Differentiation /CXCL1 Induce Chondrocyte Hypertrophic Α IL-8

IL-8/CXCL8 and Growth-Related Oncogene α/CXCL1 Induce Chondrocyte Hypertrophic Differentiation

This information is current as Denise Merz, Ru Liu, Kristen Johnson and Robert of September 25, 2021. Terkeltaub J Immunol 2003; 171:4406-4415; ; doi: 10.4049/jimmunol.171.8.4406 http://www.jimmunol.org/content/171/8/4406 Downloaded from

References This article cites 69 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/171/8/4406.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 25, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2003 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

IL-8/CXCL8 and Growth-Related Oncogene ␣/CXCL1 Induce Chondrocyte Hypertrophic Differentiation1

Denise Merz, Ru Liu, Kristen Johnson, and Robert Terkeltaub2

Foci of chondrocyte hypertrophy that commonly develop in osteoarthritic (OA) cartilage can promote dysregulated matrix repair and pathologic calcification in OA. The closely related chemokines IL-8/CXCL8 and growth-related oncogene ␣ (GRO␣)/CXCL1 and their receptors are up-regulated in OA cartilage chondrocytes. Because these chemokines regulate leukocyte activation through p38 mitogen-activated protein kinase signaling, a pathway implicated in chondrocyte hypertrophic differentiation, we tested whether IL-8 and GRO␣ promote chondrocyte hypertrophy. We observed that normal human and bovine primary articular chondrocytes expressed both IL-8Rs (CXCR1, CXCR2). IL-8 and the selective CXCR2 ligand GRO␣ (10 ng/ml) induced tissue inhibitor of metalloproteinase-3 expression, markers of hypertrophy (type X collagen and MMP-13 expression, alkaline ␣ phosphatase activity), as well as matrix calcification. IL-8 and the selective CXCR2 ligand GRO also induced increased trans- Downloaded from amidation activity of chondrocyte transglutaminases (TGs), enzymes up-regulated in chondrocyte hypertrophy that have the potential to modulate differentiation and calcification. Under these conditions, p38 mitogen-activated protein kinase pathway signaling mediated induction of both type X collagen and TG activity. Studies using mouse knee chondrocytes lacking one of the two known articular chondrocyte-expressed TG isoenzymes (TG2) demonstrated that TG2 was essential for murine GRO␣ homologue KC-induced TG activity and critically mediated induction by KC of type X collagen, matrix metalloproteinase-13,

alkaline phosphatase, and calcification. In conclusion, IL-8 and GRO␣ induce articular chondrocyte hypertrophy and calcification http://www.jimmunol.org/ through p38 and TG2. Our results suggest a novel linkage between inflammation and altered differentiation of articular chondrocytes. Furthermore, CXCR2 and TG2 may be sites for intervention in the pathogenesis of OA. The Journal of Immunology, 2003, 171: 4406Ð4415.

ow-grade inflammatory alterations within synovium Chondrocytes in OA cartilages frequently develop increased ex- and cartilage modulate pathogenesis in osteoarthritis pression of chemokines, including the CC chemokine subfamily L (OA),3 with effects of IL-1 on cartilage a prime exam- member RANTES and the CXC chemokine subfamily members ple (1Ð4). IL-1 induces expression of several matrix-degrading IL-8 (CXCL8) and GRO␣ (CXCL1) (13Ð15). Significantly, the proteases, including matrix metalloproteinases (MMPs), 1, 3, highly selective IL-8R CXCR1 and the relatively promiscuous by guest on September 25, 2021 and 13, and a disintegrin and metalloproteinase ts4/5 (1Ð5). CXC chemokine receptor CXCR2 are both expressed in situ in IL-1 also induces inducible NO synthase expression and up- normal cartilage, and CXCR1 expression may moderately increase regulated NO production, which suppresses PG synthesis and in OA cartilage (16, 17). promotes MMP activation (6Ð9). Furthermore, IL-1 induces Chemokine functions in leukocyte adhesion and migration and matrix calcification by chondrocytes (10, 11), a common patho- organization of inflammatory reactions are recognized (18). But logic event associated with OA in situ (12). However, the level RANTES induces chondrocyte expression of inducible NO syn- at which IL-1 is expressed within OA cartilage is variable, and thase, MMP-1, and IL-6, and stimulates PG depletion in cartilage chondrocyte responsiveness to IL-1 in OA can be diminished by explants (13), which has focused attention on potential pro-OA ␤ soluble IL-1R antagonist and type II IL-1 decoy receptor (1, 2, effects of chemokines exerted directly on chondrocytes. In this 4). Therefore, inflammatory cytokines other than IL-1 that can regard, high concentrations of growth-related oncogene ␣ (GRO␣) promote OA are of interest. (at Ն1 ␮g/ml) induce at least two matrix-degrading enzymes, MMP-3 and lysosomal N-acetyl-␤-D-glucosaminidase, in cartilage explants or cultured chondrocytes, respectively (16). GRO␣ (at Ն1 Veterans Affairs Medical Center, University of California at San Diego, La Jolla, CA ␮g/ml) also stimulates chondrocyte apoptosis (19). Hence, GRO␣ Received for publication March 13, 2003. Accepted for publication August 7, 2003. and IL-8 effects on chondrocytes may be pertinent in OA. The costs of publication of this article were defrayed in part by the payment of page In leukocytes, IL-8 and GRO␣ regulate adhesion and migration charges. This article must therefore be hereby marked advertisement in accordance by defined signal transduction mechanisms, including the p38 mi- with 18 U.S.C. Section 1734 solely to indicate this fact. togen-activated protein kinase (MAPK) pathway (20, 21). In chon- 1 Supported by grants to R.T. from Department of Veterans Affairs and National Institutes of Health (P01AGO7996) and VC BioSTAR program, and a National In- drocytes, p38 MAPK pathway activation promotes both hypertro- stitutes of Health R03 Award and an award from the University of California at San phic differentiation and apoptosis (22, 23). Furthermore, resting Diego Stein Institute for Aging to R.L. articular chondrocytes in OA cartilage can undergo transition to 2 Address correspondence and reprint requests to Dr. R. Terkeltaub, VA Medical hypertrophic and apoptotic cells, partially recapitulating chondro- Center, 3350 La Jolla Village Drive, San Diego, CA 92161. E-mail address: [email protected] cyte differentiation changes in endochondral mineralization (2, 3 Abbreviations used in this paper: OA, osteoarthritis; AP, alkaline phosphatase; 11). Chondrocyte hypertrophy and apoptosis in OA can contribute FXIIIA, factor XIIIA; GRO, growth-related oncogene; JNK, c-Jun N-terminal kinase; to dysregulation of matrix repair via alterations in collagen subtype MAPK, mitogen-activated protein kinase; MMP, matrix metalloproteinase; PLA2, and MMP expression, and in cell viability, respectively (24Ð26). phospholipase A2; polyHEME; polyhydroxyethylmethacrylate; TG, transglutaminase; TIMP, tissue inhibitor of metalloproteinases. Chondrocytes that have undergone hypertrophic differentiation or

FIGURE 1. Flow cytometric analysis of CXCR1 and CXCR2 expression in human and bovine articular chondrocytes. Primary articular chondrocytes of the indicated species were permeabilized, and flow cytometric analysis was performed using anti-CXCR1, anti-CXCR2, and IgG isotype control, as described in Materials and Methods. These data are representative of three independent experiments. by guest on September 25, 2021 apoptosis in OA cartilage most likely promote pathologic calcifi- Materials and Methods cation, analogous to the major contribution of chondrocyte hyper- Reagents trophy to physiologic growth plate mineralization (11, 27). Patho- Human rIL-1␤, rIL-8, rGRO␣, and murine rKC/GRO␣ were from R&D logic calcification in OA cartilage can stimulate crystal-induced Systems (Minneapolis, MN). PD98059, c-Jun N-terminal kinase (JNK) inflammation that promotes further cartilage degradation (11, 12). inhibitor II, and SB203580 were from Calbiochem (San Diego, CA). All Markers of chondrocyte hypertrophic differentiation in the other reagents were from Sigma-Aldrich (St. Louis, MO), unless otherwise growth plate include up-regulated expression of the transglutami- indicated. nase (TG) isoenzymes TG2 (also termed tissue TG and Gh) and Bovine and human chondrocyte isolation and culture factor XIIIA (FXIIIA) (10, 11). TGs are mediators of tissue repair, Normal human knee articular chondrocytes were obtained from M. Lotz and catalyze calcium-dependent transamidation, resulting in cova- (The Scripps Research Institute, La Jolla, CA), as described (29). Primary lent cross-linking of available substrate glutamine residues to a bovine chondrocytes were obtained from articular cartilage slices removed primary amino group (EC 2.3.2.13) (10, 11, 28) that can thereby from the medial and lateral condyles, the patellar grove, and the tibial stabilize matrix proteins, including several collagen subtypes and plateau of normal mature bovine knee joints (Animal Technologies, Tyler, TX) (11). Chondrocytes were isolated using collagenase digestion and fibronectin (10, 11, 28). Furthermore, TG transamidation activity, placed in monolayer culture in DMEM/high glucose supplemented with which is up-regulated in part by posttranslational modifications of 10% FCS, 1% L-glutamine, 100 U/ml penicillin, and 50 ␮g/ml streptomy- TGs (28), increases in both an OA severity-dependent and age- cin at 37¡C with 5% CO2, as described (11). For treatment of chondrocytes ϫ 6 dependent manner in joint cartilages (10). with cytokines under nonadherent conditions, aliquots of 0.1 10 cells were plated on polyhydroxylethylmethacrylate (polyHEME)-coated 96- Selective gain-of-function of either chondrocyte TG2 or FXIIIA well round-bottom plates in DMEM high glucose supplemented with only TG activity via transfection stimulates cultured chondrocytes to 1% FCS, 100 U/ml penicillin, and 50 ␮g/ml streptomycin (11). For studies calcify their matrix (10). But IL-1␤ induces TG transamidation of calcification with bovine and human cells, the medium was additionally supplemented with 2.5 mM sodium phosphate and 25 ␮g/ml of L-ascorbic activity and stimulates calcification in cultured chondrocytes in a acid 2-phosphate (11). TG2-dependent manner (10, 11). In addition, TG2 is required for retinoic acid-induced articular chondrocyte hypertrophic differen- Mouse knee chondrocyte isolation and culture tiation in vitro (11). We established a breeding colony of TG2 knockout mice (30), as well as In this study, we demonstrate that the CXCR2 ligands IL-8 and congenic wild-type control mice on a C57BL6/129SVJ background, as GRO␣ induce chondrocyte hypertrophy mediated by p38 signaling described (11). Primary mouse articular chondrocytes were isolated by and TG2. Our results suggest that chemokine-mediated inflamma- dissection of the tibial plateaus and femoral condyles of mice at 2 mo of age, as described (11). In brief, the articular cartilage was carefully peeled tion can promote altered chondrocyte differentiation and calcifica- off with a scalpel to avoid subchondral bone disruption, and chondrocytes tion in OA. extracted by collagenase digestion were plated in monolayer culture in 4408 CHEMOKINE-INDUCED CHONDROCYTE HYPERTROPHY Downloaded from http://www.jimmunol.org/

FIGURE 2. Altered matrix calciﬁcation and NO release in response to IL-8. Primary human (A) or bovine (B) articular chondrocytes (1 ϫ 105 cells/well in a 96-well plate) were stimulated with 10 ng/ml IL-1 or IL-8 in the presence of medium supplemented with 2.5 mM sodium phosphate and 25 ␮g/ml ascorbic acid. Fresh medium and stimuli were added every third day. Bound Alizarin Red S was analyzed at 14 days. The data represent the average of

8 replicates from 10 different human and 8 different bovine donors. Primary human (C) and bovine (D) articular chondrocytes were stimulated, as in A. by guest on September 25, 2021 NO release was measured via the Greiss reaction at 24 h after stimulation. The data are pooled from 5 different human and 4 different bovine donors, studied .p Ͻ 0.05 ,ء .in replicates of 8

DMEM high glucose supplemented with 10% FCS, 1% glutamine, 100 and tubulin (Sigma-Aldrich) were used. To assess MMP-13 protein, ali- ␮ o U/ml penicillin, and 50 g/ml streptomycin at 37 Cin5%CO2 for 7 days quots of 0.2 ml conditioned medium were precipitated using 15% trichlo- before initiation of each experiment. Subconfluent chondrocytes were roacetic acid and centrifuged, and the resulting pellet was resuspended in Ͼ95% type II collagen expression. Approximately 2500 primary chondro- 0.05% NaOH. Aliquots of 10 ␮g of precipitated protein were studied by cytes were obtained initially from a pair of knee joints from each mouse. SDS-PAGE/Western blotting. Mouse chondrocytes were allowed to proliferate for 5 days, yielding For RT-PCR analyses, total RNA was prepared and reversed tran- ϳ10,000 cells per pair of knees per mouse. For the described experiments, scribed, as described (29). Primers and RT-PCR conditions for the ribo- knees from 30 mice of each genotype were harvested, and the chondrocytes somal housekeeping gene L30 were described earlier (31). Type X collagen of each genotype were pooled upon isolation and plated at ϳ70% conflu- primers were sense 5Ј-TAGGAGCTAAAGGAGTGCCTGGAC-3Ј and an- ency in DMEM high glucose, with 1% FCS, 1 mM sodium phosphate, 50 tisense 5Ј-GCATACCTGTTACCCCGTGGTTAG-3Ј, which amplified a ␮g/ml of ascorbic acid, and glutamine and antibiotics, as above. 369-bp product in human and bovine chondrocytes confirmed by sequenc- ing to be type X collagen. CXCR1 and CXCR2 assessment by flow cytometry and IL-8 ELISA Primary human and bovine articular chondrocytes were permeabilized us- NO production, and MMP-13, alkaline phosphatase (AP), TG, ing the BD Cytofix/Cytoperm kit (BD PharMingen, San Diego, CA) and and kinase assays incubated for 30 min at 4oC with murine monoclonal anti-CXCR1 (Bio- NO production was measured using the Greiss reaction, as previously de- source, Camarillo, CA) and anti-CXCR2 (BD PharMingen), or murine IgG scribed (10). To assay MMP-13 activity, aliquots of 50 ␮l conditioned isotype control. Washed cells were incubated with FITC-conjugated goat medium were added to individual wells in a 96-well plate containing 25 F(abЈ) anti-mouse Ig (Biosource) for 30 min at 4oC. Fluorescence was 2 ␮M of MMP-13 fluorogenic substrate (7-methoxycoumarin-4-yl)-acetyl- detected using a FACSCaliber apparatus (BD Biosciences, San Jose, CA) Pro-Cha-Gly-Nva-His-Ala-Dpa-NH (Cha ϭ L-cyclohexylalanine; Dpa ϭ with data analyzed using CellQuest software (Purdue University, West 2 3-(2,4-dinitrophenyl)-L-2,3-diaminopropionyl; Nva ϭ L-norvaline) (Cal- Lafayette, IN). Expression of IL-8 was determined using the human IL-8 ␮ biochem) (32) in 50 l of 200 mM NaCl, 50 mM Tris-HCl, 5 mM CaCl2, ELISA kit (Biosource), according to the manufacturer protocol. ␮ 20 m ZnSO4, and 0.05% BRIJ 35, pH 7.5, for 18 h at 37¡C. Fluorescence SDS-PAGE/Western blotting and RT-PCR analyses was read at excitation 325 nm, emission 393 nm. We determined AP sp. act., as previously described, with units of AP designated as moles of Preparations of cell lysates, protein concentration determinations, and 10% substrate hydrolyzed per hour (per gram protein in each sample) (11). SDS-PAGE/Western blots were performed, as described (11). Primary and To assay TG transamidation activity, we modified a spectrophotometric secondary Ab dilutions were 1/2000. Commercial polyclonal Abs to type protocol for 5-biotinamidopentylamine binding to dimethylcasein (33), as X collagen, TG2, FXIIIA (from Calbiochem), MMP-13 (Chemicon, Te- previously described (11). One unit of TG was designated as 1 ␮M sub- mecula, CA), tissue inhibitor of metalloproteinases (TIMP-3) (Chemicon), strate catalyzed per hour. The Journal of Immunology 4409

Kinase assays for activities of extracellular signal-regulated kinase 1/2, same conditions, an absence of induction of type X collagen by JNK, and p38 mitogen-activated protein kinases were conducted using sub- IL-1 (10 ng/ml) was confirmed (11) (Fig. 3, A and B). Because strates and protocols from Cell Signaling Technology (Beverly, MA), with IL-1 can induce IL-8 expression in cultured chondrocytes (34Ð36), aliquots of 2 ϫ 106 bovine chondrocytes studied in individual wells of a polyHEME-coated 24-well culture plate. we assessed for the levels of basal vs IL-1-stimulated IL-8 expression by chondrocytes in the nonadherent culture system used in Matrix calcification this study. To do so, we exclusively used human cells because the To quantify matrix calcification, we used a previously described spectro- IL-8 ELISA used did not detect bovine IL-8. We observed that photometric assay system for Alizarin Red S binding to precipitated cal- IL-1 (10 ng/ml) significantly induced IL-8 release, but that the cium (10, 11). IL-1-induced IL-8 release achieved only modest levels (Ͻ100 Statistical analyses pg/ml in the conditioned medium) (Fig. 3C). Furthermore, IL-8- induced chondrocyte type X collagen expression, which was ro- Where indicated, error bars represent SD. Statistical analyses were per- bust in cells treated with 10 ng/ml IL-8, was modest in cells treated formed using the Student’s t test (paired two-sample testing for means). with only 0.1 and 1 ng/ml IL-8 (Fig. 3D). Results We next tested for effects of IL-8 on production of MMP-13, Effects of IL-8 on chondrocyte differentiation which parallels type X collagen during maturation of growth plate chondrocytes and mediates mineralization (37, 38). IL-8 and IL-1 Both normal human and bovine knee articular chondrocytes in first both induced MMP-13 secretion from normal bovine articular passage culture expressed both IL-8Rs (CXCR1 and CXCR2) (Fig. chondrocytes (Fig. 4A). In contrast, IL-8, but not IL-1, induced the

1). Next, we compared responses to IL-1 and IL-8 in chondrocytes MMP inhibitor TIMP-3 (Fig. 4B). Concordantly, increased Downloaded from using polyHEME-coated plates to establish nonadherent culture MMP-13 enzyme activity in conditioned medium was induced by conditions. We confirmed (10, 11) that IL-1 (10 ng/ml) stimulated IL-1, but not IL-8 (Fig. 4C). matrix calcification, and IL-1 also was confirmed (6, 8) to mark- Next, we observed that IL-8 induced two additional markers of edly increase NO release (i.e., by well over 10-fold) in chondro- hypertrophic chondrocyte differentiation, increased TG and AP ac- cytes (Fig. 2). IL-8 (10 ng/ml) stimulated calcification comparably tivity (10, 11), in normal bovine articular chondrocytes (Fig. 5). to IL-1, but induced NO release by only 3- to 6-fold in human and IL-1 was confirmed (10, 11) to induce TG activity (Fig. 5A), but http://www.jimmunol.org/ bovine chondrocytes (Fig. 2). Having observed results comparable IL-1 significantly suppressed the level of AP activity in normal in human and bovine knee chondrocytes for IL-8R expression and articular chondrocytes (Fig. 5B). Hence, there was substantial dis- responses to IL-1 and IL-8 (Figs. 1 and 2), we further examined sociation between IL-1 and IL-8 effects on chondrocyte differen- IL-8 effects on differentiation using predominantly bovine knee tiation (Figs. 2Ð5). chondrocytes, which were more readily available. Where appro- priate (e.g., for experiments using an ELISA specific for human ␣ protein), we used normal human knee chondrocytes. GRO induces chondrocyte hypertrophic differentiation IL-8 (10 ng/ml) induced the stereotypic marker of chondrocyte comparable to IL-8 hypertrophy type X collagen at the mRNA and protein expression IL-8 can differentially activate signal transduction and cell func- by guest on September 25, 2021 levels in bovine articular chondrocytes (Fig. 3, A and B). Under the tions through CXCR1 and CXCR2 (39, 40). Because GRO␣ is an

FIGURE 3. Differential induction of type X collagen in response to IL-1 and IL-8. Primary human articular chondrocytes (1 ϫ 105 cells/well in a 96-well plate) were stimulated with 10 ng/ml rIL-1 or rIL-8. A, Total RNA was harvested on days 0, 1, and 3. RT-PCR analyses for type X collagen and L30 were conducted, as described. B, SDS-PAGE and Western blotting analysis for type X collagen was performed on cell lysates at 5 days in culture. Tubulin was used as a loading control. C, IL-8 ELISA was performed on conditioned medium from cells stimulated with 10 ng/ml IL-1 at 5 days in culture. D, SDS-PAGE and Western blotting analysis for type X collagen was performed on cell lysates at 5 days in culture after stimulation with 0.1, 1, or 10 ng/ml IL-8, as indicated. Tubulin was used as the internal loading control. Results illustrated are all from human cells (representative of four independent .p Ͻ 0.05 ,ء .(experiments 4410 CHEMOKINE-INDUCED CHONDROCYTE HYPERTROPHY Downloaded from FIGURE 4. Concurrent MMP-13 and TIMP-3 induction by IL-8. Primary bovine articular chondrocytes (1 ϫ 105 cells/well in a 96-well plate) were stimulated with 10 ng/ml IL-1 or IL-8. A, SDS-PAGE and Western blot analysis for MMP-13 was performed on the precipitated conditioned medium at day 3 of stimulation, as described in Materials and Methods, with 50- and 57-kDa forms detected. B, SDS-PAGE and Western blot analysis for TIMP-3 was performed on cell lysates at 3 days of stimulation. Two alternatively glycosylated forms are visible at 24 and 30 kDa. C, MMP-13 activity was measured fluorometrically in triplicate (excitation 325 nm, emission 393 nm) at days 2, 3, 5, and 7 and expressed in picograms of substrate cleaved per 105 cells, .p Ͻ 0.05 ,ء .as described in Materials and Methods. Data representative of three independent experiments http://www.jimmunol.org/ activating ligand of CXCR2, but not CXCR1 (18), we next as- duced kinase activities of p44/42, JNK, and p38 in bovine chon- sessed whether GRO␣ shared the capacity of IL-8 to stimulate drocytes (Fig. 7A). Selective pharmacologic inhibition of the p38 hypertrophic differentiation in normal bovine articular chondro- pathway (using SB203580, 25 ␮M), but not selective PD98059 cytes (Fig. 6). We observed that comparable to IL-8, GRO␣ in- and JNK inhibitor II inhibition of the p44/42 or JNK pathways, duced matrix calcification, type X collagen, and MMP-13 expres- respectively, using inhibitor doses and conditions validated to give sion, and TG activity consistent with hypertrophic differentiation. selective suppression of each MAPK pathway (not shown), GRO␣ also induced TIMP-3 expression and did not induce in- blocked type X collagen expression in response to IL-8 (Fig. 7B). creased MMP-13 activity (Fig. 6), comparable to the effects of Similarly, selective pharmacologic p38 inhibition, but not p44/42 by guest on September 25, 2021 IL-8 (Fig. 4). Thus, CXCR2-mediated signaling appeared suffi- or JNK inhibition, blocked the capacity of not only IL-8, but also cient to promote articular chondrocyte hypertrophy. GRO␣ to induce TG activity in chondrocytes (Fig. 7C).

Role of p38 pathway signaling in IL-8-induced chondrocyte hypertrophy Central role of TG2 in chondrocyte hypertrophic differentiation ␣ Because IL-8-induced p38 pathway signaling mediates certain induced by KC/GRO functional responses in leukocytes (21), we next examined the Results to this point indicated that IL-8 and GRO␣ induced cal- roles of signaling through p38 and the related MAPKs p44/42 and ciﬁcation and multiple markers of chondrocyte hypertrophic dif- JNK in IL-8-induced chondrocyte hypertrophy. IL-8 rapidly in- ferentiation, including increased TG activity. We conﬁrmed (10,

FIGURE 5. TG activity and alkaline phosphatase (AP) activity in response to IL-8. Primary bovine articular chondrocytes (1 ϫ 105 cells/well in a 96-well plate) were stimulated with 10 ng/ml IL-1 or IL-8 in medium supplemented wth 2.5 mM sodium phosphate and 25 ␮g/ml ascorbic acid, as described in Materials and Methods. A, TG activity was assayed at 48 h, as described in Materials and Methods, with these data pooled from ﬁve different donors in replicates of three. B, AP activity was analyzed at 72 h, as described in Materials and Methods, with these data pooled from seven donors studied in .p Ͻ 0.05 ,ء .triplicate The Journal of Immunology 4411 Downloaded from http://www.jimmunol.org/

FIGURE 6. Induction of chondrocyte hypertrophy by GRO␣. Primary bovine chondrocytes (1 ϫ 105 cells/well in a 96-well plate) were stimulated with 10 ng/ml GRO␣. Using the methods described above, we performed analyses of matrix calciﬁcation (A), type X collagen expression by RT-PCR (B), and by guest on September 25, 2021 SDS-PAGE and Western blotting (C), TG activity (D), MMP-13 protein released in conditioned medium (E), MMP-13 activity (E), TIMP-3 protein (relative to tubulin as a control) in cell lysates, with results spliced from the same blot (F), and MMP-13 activity (G). Results of B are from human cells, .p Ͻ 0.05 ,ء .and the rest are from bovine cells

11) by SDS-PAGE/Western blot analysis that articular chondro- calcification in chondrocytes. Our findings were consistent with cytes expressed the TG isoenzymes FXIIIA and TG2 under the comparable states of chondrocyte hypertrophic differentiation in- culture conditions used in this study (data not shown). Because duced through CXCR2-mediated signaling that involved down- TG2 can modulate cell differentiation (11), we tested the role of stream p38 pathway activation and TG2 activation (Fig. 9). We TG2 in CXCR2 ligand-induced TG activity, hypertrophic differ- observed that articular chondrocyte CXCR2 expression was sub- entiation, and calcification. To do so, we evaluated knee articular stantially more prominent in bovine than human cells used in these chondrocytes from the TG2 null mouse (11, 30). Because mice do studies. Nevertheless, CXCR2 ligands induced hypertrophic dif- not express homologues of either IL-8 or CXCR1 (41), we tested ferentiation of both human and bovine articular chondrocytes. functional responses to the murine GRO␣ homologue KC (Fig. 8). Effects of IL-8 and GRO␣ on chondrocytes diverged from those TG2 and FXIIIA are the only TG isoenzymes detectable in nor- of IL-1. First, the CXC chemokines (but not IL-1) induced mal articular chondrocytes, and TG2Ϫ/Ϫ mouse chondrocytes re- TIMP-3. Second, IL-1 failed to induce type X collagen and AP. tain FXIIIA expression (11). In this study, we first confirmed (11) Third, IL-1, but not the CXC chemokines induced markedly in- that TG transamidation activity was not absent, but reduced by creased extracellular MMP-13 activity, despite the fact that IL-1 Ϫ Ϫ ϳ50%, in resting TG2 / chondrocytes compared with congenic and the chemokines both induced MMP-13. The relatively weak ϩ ϩ TG2 / controls (Fig. 8A). KC/GRO␣ induced a significant in- NO release induced by CXCR2 ligands, in addition to induction of crease of TG activity in wild-type, but not TG2 null chondrocytes TIMP-3 expression, most likely contributed to the negligible (Fig. 8A). Concurrently, the capacity of KC/GRO␣ to induce type MMP-13 activity in conditioned medium of IL-8- and GRO␣- X collagen and MMP-13 was markedly diminished, and induction treated cells. Ϫ Ϫ of AP activity attenuated, in TG2 / chondrocytes (Fig. 8, BÐD). Chronic inflammatory oxidative stress, and several cytokines Last, the capacity of KC/GRO␣ to induce calcification was atten- expressed in OA joints (e.g., IL-1, TNF-␣, and IL-17), have the Ϫ Ϫ uated in TG2 / mouse chondrocytes (Fig. 8E). potential to induce IL-8 and GRO␣ expression in chondrocytes (34Ð36, 42, 43). In addition, IL-1 stimulates both p38 pathway Discussion signaling (44) and a TG2-dependent increase in TG activity (11) in In this study, IL-8 and GRO␣, which are expressed in OA carti- chondrocytes. But under the conditions we used, the extracellular lage, induced type X collagen, MMP-13, AP, and TG activity, and concentration of IL-8 (Ͻ100 pg/ml) in response to 10 ng/ml IL-1 4412 CHEMOKINE-INDUCED CHONDROCYTE HYPERTROPHY Downloaded from http://www.jimmunol.org/

FIGURE 7. Role of p38 pathway signaling in IL-8-induced type X collagen expression and TG activity. A, Aliquots of 2 ϫ 106 bovine chondrocytes were stimulated with IL-8 for the times indicated, as described in Materials and Methods. The nonradioactive kinase assays for p38, JNK, and p44/42 ϫ 5 activities were performed, as described in Materials and Methods; data representative of three independent experiments. B, Aliquots of 1 10 bovine by guest on September 25, 2021 chondrocytes were pretreated for 1 h with the indicated selective pharmacological inhibitor of individual MAPK signaling pathways (PD98059, 50 ␮M for p44/42; JNK inhibitor II, 25 ␮M for JNK; SB203580, 25 ␮M for p38), then stimulated with IL-8 for 5 days. SDS-PAGE and Western blotting for type X collagen and tubulin were performed, as described above; data representative of three independent experiments. C, Aliquots of 1 ϫ 105 bovine chondrocytes were pretreated for 1 h with the selective p44/42, JNK, and p38 pathway inhibitors, as described above, then stimulated with 10 ng/ml IL-8 .p Ͻ 0.05 ,ء .or GRO␣ for 48 h. Chondrocyte TG activity was analyzed as above; data pooled from four independent experiments in triplicate was below the threshold for the robust type X collagen expression flammatory joint fluids, whereas OA synovial fluid concentrations in response to IL-8 present at 1 and 10 ng/ml. It is possible that of IL-8 have been reported to range between 1 and 5 ng/ml (52, under more sustained culture, or under different culture conditions, 53), with GRO␣ levels averaging 0.1 ng/ml in one study (54). IL-1 might induce chondrocyte hypertrophic differentiation medi- However, low-level type X collagen induction was detectable in ated by IL-8 and GRO␣ expression. But differences in IL-1 and chondrocytes treated with 0.1 ng/ml IL-8. In prior studies of chemokine signal transduction and in NO generation may have GRO␣, effects on cultured chondrocytes and cartilage explants, been contributors to the lack of chondrocyte hypertrophy in re- investigators used much higher concentrations of GRO␣ (1Ð5 ␮g/ sponse to IL-1 under the conditions used in this study. We spec- ml), which induced MMP-3 expression, N-acetyl-␤-D-glu- ulate that differing levels of expression of the broad-spectrum cosaminidase release, and apoptosis (16, 19). We did not examine MMP and a disintegrin and metalloproteinase ts4/5 inhibitor these particular responses in this study. But it should be noted that TIMP-3 (45), of MMP-13 activity, and of AP activity also were hypertrophy precedes apoptosis in sequential endochondral chon- significant factors modulating the distinct differentiation responses drocyte differentiation (55). Moreover, hypertrophic and apoptotic to IL-1 and the CXC chemokines seen in this study. In addition, cells colocalize near calcifications in association with cartilage de-

AP generates Pi (46, 47), a direct promoter of expression of type generation (27, 55). As such, CXCR2 ligands are potentially fac- X collagen and certain other genes expressed by hypertrophic tors in promoting both hypertrophy and apoptosis in OA cartilage. Ϫ/Ϫ chondrocytes (48Ð50). Significantly, AP activity and Pi metabo- Using TG2 knee chondrocytes, we discovered that TG2 was lism exert marked regulatory effects on both chondrocyte differ- centrally involved in not only GRO␣-induced increases in TG ac- entiation and organization in the endochondral growth plate (51). tivity, but also in hypertrophic differentiation and calcification. We In view of the results in this study, the capacity of inhibition of have not observed induction of TG2 or FXIIIA expression by MMP-13 to prevent prehypertrophic growth plate chondrocytes GRO␣ or IL-8 in chondrocytes (D. Merz et al., unpublished ob- from progressing to hypertrophy and mineralizing the matrix in servations). Hence, TG2 activation by the chemokines was most vitro appears paradoxical (37). likely exerted primarily at the level of posttranslational modifica- In this study, we defined relatively robust chondrocyte responses tions (11, 28, 56, 57). FXIIIA and TG2 are the only TG isoen- to doses of IL-8 and GRO␣ of 10 ng/ml. IL-8 and GRO␣ concen- zymes we have detected in articular chondrocytes (10, 11). We trations in the range of 5Ð20 ng/ml are commonly detected in in- confirmed (11) in this study that TG activity was only reduced by The Journal of Immunology 4413 Downloaded from http://www.jimmunol.org/

Ϫ Ϫ FIGURE 8. Altered effects of KC/GRO␣ on murine TG2 / chondrocytes. To determine KC/GRO␣ TG activity (A), we isolated primary knee articular by guest on September 25, 2021 chondrocytes from 2-mo-old TG2ϩ/ϩ and TG2Ϫ/Ϫ mice and plated the cells in 24-well plates (5 ϫ 104 cells/well), as described in Materials and Methods. Adherent cells were stimulated with 10 ng/ml KC/GRO␣, where indicated, for 48 h, as described in Materials and Methods, and TG activity was determined as above, with cells studied in triplicate for each condition. For studies in BÐE, aliquots of adherent primary mouse articular chondrocytes (1 ϫ 104 cells/well in a 96-well plate) were grown for up to 14 days, as described in Materials and Methods (supplemented with 1% FCS, 1 mM sodium phosphate, and 50 ␮g/ml of ascorbic acid), in the presence of 10 ng/ml KC/GRO␣, where indicated. In B, SDS-PAGE and Western blot analysis for type X collagen in cell lysates was performed at 10 days. In C, SDS-PAGE and Western blotting was performed for MMP-13 using aliquots of 10 ␮g of protein precipitated from conditioned medium after 3 days. In D, AP sp. act. was measured at 7 days, as described above. In E, the fresh medium and agonist were replaced every 3 days, and matrix calciﬁcation was assayed at 14 days. Data all are representative of three experiments, with each result pooled from chondrocytes .p Ͻ 0.05 ,ء .of 30 mice of TG2ϩ/ϩ and TG2Ϫ/Ϫ genotypes

ϳ50% in resting TG2Ϫ/Ϫ mouse knee chondrocytes, consistent growth (46, 63Ð65). Significantly, TG2 catalyzes formation of in- with a major contribution of FXIIIA to basal TG activity. Physi- tramolecular cross-links that increase phospholipase A2 (PLA2) ologically, the partial functional redundancy of TG2 and FXIIIA in activity (66). TG2-modulated PLA2 activation can modulate in- cartilages (10) appears to contribute to the grossly normal skeletal flammatory responses (66). It will be of interest to define whether Ϫ/Ϫ development of TG2 mice (11, 30). But in response to certain PLA2 activation mediates effects of CXCR2 ligands on pathologic stressors, including GRO␣ and IL-8, TG2 in articular chondrocytes. chondrocytes may function to modulate cartilage repair and Although IL-8 and GRO␣ induced TG2-dependent transamida- stability. tion activity in chondrocytes, the relative contribution of transami- Cytosolic and externalized TG2 both could regulate chondro- dation events catalyzed by TG2 to chemokine-induced chondrocyte differentiation in response to CXCR2 ligands (28, 56, 57). cyte hypertrophy is not known. TG2 is a dual function enzyme (TG Significantly, TG2-induced matrix cross-linking consistent with and GTPase/ATPase), a unique property of TG2 among human externalization of active TG2 has been detected in OA cartilage TGs (11, 28, 57). It is possible that the capacities of TG2 to reg-

(58). Cell surface TG2 cannot only promote fibronectin cross-link- ulate signal transduction and Pi production through GTPase and ing via transamidation, but also directly modulate cell attachment ATPase activities (11) modulate CXCR2 ligand-induced chondro- to fibronectin via the TG2 N-terminal fibronectin binding domain cyte hypertrophy. in TG2 (59). Interestingly, TG2 regulates adhesion and migration Our results add to a growing body of evidence that chemokine in mononuclear phagocytes and fibroblasts (60, 61). By modulat- ligands of CXCR2, such as IL-8 and GRO␣, function beyond re- ing cell adhesion, fibronectin binding, and matrix assembly (28, cruitment and activation of leukocytes in sites of tissue injury and 62), TG2 has the potential to regulate matrix signals provided to inflammation (67Ð69). For example, such chemokines exert direct chondrocytes, thereby modulating chondrocyte differentiation and effects on MMP expression, fibroblast growth, endothelial cell 4414 CHEMOKINE-INDUCED CHONDROCYTE HYPERTROPHY

15. Kanbe, K., K. Takagishi, and Q. Chen. 2002. Stimulation of matrix metallopro- tease 3 release from human chondrocytes by the interaction of stromal cell-derived factor 1 and CXC chemokine receptor 4. Arthritis Rheum. 46:130. 16. Borzi, R. M., I. Mazzetti, L. Cattini, M. Uguccioni, M. Baggiolini, and A. Facchini. 2000. Human chondrocytes express functional chemokine receptors and release matrix-degrading enzymes in response to C-X-C and C-C chemokines. Arthritis Rheum. 43:1734. 17. Silvestri, T., R. Meliconi, L. Pulsatelli, P. Dolzani, F. Zizzi, L. Frizziero, R. M. Borzi, and A. Facchini. 2003. Down-modulation of chemokine receptor cartilage expression in inflammatory arthritis. Rheumatology 42:14. 18. Olson, T. S., and K. Ley. 2002. Chemokines and chemokine receptors in leuko- FIGURE 9. Model for IL-8- and GRO␣-induced articular chondrocyte cyte trafficking. Am. J. Physiol. Regul. Integr. Comp. Physiol. 283:R7. hypertrophy. In this model, up-regulated IL-8 and GRO␣ expression in OA 19. Borzi, R. M., I. Mazzetti, G. Magagnoli, S. Paoletti, M. Uguccioni, R. Gatti, cartilage stimulates chondrocyte hypertrophic differentiation in the disease. G. Orlandini, L. Cattini, and A. Facchini. 2002. Growth-related oncogene ␣ in- Based on the data obtained in this study, the chemokine-induced chondro- duction of apoptosis in osteoarthritis chondrocytes. Arthritis Rheum. 46:3201. cyte hypertrophic differentiation is transduced through p38 MAPK-medi- 20. Takami, M., V. Terry, and L. Petruzzelli. 2002. Signaling pathways involved in IL-8-dependent activation of adhesion through Mac-1. J. Immunol. 168:4559. ated stimulation of TG2 activity. 21. Nick, J. A., S. K. Young, P. G. Arndt, J. G. Lieber, B. T. Suratt, K. R. Poch, N. J. Avdi, K. C. Malcolm, C. Taube, P. M. Henson, and G. S. Worthen. 2002. Selective suppression of neutrophil accumulation in ongoing pulmonary inflammation by systemic inhibition of p38 mitogen-activated protein kinase. J. Immu- growth, and angiogenesis, and on tumorigenesis (67Ð69). Timing nol. 169:5260. 22. Zhen, X., L. Wei, Q. Wu, Y. Zhang, and Q. Chen. 2001. Mitogen-activated of CXCR2 ligand exposure to chondrocytes, levels of CXCR2 ex- protein kinase p38 mediates regulation of chondrocyte differentiation by para- pression, and potential combinatorial effects of CXCR2 ligands thyroid hormone. J. Biol. Chem. 276:4879. Downloaded from with those of other cytokines all have the potential to modulate 23. Pelletier, J. P., J. C. Fernandes, D. V. Jovanovic, P. Reboul, and J. Martel-Pelletier. 2001. Chondrocyte death in experimental osteoarthritis is me- chondrocyte activation in vivo. Nevertheless, our identification of diated by MEK 1/2 and p38 pathways: role of cyclooxygenase-2 and inducible the capacity of CXCR2 ligands to stimulate hypertrophy and cal- nitric oxide synthase. J. Rheumatol. 28:2509. cification in cultured articular chondrocytes, and the major contri- 24. Poole, A. R., Y. Matsui, A. Hinek, and E. R. Lee. 1989. Cartilage macromole- cules and the calcification of cartilage matrix. Anat. Rec. 224:167. bution of TG2 to these events, reveals a novel mechanism by 25. Von der Mark, K., T. Kirsch, A. Nerlich, A. Kuss, G. Weseloh, K. Gluckert, and which chronic inflammation can alter the differentiation and func- H. Stoss. 1992. Type X collagen synthesis in human osteoarthritic cartilage: http://www.jimmunol.org/ tion of resident cells of mesenchymal origin in arthritis. CXCR2 indication of chondrocyte hypertrophy. Arthritis Rheum. 35:806. 26. Vignon, E., M. Arlot, D. Hartmann, B. Moyen, and G. Ville. 1983. Hypertrophic and TG2 may be novel sites for therapeutic intervention in OA. repair of articular cartilage in experimental osteoarthrosis. Ann. Rheum. Dis. 42:82. Acknowledgments 27. Ishikawa, K., I. Masuda, T. Ohira, and M. Yokoyama. 1989. A histological study of calcium pyrophosphate dihydrate crystal-deposition disease. J. Bone Jt. Surg. We gratefully acknowledge the assistance of Deborah van Etten in man- Am. 71:875. Ϫ/Ϫ aging the TG2 mouse-breeding colony. 28. Lorand, L., and R. M. Graham. 2003. Transglutaminases: cross-linking enzymes with pleiotropic functions. Nat. Rev. Mol. Cell Bio. l. 4:140. References 29. Johnson, K., S. Vaingankar, Y. Chen, A. Moffa, M. B. Goldring, K. Sano, P. Jin-Hua, A. Sali, J. Goding, and R. Terkeltaub. 1999. Differential mechanisms by guest on September 25, 2021 1. Pelletier, J. P., J. Martel-Pelletier, and S. B. Abramson. 2001. Osteoarthritis, an of PPi production by plasma cell membrane glycoprotein-1 (PC-1) and B10 in inflammatory disease: potential implication for the selection of new therapeutic chondrocytes. Arthritis Rheum. 42:1986. targets. Arthritis Rheum. 44:1237. 30. Nanda, N., S. E. Iismaa, W. A. Owens, A. Husain, F. Mackay, and R. M. Graham. 2. Goldring, M. B. 2001. Anticytokine therapy for osteoarthritis. Expert Opin. Biol. 2001. Targeted inactivation of Gh/tissue transglutaminase II. J. Biol. Chem. Ther. 1:817. 276:20673. 3. Frisbie, D. D., S. C. Ghivizzani, P. D. Robbins, C. H. Evans, and C. W. McIlwraith. 2002. Treatment of experimental equine osteoarthritis by in 31. Huang, R., M. Rosenbach, R. Vaughn, D. Provvedini, N. Rebbe, S. Hickman, vivo delivery of the equine interleukin-1 receptor antagonist gene. Gene Ther. J. Goding, and R. Terkeltaub. 1994. Expression of the murine plasma cell nu- 9:12. cleotide pyrophosphohydrolase PC-1 is shared by human liver, bone and cartilage 4. Attur, M. G., M. N. Dave, M. Y. Leung, C. Cipolletta, M. Meseck, S. L. Woo, cells: regulation of PC-1 expression in osteosarcoma cells by transforming ␤ and A. R. Amin. 2002. Functional genomic analysis of type II IL-1␤ decoy growth factor- . J. Clin. Invest. 94:560. receptor: potential for gene therapy in human arthritis and inflammation. J. Im- 32. Kna¬uper, V., C. Lo«pez-Otin, B. Smith, G. Knight, and G. Murphy. 1996. Bio- munol. 168:2001. chemical characterization of human collagenase-3. J. Biol. Chem. 271:1544. 5. Pratta, M. A., P. A. Scherle, G. Yang, R. Q. Liu, and R. C. Newton. 2003. 33. Slaughter, T. F., K. E. Achyuthan, T. S. Lai, and C. S. Greenberg. 1992. A Induction of aggrecanase 1 (ADAM-TS4) by interleukin-1 occurs through acti- microtiter plate transglutaminase assay utilizing 5-(biotinamido)pentylamine as vation of constitutively produced protein. Arthritis Rheum. 48:119. substrate. Anal. Biochem. 205:166. 6. Studer, R., D. Jaffurs, M. Stefanovic-Racic, P. D. Robbins, and C. H. Evans. 34. Lotz, M., R. Terkeltaub, and P. M. Villiger. 1992. Cartilage and joint inflamma- 1999. Nitric oxide in osteoarthritis. Osteoarthritis Cartilage 7:377. tion: regulation of IL-8 expression by human articular chondrocytes. J. Immunol. 7. Loeser, R. F., C. S. Carlson, M. Del Carlo, and A. Cole. 2002. Detection of 148:466. nitrotyrosine in aging and osteoarthritic cartilage: correlation of oxidative damage 35. Recklies, A. D., and E. E. Golds. 1992. Induction of synthesis and release of with the presence of interleukin-1␤ and with chondrocyte resistance to insulin- interleukin-8 from human articular chondrocytes and cartilage explants. Arthritis like growth factor 1. Arthritis Rheum. 46:2349. Rheum. 35:1510. 8. Lotz, M. 1999. The role of nitric oxide in articular cartilage damage. Rheum. Dis. 36. Pulsatelli, L., P. Dolzani, A. Piacentini, T. Silvestri, R. Ruggeri, G. Gualtieri, Clin. N. Am. 25:269. R. Meliconi, and A. Facchini. 1999. Chemokine production by human chondro- 9. Gu, Z., M. Kaul, B. Yan, S. J. Kridel, J. Cui, A. Strongin, J. W. Smith, cytes. J. Rheumatol. 26:1992. R. C. Liddington, and S. A. Lipton. 2002. S-nitrosylation of matrix metallopro- 37. Wu, C. W., E. V. Tchetina, F. Mwale, K. Hasty, I. Pidoux, A. Reiner, J. Chen, teinases: signaling pathway to neuronal cell death. Science 297:1186. H. E. Van Wart, and A. R. Poole. 2002. Proteolysis involving matrix metallo- 10. Johnson, K., S. Hashimoto, M. Lotz, K. Pritzker, and R. Terkeltaub. 2001. In- proteinase 13 (collagenase-3) is required for chondrocyte differentiation that is terleukin-1 induces pro-mineralizing activity of cartilage tissue transglutaminase associated with matrix mineralization. J. Bone Miner. Res. 17:639. and factor XIIIa. Am. J. Pathol. 159:149. 11. Johnson, K. A., D. Van Etten, N. Nanda, R. M. Graham, and R. A. Terkeltaub. 38. D’Angelo, M., P. C. Billings, M. Pacifici, P. S. Leboy, and T. Kirsch. 2001. 2003. Distinct transglutaminaseII/TG2-independent and TG2-dependent path- Authentic matrix vesicles contain active metalloproteases (MMP). a role for ma- ␤ ways mediate articular chondrocyte hypertrophy. J. Biol. Chem. 278:18824. trix vesicle-associated MMP-13 in activation of transforming growth factor- . 12. Terkeltaub, R. A. 2002. What does cartilage calcification tell us about osteoar- J. Biol. Chem. 276:11347. thritis?. J. Rheumatol. 29:411. 39. Jones, S. A., M. Wolf, S. Qin, C. R. Mackay, and M. Baggiolini. 1996. Different 13. Alaaeddine, N., T. Olee, S. Hashimoto, L. Creighton-Achermann, and M. Lotz. functions for the interleukin 8 receptors (IL-8R) of human neutrophil leukocytes: 2001. Production of the chemokine RANTES by articular chondrocytes and role NADPH oxidase and phospholipase D are activated through IL-8R1 but not IL- in cartilage degradation. Arthritis Rheum. 44:1633. 8R2. Proc. Natl. Acad. Sci. USA 93:6682. 14. Borzi, R. M., I. Mazzetti, S. Macor, T. Silvestri, A. Bassi, L. Cattini, and 40. L’Heureux, G. P., S. Bourgoin, N. Jean, S. R. McColl, and P. H. Naccache. 1995. A. Facchini. 1999. Flow cytometric analysis of intracellular chemokines in chon- Diverging signal transduction pathways activated by interleukin-8 and related drocytes in vivo: constitutive expression and enhancement in osteoarthritis and chemokines in human neutrophils: interleukin-8, but not NAP-2 or GRO␣, stim- rheumatoid arthritis. FEBS Lett. 455:238. ulates phospholipase D activity. Blood 85:522. The Journal of Immunology 4415

41. Terkeltaub, R., S. Baird, P. Sears, R. Santiago, and W. Boisvert. 1998. The uct ␣: a chemotactic cytokine for neutrophils in rheumatoid arthritis. J. Immunol. murine homolog of the interleukin-8 receptor CXCR-2 is essential for the occur- 155:3660. rence of neutrophilic inflammation in the air pouch model of acute urate crystal- 55. Kirsch, T., B. Swoboda, and H. Nah. 2000. Activation of annexin II and V induced gouty synovitis. Arthritis Rheum. 41:900. expression, terminal differentiation, mineralization and apoptosis in human os- 42. Josse, C., J. R. Boelaert, M. Best-Belpomme, and J. Piette. 2001. Importance of teoarthritic cartilage. Osteoarthritis Cartilage 8:294. post-transcriptional regulation of chemokine genes by oxidative stress. Biochem. 56. Lai, T. S., A. Hausladen, T. F. Slaughter, J. P. Eu, J. S. Stamler, and J. 360:32. C. S. Greenberg. 2001. Calcium regulates S-nitrosylation, denitrosylation, and 43. Honorati, M. C., M. Bovara, L. Cattini, A. Piacentini, and A. Facchini. 2002. activity of tissue transglutaminase. Biochemistry 40:4904. Contribution of interleukin 17 to human cartilage degradation and synovial in- 57. Lai, T. S., T. F. Slaughter, K. A. Peoples, J. M. Hettasch, and C. S. Greenberg. flammation in osteoarthritis. Osteoarthritis Cartilage 10:799. 1998. Regulation of human tissue transglutaminase function by magnesium-nu- 44. Thomas, B., S. Thirion, L. Humbert, L. Tan, M. B. Goldring, G. Bereziat, and cleotide complexes: identification of distinct binding sites for Mg-GTP and Mg- F. Berenbaum. 2002. Differentiation regulates interleukin-1␤-induced cyclo-ox- ATP. J. Biol. Chem. 273:1776. ygenase-2 in human articular chondrocytes: role of p38 mitogen-activated protein 58. Summey, B. T., Jr., R. D. Graff, T. S. Lai, C. S. Greenberg, and G. M. Lee. 2002. kinase. Biochem. J. 362:367. Tissue transglutaminase localization and activity regulation in the extracellular 45. Kashiwagi, M., M. Tortorella, H. Nagase, and K. Brew. 2001. TIMP-3 is a potent matrix of articular cartilage. J. Orthop. Res. 20:76. inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5). 59. Fesus, L., and M. Piacentini. 2002. Transglutaminase 2: an enigmatic enzyme J. Biol. Chem. 276:12501. with diverse functions. Trends Biochem. Sci. 27:534. 46. Mwale, F., C. Billinghurst, W. Wu, M. Alini, C. Webber, A. Reiner, M. Ionescu, 60. Akimov, S. S., and A. M. Belkin. 2001. Cell surface tissue transglutaminase is J. Poole, and A. R. Poole. 2000. Selective assembly and remodelling of collagens involved in adhesion and migration of monocytic cells on fibronectin. Blood II and IX associated with expression of the chondrocyte hypertrophic phenotype. 98:1567. Dev. Dyn. 218:648. 61. Balklava, Z., E. Verderio, R. Collighan, S. Gross, J. Adams, and M. Griffin. 2002. 47. Alini, M., D. Carey, S. Hirata, M. D. Grynpas, I. Pidoux, and A. R. Poole. 1994. Analysis of tissue transglutaminase function in the migration of Swiss 3T3 fi- Cellular and matrix changes before and at the time of calcification in the growth broblasts: the active-state conformation of the enzyme does not affect cell mo- plate studied in vitro: arrest of type X collagen synthesis and net loss of collagen tility but is important for its secretion. J. Biol. Chem. 277:16567. when calcification is initiated. J. Bone Miner. Res. 9:1077.

62. Akimov, S. S., and A. M. Belkin. 2001. Cell-surface transglutaminase promotes Downloaded from 48. Coe, M. R., T. A. Summers, S. J. Parsons, A. L. Boskey, and G. Balian. 1992. fibronectin assembly via interaction with the gelatin-binding domain of fibronec- Matrix mineralization in hypertrophic chondrocyte cultures: ␤ glycerophosphate tin: a role in TGF␤-dependent matrix deposition. J. Cell Sci. 114:2989. increases type X collagen messenger RNA and the specific activity of pp60c-src 63. Forsyth, C. B., J. Pulai, and R. F. Loeser. 2002. Fibronectin fragments and block- kinase. Bone Miner. 18:91. ␣ ␤ ␣ ␤ 49. Beck, G. R., Jr., B. Zerler, and E. Moran. 2000. Phosphate is a specific signal for ing antibodies to 2 1 and 5 1 integrins stimulate mitogen-activated protein induction of osteopontin gene expression. Proc. Natl. Acad. Sci. USA 97:8352. kinase signaling and increase collagenase 3 (matrix metalloproteinase 13) pro- 50. Gerstenfeld, L. C., and F. D. Shapiro. 1996. Expression of bone-specific genes by duction by human articular chondrocytes. Arthritis Rheum. 46:2368. hypertrophic chondrocytes: implication of the complex functions of the hyper- 64. Gemba, T., J. Valbracht, S. Alsalameh, and M. Lotz. 2002. Focal adhesion kinase and mitogen-activated protein kinases are involved in chondrocyte activation by

trophic chondrocyte during endochondral bone development. J. Cell. Biochem. http://www.jimmunol.org/ 62:1. the 29-kDa amino-terminal fibronectin fragment. J. Biol. Chem. 277:907. ␣ ␤ 51. Hessle, L., K. A. Johnson, H. C. Anderson, S. Narisawa, A. Sali, J. W. Goding, 65. Pulai, J. I., M. Del Carlo, Jr., and R. F. Loeser. 2002. The 5 1 integrin provides R. Terkeltaub, and J. L. Millan. 2002. Tissue-nonspecific alkaline phosphatase matrix survival signals for normal and osteoarthritic human articular chondro- and plasma cell membrane glycoprotein-1 are central antagonistic regulators of cytes in vitro. Arthritis Rheum. 46:1528. bone mineralization. Proc. Natl. Acad. Sci. USA 99:9445. 66. Sohn, J., T. I. Kim, Y. H. Yoon, J. Y. Kim, and S. Y. Kim. 2003. Novel trans- 52. Kaneko, S., T. Satoh, J. Chiba, C. Ju, K. Inoue, and J. Kagawa. 2000. Interleu- glutaminase inhibitors reverse the inflammation of allergic conjunctivitis. J. Clin. kin-6 and interleukin-8 levels in serum and synovial fluid of patients with osteo- Invest. 111:121. arthritis. Cytokines Cell. Mol. Ther. 6:71. 67. Strieter, R. M. 2001. Chemokines: not just leukocyte chemoattractants in the 53. Terkeltaub, R., C. Zachariae, D. Santoro, J. Martin, P. Peveri, and promotion of cancer. Nat. Immun. 2:285. K. Matsushima. 1991. Monocyte-derived neutrophil chemotactic factor/interleu- 68. Li, A., S. Dubey, M. L. Varney, and R. K. Singh. 2002. Interleukin-8-induced kin-8 is a potential mediator of crystal-induced inflammation. Arthritis Rheum. proliferation, survival, and MMP production in CXCR1 and CXCR2 expressing

34:894. human umbilical vein endothelial cells. Microvasc. Res. 64:476. by guest on September 25, 2021 54. Koch, A. E., S. L. Kunkel, M. R. Shah, S. Hosaka, M. M. Halloran, G. K. Haines, 69. Wang, D., J. Sai, and A. Richmond. 2003. Cell surface heparan sulfate partici- M. D. Burdick, R. M. Pope, and R. M. Strieter. 1995. Growth-related gene prod- pates in CXCL1-induced signaling. Biochemistry 42:1071.