Disclosures: A.C.W.V., M.E.W., I.A., M.D., A.P.V.: no conflicts of interest; G.dH.: consultant for Centocor, Shire, Teva, BMS, and Glaxo SmithKline; S.V.: Grants/Research Support: UCB; Consultancy: Astra-Zeneca, Ferring, Pfizer, Centocor, MSD; Speakers Bureau: MSD, Abbott, Ferring, UCB; Advisory Committee: Shire, Ferring; P.R.: Grant support, lecture fees, and consulting fees from Abbott, UCB, Centocor, and Schering-Plough; G.vdB.: Grants/research support from MSD, Ferring, and Guiliani; D.W.H.: Grants/research support and is a consultant for Abbott, MSD, UCB, Guiliani, Ferring, Serono, AstraZeneca, Falk, and Tramedico.
Regulatory macrophages induced by infliximab are involved in healing in vivo and in vitro†
Article first published online: 20 SEP 2011
Copyright © 2011 Crohn's & Colitis Foundation of America, Inc.
Inflammatory Bowel Diseases
Volume 18, Issue 3, pages 401–408, March 2012
How to Cite
Vos, A. C. W., Wildenberg, M. E., Arijs, I., Duijvestein, M., Verhaar, A. P., de Hertogh, G., Vermeire, S., Rutgeerts, P., van den Brink, G. R. and Hommes, D. W. (2012), Regulatory macrophages induced by infliximab are involved in healing in vivo and in vitro. Inflamm Bowel Dis, 18: 401–408. doi: 10.1002/ibd.21818
- Issue published online: 13 FEB 2012
- Article first published online: 20 SEP 2011
- Manuscript Accepted: 8 JUN 2011
- Manuscript Received: 31 MAY 2011
- regulatory macrophage;
Regulatory macrophages play an important role in wound healing and gut homeostasis and have antiinflammatory properties. Induction of this cell type (Mψind) by the anti-tumor necrosis factor (TNF) antibodies, infliximab and adalimumab, has recently been shown in vitro. Also, the superiority of infliximab/azathioprine combination therapy over infliximab or azathioprine monotherapy has recently been established, but the mechanism behind this remains unclear. The aim of this study was to examine the induction of regulatory macrophages in patients with and without mucosal healing in response to infliximab. In addition, we studied the effect of infliximab/azathioprine combination treatment on the differentiation and function of regulatory macrophages.
Inflammatory bowel disease (IBD) patients (n = 10) underwent endoscopy before and after first infliximab treatment. Immunohistochemical staining of CD68 and CD206 was performed in all patients. Mixed lymphocyte reactions (MLRs) were treated with infliximab, azathioprine, or both. Macrophage phenotype was evaluated by flow cytometry and inhibition of T-cell proliferation was measured in a secondary MLR containing macrophages and third-party lymphocytes.
A significant induction of regulatory macrophages was observed in patients with mucosal healing after treatment with infliximab; this induction was absent in patients without mucosal healing. In addition, Mψind have the ability to induce wound healing in an in vitro model, further suggesting a key role for infliximab-induced macrophages in mucosal healing. Upon infliximab/azathioprine combination treatment, an increased number of regulatory macrophages was observed. These macrophages also displayed stronger immunosuppressive properties than macrophages induced by infliximab monotherapy.
These data show that regulatory macrophages may be involved in mucosal healing and provide a rationale for the superiority of infliximab/azathioprine combination treatment observed in the clinic. (Inflamm Bowel Dis 2012;)
Crohn's disease (CD) and ulcerative colitis (UC) are chronic inflammatory bowel diseases (IBD) of unknown etiology resulting in loss of tolerance toward the mucosal flora. Traditionally, therapy to control IBD was aimed at reducing symptoms, but at the present time treatment is more and more focused on inducing mucosal healing and altering the disease course. The introduction of anti-tumor necrosis factor (TNF) agents in the 1990s has been an important breakthrough to accomplish this. Anti-TNF agents have been shown to induce mucosal healing, reduce steroid dependency, reduce the risk for surgery and hospitalization, and to improve the patient's quality of life.1–5 However, not all anti-TNF agents that have been introduced in the clinic appeared effective. For instance, the TNF receptor fusion protein etanercept6 and the IgG4 anti-TNF antibody CDP5717 lacked clinical efficacy although these agents have the ability to neutralize soluble TNF. This observation raised the question why some anti-TNF agents are effective and others are not. We have recently shown that in order for an anti-TNF to inhibit T-cell proliferation in vitro, the compound needs to bind to membrane-bound TNF on activated T cells and possess an Fc region to interact with Fc receptors.8 Upon this binding, a distinct macrophage subset is induced (Mψind) with immunosuppressive capacities, including the production of antiinflammatory cytokines and inhibition of T-cell proliferation. Furthermore, it has been shown that in vitro-derived regulatory macrophages reduce colonic inflammation in mice9 and inhibit proliferation of activated T cells.10 On top of this, regulatory macrophages play a crucial role in wound healing,11 and therefore the induction of this cell type may be of particular interest in the treatment of IBD and induction of mucosal healing.
Recently, a large randomized, placebo-controlled trial demonstrated the superiority of infliximab/azathioprine combination treatment over infliximab or azathioprine monotherapy in inducing and maintaining remission and mucosal healing in patients with CD.12 The exact mechanism underlying this effect is unknown, but it likely results from the additional effect of a second immunosuppressive. Also, it is possible that azathioprine suppresses infliximab-induced immunogenicity, thereby increasing infliximab efficacy. However, the specific mechanisms involved in this superiority and in mucosal healing in general are so far not completely understood.
In this study we aimed to investigate the induction of regulatory macrophages in patients responding to infliximab therapy versus those who did not respond. Also, we examined the wound-healing capacity of infliximab-induced regulatory macrophages in vitro. In addition, we examined the effect of infliximab/azathioprine combination treatment on the induction and function of regulatory macrophages in vitro.
MATERIALS AND METHODS
The patient material used in this study was obtained from University Hospital Gasthuisberg in Leuven (ClinicalTrials.gov number NCT00639821). All patients gave written informed consent. Ten patients with active IBD (four UC, six CD), refractory to corticosteroids and/or immunosuppression, were examined. The patients underwent endoscopy with biopsies from affected bowel (colon for UC and colonic CD, and ileum for ileal CD) within a week prior to the first infliximab infusion of 5 mg/kg body weight. Patients received a second endoscopy with biopsies at week 4–6. The biopsies were taken at sites of active inflammation but at a distance of ulcerations. When healing was observed at the second endoscopy, biopsies were obtained in the areas where lesions were present before therapy. The endoscopist was not blinded to treatment.
Next, biopsies were fixed in Carnoy's fixative for up to 5 hours and then dehydrated, cleared, and paraffin-embedded for histological examination and/or immunohistochemistry. Hematoxylin-eosin-stained slides from the paraffin block of each patient were used to score chronic intestinal inflammation using a previously reported scoring system for UC13 and for CD.14 The pathologists were blinded to treatment.
Response to infliximab was assessed 4–6 weeks after the first infliximab treatment and classified as described before.15 Briefly, for colonic CD a response was defined as complete mucosal healing with a decrease of at least 3 points on the histological score.14 For UC a response consisted of a decrease of a Mayo endoscopic subscore of 0 or 1 with a decrease to grade 0 or 1 on the histological score.13 For ileal CD (three patients), patients with a clear improvement of the ulcerations and a decrease on the histological score14 were considered responders. Patients who did not achieve this healing were considered nonresponders, although some of them showed an improvement on the histologic or endoscopic score.
For immunohistochemistry, Mψ2 cells were defined as CD206+/CD68+ cells. Mψ2 were detected using an antihuman CD206 goat polyclonal antibody (AF2534, R&D Systems, Abingdon, UK) and antihuman CD68 mouse monoclonal antibody (clone KP1, M0814, Dako Belgium, Heverlee, Belgium). Briefly, 5-μm-thick sections were cut from the paraffin blocks of Carnoy-fixed endoscopic biopsies from the patients. After drying, deparaffinization, and rehydration, epitope retrieval was performed at high pH (Dako PT Link machine, Dako Belgium). Sections were then washed 3 × 5 minutes (Envision Flex wash buffer, Dako) and Envision Flex Peroxidase-Blocking Reagent (Dako) was applied for 10 minutes at room temperature. After a second wash step, sections were incubated with the antihuman CD206 antibody (R&D Systems, dilution 1:40) or with the antihuman CD68 antibody (Dako, dilution: 1:2000) for 30 minutes at room temperature. Following a third wash step, bound primary antibody was visualized by incubating the slides for 30 minutes with Envision Flex/HRP (Dako) and application of the Envision DAB+ Chromogen (Dako) for 10 minutes at room temperature. After rinsing, the slides were counterstained with hematoxylin, dehydrated, cleared, and mounted. Negative controls (no application of primary antibody) were run together with the test samples. All the stains were evaluated by an experienced pathologist (G.D.H.). Finally, the stains were analyzed using ImageJ software.
Cell Isolation and Culture
Peripheral blood mononuclear cells (PBMCs) from healthy volunteers were isolated by Ficoll Paque density-gradient centrifugation. M1 macrophages were generated as described previously.10 Briefly, monocytes were isolated by Percoll density-gradient centrifugation and cultured in RPMI 1640 containing 10% heat-inactivated fetal calf serum (FCS) for 5 days. CD3-positive T cells were isolated from PBMCs using negative magnetic bead separation (Invitrogen, Paisley, UK). Where appropriate, T cells were activated with αCD3/αCD28 beads (Invitrogen) (1 bead / 5 cells).
Allogeneic Mixed Lymphocyte Reaction (MLR)
PBMCs from two healthy donors were cultured in a 1:1 ratio in RPMI 1640 culture medium to establish an MLR. After 48 hours of activation, cells were treated with infliximab (10 μg/mL), azathioprine (1 μM), or a combination for 5 days. Where indicated, cells were treated with 10 μg/mL certolizumab. Finally, proliferation was measured using a 3H-thymidine incorporation assay.
Isolation of Infliximab-induced Macrophages (Mψind) and Infliximab/Azathioprine-induced Macrophages (Mψind/aza)
Infliximab-induced macrophages (Mψind) and infliximab/azathioprine induced macrophages (Mψind/AZA) were isolated from MLR cultures from healthy volunteers. We have previously shown that upon infliximab treatment a distinct cell population occurs on the FCS/SSC, which is strongly positive for several markers, among which CD206 (a commonly used regulatory macrophage marker) and CD14, and that these cells have immunosuppressive properties. The induction, phenotype, and isolation of this cell subset have been extensively described before.8 In brief, Mψind and Mψind/aza were isolated from the MLR based on the expression of CD14 using CD14 microbeads according to manufacturer's protocol (Miltenyi, Bergisch Gladbach, Germany). Next, cells were counted and cultured in RPMI 1640 containing 10% heat-inactivated FCS.
In Vitro Wound-healing Assay
The wound-healing scratch assay was performed as previously described with a few modifications.16 HCT116 colon epithelial cells were cultured to ≈70% confluence in 6-well plates under standard culture conditions. The plates were marked in order to create a reference point to identify the wound at the same place at different timepoints. Next, a wound was created using a plastic tip and pictures were taken with a phase-contrast microscope. Mψind and proinflammatory type 1 macrophages from healthy volunteers were added to the cell culture (100,000 cells/well). To quantify the degree of wound healing, pictures were taken again after 24 hours and the percentage closure was calculated using ImageJ software.
FACS Analysis of Regulatory Macrophages
MLRs were treated with infliximab, azathioprine, or a combination. After 5 days cells were harvested, washed with FACS buffer, and stained for CD206 or isotype control. Finally, expression was analyzed by flow cytometry using a FACS Calibur (BD) and FlowJo software (Treestar, Ashland, OR)
Results are representative of at least three independent experiments and show means ± standard error of the mean (SEM) unless otherwise indicated. For statistical analysis, one-way analysis of variance (ANOVA) was used followed by a Bonferroni post-test. Results were considered significant when P < 0.05. A paired t-test was used to calculate statistical significance in patients before and after infliximab therapy. Results were considered significant when P < 0.05.
Infliximab Induces Regulatory Macrophages in Responders But Not in Nonresponders
We have shown previously that anti-TNF antibodies induce regulatory macrophages in vitro8 and this cell type plays an important role in wound healing.11 Here we aimed to investigate whether the induction of regulatory macrophages is also observed in vivo in patients responding to infliximab therapy. Biopsies were obtained from six CD patients (three responders, three nonresponders) and four UC patients (two responders, two nonresponders) before (week 0) and after (week 4–6) first infliximab treatment. Patient characteristics are shown in Table 1. Slides from each patient before and after therapy were stained for CD68 and CD206 (also called MRC1 or MMR). CD206 is a commonly used marker for regulatory macrophages, and CD68 is a macrophage marker. Since patients with a response to infliximab usually have a decrease in the number of CD68+ cells, we used the ratio CD206+/CD68+ to identify the induction of regulatory macrophages.
|Pt. No.||UC/CDi/CDc||R/NR||Histological Score Week 0||Histological Score Week 4||Endoscopic Score Week 0||Endoscopic Score Week 4||%CD206/CD68+ Cells Week 0||% CD206/CD68+ Cells Week 4|
|6||CDc||NR||14/16||14/16||Active colitis||Active colitis||8,4||9,6|
|7||CDc||NR||10/16||10/16||Active colitis||Incomplete healing||7,2||32,8|
|9||CDi||R||12/16||5/16||Active ileitis||Incomplete healing||44,5||71,8|
|10||CDi||NR||7/16||8/16||Active ileitis||Active ileitis||26,2||16,9|
A clear and significant increase was observed in the percentage of CD206+ macrophages in patients responding to infliximab induction therapy, indicating that regulatory macrophages were induced (Fig. 1A,B). This induction was observed in each individual responding to infliximab therapy (Fig. 1B, left panel). Importantly, induction of CD206+/CD68+ cells was absent in patients not responding to infliximab therapy (Fig. 1B, right panel). One patient (Fig. 1B, right panel, red line) was considered a nonresponder on histological basis, but did show a partial response on endoscopy. An increase of CD206+/CD68+ cells was observed in this patient.
Taken together, these data show that regulatory macrophages similar to those induced by infliximab in vitro are induced in in vivo in patients responding to infliximab therapy.
Infliximab-induced Regulatory Macrophages Induce Wound Healing In Vitro
It has been shown previously that regulatory macrophages generated in vitro by addition of IL-4 and IL-13 exhibit wound-healing capacities. The data described above show an increased presence of macrophages displaying a regulatory phenotype after infliximab therapy. However, the presence of CD206+/CD68+ cells in patients presenting with mucosal healing does not discriminate between infliximab-mediated effects and wound-healing effects induced by another mechanism such as immunosuppression in general. Therefore, we aimed to specifically assess the wound-healing capacity of macrophages induced by infliximab treatment.
To this end, macrophages were isolated from MLR cultures treated with infliximab (regulatory Mψind) or generated from monocyte cultures (inflammatory Mψ1). HCT116 colonic epithelial cells were cultured and a wound was created in the monolayer. Mψ1, Mψind, or control medium was added to the wells and images were taken at t = 0 and t = 24 hours. As expected, Mψ1 did not induce wound healing above the level of control (Fig. 2). In contrast, Mψind showed a clear capacity to enhance wound healing up to 2-fold compared to Mψ1 or medium alone. These data are in line with the results observed in patients responding to infliximab therapy, and further suggest that regulatory macrophages induced by infliximab induce wound healing.
Enhanced Induction and Function of Regulatory Macrophages Upon Infliximab/Azathioprine Combination Treatment In Vitro
Since the induction of Mψind correlates with response to infliximab (Fig. 1), and higher response rates are observed in patients receiving infliximab/azathioprine combination treatment compared to either monotherapy,12 we hypothesized that infliximab/azathioprine combination treatment might enhance the induction of Mψind in vitro. To answer this question we treated mixed lymphocyte cultures with infliximab, azathioprine, or a combination. Azathioprine and infliximab monotherapy inhibited T-cell proliferation in an MLR to the same extent (Fig. 3A). As expected, when cells in an MLR were treated with a combination of infliximab and azathioprine, T-cell proliferation was further inhibited. To investigate whether this observation was a cumulative effect of a second immunosuppressive or whether another mechanism was involved, we studied the effect of combination treatment on the induction of regulatory macrophages. As shown previously,8 anti-TNF antibodies induce a distinct subset of cells, characterized by high forward and high side scatters and expression of CD14 and CD206 when analyzed by flow cytometry. When cells were treated with azathioprine monotherapy, no induction of regulatory macrophages was observed. Strikingly, we found a significantly increased induction of this CD206+ subset upon combination treatment (Fig. 3B,C).
We have previously shown that anti-TNF agents induce regulatory macrophages in an Fc region-dependent manner,8 and that agents lacking the Fc fragment do not induce this cell type. Therefore, we also evaluated the combination azathioprine/certolizumab (lacking the Fc fragment) to examine whether the observed effect of enhanced Mψind/AZA was induced by infliximab in particular or a general effect of TNF-α neutralization. Indeed, the effect was specific for the combination treatment of azathioprine/infliximab, since the induction was absent upon azathioprine/certolizumab or azathioprine/IgG treatment (Fig. 3D).
To further examine the properties of regulatory macrophages induced by infliximab/azathioprine combination treatment (Mψind/AZA), we aimed to investigate their immunosuppressive function. Therefore, we isolated regulatory macrophages based on CD14 expression and cocultured equal numbers of these cells with activated T cells from a third donor. Strikingly, Mψind/AZA displayed a stronger immunosuppressive phenotype, since Mψind/AZA showed an enhanced ability to inhibit T-cell proliferation compared to Mψind (Fig. 3E).
These data show that combination treatment is superior to monotherapy in vitro with respect to the induction of regulatory macrophages. Not only the number of regulatory macrophages is increased, but the macrophages also display a stronger immunosuppressive phenotype. These findings may provide a partial explanation for the superiority of infliximab/azathioprine combination therapy observed in patients with CD.
The data presented here show the induction of regulatory macrophages after infliximab treatment in IBD patients. Patients responding to infliximab showed a significant induction of CD206+/CD68+ cells, whereas this induction was absent in nonresponders. Since we used endoscopic and histologic healing as a definition of response, these data suggest that the induction of regulatory macrophages may be involved in the process of mucosal healing. The fact that we observe increased numbers of CD206+/CD68+ cells in patients responding to infliximab therapy does not prove directly that this effect is mediated by infliximab. However, we have previously shown induction of CD206+ macrophages by infliximab in vitro.8 Additionally, we now show here that CD206+ macrophages induced by infliximab in vitro have wound-healing capacity, similar to what has been described for IL-4/IL-13-induced regulatory macrophages previously.17 Together these data suggest that the induction of regulatory macrophages by infliximab may contribute to mucosal healing in patients clinically responding to the treatment.
A recent clinical trial has shown that combination treatment of azathioprine and infliximab is superior to monotreatment in the induction of remission and mucosal healing in CD patients.12 In our in vitro system, azathioprine did not induce significant numbers of macrophages. However, in combination with infliximab, azathioprine potentiated both the number and immunosuppressive capacity of CD206+ macrophages. Although the superiority of combination treatment in vivo probably results at least in part from the additive effects of two immunosuppressives, our data suggest an additional layer of actual synergism which may also contribute to the increased mucosal healing observed in patients.
In order to maintain tolerance, lamina propria macrophages normally display a type 2 macrophage (Mψ2) phenotype. Mψ2 are functionally different from type 1 macrophages (Mψ1), since they have minor ability to respond to bacterial stimuli, and produce antiinflammatory cytokines rather than proinflammatory cytokines.18, 19 In IBD patients, lamina propria macrophages have a more Mψ1 phenotype, as lamina propria mononuclear cells (LPMNCs) from IBD patients spontaneously produce large amounts of proinflammatory cytokines and are hyperresponsive to bacterial stimuli.20–22 In line with this, lower amounts of Mψ2 were found in mucosal biopsies from active lesions in CD patients compared to nonaffected colon of the same patient, and compared to healthy controls.9 This suggests that lamina propria macrophages from CD patients are skewed toward an Mψ1 phenotype, thus contributing to the defect in tolerance. Since regulatory macrophages contribute to tolerance toward the mucosal flora and CD patients have decreased numbers of this cell type, the induction of Mψ2 might be an interesting target in restoring the disturbed balance.
In conclusion, our data demonstrate the induction of regulatory macrophages in IBD patients responding to infliximab therapy. In addition, we show that infliximab-induced macrophages indeed have wound-healing capacity, suggesting a potential role in mucosal healing. Finally, we show that infliximab/azathioprine combination treatment potentiates the induction of regulatory macrophages, thus providing a new rationale for the superiority of infliximab/azathioprine combination treatment observed in the clinic.
- 6Etanercept in the treatment of active refractory Crohn's disease: a single-center pilot trial. Am J Gastroenterol. 2001; 96: 2564–2568., , , et al.Direct Link:
- 7CDP571, a humanized monoclonal antibody to tumour necrosis factor-alpha, for steroid-dependent Crohn's disease: a randomized, double-blind, placebo-controlled trial. Aliment Pharmacol Ther. 2006; 23: 617–628., , , et al.