Drs. Bakalkin and Stark contributed equally to this work.
Attenuation of pain and inflammation in adjuvant-induced arthritis by the proteasome inhibitor MG132
Article first published online: 31 MAR 2010
Copyright © 2010 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 62, Issue 7, pages 2160–2169, July 2010
How to Cite
Ahmed, A. S., Li, J., Ahmed, M., Hua, L., Yakovleva, T., Ossipov, M. H., Bakalkin, G. and Stark, A. (2010), Attenuation of pain and inflammation in adjuvant-induced arthritis by the proteasome inhibitor MG132. Arthritis & Rheumatism, 62: 2160–2169. doi: 10.1002/art.27492
Drs. Bakalkin and Stark contributed equally to this work.
- Issue published online: 29 JUN 2010
- Article first published online: 31 MAR 2010
- Manuscript Accepted: 23 MAR 2010
- Manuscript Received: 25 MAR 2009
- Sven Norén Foundation
- Ulla and Gustaf af Ugglas Foundation
- AFA Insurance
- Söderberg Foundations
- Swedish Science Council
In rheumatoid arthritis (RA), pain and joint destruction are initiated and propagated by the production of proinflammatory mediators. Synthesis of these mediators is regulated by the transcription factor NF-κB, which is controlled by the ubiquitin proteasome system (UPS). The present study explored the effects of the proteasome inhibitor MG132 on inflammation, pain, joint destruction, and expression of sensory neuropeptides as markers of neuronal response in a rat model of arthritis.
Arthritis was induced in rats by injection of heat-killed Mycobacterium butyricum. Arthritis severity was scored, and nociception was evaluated by mechanical pressure applied to the hind paw. Joint destruction was assessed by radiologic and histologic analyses. NF-κB DNA-binding activity was analyzed by electromobility shift assay, and changes in the expression of the p50 NF-κB subunit and the proinflammatory neuropeptides substance P (SP) and calcitonin gene-related peptide (CGRP) were detected by immunohistochemistry.
Arthritic rats treated with MG132 demonstrated a marked reduction in inflammation, pain, and joint destruction. The elevated DNA-binding activity of the NF-κB/p50 homodimer and p50, as well as the neuronal expression of SP and CGRP, observed in the ankle joints of arthritic rats were normalized after treatment with MG132.
In arthritic rats, inhibition of proteasome reduced the severity of arthritis and reversed the pain behavior associated with joint inflammation. These effects may be mediated through the inhibition of NF-κB activation and may possibly involve the peripheral nervous system. New generations of nontoxic proteasome inhibitors may represent a novel pharmacotherapy for RA.
Rheumatoid arthritis (RA) is a chronic inflammatory joint disease. Pain and inflammation are the initial symptoms, followed by various degrees of joint destruction. Generally, inflammation is initiated and propagated by the production of cytokines, chemokines, and cell adhesion molecules (1–3). Expression of these molecules is mainly regulated by the NF-κB family of transcription factors (4, 5). Under “normal” conditions, NF-κB, a dimer of p50 and p65 subunits, is present in cells in a latent form, in complex with the inhibitory protein IκB. Cytokines, antigens, or oxidants trigger degradation of IκB, which is mediated by the ubiquitin–proteasome system (UPS), a nonlysosomal proteolytic system of selective protein degradation (6, 7). In turn, NF-κB activates the expression of cytokines and cell adhesion molecules that contribute to the pathogenesis of chronic arthritis (8). The activation of NF-κB has been observed in vascular endothelium and type A synovial lining cells from patients with RA and patients with osteoarthritis (9). Administration of the proteasome inhibitor PS341 attenuated inflammation in the rat model of bacterial cell wall–induced polyarthritis (10).
In RA, pain is generally perceived to arise directly from inflammatory processes, as well as from the activation of central and peripheral neuronal mechanisms (11, 12). The peripheral nervous system participates in nociceptive transmission and proinflammatory processes. Sensory neuropeptides in the target organs enhance vasodilatation and recruitment of inflammatory cells to sites of inflammation (13). In adjuvant-induced arthritis, the sensory neuropeptides substance P (SP) and calcitonin gene-related peptide (CGRP) have been found to be up-regulated in ankle joints and corresponding dorsal root ganglia (14). Capsaicin mitigates the development of arthritis in rats by specifically down-regulating SP and CGRP (15). Consistent with these findings, a significant increase in the levels of SP and CGRP has been detected in the synovial fluid of patients with rheumatic diseases (16).
The UPS degrades intracellular proteins in the central and peripheral nervous systems (17, 18). We and other investigators have previously demonstrated that the proteasome inhibitors MG132 (N-carbobenzoxyl-Leu-Leu-leucinal) and epoxomicin can prevent the development of behavioral signs of neuropathic pain and abolish abnormal pain induced by sustained morphine exposure (19, 20). These compounds inhibited the release of the pronociceptive neuropeptides dynorphin A and CGRP and normalized molecular changes induced by spinal nerve ligation that may contribute to central sensitization, a cellular mechanism that underlies neuropathic pain (19, 20). Thus, the development and maintenance of neuropathic pain critically depends on the regulated UPS-mediated protein degradation in the central nervous system.
The present study was designed to investigate whether the reversible proteasome inhibitor MG132 could attenuate chronic joint inflammation and inflammatory pain in a rat model of adjuvant-induced arthritis.
MATERIALS AND METHODS
A total of 48 female Lewis rats weighing 230–250 gm were used. The rats were housed at a temperature of 21°C, with a 12-hour light/dark cycle, and food pellets and water were available ad libitum. The study was approved by the Ethics Committee for Animal Experimentation, Stockholm North.
Rats were assigned into 4 groups of 12 rats each, consisting of 2 control groups (groups 1 and 2) and 2 arthritic groups (groups 3 and 4), which were treated with vehicle (groups 1 and 3) or with MG132 (groups 2 and 4). Arthritis was induced in 24 of the rats by intradermal injections of a suspension (0.05 ml) of heat-killed Mycobacterium butyricum in paraffin oil (10 mg/ml) into the base of the tail (day 0) under 3–5% isoflurane anesthesia (21). The other 24 rats (controls) received 0.05 ml of paraffin oil alone. On days 12–14, corresponding to the onset of arthritis, each control and arthritic cohort was randomly divided into 2 groups of 12 rats each. One of the control groups and one of the arthritic groups was injected subcutaneously with vehicle (DMSO); the other control group and the other arthritic group were injected with MG132 (1 mg/kg) once daily from day 14 to day 28. MG132 solution in DMSO (Sigma-Aldrich Sweden) was prepared for injections at a concentration of 2 mg/ml. Rats were killed on day 29.
Evaluation of arthritis.
Beginning at arthritis onset (day 14) and on days 20, 24, and 28, the severity of arthritis was assessed visually by two independent observers (ASA and JL) who were unaware of the treatment protocol the rats received. A modified macroscopic scoring system was used to monitor the severity of arthritis, as described previously (22). Briefly, the severity of inflammation in each paw was graded on a scale of 1–3, where 1 = detectable swelling and redness, 2 = moderate swelling and redness, and 3 = severe swelling. The total score was the cumulative value in each of the 4 paws, with a maximum of 12 for each rat. Rats with a total score of > 2 were considered to have arthritis.
Mechanical pain test.
The hind paw withdrawal threshold in response to applied mechanical pressure was assessed with an algesimeter (Ugo Basile) as described elsewhere (23). The hind paw withdrawal threshold was determined on days 0, 4, 12, 14, 16, 20, 24, and 28. A wedge-shaped pusher was applied to the dorsal surface of the hind paw at a steadily increasing pressure. The hind paw withdrawal threshold was determined when the animal removed its foot from the apparatus; the pressure (in grams) necessary to evoke paw withdrawal was recorded. A cutoff threshold of 220 gm was preset to prevent tissue damage. Before measuring the baseline values, the rats were habituated to the algesimeter. All measurements were made in a blinded manner (ASA and JL), and the experimenters had no knowledge of the treatment regimen.
Measurement of body weight.
The total body weight of each animal was measured on days 0, 4, 12, 14, 16, 20, and daily thereafter, from day 21 to day 28.
The integrity of bone tissue was evaluated radiographically on days 14 and 28 in anesthetized rats (3–5% isoflurane). A dental x-ray machine (Heliodent DS; Siemens) with a 1.25-second exposure time and 56 × 76–mm x-ray films (Ektaspeed Plus; Eastman Kodak) were used. The films were developed and scanned with a Hewlett-Packard ScanJet II scanner. Radiographs were evaluated by two observers (MA and JL) who were blinded to the treatment group, using a subjective grading scale ranging from 0 to 3 (0 = normal, 1 = mild, 2 = moderate, and 3 = severely affected joint), as described previously (24). The following 3 parameters were graded: osteoporosis (decreased density of the bone, recognized as increased radiolucency relative to control bone), cartilage loss (narrowing of the joint spaces), and erosion (destruction of bony structure, with increased radiolucency developing at the site of the erosion).
Rats were anesthetized with sodium pentobarbitone (60 mg/kg, intraperitoneally) and perfused with 0.01 moles/liter of phosphate buffered saline (PBS), followed by Zamboni's fixative, consisting of 4% paraformaldehyde in 0.2 moles/liter Sorensen's phosphate buffer, pH 7.3, containing 0.2% picric acid. Both ankle joints were dissected, fixed in Zamboni's fixative for 2 days at 4°C, and then subjected to demineralization in a 4% EDTA solution at pH 7.3 for ∼4 weeks, as described previously (25). Joints were soaked in 20% sucrose in 0.1 mole/liter Sorensen's phosphate buffer, pH 7.2, containing sodium azide and bacitracin (Sigma) for 2 days. The tissues were cut with a Leitz 1720 cryostat into sections measuring 7 μm or 12 μm and were then mounted on Superfrost Plus slides.
Sections measuring 7 μm were stained with hematoxylin and eosin to assess pathologic changes and to monitor the effects of treatment. For the semiquantitative analysis, histologic and immunohistochemical images (see below) of the tissue sections were captured with a video camera (model DEI 750; Optronics) that was attached to the microscope, and the data were stored in a computer.
For semiquantitative analysis of the histologic findings, 2 sections of every ankle joint (n = 6 rats per group) were taken at different depths, with 6 microscopic fields in each section. Thus, in all experimental groups, a total of 12 microscopic fields were analyzed in each ankle joint. The sections were examined by 2 observers (ASA and LH) who were blinded to the treatment regimen. Histologic findings were evaluated using modified grading scales, as previously described (26). Briefly, cartilage destruction and bone erosion were graded on a scale of 0–3, where 0 = no changes, 1 = detectable changes, 2 = moderate changes at few sites, and 3 = severe cartilage destruction and bone erosion. Synovial inflammation was also graded on a 0–3 scale, where 0 = no cellular infiltration, 1 = thicker-appearing synovial membrane and inflammatory cell infiltration, 2 = further thickening of the synovial membrane and moderate increase in inflammatory cells, and 3 = dramatic increase in inflammatory cells and synovial membrane forming pannus (27).
Electrophoretic mobility shift assay (EMSA).
Rats were anesthetized with sodium pentobarbitone (60 mg/kg, intraperitoneally) and decapitated. Both ankle joints were collected and immediately frozen in liquid nitrogen. Frozen tissues were homogenized with a Mikro-dismembrator S (B Braun Biotech), and mixed with buffer C supplemented with 0.2% Nonidet P40 and protease inhibitors at 4°C (28). The homogenate was centrifuged for 10 minutes at 20,000g, and the supernatant was kept at –80°C. For EMSA, 5 μg of the protein extract was incubated with 32P-labeled wild-type κB oligonucleotide; 10 ng of mutant or wild-type κB oligonucleotide was used as competitor. The protein–DNA complexes were separated on “native” polyacrylamide gel, which was then fixed, dried, and subjected to autoradiography. Images were analyzed using Fujifilm Image Gauge software.
Ankle joints were dissected, processed, and sectioned as described for histologic assessment. For assessment of NF-κB p50 staining, 7-μm sections were permeabilized in PBS, incubated with 5% normal goat serum, and then with rabbit polyclonal anti–NF-κB p50 antibody (1:100 dilution, catalog no. sc-114; Santa Cruz Biotechnology). Sections were then incubated with goat anti-rabbit secondary antibody (1:250 dilution; Vector) and with ABC reagent, followed by staining with diaminobenzidine chromogen and counterstaining with hematoxylin QS (both from Vector).
For assessment of SP and CGRP staining, 12-μm sections were incubated with antiserum to SP (1:10,000 dilution) or to CGRP (1:10,000 dilution; Bachem Peninsula Laboratories), as well as with biotinylated goat anti-rabbit antibodies (1:250 dilution; Vector). The immunoreaction was visualized with fluorochrome Cy2-conjugated avidin (1:2,000 dilution; Amersham Life Science). To test the specificity of staining, primary antibodies were omitted or sections were preincubated with blocking peptides.
For semiquantitative analysis of the immunohistochemical findings, 2 sections from each ankle joint were stained, and images of 6 microscopic fields from each tissue section were captured with a DEI 750 video camera that was attached to the microscope, and the data were then stored in a computer. The numbers of p50-positive cells in the cartilage and in the bone were manually counted. The synovium was graded according to the percentage of positive cells, as described previously (26), by two observers (ASA and LH) who were blinded to the treatment group. The immunoreactivity was measured as the area (in mm2), using Easy Analysis software (Bergström Instruments), as previously described (29). A standard lower and upper threshold of fluorescence intensity was consistently applied for positively stained nerve fibers. The results were expressed as the nerve fiber area (in mm2) in relation to the total area of each microscopic field.
The significance of the differences between experimental and control groups was analyzed by one-way analysis of variance, followed by Fisher's protected least significant difference test for quantitative variables. For qualitative variables, the Mann-Whitney U test was applied. Significance was set at P ≤ 0.05.
Reduction of arthritis severity with MG132 treatment.
In rats inoculated with M butyricum, signs of inflammation, including bilateral paw swelling, redness, and warmth, were observed on days 12–14 and beyond. The paws became swollen and gradually increased in volume. Treatment with vehicle or with MG132 was initiated on day 14 and continued until day 28. Arthritis severity reached a maximum value on day 20, and averaged 10.4 ± 1.29 (±SD) in vehicle-treated arthritic rats and 10.1 ± 1.31 in MG132-treated arthritic rats. Between days 20 and 28, the arthritis score remained elevated in the vehicle-treated arthritic rats, while in the MG132-treated arthritic group, it was reduced by 14% on day 24 (P = 0.02) and by 37% on day 28 (P < 0.0001) as compared with the vehicle-treated arthritic groups at the same time points (Figure 1A).
Reversal of pain behavior with MG132 treatment.
On day 14, a significant reduction in the hind paw withdrawal threshold was noted in comparisons between the individual vehicle-treated (P = 0.03) and MG132-treated (P = 0.05) arthritic groups and the vehicle-treated control rats. No significant difference between the control and arthritic groups treated with vehicle or with MG132 was observed on day 20. Between day 20 and day 28, a significant reduction in hind paw withdrawal threshold was seen in the arthritic rats treated with vehicle. On day 24, the hind paw withdrawal threshold was significantly decreased (15%; P = 0.006) in the vehicle-treated arthritic rats as compared with the vehicle-treated control rats and was restored in MG132-treated arthritic rats (23%; P < 0.0001) as compared with the vehicle-treated arthritic rats. On day 28, arthritic rats treated with MG132 demonstrated a 34% increase (P < 0.0001) as compared with vehicle-treated arthritic rats. In the vehicle-treated arthritic group, the decrease in paw withdrawal thresholds persisted and was significantly lower (29%; P < 0.0001) as compared with vehicle-treated control rats on day 28 (Figure 1B). No significant effects of MG132 administration were observed in control rats at any time point.
Gain in total body weight with MG132 treatment.
There was a significant decrease in body weight between days 14 and 20 in vehicle-treated (7.3%; P = 0.0002) and MG132-treated (7.6%; P < 0.0001) arthritic rats as compared with the vehicle-treated control group (Figure 2A). Continuation of MG132 administration until day 28 resulted in a significant increase (4.3%; P < 0.0001) in the weight of arthritic rats as compared with that on day 20. In the MG132-treated arthritic group, there was a 1.7% increase (P = 0.02) as compared with the vehicle-treated arthritic group on day 28 (Figure 2B).
Prevention of joint destruction with MG132 treatment.
Radiographic analysis performed on day 28 revealed no cartilage or bone destruction, nor any other changes in any of the control animals, whereas destructive changes were evident in all arthritic rats. Severe joint destruction, including osteoporosis, loss of joint space, and cartilage and bone erosion were observed in the ankle of the vehicle-treated arthritic rats. Compared with vehicle-treated arthritic rats, MG132 treatment of arthritic rats resulted in a reduction in both the osteoporosis score (by 24%; P = 0.05) and the bone erosion score (by 21%; P = 0.05) (Figures 3A and B). However, MG132 treatment of arthritic rats failed to affect joint space narrowing.
Findings of histologic analyses.
In control animals, the joints were normal, with intact bone and cartilage and no inflammatory cell infiltration in the synovial tissue in either the vehicle-treated or the MG132-treated groups. In arthritic rats, cartilage and bone erosion, as well as thickening of the synovium, with infiltration of inflammatory cells, was more prominent in the vehicle treated group. MG132 treatment of arthritic rats decrease the bone erosion score by 40% (P = 0.02) and reduced inflammatory cell infiltration and synovial thickening by 28% (P = 0.05) as compared with vehicle treatment (Figures 3C and D). However, the cartilage destruction score was essentially not affected.
Effects of MG132 on NF-κB DNA-binding activity in the joints.
EMSA of ankle joints extracts demonstrated the formation of 3 protein complexes with labeled κB oligonucleotide (Figure 4A). The 2 upper complexes corresponded to NF-κB transcription factors as we previously demonstrated in rat tissues and cell lines (30), while the third complex was formed by the protein Ku, a double-stranded DNA end-binding factor (31). In the competition experiment, wild-type, but not mutant, κB oligonucleotide inhibited the formation of NF-κB and the formation of the p50 homodimer complexes with the labeled probe, thus demonstrating the binding specificity (Figure 4B). The DNA-binding activity of NF-κB and of the p50 homodimer was significantly increased in the ankle joints of arthritic rats as compared with the control rats. We observed a 34% increase (P = 0.001) and a 38% increase (P = 0.004) in the DNA-binding activity of NF-κB and the p50 homodimer, respectively, in vehicle-treated arthritic rats as compared with vehicle-treated control rats. The administration of MG132 significantly decreased (by 29% [P = 0.003] and 22% [P = 0.05]) the DNA-binding activity of NF-κB and the p50 homodimer, respectively, in inflamed ankle joints (Figures 4C and D). No significant effect of MG132 treatment on NF-κB or the p50 homodimer was evident in the vehicle-treated control rats.
Effects of MG132 on the expression of the p50 subunit of NF-κB.
In the cartilage of control vehicle-treated and MG132-treated rats, the number of p50-positive cells was 39% and 49%, respectively. Adjuvant-induced arthritis caused a significant 24% increase in this number (P = 0.02), while MG132 treatment of arthritic rats reduced this number by 21% (P = 0.01).
In the synovium, the majority of cells expressed p50; 77% of cells in the vehicle-treated control rats and 70% of cells in the MG132-treated control rats showed positive labeling with anti–p50 antibodies. The number of p50-positive cells was increased by 22% (P = 0.006) in the arthritic rats treated with vehicle as compared with the vehicle-treated control rats. MG132 treatment of arthritic rats reduced the number of p50-positive cells by 5% (not significantly different) (Figures 5A and B). No significant effect of MG132 treatment was observed in control rats. Both the omission of the primary antibodies and the preincubation with antigenic peptide abolished labeling.
Effects of MG132 on sensory neuropeptides.
In control rats, SP-positive nerve fibers were occasionally present in the periosteum and in the synovium of the ankle joints as thin, varicose, nonvascular nerve terminals. Computerized image analysis demonstrated a robust 173% increase (P = 0.05) in SP-like immunoreactivity (SP-LI) in the arthritic ankle joint of rats given vehicle as compared with that in vehicle-treated control rats. In the vehicle-treated arthritic rats, SP-positive nerve fibers were seen in connective tissues and in bone, although at a lower intensity. A significant 73% reduction (P = 0.02) in SP immunoreactivity was observed following MG132 treatment of arthritic rats as compared with vehicle treatment (Figures 6A and B).
CGRP-positive nerve fibers were seen mainly in the periosteum and in the synovium of the ankle joints of control rats as thin, varicose nerve terminals. CGRP-like immunoreactivity (CGRP-LI) was up-regulated by 47% (P = 0.05) in the ankle joints of vehicle-treated arthritic rats as compared with vehicle-treated control rats, whereas administration of MG132 prevented this effect (by 37%; P = 0.02) (Figures 6A and B). In the periosteum of the ankle joints of vehicle-treated arthritic rats, nerve terminals were observed around the blood vessels. No significant effect of MG132 treatment was observed in the control rats. No staining was observed when primary antibody was omitted or blocked with antigenic peptide.
The main findings of this study were that repeated administration of the proteasome inhibitor MG132 reduced both joint inflammation and pain and attenuated molecular and morphologic changes in the rat model of adjuvant-induced arthritis. Effects of the proteasome inhibitor on joint inflammation have been previously described (10, 32), whereas to our knowledge, this is the first report of the actions of the proteasome inhibitor on inflammatory pain and underlying changes in sensory neuropeptides.
We (20) and other investigators (19) have previously demonstrated that proteasome inhibitors both prevent and reverse nerve injury–induced pain behavior and block pathologic pain induced by sustained administration of morphine. These results suggest that the UPS is a critical intracellular regulator of pathologic pain and that UPS-mediated protein degradation is required for maintenance of chronic pain. While the cause and neurobiologic mechanisms underlying neuropathic and inflammatory pain are different, what is common for the effects of proteasome inhibitors in these pathophysiologic conditions is (a) the robust activation of neurotransmission mediated by sensory and pronociceptive neuropeptides, including SP, CGRP, and dynorphins, and (b) the strong, virtually complete inhibition of this activation by inhibitors of the proteasome. The inhibition was evident in both the central and peripheral nervous systems.
Strong evidence from clinical and animal studies indicates that sensory neuropeptides play a role in inflammatory joint disorders, including RA, osteoarthritis, and adjuvant-induced arthritis (14, 16, 33). The elevated expression of the sensory neuropeptides SP and CGRP has been demonstrated to exacerbate pain and inflammation (14, 15, 34, 35). Consistent with these reports, the present study showed a strong up-regulation of SP and CGRP in the periosteum and synovium, structures which are pain-sensitive and prone to inflammation. Prevention of this up-regulation by the administration of MG132 resulted in normalization of pain responses.
Pronounced effects of the proteasome inhibitor MG132 on inflammation were manifested as decreases in the total arthritis index. The activation of NF-κB, the principal transcriptional regulator of inflammation, has previously been found in several animal models of arthritis (36) and in the synovium of patients with arthritis (37). Consistent with these reports, we observed an increase in the DNA-binding activity of NF-κB in inflamed ankle joints and an increase in the numbers of NF-κB/p50-positive cells in the synovium and cartilage during inflammation. It has previously been demonstrated that inhibition of NF-κB or IκB reduces the severity of arthritis (38). Similarly, in the present study, administration of MG132 down-regulated the expression of the p50 subunit of NF-κB, as well as the DNA-binding activity of both NF-κB and the p50 homodimer, in arthritic rats. MG132 may inhibit either IκB degradation (7) or the UPS-mediated processing of the p105 and p100 precursors to mature p50 and p52 NF-κB subunits (39). Both mechanisms are consistent with our observations.
Our radiographic and histologic analyses of the ankle joints revealed that the subchondral bone resorption, a characteristic feature of adjuvant-induced arthritis, was mitigated by MG132. Bone resorption is a collective result of osteoclast stimulation and suppression of osteoblast precursors within the bone marrow. Previous studies have shown that NF-κB controls osteoclast activation through RANKL signaling (40), while inhibition or deletion of RANKL prevents bone destruction (41, 42). MG132 may prevent bone resorption by interfering with NF-κB–mediated osteoclast activation through the RANKL signaling pathway or by enhancing osteoblast activity. This idea is supported by the findings that another proteasome inhibitor, bortezomib, suppressed the formation of human osteoclasts and promoted the maturation of osteoblasts (43, 44).
The rat model of arthritis used in our study has several clinical and pathologic similarities to human RA with regard to pain, swelling, synovial hyperplasia, inflammation, and destruction of cartilage and bone (45). In the present study, treatment with MG132 during adjuvant-induced arthritis had antinociceptive and antiinflammatory effects that resulted in the reduction of cartilage and bone destruction. These effects may be translated into a potential clinical benefit. The clinical application of proteasome inhibitors might, however, be limited because of the potential neurotoxicity of available compounds when administered long-term. Indeed, the proteasome inhibitor bortezomib induced mild-to-moderate neurotoxic effects in rats (46) and peripheral neuropathy in cancer patients who underwent prolonged treatment with this compound (47).
Several observations, however, provide evidence against these possibilities. First, in rats, the neurotoxic effects were observed following administration of multiple intravenous doses of the maximum tolerated, sublethal doses of bortezomib. Bortezomib is 100–1,000 times more potent than MG132 (48, 49). Therefore, the MG132 dose (1 mg/kg/day given subcutaneously) administered in our experiments may be regarded as much lower than the cumulative doses of intravenous bortezomib (0.15 mg/kg/day, given twice weekly for 4 weeks), which did not produce any pathologic changes (46). Second, bortezomib apparently induces neurotoxic effects through a component in its activity that is blocked by the polyhydroxyl compound Tiron, a radical spin trap. The actions of MG132 lack this component (50). Third, no effects of MG132, which was injected for 7 days at a dose 10 times higher than that used in the present experiments, were seen in our previous experiments of motor performance on the rotarod test, posture, gait, exploratory locomotor activity, and cell death in the spinal cord (20). Fourth, in the present study, MG132 treatment did not produce apparent toxic effects and was well tolerated. This was reflected in the fact that the MG132-treated arthritic rats gained more body weight than did the vehicle-treated arthritic rats.
In conclusion, the findings of the present study demonstrate that the UPS is apparently one of the intracellular pathways that control the development of both joint inflammation and inflammatory pain. Based on our observations, we propose that the proteasome represents a novel therapeutic target for RA. From a clinical perspective, this might be an exciting opportunity, considering that novel, safe UPS inhibitors with limited adverse effects are expected to be available soon.
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. Ms A. S. Ahmed 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. A. S. Ahmed, Li, M. Ahmed, Bakalkin, Stark.
Acquisition of data. A. S. Ahmed, Li, M. Ahmed, Hua, Yakovleva, Bakalkin, Stark.
Analysis and interpretation of data. A. S. Ahmed, Li, M. Ahmed, Yakovleva, Ossipov, Bakalkin, Stark.
- 22Homologous type II collagen–induced arthritis in rats: characterization of the disease and demonstration of clinically distinct forms of arthritis in two strains of rats after immunization with the same collagen preparation. Arthritis Rheum 1990; 33: 693–701., , , .
- 41Single and combined inhibition of tumor necrosis factor, interleukin-1, and RANKL pathways in tumor necrosis factor–induced arthritis: effects on synovial inflammation, bone erosion, and cartilage destruction. Arthritis Rheum 2004; 50: 277–90., , , , , , et al.