non-proliferative activation (CD25) of CD4+ α/β cells in the lamina propria but proliferation (Ki-67) of α/β and γ/δ cells in the epithelium.
        Eur J Immunol. 1993; 23: 505-510
        • 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: 581-590
        • Lundin KEA
        • Scott H
        • Hansen T
        • Paulsen G
        • Halstensen TS
        • Fausa O
        • Thorsby E
        • Sollid LM
        Gliadin-specific, HLA-DQ (α1*0501,β1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients.
        J Exp Med. 1993; 178: 187-196
        • Lundin KEA
        • Scott H
        • Fausa O
        • Thorsby E
        • Sollid LM
        T cells from the small intestinal mucosa of a DR4, DQ7/DR4, DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8.
        Hum Immunol. 1994; 41: 285-291
        • Nilsen EM
        • Lundin KEA
        • Krajci P
        • Scott H
        • Sollid LM
        • Brandtzaeg P
        Gluten specific, HLA-DQ restricted T cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by interferon γ.
        Gut. 1995; 37: 766-776
        • Deem RL
        • Shanahan F
        • Targan SR
        Triggered human mucosal T cells release tumour necrosis factor-alpha and interferon-gamma which kill human colonic epithelial cells.
        Clin Exp Immunol. 1991; 83: 79-84
        • Scott P
        Selective differentiation of CD4+ T helper cell subsets.
        Curr Opin Immunol. 1993; 5: 391-397
        • Kelso A
        Th1 and Th2 subsets: paradigms lost?.
        Immunol Today. 1995; 16: 374-379
        • Romagnani S
        Development of Th1- or Th2-dominated immune responses: what about the polarizing signals?.
        Int J Clin Lab Res. 1996; 26: 83-98
        • Mosmann TR
        • Sad S
        The expanding universe of T-cell subsets: Th1, Th2 and more.
        Immunol Today. 1996; 17: 138-146
        • Walker-Smith JA
        • Guandalini S
        • Schmitz J
        • Shmerling DH
        • Visakorpi JK
        Revised criteria for diagnosis of coeliac disease.
        Arch Dis Child. 1990; 65: 909-911
        • Frazer AC
        • Fletcher RF
        • Ross CAC
        • Shaw B
        • Sammons HG
        • Schneider R
        Gluten-induced enteropathy. The effects of partially digested gluten.
        Lancet. 1959; 5: 252
        • MacDonald RJ
        • Swift GH
        • Przybyla AE
        • Rutters WJ
        • Chirgwin JM
        Isolation of RNA using guanidium salts.
        Methods Enzymol. 1979; 152: 219-234
        • Jung HC
        • Eckmann L
        • Yang SK
        • Panja A
        • Fierer J
        • Morzycka-Wroblewska E
        • Kagnoff MF
        A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion.
        J Clin Invest. 1995; 95: 55-65
        • Andersson J
        • Abrams J
        • Björk L
        • Funa K
        • Litton M
        • Ågren K
        Concomitant in vivo production of 19 different cytokines in human tonsils.
        Immunology. 1994; 83: 16-24
        • Jung T
        • Schauer U
        • Heusser C
        • Neumann C
        • Rieger C
        Detection of intracellular cytokines by flow cytometry.
        J Immunol Methods. 1993; 159: 197-207
        • Kontakou M
        • Sturgess RP
        • Przemioslo RT
        • Limb GA
        • Nelufer JM
        • Ciclitira PJ
        Detection of interferon gamma mRNA in the mucosa of patients with coeliac disease by in situ hybridisation.
        Gut. 1994; 35: 1037-1041
        • 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: 1368-1378
        • Oberhuber G
        • Vogelsang H
        • Stolte M
        • Muthenhaler S
        • Kummer AJ
        • Radaszkiewicz T
        Evidence that intestinal intraepithelial lymphocytes are activated cytotoxic T cells in celiac disease but not in giardiasis.
        Am J Pathol. 1996; 148: 1351-1357
        • Niessner M
        • Volk BA
        Altered Th1/Th2 cytokine profiles in the intestinal mucosa of patients with inflammatory bowel disease as assessed by quantitative reversed transcribed polymerase chain reaction (RT-PCR).
        Clin Exp Immunol. 1995; 101: 428-435
        • Lionetti P
        • Breese E
        • Braegger CP
        • Murch SH
        • Taylor J
        • MacDonald TT
        T-cell activation can induce either mucosal destruction or adaptation in cultured human fetal small intestine.
        Gastroenterology. 1993; 105: 373-381
        • Przemioslo RT
        • Lundin KEA
        • Sollid LM
        • Nelufer J
        • Ciclitira PJ
        Histological changes in small bowel mucosa induced by gliadin sensitive T lymphocytes can be blocked by anti-interferon γ antibody.
        Gut. 1995; 36: 874-879
        • Doherty TM
        T-cell regulation of macrophage function.
        Curr Opin Immunol. 1995; 7: 400-404
        • Pender SLF
        • Lionetti P
        • Murch SH
        • Wathan N
        • MacDonald TT
        Proteolytic degradation of intestinal mucosal extracellular matrix after lamina propria T cell activation.
        Gut. 1996; 39: 284-290
        • Breese E
        • Braegger CP
        • Corrigan CJ
        • Walker-Smith JA
        • MacDonald TT
        Interleukin-2 and interferon-γ secreting T cells in normal and diseased human intestinal mucosa.
        Immunology. 1993; 78: 127-131
        • McCabe RP
        • Secrist H
        • Botney M
        • Egan M
        • Peters M
        Cytokine mRNA expression in intestine from normal and inflammatory bowel disease patients.
        Clin Immunol Immunopathol. 1993; 66: 52-58
        • Fuss IJ
        • Neurath M
        • Boirivant M
        • Klein JS
        • Motte Cdl
        • Strong SA
        • Fiocchi C
        • Strober W
        Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease.
        J Immunol. 1996; 157: 1261-1270
        • Sollid LM
        • Kvale D
        • Brandtzaeg P
        • Markussen G
        • Thorsby E
        Interferon-γ enhances expression of secretory component, the epithelial receptor for polymeric immunoglobulins.
        J Immunol. 1987; 138: 4303-4306
        • Kvale D
        • Brandtzaeg P
        • Løvhaug D
        Up-regulation of the expression of secretory component and HLA molecules in a human colonic cell line by tumour necrosis factor–α and gamma interferon.
        Scand J Immunol. 1988; 28: 351-357
        • Phillips JO
        • Everson MP
        • Moldoveanu Z
        • Lue C
        • Mestecky J
        Synergistic effect of IL-4 and IFN-γ on the expression of polymeric Ig receptor (secretory component) and IgA binding by human epithelial cells.
        J Immunol. 1990; 145: 1740-1744
        • Krajci P
        • Taskén K
        • Kvale D
        • Brandtzaeg P
        Interferon-γ stimulation of messenger RNA for human secretory component (poly-Ig receptor) depends on continuous intermediate protein synthesis.
        Scand J Immunol. 1993; 37: 251-256
        • Fiorentino DF
        • Zlotnik A
        • Vieira P
        • Mosmann TR
        • Howard M
        • Moore KW
        • O'Garra A
        IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Th1 cells.
        J Immunol. 1991; 146: 3444-3451
        • Defrance T
        • Vanbervliet B
        • Briére F
        • Durand I
        • Rousset F
        • Banchereau J
        Interleukin 10 and transforming growth factor β cooperate to induce anti-CD40–activated naive human B cells to secrete immunoglobulin A.
        J Exp Med. 1992; 175: 671-682
        • Weiner HL
        Oral tolerance.
        Proc Natl Acad Sci USA. 1994; 91: 10762-10765
        • Seder RA
        • Paul WE
        Acquisition of lymphokine-producing phenotype by CD4+ T cells.
        Annu Rev Immunol. 1994; 12: 635-673
        • Manetti R
        • Parronchi P
        • Giudizi MG
        • Piccinni MP
        • Maggi E
        • Trinchieri G
        • Romagnani S
        Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper type 1 (Th1)-specific immune responses and inhibits the development of IL-4–producing Th cells.
        J Exp Med. 1993; 177: 1199-1204
        • Macatonia SE
        • Hosken NA
        • Litton M
        • Vieira P
        • Hsieh CS
        • Culpepper JA
        • Wysocka M
        • Trinchieri G
        • Murphy KM
        • O'Garra A
        Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells.
        J Immunol. 1995; 154: 5071-5079
        • Kato T
        • Hakamada R
        • Yamane H
        • Nariuchi H
        Induction of IL-12 p40 messenger RNA expression and IL-12 production of macrophages via CD40-CD40 ligand interaction.
        J Immunol. 1996; 156: 3932-3938
        • Gerosa F
        • Paganin C
        • Peritt D
        • Paiola F
        • Scupoli MT
        • Aste-Amezaga M
        • Frank I
        • Trinchieri G
        Interleukin-12 primes human CD4 and CD8 T cell clones for high production of both interferon-γ and interleukin-10.
        J Exp Med. 1996; 183: 2559-2569
        • Kobayashi M
        • Fitz L
        • Ryan M
        • Hewick RM
        • Clark SC
        • Chan S
        • Loudon R
        • Sherman F
        • Perussia B
        • Trinchieri G
        Identification and purification of natural killer cell stimulatory factor (NKSF), a cytokine with multiple biologic effects on human lymphocytes.
        J Exp Med. 1989; 170: 827-845
        • Trinchieri G
        Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive immunity.
        Annu Rev Immunol. 1995; 13: 251-276
        • Hsieh CS
        • Macatonia SE
        • Tripp CS
        • Wolf SF
        • O'Garra A
        • Murphy KM
        Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages.
        Science. 1993; 260: 547-549
        • Seder RA
        • Kelsall BL
        • Jankovic D
        Differential roles for IL-12 in the maintenance of immune responses in infectious versus autoimmune disease.
        J Immunol. 1996; 157: 2745-2748
        • Scott H
        • Nilsen EM
        • Sollid LM
        • Lundin KEA
        • Rugtveit J
        • Molberg Ø
        • Thorsby E
        • Brandtzaeg P
        Immunopathology of gluten-sensitive enteropathy.
        Springer Semin Immunopathol. 1997; 18: 535-553
        • Molberg Ø
        • Nilsen EM
        • Sollid LM
        • Scott H
        • Brandtzaeg P
        • Thorsby E
        • Lundin KEA
        CD4+ T-cells with specific reactivity against astrovirus isolated from normal human small intestine.
        Gastroenterology. 1998; 114: 115-122
        • Hauer AC
        • Breese EJ
        • Walker-Smith JA
        • MacDonald TT
        The frequency of cells secreting interferon-γ and interleukin-4, -5, and -10 in the blood and duodenal mucosa of children with cow's milk hypersensitivity.
        Pediatr Res. 1997; 42: 629-638