Detection of matrix metalloproteinase (MMP) activity in supernatants from in vitro–activated T lymphocytes from patients with multiple sclerosis by means of gelatin zymography. Zone of clearing represents proteinase activity at levels of 92 and 72 kDa, indicative of MMP-9 and MMP-2, respectively. The MMPs incubated with simvastatin (statin) during development show increased proteolytic activity, whereas incubation with interferon beta-1b reduced enzymatic gelatinolysis. Control specimens were developed in conventional buffer only.
Quantitative analysis of gelatinolytic activity in supernatants after stimulation with phytohemagglutinin (PHA) or an antibody to CD3 (CD3). Peripheral blood mononuclear cells were obtained from patients with multiple sclerosis and, for comparison, from healthy individuals. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis zymograms were developed with simvastatin (statin), interferon beta-1b, or buffer (control). Densitometric quantitation was performed at the sizes of 72 and 92 kDa, indicative of matrix metalloproteinase (MMP)–2 and MMP-9. Asterisk indicates P<.01; dagger, P<.05; OD, optical density.
Bernd C. Kieseier, Juan J. Archelos, Hans-Peter Hartung. Different Effects of Simvastatin and Interferon Beta on the Proteolytic Activity of Matrix Metalloproteinases. Arch Neurol. 2004;61(6):929–932. doi:10.1001/archneur.61.6.929
Interferon beta and statins are known to exert immunomodulatory actions by inhibiting gene transcription and translation of various proinflammatory molecules. Although interferon beta represents a mainstay in therapy for multiple sclerosis (MS), statins are considered a potential new approach in treating MS.
To investigate the effect of interferon beta and statins on the posttranslational activity of matrix metalloproteinases (MMPs) released from mononuclear cells in vitro that were obtained from patients with MS and healthy donors.
Blood samples from 10 patients with a relapsing-remitting course of MS and 5 matched healthy individuals were studied. Peripheral blood mononuclear cells were isolated and stimulated with an antibody to CD3, phytohemagglutinin, or medium alone as a negative control. Proteolytic activity was investigated in the supernatants by means of sodium dodecyl sulfate–polyacrylamide gel electrophoresis zymography in which gels were incubated with interferon beta or simvastatin during development. Zones of gelatin digestion were visualized and quantified.
Incubation with interferon beta resulted in an inhibition of the gelatinolytic activity of MMP-9 and MMP-2. In contrast, simvastatin enhanced the proteolytic capacity of MMP-2 and MMP-9, with a statistically significant increase of MMP-2 activity when compared with findings in controls. There were no differences in the enzymatic response between patients with MS and healthy individuals.
Interferon beta exhibits inhibitory effects at the posttranslational level of MMP activity, whereas simvastatin augments the proteolytic activity of MMP-2 and MMP-9, suggesting that statins exert anti-inflammatory and proinflammatory effects. This dual mechanism of action should be considered, given the recent interest in developing these drugs for treatment of MS.
Interferon beta represents a mainstay in the current management of multiple sclerosis (MS). The annualized relapse rate of drug-treated patients is lower compared with that of placebo-treated control subjects, and more drug-treated patients remain relapse free for several years relative to untreated cohorts.1 Although interferon beta ultimately decreases inflammation within the central nervous system (CNS) and normalizes some of the aberrant immune responses thought to contribute to MS pathogenesis, its precise mechanism of action remains elusive.2 Since interferon beta does not stop progression of the disease, increasing scientific efforts are invested to develop more effective therapeutic strategies. Recent evidence suggests that the 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) exhibit immunomodulatory activities that might be beneficial in MS,3,4 and that statins may even be neuroprotective.5 In experimental autoimmune encephalomyelitis, an animal model of MS, lovastatin partially suppressed acute disease, and atorvastatin calcium prevented or reversed chronic and relapsing disease.6- 9 Recent studies on in vitro–activated T lymphocytes from patients with MS and healthy individuals showed that simvastatin possesses a broad spectrum of anti-inflammatory properties.10 However, the molecular mechanisms subserving these immunomodulatory activities have not been identified. Most of our present knowledge is based on findings along the level of the transcription or translation of various proinflammatory and anti-inflammatory molecules. In this study, we investigated the direct effects of interferon beta and simvastatin on the protein level by studying the proteolytic activity of matrix metalloproteinases (MMPs), a group of enzymes implicated in critical steps during the pathogenetic evolution of MS such as blood-brain barrier disruption, the trafficking of immunocompetent cells into the CNS, and demyelination.11,12 Several investigators detected elevated levels of MMP messenger RNAs as well as proteins in tissue samples, blood, and cerebrospinal fluid of patients with MS,13 raising the possibility of using MMP levels, especially MMP-9 levels, as a surrogate marker for disease activity in MS, and defining MMPs as specific targets for treatment in MS.14
All patients were recruited from the outpatient Multiple Sclerosis Clinic of the Department of Neurology, Heinrich-Heine-University, Düsseldorf, Germany. Ten patients with MS who were diagnosed according to the criteria of McDonald et al15 and who had a relapsing-remitting course of the disease were included (female-male ratio, 8:2; mean ± SE age, 30.4 ± 3.1 years; mean ± SE Kurtzke Expanded Disability Status Scale score,16 2.5 ± 0.7; mean ± SE disease duration, 4.4 ± 1.4 years) and had not received immunosuppressive therapy or steroid treatment in the 6 months before blood drawing. The control group consisted of 5 age- and sex-matched healthy individuals who were not receiving any medication known to affect the immune system. All individuals gave informed consent.
Peripheral blood specimens were aseptically collected by means of standard venipuncture into vacuum tubes containing heparin sodium as anticoagulant. Systemic infection was excluded by normal findings for white blood cell counts, C-reactive protein levels, and serologic markers for infection.
Peripheral blood mononuclear cells were isolated on a discontinuous density gradient (Lymphoprep; Nycomed, Oslo, Norway). Cells were plated in 96-well microtiter plates at a density of 2 × 106 cells/mL at 100 µL/well in serum-free defined medium (Aim V; Life Technologies GmbH, Karlsruhe, Germany) and stimulated with an antibody to CD3 (0.1 µg/mL; Immunotech, Marseille, France), phytohemagglutinin (5 µg/mL; Sigma-Aldrich Chemie GmbH, Munich, Germany), or medium alone as a negative control as described elsewhere.10 After 24 hours, cell-free supernatants were subjected to gelatin zymography.
Gelatinase activity was determined by means of sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis zymography as described elsewhere.17 Briefly, 25 µL of cell-free supernatants was incubated with 25 µL of Tris/glycine-SDS sample buffer (Novex, San Diego, Calif) and applied to a 10% (wt/vol) polyacrylamide-resolving gel containing 0.1% SDS and 0.1% gelatin type A from porcine skin (Sigma-Aldrich Corp, St Louis, Mo). Stacking gels were 3% (wt/vol) polyacrylamide. After electrophoresis, the gels were washed in renaturing buffer (Novex) containing Triton X-100 to remove any SDS and incubated in developing buffer (Novex) for 18 hours at 37°C. Gels loaded with the same samples were incubated in developing buffer without or with 10µM simvastatin or interferon beta-1b (2000 U/mL; Betaferon; Schering AG, Berlin, Germany). Gels were stained for 6 hours in 30% methanol/10% acetic acid containing 0.5% (wt/vol) Coomassie brilliant blue G-250 and destained in the same buffer without dye. Gelatinase activity was detected as unstained bands on a blue background representing areas of gelatin digestion. As a negative control, gels were incubated with 10M EDTA before the activation of gelatinases in parallel.
Images of gels were captured by scanning on a flatbed scanner. A standard curve was obtained for densitometric quantitation of MMP activity using purified MMP-2 and MMP-9 (Chemicon International, Inc, Temecula, Calif). In each sample, MMP-2 and MMP-9 activities were calculated using electrophoretic gel lane calculation software (TINA 2.0; Raytest, Straubenhardt, Germany) after image inversion.
We used 1-way analysis of variance and the Newman-Keuls multiple comparisons test as principal statistical tests. A P value of less than .05 was considered significant. Unless otherwise indicated, data are expressed as mean ± SE.
Proteolytic activity was detectable at molecular weights of 92 and 72 kDa, indicative of MMP-9 and MMP-2, respectively, in the supernatants from all samples studied. Incubation with interferon beta-1b markedly reduced the proteolytic activity of MMP-9, whereas the effect on MMP-2 did not reach statistical significance. In contrast, incubation with simvastatin resulted in increased sizes of the bands at 92 and 72 kDa, pointing to enhanced activation of MMP-9 and MMP-2 in comparison with controls (Figure 1). Quantitative analysis of the zones of gelatinolysis confirmed the rise in proteolytic activity, especially of MMP-2, which was statistically significant compared with findings in controls, whereas treatment with interferon beta-1b clearly diminished MMP activity (Figure 2). This observation was consistent, and no differences between patients and healthy controls could be discerned. Moreover, the results were similar when comparing supernatants from cells stimulated with phytohemagglutinin or an antibody to CD3. Negative controls, ie, gels incubated with EDTA before the activation of gelatinases, showed no lysis areas, indicating total abrogation of gelatinase activity.
The MMPs represent a family of enzymes involved in extracellular matrix remodeling in a variety of physiologic and pathologic conditions. Expression and proteolytic activity of MMPs are strictly regulated at the transcriptional level by various stimuli, such as inflammatory cytokines and growth factors, whereas posttranscriptional regulation includes zymogen activation and inhibition of MMP activation and proteolytic activity through binding to endogenous tissue inhibitors of metalloproteinases.18
Recent in vitro studies suggest that MMP-9 promotes migration of T lymphocytes across basement membranes,19 emphasizing an important role of this MMP in the process of T-cell migration from the blood into the CNS. Two studies independently demonstrated that interferon beta ameliorates T-cell migration across an in vitro blood-brain barrier model by suppressing MMP-9 production.20,21 Therefore, interferon beta exhibits down-regulatory effects on MMP-9 gene transcription/translation, because the total amount of the protease secreted was found to be reduced. In the present study, we demonstrate that interferon beta exerts direct effects on secreted MMP-9 by further reducing the gelatinolytic activity of this protease. Hence, interferon beta inhibits MMP-9 activity at the level of gene transcription as well as posttranslationally. The effect of interferon beta on the gelatinolytic activity of MMP-2, as measured in our study, did not reach statistical significance compared with that in controls. Thus, interferon beta seems to selectively inhibit MMP-9 but not MMP-2 when proteolytic activity is measured by gelatin zymography. Our present observations support the notion that this drug mediates its effect in part by reducing MMP activity and, as such, modulates the trafficking of inflammatory cells through the blood-brain barrier into the CNS. Ex vivo data support this notion, providing evidence that interferon beta therapy results in decreased serum contents of MMP-9 in patients with MS.22
Statins have a large number of immunoregulatory properties and, like interferon beta, have been reported to diminish the secretion of MMP-9 by monocytes.23,24 We were able to confirm this observation on the level of MMP-9 gene transcription, demonstrating a reduction in the release of inactive MMP proforms.10 In contrast, our present data show that on the posttranslational level simvastatin augments the proteolytic activity of MMPs, in particular of MMP-2, suggesting that statins exhibit not only anti-inflammatory but also proinflammatory effects. This observation is puzzling. Why statins on the one hand reduce MMP secretion but on the other hand increase their proteolytic activity cannot be explained. However, along this line, statins have been reported to increase the expression of proinflammatory cytokines such as interferon γ25 and to lower the expression levels of anti-inflammatory cytokines such as interleukin 10.10 Other investigators, however, found that statins precipitate the secretion of anti-inflammatory helper T cell (TH)2–type cytokines and promote the differentiation of TH0 cells into TH2 cells.7 It is unclear at present whether technical issues, species differences, or differential regulatory events in diverse biological situations account for these discrepancies. Our in vitro data find support in a recently published study of patients with hypercholesterolemia, in which while receiving therapy with pitavastatin, a newly developed statin, a significant elevation in MMP-2 plasma levels was observed, whereas no alterations were detectable for MMP-9.26
The inhibitory effect of interferon beta-1b was most pronounced at concentrations above 1000 U/mL, as we determined in dose-response experiments (data not shown). A concentration of 2000 U/mL, as applied in the present study, corresponds to 3nM or to the 40-fold steady state concentration of the drug, which usually ranges from 40 to 80 U/mL in patients receiving 8 × 106 U subcutaneously 3 times a week.27,28 In contrast, no large differences were observed in the efficacy of simvastatin at concentrations in the range of 1µM to 10µM, as measured in the same dose-response experiments. The 10µM concentration used in the present study corresponds to the 250-fold steady-state plasma concentration of statins after oral intake.29 It is conceivable that simvastatin might cross the blood-brain barrier because of its small size of 4.19 kDa, via passive diffusion, whereas interferon beta, with a size of 20 kDa, might enter the CNS only at the time of acute blood-brain barrier damage. Thus, statins might exhibit their proproteolytic effect on MMPs at any time during the course of the disease within the CNS.
How simvastatin and interferon beta mediate their direct effects on released MMPs remains unclear at present. We were unable to determine the specific underlying mechanism; we could only speculate as to whether the effects are mediated by a direct binding to MMPs or by other mediators, although the cell-culture medium used was serum free. The effects of simvastatin and interferon beta were completely abolished by the chelating properties of EDTA, indicating that the zinc-binding site of the MMPs remains available and does not represent the target site of simvastatin or of interferon beta. Nevertheless, the effect seen in our study was reproducible in patients with MS and in controls, pointing to a general effect of these drugs not specifically related to disease. To what extent such a direct effect of statins on released MMPs might be relevant in propagating the inflammatory immune cascade in MS is difficult to determine. However, when exploring statins as a monotherapeutic approach in treating MS, such potentially deleterious effects should be taken into consideration. Similarly, combination therapy for MS using interferon beta and statins should be approached with caution. Our data suggest that, at least on the level of MMP activity, the immunological effects of both drugs may be antagonistic. Recently, MMP-9 has been reported to proteolytically cleave interferon beta, resulting in a loss of its antiviral and immunomodulatory activity.30 Although this has only been shown for MMP-9, it remains possible that other MMPs, especially other gelatinases such as MMP-2, might exhibit similar effects because of their broad spectrum and overlap in substrates. Thus, any increase in MMP-2 activity would diminish the effect of interferon beta. Further studies are clearly warranted to elucidate the immunomodulatory potential of statins, and large clinical trials are needed to validate their role as a treatment of inflammatory diseases of the CNS.
Corresponding author and reprints: Bernd C. Kieseier, MD, Research Group for Clinical and Experimental Neuroimmunology, Department of Neurology, Heinrich-Heine-University, Moorenstrasse 5, 40225 Düsseldorf, Germany (e-mail: firstname.lastname@example.org).
Accepted for publication January 16, 2004.
Author contributions: Study concept and design (Drs Kieseier, Archelos, and Hartung); acquisition of data (Drs Kieseier and Archelos); analysis and interpretation of data (Dr Kieseier); drafting of the manuscript (Dr Kieseier); critical revision of the manuscript for important intellectual content (Drs Kieseier, Archelos, and Hartung); statistical expertise (Dr Kieseier); administrative, technical, and material support (Drs Kieseier, Archelos, and Hartung); study supervision (Dr Kieseier).
We thank Heidrun Pischel for the technical assistance. Interferon beta-1b was kindly provided by Schering AG Berlin, Germany. We also thank our patients for their support.