Experimental immunology
High-dose methotrexate ameliorates collagen-induced arthritis but does not inhibit the release of proinflammatory cytokines by peritoneal macrophages in mice

Introduction

Methotrexate (MTX), an anti-inflammatory and anti-cancer agent, was shown to improve clinical symptoms of rheumatoid arthritis (RA) as shown for the first time in 1951 [1].
Recently, methotrexate (4-amino-N10-methyl­ptero­y­l­glutamic acid) is the most frequently used anti-rheumatic drug [2, 3]. Therapeutic effect of MTX in inflammatory rheumatic conditions and in animal models of experimental arthritis seems to be associated with its anti-inflammatory properties rather than with anti-proliferative mechanism of action. Several pharmacological mechanisms of action of MTX have been suggested so far, however, despite well documented facts there are still many controversies. There is some evidence, that MTX, a folate antagonist, may prevent de novo pyrimidine and purine synthesis consequently inhibiting production of DNA, RNA and proliferation of cells [4]. Most studies describe a dose-dependent induction of apoptosis and/or reduction of cell proliferation by MTX [5]. It has been suggested that MTX induces apoptosis only in activated cells [6]. Interestingly, MTX differentially affects monocytic cell lines and lymphocytes [5, 7].
In RA the rapid clinical remission after MTX application may indicate endogenous alterations in the production of cytokines [4, 8, 9] and/or humoral responses [10]. Methotrexate reduces in vivo the production of many cytokines (TNF-, IL-4, IL-6, IFN-, and GM-CSF) [4, 8, 9]. On the other hand, MTX hardly affects cytokine production by monocytes in vitro [8, 11].
Moreover, it has been shown that low-dose MTX-treatment of RA patients results in the reduced serum concentration of TNF-, while ex vivo there is no difference in the production of TNF- by PHA-stimulated peripheral blood mononuclear cell (PBMC) taken from RA patients and controls [12]. In addition, Aggarwal and coworkers [13] showed that 4 weeks MTX treatment has no effect on the spontaneous production of cytokines in whole-blood cultures, however after LPS stimulation IL-6 levels were lower in MTX treated patients. MTX does not affect LPS-induced release of IL-1 and TNF- by monocyte cell lines [14]. Methotrexate also does not control in vitro LPS induced IL-1 synthesis by resident peritoneal macrophages [15].
As many studies on methotrexate immunosuppressive properties seem to be inconclusive and generally demonstrate little or no effect of MTX on cytokine production in vitro, it rises the questions whether is it a matter of:
• different target cells for MTX (macrophages vs. T cells or synoviocytes as a source of cytokines),
• expression of different receptors for MTX in distinct inflammatory loci,
• different regimes of MTX treatment used; e.g. low-dose (anti-inflammatory) vs. high-dose (anti-cancer) treatment.
Therefore, in our study we have addressed the issue whether expected beneficial therapeutic effect of high-dose MTX treatment in CIA, a murine model of RA, will be accompanied by suppression of cytokine production by any inflammatory cells, including cells not involved directly in the pathogenesis of arthritis. First of all, we wanted to explain whether the production of proinflammatory cytokines by peritoneal macrophages is resistant to MTX, both in vitro and in vivo.

Material and methods

Reagents

Collagen type II (CII) from chicken sternal cartilage, Complete Freund’s Adjuvant (CFA), methotrexate (MTX), lipopolysaccharide (LPS) from Escherichia coli 0111:B4, myeloperoxidase (MPO) from human leukocytes, hexade­cyltrimethylammonium bromide (HTAB), o-dianisidine dihydrochloride (OPD), hydrogen peroxide were all from Sigma-Aldrich (USA). Horseradish peroxidase (HRP) conjugated streptavidin was obtained from Vector (USA) and bovine serum albumin (BSA) fraction V was from Roche Diagnostics (Germany).
Anti-mouse cytokine primary monoclonal antibodies: rat anti-mouse IL-6 monoclonal antibody, rat anti-TNF- monoclonal antibody, rat anti-mouse IL-12p40 monoclonal antibody were all from e-Bioscience (USA). OptEIA mouse IL-10 set was from BD PharMingen (USA). Biotin conjugated antibodies: anti-IgM, anti-IgG2a were from SouthernBiotech (USA), anti-IgG1 from MP Biomedicals (USA) and anti-IgG was from Sigma-Aldrich (USA). Cytokines: recombinant mouse IL-6, recombinant mouse IL-12p40, recombinant mouse TNF- were all from e-Bioscience (USA).

Mice

Inbred DBA/1J male mice and CBA mice were bred in the Animal Breeding Unit, Department of Immunology, Jagiellonian University College of Medicine, Krakow. Mice were housed 4-5 per cage and maintained under clean conventional conditions with free access to standard rodent diet and water. Mice were used at 8 to 10 weeks of age. The authors were granted permission by the Local Bioethical Committee to use mice in presented study. Experiments were conducted according to ethical guidance of Local Bioethical Committee.

Induction and evaluation of collagen induced arthritis

To induce collagen induced arthritis (CIA), DBA/1J mice were injected intradermally with chicken collagen type II (CII, 200 µg/mouse) emulsified in Complete Freund’s Adjuvant (CFA). 21 days later mice received boost intraperitoneal injection of CII (100 µg/mouse) in the presence of LPS (5 µg/mouse). Starting from the second (booster) injection of collagen, mice were examined visually every other day for the incidence and severity of arthritis (joint swelling and redness). Paw thickness was measured using Mitutoyo micrometer. The lesions of the four paws were each graded from 0 to 4 (CIA index) according to the increasing extend of erythema and edema of the periarticular tissues.

Methotrexate treatment of collagen induced arthritis

Methotrexate (50 µg/mouse = ~2,5 mg/kg) was given by intravenous injections three times a week for 3 weeks, starting on day 21, the day of second immunization (see Protocol – Fig. 1). Control animals (referred to as untreated) received intravenous injection of saline. The regimen was chosen to test the effect of high-dose MTX on cytokine production by macrophages and was based on the protocol described by Neurath et al. [16].

Measurement of myeloperoxidase activity

On day 42 (the end of experiment) hind paws were collected and homogenized. Myeloperoxidase (MPO) activity in articular tissues was measured according to Bradley’s method [17]. The activity of MPO was calculated from the MPO standard curve and expressed in units. One unit of MPO activity was defined as that degrading 1 µmol of hydrogen peroxide per minute at room temperature. Each sample was measured in duplicate.

Measurement of serum anti-collagen antibody titers

The blood was collected from mice on day 35 and 42 and the serum was prepared. Anti-CII antibodies in serum were measured by sandwich ELISA as described previously [18]. Shortly, plates (Costar EIA/RIA High Binding plates, Corning Incorporated, USA) were coated with native chicken collagen type II (5 µg/ml). Mouse serum diluted geometrically in PBS was applied into collagen coated wells followed with biotin-conjugated antibodies against IgM, IgG, IgG1 and IgG2a. Horseradish peroxidase conjugated streptavidin was used as detection reagent followed with the o-phenylenediamine dihydrochloride (OPD) peroxidase substrate. The antibody levels were expressed in arbitrary ELISA units calculated from anti-CII immunoglobulin titer: 1 Unit = 1/100 titer of immunoglobulin specific to native chicken collagen type II.

Measurement of cytokine production by peritoneal residual cells taken from DBA/1J mice

On day 42 (the end of experiment) peritoneal residual cells were collected and cultured at a cell density of 5 × 105/ml for 24 h with or without LPS (100 ng/ml) (E. coli 0111:B4 cell wall component). The concentration of cytokines (TNF-, IL-6, IL-10, IL-12p40) in culture supernatants was measured by sandwich cytokine ELISA.

Short-term in vivo methotrexate treatment and its effect on ex vivo cytokine production by peritoneal macrophages of CBA mice

Methotrexate (50 µg/mouse) was injected intravenously on day 0, 2, 4 (7.5 mg/kg/week) to CBA mice, (referred to as MTX treated mice). Control mice were injected with saline (referred to as untreated mice). Thioglycollate was given by intraperitoneal injection on day 4 (the day of the last injection of MTX). Three days later peritoneal exuded cells (mostly macrophages) were collected and cultured 24 h in vitro with or without LPS (100 ng/ml). The cytokine production was tested as described above.

In vitro effect of MTX on cytokine production by macrophages

A. Peritoneal macrophages

Naïve CBA mice were injected intraperitoneally with thioglycollate. Three days later peritoneal macrophages were collected and cultured 24 h in vitro with different concentrations of MTX (0, 1, 10, 100 µg/ml), with or without LPS (100 ng/ml) or cells were cultured with indicated amounts of MTX for 24 h and then medium was replaced and LPS was added to the cultured cells for additional 24 h. Cytokine production was measured in culture supernatant by ELISA.

B. RAW 264.7 – macrophage cell line

RAW 264.7 cells were cultured in vitro in the presence of LPS (100 ng/ml) and different concentrations of MTX (0, 1, 10, 100 µg/ml) for 24 h or cells were cultured for 24 h in the presence of MTX and then stimulated with LPS for additional 24 h. Cytokines were measured in the culture supernatant by ELISA.

Analysis of cytokine production in macrophage culture supernatant by ELISA

Cytokines in the culture supernatant were measured by sandwich cytokine ELISA.

Statistical analysis

Student’s t-test was used to determine the significance of group differences. A p-value less than 0.05 was considered significant.

Results

The effect of methotrexate treatment on the development of CIA in DBA/1J mice

As expected, in mice treated with MTX (7.5 mg/kg/week) the incidence of disease and the severity of arthritis evaluated by the mean arthritis index, paw thickness and MPO activity in the articular tissues were significantly reduced (Fig. 2). Methotrexate treatment markedly decreased the serum levels of anti-CII antibodies (IgM, IgG, IgG2a and IgG1). The production of both anti-CII IgG2a, the indicator of Th1 response, and anti-IgG1, the indicator of Th2 response, were inhibited by high dose of MTX to the same extend (Fig. 3).

The effect of methotrexate treatment on the production of cytokines by residual peritoneal cells of DBA/1J mice

Methotrexate treatment did not affect the spontaneous ex vivo cytokine production by residual peritoneal cells of DBA/1J mice (data not shown). Surprisingly, cells taken from MTX-treated mice and stimulated in vitro with LPS, showed higher production of proinflammatory cytokines (TNF-, IL-6 and IL-12p40) as compared to the peritoneal cells of untreated mice, while IL-10 production was not affected by MTX-treatment (Fig. 4).

The effect of short time in vivo methotrexate treatment on the cytokine production by thioglycollate-induced peritoneal macrophages of CBA mice

Short term in vivo MTX treatment did not affect the spontaneous ex vivo cytokine production by thioglycollate-induced peritoneal macrophages. However, in response to in vitro stimulation with LPS of cells taken from mice treated with MTX in vivo (3 intravenous injections of MTX, 50 µg/mouse) produced more proinflammatory cytokines (IL-6, TNF- and IL-12p40) than cells taken from untreated mice (saline injected mice) (Fig. 5).

In vitro effect of methotrexate on cytokine production by thioglycollate-induced CBA peritoneal macrophages

When thioglycollate-induced peritoneal macrophages from naïve CBA mice were cultured in vitro with LPS and different concentrations of MTX (1-100 µg/ml), MTX had no effect on LPS-induced cytokine production by these cells (Fig. 6A). However, when the thioglycollate-induced peritoneal macrophages were pretreated with MTX in vitro for 24 h and then stimulated with LPS, they produced more proinflammatory cytokines (IL-6, TNF- and IL-12p40). The level of IL-10 was not affected by MTX treatment in vitro (Fig. 6B).

In vitro effect of methotrexate on cytokine production by RAW 246.7 macrophage cell line

In the culture of RAW 264.7 cells stimulated in vitro with LPS MTX at low concentration (1 or 10 µg/ml) did not affect the production of proinflammatory cytokines (IL-6 and TNF-) by that macrophage cell line. However, in the presence of high concentration of MTX (100 µg/ml) LPS stimulated RAW 264.7 cells produced less TNF- and IL-6 as compared to MTX-untreated cells. The level of IL-10 production by RAW 264.7 cells was decreased by MTX in a dose dependent manner (Fig. 7A). When the RAW 264.7 cells were treated with MTX (24 h culture in vitro) prior to stimulation with LPS they produced significantly less cytokines tested (IL-6, TNF- and IL-10; IL-12p40 was not detectable) (Fig. 7B).

Discussion

Methotrexate is the most commonly used slow acting immunosuppressive drug for RA and other autoimmune diseases. In contrast to anticancer treatment (e.g. in acute lymphoblastic leukemia), very low doses of MTX are given weekly in RA (15-30 mg/week  0.3 mg/kg), as a standard procedure [2]. However, high-dose MTX has been also used in arthritis in the treatment of refractory juvenile rheumatoid arthritis [19]. The beneficial therapeutic effect of MTX in RA has been shown to be associated with anti-inflammatory properties of the drug, especially with the inhibition of proinflammatory cytokines such as TNF- [8].
Collagen-induced arthritis (CIA) in mice is a commonly used animal experimental model for studying arthritis, the most relevant model to RA in humans. Collagen-induced arthritis has been also used to examine the immuno­suppresive and anti-inflammatory properties of MTX. The effects of MTX treatment on both clinical symptoms and ex vivo activities of immune cells (macro­phages, T cells, synoviocytes, epithelial cells) has been studied in many laboratories. It is a common agreement that MTX ameliorates the development CIA and is accompanied with inhibition of the production of proinflammatory mediators (cytokines, eicosanoids, NO). However, the major target cell for MTX, the producer of cytokines, is still controversial. Especially the data showing the effect of MTX on macrophage activity ex vivo and in vitro are not consistent. It has been shown that MTX does not affect, or only slightly inhibits, the cytokine production by macrophages, while it strongly inhibits the production of T-cell cytokines [4, 15]. On the other hand, low-doses of MTX significantly diminished the production of TNF-, IL-1 and IL-6 by human synovial macrophages [20]. The differences between in vivo effects of MTX on human (RA) and murine (CIA) macrophages may be explained by treatment of mice with the doses of MTX not relevant to those used for treatment of RA patients.
In this study we have tested the MTX effect on cytokine production by macrophages in three experimental systems:
• MTX treatment of CIA in DBA/1J mice and ex vivo cytokine release by residual peritoneal cells,
• short-term in vivo MTX treatment of CBA mice and in vitro examination of thioglycollate-induced peritoneal macrophages,
• in vitro incubation of MTX with either CBA mice peritoneal macrophages or with RAW 264.7 cells (murine macrophage cell line).
We have used high-dose MTX treatment in vivo (7.5 mg MTX/kg/week) and in vitro (1, 10, 100 µg of MTX/ml), according to the protocols used previously by Neurath, to test MTX specific effects on proinflammatory cytokine production in healthy and arthritic (CIA) mice [16].
Our studies clearly indicate that intravenous MTX administration markedly ameliorates arthritis in CIA-induced mice. The attenuation of clinical symptoms was associated with the profound inhibition of anti-collagen II immunoglobulin production. This therapeutic effect was similar to that observed by Neurath et al. after intraperitoneal administration of MTX [16]. There was no difference between the effect of MTX on Th1 and Th2 related IgG subclasses, IgG2a and IgG1, respectively, in contrast to other studies, in which MTX selectively inhibited Th1 response [21, 22]. It has been found that MTX suppresses production of proinflammatory cytokines such as IL-6, IL-1, TNF- and IFN- in vitro, ex vivo and in vivo, while production of the anti-inflammatory Th2 IL-4 cytokine was less affected [11]. Moreover, the production of T-cell cytokines (TNF-) was more affected than the production of TNF- by macrophages. It was the effect of MTX on spleen cells taken from arthritic mice and activated in vitro with either T-cell (anti-CD3/CD28 antibodies) or macrophage (LPS – TLR4 ligand) stimuli [16, 23]. These data suggest that TNF- production by T cells is an important target of MTX. In the present study we addressed the question whether macrophages are the target cells for MTX in arthritis. Surprisingly, we have found that MTX treatment, starting prior to the CIA onset, resulted in the enhancement of cytokine production (TNF-, IL-6, IL-12) by peritoneal cells stimulated in vitro with LPS, as compared to cytokine production by cells taken from untreated DBA/1J mice. This effect was also observed when thioglycollate-induced peritoneal macrophages were taken from CBA mice treated with MTX and compared with control macrophages. Interestingly, the production of proinflammatory cytokines (TNF-, IL-6 and IL-12) was more affected than that of anti-inflammatory IL-10. The effect was not limited to LPS stimulation because the same results were seen after the stimulation of macrophages with zymosan (data not shown). These data suggest indirect in vivo effect of MTX on macrophages. One may speculate that in vivo MTX selectively inhibited some undetected macrophage suppressor(s), what resulted in an enhancement of the production of proinflammatory mediators such as TNF-, IL-6 and NO (not shown). On the other hand, there exists a great body of evidence that MTX directly inhibits proinflammatory functions of macrophages. It has been shown that suppression of inflammation (thioglycollate-induced peritonitis) by low-dose MTX is mediated by activation of adenosine A2A receptor in peritoneal cells and by reduced leukocyte accumulation at a site of inflammation. However, it has been also postulated that the anti-inflammatory effects of MTX are mediated by different receptors in different inflammatory loci or on different inflammatory cells [24]. The latter hypothesis may explain the observed dual effect of MTX treatment of CIA, the anti-inflammatory effect on the development of arthritis and the “proinflammatory” effect on peritoneal cells in DBA/1J mice. To investigate the direct effect of high-dose MTX on cytokine production by macrophages we extended our studies on in vitro experimental model. Interestingly, MTX (1-100 µg/ml), added to CBA mice peritoneal macrophages simultaneously with LPS, did not inhibit the production of proinflammatory cytokines and in some experiments it even slightly enhanced this production. Finally, we have tested in vitro the influence of MTX on RAW macrophages. We have used macrophage cell line for two reasons. First, we wanted to avoid the presence of additional inflammatory cells, which could be the primarily target of MTX. Second, in contrast to peritoneal macrophages, RAW macrophages are proliferating cells. Therefore, MTX may halt proliferation of these dividing cells at G1 phase of the cell cycle, similarly to its action on macrophage synoviocytes in arthritic joints. We have found that MTX markedly suppresses production of proinflammatory cytokines when pre-incubated with proliferating RAW cells for 24 h before their stimulation with either LPS or zymosan (data not shown).
In conclusion, the data reported here provide evidence that high-dose MTX-treatment markedly ameliorates arthritis while it does not suppress the cytokine release from peritoneal macrophages. It may suggest different biological targets for MTX (e.g. synoviocytes vs. peritoneal macrophages). Moreover, the in vitro study clearly indicates that MTX can inhibit the cytokine production if target cells proliferate. However, further studies are necessary to explain the mechanism of MTX-dependent sensitization of peritoneal macrophages to enhanced production of proinflammatory mediators.

Acknowledgments

This study was supported by Jagiellonian University College of Medicine grants: No K/ZDS/000684 and No K/ZDS/000678.

References

 1. Gubner R, August S, Ginsberg V (1951): Therapeutic suppression of tissue reactivity. II. Effect of aminophen in rheumatoid arthritis and psoriasis. Am J Med Sci 221: 176-182.  
2. Swierkot J, Szechinski J (2006): Methotrexate in rheumatoid arthritis. Pharmacol Rep 58: 473-492.  
3. O’Dell JR (1997): Methotrexate use in rheumatoid arthritis. Rheum Dis Clin North Am 23: 779-796.  
4. de Lathouder S, Gerards AH, Dijkmans BA, Aarden LA (2004): Two inhibitors of DNA-synthesis lead to inhibition of cytokine production via a different mechanism. Nucleosides Nucleotides Nucleic Acids 23: 1089-1100.  
5. Phillips DC, Woollard KJ, Griffiths HR (2003): The anti-inflammatory actions of methotrexate are critically dependent upon the production of reactive oxygen species. Br J Pharmacol 138: 501-511.  
6. Genestier L, Paillot R, Fournel S, et al. (1998): Immunosuppressive properties of methotrexate: apoptosis and clonal deletion of activated peripheral T cells. J Clin Invest 102: 322-328.  
7. Herman S, Zurgil N, Deutsch M (2005): Low dose methotrexate induces apoptosis with reactive oxygen species involvement in T lymphocytic cell lines to a greater extend than in monocytic lines. Inflamm Res 54: 273-280.  
8. Gerards AH, de Lathouder S, de Groot ER, et al. (2003): Inhibition of cytokine production by methotrexate. Studies in healthy volunteers and patients with rheumatoid arthritis. Rheumatology 42: 1189-1196.  
9. Morita Y, Fukazawa T, Hirashima M, et al. (2006): The effect of methotrexate (MTX) on expression of signaling lymphocytic activation molecule (SLAM) in patients with rheumatoid arthritis (RA) and its role in the regulation of cytokine production. Scand J Rheumatol 35: 268-272.
10. Wijngaarden S, van Roon JA, van de Winkel JG, et al. (2005): Down-regulation of activating Fc-gamma receptors on monocytes of patients with rheumatoid arthritis upon methotrexate treatment. Reumatology 44: 729-734.
11. de Lathouder S, Gerards AH, de Groot ER, et al. (2002): Mycophenolic acid and methotrexate inhibit lymphocyte cytokine production via different mechanisms. Eur Cytokine Netw 13: 317-323.
12. Giacomelli R, Cipriani P, Matucci CM, et al. (2002): Combination therapy with cyclosporine and methotrexate in patients with early rheumatoid arthritis soon exhibits TNF-αlpha production without decreasing TNF-αlpha levels. An in vivo and in vitro study. Clin Exp Rheumatol 20: 365-372.
13. Aggarwal A, Misra R (2003): Methotrexate inhibits interleukin-6 production in patients with juvenile rheumatoid arthritis. Rheumatol Int 23: 134-137.
14. Seitz M, Valbracht J, Quach J, Lotz M (2003): Gold sodium thiomalate and chloroquine inhibit cytokine production in monocytic THP-1 cells through distinct transcriptional and posttranslational mechanisms. J Clin Immnol 23: 477-484.
15. Hu SK, Mitcho YL, Oronsky AL, Kerwar SS (1988): Studies on the effect of methotrexate on macrophage function. J Reumatol 15: 206-209.
16. Neurath MF, Hildner K, Becker C, et al. (1999): Methotrexate specifically modulates cytokine production by T cells and macrophages in murine collagen-induced arthritis (CIA): a mechanism for methotrexate-mediated immunosuppression. Clin Exp Immunol 115: 42-55.
17. Bradley PP, Priebat DA, Christensen RD, Rothstein G (1982): Measurement of coetaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 78: 206-209.
18. Kwaśny-Krochin B, Bobek M, Kontny E, et al. (2002): Effect of taurine chloramine, the product of activated neutrophils, on the development of collagen-induced arthritis in DBA/1J mice. Amino Acids 23: 419-426.
19. Reiff A, Shaham B, Wood BP, et al. (1995): High dose methotrexate in the treatment of refractory juvenile rheumatoid arthritis. Clin Exp Rheumatol 13: 113-118.
20. Cutolo M, Capellino S, Montagna P, et al. (2006): Anti-inflammatory effects of leflunomide in combination with methotrexate on co-culture of T lymphocytes and synovial macrophages from rheumatoid arthritis patients. Ann Rheum Dis 65: 728-735.
21. Yamaki K, Uchida H, Harada Y, et al (2003): Effect of methotrexate on Th1 and Th2 immune responses in mice. J Pharm Pharmacol 55: 1661-1666.
22. Herman S, Zurgil N, Langevitz P, et al. (2008): Methotrexate selectively modulates Th1/Th2 balance in active rheumatoid arthritis patients. Clin Exp Rheumatol 26: 317-323.
23. Hildner K, Finotto S, Becker C, et al. (1999): Tumor necrosis factor (TNF) production by T cell receptor-primed T lymphocytes is a target for low dose methotrexate in rheumatoid arthritis. Clin Exp Immunol 118: 137-146.
24. Montesinos CM, Desai A, Cronstein BN (2006): Suppression of inflammation by low-dose methotrexate is mediated by adenosine A2A receptor but not A3 receptor activation in thioglycollate-induced peritonitis. Arthritis Res Therapy 8: 1-7.