Abstract
Interleukin 6 [IL-6] or its receptor is currently a candidate for targeted biological therapy of inflammatory bowel disease [IBD]. Thus, a comprehensive understanding of the consequences of blocking IL-6 is imperative. We investigated this by evaluating the effects of IL-6 deletion on the spontaneous colitis of IL-10-deficient mice [IL-10−/−].
IL-6/IL-10 double-deficient mice [IL-6−/−/IL-10−/−] were generated and analysed for intestinal inflammation, general phenotypes and molecular/biochemical changes in the colonic mucosa compared with wild-type and IL-10−/− mice.
Unexpectedly, the IL-6−/−/IL-10−/− mice showed more pronounced gut inflammation and earlier disease onset than IL-10−/− mice, both locally [colon and small bowel] and systemically [splenomegaly, ulcerative dermatitis, leukocytosis, neutrophilia and monocytosis]. IL-6−/−/IL-10−/− mice exhibited elevations of multiple cytokines [IL-1β, IL-4, IL-12, TNFα] and chemokines [MCP-1 and MIG], but not IFN-γ [Th1], IL-17A and IL-17G [Th17], or IL-22 [Th22]. FOXP3 and TGF-β, two key factors for regulatory T [Treg] cell differentiation, were significantly down-regulated in the colonic mucosa, but not in the thymus or mesenteric lymph nodes, of IL-6−/−/IL-10−/− mice. CTLA-4 was diminished while iNOS was up-regulated in the colonic mucosa of IL-6−/−/IL-10−/− mice.
In IL-10−/− mice, complete IL-6 blockade significantly aggravates gut inflammation, at least in part by suppressing Treg/CTLA-4 and promoting the IL-1β/Th2 pathway. In addition, the double mutant exhibits signs of severe systemic inflammation. Our data define a new function of IL-6 and suggest that caution should be exercised when targeting IL-6 in IBD patients, particularly those with defects in IL-10 signalling.
1. Introduction
Inflammatory bowel disease [IBD], encompassing ulcerative colitis [UC] and Crohn’s disease [CD], is a disorder of the gastrointestinal tract characterized by chronic, remitting and relapsing inflammation.1,2 IBD results from a dysregulated immune response [both innate and adaptive] to gut commensal bacteria, and is modulated by environmental factors in genetically susceptible hosts.2–5 The innate immune system, primarily including dendritic cells, macrophages and epithelial cells, senses microbes and initiates both a rapid and a sustained response.6,7 In contrast, the adaptive immune system functions mainly through CD4+ T cells in IBD, consisting of effector subsets [Th1, Th2 and Th17], which promote the inflammation, and/or regulatory T [Treg] cells, which are immunosuppressive and maintain homeostasis of the intestine.8–10
Interleukin 6 [IL-6] is produced by various cell types.11,12 IL-6 signalling is thought to be achieved by two mechanisms: [1] regenerative and/or anti-inflammatory functions of IL-6, primarily mediated by a classical pathway through binding to the membrane-bound IL-6 receptor [IL-6R]; or [2] the proinflammatory responses, regulated by a trans-signalling pathway through binding to a soluble form of IL-6R and subsequently recruiting membrane-bound glycoprotein 130 [gp130].13
Previous studies have demonstrated that IL-6 signalling plays a key role in the pathogenesis of IBD.14,15 Elevated levels of IL-6 are found in both the blood and colonic mucosa of IBD patients compared to those of healthy controls.16 IL-6 augments pathogenic cytokine production by colonic innate CD3+ IL17R+ cells from IBD patients14. Furthermore, serum levels of IL-6 were shown to be positively associated with disease activity in both CD and UC.17 The importance of IL-6 in IBD pathogenesis was further demonstrated in murine models of colitis, including T-cell-reconstituted severe combined immunodeficiency [SCID] mice, and mice with TNBS [2,4,6-trinitrobenzene sulfonic acid]- or DSS [Dextran sulfate sodium]-induced colitis.18–21 In recent years, the therapeutic potential of targeting IL-6 as a strategy to treat autoimmune disease and other chronic inflammatory diseases has been explored.22,23 Agents that inhibit IL-6 or its receptor have shown promise in phase I and II clinical trials in CD patients.24 Recently, an anti-IL-6 antibody [PF-04236921] exhibited clinical response and remission in refractory patients with moderate to severe CD following the failure of anti-tumour necrosis factor [anti-TNF] therapy.25
However, the contribution of IL-6 in the pathogenesis of IBD has not been completely elucidated. Conflicting in vivo studies in mouse models have suggested both detrimental and protective effects of IL-6.26 Several reports have suggested that IL-6 plays a potential role in wounding healing of the injury and intestinal homeostasis.27–29 Thus, it is conceivable that IL-6 can play both a pro-inflammatory and a protective/anti-inflammatory role during the inflammatory process depending on the models or models of regulations. Therefore, a better understanding of the function and regulation of IL-6 in experimentally induced colitis is needed if IL-6 blockade is to be considered as a treatment for IBD.
IL-10 is an immunomodulatory cytokine and its main biological function is to limit and terminate inflammatory responses. Patients carrying mutations of the IL-10 receptor that abrogate IL-10 signalling develop a more aggressive disease at very early age.30 Furthermore, IL-10−/− and IL-10R−/− mice spontaneously develop colitis and such mice have long been considered as a classical animal model of IBD, in which IL-6 is increased and plays a key role during the inflammation process.31 All these studies are consistent with the notion that a lack of IL-10-mediated negative feedback signalling perturbs homeostasis of the intestinal immune system, leading to IBD.
Given that IL-10−/− mice spontaneously develop colitis with elevated levels of IL-6, it is conceivable that the development of colitis may be at least in part due to up-regulated IL-6. Therefore, we generated IL-6/IL-10 double-deficient mice [IL-6−/−/IL-10−/−] to investigate the in vivo functional interplay between IL-10 and IL-6.
2. Materials and Methods
2.1. Animals
Wild type [WT] and IL-10−/− mice [all in C57BL/6 background] were originally obtained from The Jackson Laboratory. IL-6-deficient [IL-6−/−] mice [homozygous deletion of Il-6, also in C57BL/6] were generously provided by Dr Bin Gao, National Institute on Alcohol Abuse and Alcoholism, NIH, Bethesda, MD, USA. IL-6−/− mice were crossbred with IL-10−/− mice. Their offspring were bred >10 times to generate IL-6/IL-10 double-deficient mice [IL-10−/−/IL-6−/−]. The genotypes of the mice were confirmed by analysis of tail tip digests by PCR. All mice were raised in exactly the same environmental conditions and under strict care at the Johns Hopkins Animal Facility; all were group-housed in the same room, under the same controlled temperature [25°C] and photoperiods [12:12-h light–dark cycle], and fed with the same chow and water. Care and experimentation of mice were performed according to institutional guidelines using protocols approved by the Johns Hopkins Institutional Animal Care and Use Committee. Mice were kept in Helicobacter-positive specific-pathogen-free conditions.
2.2. Antibodies
Antibodies used include IL-6 mAb [Transduction Laboratories], iNOS mAb [Abcam] and actin pAb [Sigma-Aldrich].
2.3. Evaluation of colitis
Animals were observed once daily for weight, water/food consumption, morbidity, stool consistency, piloerection, the presence of gross blood in faeces and at the anus, and prolapse at the anus. Disease severity was evaluated as described previously.32 At the end point of experiments, before the animals were killed, the mice were anaesthetized and blood was collected in EDTA-containing tubes for preparation of serum. The animals were then killed via CO2 inhalation, rapidly dissected, and the entire gastrointestinal tract [caecum, colon, and the whole small bowel including jejunum and ileum] was quickly excised and photographed. The colon was gently cleared of faeces with 4°C saline and weighed. The colonic weight was previously shown to correlate strongly with the clinical and histological severity of the disease32. Small segments of the colon and small bowel [jejunum and ileum] taken for histopathology and immunohistochemistry were fixed in 10% normal buffered formalin. Sections [4 μm] were stained with H&E [Richard Allen Scientific]. Histological scores of the bowel were determined by an experienced pathologist who was blinded to treatment, and histological activity index [HAI] was calculated as described previously.33
2.4. Harvest of other organs/tissues
Several other organs or tissues were harvested from experimental animals, including liver, spleen, thymus, kidney and mesenteric lymph nodes. These tissues/organs were stored at −80°C for further PCR or Western blot analyses. Liver, spleen and kidneys were weighed and photographed.
2.5. Haematological profiling for evaluation of systemic inflammation
Blood was collected from the periorbital venous plexus for haematological profiling, which was performed at the Phenotyping Core, Johns Hopkins School of Medicine, Department of Molecular and Comparative Medicine.
2.6. RNA isolation and RT-PCR analysis
RNA in the colonic mucosa, spleen, thymes or mesenteric lymph nodes was extracted with Trizol [Invitrogen], and reverse-transcribed to cDNA using a high-capacity cDNA reverse transcription Kit [Applied Biosystems]. PCR amplification was carried out using 12.5 µl DreamTaq Green PCR Master Mix [Fermentas], and PCR products were analysed on 2% agarose gels.
2.7. Isolation of colonic mucosa and extraction of proteins for SDS-PAGE/Western blot and multiplex ELISA analysis
Mucosal protein extraction was performed as described previously.33 At 4°C, the intestinal mucosa [jejunum, ileum and colon] were scraped, snap-frozen in liquid N2, and stored at −80°C for the remaining experiments. Frozen tissues were homogenized in homogenization buffer (50 mM Tris-HCl [pH 7.2]) containing Na3VO4 and a protease inhibitor mixture [Sigma-Aldrich] using an Omni TH homogenizer [Omni International]. Following sonication, the homogenate was centrifuged at 2000 g for 10 min. Supernatants were collected as total mucosal proteins, and protein concentrations were measured using the Bio-Rad Protein Assay. Protein extraction, sodium dodecyl sulfate polyacrylamide gel electrophoresis [SDS-PAGE] and Western blots were performed as described previously.33
2.8. Biometric multiplex cytokine profiling
Multiplex enzyme-linked immunosorbent assay [ELISA] measurement of cytokines/chemokines on extracted mucosal proteins was performed using a Bio-Plex 200 System+HTF [Bio-Rad] as described previously.34 Cytokine data were analysed by Bio-Plex Manager software [Bio-Rad]. Multiplex magnetic beads include 20 analytes: IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12[p40], IL-13, IL-17, interferon-γ [IFN-γ], TNF-α, granulocyte-macrophage colony-stimulating factor [GM-CSF], interferon-inducible protein 10 [IP-10], keratinocyte-derived chemokine [KC], MIG, MCP-1, VGEF, FGF-Basic, monokine-induced by IFN-γ and MIP-1α [Invitrogen].
2.9. Statistical analysis
Statistical analysis within and between each group was done using a Student t-test, and descriptive results are presented as mean ± SD. p values <0.05 were considered statistically significant.
3. Results
3.1. IL-6−/−/IL-10−/− mice exhibit more severe colitis with much earlier disease onset than IL-10−/− mice
Contrary to our expectations, while IL-10−/− mice hosted in our animal facility generally started to develop macroscopic colitis when ~2 months old, IL-6−/−/IL-10−/− mice developed bowel disease as early as 1 month [earlier-onset] [Figure 1]. The IL-6−/−/IL-10−/− mice exhibited much more severe colitis and more aggressive disease progression than IL-10−/− mice, as evidenced by accelerated weight loss, the earlier appearance of diarrhoea/loose faeces, and thickening of the bowel wall [Figure 1A, B]. In general, mortality among >250 IL-6−/−/IL-10−/− mice was ~1.2% by the end of 3 months. We did not know the overall lifespan of the double knockout mice.
![IL-6−/−/10−/− mice exhibit more severe and earlier onset colitis in both colon and small bowel compared to IL-10−/− mice. [A] Morphology of the entire intestine of mice [1 month old]: inflammation is clearly visible in both colon and small bowel of IL-6−/−/10−/− mice, as indicated by irregularly loose stools in the colon, wall thickening of both colon and small bowel, and rectal prolapse [see Supplementary Figure 3]. [B] Half of 1-month-old mice display macroscopic disease, compared to 0% in the age-matched IL-10−/− mice. [C] As a reliable indicator of macroscopic disease, increased fresh colon weight was analysed in all age groups of IL-6−/−/10−/− mice studied [1, 2 and 3 months old] when compared to WT and IL-10−/− mice. At least ten mice were used in each experimental group.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0001.jpeg?Expires=1749481952&Signature=Ie-0H4Il3hWs9g-Qdbt4GLYf0a~f2JuWNJNlJwPG0hxsp0TU7WRfHJDNwcWa46xAgm17j0UWfyiN2squ4kxE1C-qeyC682CxVmDNc10UpjFYJqcX4Mo22uiZ1HcQIVvKZn8vJRZYLSrCMJ6YzjkponuWzHA7~cCQMVg67i2c9zpeGQf9VxEV1Q~2QaVKRyhLWHfD0G9iBUEJqAQ--aJRb1RRpDtCNUCA2ul9Qza75uzBV3VW7ANwqrxyZBvAs~8mdZlgqjx07AFXIhkUi9LubwwGG4RSirIDEsh9CMK31vzwTwfvZQkwdIlxEHf~byYWR5p9TL2TRsz~xGqaU920KQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
IL-6−/−/10−/− mice exhibit more severe and earlier onset colitis in both colon and small bowel compared to IL-10−/− mice. [A] Morphology of the entire intestine of mice [1 month old]: inflammation is clearly visible in both colon and small bowel of IL-6−/−/10−/− mice, as indicated by irregularly loose stools in the colon, wall thickening of both colon and small bowel, and rectal prolapse [see Supplementary Figure 3]. [B] Half of 1-month-old mice display macroscopic disease, compared to 0% in the age-matched IL-10−/− mice. [C] As a reliable indicator of macroscopic disease, increased fresh colon weight was analysed in all age groups of IL-6−/−/10−/− mice studied [1, 2 and 3 months old] when compared to WT and IL-10−/− mice. At least ten mice were used in each experimental group.
Macroscopic disease was observed in approximately 50% of IL-6−/−/IL-10−/− mice at 1 month, and 100% at 2 months, whereas IL-10−/− mice exhibited no macroscopic disease at 1 month and only 19% mice exhibited macroscopic disease at 2 months [Figure 1B]. Remarkably, IL-6−/−/IL-10−/− mice also displayed visible macroscopic disease in the small bowel [both ileum and jejunum] with thickening of the bowel wall that was not observed in IL-10−/− mice [Figure 1A]. Further examination of the colon of both IL-6−/−/IL-10−/− mice and IL-10−/− mice revealed that there was a progressive increase in colon weight in IL-6−/−/IL-10−/− mice starting at 1 month old [Figure 1C], when there was no difference in colon weights between WT and IL-10−/− mice.
3.2. Crypt lesion/abscess, loss of goblet cells and leukocyte infiltration were observed in the intestine of 1-month-old IL-6−/−/IL-10−/− mice
Consistent with the macroscopic disease phenotype, histologically IL-6−/−/IL-10−/− mice exhibited much more visible leukocyte infiltration in both colon and small bowel [jejunum and ileum], as well as more frequent prolapse, as early as 1 month old [Figure 2; Supplementary Fig. 3]. In addition, bowel alteration [abnormality] was characterized by focal crypt lesions, goblet cell loss and submucosal oedema. The histological score was significantly elevated in the colon of IL-6−/−/IL-10−/− mice compared with 2-month-old IL-10−/− mice [Figure 2B]
![Intestinal inflammation appears in both colon and small bowel [jejunum and ileum] of IL-6−/−/10−/− mice as early as 1 month old. Microscopic evaluation by H&E staining of various segments of the gut in 1-month-old mice shows lost of crypts, leukocyte infiltration and thickening of the bowel wall. [A] H&E staining of the colon [1-month-old mice]. [B] Histological scores [2-month-old mice] as an evaluation of disease activity. [C] H&E staining of rectum prolapse [1-month-old mice]. [D] H&E staining of jejunum and ileum [1-month-old mice]. Representative images from at least ten mice per group are shown.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0002.jpeg?Expires=1749481952&Signature=3egSBfDjafhEseVxVQGs-MB4LEXQUk9TeL-92-vBbXKs7ANyBhIQicFVxSME8atC6GI84AeWIiI-lp4WOlAy3L57IH7rufnHx4QqkSeiZMHqrYpHgjWGWqzJnqTgkFWn90p5d8QvGHKmhATMGJT9sT1q3MGVVZUzuy~Tgawz7Vb8BEa6oEzPBUz6qRfR~jYVdMnquSoa8RVV5mhyzYCjzKPR1aF61T6MIUtdeDlX8IZ-eGLSy4LEzCz6bG-sOxue5qWawOhTh8VAM6VQM4vIgKcJYIbUYFjW03-Qxo~oCCjMe~VXVQf1E1kYFRm4V9tQrLp~XXX9ueuWtGTdgpbIaQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Intestinal inflammation appears in both colon and small bowel [jejunum and ileum] of IL-6−/−/10−/− mice as early as 1 month old. Microscopic evaluation by H&E staining of various segments of the gut in 1-month-old mice shows lost of crypts, leukocyte infiltration and thickening of the bowel wall. [A] H&E staining of the colon [1-month-old mice]. [B] Histological scores [2-month-old mice] as an evaluation of disease activity. [C] H&E staining of rectum prolapse [1-month-old mice]. [D] H&E staining of jejunum and ileum [1-month-old mice]. Representative images from at least ten mice per group are shown.
3.3. IL-6−/−/IL-10−/− mice show extensive systemic inflammation that was not seen in the IL-10−/− mice
IL-6−/−IL-10−/− mice showed dramatic systemic inflammation characterized by splenomegaly, with histology revealing increased granulopoiesis, as well as increased liver and kidney weights compared to IL-6−/− and IL-10−/− mice; IL-10−/− mice showed only a slightly enlarged spleen and liver, while IL-6−/− mice exhibited normal sized spleen, liver and kidney when compared to WT mice [Figure 3A–D]. The splenomegaly was primarily caused by a large expansion in the numbers of granulocytes [Figure 3B]. Histological analysis of the spleen showed marked granulopoiesis in red and white pulp with a red/white pulp ratio ≥ 5 [Figure 3B], which was not significantly altered in the spleen of IL-10−/− mice. Haematology profiles further indicated systemic inflammation in the IL-6−/−/IL-10−/− mice [5 months old], including leukocytosis, neutrophilia and monocytosis [Figure 3E]. These double deficiency mice were also anaemic with significantly lower hemoglobin [Hb] and hematocrit [HCT] compared with IL-10−/− mice [Figure 3E].
![IL-6−/−/10−/− mice exhibited systemic inflammation. [A,B] Splenomegaly with histology of increased granulopoiesis. [C,D] Increased size of the liver and kidney. [E] Haematology profiles [mean ± SD] indicate leukocytosis, neutrophilia and anaemia in IL-6−/−/10−/− mice. Representative data from at least ten 5-month-old mice per group are shown. When comparing between WT, IL-10−/− and IL-6−/−/10−/− mice, the difference in each item in E is statistically significant [p < 0.05], except for the comparison of leukocytes between WT and IL-10−/− mice. WP, white pulp.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0003.jpeg?Expires=1749481952&Signature=SBQJFQuM~i143~LaBQK-ypBPm~CJ1h644wRwpfD0c8TEEpwfx92R6aHsbPr~figWpJ2DuvkmnKl-hXvkbwvshib0l5Y31JBmNtnZpa0JQG0gNjdGJYFfnWFYM9FJDG9vPLN3a~hohLIJrQbXeKoGXrjacSqIufwthCaPAIM8QH5MX2hIvaKjKO93KatXgVu7l2lu4aFS5cQdoL0u6s31GwSHjxA~WB6iSZIi3UMqNUe9wySJfE~d0XkeALwPW~DE4xPBHpv7-y5a6gLnhYipgAsAhoTCHdgQ-xQ6V70hlAxOEOMvOB73jcojBcgV9Td7KOXH-TAJVZPQTwbqvsVvGw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
IL-6−/−/10−/− mice exhibited systemic inflammation. [A,B] Splenomegaly with histology of increased granulopoiesis. [C,D] Increased size of the liver and kidney. [E] Haematology profiles [mean ± SD] indicate leukocytosis, neutrophilia and anaemia in IL-6−/−/10−/− mice. Representative data from at least ten 5-month-old mice per group are shown. When comparing between WT, IL-10−/− and IL-6−/−/10−/− mice, the difference in each item in E is statistically significant [p < 0.05], except for the comparison of leukocytes between WT and IL-10−/− mice. WP, white pulp.
Another small fraction [<10%] of IL-6−/−/IL-10−/− mice developed an extraintestinal manifestation, namely ulcerative dermatitis [Supplementary Figure 3] as early as 1 month old; those mice were terminated early to prevent further suffering.
3.4. Other phenotypes associated with IL-6−/−/IL-10−/− mice
In addition to the overall systematic and intestinal inflammation described above, a small fraction [3.75%; three out of 80] of IL-6−/−/IL-10−/− mice exhibited development retardation, with a body mass about one-third [or less] of the age-matched WT or IL-10−/− counterparts [Supplementary Figure 1]. Moreover, 6.67% of IL-6−/−/IL-10−/− mice developed cataract-like phenotypes in one or both eyes by the age of 3 months [Supplementary Figure 2]. Finally, by the age of 5 months old, most of the IL-6−/−/IL-10−/− mice had developed various degrees of pancreatitis [Supplementary Figure 4].
3.5. IL-6/IL-10 double-deficiency led to down-regulation of both FOXP3 and TGF-β in the colon, but not in the thymus or mesenteric lymph nodes
Surprisingly, we noted a dramatic down-regulation of both FOXP3 and transforming growth factor-β [TGF-β] in the colonic mucosa, but not in the mesenteric lymph nodes or thymus, of IL-6−/−/IL-10−/− mice [Figure 4A, B]. FOXP3, but not TGF-β, was also down-regulated in the spleen of IL-6−/−/IL-10−/− mice. Our data appear to conflict with the general concept that IL-6 inhibits FOXP3 expression to suppress Treg development and promote Th17 differentiation.
![PCR analysis reveals a significantly suppressed FOXP3 expression in the colon and spleen [but not mesenteric lymph nodes or thymus] of IL-6−/−/10−/− mice compared to that in IL-10−/− mice. [A] PCR products. [B] Qualitative analyses of selected genes (only genes whose expression was significantly altered [marked by arrows] are shown). Double arrows indicate the size-shifts [two different sizes] of the PCR products of RORγt in IL-6−/−/10−/− mice, compared to the single-sized band in IL-10−/− mice. [C] Expression of several major cytokines. Representative data from at least 4–6 mice per group [3 months old] are shown. *p < 10–4.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0004.jpeg?Expires=1749481952&Signature=M~5bCn~SSze9zG6oYg7JfANnFX1YIL~yFNAhSbwwZHQO4z6puvCVuWOnfXcB11f51l3gdyiUDFJ7p5h79U~XZjMgtfkQQfIHcTmZJtNpenxe1kFqC4z1la9m8o4IhZsfxYwiECvXSxstNlkHSxqEq~cbmbd4JLP8lpS2FXZzXMMgTyRFkph430NQhTGonp1B6dCrTYC9oOpX7LAEIu2Q1p2peJ20nbMaFVJMS2b8s-y5EP5LvLJAooTdJfOy9p1g4PfxmsTRA81YIxkI0UBXx81ulxd9nhitf06CnITly8ZQGTi-cAEheDVij1qgi3WdRA99LTI~~CC4UGHwI4sRlw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
PCR analysis reveals a significantly suppressed FOXP3 expression in the colon and spleen [but not mesenteric lymph nodes or thymus] of IL-6−/−/10−/− mice compared to that in IL-10−/− mice. [A] PCR products. [B] Qualitative analyses of selected genes (only genes whose expression was significantly altered [marked by arrows] are shown). Double arrows indicate the size-shifts [two different sizes] of the PCR products of RORγt in IL-6−/−/10−/− mice, compared to the single-sized band in IL-10−/− mice. [C] Expression of several major cytokines. Representative data from at least 4–6 mice per group [3 months old] are shown. *p < 10–4.
3.6. Effects of IL-6/IL-10 double-deficiency on the expression of other transcription factors of CD4 T cells
We also examined the expression of other transcription factors including T-bet [Th1], GATA-3 [Th2] and RORγt [Th17] [Figure 4A, B]. Together with FOXP3, these transcription factors influence and determine largely T cell differentiation and play a critical role in the immune response during the inflammatory process. T-bet was found to be up-regulated only in the thymus [not colon, spleen or mesenteric lymph nodes] of IL-6−/−/IL-10−/− mice when compared to that of IL-10−/− mice [Figure 4A, B]. Intriguingly, we observed two different sizes of the PCR products of RORγt in the spleen of IL-6−/−/IL-10−/− mice, compared to the single-sized band in that of IL-10−/− mice [Figure 4A; indicated by the double-arrows]. The nature of the larger-sized band of RORγt is unknown. There was no alteration of RORγt expression in other tissues tested. There was no change in GATA-3 expression in any of the tissues either.
3.7. Evaluation of changes in the expression of major cytokines and CTLA-4 by RT-PCR
Among eight major cytokines examined by reverse transcription PCR [RT-PCR], representing four major pathways of CD4+ T cells, the Th1 [IFN-γ, IL-2], Th2 [IL-4], Th17 [IL-17A, IL-17G] and Th22 [IL-22] pathways, only two cytokines, IL-1β and IL-4 exhibited elevated expression in the colonic mucosa of IL-6−/−/IL-10−/− mice compared to that of IL-10−/− mice [Figure 4C]. Cytotoxic T-lymphocyte antigen 4 [CTLA-4], a co-inhibitory molecule expressed by activated T cells and particularly FOXP3+ Treg cells,35 was drastically diminished [undetectable] in the colonic mucosa of IL-6−/−/IL-10−/− mice compared to that of IL-10−/− mice [Figure 4C].
3.8. Multiplex pro-inflammatory cytokines and chemokines were elevated in the IL-6−/−/IL-10−/− mice
To evaluate the alteration of more cytokines and chemokines as a consequence of IL-6 deletion, we measured a panel of 20 cytokines/chemokines in colonic mucosa by multiplex ELISA, covering a broad spectrum of immune and inflammatory mechanisms. IL-6−/−/IL-10−/− mice exhibited significantly elevated levels of IL-1β, IL-4, IL-12, TNF-α, MIG [CXCL9], MCP-1 [CCL2], GM-CSF and FGF-Basic when compared to IL-10−/− mice [Figure 5].
![Multiple pro-inflammatory cytokines and chemokines were elevated in IL-6−/−/10−/− mice compared to those in IL-10−/− mice of the same age [3 months old]. Cytokines and chemokines in colonic mucosa were measured and analysed by multiplex ELISA that simultaneously profiled 20 cytokines and chemokines. Eight cytokines or chemokines were significantly elevated in the colonic mucosa of IL-6−/−/10−/− mice: TNFα, IL-1β, IL-4, IL-12, MCP-1, MIG, GM-CSF and FGFBase. At least six mice were used per group.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0005.jpeg?Expires=1749481952&Signature=ivaFPEyaAT6OXiw1srrspFwEgoSn10y4cG6jYlf3ebBHpzlWucgb6cAQEXNwO1i7BDyBWrjkDcjm3ex8falo8HKXL5RorSTiawMMBo-e35UmcIuqf5myejUqS0s53tAnllwxFL4ifDa0mEo7gihskfHhckZU7h6YpzxW3Ma5dL1crNnzJCSFDp5~rkaPsmeTqSUzbJaKBDL7tEq7efWLu0oqREWzcuMw5ZORBx9CKaojDUyAV8qeHkIdpKoTqVdaj5B0dOjpZ6jsZyOsXT6cE5vIySHoLjsuv8KyNqIPpL1sbimh5EYXg9ItTGNecMDcQo91rINO1mJ91-K6wOztPA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Multiple pro-inflammatory cytokines and chemokines were elevated in IL-6−/−/10−/− mice compared to those in IL-10−/− mice of the same age [3 months old]. Cytokines and chemokines in colonic mucosa were measured and analysed by multiplex ELISA that simultaneously profiled 20 cytokines and chemokines. Eight cytokines or chemokines were significantly elevated in the colonic mucosa of IL-6−/−/10−/− mice: TNFα, IL-1β, IL-4, IL-12, MCP-1, MIG, GM-CSF and FGFBase. At least six mice were used per group.
3.9. iNOS was elevated in the colonic mucosa of IL-6−/−/IL-10−/− mice
Inducible nitric oxide synthase [iNOS] has recently been demonstrated to contribute to the very early onset of IBD.36 Because IL-6−/−/IL-10−/− mice develop the much earlier onset of colitis than IL-10−/− mice, we investigated whether iNOS might play a role in this phenomenon. Indeed, while iNOS was markedly increased in IL-10−/− mice compared to WT mice, it was even more elevated in IL-6−/−/IL-10−/− mice than in IL-10−/− mice [Figure 6].
![iNOS expression is highly elevated in IL-6−/−/10−/− mice compared to that of IL-10−/− mice. All mice used in this experiment were 3 months old. The colonic mucosa of WT, IL-10−/− and IL-6−/−/IL-10−/− mice were analysed by SDS-PAGE and Western blot for protein expression levels of IL-6, iNOS and actin [loading control]. Representatives of three independent experiments are shown.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/ecco-jcc/14/6/10.1093_ecco-jcc_jjz176/1/m_jjz176f0006.jpeg?Expires=1749481952&Signature=g9pP~2hg8vY4saXYFIKT3m8yiGvEoxBes2jRx1knC8VUbJ9p4Rs~PAuutIJUKY-GTwt~aCONIwQB3dykSpbqsQxhSwwOkXspMXJtJAN0cG-bcwo6EPe0eOBCCDgH74SnyT69Je6xt51Qa570rAjDJW6CEDxzFJi73Jh19EkTcdVkAL9R41g30N0S8c8hfg6oUPds7Ydn-gsBLFX8-EEu6-brNOyF1LThajlPrsrhL6t4M6fhOvfz4AR~RJtNg9NaXw8wR3MwbyAhmAQOYmLgBFv6VRw~lbBjBW6KHt8TGXxGOyRW7mGxIn4aKcmWSbK2UF2Jn00TB3Bsl0a8f77eqg__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
iNOS expression is highly elevated in IL-6−/−/10−/− mice compared to that of IL-10−/− mice. All mice used in this experiment were 3 months old. The colonic mucosa of WT, IL-10−/− and IL-6−/−/IL-10−/− mice were analysed by SDS-PAGE and Western blot for protein expression levels of IL-6, iNOS and actin [loading control]. Representatives of three independent experiments are shown.
4. Discussion
IL-6 is a pleiotropic cytokine. It plays an important role in the connection between the innate and adaptive immune responses by promoting the specific differentiation of naive CD4+ T cells. Consistent with most previous reports,31,32,37 our current study demonstrated that IL-6 expression was up-regulated in the colonic mucosa of IL-10-deficient mice, suggesting that IL-6 may at least in part exert a pro-inflammatory role in the development of spontaneous colitis in IL-10−/− mice.
Surprisingly, however, we found that IL-6 deficiency in IL-10-knockout mice not only aggravated the spontaneous colitis but also led to an early onset of the disease when compared to IL-10−/− controls. Moreover, IL-6-deficiency in IL-10−/− mice led to extended inflammation in the small bowel [both jejunum and ileum] at 1 month old, a phenomenon that has not been seen in IL-10−/− mice. Our findings are entirely at odds with previous relevant reports regarding IL-6. It is possible that by mating knock-out animals to produce this combination genotype, we have produced a setting in which IL-6 functions have been ‘isolated’ and so critical roles can be more clearly determined. In fact, the anti-inflammatory property of IL-6 has been reported in the literature. First, IL-6 was shown to induce several factors with anti-inflammatory properties, including IL-10, soluble TNF receptors and IL-1R antagonists.38–40 Second, the anti-inflammatory effects of IL-6 have been shown in multiple inflammatory conditions, including asthma and other inflammatory lung diseases41, acute kidney injury [AKI],42 and in the intestine with intestinal ischaemia/reperfusion injury.26 Third, there is evidence that IL-6 protects intestinal epithelial cells from apoptosis during acute DSS-induced injury,27,43 and promotes intestinal epithelial proliferation and repair in the early stages of murine models of acute intestinal injury induced by either biopsy or bacteria.44
Because IL-6 and IL-10 are mainly involved in enteritis by regulating T cell immunity, we further examined the expression of major transcription/regulatory factors in T cells, including Th1, Th2, Th17 and Treg cells.45,46 The regulatory function of T cells is known to be critical in maintaining immune homeostasis in the gut.47 FOXP3 and TGF-β are two key factors for Treg-mediated regulatory function, and Treg cells integrate environmental signals or factors to inhibit specific types of inflammation. IL-6 and IL-10 are two of the major cytokines in regulating inflammation. On the one hand, IL-6 inhibits TGF-β-induced production of FOXP3+ Treg cells during inflammation.48 On the other, ablation of the IL-10 receptor in Treg cells resulted in selective dysregulation of Th17 cell responses and colitis.49 Moreover, the combination of IL-6 and TGF-β is essential for the differentiation of Th17 from naive CD4+ T cells.50 These data suggest that IL-6 and the interplay between IL-6 and IL-10 play a critical role in regulating the balance between Treg and Th17 cells in autoimmune diseases and chronic inflammatory diseases.51 Unexpectedly, our experiments demonstrate that complete deletion of IL-6 significantly down-regulated both FOXP3 and TGF-β in the colonic mucosa [but not the thymus or mesenteric lymph nodes] of IL-6−/−/IL-10−/− mice, suggesting that the differentiation of immunosuppressive FOXP3-Treg cells in the gut is inhibited. This may at least in part explain the more severe gut inflammation in IL-6−/−/IL-10−/− mice. While the obvious discrepancy between our data and previous reports remains to be explored, it may arise from alteration of the gut mucosal microenvironment as a result of IL-10 deficiency, so that the usual synergy among IL-6 and other regulatory/inflammatory factors in the immune network become disordered.
Given that the development and resolution of inflammation are mediated by the interplay among a host of anti- and pro-inflammatory cytokines and chemokines, to further explore what other cytokines or chemokines are involved in this interplay, we used multiplex ELISA technology to evaluate 20 cytokines/chemokines in the colonic mucosa of IL-6−/−/IL-10−/− mice. The absence of IL-6 in IL-10-deficient mice led to elevations of multiple cytokines/chemokines, including IL-1β, IL-4, IL-12, TNFα, MCP-1 and MIG (but not IFN-γ [Th1], IL-17A and IL-17G [Th17], or IL-22), compared with IL-10-deficient mice. RT-PCR data also showed regulations of IL-1β and IL-4. Based on these results, along with the FOXP3/TGF-β data described above, we speculate that two potential proinflammatory events occur in IL-6−/−/IL-10−/− mice: [1] a pro-inflammatory Th2 pathway is enhanced through a self-feedback loop mediated by IL-4; 52 and [2] because little effect was observed on the Th1 or Th17 pathways, it is likely that IL-6 plays a regulatory role when IL-10 is absent by promoting the differentiation of immunosuppressive Treg cells.
Because the IL-6−/−/IL-10−/− mice exhibited a much earlier onset of colitis than IL-10−/− mice, we explored the mechanism behind this observation. NOS2, which encodes iNOS, was found to associate strongly with susceptibility to very early onset IBD [VEO-IBD].36 Therefore, we first evaluated and compared the expression of iNOS in the colonic mucosa of IL-6−/−/IL-10−/− and IL-10−/− mice. We found that iNOS was dramatically elevated in both mice when compared to that in the WT controls. Moreover, IL-6−/−/IL-10−/− mice expressed three-fold more iNOS compared with IL-10−/− mice.
iNOS is an inducible enzyme known to be up-regulated in response to inflammation and injury.53 Early studies found a high expression of iNOS in the inflamed colonic mucosa in patients with IBD while it was undetectable in uninflamed epithelium.54 A later study confirmed that excessive iNOS-derived nitric oxide [NO] production is correlated with disease activity and tissue damage in IBD.36 iNOS expression in intestinal epithelial cells has been shown to be induced by several major pro-inflammatory cytokines, including IL-1β, IFN-γ and TNF-α.55–57 Enhanced expression of TNF-α in the colon was required for increased iNOS expression57. IL-10, by contrast, down-regulated the expression of TNF-α and iNOS, and subsequently suppressed colitis and cancer.57 These data provide a strong molecular basis for our explanation regarding the current observations on iNOS and cytokine expression: [1] in the absence of IL-10, as in the case of IL-10-deficient mice, iNOS was dramatically up-regulated; and [2] IL-6−/−/IL-10−/− mice expressed a much higher level of iNOS than IL-10−/− mice as a result of elevated levels of the pro-inflammatory cytokines IL-1β and TNF-α.
The relationship between enhanced expression of iNOS and the increased levels of other cytokines and chemokines, including IL-4, IL-12, and MCP-1 and MIG, in IL-6−/−/IL-10−/− mice is not known and warrants further investigation. Most of these cytokines/chemokines are also over-expressed in the inflamed colonic mucosa of IL-10−/− mice than in WT mice,32 suggesting a role in the pathogenesis of colitis. Therefore, the more elevated expression of these pro-inflammatory factors in IL-6−/−/IL-10−/− mice may contribute further to the more severe gut inflammation and earlier onset of colitis.
In addition to severe intestinal inflammation, IL-6−/−/IL-10−/− mice also showed extensive systemic inflammation, including splenomegaly/pancreatitis, ulcerative dermatitis, leukocytosis, neutrophilia and monocytosis, and in some cases growth retardation and cataract-like phenotypes. These data confirmed that IL-6 also plays an essential role in the homeostasis of multiple organ systems, including anti-inflammatory functions [discussed above] and in promoting wound healing/tissue repair.11,12,58 The reasons for the growth retardation and cataract-like phenotypes in the small fraction of IL-6−/−/IL-10−/− mice are unclear; one possible cause may be related to the severe systemic inflammation.
Together, the current study provides a more comprehensive understanding of IL-6 in the pathogenesis of IBD and systemic homeostasis. Our data suggest that a global or complete blockade of IL-6 signalling may lead to unintended side effects as a result of its homeostatic or regenerative/anti-inflammatory functions in immune-compromised IBD patients [analogous to the immune-compromised IL-10/IL-10R-deficient mice]. Although the animal model remains critical and valuable for IBD study and the phenotype of our double knockout mice used is highly homogeneous, it does not reflect the human disease or the effects of monoclonal antibody blockade in the clinic. Therefore, further investigation is warranted. Nevertheless, the implication of the study is that caution is needed when blocking IL-6 in IBD patients, particularly those with a defective IL-10/IL-10R signalling pathway that results in severe and/or very early-onset IBD [VEO-IBD].59 More specifically, our study suggest monitoring levels of IL-10 or mutational status of IL-10/IL-10R in IBD patients before choosing anti-IL-6 as a therapeutic agent. Strategies by which either preserving functional levels of IL-6 homeostatic/anti-inflammatory functions or targeting the IL-6 trans-signalling that is specifically responsible for tissue inflammation [pro-inflammatory function] are more desirable for developing IL-6 pathway-based IBD therapy. Targeting gp130, a membrane protein that can form a complex with soluble IL-6R and specifically activate pro-inflammatory responses without compromising the general homeostatic functions, is perhaps the first example of such a strategy.60,61
Funding
This work was supported by grants from the National Institute of Health [DK077064 and P30DK089502] to X.L. And also supported by grants from the National Natural Science Foundation of China [Nos. 81270468, 81870391] to M.Y.
Conflict of Interest
None declared.
Author Contributions
MY performed most of the experiments, and played a key role in preparing the manuscript; MEJ performed a significant part in the experiments. LL, YS, YD, CC, MS and JZ contributed to animal colony management and/or some specific experiments. SRB and ML are long-time collaborators on XL’s project. XL was involved in project inception, design and supervision, and manuscript writing and revision.
Acknowledgments
We thank Dr Bin Gao, chief of Laboratory of Liver Diseases, National Institute on Alcohol Abuse and Alcoholism, NIH, Bethesda, for providing the IL-6−/− mice, which were used to crossbreed with IL-10−/− mice to generate IL-6−/−/IL-10−/− double-deficient mice. We also thank the Phenotyping Core, Johns Hopkins School of Medicine, Department of Molecular and Comparative Medicine, as well as The Image Core of the Johns Hopkins Digestive Diseases Basic and Translational Research Core Center funded by NIH/NIDDK. We thank Dr Alan Hofmann for his comments on, and editing of, the manuscript.
