Induction of cartilage damage by overexpression of T cell interleukin-17A in experimental arthritis in mice deficient in interleukin-1

Authors


Abstract

Objective

To examine the capacity of T cell interleukin-17A (IL-17A; referred to hereinafter as IL-17) to induce cartilage damage during experimental arthritis in the absence of IL-1.

Methods

Local IL-17 gene transfer was performed in the knee joint of IL-1–deficient mice and wild-type controls during streptococcal cell wall (SCW)–induced arthritis. Knee joints were isolated at various time points for histologic analysis of cartilage proteoglycan (PG) depletion. Expression of messenger RNA for inducible nitric oxide synthase, matrix metalloproteinases (MMPs) 3, 9, and 13, and ADAMTS-4 was determined by quantitative polymerase chain reaction analysis. VDIPEN staining was analyzed to study MMP-mediated cartilage damage. In addition, systemic anti–IL-1α/β antibody treatment was performed in mice immunized with type II collagen and injected locally with an adenoviral vector expressing IL-17 or with control adenovirus. Knee joints were isolated and analyzed for cartilage PG depletion, chondrocyte death, and cartilage surface erosion.

Results

During SCW-induced arthritis, local T cell IL-17 gene transfer turned this acute, macrophage-driven joint inflammation into a severe, chronic arthritis accompanied by aggravated cartilage damage. Of high interest, the IL-1 dependency of cartilage PG depletion was fully abrogated when IL-17 was locally overexpressed in the joint. Moreover, local IL-17 gene transfer increased MMP expression without the need for IL-1, although IL-1 remained essential for part of the cartilage VDIPEN expression. Furthermore, when IL-17 was overexpressed in the knee joints of mice with collagen-induced arthritis, anti–IL-1 treatment did not reduce the degree of chondrocyte death or cartilage surface erosion.

Conclusion

These data show the capacity of IL-17 to replace the catabolic function of IL-1 in cartilage damage during experimental arthritis.

Interleukin-17A (IL-17A), the best-known member of the IL-17 family (IL-17A–F), is a likely contributor to the pathogenesis of arthritis. IL-17A (referred to hereinafter as IL-17) is produced by activated (memory) T cells and is present at high levels in the synovium and synovial fluid of patients with rheumatoid arthritis (RA) (1–3). T cell IL-17 has many IL-1–like activities, although IL-17 and its receptor (IL-17R) do not belong to the IL-1/IL-1R family. Both cytokines stimulate cells by the recruitment of intracellular tumor necrosis factor receptor–associated factor 6 (TRAF6) after ligand binding, which subsequently leads to the activation of NF-κB (4, 5). IL-17, like IL-1, can induce the production of other cytokines and chemokines (6–9), but can also act on chondrocyte metabolism by reducing proteoglycan (PG) synthesis (10–12) and enhancing cartilage degradation (10, 13, 14). In vitro studies have shown that IL-17 is a direct and potent inducer of matrix breakdown independently of IL-1 (11), and IL-17–induced production of matrix metalloproteinase 9 (MMP-9) in human monocytes and macrophages is not inhibited by IL-1 receptor antagonist (IL-1Ra) (15).

In vivo studies in different experimental models of arthritis have implicated an important role of T cell IL-17 in both joint inflammation and joint destruction. Neutralization of IL-17 during collagen-induced arthritis (CIA) showed that IL-17 plays a role both in the early and late stages of CIA (16, 17). Moreover, crossing IL-17–deficient mice with IL-1Ra–deficient mice prevented the spontaneous development of arthritis normally found in mice lacking IL-1Ra (18). However, systemic as well as local overexpression of IL-17 in CIA accelerated the onset of CIA and aggravated the joint pathology (16). High levels of synovial IL-1 were detected after local IL-17 gene transfer in mice with CIA, and, interestingly, treatment with neutralizing anti–IL-1α/β antibodies had no effect on the local IL-17–induced joint inflammation or bone erosion in CIA (16). However, whether IL-17 administered during experimental arthritis can induce cartilage damage in mice independently of IL-1 remains to be elucidated.

Cartilage destruction is a major complication in inflammatory joint diseases such as RA. Different degrees of cartilage damage are distinguishable, varying from reversible cartilage PG depletion to irreversible chondrocyte death and cartilage surface erosion. Both in vitro and in vivo studies have shown that IL-1 is a key mediator in cartilage destruction (18). IL-17 is clearly less potent than IL-1 in inducing cartilage damage (10). Local overexpression of IL-17 in a naive knee joint induces PG depletion, but not irreversible cartilage damage (16). However, in the autoimmune CIA model, IL-17 overexpression also enhances chondrocyte death and cartilage erosion (16).

In the present study, we investigated whether the T cell cytokine IL-17 could aggravate an acute, macrophage-driven joint inflammation, and we explored the role of IL-1 in this process. We show that local overexpression of T cell IL-17 during streptococcal cell wall (SCW)–induced arthritis leads to more-severe and chronic arthritis. Remarkably, IL-17 replaced the catabolic function of IL-1 in cartilage damage during SCW-induced arthritis. Furthermore, in the presence of IL-17, IL-1 dependency of the expression of major MMPs was lost, although IL-1 remained essential for part of the cartilage VDIPEN expression. Neutralization of IL-1 during local overexpression of IL-17 in the knee joint of mice immunized with type II collagen did not reduce the degree of chondrocyte death and cartilage surface erosion. These findings suggest that IL-17 is an important catabolic factor in cartilage damage through its potential to replace the catabolic function of IL-1, either directly or via interplay with other macrophage-derived factors.

MATERIALS AND METHODS

Animals.

Male C57BL/6 and DBA/1J mice were obtained from Janvier-Elevage (Le Genest Saint Isle, France). IL-1α/β–deficient mice (C57BL/6 background) were a kind gift of Dr. Y. Iwakura (University of Tokyo, Center of Experimental Medicine, Tokyo, Japan [19]). All mice were housed in filter-top cages under specific pathogen-free conditions, and a standard diet and water were provided ad libitum. The mice were used in the experiments when they were between the ages of 10 and 12 weeks and were housed in isolators after adenovirus injection. All animal procedures were approved by the institutional ethics committee.

Materials.

Bovine serum albumin was purchased from Sigma-Aldrich (St. Louis, MO). TRIzol reagent, oligo(dT) primers, and reverse transcriptase derived from Moloney murine leukemia virus (MMLV) were obtained from Life Technologies (Breda, The Netherlands). Primers were purchased from Biolegio (Malden, The Netherlands). SYBR Green Master Mix was purchased from Applied Biosystems (Foster City, CA). Freund's complete adjuvant and Mycobacterium tuberculosis (strain H37Ra) were obtained from Difco (Detroit, MI). Bovine type II collagen was prepared as described previously (20). Rabbit anti-murine IL-1α/β polyclonal antibodies were prepared in our laboratory (21).

SCW preparation and induction of SCW arthritis.

Streptococcus pyogenes T12 organisms were cultured overnight in Todd-Hewitt broth. Cell walls were prepared as described previously (22). The supernatant obtained after centrifugation at 10,000g was used throughout the experiments. These preparations contained 11% muramic acid. Unilateral arthritis was induced by intraarticular (IA) injection of 25 μg of SCW (rhamnose content) in 6 μl of pyrogen-free saline into the right knee joints of naive mice. As a control, phosphate buffered saline (PBS) was injected into the left knee joint.

Adenovirus vectors.

AdIL-17 was kindly provided by one of us (JKK) and was constructed as reported previously (23). The replication-deficient empty virus vector AdDel70-3 (AdControl) was used as a control vector throughout the study. For local overexpression of IL-17 during SCW-induced arthritis, 107 plaque-forming units (PFU) of the adenoviral vector in 6 μl of PBS was injected IA into the right knee joint 18 hours prior to the induction of SCW arthritis. As a control, AdDel70-3 was injected into the left knee joint. We have previously shown that this dose of adenoviral vector does not induce any inflammatory response after IA injection into the mouse knee joint (24). Several days thereafter, mice were killed by cervical dislocation, and the knee joints were harvested for further processing.

Assessment of histologic changes.

Total knee joints of mice were isolated at several time points after the injection of adenovirus and SCW fragments. For standard histologic assessment, tissue was fixed for 4 days in 10% formalin, decalcified in 5% formic acid, and subsequently dehydrated and embedded in paraffin. Standard frontal sections measuring 7 μm were mounted on Superfrost slides (Menzel-Gläser, Braunschweig, Germany). Histopathologic changes in the knee joints were scored in the patella and femur/tibia regions on 5 semiserial sections of the joint spaced 70 μm apart. Scoring was performed in a blinded manner by 2 independent observers (MIK and EL).

Hematoxylin and eosin staining was performed to study joint inflammation. The severity of inflammation in the knee joints was scored on a scale of 0–3, where 0 = no cells, 1 = mild cellularity, 2 = moderate cellularity, and 3 = maximal cellularity. To study PG depletion from the cartilage matrix, sections were stained with Safranin O followed by counterstaining with fast green. PG depletion was scored on an arbitrary scale of 0–3, where 0 = normal, 1 = fully stained cartilage, 2 = destained cartilage, and 3 = completely depleted of PGs. The degree of chondrocyte death was scored on a scale of 0–3, where 0 = no empty lacunae and 3 = complete loss of chondrocytes in the cartilage layer. Cartilage surface erosion was scored on a scale of 0–3, where 0 = no cartilage loss and 3 = complete loss of articular cartilage.

RNA isolation and reverse transcriptase reaction.

Mice were killed by cervical dislocation, and the patella and adjacent synovium were harvested immediately. Two synovial tissue samples, 1 from the lateral side and 1 from the medial side, with a diameter of 3 mm were punched out using a biopsy punch (Stiefel, Wachtersbach, Germany). Six patella specimens per experimental group were taken, and 3 lateral and 3 medial samples were pooled to yield 2 samples per group. Samples of synovium were immediately frozen in liquid nitrogen and ground to powder using a Microdismembrator II (Braun, Melsungen, Germany).

For the study of messenger RNA (mRNA) levels in murine articular cartilage, patellae were dissected and immediately decalcified in 10% EDTA for 16 hours at 4°C. Following decalcification, the complete articular cartilage was stripped from the underlying bone and immediately placed in TRIzol reagent. Cartilage obtained from 6 patellae was pooled.

Total RNA was extracted in 1 ml of TRIzol reagent, an improved single-step RNA isolation method based on the method described by Chomczynski and Sacchi (25). Thereafter, RNA was precipitated with isopropanol, washed with 70% ethanol, and redissolved in water. Isolated RNA was treated with DNase before being reverse transcribed into complementary DNA using oligo(dT) primers and MMLV reverse transcriptase.

Quantitative polymerase chain reaction (PCR).

Quantitative real-time PCR was performed using the ABI Prism 7000 Sequence Detection system (Applied Biosystems) for quantification with SYBR Green and melting curve analysis. Primer sequences for the genes were as follows: for GAPDH (reference gene), 5′-GGC-AAA-TTC-AAC-GGC-ACA-3′ (forward) and 5′-GTT-AGT-GGG-GTC-TCG-CTC-TG-3′ (reverse); for inducible nitric oxide synthase (iNOS), 5′-GGG-CAG-CCT-GTG-AGA-CCT-T-3′ (forward) and 5′-CGT-TTC-GGG-ATC-TGA-ATG-TGA-3′ (reverse); for MMP-3, 5′-TGG-AGC-TGA-TGC-ATA-AGCCC-3′ (forward) and 5′-TGA-AGC-CAC-CAA-CAT-CAG-GA-3′ (reverse); for MMP-9, 5′-GGA-ACT-CAC-ACG-ACA-TCT-TCC-A-3′ (forward) and 5′-GAA-ACT-CAC-ACG-CCA-GAA-GAA-TTT-3′ (reverse); for MMP-13, 5′-ACC-TTG-TGT-TTG-CAG-AGC-ACT-AAC-TT-3′ (forward) and 5′-CTT-CAG-GAT-TCC-CGC-AAG-AGT-3′ (reverse); and for ADAMTS-4, 5′-CAC-TGA-CTT-CCT-GGA-CAA-TGG-TTA-T-3′ (forward) and 5′-GGA-AAA-GTC-GCT-GGT-AGA-TGG-A-3′ (reverse).

PCR conditions were as follows: 2 minutes at 50°C and 10 minutes at 95°C, followed by 40 cycles of 15 seconds at 95°C and 1 minute at 60°C, with data collection during the last 30 seconds. For all PCRs, SYBR Green Master Mix was used in the reaction. Primer concentrations were 300 nmoles/liter. All PCRs were performed in a total volume of 25 μl. Relative quantification of the PCR signals was performed by comparing the threshold cycle (Ct) value, in duplicate, of the gene of interest of each sample with the Ct values of the reference gene GAPDH. Quantitative PCR analysis for each sample was performed in duplicate.

VDIPEN neoepitope staining.

Irreversible PG damage by MMP activity in the cartilage was assessed by immunohistochemistry for the VDIPEN neoepitope. Joint sections were deparaffinized, rehydrated, and digested for 1 hour at 37°C in 0.25 units/ml of chondroitinase ABC in 0.1M Tris HCl, pH 8.0 (Sigma, Zwijndrecht, The Netherlands) to remove chondroitin sulfate from the PGs. Sections were treated with 1% H2O2 in methanol for 20 minutes, followed by treatment with 0.1% Triton X-100 in PBS for 5 minutes. After incubation with 1.5% normal goat serum for 20 minutes, sections were incubated overnight at 4°C with affinity-purified rabbit anti-VDIPEN IgG (a kind gift from Dr. I. Singer, Merck Research Laboratories, Rahway, NJ) or with normal rabbit IgG. The next day, the sections were incubated with biotinylated goat anti-rabbit IgG followed by labeling with avidin–peroxidase (Elite kit; Vector, Burlingame, CA). Peroxidase development was performed using nickel enhancement to increase sensitivity. Sections were counterstained with 2% orange G for 5 minutes.

Induction of CIA.

Bovine type II collagen was diluted in 0.05M acetic acid to a concentration of 2 mg/ml and was emulsified in equal volumes of Freund's complete adjuvant (2 mg/ml of M tuberculosis). The mice were immunized intradermally at the base of the tail with 100 μl of emulsion (50 μg of collagen). On day 21, mice were given an intraperitoneal booster injection of 100 μg of type II collagen dissolved in PBS. The first signs of arthritis onset, mainly in the ankles, typically occur around days 25–28.

Study protocol.

CIA was induced in male DBA-1 mice as described above. Just before the expected onset of CIA (day 25), mice were scored visually for the appearance of arthritis. Mice without macroscopic signs of arthritis in the paws were selected. At this time, before the expected onset of CIA (day 25), mice were anesthetized with ether, and a small aperture was cut in the skin of the knee for the IA injection procedure. When the absence of arthritis in the knee joint was confirmed visually, IA injection of 107 PFU per 6 μl of either an IL-17–expressing adenoviral vector (AdIL-17) or a control adenoviral vector (AdControl) was performed. Five days after the IA injection of the virus vector, mice were killed by cervical dislocation, and the knee joints were obtained for histologic analysis.

Statistical analysis.

Differences between experimental groups were tested using the Mann-Whitney U test, unless indicated otherwise. P values less than 0.05 were considered significant. Results are expressed as the mean ± SEM.

RESULTS

IL-1 independency of cartilage damage in SCW arthritis induced by T cell IL-17.

In the SCW-induced arthritis model, a single IA injection of 25 μg SCW fragments into the knee joint of C57BL/6 mice resulted in a clear influx of inflammatory cells and marked cartilage PG depletion (Table 1). However, IL-1–deficient mice showed significantly suppressed cell influx and PG depletion on day 7 (Table 1) compared with the wild-type mice, suggesting that IL-1 plays an important role in both sustained cellular infiltration and cartilage damage during SCW-induced arthritis.

Table 1. IL-1 dependency of inflammation and cartilage proteoglycan depletion in SCW-induced arthritis*
Mouse groupInfiltratePG depletion
Day 4Day 7Day 4Day 7
  • *

    Interleukin-1 (IL-1)–deficient and wild-type mice were injected intraarticularly with 25 μg of streptococcal cell wall (SCW) fragments. Injected knee joints were harvested and histologic sections were scored on a 0–3 scale for synovial infiltrates and cartilage proteoglycan (PG) depletion (see Materials and Methods for details). Values are the mean ± SEM score.

  • P < 0.005 versus wild-type mice, by Mann-Whitney U test.

  • P < 0.05 versus wild-type mice, by Mann-Whitney U test.

Wild-type1.10 ± 0.271.96 ± 0.310.92 ± 0.111.21 ± 0.21
IL-1–deficient0.88 ± 0.290.58 ± 0.170.50 ± 0.330.63 ± 0.42

We next examined whether the presence of the T cell factor IL-17 in this model of acute, macrophage-driven arthritis could aggravate cartilage damage and whether IL-17 affects the IL-1 dependency of this process. Wild-type and IL-1–deficient mice were injected IA with 107 PFU of AdIL-17 or a control virus vector before the induction of arthritis. In wild-type mice, overexpression of IL-17 during SCW-induced arthritis resulted in more-severe arthritis, with significantly increased joint inflammation and cartilage PG depletion on days 4 and 7 compared with the control group (Figures 1 and 2A and C). In contrast, IL-17 overexpression in the naive knee joint resulted in only mild inflammation (mean ± SEM 0.95 ± 0.14) and cartilage PG depletion (0.95 ± 0.08) on day 2. On day 10 of SCW-induced arthritis, inflammatory cell influx and cartilage damage declined quickly in the control group, while remaining high in the IL-17–injected group (Figure 1). Even at 21 days after arthritis induction, inflammation and PG depletion could still be found in IL-17–injected knee joints (data not shown). These results show enhanced chronicity of the macrophage-mediated SCW-induced arthritis model, with an increased severity of cartilage damage by overexpression of T cell IL-17.

Figure 1.

Loss of interleukin-1 (IL-1) dependency of inflammation and cartilage proteoglycan (PG) depletion by overexpression of IL-17 in mice with streptococcal cell wall (SCW)–induced arthritis. Histologic scores for infiltrates (A) and depletion of cartilage proteoglycan (PG) (B) in IL-1–deficient (IL-1–/–) and wild-type (WT) mice with SCW-induced arthritis after intraarticular injection of 107 plaque-forming units of adenoviral vector AdControl (AdC) or AdIL-17 are shown. Histologic features scored on a scale of 0–3 (see Materials and Methods for details). Values are the mean ± SEM. ∗ = P < 0.05 versus wild-type mice, by Mann-Whitney U test.

Figure 2.

Histologic sections of mouse knee joints showing loss of IL-1 dependency of joint inflammation and cartilage PG depletion by overexpression of IL-17 in mice with SCW-induced arthritis. Knee joint sections from wild-type (A and C) and IL-1–deficient (B and D) mice were obtained 7 days after intraarticular injection of 107 plaque-forming units of adenoviral vector AdControl (A and B) or AdIL-17 (C and D) and subsequent induction of SCW arthritis. P = patella; C = cartilage; S = synovium; JS = joint space; F = femur. (Safranin O stained; original magnification × 100.) See Figure 1 for other definitions.

Consistent with the findings shown in Table 1, IL-1 deficiency resulted in significantly reduced joint inflammation and cartilage PG depletion on day 7 (Figures 1 and 2A and B). However, IL-17 overexpression enhanced joint inflammation both in IL-1–deficient mice and their wild-types (Figures 1 and 2A–D). Interestingly, histologic analysis revealed no suppression of cartilage damage in IL-1–deficient mice injected with AdIL-17 (Figures 1 and 2C and D). These data indicate that in the presence of IL-17, cartilage PG depletion in SCW-induced arthritis is no longer IL-1–dependent, suggesting that IL-17 can replace the need for IL-1 in cartilage damage.

Loss of IL-1 dependency of cartilage-degradative genes in the presence of IL-17.

The above studies of IL-17 overexpression during SCW-induced arthritis suggested that IL-17 is able to replace the function of IL-1 in cartilage degradation. We therefore investigated the effect of IL-17 overexpression during SCW-induced arthritis on the IL-1 dependency of the mRNA expression of the cartilage-degradative genes iNOS, MMPs 3, 9, and 13, and ADAMTS-4. Consistent with the importance of IL-1 in cartilage damage, we found that IL-1–deficient mice from the control adenovirus group had reduced levels of mRNA for these cartilage-degradative genes compared with wild-type mice (Figure 3). The levels of mRNA for iNOS, stromelysin (MMP-3), gelatinase B (MMP-9), collagenase-3 (MMP-13), and gelatinase-1 (ADAMTS-4) were clearly reduced on day 4 both in the cartilage layers and in the synovial biopsy tissues (Figure 3). However, when IL-17 was overexpressed during SCW-induced arthritis, the levels of mRNA for iNOS, MMPs 3, 9, and 13, and ADAMTS-4 were similar between IL-1–deficient mice and their wild-types (Figure 3). These data show that the role of IL-1 in the expression of cartilage-regulating genes can be circumvented by IL-17.

Figure 3.

Quantitative polymerase chain reaction analysis of patellar cartilage (A) and synovium (B) obtained from IL-1–deficient (IL-1–/–) and wild-type mice, showing reduced levels of mRNA for cartilage-regulating genes in IL-1–deficient mice on day 4 of SCW-induced arthritis. This IL-1 dependency is counteracted by local overexpression of IL-17. IL-1–deficient and wild-type mice were injected intraarticularly with 107 plaque-forming units of adenoviral vector AdControl (AdC) or AdIL-17 prior to the induction of SCW arthritis. Samples of synovium were obtained at different time points for reverse transcription–polymerase chain reaction analysis. Expression levels were corrected against the expression of GAPDH. iNOS = inducible nitric oxide synthase; MMP = matrix metalloproteinase (see Figure 1 for other definitions).

Partial replacement by IL-17 of the role of IL-1 in MMP-mediated VDIPEN expression.

In addition to the regulation of MMP mRNA levels, we studied the capacity of IL-17 to affect the activity of MMPs in cartilage independently of IL-1. Degradation of aggrecan by MMPs generates specific neoepitopes in cartilage that end with the amino acid sequence VDIPEN, which can be detected by specific antibodies. In the arthritic joints of wild-type mice that had been locally injected with IL-17 adenovirus, VDIPEN expression could clearly be found (Figures 4A and B). IL-1 deficiency resulted in significantly suppressed levels of VDIPEN expression (Figures 4A and C), a reduction of ∼45%. These results indicate that cartilage VDIPEN expression remains partially IL-1–dependent in the presence of IL-17.

Figure 4.

Partial dependence of IL-1 on cartilage VDIPEN expression in the presence of IL-17 overexpression in mice with SCW-induced arthritis. SCW arthritis was induced in IL-1–deficient (IL-1–/–) and wild-type mice after intraarticular injection of 107 plaque-forming units of the adenoviral vector AdIL-17. On day 10 after arthritis induction, knee joints were studied for VDIPEN expression as a measure of matrix metalloproteinase–mediated irreversible cartilage destruction. Immunostaining was measured by image analysis (A), using Leica Qwin software (Leica Microsystems, Rijswijk, The Netherlands). Values are the mean ± SEM. ∗ = P < 0.05; ∗∗ = P < 0.01 versus wild-type mice, by Mann-Whitney U test. Immunostaining of VDIPEN epitopes in the cartilage of IL-1–deficient (B) and wild-type (C) mice with SCW-induced arthritis 7 days after IL-17 overexpression. P = patella; M = meniscus; F = femur. (Original magnification × 200.) See Figure 1 for other definitions.

IL-1–independent cartilage damage in CIA in the presence of IL-17.

We further examined the potential of IL-17 to bypass the need for IL-1 in cartilage damage in the more erosive CIA model. CIA is an autoimmune model of arthritis that is characterized by progressive cartilage destruction, with chondrocyte death and surface erosion. The importance of IL-1 in the pathologic changes of CIA has been demonstrated previously (18, 26). Consistent with the results of the experiments with IL-17 overexpression during SCW-induced arthritis, neutralizing IL-1 did not suppress PG depletion in IL-17–aggravated CIA (Table 2). Moreover, irreversible cartilage damage, which was scored as the degree of chondrocyte death and cartilage surface erosion, was not reduced after anti–IL-1 antibody treatment in the presence of IL-17 (Table 2). These data strongly suggest that when IL-17 is present, neutralization of IL-1 in the synovial infiltrate is not sufficient to prevent severe cartilage destruction.

Table 2. Effects of anti–IL-1 treatment on cartilage damage during collagen-induced arthritis*
TreatmentPG depletionChondrocyte deathCartilage erosion
  • *

    Mice immunized with type II collagen were injected intraarticularly with 107 plaque-forming units of the adenoviral vector AdIL-17. Two hours before virus injection, mice were injected intraperitoneally with anti–interleukin-1 (anti–IL-1) or normal rabbit serum (control). Five days later, knee joints were harvested, and histologic sections were scored on a 0–3 scale for cartilage proteoglycan (PG) depletion, chondrocyte death, and cartilage surface erosion (see Materials and Methods for details). Values are the mean ± SEM score.

Control2.66 ± 0.161.04 ± 0.360.68 ± 0.51
Anti–IL-12.86 ± 0.121.11 ± 0.300.63 ± 0.25

DISCUSSION

In this study, we clearly demonstrated the loss of IL-1 dependency of cartilage degradation in the presence of IL-17 during experimental arthritis. In the macrophage-driven SCW-induced arthritis model, IL-17 induced cartilage PG depletion in the absence of IL-1. Moreover, during IL-1 deficiency, IL-17 increased MMP expression independently of IL-1 and partly circumvented the need for IL-1 in MMP-mediated VDIPEN expression. Furthermore, neutralization of IL-1 during CIA in the presence of IL-17 did not reduce chondrocyte death and cartilage surface erosion. These data show the capacity of IL-17 to replace the catabolic function of IL-1 in cartilage damage during experimental arthritis.

T cell IL-17 is a proinflammatory cytokine. A previous study demonstrated that in a naive knee joint, IL-17 overexpression results in the influx of proinflammatory cells, predominantly polymorphonuclear neutrophils, and significant PG depletion (16). In the present study, we showed that overexpression of T cell IL-17 during SCW-induced arthritis resulted in aggravated joint inflammation and enhanced cartilage PG depletion. Normally, in the acute SCW-induced arthritis model, joint inflammation peaks on day 1, and by day 7, hardly any joint inflammation can be detected anymore. However, local IL-17 gene transfer resulted in the aggravation of SCW-induced arthritis during the first days after arthritis induction, and joint inflammation remained high for more than 3 weeks (data not shown). This suggests that the presence of IL-17 turns an acute joint inflammation into a more severe and chronic synovitis, probably through synergy with IL-1, tumor necrosis factor (TNF), or other mediators, continuing the joint inflammation and cartilage degradation.

In vitro studies have shown that IL-17, acting in synergy with IL-1 or TNF, has powerful effects on cytokine (9, 27, 28) and chemokine (8) induction and cartilage degradation (14, 29). The aggravation of arthritis by IL-17 during the first days of SCW-induced arthritis could be caused by additional, or even synergistic, effects of IL-17 in the presence of IL-1 and TNF, both of which are abundantly produced just after the onset of SCW-induced arthritis (30, 31). In the absence of IL-1, TNF seems to be the first candidate to synergize with IL-17. Remarkably, local overexpression of IL-17 in combination with the blocking of IL-1 and TNF (using TNFα-deficient mice and neutralizing antibodies for IL-1) resulted in joint pathology comparable to that produced by IL-1 deficiency alone (data not shown). However, neutralizing antibodies might not be 100% effective in blocking IL-1, which would give small amounts of IL-1 the opportunity to synergize with IL-17. Studies of mice deficient in both IL-1 and TNF will be necessary to show that IL-17 can aggravate and prolong SCW-induced arthritis independently of both IL-1 and TNF. Our preliminary data also suggest a role of IL-6 in prolonging the effects of IL-17, but more research is necessary to define which cytokines interact with IL-17 in the continuation of SCW-induced inflammation.

While IL-17 has many IL-1–like activities, it is not a member of the IL-1 family of cytokines. IL-17 and IL-1 show many similarities with regard to their actions on cytokine induction and joint pathology. IL-17 has similar effects on chondrocyte metabolism and cartilage degradation, although it seems less potent than IL-1 (10, 12). Both cytokines can recruit TRAF6 and activate NF-κB and activator protein 1 (4, 5), although differences in the effects of these transcription factors have also been described (32). These differences in intracellular signaling by IL-17 and IL-1 might be reflected in their potencies to induce joint pathology (11) and might also explain their different modes of action.

We recently showed an IL-1β–independent role of IL-17 in joint inflammation (16). However, from that study, we could not exclude the role of IL-1 in cartilage damage and the proinflammatory role of IL-1α. In the present study, using IL-1α/β–deficient mice, it became clear that IL-1α is not a critical factor in the IL-1–independent effect of IL-17. Moreover, in the present study, we showed that cartilage PG depletion in SCW-induced arthritis in the presence of IL-17 can be aggravated independently of IL-1 and, more intriguingly, that in the presence of IL-17, this cartilage damage itself is no longer dependent on IL-1. Overexpression of interferon-γ, another T cell cytokine, during SCW-induced arthritis did not affect the role of IL-1 in cartilage damage (data not shown). These data demonstrate that in experimental arthritis, IL-17 can induce cartilage damage in the absence of IL-1, either directly or by interaction with other macrophage-derived factors such as TNF. Van Bezooijen et al (29) have shown that IL-17 synergizes with TNFα to induce cartilage damage in vitro. Although TNF seems to be an obvious candidate for synergy with IL-17 to circumvent the necessity for IL-1 in cartilage damage, a combination blocking of IL-1 and TNF using TNFα-deficient mice and neutralizing antibodies for IL-1α and IL-1β did not stop the capacity of IL-17 to induce cartilage damage in the absence of IL-1 (data not shown).

To elucidate the mechanism by which IL-17 circumvents IL-1 in the production of cartilage damage, we studied the expression of cartilage degradative genes, such as iNOS, MMPs, and ADAMTS. Many studies have shown the importance of IL-1 in cartilage metabolism and in the regulation of these genes (33–35), although IL-17 also influences the expression of these genes. IL-17 increases the spontaneous production of MMP-1 by synoviocytes (36) and induces MMP-9 production in human monocyte/macrophages (15). In IL-17–treated chondrocytes, expression of mRNA for MMPs 1, 3, and 13 was detected, and a synergistic induction of these MMPs was seen when IL-17 was combined with other proinflammatory cytokines (14). Stimulation of normal human articular chondrocytes with IL-17 was shown to induce nitric oxide production and increase iNOS expression (37). In this study, we showed that the expression of mRNA for iNOS, MMPs, and ADAMTS in both cartilage and synovium was reduced in IL-1–deficient mice after induction of SCW arthritis as compared with wild-type mice. However, in the presence of IL-17, IL-1–deficient mice did not show reduced levels of expression of these cartilage-regulating genes.

The effect of IL-17 on the role of IL-1 in MMP activity was studied by immunodetection of MMP-specific neoepitopes in aggrecan (VDIPEN). IL-1 is a strong inducer of VDIPEN expression, and previous studies have shown that the neutralization of IL-1 during experimental arthritis results in an almost complete loss of VDIPEN staining (26, 38). IL-1–induced VDIPEN expression is mediated by MMP-3, as has been shown in vitro using cartilage from MMP-3–deficient mice (34). Although IL-17 circumvented the need for IL-1 in MMP-3 expression, MMP-mediated VDIPEN staining remained partly IL-1–dependent. During IL-17 overexpression in SCW-induced arthritis, IL-1 deficiency reduced VDIPEN staining to only 45% on day 10, indicating that IL-17 partly replaced the role of IL-1 in this marker of irreversible cartilage destruction.

Several factors are believed to explain this incapacity of IL-17 to completely replace the role of IL-1 in VDIPEN expression. First, although IL-1–deficient mice showed comparable levels of mRNA for the most important MMPs as their wild-type controls, differences between IL-17 and IL-1 in the induction of other MMPs and enzymes might also play a role. Second, because of the brief viral expression of IL-17 after local injection (16), we cannot exclude the possibility that the levels of IL-17 remain sufficient on day 10 to completely replace the role of IL-1 in VDIPEN expression. Last, although the IL-1 and IL-17 signaling pathways have much in common, differences in the use of transcription factors or gene regulations (32) might cause differences in the ability to activate MMPs. Further research to elucidate the exact signaling pathways for IL-17 and IL-1 is presently under way.

In addition to our findings in SCW-induced arthritis, showing a loss of IL-1 dependency in cartilage damage by IL-17 overexpression, anti–IL-1 therapy in the more erosive CIA model could not prevent irreversible cartilage destruction in the presence of IL-17. Although from this blocking study, we could not exclude the possibility that all IL-1 produced by the synovial infiltrate was neutralized, excess amounts of anti–IL-1 antibodies were injected that clearly reduced the arthritis scores in the control group, a finding comparable to that of our previous IL-1–blocking studies in CIA (18, 26). However, the anti–IL-1 antibodies will have difficulty penetrating the cartilage matrix and blocking all IL-1 production by chondrocytes. Therefore, the combination of a relatively small amount of IL-1 produced by chondrocytes and the presence of IL-17 seems to be responsible for irreversible cartilage destruction, which is consistent with our observation that some IL-1 in combination with IL-17 was needed for more pronounced MMP-mediated cartilage damage during SCW-induced arthritis.

This study is the first to show that the dominant role of IL-1 in cartilage damage during experimental arthritis can be circumvented by IL-17. The importance of IL-1 in cartilage damage in RA has been shown in many studies (39–41), and IL-1 has recently become a therapeutic target through the introduction of anakinra, a recombinant IL-1Ra. Anti–IL-1 treatment in various murine models of experimental arthritis showed protection against cartilage damage and PG synthesis inhibition (18, 42). Local intraarticular expression of human IL-1 following gene transfer has been shown to result in all major pathologic changes of human RA, with severe joint inflammation and erosions of articular cartilage and periarticular bone (43). Recently, it was shown that deficiency of the IL-1Ra gene causes autoimmunity and chronic inflammatory arthropathy that resembles RA (44). Those studies indicate that under certain circumstances, IL-1 might be an appropriate therapeutic target in the protection of cartilage damage during RA.

Our study, however, suggests that in the presence of T cell IL-17, cartilage damage might become less IL-1–dependent. Our data also indicate that the circumvention of the necessity of IL-1 that is caused by IL-17–producing T cells might result in unresponsiveness to treatment with anti–IL-1. Therefore, screening of individual RA patients for cytokines critical to the pathogenesis of RA may lead to the conclusion that combination therapy that includes IL-17 blocking might result in better protection against cartilage degradation during destructive arthritis than anti–IL-1 therapy alone.

Ancillary