Mechanical injury of cartilage explants causes specific time-dependent changes in chondrocyte gene expression.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 52, No. 8, August 2005, pp 2386–2395 DOI 10.1002/art.21215 © 2005, American College of Rheumatology Mechanical Injury of Cartilage Explants Causes Specific Time-Dependent Changes in Chondrocyte Gene Expression Jennifer H. Lee, Jonathan B. Fitzgerald, Michael A. DiMicco, and Alan J. Grodzinsky Objective. Joint injury in young adults leads to an increased risk of developing osteoarthritis (OA) later in life. This study was undertaken to determine if injurious mechanical compression of cartilage explants results in changes at the level of gene transcription that may lead to subsequent degradation of the cartilage. Methods. Cartilage was explanted from the femoropatellar groove of newborn calves. Levels of messenger RNA encoding matrix molecules, proteases, their natural inhibitors, transcription factors, and cytokines were assessed in free swelling control cultures as compared with cartilage cultures at 1, 2, 4, 6, 12, and 24 hours after application of a single injurious compression. Results. Gene-expression levels measured in noninjured, free swelling cartilage varied over 5 orders of magnitude. Matrix molecules were the most highly expressed of the genes tested, while cytokines, matrix metalloproteinases (MMPs), aggrecanases (ADAMTS5), and transcription factors showed lower expression levels. Matrix molecules showed little change in expression after injurious compression, whereas MMP-3 increased ⬃250-fold, ADAMTS-5 increased ⬃40-fold, and tissue inhibitor of metalloproteinases 1 increased ⬃12fold above the levels in free swelling cultures. Genes typically used as internal controls, GAPDH and ␤-actin, increased expression levels ⬃4-fold after injury, making them unsuitable for use as normalization genes in this study. The expression levels of tumor necrosis factor ␣ and interleukin-1␤, cytokines known to be involved in the progression of OA, did not change in the chondrocytes after injury. Conclusion. Changes in the level of gene expression after mechanical injury are gene specific and time dependent. The quantity of specific proteins may be altered as a result of these changes in gene expression, which may eventually lead to degradation at the tissue level and cause a compromise in cartilage structure and function. Acute traumatic joint injury in young adults leads to an increased risk of developing osteoarthritis (OA) later in life (1–3), despite efforts to intervene in this process by surgically stabilizing injured joints (4). Although the mechanism by which injury leads to tissue degeneration remains to be elucidated, several injuryrelated factors may contribute to the development of OA. These factors include, but are not limited to, instability in the joint due to ligament, tendon, or meniscal tear, and/or initiation of a cellular response in cartilage or other joint tissues at the time of the injury. Previous clinical studies have shown an increase in the protein levels of matrix metalloproteinase 3 (MMP-3) and tissue inhibitor of metalloproteinases 1 (TIMP-1) as well as an increase in proteoglycan and type II collagen fragments in the synovial fluid of patients following a tear in the anterior cruciate ligament (ACL) or meniscus from 1 day to 20 years after the injury (5,6). During the first week after ACL injury, a significant increase in synovial fluid levels of tumor necrosis factor ␣ (TNF␣) and interleukin-1␤ (IL-1␤) can also occur. By 3 weeks after injury, the levels of these cytokines decreased to the levels observed in samples from patients with chronic arthritis (7). Studies of gene expression in normal and OA cartilage have shown up-regulation of MMP-13 in late-stage OA, whereas MMP-3 is downregulated (8). In addition, bone morphogenetic protein 2 is increased in OA cartilage and colocalizes with newly synthesized type II procollagen, which suggests that anabolic remodeling of the tissue is taking place (9). With the use of the lapine ACL transection Supported by the NIH (grant AR-45779). Dr. Lee’s work was supported by a National Science Foundation predoctoral fellowship. Jennifer H. Lee, PhD, Jonathan B. Fitzgerald, BSE, Michael A. DiMicco, PhD, Alan J. Grodzinsky, ScD: Massachusetts Institute of Technology, Cambridge. Address correspondence and reprint requests to Alan J. Grodzinsky, MD, Massachusetts Institute of Technology NE47-377, 77 Massachusetts Avenue, Cambridge, MA 02139. E-mail: [email protected] Submitted for publication December 16, 2004; accepted in revised form May 3, 2005. 2386 GENE EXPRESSION AFTER CARTILAGE INJURY model to study the pathogenesis of OA in vivo, investigators have observed an increase in MMP-3 expression after 9 weeks (10) and location-specific changes in messenger RNA (mRNA) levels of several genes after 3 weeks and 8 weeks (11). Expression of type II collagen, aggrecan, biglycan, MMPs 1, 3, and 13, and TIMP-1 increased during the development of OA in this animal model, whereas decorin and fibromodulin showed decreased expression (11). Because loading variables are difficult to control in vivo, a number of investigators have developed in vitro models to isolate cartilage and study tissue- and cellular-level effects of mechanical injury. Mechanical loads applied in vitro range from single compressions of up to 50% strain (12–17) to large-amplitude cyclic compression at varying frequencies (⬃0.05–0.3 Hz) for up to 2 hours (18–21). Injurious mechanical compression of cartilage in vitro can damage the extracellular matrix (ECM), leading to increased water content (12,16,18,22), decreased stiffness (12,22), increased hydraulic permeability (20), loss of glycosaminoglycan (GAG) to the culture medium (12–15,17,20,22,23), loss of collagen to the medium (20), and temporary denaturation of collagen in the tissue (16,18,20,21). In addition, injurious mechanical compression can lead to cell death by both apoptosis and necrosis (14,16,19,21–23), as well as decreased matrix biosynthesis rates in the remaining viable cells after injury (12). Although many studies have focused on the effects of in vitro injurious compression on cartilage tissue, the resulting modulation of chondrocyte gene transcription has not been fully elucidated. The objective of this study was to quantify the effects of cartilage injury in vitro on 24 genes central to cartilage maintenance, including genes encoding macromolecules of the ECM, proteases that can cleave ECM proteins and their natural inhibitors, transcription factors, and cytokines known to affect cartilage metabolism. Using real-time polymerase chain reaction (PCR), we measured the levels of mRNA of these molecules at 6 time points after acute mechanical injury. We observed distinct changes in the pattern and kinetics of expression that may suggest a role for certain catabolic processes associated with eventual cartilage degradation. MATERIALS AND METHODS Tissue harvest. Articular cartilage explant disks were harvested from the femoropatellar grooves of 1–2-week-old calves using previously developed methods (24). Briefly, 9-mm–diameter cartilage-bone cylinders were drilled perpen- 2387 Figure 1. Loading device used to induce injurious compression of bovine cartilage explants, and example of compression waveforms. A, An incubator-housed loading apparatus was used to apply injurious compression in displacement control to individual cartilage disks. The load and displacement were recorded by transducers during loading. B, Polysulfone chamber used to hold cartilage disks during loading in unconfined compression. C, Representative data acquired during compression to 50% strain at a strain rate of 1.0/second. Peak stress reached a maximum value of 20.7 MPa. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org. dicular to the cartilage surface. These cylinders were then placed in a microtome holder and the most superficial, ⬃200-m layer was removed to obtain a level surface. Up to 3 sequential 1-mm slices were cut from each cylinder, and 4 disks (1-mm thick, 3-mm diameter) were cored from each slice using a dermal punch, yielding a total of 48 disks from each joint. These disks were then equilibrated in culture medium for 2 days (low-glucose Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 10 mM HEPES buffer, 0.1 mM nonessential amino acids, 0.4 mM proline, 20 g/ml ascorbic acid, 100 units/ml penicillin G, 100 g/ml streptomycin, and 0.25 g/ml amphotericin B) in a 37°C, 5% CO2 environment. Injurious compression. After equilibration of the explants, a custom-designed incubator-housed loading apparatus (25) was used (Figure 1A) to injuriously compress 36 cartilage disks from each joint, while the remaining 12 disks served as free swelling controls. Cartilage samples to be injured were placed individually into a polysulfone chamber (Figure 1B), which allows radially unconfined compression of the disk by impermeable platens (12,13,15). The measured thickness of the cartilage disk just prior to loading was recorded, and the zero-strain position was identified by the point of first contact between the loading platen and the cartilage surface. The injury protocol consisted of a single displacement ramp to a final strain of 50% at a velocity of 1 mm/second (strain rate 1.0/second in displacement control), followed by immediate removal of the displacement at the same rate (Figure 1C). Application of these strain and strain rate parameters resulted in an average peak stress of ⬃20 MPa; this loading protocol has previously been shown to produce damage to the ECM, a significant decrease in cell viability, a decrease in cell biosynthesis by the remaining viable cells, and an increase in GAG 2388 loss to the medium in similar bovine cartilage explants (12,13,15,22,23). After injury, the disks were placed into fresh culture medium. Groups of 6 cartilage disks were removed from culture at 1, 2, 4, 6, 12, and 24 hours and then flash-frozen in liquid nitrogen and stored at ⫺80°C. Two groups, each comprising 6 free swelling disks, were frozen at 4 and 24 hours to serve as controls. Explant disks in each group of 6 specimens were purposely matched across depth and location along the joint surface, to prevent bias based on location; as a result, in each experimental condition, the specimens used are representative of the specimens within the joint surface. RNA extraction. RNA was extracted from the 6 pooled cartilage disks by first pulverizing the tissue and then homogenizing in TRIzol reagent (Invitrogen, San Diego, CA) to lyse the cells. Extracts were then transferred to Phase Gel tubes (Eppendorf, Hamburg, Germany) with 10% volume/volume chloroform and spun at 13,000g for 10 minutes. The clear liquid was removed from the phase gel, and RNA was isolated from the sample using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). Genomic DNA was removed by a DNase digestion step (Qiagen) during purification. Absorbance measurements were read at 260 nm and 280 nm to determine the concentration of RNA extracted from the tissue and the purity of the extract. The mean ⫾ SD 260 nm:280 nm ratio of absorbencies was 1.86 ⫾ 0.12. Reverse transcription of equal quantities of RNA (2.5 g) from each sample was performed using the AmpliTaq-Gold Reverse Transcription Kit (Applied Biosystems, Foster City, CA). Real-time PCR. Real-time PCR was performed using the Applied Biosystems 7700 instrument and SYBR Green Master Mix (Applied Biosystems). Primers were designed on the basis of bovine sequences for matrix molecules (type II collagen, aggrecan, link protein, fibronectin, fibromodulin, and type I collagen), proteases (MMP-1, MMP-3, MMP-9, MMP13, ADAMTS-4, and ADAMTS-5), protease inhibitors (TIMP-1 and TIMP-2), cytokines (TNF␣ and IL-1␤), housekeeping genes (␤-actin and GAPDH), transcription factors (c-fos, c-jun, and SOX9), and growth factors (insulin-like growth factor 1 [IGF-1], IGF-2, and transforming growth factor ␤ [TGF␤]), using Primer Express software (Applied Biosystems). Standard curves for amplification using these primers were generated; all primers demonstrated approximately equal efficiency, with standard curve slopes of ⬃1, indicating a doubling in complementary DNA quantity in each cycle. Expression levels in injured samples were normalized to those in free swelling control samples for each gene. Statistical analysis. Changes in gene-expression levels. In each experiment, expression levels measured in injured sample groups were normalized to those in free swelling control groups for each gene; expression data are presented as the mean (⫾SEM) of 3 replicate experiments. Changes in gene-expression levels in the injured samples with respect to the levels in free swelling controls at the 4-hour and 24-hour time points were examined using a nonparametric t-test (26). The t-test was made nonparametric by estimating the P values from permuted data sets (27); the t statistic was calculated from each of the permuted data sets to create a distribution of possible values. Using this method, all changes in expression that were ⱖ5-fold were found to be statistically significant. Changes between 2-fold and 5-fold were also found to be LEE ET AL significant, with 3 exceptions (c-fos at 4 hours and 24 hours, and c-jun at 24 hours); in certain instances, a lower magnitude of change was found to be significant. Gene clustering. To distinguish the main expression trends, a k-means clustering algorithm was applied to the injury time-course data (28–31). Each gene was grouped on the basis of correlations between the time-course expression profile and a set of randomly chosen starting genes. Group profiles were then calculated as the mean of the expression profiles of the genes in each group. The correlation between each gene and each group profile was calculated and the genes were then regrouped in an iterative manner until convergence. To ensure that an optimal clustering solution for the 24 genes was found, the algorithm was run a sufficient number of times to cover every possible selection of starting genes. Each set of randomly chosen starting genes produced a deterministic grouping of the genes, with each gene paired with the group profile showing the highest correlation. The optimal solution was chosen as the grouping that had the highest overall correlation of genes with group profiles, by averaging over all of the genes (for details, see ref. 31). The number of groups was varied from 3 to 6. Ultimately, 5 groups were chosen to best represent the trends. To determine the distinctiveness of the main expression trends, the final group profiles were compared using a comparison-means Student’s t-test. The Euclidean distance between 2 group profiles represents the difference of means, and the average squared distance of the genes within a group to the group profile represents the variance. The number of genes in each group corresponds to the degrees of freedom for that group. RESULTS Gene-expression levels in noninjured control cartilage disks. Real-time PCR was used to determine the expression levels of 24 genes of interest in noninjured control cartilage disks for comparison with mechanically injured disks. Levels of expression of the tested genes varied over 5 orders of magnitude, as seen in Figure 2, with data normalized to the level of the lowest expressed gene, ADAMTS-4 (aggrecanase 1). ECM molecules, as well as SOX9, a transcription factor promoting expression of matrix molecules in cartilage, showed the highest levels of expression. Genes typically used as internal controls (GAPDH and ␤-actin) showed intermediate levels of expression, whereas certain cytokines, MMPs, and transcription factors displayed relatively lower levels of expression. ADAMTS-4 and ADAMTS-5 (aggrecanase 2) showed the lowest levels of expression of the genes tested. Effects of injurious compression on gene expression. Levels of gene expression in noninjured free swelling controls (shown in Figure 2) changed selectively in response to injurious compression. P values comparing expression levels after injury with expression levels in noninjured controls were calculated at the 4-hour and GENE EXPRESSION AFTER CARTILAGE INJURY Figure 2. Expression levels of 24 genes in free swelling control cartilage cultures, ranked by relative abundance. Medium was changed 2 days after harvest and samples were obtained at 4 and 24 hours after medium change for gene-expression quantification. Levels at the 2 time points were averaged to give a single value for each tissue sample. Expression levels were normalized to expression of ADAMTS-4, the least abundant gene measured. Data are reported as the mean and SEM of 3 replicate experiments using tissue from 3 different joints. IGF-2 ⫽ insulin-like growth factor 2; MMP-3 ⫽ matrix metalloproteinase 3; IL-1␤ ⫽ interleukin-1␤; TIMP-2 ⫽ tissue inhibitor of metalloproteinases 2; TGF␤ ⫽ transforming growth factor ␤; TNF␣ ⫽ tumor necrosis factor ␣. 24-hour time points (Table 1). Although expression levels of some genes remained unchanged in response to injury, others exhibited dramatic differences compared with their free swelling controls. After compression, GAPDH and ␤-actin increased in expression ⬃4-fold above the levels in free swelling controls (Figure 3A). Because these housekeeping genes showed variations in expression levels within the 24 hours after loading, they were not used as internal controls to normalize the data acquired on the other genes; instead, all expression levels were normalized by using a fixed quantity of extracted RNA for reverse transcription. By using a fixed quantity of RNA from each sample, decreased cell viability in injuriously compressed cartilage should not affect the observed levels of gene expression; rather, changes in expression should represent the changes occurring within the remaining viable cells in the tissue. Matrix molecules showed no more than a 2-fold change in gene-expression levels during the 24 hours immediately following compression. Levels of type II collagen and aggrecan (Figure 3A), as well as fibromodulin and link protein (results not shown), did not fluctuate during the 24 hours immediately following injury. Fibronectin increased ⬃2-fold in gene expression at the 12- and 24-hour time points (results not shown). 2389 The most dramatically changing gene in this study was MMP-3, which, following injurious compression, increased in expression ⬃250-fold above the levels in free swelling controls (Figure 3B). MMP-3 expression began to increase within 2 hours after injury, peaked by 12 hours, and then declined to an ⬃50-fold increase above free swelling levels by 24 hours. MMP-13, in contrast, showed only an ⬃2-fold increase above the level in free swelling controls during the 24 hours after injury (Figure 3B). Moreover, MMP-1 and MMP-9 increased by ⬃6-fold and ⬃4-fold, respectively, above the levels in their free swelling controls (results not shown). Similar to MMP-3, ADAMTS-5 showed a dramatic increase in gene expression, to ⬃40-fold above the levels in free swelling controls by 12 hours after injury, which, by 24 hours, remained elevated by ⬃10-fold above Table 1. List of group members with distinct temporal geneexpression profiles induced by injury of cartilage explants, as determined by k-means clustering* Time point Group 1 MMP-3 ADAMTS-5 TGF␤ Group 2 c-fos c-jun Group 3 MMP-1 MMP-9 MMP-13 Type I collagen TIMP-1 TIMP-2 Fibronectin SOX9 GAPDH ␤-actin TNF␣ Group 4 IGF-1 IGF-2 ADAMTS-4 Group 5 Type II collagen Aggrecan Fibromodulin Link protein IL-1␤ 4 hours 24 hours 0.01 ⬍0.01 ⬍0.01 0.05 ⬍0.01 0.03 0.53 0.03 0.30 0.26 0.07 0.83 0.03 0.02 0.02 0.69 0.49 ⬍0.01 0.14 0.08 0.91 0.01 0.04 ⬍0.01 0.05 0.03 ⬍0.01 0.04 0.92 ⬍0.01 0.03 0.33 0.16 0.15 0.03 0.08 0.49 0.02 0.46 0.05 0.09 0.12 0.09 0.09 0.02 0.72 0.18 0.25 * Values are P values in comparison with noninjured control cartilage, calculated from t-tests performed at the 4-hour and 24-hour time points after injury. Groups were formed on the basis of gene–group profile correlations. MMP-3 ⫽ matrix metalloproteinase 3; TGF␤ ⫽ transforming growth factor ␤; TIMP-1 ⫽ tissue inhibitor of metalloproteinases 1; TNF␣ ⫽ tumor necrosis factor ␣; IGF-1 ⫽ insulin-like growth factor 1; IL-1␤ ⫽ interleukin-1␤. 2390 Figure 3. Changes in expression level of A, matrix molecules, ␤-actin, and GAPDH genes and B, matrix metalloproteinases and tissue inhibitor of metalloproteinases 1 after injurious compression. Results are the fold change (⫻) from free swelling levels, with a value of 1 (broken line) indicating similar expression after injury to the level measured in free swelling conditions. Six cartilage disks were pooled for each time point for each experiment. All samples were normalized to total RNA at the reverse transcription step. Data are reported as the mean ⫾ SEM of 3 replicate experiments. See Figure 2 for definitions. control levels (Figure 3B). In contrast, ADAMTS-4 increased only ⬃2–3-fold above the free swelling levels and showed little variation within the time period up to 24 hours after injury (results not shown). TIMPs, the endog- LEE ET AL Figure 4. Changes in expression level of A, transcription factors, B, growth factors, and C, cytokines after injurious compression. Results are the fold change (⫻) from free swelling levels, with a value of 1 (broken line) indicating similar expression after injury to the level measured in free swelling conditions. Six cartilage disks were pooled for each time point for each experiment. All samples were normalized to total RNA at the reverse transcription step. Data are reported as the mean ⫾ SEM of 3 replicate experiments. See Figure 2 for definitions. GENE EXPRESSION AFTER CARTILAGE INJURY enous tissue inhibitors of metalloproteinases, were also affected by injurious compression. TIMP-1 increased to ⬃12-fold over free swelling levels by 12 hours and remained elevated by 24 hours after injury (Figure 3B). TIMP-2, which was expressed at an overall higher level in free swelling cartilage (Figure 2), was increased by only ⬃2-fold at 12 hours and 24 hours after injury (results not shown). The immediate-response transcription factors c-fos and c-jun responded to injury with a rapid increase in gene expression (⬃120-fold for c-fos and ⬃40-fold for c-jun) within the first hour after injury (Figure 4A). By 4 hours, both genes returned to an ⬃3-fold increase over free swelling levels and remained moderately elevated for 24 hours. Another transcription factor, SOX9, which promotes transcription of matrix molecules, did not change expression levels significantly during the 24 hours following injurious compression (Figure 4A). This is consistent with the observed lack of change in geneexpression levels of the matrix molecules shown in Figure 3A. Selected growth factors of interest also showed specific changes in gene-expression levels in response to injurious compression. TGF␤ increased expression in the first 4 hours after injury to a peak value ⬃7-fold above the levels in free swelling controls, which remained elevated through 12 hours and then decreased to ⬃4-fold over the free swelling value by 24 hours (Figure 4B). Insulin-like growth factors IGF-1 and IGF-2 (Figure 4B) and, similarly, the cytokines IL-1␤ and TNF␣ (Figure 4C) showed little variation with time (not exceeding 2-fold increased or decreased levels compared with noninjured controls) in the 24 hours immediately following injurious compression. Clustering analyses of gene-expression profiles. Clustering analysis revealed 5 groups with distinct temporal expression profiles induced by injury. The group expression profiles are shown in Figure 5, and the corresponding group members along with their associated level of significance (P values calculated by t-test, performed at 4 hours and 24 hours after injury) in comparison with noninjured control cartilage are listed in Table 1. In general, the group expression profiles are a reflection of the main traits of the individual genes within each group, with mean correlation coefficients of 0.90, 1.00, 0.89, 0.77, and 0.88 for groups 1, 2, 3, 4, and 5, respectively. Comparison of means by Student’s t-test revealed that the group expression profiles were distinct (Figure 5). The unique profile of group 2 was significantly different from the expression profiles of groups 3, 4, and 5 (P ⬍ 0.05), and the expression profiles of groups 2391 Figure 5. Group expression profiles generated by k-means clustering, showing the main temporal gene-expression patterns induced by injury of cartilage explants. Group profiles were calculated by averaging the expression profiles of genes within each group. Results are the mean change (⫻) from free swelling levels, with a value of 1 (broken line) indicating similar expression after injury to the level measured in free swelling conditions. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org. 1 and 3 were also significantly different (P ⫽ 0.006). The expression profiles of groups 1 and 2 were found to be not significantly different from each other, primarily due to the low number of genes within each of these groups. DISCUSSION A single injurious compression of cartilage has been shown previously to decrease ECM biosynthesis rates, compromise mechanical properties, and reduce chondrocyte viability (12–16,22). We undertook this study to determine if changes also occur at the level of gene expression, and to determine whether the changes are general or are specific to certain genes. Analysis of samples was performed using real-time PCR, which allows the measurement of many genes to be achieved in a high-throughput manner using a relatively small sample volume. We observed significant changes in the expression of several catabolic and anabolic genes in response to mechanical injury, and used k-means clustering (31) to further analyze gene-expression patterns and coregulation of specific genes that may result from injury. Previous investigators have used similar cluster- 2392 ing techniques to analyze changes in expression caused by noninjurious static compression (31). Analysis of the behavior of the gene groups by clustering resulted in separation of the genes into 5 groups that displayed distinct patterns of behavior after injury (Table 1 and Figure 5). Group 1 contained MMP-3, ADAMTS-5, and TGF␤, which all displayed large changes in expression levels at early time points (within 4 hours) following injury. In addition to directly cleaving matrix molecules, MMP-3 has been implicated as a member of the activation cascades of matrixdegrading enzymes, including other MMPs. Stimulation of these 3 genes immediately after injury may represent an attempt to remodel the damaged matrix by removing some of the matrix molecules or by activating latent molecules in the matrix. The transcription factors c-fos and c-jun (group 2) showed an immediate transient up-regulation followed by a rapid decline within 4 hours; c-fos and c-jun are members of the activator protein 1 family of genes, which were previously shown to activate MMPs in a chondrocyte cell line after IL-1␤ treatment (32). This is consistent with the activation of several MMPs in groups 1 and 3 of this study, observed at time points subsequent to the increased expression of c-fos and c-jun immediately after injury. Group 3 represents the slowly increasing expression pattern seen for MMPs (other than MMP-3) and their inhibitors, as well as TNF␣, fibronectin, type I collagen, GAPDH, and ␤-actin. Although gene expression of SOX9 remained below free swelling levels for all time points tested (Figure 4A), this gene clustered into group 3 because its expression increased from 0.6-fold to 1.0-fold the level of free swelling controls during the 24 hours after injury, as was the case for group 3 overall. Further investigation is required to determine the extent to which these molecules may affect cartilage behavior after injury in this system, since their changes in expression were relatively low. Group 4 (IGF-1, IGF-2, and ADAMTS-4) and group 5 (type II collagen, aggrecan, fibromodulin, link protein, and IL-1␤) showed expression patterns that did not vary significantly with time after injury. Thus, any immediate alterations in ECM biosynthesis that may result from mechanical injury are unlikely to be related to events at the level of matrix gene (group 5) transcription. Any rapid initial repair of the matrix immediately after acute mechanical injury is not likely to be associated with changes in expression of group 4 genes. The effects of mechanical injury on the expression of MMP-3 and MMP-13 (Figure 3B) are similar to the trends reported by Patwari et al (13) in a study in LEE ET AL which Northern analysis was used to determine the expression levels in similar cartilage explants that were subjected to the same injury protocol as depicted in Figure 1C. In that study, injury caused a significant increase in MMP-3 expression (10-fold) above the level in controls but no change in MMP-13 in the first 24 hours. In comparison, when our data (shown in Figure 3B) are averaged over the full 24-hour period, MMP-3 appears to be up-regulated 80-fold, while MMP-13 expression increases no more than 2-fold above the control levels. Taken together, these studies show similar differential changes in the expression of MMPs 3 and 13, obtained using both Northern analysis and real-time PCR techniques. In vivo studies of OA progression following joint injury have also demonstrated changes in MMP gene-expression levels. Le Graverand et al, using a lapine ACL transection model of OA, found 2–3-fold increased chondrocyte expression of MMP-3 and 10–30fold increased expression of MMP-13 (11), which differs in their relative increases as compared with the changes of MMPs 3 and 13 seen in the present study and in previous studies (13). This difference may be related, in part, to the presence of other tissues in the in vivo model, such as ligaments, tendons, and synovium, that are not included in the present in vitro model of injury to cartilage alone. It is also informative to compare the observed changes in gene expression reported herein to the changes in protein levels reported previously in response to mechanical injury of cartilage in vitro. Immunohistochemical analyses of adult (23-month-old) bovine articular cartilage disks subjected to a rapid ramp displacement to 50% strain and held for 5 minutes revealed an increase in MMP-1, MMP-3, and MMP-13 as well as a decrease in TIMP-1 and TIMP-2 (33). A study applying compressive loading to immature bovine tissue measured increased synthesis and activity of MMP-2 and MMP-9 after 1–16 hours of loading, while no change was measured in TIMP-1 or TIMP-2 synthesis (34). Porcine cartilage disks from 3–6-month-old animals subjected to a cutting injury showed an increase in synthesis of MMP-1, MMP-3, and TIMP-1, while collagen synthesis remained unchanged (35). Interestingly, our study revealed changes in the expression of MMPs and TIMPs, but no change in type II collagen expression. In addition, compression injury was found previously to increase fibronectin protein synthesis (18); the compression injury used in the present study caused an increase in fibronectin gene expression (Table 1). Specific differences found in these studies may be due to differences in regulation at the level of translation, as well as to the GENE EXPRESSION AFTER CARTILAGE INJURY different injury models used (e.g., cutting versus compression). Findings from in vivo studies at the protein level also have certain parallels to the results reported herein. Lohmander et al analyzed human synovial fluid after ACL or meniscus injury and found increases in MMP-3 and TIMP-1 protein levels within 1 day after injury, which persisted for 20 years (5). Similarly, increased chondrocyte mRNA levels of MMP-3 and TIMP-1 were found in our study after cartilage injury and in the lapine ACL transection model (11). Irie et al measured elevated levels of the inflammatory cytokines IL-1␤ and TNF␣ in human joints within 24 hours after ACL injury (7). In the current study, cartilage injury did not cause an increase in chondrocyte expression of IL-1␤ and TNF␣. Although the major source of the increased levels of cytokines, MMPs, and TIMPs seen in the synovial fluid of injured human joints could be the synovium or tissues other than cartilage, it is informative to be able to identify specific changes in chondrocyte gene and protein expression for comparison. Changes in expression of proteases and cytokines have been found during the progression of OA. Bau et al (8) compared chondrocytes isolated from patients with normal articular cartilage with chondrocytes from patients with early and late-stage OA. MMP-13 and ADAMTS-4 expression increased in late-stage OA, whereas MMP-3 expression was the highest of the gene levels tested and was down-regulated in OA (8). Murata et al (36) measured IL-1␣ and IL-1␤ gene expression (by reverse transcription–PCR) and protein levels (by enzyme-linked immunosorbent assay) in OA chondrocytes isolated from cartilage obtained during joint arthroplasty. They reported a decrease in IL-1␣ and IL-1␤ transcript in cells from advanced OA tissue compared with cells from tissue displaying only moderate degeneration. This decrease in expression was accompanied by a decrease in protein level in advanced OA (36). In contrast, we found that the expression of IL-1␤ was not significantly altered by acute compression injury in vitro (Table 1). Other distinct and important differences in gene expression exhibited by OA tissues versus that in explants subjected to acute mechanical injury have been observed. For example, types I and II collagen exhibit increased expression levels with the progression of OA (37), whereas in our study no significant change in the expression of type II collagen was observed following compression injury and type I collagen significantly increased expression 2.5-fold by 24 hours after injury. It should be emphasized that the focus of the present study 2393 is on immediate changes after injury (within the first 24 hours), while OA develops over a time span of many years and involves pathologic processes of the whole joint (38). It will be important to expand such in vitro studies to include longer culture periods after injury. In addition, in vitro models of whole-joint injury that involve injured cartilage in the presence of exogenous cytokines (13) or injured cartilage cocultured with other injured joint tissues (39) may give additional insight into the cellular pathways underlying chondrocyte response to injury. Investigators have studied the effects of static compression on cartilage explants as well as chondrocyte-seeded gels to determine if changes occur at the levels of gene expression and protein synthesis in these model systems. Cartilage explants were compressed very slowly to 25% and 50% strain and maintained in compression for up to 24 hours. This low strain rate protocol does not alter cell viability (40), in contrast to the marked increase in cell death and matrix damage typically observed after injurious loading. These samples were compared with unloaded controls. Results by realtime PCR showed a transient increase in mRNA levels for aggrecan and type II collagen, as well as other matrix proteins, followed by a down-regulation below control levels by 24 hours (31). The down-regulation of aggrecan and type II collagen expression was shown by Northern analysis to be dose dependent, with 50% compression causing a greater decrease than that caused by 25% compression (41). Radiolabel incorporation into proteoglycans and collagen also decreased with increasing static compression (41). Transcription of many matrix proteases, including MMP-3, increased with loading duration, with elevations ranging from 3-fold to 16-fold by 24 hours of loading (31). Transcription factors c-fos and c-jun were transiently up-regulated by 6- to 35-fold after 1 hour of loading (31). Results similar to those seen for matrix molecule expression in cartilage explants were obtained in a cell-seeded construct. Primary chondrocytes were seeded in type I collagen gels and subjected to static compression of 0%, 25%, or 50% for up to 24 hours (42). Results using competitive and real-time PCR showed inhibition of type I collagen, type II collagen, and aggrecan mRNA expression. Radiolabel incorporation of proline and sulfate were also inhibited by the application of static compression (42). In both of these experiments, changes in mRNA expression levels correlated with changes occurring at the protein level. Notably, changes seen in response to static compression were markedly different from those observed in injurious 2394 loading scenarios. In the current study, matrix molecules did not change in expression level, in contrast to that seen in response to static compression; also, the magnitude of increased expression of degradative enzymes and transcription factors was higher in response to injurious compression compared with noninjurious static compression. Thus, the response of chondrocytes to mechanical compression appears to depend on the specific parameters (rate, amplitude, and duration) of the applied compression. An unexpected result in the current study was the relatively high level of type I collagen expression in free swelling control tissue (Figure 2). Type I collagen molecules can be found in diseased or damaged articular cartilage; however, it is not abundant in healthy cartilage. Although type I collagen expression was, indeed, ⬃60-fold lower than that of type II collagen, mRNA levels were higher than most of the other non-ECM genes studied. Relatively high levels of type I collagen expression were previously reported in 6-month-old porcine cartilage and found to be ⬃3-fold lower than the expression levels of type II collagen (43). The tissue used in the current study was obtained from newborn bovines, and type I collagen expression may vary widely with age and species. In addition, cells from the sparse blood vessels present in newborn cartilage tissue could contribute to the expression of type I collagen, as seen in the control data in Figure 2. One limitation of our study is that samples from different locations within the femoropatellar groove were pooled; thus, it was not possible to determine whether tissue from different depths and locations along the groove would react differently to injurious compression. Explant disks in each group of 6 specimens were purposely matched across depth and location along the joint surface to prevent bias based on location; therefore, the gene-expression results represent an average of specimens within the joint surface. Another limitation is the use of newborn tissue. We and other investigators have previously studied the effects of injury on cell viability and ECM degradation in the presence and absence of exogenous cytokines, using both immature and adult cartilage from bovine and human joint surfaces, with the finding that certain responses vary with age (12,13,44). It will be important to extend the present study to identify any age- or disease-dependence of changes in gene expression caused by mechanical injury. In summary, injurious compression caused timedependent changes within 24 hours in the expression of specific catabolic and anabolic genes that can regulate matrix remodeling and turnover, whereas many ECM LEE ET AL molecules were unaffected. Ongoing studies are focused on determining whether these changes at the level of gene expression result in changes in protein levels in the cartilage, and whether the high up-regulation of ADAMTS-5 and MMP-3 in response to injurious mechanical compression may be associated with cellmediated changes in the proteolytic cleavage of ECM molecules over extended times after injury. ACKNOWLEDGMENT We thank Dr. Moonsoo Jin for design of certain primers used in this study. REFERENCES 1. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Joint injury in young adults and risk for subsequent knee and hip osteoarthritis. Ann Intern Med 2000;133:321–8. 2. Davis MA, Ettinger WH, Neuhaus JM, Cho SA, Hauck WW. The association of knee injury and obesity with unilateral and bilateral osteoarthritis of the knee. Am J Epidemiol 1989;130:278–88. 3. Von Porat A, Roos EM, Roos H. High prevalence of osteoarthritis 14 years after an anterior cruciate ligament tear in male soccer players: a study of radiographic and patient relevant outcomes. Ann Rheum Dis 2004;63:269–73. 4. Roos H, Adalberth T, Dahlberg L, Lohmander LS. Osteoarthritis of the knee after injury to the anterior cruciate ligament or meniscus: the influence of time and age. Osteoarthritis Cartilage 1995;3:261–7. 5. Lohmander LS, Roos H, Dahlberg L, Hoerrner LA, Lark MW. Temporal patterns of stromelysin-1, tissue inhibitor, and proteoglycan fragments in human knee joint fluid after injury to the cruciate ligament or meniscus. J Orthop Res 1994;12:21–8. 6. Lohmander LS, Atley LM, Pietka TA, Eyre DR. The release of crosslinked peptides from type II collagen into human synovial fluid is increased soon after joint injury and in osteoarthritis. Arthritis Rheum 2003;48:3130–9. 7. Irie K, Uchiyama E, Iwaso H. Intraarticular inflammatory cytokines in acute anterior cruciate ligament injured knee. Knee 2003;10:93–6. 8. Bau B, Gebhard PM, Haag J, Knorr T, Bartnik E, Aigner T. Relative messenger RNA expression profiling of collagenases and aggrecanases in human articular chondrocytes in vivo and in vitro. Arthritis Rheum 2002;46:2648–57. 9. Fukui N, Zhu Y, Maloney WJ, Clohisy J, Sandell LJ. Stimulation of BMP-2 expression by pro-inflammatory cytokines IL-1 and TNF-␣ in normal and osteoarthritic chondrocytes. J Bone Joint Surg Am 2003;85 Suppl 3:59–66. 10. Takahashi K, Goomer RS, Harwood F, Kubo T, Hirasawa Y, Amiel D. The effects of hyaluronan on matrix metalloproteinase-3 (MMP-3), interleukin-1␤(IL-1␤), and tissue inhibitor of metalloproteinase-1 (TIMP-1) gene expression during the development of osteoarthritis. Osteoarthritis Cartilage 1999;7:182–90. 11. Le Graverand MP, Eggerer J, Vignon E, Otterness IG, Barclay L, Hart DA. Assessment of specific mRNA levels in cartilage regions in a lapine model of osteoarthritis. J Orthop Res 2002;20:535–44. 12. Kurz B, Jin M, Patwari P, Cheng DM, Lark MW, Grodzinsky AJ. Biosynthetic response and mechanical properties of articular cartilage after injurious compression. J Orthop Res 2001;19: 1140–6. 13. Patwari P, Cook MN, DiMicco MA, Blake SM, James IE, Kumar S, et al. Proteoglycan degradation after injurious compression of GENE EXPRESSION AFTER CARTILAGE INJURY 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. bovine and human articular cartilage in vitro: interaction with exogenous cytokines. Arthritis Rheum 2003;48:1292–301. D’Lima DD, Hashimoto S, Chen PC, Colwell CW Jr, Lotz MK. Human chondrocyte apoptosis in response to mechanical injury. Osteoarthritis Cartilage 2001;9:712–9. DiMicco MA, Patwari P, Siparsky PN, Kumar S, Pratta MA, Lark MW, et al. Mechanisms and kinetics of glycosaminoglycan release following in vitro cartilage injury. Arthritis Rheum 2004;50:840–8. Torzilli PA, Grigiene R, Borrelli J Jr, Helfet DL. Effect of impact load on articular cartilage: cell metabolism and viability, and matrix water content. J Biomech Eng 1999;121:433–41. Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC. The extent of matrix damage and chondrocyte death in mechanically traumatized articular cartilage explants depends on rate of loading. J Orthop Res 2001;19:779–84. Chen CT, Burton-Wurster N, Lust G, Bank RA, Tekoppele JM. Compositional and metabolic changes in damaged cartilage are peak-stress, stress-rate, and loading-duration dependent. J Orthop Res 1999;17:870–9. Chen CT, Burton-Wurster N, Borden C, Hueffer K, Bloom SE, Lust G. Chondrocyte necrosis and apoptosis in impact damaged articular cartilage. J Orthop Res 2001;19:703–11. Thibault M, Poole AR, Buschmann MD. Cyclic compression of cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of collagen and release of matrix fragments. J Orthop Res 2002;20:1265–73. Chen CT, Bhargava M, Lin PM, Torzilli PA. Time, stress, and location dependent chondrocyte death and collagen damage in cyclically loaded articular cartilage. J Orthop Res 2003;21:888–98. Loening AM, James IE, Levenston ME, Badger AM, Frank EH, Kurz B, et al. Injurious mechanical compression of bovine articular cartilage induces chondrocyte apoptosis. Arch Biochem Biophys 2000;381:205–12. Quinn TM, Grodzinsky AJ, Hunziker EB, Sandy JD. Effects of injurious compression on matrix turnover around individual cells in calf articular cartilage explants. J Orthop Res 1998;16:490–9. Sah RL, Kim YJ, Doong JY, Grodzinsky AJ, Plaas AH, Sandy JD. Biosynthetic response of cartilage explants to dynamic compression. J Orthop Res 1989;7:619–36. Frank EH, Jin M, Loening AM, Levenston ME, Grodzinsky AJ. A versatile shear and compression apparatus for mechanical stimulation of tissue culture explants. J Biomech 2000;33:1523–7. Troyanskaya OG, Garber ME, Brown PO, Botstein D, Altman RB. Nonparametric methods for identifying differentially expressed genes in microarray data. Bioinformatics 2002;18:1454–61. Good P. Permutation tests: a practical guide to resampling methods for testing hypotheses. New York: Springer; 2000. Jain A, Murty MN, Flynn PJ. Data clustering: a review. ACM Computing Surveys 1999;31:264–323. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A 1998;95:14863–8. Dougherty ER, Barrera J, Brun M, Kim S, Cesar RM, Chen Y, et al. Inference from clustering with application to gene-expression 2395 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. microarrays [review][published erratum appears in J Comput Biol 2002;9:594–5]. J Comput Biol 2002;9:105–26. Fitzgerald JB, Jin M, Dean D, Wood DJ, Zheng MH, Grodzinsky AJ. Mechanical compression of cartilage explants induces multiple time-dependent gene expression patterns and involves intracellular calcium and cyclic AMP. J Biol Chem 2004;279:19502–11. Vincenti MP, Brinckerhoff CE. Transcriptional regulation of collagenase (MMP-1, MMP-13) genes in arthritis: integration of complex signaling pathways for the recruitment of gene-specific transcription factors. Arthritis Res 2002;4:157–64. Pufe T, Lemke A, Kurz B, Petersen W, Tillmann B, Grodzinsky AJ, et al. Mechanical overload induces VEGF in cartilage discs via hypoxia-inducible factor. Am J Pathol 2004;164:185–92. Blain EJ, Gilbert SJ, Wardale RJ, Capper SJ, Mason DJ, Duance VC. Up-regulation of matrix metalloproteinase expression and activation following cyclical compressive loading of articular cartilage in vitro. Arch Biochem Biophys 2001;396:49–55. Vincent T, Hermansson M, Bolton M, Wait R, Saklatvala J. Basic FGF mediates an immediate response of articular cartilage to mechanical injury. Proc Natl Acad Sci U S A 2002;99:8259–64. Murata M, Trahan C, Hirahashi J, Mankin HJ, Towle CA. Intracellular interleukin-1 receptor antagonist in osteoarthritis chondrocytes. Clin Orthop Relat Res 2003;409:285–95. Gebhard PM, Gehrsitz A, Bau B, Soder S, Eger W, Aigner T. Quantification of expression levels of cellular differentiation markers does not support a general shift in the cellular phenotype of osteoarthritic chondrocytes. J Orthop Res 2003;21:96–101. Buckwalter JA, Mankin HJ. Articular cartilage: degeneration and osteoarthritis, repair, regeneration, and transplantation. Instr Course Lect 1998;47:487–504. Patwari P, Fay J, Cook MN, Badger AM, Kerin AJ, Lark MW, et al. In vitro models for investigation of the effects of acute mechanical injury on cartilage [review]. Clin Orthop Relat Res 2001; 391 Suppl:S61–71. Kim YJ, Grodzinsky AJ, Plaas AH. Compression of cartilage results in differential effects on biosynthetic pathways for aggrecan, link protein, and hyaluronan. Arch Biochem Biophys 1996; 328:331–40. Ragan PM, Badger AM, Cook M, Chin VI, Gowen M, Grodzinsky AJ, et al. Down-regulation of chondrocyte aggrecan and type-II collagen gene expression correlates with increases in static compression magnitude and duration. J Orthop Res 1999;17:836–42. Hunter CJ, Imler SM, Malaviya P, Nerem RM, Levenston ME. Mechanical compression alters gene expression and extracellular matrix synthesis by chondrocytes cultured in collagen I gels. Biomaterials 2002;23:1249–59. Fehrenbacher A, Steck E, Rickert M, Roth W, Richter W. Rapid regulation of collagen but not metalloproteinase 1, 3, 13, 14 and tissue inhibitor of metalloproteinase 1, 2, 3 expression in response to mechanical loading of cartilage explants in vitro. Arch Biochem Biophys 2003;410:39–47. Kurz B, Lemke A, Kehn M, Domm C, Patwari P, Frank EH, et al. Influence of tissue maturation and antioxidants on the apoptotic response of articular cartilage after injurious compression. Arthritis Rheum 2004;50:123–30.