John Libbey Eurotext

The mucosal lesions in Crohn's disease (CD) and ulcerative colitis (UC) are characterized by an infiltration with inflammatory cells predominantly represented by neutrophils, macrophages and lymphocytes [1]. Phenotypic and limited functional studies indicate that macrophages present in these lesions are activated to produce proinflammatory mediators, oxygen free radicals and proteolytic enzymes [2-4].

Cytokines are key factors in the regulation of intestinal immune activation and in the mediation of IBD tissue damage. Changes in the local expression of cytokines, such as interleukin-2 (IL-2), IL-4, IL-10 and interferon-gamma (IFN-gamma), have been described in IBD, indicating that a dysregulation of the intestinal immune system is probably associated with pathogenic events [5]. Although, the etiology of IBD remains largely unknown, an impaired balance between proinflammatory and anti-inflammatory cytokines could be critical in determining the chronicity of inflammation and the pattern of remissions and reactivations characteristic of these diseases [6]. In addition to its classical functions as a neurotransmitter and vasodilatator, nitric oxide (NO) also plays an important role in inflammatory processes, as a mediator of macrophage function. It becomes quite evident that NO contributes to the pathophysiology of IBD, and inducible NO synthase (iNOS) is thought to exert a proinflammatory role in these disorders [7-9]. Furthermore, it is now well documented that cytokines mediating the inflammatory response such as IFN-gamma, tumor necrosis factor-alpha (TNF-alpha), and IL-1ß, are potent inducers of iNOS [10].

The multifunctional anti-inflammatory cytokine, IL-10 [11], originally described as cytokine synthesis inhibitory factor, is a product of a T-helper 2 (Th2) subset of T cells [12, 13], B cells, macrophages, thymocytes, keratinocytes and Kupffer cells [14, 15].
IL-10 strongly inhibits the production of cytokines by Th1 lymphocytes and by cells of the monocyte/macrophage lineage [16, 17]. Indeed, IL-10 inhibits IFN-gamma and IL-2 release by Th1 cells and down-regulates the enhanced secretion as well as the mRNA levels of IL-1, IL-6, IL-8, granulocyte-macrophage colony-stimulating factor and TNF-alpha from activated macrophages and neutrophils in vitro [18-20]. IL-10 also inhibits the formation of reactive oxygen intermediates and cellular functions such as macrophage-mediated cytotoxicity [21], specially NO production by IFN-gamma-activated macrophages [22].

Although IBD is characterized by an increased local production of Th1 cytokines in the gut mucosa, the cause of this pathogenic activation of the immune system remains unknown and is thought to result either from insufficient synthesis of inhibitory cytokines or from a resistance of the intestine cells to such cytokines. Interestingly, human IL-10 capable of down-regulating its own synthesis by monocytes via an auto-regulative, negative feedback mechanism [23] is elevated only in serum of patients with active IBD suggesting that IL-10 acts as a naturally occurring damper in the acute inflammatory process of IBD [24]. Furthermore, inactivation of the IL-10 gene by homologous recombination induces enterocolitis in mice [25], thereby indicating a potential role for IL-10 in the maintenance of a non-inflammatory state in the intestine.

Several human diseases are genetically linked to particular major histocompatibility complex (MHC) class I or II alleles. One of the most striking of these associations is that of the class I allele HLA-B27 with a group of human disorders termed spondylarthropathies in which inflammatory involvement of the gastrointestinal tract sometimes occurs. Although no increase in the frequency of the HLA-B27 gene has been reported in chronic IBD [26], HLA-B27 patients more frequently develop IBD [27, 28]. Several lines of rats transgenic for HLA-B27 and human ß2-microglobulin (hß2m) develop a spontaneous multisystem inflammatory disease [29] that strikingly resembles human B27-associated disease: histologic analysis showed gut lesions reminiscent of those described in ulcerative colitis, dependent on the presence of a normal gut flora [30]; the disease seems to be mediated by T lymphocytes and is transmittable to normal rats by bone marrow engraftment [31, 32]. Finally, a study has shown increased plasma levels of NO-derived metabolites [33].

The objectives of this study were to further investigate the mechanisms of IBD in HLA-B27 transgenic rats and to determine whether administration of murine recombinant IL-10 (mrIL-10) could affect established IBD in this model.

Purchased reagents and suppliers were as follows: 100 bp DNA ladder were from Gibco-BRL, Gaithersburg, MD; hexadecyltrimethylammonium bromide (HTAB), guaïcol, stabilised hydrogen peroxide, guanidium isothiocyanate, bovine serum albumin (BSA), phenylmethylsulfoxylfluoride (PMSF), aprotinin, leupeptin, diethylpyrocarbonate (DEPC), ethidium bromide, dithiotreitol (DTT), FAD, FMN, NADPH, tetrahydro-1-biopterin (BH₄) and horseradish peroxidase were from Sigma Chemical, Co, St Louis, MO; Taq polymerase and Aquaphenol were from Appligene, France; RNAsin, dXTP, oligo-dT and AMV-reverse transcriptase were from Promega, Madison, USA; primers were synthezised by Genosys, Cambridge, UK; L-¹⁴C-arginine monohydrochloride was from Amersham International Plc, Buckinghamshire, UK; all other chemicals and reagents were of analytical grade.

HLA-B27/hß2m transgenic female rats of the 33-3 line (F344 background) [29], 12 to 20 week-old (weight range: 150-200 g) bred under specific pathogen-free conditions were either kindly provided by J.D. Taurog and R.E. Hammer (Dallas, Texas) (first experiment) or purchased from Genpharm Int, Mountain View, CA (second experiment). Sex- and age-matched, F344 nontransgenic rats (n = 5) were purchased from Iffa-Credo, L'Arbresle, France. All rats were housed in rack-mounted cages with a maximum of 5 rats per cage: rats were randomized three days before initiation of the treatment, with a permutation table (each cage contained at least one rat from each experimental group). They were housed under conventional conditions and were fed standard laboratory chow and tap water ad libitum.

Murine rIL-10, kindly provided by Shering-Plough Research Institute, was administered intraperitonally, 5 days per week (first experiment: 10 and 100 µg/kg) or every day (second experiment: 200 µg/kg) for 5 consecutive weeks. Each experiment included groups of 33-3 rats that were left untreated or that received the vehicle of IL-10 only.

Rats were examined three times a week for clinical symptoms of colitis (loose stools and/or frank diarrhea), arthritis and alopecia and were assigned for each of these symptoms a score on a scale ranging from 0 (normal) to 4. The average daily total of these scores was calculated each week (maximum score per rat: 12). Body weight was monitored twice a week.

After completion of the 5 weeks of treatment, the rats were sacrificed: spleen, thymus, liver, colon and peri-pheral (PLN) and mesenteric (MLN) lymph nodes were weighed. The length of the colon was measured. In addition, each colon was assigned a macroscopic score on a scale ranging from 0 (normal) to 5 (severe), based on a classification adapted from Morris et al. [34] and Boughton-Smith et al. [35], which evaluates the presence of ulcerations, inflammation, depth of the lesions and fibrosis of the colonic wall.

Tissue samples were fixed in 10% (v/v) phosphate buffered saline formalin, ethanol dehydrated, and embedded in paraffin. Five microns sections were cut. The sections were stained routinely with hematoxylin-eosin. The severity of colonic lesions was scored, using a scoring system described in Table 1. The procedure for assigning a score was the following: slides from all rats were examined once; thereafter, all slides were reviewed blindly (M. Tulliez and S. Quéré) to determine the score, the maximum possible score per slide being 19.

Single flow cytometry was carried out as previously described [31]. The following monoclonal antibodies (mAb) and their specificities were used, references of which are cited in [31]: B1.23.2, IgG2a; R73, IgG1; OX35, IgG2a; OX8, IgG1; OX33, IgG1 and the mAb OX34, IgG2a [36]. The mAb V65, IgG1 was a kind gift of T. Hünig [37]. All procedures were carried out at 4° C in Dulbecco's PBS/4% FCS/0.05% NaN₃. Cells were incubated with saturating concentrations of the appropriate mAb for 30 min, washed, then incubated with FITC-conjugated monoclonal goat anti-mouse IgG (Eurobio, Les Ulis, France) for 30 min. After washing, the cells were analyzed using a FACScan flow cytometer and the LYSIS software (Becton Dickinson). The proportions of positive cells were determined, in comparison to isotype-matched irrelevant mAb staining.

MPO activity was measured according to the method of Maehly and Chance [38]. MPO was extracted from the colon by suspending tissue samples in 1 ml of lysis buffer containing 20 mM KH₂PO₄ and 1.4 mM hexadecyltrimethyl ammonium bromide (HTAB) (pH = 6.0), before homogeneization in an ice-bath with a polytron homogenizer. The suspension was assayed for MPO activity using a spectrophotometric method: 500 µl of suspension were mixed with 2,500 µl of buffer (pH = 6.0) containing 0.16 mM Na₂HPO₄, 18.4 mM KH₂PO₄, 44.8 µM guaïcol and 0.25 x 10^-3% hydrogen peroxide. The kinetics of absorbance at 470 nm was assessed with a spectrophotometer (Secomam S 1000, Sarcelles, France) set at 40° C. One unit of MPO activity is defined as that degrading 1 µmole of peroxide per minute [39]. The change in absorbance for each sample was expressed in international units (U) according to a standard range of horseradish peroxidase activity established under the same conditions. MPO activity was expressed in U/g of total protein contained in the colonic sample, as determined by the method of Lowry [40].

Colonic iNOS activity was estimated by measuring the conversion of L-¹⁴C-arginine to L-¹⁴C-citrulline [41]. Colonic tissue samples were homogenized in buffer containing 10 mM HEPES (pH = 7.4), 0.1 mM EDTA, 1 mM DTT, 23 µM leupeptin, 2.63 µM aprotinine, and 57 µM PMSF. After centrifugation (10,000 g, 30 min at 4° C), an aliquot of the supernatant was kept for measuring protein concentration and the remaining supernatant was centrifuged for 5 min on pre-equilibrated Dowex AG50W-X8. An aliquot of this supernatant was added to a reaction mixture containing 50 mM KH₂PO₄ (pH = 7.4), 157 pM L-¹⁴C-arginine, 15.54 nM L-arginine, 1 mM citrulline, 0.3 mM NADPH, 3 µM FMN, 1mM DTT, 3 µM BH₄, 50 mM valine, 0.2 mM CaCl₂, 1 mM MgCl₂, supplemented with 1 mM EDTA (in the tube iNOS). The mixture was incubated for 10 min at 37° C. L-¹⁴C-citrulline was separated by adding 500 µl of pre-equilibrated Dowex AG50W-X8 and centrifugation for 5 min at 10,000 g. After washing the Dowex twice with water, the amount of radioactivity was determined by scintillation counting. Data are expressed in pmole/mg of protein/min.

Extraction of total RNA. RNA was extracted from colon, using a modified guanidium-thiocyanate-phenol-chloroform extraction method [42]. Briefly, tissue samples were homogenized and lysed in guanidium solution (4 mol/l guanidinium-thiocyanate, 25 mmol/l sodium citrate, 0.5% sarcosyl, 0.1 mol/l ß2-mercaptoethanol, pH = 7.0). Then, 0.1 volumes of 1 mol/l sodium acetate (pH = 4.0), 1.25 volumes of H₂O-saturated phenol and 0.25 volumes of chloroform/isoamylalcohol (49v/1v) were added sequentially. The mixture was incubated on ice (10 min) and then centrifuged at 4° C (20 min x 12,000 g). RNA contained in the aqueous phase was precipitated overnight with 3 volumes of absolute ethanol ( 80° C). After centrifugation at 4° C (20 min x 12,000 g), the pellet was washed with 70% ethanol and RNA was resuspended in sterile H₂O treated with 0.1% diethyl pyrocarbonate (DEPC). The RNA contained in the extract was quantitated by absorbance at 260 nm. Purity and RNA degradation were assessed by electrophoresis on a 1.5% agarose gel-containing formaldehyde.

Reverse transcription (RT). RNA was reverse-transcribed into complementary DNA (cDNA): for interleukin detection, 2.5 µg of total RNA in 16 µl of DEPC water and mixed with reverse transcriptase reaction mixture containing 4.4 U RNAsin, 1.25 mM of each dXTP, 0.1 µg oligo-dT, 1.5 U AMV-reverse transcriptase and 4 µl 5X RT-buffer, were incubated at 42° C for 2 hours; for iNOS detection 2 µg of total RNA were incubated with 200 U murine Moloney leukemia virus reverse transcriptase and 100 µM random hexanucleotides for 1 hour at 37° C in a final volume of 20 µl.

Polymerase chain amplification (PCR). Cytokine cDNA amplification: cDNA equivalent to 250 ng total RNA was amplified in a 100 µl reaction volume containing 125 µM of each dXTP, 2 U of Taq DNA polymerase, 50 ng of the appropriate primer (Table 2) and 10 µl of 10X PCR buffer. After an initial denaturation step (94° C for 3 min), cDNA samples were subjected to rounds of denaturation (94° C for 30 sec), annealing (57° C for 30 sec) and extension (72° C for 40 sec) using the thermal cycler 9600 (Perkin-Elmer/Cetus). The samples were then submitted to a final extension (72° C for 10 min). Control samples containing no cDNA or 1 pg of a multispecific plasmid kindly provided by H.D. Volk (Institute for Medical Immunology, Berlin, Germany) [43] were included in all experiments to respectively exclude contamination and provide evidence of specific amplification. Fifty microliters of PCR product were mixed with 5 µl Sybr green nucleic acid gel and electrophoresed on a 2% agarose gel in Tris-ethylene-diaminetetraacetic acid buffer. A 100 bp ladder was used to assess sample size. The cDNA products were visualised by ultra-violet fluorescence and photographs of the gels were taken with a polaroïd negative film. The negative was scanned with a personal densitometer using Image Quant 3.3 software (Molecular Dynamics, Sunnyvale, USA). Semi-quantitative data are expressed in arbitrary units defined as the ratio between the OD obtained for the specific cDNA and the one for ß-actin on the same sample.

iNOS cDNA amplification: PCR was performed in a 50 µl reaction volume containing 200 µM of each dXTP, 2.5 U of Taq DNA polymerase, 0.2 to 0.5 µM specific primers and 1.5 mM Mg²⁺. Identity of the PCR product was confirmed by comparison with the expected size on agarose gel electrophoresis and by restriction mapping. Products were purified on Chroma Spin+TE-200 columns (Clontech). A negative and a positive control were included in each RT-PCR reaction. The positive control for iNOS was 1 µg of total RNA from LPS-treated rat liver. 1 µl of denatured PCR products was applied onto a Hybond-N⁺ nylon membrane (Amersham) and crosslinked for 5 min. The membranes were hybridized overnight at 60° C with fluoresceine-labeled probes and washed at 65° C (1X SSC, 0.1% SDS for 15 min and 0.5X SSC, 0.1% SDS for 15 min). The detection was performed according to the manufacturer's protocol (Fluorescein Gene Images, Amersham). Autoradiography was performed by exposure to X-Omat film (KodaK) and the intensity of high density scans was determined using NIH1.44 Image Software as previously described [44]. ß-actin was used as an endogenous internal standard. Semi-logarithmic plots of densitometry versus cycle number (15 to 35) were constructed. Before reaching the plateau effect and at the same PCR efficiency, the ratio between the logarithm of PCR products was equivalent to that between corresponding target mRNA [45-47]. Data are expressed in optic density (OD).

The data are expressed as median values and range [min-max]. For clinical and histological scores, the data are expressed as mean ± SEM. The statistical significance of differences was tested with non-parametric tests (Kruskal-Wallis and Mann Whitney). P values less than 0.05 were considered significant.

The HLA-B27 transgenic rats of the 33-3 line that were used in the two experiments reported here suffered from established disease. Chronic diarrhea was the most common and persistent finding, being present in all transgenic rats except one of those used in the second experiment. Arthritis and alopecia developed only in those rats used in the first experiment. Nontransgenic F344 rats remained healthy during the whole study period (clinical score: 0). The clinical disease score progressively worsened during the five weeks of investigation in all groups of transgenic rats, although none of the three clinical parameters was found to be different between the two control groups (that were therefore analyzed together) or any of the IL-10-treated groups (Table 3). The weight of transgenic rats did not vary significantly during the five weeks of experiment (not shown). Although the 33-3 rats purchased from Genpharm that were used in this second experiment were generally less severely affected than those generally provided by J.D.T. (first experiment), there was again no difference in clinical disease severity between the three groups of rats (Table 3).

The weights of PLN, spleen, liver and thymus were similar in nontransgenic and in all groups of 33-3 rats, whether receiving IL-10 or not (not shown). MLN were hyperplastic in HLA-B27 transgenic rats as demonstrated by a significant increase of weight, as compared to nontransgenic rats. However, there was no influence of IL-10 treatment on this parameter (Table 4). We also observed a significant increase of approximately two fold in the weight/length ratio of the colon in B27-transgenic rats of the control as compared to nontransgenic rats (p < 0.05). This ratio was not modified by the treatment with IL-10 (Table 4).

Macroscopically, nontransgenic rats had a score equal to 0. The macroscopic damage observed in B27-transgenic rats were small ulcerations (< 3 mm) surrounded by thickened inflamed tissue. The score increased to about 1 in all transgenic groups (not shown).

Microscopically, all nontransgenic rats had a normal intestinal mucosa (Figure 1A). As previously described [29], in HLA-B27 transgenic rats, the colon was the site most consistently and prominently affected by IBD (Figure 1B and C). As shown in Table 4, the median histological grade of inflammation was quite elevated in transgenic rats without significant differences between IL-10-treated (10 or 100 µg/kg/d) and control rats (the latter referring to combined vehicle-treated and untreated groups). In the second experiment, histologic evidence of inflammation in the colon was found in all 33-3 rats examined apart from the one that had remained free of diarrhea, in the group treated with 200 µg/kg/d of IL-10. Albeit the histologic score of severity was generally lower in these rats, it was not modified in rats treated with the highest dose of mrIL-10, confirming the clinical findings.

MLN cells were analyzed by flow cytometry, for surface expression of HLA-B27 and several lymphoid markers, using a panel of murine mAbs. Neither the absolute numbers, nor the proportions of lymphoid cell subsets were modified in 33-3 rats treated with the highest dose of mrIL-10 (Table 5).

In an attempt to quantify the infiltration of colonic mucosa by neutrophils, we assessed colonic MPO activity. In the first experiment, colonic MPO activity of the HLA-B27 transgenic control rats was increased by approximately three fold in comparison with nontransgenic rats (median: 3.91 U/g of protein, range [2.05-5.12] versus median: 1.15 U/g of protein, range [0.98-1.4]; p <0.05). This increased level of MPO activity was not reduced in rats treated with 10 µg/kg/d (median: 3.27 U/g of protein, range [2.19-9.25]) or 100 µg/kg/d (median: 3.20 U/g of protein, range [2.43-5.88]) of mrIL-10. In the second experiment, the increased level of MPO activity in B27-transgenic rats was less pronounced, reflecting an overall milder disease status, which was nevertheless not
reduced by the highest dose of IL-10 administered (200 µg/kg/d).

The activity of the iNOS the Ca²⁺-independent fraction was detectable in the colonic mucosa in rats of all groups. However, it was very low or undetectable in samples from nontransgenic rats, median (0.067 pmol/min/mg of protein [range: 0-0.064]). In contrast, this Ca²⁺-independent activity was dramatically enhanced in all clinically-affected transgenic rats median (0.749 pmol/min/mg of protein [range: 0.115-3.309]). The iNOS activity was not modified by the treatment with any of the mrIL-10 dosages either in the first or the second experiment.

RNA extracts of colonic mucosa were subjected to RT-PCR, using primers specific for rat TNF-alpha, IFN-gamma, CD3, and alpha-actin, yielding the expected corresponding 468, 419, 253 and 762 base-paired products. As shown in Table 6, CD3 mRNA was expressed in all groups of rats, but at higher levels in groups of 33-3 rats than in nontransgenic rats. However, its expression was not significantly modified by the treatment with IL-10. TNF-alpha mRNA was also expressed in all groups of rats. However, it is likely that TNF-alpha mRNA was decreased by the treatment with IL-10, since TNF-alpha mRNA was no longer detectable in 3/6 rats and 4/5 rats, in the groups treated with 100 µg/kg/d and 200 µg/kg/d respectively, and the level detected was lower in the three (0.087 ± 0.03) and the one (0.05) remaining rats of these two groups, than in control transgenic rats.

IFN-gamma mRNA was not detected in any nontransgenic rat. In contrast, this cytokine mRNA was over-expressed in control transgenic rats (Table 6). In the group treated with IL-10 at 10 µg/kg, IFN-gamma mRNA expression was not different than in the control rats. In the group treated with IL-10 at 100 µg/kg, IFN-gamma mRNA was not detected in 3/6 rats although it was not significantly decreased in the 3 remaining rats. Finally, in the group treated with IL-10 at 200 µg/kg, IFN-gamma mRNA was not detected in 3/5 rats and was very low in the 2 remaining rats (0.03 and 0.07), suggesting an almost complete inhibitory effect of IL-10 on IFN-gamma mRNA expression at this highest dose of IL-10.

RNA extracts from colonic mucosa were subjected to RT-PCR, yielding a 499 base-paired product that was identified as the expected iNOS gene product. iNOS mRNA was not detected in nontransgenic rats and its expression was up-regulated in control transgenic rats. There was no obvious effect of the treatment with IL-10 on the expression level of iNOS mRNA.

Rats transgenic for HLA-B27 and hß2m are a model of human IBD in which gastrointestinal attack, in particular at the colonic level, predominates [29]. Our present study confirms that overt diarrhea or loose stools are the most common and persistent clinical findings. Furthermore, spontaneous colonic inflammation is associated with hyperplasia of MLN and thickening of the colonic mucosa that is reflected by an increase in the weight/length ratio of the colon. It is worth noting that, during the five week study period the weight of the rats did not significantly change regardless of IL-10 treatment. On the whole, clinical signs were only slightly modified during the experimental period.

The immunological process by which chronic inflammation and progressive destruction of the colonic mucosa are induced in HLA-B27 rats, remains un-known. However, the histological lesions in the colonic mucosa are similar, on the one hand to those previously described in IL-10- or IL-2-knockout mice [25, 48] and on the other hand to those described in human IBD such as ulcerative colitis. On histologic examination, the colonic mucosa of HLA-B27 rats is characterized by an extensive inflammatory infiltrate consisting mainly of lymphocytes, polynuclear neutrophils, plasmocytes, mastocytes and macrophages. However, we did not notice any modification of this inflammatory process in the groups treated with IL-10. Moreover, in an attempt to quantify the neutrophilic infiltrate in the colonic mucosa we assessed colonic MPO activity and report here that IL-10 does not affect this parameter. Taken together, these results suggest that IL-10 does not inhibit the influx of inflammatory cells into the colonic mucosa.

In human IBD, both TNF-alpha and IFN-gamma have been suggested as critical proinflammatory mediators involved in the initiation and/or progression of IBD. We also found that, as previously reported [49], in HLA-B27 transgenic rat, colonic inflammation in established disease is associated with an enhanced local expression of IFN-gamma mRNA. Moreover, the high level of NO in plasma reported by others [33] as assessed for an increased NO metabolism is in good agreement with regulation the up- of iNOS activity and mRNA level in transgenic rats reported here.

Recent evidence suggests that the outcome of experimental IBD may be the result of an imbalance between host-protective Th2 and disease-promoting Th1 responses. The cytokine IL-10 was initially detected as a product of murine Th2 cells that could inhibit cytokine synthesis (IL-2 and IFN-gamma) by activated cells of the Th1 subset. The capacity of IL-10 to interfere with the production of Th1-derived cytokines is a contributing factor to its modulatory effects on cell-mediated inflammatory responses. We demonstrate here that IL-10 effectively downregulates IFN-gamma at the mRNA level consistent with the results obtained of the in vivo protective effects of exogenous IL-10 on experimental granulomatous inflammatory responses [50]. These data indicate that IL-10 may exert a significant influence on the cytokine production in the IBD of B27-transgenic rats. These results also suggest that IL-10 and IFN-gamma may be reciprocally regulated in this model, as already described in rheumatoid synovium [51]. However, despite the fact that IFN-gamma mRNA expression was down-regulated, there was no reversal of clinical and histological findings. Indeed, the doses of IL-10 used here could be responsible for the absence of effects, the highest dose of IL-10 (200 µg/kg) we used in this study is similar to the one (250 µg/kg) Herfarth et al. described as relatively successfully attenuating acute and chronic granulomatous inflammation induced by bacterial cell wall polymers [50]. Moreover, IL-10 has been shown to successfully prevent the onset of diabetes in the non-obese diabetic mouse [52] and in experimental allergic encephalomyelitis in rats [39] at a dose range lower than 100 µg/kg. Negative regulation of IFN-gamma by IL-10 is the suggested mechanism of disease prevention in the experimental allergic encephalomyelitis model [53] and in the non-obese diabetic mouse model [52]. Another variable is tissue concentration of the administered IL-10 within the colonic inflammed tissue which could not be measured in this study. For this purpose, topical administration of recombinant IL-10 provided benefit in experimental models [54] as well as in human IBD [5]. Importantly, a number of therapeutic interventions known to inhibit Th1 responses, inhibited disease induction, but the efficacy and mechanism of action of these treatments is still under discussion [55-57].

The iNOS mRNA and enzymatic activity were markedly induced in the colonic mucosa of HLA-B27 transgenic rats: these results are in agreement with the increased NO production in plasma previously described by Aiko and Grisham in this model [33]. This enhanced NO metabolism in plasma presumably reflects what happens at the colonic level. The functional significance of NO produced by mucosal iNOS activity in IBD is unknown. However, some modifications associated with UC and CD could be ascribed to this NO production. This is the case for mucosal vasodilation, resulting in mucosal erythema and increased vascular permeability which results in mucosal edema [58]. IBD is also marked by enhanced epithelial permeability as measured by the passage of tracer molecules. Finally, the diarrhea that is frequently the major symptom of patients with UC or CD and in HLA-B27 transgenic rats usually has a secretory component and NO has been shown to induce chloride secretion in rabbit colon through a prostaglandin, and partially neural-dependent mechanism that may involve guanylate cyclase. Moreover, NO has been known for the past six years to mediate some aspects of macrophage cytotoxicity. In contrast to constitutive NOS that lies dormant until and so long a rise in intracellular Ca²⁺ sustains the binding of calmodulin leading to NO release over several minutes, iNOS is expressed in many cell types after challenge with immunologic or inflammatory stimuli and thereupon generates large amounts of NO over periods as long as five days. The prolonged administration of IL-10 did not markedly affect the expression of iNOS in B27-transgenic rat colonic mucosa. Although IFN-gamma is known to synergise with LPS to induce transcription and also to stabilize iNOS mRNA, alternative pathways are probably implicated in the over-expression of iNOS mRNA in HLA-B27 transgenic rat colonic mucosa.

While several studies in small groups of humans showed a benefit of treatment with IL-10, either via the parenteral route in patients with CD or as a topical treatment in patients with UC [5], the chronic IBD in HLA-B27 transgenic rats was not ameliorated by a curative treatment with IL-10. Our present results and those of Berg et al. demonstrating that IL-10 completely prevented the development of enterocolitis in IL-10 deficient without reversing established disease [59], are in favour of an early key role for IL-10 in the immune response. Moreover, given the redundancy of many immunological processes, treatment with IL-10 alone might not be sufficient to completely suppress established inflammatory bowel disease. For this reason it becomes evident that an additional anti-inflammatory agent may be necessary to provide synergistic benefit. For this purpose, Powrie et al. showed in a delayed type hypersensitivity model that the combination of IL-4 and IL-10 is more effective than monotherapy [60] and Herfarth et al. demonstrated additive activities of IL-10 and corticoïds in the PG-APS model [50].

The IBD in HLA-B27 transgenic rats is characterized by a local up-regulation of IFN-gamma in the colonic mucosa, indicating that in this model, the disease process could be mediated by Th1 lymphocytes. Although prolonged administration of IL-10 efficiently inhibited the expression of INF-gamma, it failed to induce clinical, histological or biological improvement of the IBD disease of HLA-B27 transgenic rats. These results allow us to speculate that IFN-gamma expression is an early event that could be critically involved in the appearance, but probably not in the maintenance of the disease. This particular point will require further investigation using younger or germ-free HLA-B27 rats colonized with a normal flora. Moreover, IL-10 was rather inefficient at interfering with iNOS activity in the colonic mucosa, suggesting that local NO production could be another key factor responsible for the induction and/or maintenance of colonic lesions in this model. IFN-gamma production is involved in the pathogenesis of this murine model of IBD that bears some similarity to UC in humans. Future studies with associated treatments and other animal models of IBD will undoubtelly lead to a clearer understanding of the entire pathogenic process and improve chances of developing effective treatments for human IBD.

We thank Dr. J.D. Taurog and Dr. R.E. Hammer for kindly providing us with HLA-B27 transgenic rats of the 33-3 line. This research was supported by a grant from Schering-Plough Research Institute and by a grant from Ferring Laboratories.

1. Stenson W F. 1995. Inflammatory bowel disease. In: Yamada, T. Ed. Philadelphia. Lippincott, 1748.

2. Allison M C, Poulter W. 1991. Changes in phenotypically distinct mucosal macrophage population may be a prerequisite for development of inflammatory bowel disease. Clin. Exp. Immunol. 85: 504.

3. Murch S H, Braegger C P, Walker-Smith J A, Mac Donald T T. 1993. Location of tumour necrosis factor-alpha by immunochemistry in chronic inflammatory bowel disease. Gut 34: 1705.

4. Raab Y, Hällgren R, Gerdin B. 1994. Enhanced intestinal synthesis of interleukin-6 is related to the disease severity and activity in ulcerative colitis. Digestion 55: 44.

5. Schreiber S, Heinig T, Thiele H G, Raedler A. 1995. Immunoregulatory role of interleukin-10 in patients with inflammatory bowel disease. Gastroenterology 108: 1434.

6. Niessner M, Volk B A. 1995. 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. 104: 428.

7. Miller M J S, Clark D A. 1994. Nitric oxide synthase inhibition can initiate or prevent gut inflammation: role of enzyme source. Agents Actions 41: C231 (abstract).

8. Mourelle M, Casellas F, Guarner F, Salas A, Riveros-Moreno V, Moncada S, Malagelada J R. 1995. Induction of nitric oxide synthase in colonic smooth muscle from patients with toxic megacolon. Gastroenterology 109: 1497.

9. Rachmilewitz D, Stamler J S, Bachwich D, Karmeli F, Ackerman Z, Podolsky D K. 1995. Enhanced colonic nitric oxide generation and nitric oxid synthase activity in ulcerative colitis and Crohn's disease. Gut 36: 718.

10. Nathan C, Xie Q W. 1994. Nitric oxide synthases: roles, tolls and controls. (Review) Cell 78: 915.

11. Fiorentino D F, Bond M W, Mosmann T R. 1989. Two types of mouse helper T cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones. J. Exp. Med. 170: 2081.

12. De Waal Malefyt R, Yssel H, Roncarolo M G, Spits H, De Vries J E. 1992. Interleukin-10. (Review) Curr. Opin. Immunol. 4: 314.

13. Moore K W, O'Garra A, De Waal Malefyt R, Vieira P, Mosmann T R. 1993. Interleukin-10. (Review) Annu. Rev. Immunol. 11: 165.

14. O'Garra A, Stapleton G, Dhar V, Pearce M, Schumacher J, Rugo H, Barbis D, Stall A, Cupp J, Moore K, Vieira P, Mosmann T, Whitmore A, Arnold L, Haughton G, Howard M. 1990. Production of cytokines by mouse B cells: B lymphomas and normal B cells produce interleukin-10. Int. Immunol. 2: 821.

15. Knolle P, Schlaak J, Uhrig A, Kempf P, Meyer zum Buschenfelde K H, Gerken G. 1995. Human Kupffer cells secrete IL-10 in response to lipopolysaccharide (LPS) challenge. J. Hepatol. 22: 226.

16. Mosmann T R, Moore K W. 1991. The role of IL-10 in crossregulation of Th1 and Th2 responses. (Review) Immunol. Today 12: 49.

17. Howard M, O'Garra A. 1992. Biological properties of interleukin-10. (Review) Immunol. Today 13: 198.

18. Fiorentino D F, Zlotnik A, Mosmann T R, Howard M, O'Garra A. 1991. Interleukin-10 inhibits cytokine production by activated macrophages. J. Immunol. 147: 3815.

19. Cassatella M A, Meda L, Bonora S, Ceska M, Constantin G. 1993. I- Interleukin-10 (IL-10) inhibits the release of proinflammatory cytokines from human polymorphonuclear leukocytes. II- Evidence for an autocrine role of tumour necrosis factor and IL-1ß in mediating the production of IL-8 triggered by lipopolysaccharide. J. Exp. Med. 178: 2207.

20. De Vries J E. 1995. Immunosuppressive and anti-inflammatory properties of interleukin-10. (Review) Ann. Med. 27: 537.

21. Bogdan C, Nathan C. 1993. Modulation of macrophage function by transforming growth factor ß, interleukin-4 and interleukin-10. Ann. NY Acad. Sci. 685: 713.

22. Gazzinelli R T, Oswald I P, James S L, Sher A. 1992. IL-10 inhibits parasite killing and nitrogen oxide production by IFN-gamma activated macrophages. J. Immunol. 148: 1792.

23. De Waal Malefyt R, Abrams J, Bennett B, Figdor G G, De Vries J E. 1991. Interleukin-10 (IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes. J. Exp. Med. 174: 1209.

24. Kucharzik T, Stoll R, Lugering N, Domschke W. 1995. Circulating anti-inflammatory cytokine IL-10 in patients with inflammatory bowel disease (IBD). Clin. Exp. Immunol. 100: 452.

25. Kühn R, Löhler J, Rennick D, Rajewsky K, Müller W. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75: 263.

26. Leirisalo Repo M, Turunen U, Stenmans S, Helenius P, Seppälä K. 1994. High frequency of silent inflammatory bowel disease in spondylarthropathy. Arthritis Rheum. 37: 23.

27. Mielants H, Veys E M, Cuvelier C, De Vos M, Botelberghe L. 1985. HLA-B27 related arthritis and bowel inflammation. Ileocolonoscopy and bowel histology in patients with HLA-B27 related arthritis. J. Rheumatol. 12: 244.

28. Mielants H, Veys E M, Goemaere S, Cuvelier C, De Vos M. 1993. A prospective study of patients with spondylarthropathy with special reference to HLA-B27 and the gut histology. J. Rheumatol. 20: 1353.

29. Hammer R E, Maika S D, Richardson J A, Tang J P, Taurog J D. 1990. Spontaneous inflammatory disease in transgenic rats expressing HLA-B27 and human ß2-microglobulin: an animal model of HLA-B27 associated human disorders. Cell 63: 1099.

30. Taurog J D, Richardson J A, Croft J T, Simmons W A, Zhou M, Fernandez-Sueiro J L, Balish E, Hammer R E. 1994. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J. Exp. Med. 180: 2359.

31. Breban M, Fernandez-Sueiro J L, Richardson J A, Hadavand R R, Maika S D, Hammer R E, Taurog J D. 1996. T cells, but not thymic exposure to HLA-B27, are required for the inflammatory disease of HLA-B27 transgenic rats. J. Immunol. 156: 794.

32. Breban M, Hammer R E, Richardson J A, Taurog J D. 1993. Transfer of inflammatory disease of HLA-B27 transgenic rats by bone marrow engraftment. J. Exp. Med. 178: 1607.

33. Aiko S, Grisham M B. 1995. Spontaneous intestinal inflammation and nitric oxide metabolism in HLA-B27 transgenic rats. Gastroenterology 109: 142.

34. Morris G P, Bleck P L, Herridge M S, Depew W T, Szewozuk M R, Wallace J L. 1989. Hapten-induced model of chronic inflammation and ulceration in the rat colon. Gastroenterology 96: 795.

35. Boughton-Smith N K, Wallace J L, Morris G P, Whittle B J. 1988. The effect of anti-inflammatory drugs on eicosamoid formation in a chronic model of inflammatory bowel disease in the rat. Br. J. Pharmacol. 94: 65.

36. Beyers A D, Barclay A N, Law D A, Williams A F. 1989. Activation of T lymphocytes via monoclonal antibodies against rat cell surface antigens with particular reference to CD2 antigen. (Review) Immunol. Rev. 111: 59.

37. Kühnlein P, Park J H, Herrmann T, Elbe A, Hünig T. 1994. Identification and characterization of rat gamma/delta T lymphocytes in peripheral lymphoid organs, small intestine, and skin with a monoclonal antibody to a constant determinant of the gamma/delta T cell receptor. J. Immunol. 153: 979.

38. Maehly A C, Chance B. 1954. The assay of catalases and peroxidases. In: Glick D. Eds. Methods of Biochemical analysis. Interscience, New-York, 357.

39. Freehold N J. 1972. Worthington enzyme manual. Worthington Biochemical Corp., 43.

40. Lowry O H. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193: 265.

41. Bush P A, Gonzalez N E, Griscavage J M, Ignarro L J. 1992. Nitric oxide synthase from cerebellum catalyzes the formation of equimolar quantities of nitric oxide and citrulline from L-arginine. Biochem. Biophys. Res. Commun 185: 960.

42. Chromczynski P, Sacchi N. 1987. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156.

43. Siegling A, Lehmann M, Platzer C, Emmrich F, Volk H D. 1994. A novel multispecific competitor fragment for quantitative PCR analysis of cytokine gene expression in rats. J. Immunol. Methods 177: 23.

44. Kawai N, Bloch D B, Filippov G, Rabkina D, Suen H, Losty P D, Janssens S P, Zapol W M, de la Monte S, Bloch K D. 1995. Constitutive endothelial nitric oxide gene expression is regulated during lung development. Am. J. Physiol. 268: L589.

45. Chelly J, Kaplan J C, Maire P, Gautron S, Kahn A. 1988. Transcription of the dystrophin gene in human muscle and non-muscle tissues. Nature 333: 858.

46. Salomon R N, Underwood R, Doyle M V, Wang A, Libby P. 1992. Increased apolipoprotein E and c-fos gene expression without elevated interleukin-1 or 6 mRNA levels indicates selective activation of macrophage functions in advanced human atheroma. Proc. Natl. Acad. Sci. USA 89: 2814.

47. Wang A M, Doyle M V, Mark D F. 1989. Quantification of mRNA by polymerase chain reaction. Proc. Natl. Acad. Sci. USA 86: 9717.

48. Sadlack B, Merz H, Schorle H, Schimpl A, Feller A C, Horak I. 1993. Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 75: 253.

49. McLean L, Hammer R E, Taurog J D. 1994. Cytokine and
T cell receptor VßmRNA profiles in the inflammatory disease of HLA-B27 transgenic rats. Arthritis Rheum. 37: A384
(abstract).

50. Herfarth H H, Mohanty S P, Rath H C, Tonkonogy S, Sartor R B. 1996. IL-10 suppresses experimental chronic, granulomatous inflammation induced by bacterial cell wall polymers. Gut 39: 836.

51. Katsikis P D, Chu C Q, Brennan F M, Maini R N, Feldmann M. 1994. Immunoregulatory role of interleukin-10 in rheumatoid arthritis. J. Exp. Med. 179: 1517.

52. Pennline K J, Roque-Gaffney E, Monahan M. 1994. Recombinant human IL-10 prevents the onset of diabetes in the non obese diabetic mouse. Clin. Immunol. Immunopathol. 71: 169.

53. Rott O, Fleischer B, Cash E. 1994. Interleukin-10 prevents experimental allergic encephalomyelitis in rats. Eur. J. Immunol. 24: 1434.

54. Mulligan M S, Jones M L, Vaporciyan A A, Howard M C, Ward P A. 1993. Protective effects of IL-4 and IL-10 against immune complexe-induced lung injury. J. Immunol. 151: 5666.

55. Itoh S, Matsuzaki Y, Ikegami T, Kimura T, Shoda J, Tanaka N. 1997. Cytokine profile of purified CD4 T lymphocytes and induction of hepatic lesion in murine model of primary biliary cirrhosis. Gastroenterology 112: A1291.

56. Barbara G, Hogabam C M, Gauldie J, Collins S M. 1997. Interleukin-10 gene therapy prevents the occurence but doesn't reverse experimental colitis. Gastroenterology 112: A927 (abstract).

57. Steele-Norwood D, Stallwood N, Borojevic R, Snider D, Croitoru K. 1997. Alterations in Th1 cytokine profiles in peripheral blood and lamina propria lymphocyte subsets in IBD. Gastroenterology 112: A1096 (abstract).

58. Mielants H, De Vos M, Goemaere S, Schelstraete K, Cuvelier C, Goethals K, Maertens M, Ackerman C, Veys E M. 1991. Intestinal mucosal permeability in inflammatory rheumatic diseases. II. Role of disease. J. Rheumatol. 18: 394.

59. Berg D J, Davidson N, Kühn R, Müller W, Menon S, Holland G, Thompson-Snipes L, Leach M W, Rennick D. 1996. Enterocolitis and colon cancer in IL-10 deficient mice are associated with aberrant cytokine production and CD4⁺Th1-like responses. J. Clin. Invest. 98: 1010.

60. Powrie F, Menon S, Coffman R L. 1993. IL-4 and IL-10 synergize to inhibit cell-mediated immunity in vivo. Eur. J. Immunol. 23: 2223.

Effects of murine recombinant interleukin-10 on the inflammatory disease of rats transgenic for HLA-B27 and human 2-microglobulin