Infection with an intestinal helminth parasite reduces Freund's complete adjuvantinduced monoarthritis in mice.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 63, No. 2, February 2011, pp 434–444 DOI 10.1002/art.30098 © 2011, American College of Rheumatology Infection With an Intestinal Helminth Parasite Reduces Freund’s Complete Adjuvant–Induced Monoarthritis in Mice Meiqing Shi, Arthur Wang, David Prescott, Christopher C. M. Waterhouse, Shuren Zhang, Jason J. McDougall, Keith A. Sharkey, and Derek M. McKay Objective. Assessment of infection with helminth parasites in murine models of disease could identify antiinflammatory mechanisms that translate into treatments for arthritic disease. The aim of this study was to test the ability of infection with the tapeworm Hymenolepis diminuta to ameliorate Freund’s complete adjuvant (CFA)–induced monoarthritis in mice. Methods. Mice received CFA with or without H diminuta, and knee swelling, pain, and measures of inflammation were assessed. Results. Injection of CFA resulted in rapid (within 24 hours) and sustained (lasting 20 days) knee swelling, a decreased pain threshold, increased blood flow to the knee, and increased production of tumor necrosis factor ␣ and interleukin-12p40 (IL-12p40). In mice that were infected with H diminuta 8 days prior to receiving CFA, the severity of arthritis was reduced as assessed by these indices of inflammation and infection 2 days after CFA injection and resulted in more rapid resolution of knee swelling. This antiarthritic effect required a viable infection and was dependent on adaptive immunity, because infection with H diminuta did not protect mice lacking T cells and B cells or the IL-4 receptor ␣ chain from CFA-induced inflammation. Interleukin-10 was of prime importance in the antiarthritic effect, because IL-10–knockout mice were not protected by infection, the antiarthritic effect was ablated by use of neutralizing IL-10 antibodies, and transfer of CD4ⴙ cells from infected wild-type mice but not IL-10–knockout mice significantly reduced CFAinduced knee swelling. Conclusion. In mice, the adaptive immune response to infection with H diminuta involves mobilization of IL-10, which has the concomitant advantage of dampening the innate immune responses that drive CFA-induced joint inflammation. There have been rapid increases in the incidence of autoimmune and inflammatory diseases such as diabetes, inflammatory bowel disease, and rheumatoid arthritis (RA) in Westernized societies (1). Inflammatory joint diseases are painful disorders that are particularly prevalent in the elderly; with life expectancies increasing, an ever larger proportion of the population will be affected by arthritis (2,3). Current treatments aimed at symptom relief can be effective in some patients, but they elicit undesirable side effects (e.g., steroids) or are expensive (e.g., anti–tumor necrosis factor ␣ [antiTNF␣]). Thus, there is an urgent need to develop better therapeutics to manage and ultimately cure arthritis and related musculoskeletal disorders. Chronic inflammation is often the result of an imbalance in the production of proinflammatory and antiinflammatory/proresolution mediators, and the reason(s) for this imbalance frequently are poorly understood. Therefore, rather than attempting to specifically intervene in the activity of a given molecule, the question arises: can the immune system be provoked to Supported by an operating grant from the Canadian Institutes for Health Research to Drs. McDougall, Sharkey, and McKay (MOP218990). Mr. Prescott is recipient of doctoral awards from the Alberta Heritage Foundation for Medical Research and the Natural Sciences and Engineering Research Council of Canada. Dr. McDougall is an Arthritis Society of Canada Investigator and an Alberta Heritage Foundation for Medical Research Senior Scholar. Drs. Sharkey and McKay are recipients of Scientist Awards from the Alberta Heritage Foundation for Medical Research. Dr. Sharkey holds the Crohn’s and Colitis Foundation of Canada Chair in Inflammatory Bowel Disease Research at the University of Calgary. Dr. McKay holds a Tier 1 Canada Research Chair in Intestinal Immunophysiology in Health and Disease at the University of Calgary. Meiqing Shi, PhD, Arthur Wang, MD, MSc, David Prescott, BSc, Christopher C. M. Waterhouse, MD, PhD, Shuren Zhang, MD, Jason J. McDougall, PhD, Keith A. Sharkey, PhD, Derek M. McKay, PhD: University of Calgary, Alberta, Canada. Address correspondence to Derek M. McKay, PhD, HSc 1877, Department of Physiology and Pharmacology, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta T2N 4N1, Canada. E-mail: firstname.lastname@example.org. Submitted for publication August 13, 2010; accepted in revised form October 12, 2010. 434 HELMINTH INHIBITION OF JOINT INFLAMMATION produce antiinflammatory signals that would downregulate joint inflammation? Proof-of-principal studies with murine models of colitis, airway inflammation, diabetes, allergic responses, and experimental autoimmune encephalitis (4–6) have shown that infection with helminth parasites can protect against concomitant disease. These studies have been complemented by 2 small clinical trials that provided the provocative findings that some patients with colitis treated with viable ova of the pig whipworm, Trichuris suis, had amelioration of their disease activity (7). This approach has not been widely applied to models of arthritis. According to a serendipitous observation 35 years ago, rats infected with the nematode parasite Syphacia oblevata were less susceptible to adjuvant-induced arthritis (8); a mechanism of protection was not presented. Our group observed that mice infected with the rat tapeworm Hymenolepis diminuta were protected against chemically induced colitis (9). Because arthritis is often an extraintestinal manifestation associated with colitis (10), and given that current treatments for arthritis and colitis often target the same pathway (11,12), we reasoned that infection with H diminuta could exert an antiarthritic effect. Here we show that mice infected with H diminuta experience significantly less joint inflammation following intraarticular injection of Freund’s complete adjuvant (CFA) compared with noninfected CFA-treated mice, and that interleukin-10 (IL-10) and CD4⫹ T cells are important components of the antiarthritic effect. Thus, mobilization of an adaptive immune response to combat infection with a tapeworm has the advantage of down-regulating the innate immune responses that drive joint inflammation in the murine model of CFA-induced monoarthritis. MATERIALS AND METHODS Animals. Male 7–9-week-old BALB/c mice, C57BL/6 mice (Charles River), NOD/SCID mice (BALB/c background; The Jackson Laboratory), IL-10–knockout mice (C57BL/6 background; breeding pairs were from The Jackson Laboratory), and IL-4 receptor ␣ chain (IL-4R␣)–knockout mice (BALB/c background; breeding pairs were a gift from Dr. F. Brombacher, University of Capetown, Capetown, South Africa) (13) were housed with free access to food and water. This study conformed to the Canadian Guidelines for the use of animals in biomedical research. Arthritis induction and H diminuta infection. Fur was removed from the knees of anesthetized mice using hair removal lotion (Nair; Church & Dwight Canada), and CFA (Sigma) was injected into the articular joint space via a 30G1/2-gauge needle, as previously described (14,15). Mice were infected with H diminuta (10 cysticercoids in 435 100 m phosphate buffered saline [PBS]) 8 days prior to the administration of CFA (i.e., a time when the helminths are being expelled from the mouse) (16); we have used this dose in other experiments, and it is known to elicit a reproducible immune response in mice (9). Additional mice were gavaged with 10 H diminuta cysticercoids that had been destroyed by sonication and boiled for 10 minutes. In other experiments, mice received CFA and 2 days later were infected with H diminuta. Other mice were infected with H diminuta and 28 days later received CFA with or without a simultaneous secondary infection or PBS as a control (by this time, immunologic memory against H diminuta has been generated) (16). Some mice were treated with indomethacin (250 g/ mouse) (17) or dexamethasone (25 g/mouse) (18,19) (both from Sigma) on days ⫺1, 0, 1, and 2 after receiving CFA by subcutaneous injection, for comparison with mice infected with H diminuta. Peripheral blood cell counts. Blood (1–2 l) was smeared onto a glass slide, differential staining was performed (Modified Wright-Giemsa Stain Pack; Fisher Diagnostics), and 300 immune cells were classified as mononuclear cells, neutrophils, or eosinophils (3% eosinophils was used as an indicator of successful infection with H diminuta [16,20,21]). Knee swelling. Following the injection of CFA or PBS, the knee diameters of anesthetized mice were measured, using microcalipers (Model 62379-531; Control Company) oriented in a mediolateral plane across the joint line, with minimal or no compression exerted on the joint. The average value from 3 separate measurements at each time point was recorded, with the data presented in millimeters or as the percentage knee swelling for each individual mouse. In all experiments, the patterns of knee swelling were similar in PBS-treated mice, H diminuta–treated mice, CFA-treated mice, and mice treated with H diminuta plus CFA. Pain behavior assessment. Secondary mechanical hyperalgesia was assessed using a dynamic plantar algesiometer (Ugo Basile model 37400), as previously described (22). Briefly, an unrestrained mouse was placed on a metal mesh surface, and a touch stimulator unit was positioned below the plantar surface of the hind paw. Activation of the unit causes a metal monofilament (0.5 mm diameter) to advance with constant speed and touch the paw in the proximal metatarsal region. The filament exerts a gradually increasing force to the plantar surface, starting below the threshold of detection and increasing until the stimulus is sensed and the mouse removes its paw. The threshold force (in grams) at which paw withdrawal occurs is recorded. The mice are habituated to the test apparatus over a 3-day period prior to assessment. This is a standard approach to assessing referred pain, which gives an indication of pain processing changes in the central nervous system. Assessment of synovial blood flow. Knee joint blood flow was determined by laser Doppler perfusion imaging (Moor Laser Doppler Imager V2; Moor Instruments), using a validated protocol (23). Briefly, mice were maintained under deep general anesthesia, an ellipse of skin covering the medial aspect of the knee was removed, a red He-Ne laser (633 nm) scanned over the surface of the joint capsulem, and basal blood flow to the region was calculated. At the end of the experiment, the mouse was killed by sodium pentobarbital overdose (120 mg/mouse, intracardiac administration), and a dead scan 436 of the joint was performed. This “biological zero” value corresponding to a tissue reflection artifact was subtracted from the basal blood flow value, which is presented in arbitrary perfusion units. Myeloperoxidase (MPO) activity. Knees and segments of colon and lung were excised and homogenized, and MPO activity was determined by a kinetic assay (9), in which 1 unit of MPO activity is the amount of enzyme required to degrade 1 M H2O2 per minute. Cytokine production. Messenger RNA (mRNA) measured by real-time polymerase chain reaction (PCR). RNA was extracted from knee homogenates and reverse transcribed using the iScript reverse transcriptase kit (Bio-Rad) in a MyCycler thermal cycler (Bio-Rad). One microliter of complementary DNA was mixed with a reaction buffer containing 1⫻ SYBR Green (Bio-Rad) and 300 nM gene primers. Quantitative PCR was conducted in a Mastercycler real-time thermal cycler (Eppendorf), and the data were analyzed using realplex software (Eppendorf). The expression of TNF␣ and IL-10 mRNA is presented relative to the expression of the 18S RNA housekeeping gene. The primer sequences used are as follows: for IL-10, forward CCAAGCCTTATCGGAAATGA, reverse TTTTCACAGGGGAGAAATCG; for TNF␣, forward AGCCCCCAGTCTGTATCCTT, reverse CTCCCTTTGCAGAACTCAGG); for 18S RNA, forward CAGTTATGGTTCCTTTGGTCG, reverse ATCTAGAGTCACCAAGCCGC. Protein levels measured by enzyme-linked immunosorbent assay (ELISA). The spleen and inguinal lymph nodes (LNs) were excised separately, and 5 ⫻ 106 cells (inguinal LN cells from 3 mice were grouped) were cultured with 2 g/ml of the T cell mitogen concanavalin A (ConA; Sigma) for 48 hours. Levels of IL-4, IL-10, IL-12p40, and TNF␣ were determined by ELISA, following the manufacturer’s instructions (R&D Systems). Each ELISA had a detection limit of 2–9 pg/ml. Data are presented as either pg/ml or as an IL-10:TNF␣ ratio, with a higher value for the ratio indicating a relative increase in IL-10 over TNF␣. Anti–IL-10 antibody treatment. BALB/c mice were infected with H diminuta, and CFA was administered 8 days later. A time-matched group of mice received intraperitoneal injections of 50 g, 100 g, and 50 g rat anti-mouse neutralizing IL-10 antibody (rat IgG1; Thermo Fisher Scientific) on days 7, 8 (i.e., 30 minutes after administration of CFA), and 9 after H diminuta infection, respectively (previous studies showed that isotype-matched IgG did not affect the outcome of colitis) (9). In these studies and those involving IL-10– deficient mice, tapeworms were not observed in the small bowel or in intestinal washings, suggesting that normal antiworm expulsion events had occurred (which is in contrast to the situation in SCID and IL-4R␣–knockout mice [see below]). Flow cytometry analysis. Spleen cells (5 ⫻ 106) were stimulated with ConA (2 g/ml) and processed for flow cytometry (24). Briefly, 43 hours after stimulation, 1 l GolgiPlug (BD PharMingen) was added, and 5 hours later cells were harvested, washed with PBS, and resuspended at 1 ⫻ 106 cells/100 l PBS containing 0.5% fetal calf serum. Cells were incubated with phycoerythrin (PE)–conjugated rat anti-mouse CD4, CD8, CD220, or F4/80, or with PE-conjugated matched control antibodies (BD PharMingen) for 30 minutes at 4°C in SHI ET AL the dark. Following 2 rinses with PBS, the cells were fixed with 1% formalin (20 minutes at room temperature), washed with PBS, and permeabilized with a solution of PBS and 0.3% saponin. Cells were incubated with fluorescein isothiocyanate (FITC)–conjugated rat anti-mouse IL-10 monoclonal antibodies (mAb; BD PharMingen) or FITC-conjugated rat IgG2b (isotype control) for 30 minutes at room temperature and washed twice in 500 l PBS–0.3% saponin. Fluorescence was detected on a FACScan flow cytometer using CellQuest Pro (both from BD PharMingen) and was analyzed with FlowJo software (TreeStar). Isolation and adoptive transfer of CD4ⴙ T cells. Eight days postinfection, the spleens were removed from wild-type or IL-10–knockout mice, disrupted into 1 ⫻ 108 cells/ml PBS suspension, and CD4⫹ cells were purified by negative selection using a mouse CD4⫹ T cell enrichment kit according to the manufacturer’s instructions (StemCell Technologies). The purity of the population was assessed using PE-conjugated rat anti-mouse CD4 mAb (BioLegend), FITC-conjugated rat antimouse CD8 mAb (Abcam), or FITC-conjugated rat antimouse CD14 mAb (Abcam) and was 98% CD4⫹ cells with a size and granularity pattern typical of lymphocytes. The purified CD4⫹ cell population was resuspended, and 1 ⫻ 107 cells were injected intravenously into each donor mouse. Two days later, the mice were injected with CFA. Statistical analysis. Data are presented as the mean ⫾ SEM. Two-group comparisons were performed using Student’s t-test, and multiple-group comparisons were performed by one-way analysis of variance followed by Tukey’s post hoc test. P values less than 0.05 were considered significant. RESULTS No CFA-induced blood neutrophilia in H diminuta– infected mice. On days 1, 2, and 7 after the administration of CFA, the number of blood neutrophils was increased, with a proportional reduction in the number of mononuclear cells; this was not observed in mice treated with H diminuta plus CFA (data not shown). For instance, 2 days posttreatment, neutrophils represented 47.3 ⫾ 3.9% (mean ⫾ SEM) of blood leukocytes in CFA-treated mice compared with 28.6 ⫾ 2.0% and 30.6 ⫾ 4.4% of blood leukocytes in PBS-treated mice and mice treated with H diminuta plus CFA, respectively (n ⫽ 4 mice per group; P ⬍ 0.05, CFA-treated mice versus other groups). H diminuta–infected mice displayed ⬎3% blood eosinophils, indicating successful infection (25), whereas control mice and CFA-treated mice displayed 0.5–2.5% blood eosinophils. Effect of infection with H diminuta on CFAinduced arthritis. The injection of sterile PBS into the articular space caused increases in knee diameter, typically ⱕ10% of the preinjection size (Figure 1A), and infection with H diminuta did not result in any knee swelling. In contrast, the injection of CFA evoked HELMINTH INHIBITION OF JOINT INFLAMMATION marked swelling that peaked within 1–3 days and was sustained for 20 days (e.g., 2 days after the injection of CFA, the mean ⫾ SEM size of the injected knees increased from 3.76 ⫾ 0.06 mm to 4.57 ⫾ 0.07 mm [n ⫽ 15 mice]), and 20 days after the injection of CFA the size was 4.22 ⫾ 0.08 mm (Figure 1A). Swelling was not observed in the noninjected contralateral knees. The response to CFA was similar in C57BL/6 mice (Figure 1B). In mice infected with H diminuta, the impact of CFA was significantly reduced, as gauged by a reduction in maximum knee swelling and enhanced resolution of swelling (Figures 1A and B). The injection of CFA resulted in increased pain sensitivity, whereas the threshold for paw withdrawal was not different between control mice and those that received H diminuta plus CFA (Figure 1C). Similarly, the increase in joint blood flow that followed CFA treatment was significantly reduced in mice infected with H diminuta (Figure 1D). MPO activity reflects acute inflammation and is characteristic of granulocyte infiltration, particularly neutrophils. There was a small but statistically significant increase in MPO activity in knee extracts obtained 24 hours after CFA treatment, and this activity was reduced by ⬃40% in the mice treated with H diminuta plus CFA (n ⫽ 6). Moreover, and in accordance with the sustained knee swelling in the mice treated with CFA only, MPO levels were still elevated 7 days after treatment with CFA and were significantly reduced in mice treated with H diminuta plus CFA (Figure 1E). Analysis of MPO levels in the lung and colon revealed no differences between mice treated with CFA and those treated with H diminuta plus CFA (n ⫽ 4 mice per group; results not shown), supporting a local effect of CFA. Based on these findings and the robustness of the blockade of CFA-induced knee swelling by H diminuta, joint swelling was used as a measure of joint inflammation throughout the remainder of this study. Additional studies revealed that infection with H diminuta was as effective at reducing CFA-induced knee swelling as was dexamethasone or indomethacin, which are representative steroid and nonsteroidal antiinflammatory drugs (NSAIDs) (Figure 1F). Role of H diminuta infection in inhibition of CFA-induced knee swelling. Oral administration of disrupted cysticercoids (confirmed microscopically), which provides the same antigen load and antigen complexity as a viable infection, did not prevent CFA-induced knee swelling, suggesting that the antiarthritic effect required 437 Figure 1. Infection with Hymenolepis diminuta (H. dim; H.d) protects against Freund’s complete adjuvant (CFA)–induced arthritis. A, Knees of BALB/c mice injected with CFA displayed significant swelling that was significantly less severe in mice infected 8 days previously with H diminuta. Time-matched control and H diminuta–infected mice injected with phosphate buffered saline (PBS) are shown (n ⫽ 8–16 mice for days 2–7 after CFA treatment [i.e., days 10–15 postinfection] and n ⫽ 4 mice thereafter [except for PBS, where n ⫽ 2]; the average starting knee diameter was 3.40–4.28 mm). B, The effects of H diminuta infection in C57BL/6 mice (n ⫽ 5–8) were similar to those observed in BALB/c mice. The numbers in parentheses represent the average inhibition of knee swelling. C, BALB/c mice receiving CFA had a reduced pain threshold compared with PBS-treated controls or mice infected with H diminuta (algesiometry was performed 7 days after injection of CFA or PBS; n ⫽ 4 mice per group). D, The increase in joint blood flow that followed CFA treatment was significantly reduced in mice infected with H diminuta (n ⫽ 4 mice per group). E, Myeloperoxidase (MPO) levels were significantly reduced in mice treated with H diminuta plus CFA (n ⫽ 4 mice per group; assessed 7 days after CFA treatment). F, Treatment with H diminuta infection was as effective as treatment with dexamethasone (Dex) or indomethacin (Indo) in preventing CFA-induced knee swelling (n ⫽ 4 mice per group). Values are the mean ⫾ SEM. In A and B, ⴱ ⫽ P ⬍ 0.05 versus starting knee size; # ⫽ P ⬍ 0.05 versus PBS/CFA. In C–F, ⴱ ⫽ P ⬍ 0.05 versus control or H diminuta alone; # ⫽ P ⬍ 0.05 versus CFA alone. PU ⫽ perfusion units. 438 SHI ET AL Figure 2. Adaptive immune responses are required for the antiarthritic effect of infection with H diminuta. A, Oral administration of disrupted cysticercoids (dead H.d⫹CFA) did not prevent CFAinduced knee swelling, suggesting that the antiarthritic effect of H diminuta requires a viable infection. B, Secondary infection with H diminuta at the time of CFA injection (H.d[2°]⫹CFA) evoked increased resolution of knee swelling in wild-type (WT) BALB/c mice, whereas mice infected with H diminuta 28 days prior to CFA challenge (primary infection; H.d[1°]⫹CFA) were not protected. C and D, Lymphocytes (which are absent in SCID mice) (C) and functional interleukin-4 (IL-4)/IL-13 receptors (which are absent in IL-4 receptor ␣-chain–knockout [IL-4R␣ KO] mice) (D) are required for H diminuta inhibition of CFA-induced knee swelling. Values are the mean ⫾ SEM results from 3–4 mice per group. ⴱ ⫽ P ⬍ 0.05 versus CFA. See Figure 1 for other definitions. a viable infection (Figure 2A). Moreover, mice infected 14 days prior to CFA challenge were protected, with knee swelling being 35–40% less than that in mice treated with CFA only (n ⫽ 4 mice per group), but this effect waned such that mice infected 28 days prior to CFA treatment were not protected (Figure 2B). However, a second infection of mice at the time of CFA injection (i.e., 28 days after the primary infection) led to a more rapid reduction in knee swelling (Figure 2B); these data indicate an immunologic memory response. SCID mice, which lack T cells and B cells, and IL-4R␣–knockout mice, which cannot respond to IL-4 or IL-13, developed substantial knee swelling that was not affected by infection with H diminuta (Figures 2C and D). In the latter instance, the possibility was considered that any beneficial effect of infection with H diminuta might be delayed; this proved not to be so, because the degree of knee swelling observed 7 days after CFA injection was virtually identical in IL-4R␣– knockout mice treated with CFA only and in mice treated with H diminuta plus CFA (Figure 2D). At autopsy, H diminuta was observed in the small intestines of all infected SCID and IL-4R␣–knockout mice but not in the intestines of wild-type mice. Effect of treatment with H diminuta infection on recovery from CFA-induced arthritis. Mice infected with H diminuta 2 days after CFA treatment displayed enhanced recovery, with a statistically significant reduc- HELMINTH INHIBITION OF JOINT INFLAMMATION 439 mice, which was not apparent in mice infected with H diminuta (Figure 4). IL-10 mRNA levels in the injected joints were not different between groups. Cytokine production from mitogen-activated spleen cells is shown in Figure 5. One day after the injection of CFA, splenocytes from CFA-treated mice produced more IL-12p40 and TNF␣ than did mice treated with PBS or H diminuta plus CFA (Figure 5A). In contrast, splenocytes from mice treated with H diminuta, with or without CFA, produced increased amounts of IL-4 and IL-10 compared with control mice and mice treated with CFA only. This pattern of cytokine responsiveness to ConA was also evident 2 days (results not shown) and 7 days after CFA treatment (Figure 5B) and in the draining inguinal LNs (results not shown) following the H diminuta treatment regimen (Figure 3B), and in the memory experiment (Figure 5C). These data illustrate a shift in the cytokine profile away from TNF␣ and in favor of IL-10 synthesis and hence a substantial relative increase in IL-10. For example, the mean ⫾ SEM IL-10:TNF␣ ratio 18 hours after CFA treatment was 0.8 ⫾ 0.1 for mice treated with CFA only compared with 5.8 ⫾ 1.4 for mice treated with H diminuta plus CFA (P ⬍ 0.05, by Student’s t-test). For control mice treated with PBS, the IL-10:TNF␣ ratio was 1.4 ⫾ 0.2 (n ⫽ 3–4 mice per group). The levels of IL-10 and IL-12p40 in supernatants Figure 3. Infection with Hymenolepis diminuta (H.d) as a treatment blocks the effect of Freund’s complete adjuvant (CFA) injection. A and B, Mice infected with H diminuta 2 days after injection of CFA had a hastened recovery as assessed by knee swelling (A), and this was associated with reduced tumor necrosis factor ␣ (TNF␣) production and increased interleukin-4 (IL-4) and IL-10 production by splenocytes stimulated with 2 g/ml concanavalin A that were excised 9 days after CFA injection (B). Cytokines were measured 48 hours after treatment with concanavalin A. Values are the mean ⫾ SEM results from 4 mice per group. ⴱ ⫽ P ⬍ 0.05 versus CFA. tion in knee swelling at 5 days postinfection (Figure 3A) and skewing of stimulated splenocyte cytokine production in favor of IL-4 and IL-10 and away from TNF␣ (Figure 3B). Effect of H diminuta infection on immune responses in favor of IL-10. Quantitative PCR revealed an increase in TNF␣ mRNA in the knees of CFA-treated Figure 4. Increased TNF␣ mRNA expression in the CFA-injected knees of mice. Quantitative real-time polymerase chain reaction revealed an increase in TNF␣ mRNA expression in the knees of mice injected with CFA (n ⫽ 3). Such an increase was not observed in mice infected with 10 H diminuta cysticercoids 8 days before the CFA injection (n ⫽ 4). Three mice were treated with phosphate buffered saline (PBS) as control. Bars show the mean ⫾ SEM. See Figure 3 for other definitions. 440 from ConA-stimulated splenocytes from mice treated with killed H diminuta plus CFA were not different from those in mice treated with CFA only (data not shown), mirroring the inability of this treatment to block CFAinduced knee swelling. Similarly, splenocytes from H diminuta–infected and CFA-challenged SCID mice or IL-4R␣–knockout mice did not produce increased amounts of IL-10 compared with cells from mice treated with CFA only (data not shown). Role of IL-10 in the antiarthritic effect of H diminuta infection. Wild-type C57BL/6 and IL-10– knockout mice had similar knee swelling in response to CFA and displayed a ⬃2.5-fold increase in MPO activity. Infection with H diminuta protected the wild-type but not the IL-10–knockout mice from CFA-induced knee swelling (Figure 6A). The mean ⫾ SEM cytokine production was not different when splenocytes from infected and noninfected IL-10–knockout mice were compared: for CFA-treated IL-10–knockout mice (n ⫽ 3) versus IL-10–knockout mice treated with H diminuta plus CFA (n ⫽ 4), 498 ⫾ 28 versus 539 ⫾ 115 pg/ml TNF␣; for CFA-treated IL-10–knockout mice versus IL-10–knockout mice treated with H diminuta plus CFA, 276 ⫾ 36 versus 279 ⫾ 59 pg/ml IL-12p40. Because IL-10–knockout mice develop colitis (at ⬃12 weeks of age in the University of Calgary facility), it is possible that intestinal inflammation affected the outcome of these experiments. However, the mice used in this study had negligible colitis, and there were no intestine differences between mice treated with CFA only and those treated with H diminuta plus CFA, as gauged by colon histology or MPO levels (data not shown). The abrogation of the antiarthritic effect of infection with H diminuta by neutralizing anti–IL-10 antibodies corroborated the findings in the IL-10–knockout mice (Figure 6B). Effect of adoptive transfer of CD4ⴙ spleen cells from H diminuta–infected wild-type mice on the effect of CFA. Intracellular staining of splenocytes from H diminuta–infected mice revealed that subpopulations of CD4⫹ T cells, CD8⫹ T cells, and B220⫹ B cells expressed IL-10: 14.1 ⫾ 2.8% (mean ⫾ SEM) of CD4⫹ cells (n ⫽ 3 mice) were IL-10 positive, while ⱕ0.5% of F4/80-positive macrophages displayed IL-10 immunoreactivity. Adoptive transfer of spleen-derived CD4⫹ T cells from H diminuta–infected wild-type mice but not IL-10–knockout mice protected recipients against the proarthritic effects of CFA, as measured by knee swelling (Figure 6C). SHI ET AL Figure 5. Infection with H diminuta results in increased production of IL-10 by immune cells. A, One day after injection of CFA, splenocytes from CFA-treated mice produced more IL-12p40 and TNF␣ than did splenocytes from mice treated with phosphate buffered saline (PBS) or H diminuta plus CFA. Splenocytes from mice treated with H diminuta, with or without CFA, produced increased amounts of IL-4 and IL-10 compared with control mice and mice treated with CFA only. B and C, The pattern of cytokine responsiveness observed 1 day after injection of CFA was also evident 7 days after CFA treatment, following the 8-day prophylactic regimen (B) and in the memory experiment (C). Values beside the bars in B are the actual values. Bars show the mean ⫾ SEM results from 3–4 mice and are representative of 1 of 2 experiments performed. ⴱ ⫽ P ⬍ 0.05 versus PBS control. ND ⫽ not detectable (see Figure 3 for other definitions). HELMINTH INHIBITION OF JOINT INFLAMMATION 441 DISCUSSION Figure 6. Interleukin-10 (IL-10) and CD4⫹ cells are important in the antiarthritic effect of infection with Hymenolepis diminuta (H.d). A, C57BL/6 IL-10–knockout (KO) mice infected with H diminuta 8 days prior to intraarticular injection of Freund’s complete adjuvant (CFA) were not protected against knee swelling (n ⫽ 4 mice per group). CFA-induced knee swelling in wild-type (WT) C57BL/6 mice was reduced by infection with H diminuta (see Figure 1B). B, In vivo neutralization of IL-10 by intraperitoneal administration of anti–IL-10 antibodies (aIL-10) blocked the antiarthritic effect of infection with H diminuta in CFA-treated wild-type BALB/c mice (n ⫽ 4). ⴱ ⫽ P ⬍ 0.05 versus phosphate buffered saline (PBS) (control); # ⫽ P ⬍ 0.05 versus CFA and versus H diminuta ⫹ CFA ⫹ anti–IL-10. C, Adoptive transfer of CD4⫹ T cells from the spleens of wild-type mice infected 8 days previously with H diminuta protected against the knee swelling induced by CFA injection 24 hours previously; transfer of CD4⫹ T cells from infected IL-10–knockout mice did not confer protection (n ⫽ 3 mice per group). ⴱ ⫽ P ⬍ 0.05 versus PBS control; # ⫽ P ⬍ 0.05 versus CFA alone. Values in A–C are the mean ⫾ SEM. There is renewed interest in the ability of the immune response mobilized to combat infection with helminth parasites to have a concomitant health benefit (6,26). Data in support of this have been presented, particularly in the context of colitis and airway inflammation (27–29). The data presented here support 3 conclusions: 1) infection with H diminuta significantly improves the outcome of adjuvant-induced monoarthritis, 2) an intact adaptive immune response involving T cells and IL-4R␣ signaling is required for the antiarthritic effect of H diminuta infection, and 3) IL-10 and CD4⫹ T cells are important elements of the antiarthritic effect. Thus, the adaptive immune response raised against H diminuta can reduce joint inflammation that occurs as a consequence of local innate immune reactions. Intraarticular injection of CFA results in rapid and sustained knee swelling that is accompanied by increased MPO and TNF␣ levels in the affected knee (14,15,30,31); the contralateral, noninjected knee and other joints are not involved. Using this model of monoarthritis, we observed that mice infected with H diminuta were significantly protected against CFAinduced arthritis: knee swelling was reduced, the knee more rapidly returned to normal size, the enhanced blood flow and MPO levels were reduced, and mice displayed less sensitivity to referred pain. When used as a treatment regimen, infection with H diminuta resulted in enhanced recovery following injection of CFA. Moreover, infection with H diminuta was as potent as dexamethasone and indomethacin in preventing CFAinduced knee swelling; this finding is notable because steroids and NSAIDs are mainstay treatments for arthritis, and both classes of drug have many undesirable side effects. The potential for infection with helminth parasites to reduce the severity of arthritic disease is not unprecedented. Rats infected with S oblevata were protected against CFA-induced paw inflammation (8), and the spontaneous joint inflammation that develops in MRL/lpr mice was slightly reduced by infection with the nematode Nippostrongylus brasiliensis, as gauged by synovial hyperplasia only (32). The only other study in this area showed that infection with Schistosoma mansoni reduced the severity of collagen-induced arthritis in mice, and this correlated with increased IL-4 and IL-10 production and decreased IL-17A, IFN␥, and TNF␣ production by spleen cells (33). However, with the exception of a correlation between cytokine profile skewing and the severity of arthritis, a mechanism for 442 helminth-induced modulation of joint inflammation has not been elucidated. Protection against arthritis by a gut-dwelling helminth must be mediated by a systemic event, as supported by normalization of the blood leukocyte composition in mice treated with H diminuta plus CFA. This concept is supported by the finding that oral administration of probiotics suppressed atopic dermatitis and arthritis in mouse models (34). The presence of T cells and IL-4 signaling are prerequisites for the immunologic expulsion of H diminuta from mice (16,35), and all of the following point to the antiarthritic effect of H diminuta occurring via an active host immune response involving T cells: 1) destroyed H diminuta cysticercoids did not have an antiarthritic effect, 2) mice lacking T cells and B cells or competent IL-4 signaling (IL-4R␣–knockout mice) were not protected against CFA-induced arthritis by infection with H diminuta, and 3) a second infection at the time of CFA challenge resulted in significantly less knee swelling, consistent with triggering an anti– H diminuta memory response. There is a paucity of data on the local or systemic cytokine response following injection of CFA into the mouse knee, with one study showing increased expression of TNF␣ (14). Analyses of models of arthritis and patient studies support proarthritic roles for TNF␣, which also participates in pain perception (36–38), IL12, IL-17, and IL-23 (19,39–41), with IL-10 and FoxP3⫹ regulatory T cells exerting antiarthritic effects (34,42– 46). Immune cells from CFA-treated mice produced more proinflammatory cytokines than were produced by cells from control mice or mice treated with H diminuta plus CFA. In vitro assays revealed that immune cells from H diminuta–infected mice produce substantial amounts of IL-4 and IL-10, corroborating studies with this and other parasitic helminths (20,47). Many components of the reaction to H diminuta could contribute to an antiinflammatory response (e.g., liberation of IL-4 to block Th1 events). However, based on skewing of the IL-10:TNF␣ ratio in favor of IL-10 in mice treated with H diminuta plus CFA, mechanistic studies focused on the putative involvement of IL-10 in the suppression of joint inflammation. Experiments using the complementary approaches of in vivo IL-10 neutralization, CFA treatment of IL-10–knockout mice, and adoptive transfer of CD4⫹ T cells from infected wild-type or IL-10–knockout mice support the conclusion that IL-10 is an important factor in the inhibition of CFA-induced inflammation. Given the role of IL-10 in suppressing this adjuvant-induced monoarthritis and data showing increased IL-10 production in S mansoni– SHI ET AL infected mice sensitized to collagen (33), additional studies are needed to assess the effect of infection with H diminuta in the type II collagen–induced polyarthritis model, which is somewhat reminiscent of RA. The notion that CD4⫹ T cells are a source of IL-10 that mediates the antiarthritic effect is in accordance with the presence of an IL-10–positive Treg cell population in patients responding to therapy for RA (44). However, the presence of IL-10⫹/CD8⫹ and IL10⫹/B220⫹ cells in H diminuta–infected mice suggests that these cells can participate in antiarthritic effects; this possibility remains to be tested. Similarly, numerous other putative mechanisms need to be assessed to precisely define the role of IL-10 in this model system, as follows: Is the antiinflammatory effect of CD4⫹/IL-10⫹ cells or IL-10 exerted locally or systemically? Is the inhibition of TNF␣ production of paramount importance, and is this accomplished by inhibition of macrophage activity by IL-10 or another signal elicited by IL-10? Is the ability of IL-10 to block mast cell activity important in the CFA model? What role, if any, does IL-10 play in the modulation of blood flow and pain perception, and is this integrated with local or central neural events? As these questions are answered, a precise picture of the antiarthritic mechanism of IL-10 will emerge. As a caveat, helminth therapy or the delivery of excess amounts of IL-10 has, in theory, the ability to promote autoimmune disease or cause undesirable side effects (e.g., fibrosis) (6). However, the experiments conducted with helminths in animal models have not demonstrated side effects (6), and, in the context of antibody-driven autoimmunity, infection with S mansoni reduced type II collagen–induced arthritis (33). In summary, the immune response elicited by mice in reaction to infection with H diminuta (a gut lumen–dwelling tapeworm) significantly reduced (prophylactically and therapeutically) CFA-induced monoarthritis, and IL-10 was an important element of the antiarthritic mechanism. However, recombinant IL-10 has been of little value in treating RA (47). This suggests either that natural production of IL-10 in the host is more effective than a bolus of recombinant IL-10 or that IL-10 operates in concert with other factors mobilized in response to helminth infection. Finally, the potential of helminth therapy in patients with arthritic disease in whom conventional treatments have failed is intriguing, particularly if antiinflammatory immunologic memory responses could be evoked. HELMINTH INHIBITION OF JOINT INFLAMMATION 443 AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. McKay had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. McDougall, Sharkey, McKay. Acquisition of data. Shi, Wang, Prescott, Waterhouse, Zhang. Analysis and interpretation of data. Shi, Prescott, Waterhouse, McDougall, Sharkey, McKay. REFERENCES 1. Bach JF. The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med 2002;347:911–20. 2. Helmick CG, Felson DT, Lawrence RC, Gabriel S, Hirsch R, Kwoh CK, et al, for the National Arthritis Data Workgroup. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part I. Arthritis Rheum 2008;58: 15–25. 3. Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H, Deyo RA, et al, for the National Arthritis Data Workgroup. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum 2008;58: 26–35. 4. Rook GA. Review series on helminths, immune modulation and the hygiene hypothesis: the broader implications of the hygiene hypothesis. Immunology 2009;126:3–11. 5. Elliott DE, Summers RW, Weinstock JV. Helminths as governors of immune-mediated inflammation. Int J Parasitol 2007;37: 457–64. 6. McKay DM. The therapeutic helminth? Trends Parasitol 2009;25: 109–14. 7. Summers RW, Elliott DE, Urban JF Jr, Thompson R, Weinstock JV. Trichuris suis therapy in Crohn’s disease. Gut 2005;54:87–90. 8. Pearson DJ, Taylor G. The influence of the nematode Syphacia oblevata on adjuvant arthritis in the rat. Immunology 1975;29: 391–6. 9. Hunter MM, Wang A, Hirota CL, McKay DM. Neutralizing anti-IL-10 antibody blocks the protective effect of tapeworm infection in a murine model of chemically induced colitis. J Immunol 2005;174:7368–75. 10. Larsen S, Bendtzen K, Nielsen OH. Extraintestinal manifestations of inflammatory bowel disease: epidemiology, diagnosis, and management. Ann Med 2010;42:97–114. 11. Atzeni F, Ardizzone S, Bertani L, Antivalle M, Batticciotto A, Sarzi-Puttini P. Combined therapeutic approach: inflammatory bowel diseases and peripheral or axial arthritis. World J Gastroenterol 2009;15:2469–71. 12. Armaka M, Apostolaki M, Jacques P, Kontoyiannis DL, Elewaut D, Kollias G. Mesenchymal cell targeting by TNF as a common pathogenic principle in chronic inflammatory joint and intestinal diseases. J Exp Med 2008;205:331–7. 13. Mohrs M, Holscher C, Brombacher F. Interleukin-4 receptor ␣-deficient BALB/c mice show an unimpaired T helper 2 polarization in response to Leishmania major infection. Infect Immun 2000;68:1773–80. 14. Keeble J, Blades M, Pitzalis C, Castro da Rocha FA, Brain SD. The role of substance P in microvascular responses in murine joint inflammation. Br J Pharmacol 2005;144:1059–66. 15. Keeble J, Russell F, Curtis B, Starr A, Pinter E, Brain SD. Involvement of transient receptor potential vanilloid 1 in the vascular and hyperalgesic components of joint inflammation. Arthritis Rheum 2005;52:3248–56. 16. McKay DM, Halton DW, McCaigue MD, Johnston CF, Fair- 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. weather I, Shaw C. Hymenolepis diminuta: intestinal goblet cell response in male C57 mice. Exp Parasitol 1990;71:9–20. Gauldie SD, McQueen DS, Clarke CJ, Chessell IP. A robust model of adjuvant-induced chronic unilateral arthritis in two mouse strains. J Neurosci Methods 2004;139:281–91. Lesniak MS, Gabikian P, Tyler BM, Pardoll DM, Brem H. Dexamethasone mediated inhibition of local IL-2 immunotherapy is dose dependent in experimental brain tumors. J Neurooncol 2004;70:23–8. Monari C, Bevilacqua S, Piccioni M, Pericolini E, Perito S, Calvitti M, et al. A microbial polysaccharide reduces the severity of rheumatoid arthritis by influencing Th17 differentiation and proinflammatory cytokines production. J Immunol 2009;183:191–200. Persaud R, Wang A, Reardon C, McKay DM. Characterization of the immuno-regulatory response to the tapeworm Hymenolepis diminuta in the non-permissive mouse host. Int J Parasitol 2007; 37:393–403. Reardon C, Sanchez A, Hogaboam CM, McKay DM. Tapeworm infection reduces the ion transport abnormalities induced by dextran sulphate sodium (DSS) colitis. Infect Immunity 2001;69: 4417–23. McDougall JJ, Watkins L, Li Z. Vasoactive intestinal peptide (VIP) is a modulator of joint pain in a rat model of osteoarthritis. Pain 2006;123:98–105. Karimian SM, McDougall JJ, Ferrell WR. Neuropeptidergic and autonomic control of the vasculature of the rat knee joint revealed by laser Doppler perfusion imaging. Exp Physiol 1995;80:341–8. Shi M, Wei G, Pan W, Tabel H. Experimental African trypanosomiasis: a subset of pathogenic, IFN-␥-producing, MHC class II-restricted CD4⫹ T cells mediates early mortality in highly susceptible mice. J Immunol 2006;176:1724–32. Hunter MM, Wang A, McKay DM. Helminth infection enhances disease in a murine TH2 model of colitis. Gastroenterology 2007;132:1320–30. Maizels RM, Yazdanbakhsh M. T-cell regulation in helminth parasite infections: implications for inflammatory diseases. Chem Immunol Allergy 2008;94:112–23. Wilson MS, Taylor MD, Balic A, Finney CA, Lamb JR, Maizels RM. Suppression of allergic airway inflammation by helminthinduced regulatory T cells. J Exp Med 2005;202:1199–212. Elliott DE, Setiawan T, Metwali A, Blum A, Urban JF Jr, Weinstock JV. Heligmosomoides polygyrus inhibits established colitis in IL-10-deficient mice. Eur J Immunol 2004;34:2690–8. Ince MN, Elliott DE, Setiawan T, Metwali A, Blum A, Chen HL, et al. Role of T cell TGF-␤ signaling in intestinal cytokine responses and helminthic immune modulation. Eur J Immunol 2009;39:1870–8. Ferrell WR, Lockhart JC, Kelso EB, Dunning L, Plevin R, Meek SE, et al. Essential role for proteinase-activated receptor-2 in arthritis. J Clin Invest 2003;111:35–41. Kelso EB, Dunning L, Lockhart JC, Ferrell WR. Strain dependence in murine models of monoarthritis. Inflamm Res 2007;56: 511–4. Salinas-Carmona MC, de la Cruz-Galicia G, Perez-Rivera I, Solis-Soto JM, Segoviano-Ramirez JC, Vazquez AV, et al. Spontaneous arthritis in MRL/lpr mice is aggravated by Staphylococcus aureus and ameliorated by Nippostrongylus brasiliensis infections. Autoimmunity 2009;42:25–32. Osada Y, Shimizu S, Kumagai T, Yamada S, Kanazawa T. Schistosoma mansoni infection reduces severity of collagen-induced arthritis via down-regulation of pro-inflammatory mediators. Int J Parasitol 2009;39:457–64. Kwon HK, Lee CG, So JS, Chae CS, Hwang JS, Sahoo A, et al. Generation of regulatory dendritic cells and CD4⫹Foxp3⫹ T cells by probiotics administration suppresses immune disorders. Proc Natl Acad Sci U S A 2010;107:2159–64. 444 35. McKay DM. The immune response to and immunomodulation by Hymenolepis diminuta. Parasitology 2010;137:385–94. 36. Bowen CJ, Edwards CJ, Lindsey H, Dewbury K, Sampson M, Sawyer S, et al. Improvement in symptoms and signs in the forefoot of patients with rheumatoid arthritis treated with antiTNF therapy. J Foot Ankle Res 2010;3:10–5. 37. Geboes L, De Klerck B, Van Balen M, Kelchtermans H, Mitera T, Boon L, et al. Freund’s complete adjuvant induces arthritis in mice lacking a functional interferon-␥ receptor by triggering tumor necrosis factor ␣–driven osteoclastogenesis. Arthritis Rheum 2007;56:2595–607. 38. Inglis JJ, Nissim A, Lees DM, Hunt SP, Chernajovsky Y, Kidd BL. The differential contribution of tumour necrosis factor to thermal and mechanical hyperalgesia during chronic inflammation. Arthritis Res Ther 2005;7:R807–16. 39. Geboes L, Dumoutier L, Kelchtermans H, Schurgers E, Mitera T, Renauld JC, et al. Proinflammatory role of the Th17 cytokine interleukin-22 in collagen-induced arthritis in C57BL/6 mice. Arthritis Rheum 2009;60:390–5. 40. Hossain A, Zheng CL, Kukita A, Kohashi O. Balance of Th1/Th2 cytokines associated with the preventive effect of incomplete Freund’s adjuvant on the development of adjuvant arthritis in LEW rats. J Autoimmun 2001;17:289–95. 41. Murphy CA, Langrish CL, Chen Y, Blumenschein W, McClanahan T, Kastelein RA, et al. Divergent pro- and antiinflammatory roles for IL-23 and IL-12 in joint autoimmune inflammation. J Exp Med 2003;198:1951–7. SHI ET AL 42. An H, Chandra V, Piraino B, Borges L, Geczy C, McNeil HP, et al. Soluble LILRA3, a potential natural antiinflammatory protein, is increased in patients with rheumatoid arthritis and is tightly regulated by interleukin 10, tumor necrosis factor-␣ and interferon-␥. J Rheumatol 2010;37:1596–606. 43. Zimmerman DH, Taylor P, Bendele A, Carambula R, Duzant Y, Lowe V, et al. CEL-2000: a therapeutic vaccine for rheumatoid arthritis arrests disease development and alters serum cytokine/ chemokine patterns in the bovine collagen type II induced arthritis in the DBA mouse model. Int Immunopharmacol 2010;10:412–21. 44. Kavousanaki M, Makrigiannakis A, Boumpas D, Verginis P. Novel role of plasmacytoid dendritic cells in humans: induction of interleukin-10–producing Treg cells by plasmacytoid dendritic cells in patients with rheumatoid arthritis responding to therapy. Arthritis Rheum 2010;62:53–63. 45. Wieten L, Berlo SE, ten Brink CB, van Kooten PJ, Singh M, van der Zee R, et al. IL-10 is critically involved in mycobacterial HSP70 induced suppression of proteoglycan-induced arthritis. PLoS ONE 2009;4:e4186. 46. Paul WE, Zhu J. How are TH2-type immune responses initiated and amplified? Nat Rev Immunol 2010;10:225–35. 47. Van Roon J, Wijngaarden S, Lafeber FP, Damen C, van de Winkel J, Bijlsma JW. Interleukin 10 treatment of patients with rheumatoid arthritis enhances Fc␥ receptor expression on monocytes and responsiveness to immune complex stimulation. J Rheumatol 2003;30:648–51.