The Modulation of Matrix Metalloproteinase and ADAM Gene Expression in Human Chondrocytes by Interleukin-1 and Oncostatin M

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 46, No. 4, April 2002, pp DOI /art , American College of Rheumatology The Modulation of Matrix Metalloproteinase and ADAM Gene Expression in Human Chondrocytes by Interleukin-1 and Oncostatin M A Time-Course Study Using Real-Time Quantitative Reverse Transcription Polymerase Chain Reaction P. J. T. Koshy, 1 C. J. Lundy, 2 A. D. Rowan, 1 S. Porter, 2 D. R. Edwards, 2 A. Hogan, 2 I. M. Clark, 2 and T. E. Cawston 1 Objective. Previous studies have reported elevated levels of interleukin-1 (IL-1) and oncostatin M (OSM) in rheumatoid joints, as well as the synergistic degradation of human articular cartilage by this cytokine combination. The present study was undertaken to investigate the ability of IL-1 and OSM to modulate gene expression of matrix metalloproteinase (MMP), ADAM, and ADAM-TS (ADAM with thrombospondin motifs) family members in human chondrocytes. Methods. T/C28a4 human chondrocytes were stimulated for 2 48 hours with IL-1 and/or OSM. Total RNA was harvested, reverse transcribed, and assessed by real-time polymerase chain reaction for the expression of various MMP, ADAM, and ADAM-TS messenger RNAs (mrna). Results were normalized to 18S ribosomal RNA. Results. IL-1 and OSM synergized to markedly induce the expression of the collagenases MMP-1, MMP-8, and MMP-13 as well as MMP-3, an activator of prommps. Expression of mrna for MMPs 1, 3, and 13 was induced early, whereas that of MMP-8 mrna Supported by the Arthritis Research Campaign and the Medical Research Council of Great Britain. 1 P. J. T. Koshy, BSc, PhD, A. D. Rowan, BSc, PhD, T. E. Cawston, BSc, PhD: University of Newcastle, Newcastle upon Tyne, UK; 2 C. J. Lundy, BSc, PhD, S. Porter, BA, MSc, DipACC, D. R. Edwards, BA, PhD, A. Hogan, BSc, PhD, I. M. Clark, BSc, PhD: University of East Anglia, Norwich, UK. Address correspondence and reprint requests to A. D. Rowan, BSc, PhD, Department of Rheumatology, Cookson Building, The Medical School, University of Newcastle, Framlington Place, Newcastle upon Tyne NE2 4HH, UK. A.D.Rowan@ncl.ac.uk. Submitted for publication June 11, 2001; accepted in revised form December 5, occurred late. Gene expression of MMP-14, an MMP that degrades collagen and activates prommp-13, was elevated by this combination. IL-1 and OSM also synergized to induce gene expression of the aggrecanase ADAM-TS4, but not ADAM-TS5. Conclusion. These data indicate that the potent cartilage-degrading properties of the combination of IL-1 and OSM are potentially mediated by a synergistic induction of the aggrecan-degrading enzyme ADAM- TS4 and the collagen-degrading enzymes MMP-1, MMP-8, MMP-13, and MMP-14, although differences in the magnitude of response and in the time course of induction were observed. A role for MMPs 3 and 14 in the activation of prommps may also be implicated. Articular cartilage is a highly specialized tissue that covers the surface of synovial joints, allowing its smooth articulation. Cartilage is synthesized and maintained by chondrocytes, and is composed primarily of water, proteoglycan (principally aggrecan), and collagen (1). Type II collagen fibrils are arranged to form a fibrillar network within which are trapped aggrecan molecules that pull water into the tissue. These two components provide cartilage with its tensile strength and the ability to resist compression. The degradation of cartilage is one of the key features of the arthritides and results in disability. The degradation of cartilage in vitro and in vivo is believed to be mediated by neutral endopeptidases of the metalloproteinase class of enzymes (2). Two families in this class that have been implicated are the matrix metalloproteinase (MMP) and ADAM families (2,3). 961

2 962 KOSHY ET AL Recently, two members of an ADAM-related family known as ADAMs with thrombospondin motifs (ADAM-TS) (4), ADAM-TS4 and ADAM-TS5 (initially named ADAM-TS11), were the first identified enzymes (5,6) possessing the ability to cleave aggrecan at what was shown (7) to be the physiologically and pathologically relevant site. Collagen-degrading enzymes have been identified within the MMP family. These include the collagenases (MMP-1, MMP-8, and MMP-13), gelatinase A (MMP-2), and MMP-14 (a membrane-type MMP) (8). MMPs are synthesized as proenzymes that require proteolytic processing into the active form (2). Important in this critical step may be activator enzymes such as MMP-3 and MMP-14, which are able to activate one or more of the procollagenases (8). We previously reported that the combination of interleukin-1 (IL-1) and oncostatin M (OSM) promotes synergistic cartilage degradation, stimulating the reproducible loss of proteoglycan and collagen from bovine, porcine, and human cartilage ex vivo (9). In addition, both IL-1 levels and OSM levels are elevated in rheumatoid arthritis (RA) synovial fluid (9,10). In this study, real-time quantitative reverse transcription polymerase chain reaction (RT-PCR) was used to investigate the ability of IL-1 in combination with OSM to modulate the expression of MMP, ADAM, and ADAM-TS family members in the T/C28a4 human chondrocyte cell line, in order to identify mediators of IL-1/OSM induced cartilage degradation. MATERIALS AND METHODS Recombinant human IL-1 and OSM were used as previously described (9). Cytokines were stored at 70 C, and were diluted into culture medium immediately prior to use. Tissue culture reagents and all other chemicals have been described previously (9). Cell culture and RNA extraction. Human immortalized chondrocytes (T/C28a4) (11) were grown to 70% confluence and serum-starved overnight as previously described (9). Cells were stimulated with medium containing test reagents for 2 48 hours. Following stimulation, cells were lysed in RNeasy lysis buffer (Qiagen, Crawley, UK) and total cellular RNA was prepared according to the manufacturer s instructions. Isolated RNA was stored at 80 C until required for further experiments. Reverse transcription. Total RNA (1 g) was reverse transcribed in a 20- l reaction using 2 g of random hexamers and Superscript II reverse transcriptase (Life Technologies, Paisley, UK) according to the manufacturer s instructions. Design of primers and probes for TaqMan PCR. Oligonucleotide primers and fluorescent-labeled TaqMan probes were designed using Primer Express 1.0 software (Applied Biosystems, Warrington, UK). Sequences for the MMP primers and probes are under the copyright of Applied Biosystems. Sequences for the ADAM and ADAM-TS primers and probes are shown below: ADAM-10 forward primer 5 -AGCGGCCCCGAGAGAGT-3, reverse primer 5 - AGGAAGAACCAAGGCAAAAGC-3, probe 5 -FAM (6-carboxy-fluorescein) ATCAAATGGGACACATGAGAC- GCTAACTGC TAMRA (6-carboxy-N,N,N N -tetramethylrhodamine) 3 ; ADAM-17 forward 5 -GAAGTGCCA- GGAGGCGATTA-3, reverse 5 -CGGGCACTCACTGCT- ATTACC-3, probe 5 -FAM-TGCTACTTGCAAAGGCGT- GTCCTACTGC-TAMRA-3 ; ADAM-TS1 forward 5 - GGACAGGTGCAAGCTCATCTG-3, reverse 5 -TCTAC- AACCTTGGGCTGCAAA-3, probe 5 -FAM-CAAGC- CAAAGGCATTGGCTACTTCTTCG-TAMRA-3 ; ADAM- TS4 forward 5 -CAAGGTCCCATGTGCAACGT-3, reverse 5 -CATCTGCCACCACCAGTGTCT-3, probe 5 -FAM- CCGAAGAGCCAAGCGCTTTGCTTC-TAMRA-3 ; AD- AM-TS5 forward 5 -TGTCCTGCCAGCGGATGT-3, reverse 5 -ACGGAATTACTGTACGGCCTACA-3 ; probe 5 -FAM- TTCTCCAAAGGTGACCGATGGCACTG-TAMRA-3. In order to control against amplification of genomic DNA and to ensure that the PCR signal was generated from complementary DNA (cdna), primers were placed within different exons, close to intron exon boundaries. BLASTN searches were conducted on all primer/probe nucleotide sequences to ensure gene specificity. The identity of PCR products was confirmed by direct sequencing of amplicons. The 18S ribosomal RNA gene was used as an endogenous control to normalize for differences in the amount of total RNA in each sample. TaqMan 18S ribosomal RNA primers and a 5 -VIC labeled probe were used according to the manufacturer s instructions (Applied Biosystems). TaqMan PCR. Relative quantitation of gene expression was performed using the Applied Biosystems ABI Prism 7700 sequence detection system (TaqMan). PCR reactions for all samples were performed in duplicate in 96-well optical plates using 5 ng of cdna, 12.5 l2 TaqMan Universal PCR mastermix (Applied Biosystems), 100 nm probe, 200 nm of each primer, and water to a 25 l final volume. Thermocycler conditions comprised an initial holding at 50 C for 2 minutes, then 95 C for 10 minutes. This was followed by a 2-step TaqMan PCR program consisting of 95 C for 15 seconds, and 60 C for 60 seconds for 40 cycles. Statistical analysis. During PCR, the TaqMan probe emits a fluorescence signal that increases in intensity in direct proportion to the amount of specific amplified product. The ABI Prism 7700 instrument measures the cycle-to-cycle changes in fluorescence in each sample. Data are initially expressed as a threshold cycle (CT), defined as the point at which an increase in fluorescence above a baseline signal can first be detected. The fewer cycles it takes to reach the CT, the greater the initial template copy number. The CT values generated were used to calculate relative input amounts of template cdna, using the standard curve method as described by the manufacturer (1997 User Bulletin no. 2; Applied Biosystems). The input cdna levels measured corresponded directly to the levels of RNA reverse transcribed. Thus, data are presented graphically as relative levels of messenger RNA (mrna) for each primer/probe set. Direct comparisons between levels cannot be made between different primer/probe sets.

3 MODULATION OF MMP, ADAM, AND ADAM-TS IN CHONDROCYTES BY IL-1 AND OSM 963 Figure 1. The effect of interleukin-1 (IL-1) and oncostatin M (OSM) on the gene expression of matrix metalloproteinase 1 (MMP-1), MMP-8, and MMP-13 in human chondrocytes. T/C28a4 chondrocytes were treated for 2 48 hours with medium alone (control) (diagonally hatched bars), IL-1 (1 ng/ml) (open bars), OSM (10 ng/ml) (horizontally hatched bars), and IL-1 (1 ng/ml) in combination with OSM (10 ng/ml) (solid bars). Following treatment, total RNA was isolated from cells and reverse transcribed, and the resulting cdna was used in separate real-time quantitative polymerase chain reaction assays with specific primers and probes for MMP-1 (A), MMP-8 (B), and MMP-13 (C). Results were normalized to 18S ribosomal RNA and presented graphically as relative mrna levels (see Materials and Methods). RESULTS Effect of IL-1 in combination with OSM on the expression of MMP genes. The expression of the collagenases MMP-1, MMP-8, and MMP-13 was highly induced in T/C28a4 human chondrocytes following stimulation with IL-1 in combination with OSM (Figure 1). This cytokine combination also highly induced MMP-3 mrna expression (Figure 2A). Maximal induction of mrna for MMPs 1, 3, and 13 was observed early (8 12 hours), whereas MMP-8 mrna was induced late (max-

4 964 KOSHY ET AL Figure 2. The effect of IL-1 and OSM on the gene expression of MMP-3, MMP-9, and MMP-14 in human chondrocytes. T/C28a4 chondrocytes were treated for 2 48 hours with medium alone (control) (diagonally hatched bars), IL-1 (1 ng/ml) (open bars), OSM (10 ng/ml) (horizontally hatched bars), and IL-1 (1 ng/ml) in combination with OSM (10 ng/ml) (solid bars). Following treatment, total RNA was isolated from cells and reverse transcribed, and the resulting cdna was used in separate real-time quantitative polymerase chain reactions with specific primers and probes for MMP-3 (A), MMP-9 (B), and MMP-14 (C). Results were normalized to 18S ribosomal RNA and presented graphically as relative mrna levels (see Materials and Methods). See Figure 1 for definitions. imal at 48 hours) (Figures 1A C and 2A). A smaller synergistic induction of MMP-14 was also observed following IL-1 /OSM stimulation (Figure 2C). Expression of mrna for MMPs 1, 3, 8, 13, and 14 was also induced by IL-1 alone (Figures 1 and 2). In addition, MMP-9 gene expression was induced by IL-1, but both basal and IL-1 induced levels were inhibited by the addition of OSM (Figure 2B). No clear modulation of MMP-2, MMP-15, MMP-16, or MMP-17 gene expression was observed (data not shown).

5 MODULATION OF MMP, ADAM, AND ADAM-TS IN CHONDROCYTES BY IL-1 AND OSM 965 Figure 3. The effect of IL-1 and OSM on the gene expression of ADAM-10 and ADAM with thrombospondin motifs family members (ADAM-TS1, ADAM-TS4, and ADAM-TS5) in human chondrocytes. T/C28a4 chondrocytes were treated for 2 48 hours with medium alone (control) (diagonally hatched bars), IL-1 (1 ng/ml) (open bars), OSM (10 ng/ml) (horizontally hatched bars), and IL-1 (1 ng/ml) in combination with OSM (10 ng/ml) (solid bars). Following treatment, total RNA was isolated from cells and reverse transcribed, and the resulting cdna was used in separate real-time quantitative polymerase chain reactions with specific primers and probes for ADAM-10 (A), ADAM-TS1 (B), ADAM-TS4 (C), and ADAM-TS5 (D). Results were normalized to 18S ribosomal RNA and presented graphically as relative mrna levels (see Materials and Methods). See Figure 1 for definitions.

6 966 KOSHY ET AL Effect of IL-1 in combination with OSM on the expression of ADAM and ADAM-TS genes. ADAM-10 mrna expression appeared to be moderately upregulated by IL-1 /OSM between 12 hours and 24 hours poststimulation (Figure 3A). No clear modulation of ADAM-17 gene expression was observed (data not shown). A minor, but consistent, induction of ADAM- TS1 mrna was seen following IL-1 stimulation between 12 hours and 24 hours, but was not enhanced by costimulation with OSM (Figure 3B). IL-1 alone and OSM alone failed to increase ADAM-TS4 gene expression, but these cytokines in combination synergized to up-regulate its expression over 4 24 hours (Figure 3C). IL-1 enhanced ADAM-TS5 expression over 4 48 hours; however, the addition of OSM appeared to partially inhibit this increase (Figure 3D). DISCUSSION Aggrecan and collagen loss from cartilage results in the loss of functionality of the tissue and joint disability. It is therefore important to identify the enzymes that are potentially involved in the turnover of these matrix components, and the cytokines that regulate the production of these proteases. We have previously reported that IL-1 and OSM, both present at elevated levels in RA synovial fluid (9,10), are a potent combination in promoting cartilage degradation (9). In this study, we have used real-time PCR to investigate the ability of IL-1 in combination with OSM to modulate the gene expression of aggrecan- and collagen-degrading enzymes, as well as that of other family members, in the T/C28a4 human chondrocyte cell line. The T/C28a4 cell line has previously been characterized for its use as a model of primary chondrocytes (11). Consistent with the data reported in the present study, we have previously shown qualitatively by Northern blotting that IL-1/OSM synergistically induced the expression of MMP-1 mrna in both primary human articular chondrocytes and the T/C28a4 cell line (12). The present study clearly demonstrates the ability of IL-1 and OSM to synergize to potently induce the mrna expression of the collagenases MMP-1, MMP-8, and MMP-13. Interestingly, there appeared to be a differential regulation of these genes with respect to time. Expression of mrna for MMPs 1 and 13 increased early, rapidly peaking at 12 hours before decreasing slowly. These results using real-time RT- PCR are consistent with previous data, in which timecourse studies assessing MMP-1 expression have qualitatively shown by Northern blotting that IL-1/OSM induced MMP-1 expression had clearly decreased by 48 hours following a maximal increase between 8 hours and 24 hours (12). In contrast with the early induction of MMP-1 and MMP-13 shown in the present study, MMP-8 mrna levels were induced slowly and were still increasing by 48 hours. These data may implicate MMP-1 and MMP-13 in the early phase of chondrocytemediated cartilage collagen breakdown, with MMP-8 having a significant role at a later stage. These observations suggest similarities, but also real differences, in the way these genes are regulated. MMPs are synthesized as proenzymes that require activation (2). Thus, the marked induction of MMP-3 mrna, concurrent with that of MMP-1 and MMP-13 and prior to that of MMP-8 mrna, may be of significance since this enzyme is able to activate proforms of all 3 collagenases (8). In addition, the gene expression of MMP-14, an enzyme shown to possess both collagen-degrading activity and the ability to activate prommp-13 (8), was moderately up-regulated by the combination of IL-1 and OSM. MMP-14 is a member of the membrane-type MMPs that can undergo activation within the Golgi by furin-type enzymes during cellular export. Thus, it can be secreted and expressed at the cell surface in the active form (13). MMP-14 may, therefore, play an important role in IL-1/OSM induced cartilage destruction, either directly via pericellular collagen degradation or indirectly via the activation of procollagenases. IL-1 alone was also able to up-regulate gene expression of MMPs 1, 3, 8, 13, and 14. However, costimulation with OSM resulted in the synergistic induction of these mrna. In contrast, whereas IL-1 stimulated MMP-9 mrna expression, OSM inhibited both basal and IL-1 induced expression. IL-1/OSM failed to modulate the gene expression of the gelatinase MMP-2 and that of the membrane type MMPs, MMP- 15, MMP-16, and MMP-17. The marked differences observed in the effect of IL-1 and OSM on the transcriptional regulation of the MMP family suggest that a comparison of the respective gene promoter regions may be revealing. For example, previous studies have shown significant differences between the promoters of the highly inducible MMP-1 and the less responsive MMP-2 gene (see ref. 14 and references therein). It is uncertain, however, what effect this cytokine combination has on the posttranscriptional regulation of these genes. The effect of IL-1 and OSM on the regulation of genes encoding the aggrecan-degrading enzymes ADAM-TS4 and ADAM-TS5 (5,6) was also investigated. Our data show that although IL-1 alone failed to

7 MODULATION OF MMP, ADAM, AND ADAM-TS IN CHONDROCYTES BY IL-1 AND OSM 967 increase ADAM-TS4 mrna levels, the combination of IL-1 and OSM synergistically induced the expression of this gene. In contrast, ADAM-TS5 expression was elevated by IL-1, but IL-1 induced expression was partially inhibited by OSM. This differential regulation may be explained by a close examination of the promoter regions of these genes. These data perhaps implicate ADAM-TS4 as the aggrecanase responsible for the synergistic degradation of aggrecan that occurs following IL-1/OSM stimulation. ADAM-TS5, however, may be important in IL-1 induced aggrecan loss, and may also contribute to this process in IL-1/OSM stimulated cartilage. Compared with ADAM-TS4 and -TS5, ADAM- TS1 mrna regulation showed an intermediate profile. Its expression was induced by IL-1 but was not further modulated by the addition of OSM. ADAM-10, but not ADAM-17, mrna levels were moderately increased following IL-1/OSM treatment. ADAM-10 is known to possess the ability to process pro tumor necrosis factor (TNF ) to its active form (15). Thus, the generation of active TNF may further promote a proinflammatory cycle of events. Several limitations of this type of study must be noted. First, the results show relative levels of mrna expression and thus, although the effect of different treatments on the levels of expression of a given gene can be assessed, comparisons between the levels of different mrna cannot be made. For example, the levels of one collagenase mrna may be far in excess of another, even though both are equally inducible by proinflammatory cytokines. For a determination of absolute levels of mrna, standards for all mrna would be required. Second, mrna expression does not necessarily reflect protein expression and activity. However, despite the limitations, mrna measurements are potentially a good indicator of important downstream mediators of the processes under investigation. In summary, this study has made the following novel observations. First, a marked differential regulation of the genes assessed was demonstrated. This was observed with respect to differences in gene inducibility (e.g., MMP-1 MMP-14 MMP-2), temporal expression (e.g., MMP-8 versus MMP-1 and MMP-13), and responsiveness to cytokine stimulation (e.g., ADAM-TS4 versus ADAM-TS5). Comparative studies of these gene promoters are in progress to identify the mechanisms underlying the apparent differences in the transcriptional regulation of these genes. Second, this study has identified a number of potential mediators of IL-1/OSM induced cartilage destruction. All 3 human collagenases MMP-1, MMP-8, and MMP-13, as well as MMP-14, may be implicated in IL-1/OSM induced collagen degradation, while a role for ADAM-TS4 was suggested in the breakdown of aggrecan. The data also suggested that MMPs 3 and 14 and ADAM-10 may be important in IL-1/OSM induced cartilage breakdown, perhaps via the activation of prommps and protnf, respectively. Thus, the combination of IL-1 and OSM, both of which are up-regulated in the rheumatoid joint, has been shown to induce the chondrocyte expression of genes encoding enzymes that can degrade the principal components of articular cartilage. REFERENCES 1. Muir H. The chondrocyte, architect of cartilage: biomechanics, structure, function and molecular biology of cartilage matrix macromolecules. BioEssays 1995;17: Cawston TE. Metalloproteinase inhibitors and the prevention of connective tissue breakdown. Pharmacol Ther 1996;70: Primakoff P, Myles DG. The ADAM gene family: surface proteins with adhesion and protease activity. Trends Genet 2000;16: Kaushal GP, Shah SV. The new kids on the block: ADAMTSs, potentially multifunctional metalloproteinases of the ADAM family. J Clin Invest 2000;105: Tortorella MD, Burn TC, Pratta MA, Abbaszade I, Hollis JM, Liu R, et al. Purification and cloning of aggrecanase-1: a member of the ADAMTS family of proteins. Science 1999;284: Abbaszade I, Liu R-Q, Yang F, Rosenfield SA, Ross OH, Link JR, et al. Cloning and characterization of ADAMTS11, an aggrecanase from the ADAMTS family. J Biol Chem 1999;274: Sandy JD, Flannery CR, Neame PJ, Lohmander LS. The structure of aggrecan fragments in human synovial fluid: evidence for the involvement in osteoarthritis of a novel proteinase which cleaves the Glu373-Ala374 bond of the interglobular domain. J Clin Invest 1992;89: Bigg HF, Rowan AD. The inhibition of metalloproteinases as a therapeutic target in rheumatoid arthritis and osteoarthritis. Curr Opin Pharmacol 2001;1: Cawston TE, Curry VA, Summers CA, Clark IM, Riley GP, Life PF, et al. The role of oncostatin M in animal and human connective tissue collagen turnover and its localization within the rheumatoid joint. Arthritis Rheum 1998;41: Westacott CI, Whicher JT, Barnes IC, Thompson D, Swan AJ, Dieppe PA. Synovial fluid concentration of five different cytokines in rheumatic diseases. Ann Rheum Dis 1990;49: Goldring MB, Birkhead JR, Suen L-F, Yamin R, Mizuno S, Glowacki J, et al. Interleukin-1 modulated gene expression in immortalized human chondrocytes. J Clin Invest 1994;94: Rowan AD, Koshy PJT, Shingleton WD, Degnan BA, Heath J, Vernallis AB, et al. Synergistic effects of gp130 binding cytokines in combination with interleukin-1 on cartilage collagen breakdown. Arthritis Rheum 2001;44: Sato H, Kinoshita T, Takino T, Nakayama K, Seiki M. Activation of a recombinant membrane type 1-matrix metalloproteinase (MT1-MMP) by furin and its interaction with tissue inhibitor of metalloproteinase (TIMP)-2. FEBS Lett 1996;393: Mauviel A. Cytokine regulation of metalloproteinase gene expression. J Cell Biochem 1993;53: Rosendahl MS, Ko SC, Long DL, Brewer MT, Rosenzweig B, Hedl E, et al. Identification and characterization of a pro-tumor necrosis factor-alpha-processing enzyme from the ADAM family of zinc metalloproteases. J Biol Chem 1997;272: