U.I. Chung

p63 Regulates Chondrocyte Survival in Articular Cartilage

Transcription factor p63, of the p53 family, regulates cell proliferation, survival, and apoptosis in various cells and tissues. This study was undertaken to examine the expression and roles of p63 transcript variants in the mouse growth plate and articular chondrocytes.

71 NOTCH/RBPJ/HES1 SIGNAL IN CHONDROCYTES MODULATES TERMINAL STAGE OF ENDOCHONDRAL OSSIFICATION DURING SKELETAL GROWTH AND OSTEOARTHRITIS DEVELOPMENT

accelerated synovial joint formation at E14.5, although the joint structure appeared similar to controls at E18.5. By 2-weeks of age the mutant mice exhibit delays in the formation of the secondary center of ossification as assessed by histology and ISH, although the surface articular cartilage appears mostly normal. Micro-CT analyses at 2-, 4-, and 8-months of age demonstrated an early increase in mineralization in both the subchondral and trabecular bone regions that was followed by a progressive loss of bone. Additionally, mutant mice developed osteophytes and excessive mineralization of the menisci as compared to controls. Histology, histomorphometric, ISH, and IHC analyses of 2-, 4-, and 8-month old knee joint sections revealed that mutant mice show early and progressive signs of OA. Analyses of alcian blue stained sections from mutant mice displayed a progressive degeneration and fibrosis of the articular cartilage and menisci, subchondral bone sclerosis, osteophyte formation, and synovial expansion. By 8-months of age the articular cartilage of mutant mice exhibit severe cellular and molecular changes consistent with OA, such as fibrosis and fibrillation, near complete loss of Prg4 expression, decreased Col2a1 and enhanced Col10a1 and Mmp13 immunoreactivity, and enhanced cell death. Polarized light microscopy on all mutant and control adult knee sections further demonstrated a loss of normal collagen fibrils within the subchondral bone and articular cartilage extracellular matrix of mutant mice. Conclusions: Our in vivo data demonstrate for the first time that RBPjkdependent Notch signaling is critical in maintaining post-natal articular cartilage and joint maintenance, but is dispensable for the formation of synovial joints during embryonic development.

072 AKT1 IN CHONDROCYTES CONTROLS CARTILAGE CALCIFICATION DURING SKELETAL GROWTH AND OSTEOPHYTE FORMATION IN OSTEOARTHRITIS

INTRODUCTION Endochondral ossification plays crucial roles in skeletal growth and osteoarthritis progression. Since the phosphoinositide-dependent serinethreonine protein kinase Akt is known to be a pivotal signaling molecule for several factors regulating cartilage metabolism, this study examined the possible involvement of Akt in the endochondral ossification process under physiological and pathological conditions.

158 AKT1 CONTRIBUTES TO CHONDROCYTE CALCIFICATION DURING ENDOCHONDRAL OSSIFICATION

Purpose: Endochondral ossification plays crucial roles in skeletal disorders like osteoarthritis as well as in skeletal growth. Since the phosphoinositide-dependent serine-threonine protein kinase Akt has been proposed to be a pivotal signaling molecule for several factors regulating cartilage metabolism, this study examined the possible involvement of Akt in the processes of endochondral ossification. Methods: Expressions of the three isoforms of Akt (Akt1, 2 & 3) were examined by real-time RT-PCR in chondrocytes derived from ribs of neonatal mice. To know the physiological function of Akt1, we compared the skeletal phenotypes of homozygous Akt1-deficient (Akt1−/−) mice with those of the wild-type littermates by plain radiograph, 3D-mCT, and histological analyses including toluidine blue, von Kossa stainings, type X collagen (COL10) immunostaining, and BrdU labeling. For gainand lossof-function analyses, we established stable lines of mouse chondrogenic ATDC5 cells with retroviral overexpression of constitutively active Akt1 (CA-Akt1) and small interfering RNA (siRNA) of Akt1, respectively, and compared the functions with respective empty-vector controls. Cell proliferation was assessed by CCK-8 assay. The differentiation was determined by Alcian blue and alkaline phosphatase (ALP) stainings under the stimulation of insulin. The hypertrophic differentiation was determined by luciferase assays using ATDC5 cells transfected with a luciferase-reporter gene construct containing a 4.5-kb fragment of the COL10 promoter. Calcification was assessed by Alizarin red staining under the stimulation of insulin and inorganic phosphate ion (Pi). Expressions of inorganic pyrophosphate (PPi)-related factors were determined by semi-quantitative RT-PCR. Results: Since Akt1 was most highly expressed in chondrocytes among the three isoforms, we generated Akt1−/− mice and examined the skeletal phenotype. The Akt1−/− mice exhibited dwarfism with about 20% shorter limbs and trunks than the wild-type littermates during observation periods from embryos to 12 weeks postnatal. The entire width and columnar structure of the growth plate cartilage were normal, and BrdU-positive proliferative zone and COL10-positive hypertrophic zones were also unaffected by the Akt1 deficiency. However, the width of the calcified layer and the number of calcified chondrocytes determined by the von Kossa staining were significantly decreased at the bottom of the Akt1−/− growth plate. In the ATDC5 cell culture, neither the CA-Akt1 nor the Akt1 siRNA overexpression altered cell proliferation, differentiation, or the COL10 promoter activity. Contrarily, calcification of ATDC5 cells cultured in the presence of insulin/Pi was significantly enhanced by the CA-Akt1 and suppressed by the Akt1 si-RNA. These in vivo and in vitro findings demonstrate that Akt1 is essential for chondrocyte calcification without affecting the prior processes of endochondral ossification. As PPi has been known to be a crucial regulator of chondrocyte calcification, we finally examined expressions of principal modulators of PPi: ANK, NPP1 and Pit1. Although all increased during chondrocyte calcification under the insulin/Pi stimulation, none of them was affected by the CA-Akt1 or the Akt1 si-RNA. Conclusions: Akt1 controls skeletal growth by maintaining chondrocyte calcification during endochondral ossification, without affecting the proliferation or differentiation of chondrocytes. Further understanding of the molecular network related to Akt1, probably independent of the putative PPi pathway, will quite probably lead to a breakthrough for the treatment of skeletal disorders like osteoarthritis.

A4 HYPOXIA-INDUCIBLE FACTOR 2A (HIF2A) CONTROLS SEQUENTIAL STEPS IN THE LATE STAGE OF ENDOCHONDRAL OSSIFICATION

Purpose: Demonstrating efficacy of potential disease-modifying OA drugs (DMOAD) requires long treatment times and a large number of subjects, resulting in expensive clinical trials. An important risk of such studies is the absence of beneficial effect of the drug. Accurate prediction of the doses that would result in cartilage protection and joint preservation would help minimize this risk. Although dose may be extrapolated from pre-clinical studies, the lack of approved DMOADs obscures the clinical predictive capability of the OA animal models. Here we describe a translational pharmacology strategy using TIINE (a type II collagen neoepitope assay) to evaluate MMP-13 inhibitors for the treatment of OA Methods: The MMP-13 inhibitors belong to the class of compounds that bind the S1 pocket of the enzyme providing good potency (Ki 2−6 nM) and selectivity (>4000 fold over 15 other MMPs-, ADAMTS-4 and -5, and TACE). TIINE was measured using a sandwich immunoassay and an LCMS/MS assay specific for the 45-mer peptide. Rat medial meniscal tear (MMT) and dog partial medial meniscectomy (pMx) were performed at Bolder BioPath as previously described. All studies were approved by the Institutional Animal Care and Use Committees. Results: Studies in vitro showed that TIINE was generated by addition of active MMP-13 to de-vitalized cartilage in a timeand concentrationdependent manner, which was inhibited by MMP-13 inhibitors. Because TIINE was also generated by addition of other MMPs, this biomarker is not specific for MMP-13 activity. In vivo, TIINE was upregulated about two fold in the urine from OA patients in contrast to age-matched controls. To understand the role of MMP-13 in the generation of such OA-associated TIINE elevation, TIINE modulation was evaluated in animal models of OA, and in response to MMP-13 selective inhibitors. In the rat MMT, TIINE was elevated in the synovial fluid (SF) from the operated knee, but not the contralateral knee, and it was inhibited in a timeand dose-dependent manner following oral administration of an MMP-13 inhibitor to a maximum of ~50%. Similar TIINE inhibition in the SF and urine was observed at the end of a 4-wk study (BID, prophylactic), and maximal inhibition correlated with cartilage protection as evaluated by histology. All animals and human subjects evaluated to date excrete some amount of TIINE in their urine (which is age-dependent). Single dose oral administration of MMP-13 inhibitor to skeletally-mature beagle dogs (naive) inhibited baseline urinary TIINE in a timeand dose-dependent manner up to ~50%. In the 4-wk dog pMx model, MMP-13 inhibitors (BID, prophylactic) protected cartilage degeneration at exposures that inhibited TIINE ~45% at 2, 3 and 4 wks. These results suggest that doses that result in ~45% inhibition of TIINE in early clinical trials may protect cartilage degradation and joint structure in Phase 2−3 studies. In previous clinical studies with broader spectrum MMP inhibitors, urinary TIINE was found to be inhibited in both human healthy volunteers and OA patients. Conclusions: The proposed translational pharmacology strategy includes: (1) determine drug plasma and/or SF concentration required for maximum TIINE inhibition in urine of naive and/or OA animals, (2) establish correlation between TIINE inhibition and cartilage protection in animal models of OA, (3) evaluate TIINE inhibition in healthy volunteers and/or OA patients in early clinical trials, and (4) in POC studies, use the doses that resulted in sufficient inhibition of TIINE from the FIH studies and that was associated with cartilage protection in pre-clinical studies.

A1 A NOVEL SMALL THIENOINDAZOLE-DERIVATIVE COMPOUND PROMOTES CHONDROGENIC DIFFERENTIATION WITHOUT INDUCING HYPERTROPHY THROUGH PRODUCTION OF RUNX1

Purpose: Aiming at regeneration of permanent cartilage like joint cartilage, this study screened natural and synthetic compound libraries to discover a novel compound inducing chondrogenic differentiation without hypertrophy. We further investigated the underlying molecular mechanism. Methods: As an efficient monitoring system for chondrogenic differentiation, we established stable lines of mouse chondrogenic ATDC5 cells expressing green fluorescent protein under the control of type II collagen promoter fused with four repeats of a Sox9 enhancer (COL2GFP). Chondrogenic differentiation was assessed by real-time RT-PCR for COL2, aggrecan, chondromodulin-1 and COL10, toluidine blue and Alcian blue stainings, and quantitative GAG assay in cultures of mouse embryonic stem cells or immature mesenchymal C3H10T1/2 cells. The downstream molecules were screened by a microarray analysis using C3H10T1/2 cells. The COL2 promoter activity was determined using HuH-7 cells transfected with a luciferase-reporter gene construct containing the COL2 promoter above, and the specific binding to the identified region was verified by electrophoretic mobility shift assay (EMSA) and chromatin immunoprecipitation (ChIP). For functional analyses, we performed adenoviral overexpression of the gene or the small interfering RNA in C3H10T1/2 cells, and compared the chondrogenic differentiation as described above with respective empty-vector controls. Molecular interactions were examined by immunoprecipitation, mammalian two-hybrid system, and immunohistochemistry in the mouse growth plate. Finally, cartilage formation in full-thickness defects of mouse knee cartilage was histologically evaluated after transplantation of cell-sheets of mouse rib chondrocytes with or without TM treatment. Results: The COL2-GFP system showed that a small thienoindazolederivative compound T-198946 (TM) most strongly induced the GFP fluorescence as early as after 48 h of treatment. TM was confirmed to enhance chondrogenic differentiation but inhibit the further hypertrophic differentiation in the cultures of precursor cells. The microarray screening revealed that Runx1 was most strongly induced by TM among 581 up-regulated genes including Sox5 and Sox6. Deletion, mutagenesis, and tandem-repeat analyses of the luciferase assay identified the core responsive element of Runx1 in the COL2 promoter to be between the −293 and −288 bp region containing a putative Runx-binding motif. EMSA and ChIP assays confirmed the specific binding of Runx1 to this region. Although chondrogenic differentiation of C3H10T1/2 cells was little enhanced by the Runx1 overexpression alone, it was much enhanced by co-transfection with Sox5, 6, and 9 (the Sox trio), without inducing the hypertrophy, as true of the effect of TM treatment. Gene-silencing of Runx1, Sox5/6, or Sox9 suppressed the TM effect on chondrogenic differentiation. In fact, Runx1 and the Sox trio were co-localized in the proliferative and pre-hypertrophic chondrocytes of the mouse growth plate, and their physical interaction was confirmed by immunoprecipitation and two-hybrid analysis. Finally, cell-sheets of TM-treated chondrocytes filled the defects with cartilaginous tissue, while the control cell-sheets did not. Conclusions: A novel small compound TM promotes chondrogenic differentiation without inducing hypertrophy, through production of Runx1 that cooperatively functions with the Sox trio. TM will herald a new era of regenerative medicine of permanent cartilage, thus realizing an epochal treatment of osteoarthritis. A2 IMPACT OF GENDER ON QUALITY-ADJUSTED LIFE EXPECTANCY LOSSES DUE TO KNEE OSTEOARTHRITIS IN THE US ELDERLY POPULATION