Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on March 3, 2005
Human Molecular Genetics 2005 14(8):1009-1017; doi:10.1093/hmg/ddi093

This Article Abstract FREE Full Text (PDF) Supplementary Material All Versions of this Article: 14/8/1009    most recent ddi093v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (26) Request Permissions Google Scholar Articles by Mototani, H. Articles by Ikegawa, S. Search for Related Content PubMed PubMed Citation Articles by Mototani, H. Articles by Ikegawa, S. Social Bookmarking          What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

A functional single nucleotide polymorphism in the core promoter region of CALM1 is associated with hip osteoarthritis in Japanese

Hideyuki Mototani^1,3, Akihiko Mabuchi¹, Susumu Saito^4,, Mikihiro Fujioka⁵, Aritoshi Iida², Yoshio Takatori⁶, Akihiro Kotani⁷, Toshikazu Kubo⁵, Kozo Nakamura⁶, Akihiro Sekine⁸, Yoshinori Murakami¹⁰, Tatsuhiko Tsunoda⁹, Kohei Notoya³, Yusuke Nakamura^2,8 and Shiro Ikegawa^1,*

¹Laboratory for Bone and Joint Diseases and ²Laboratory for SNP Analysis, Research Groups of Personalized Medicine, SNP Research Center, The Institute of Physical and Chemical Research (RIKEN), 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan, ³Pharmaceutical Research Laboratories I, Pharmaceutical Research Division, Takeda Pharmaceutical Company, Ltd, Osaka 532-8686, Japan, ⁴Department of Orthopaedic Surgery, Sumitomo Hospital, Nakanoshima 5-3-20, Kita-ku, Osaka 530-0005, Japan, ⁵Department of Orthopaedic Surgery, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan, ⁶Department of Orthopaedic Surgery, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan, ⁷Department of Orthopaedic Surgery, Kyorin University, School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, Japan, ⁸Laboratory for Genotyping and ⁹Laboratory for Medical Informatics, SNP Research Center, RIKEN, 1-7-22 Suehirocho, Tsurumi-ku, Yokohama 230-0045, Japan and ¹⁰Tumor Suppression and the Functional Genomics Project, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan

^* To whom correspondence should be addressed. Tel: +81 354495393; Fax: +81 354495393; Email: sikegawa{at}ims.u-tokyo.ac.jp

Received December 27, 2004; Revised February 14, 2005; Accepted February 22, 2005

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Osteoarthritis (OA), a common skeletal disease, is a leading cause of disability among the elderly populations. OA is characterized by gradual loss of articular cartilage, but the etiology and pathogenesis of OA are largely unknown. Epidemiological and genetic studies have demonstrated that genetic factors play an important role in OA. To identify susceptibility genes for OA, we performed a large-scale, case–control association study using gene-based single nucleotide polymorphisms (SNPs). In two independent case–control populations, we found significant association (P=9.8x10^–7) between hip OA and a SNP (IVS3–293C>T) located in intron 3 of the calmodulin (CaM) 1 gene (CALM1). CALM1 was expressed in cultured chondrocytes and articular cartilage, and its expression was increased in OA. Subsequent linkage-disequilibrium mapping identified five SNPs showing significant association equivalent to IVS3–293C>T. One of these (–16C>T) is located in the core promoter region of CALM1. Functional analyses indicate that the susceptibility –16T allele decreases CALM1 transcription in vitro and in vivo. Inhibition of CaM in chondrogenic cells reduced the expression of the major cartilage matrix genes Col2a1 and Agc1. These results suggest that the transcriptional level of CALM1 is associated with susceptibility for hip OA through modulation of chondrogenic activity. Our findings reveal the CALM1-mediated signaling pathway in chondrocytes as a novel potential target for treatment of OA.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Osteoarthritis (OA; MIM165720) is a common disease and a leading cause of disability among the elderly populations. OA causes pain, deformity and loss of joint function, leading to reduced activity in daily living and decreased quality of life. OA affects various joints, most commonly spine, hand, knee and hip, with clinical features that are unique to each joint. About 20 million patients suffer from OA in the United States (1), and in Japan, the prevalence of hip OA has been reported as 3.5% (women) and 1.4% (men) (2). The cost of OA to society is estimated to exceed $15 billion in the United States, an amount roughly three times that of rheumatoid arthritis (3). Thus, OA is a serious medical and economical issue, particularly in the aging populations of developed countries.

OA is a degenerative disease characterized by gradual loss of articular cartilage. During the course of life, articular cartilage is lost gradually because of excessive mechanical stress and/or injuries to the joint. Decompensation of the capacity for cartilage repair by chondrocytes has been suggested as potential pathogenic factor in OA (4). In addition, aberration of cartilage matrix metabolism ultimately results in failure of its ability to respond to and withstand mechanical stress on the joint and hence contributing to OA.

The etiology and pathogenesis of OA remain largely undetermined. Epidemiological studies have demonstrated that OA is a genetic disease, with its major component transmitted as a complex, multifactorial trait (5). Many genetic studies have sought to identify susceptibility genes for OA, including candidate gene association studies and genome-wide linkage studies of hand, knee and hip OA (6–8). These efforts have resulted in the identification of several susceptibility loci for each type of OA, confirming that OA has a genetic component. However, the genetic components of OA remain largely uncharacterized because of limitations in current genetic approaches. Candidate gene association studies are severely limited by the requirement for a priori knowledge of ‘candidacy’. The utility of linkage studies is limited by the fact that defined genomic regions, which contain hundreds of genes, are very broad to pinpoint single disease genes. Alternative methods for identifying susceptibility genes in common disease include the genome-wide association study, followed by linkage-disequilibrium (LD) mapping using single nucleotide polymorphisms (SNPs). This approach has been used successfully to identify susceptibility genes for myocardial infarction (9).

Through a large-scale association study followed by LD mapping, we show that the calmodulin (CaM) 1 gene (CALM1) is associated with hip OA in the Japanese population. We identified a functional SNP in the core promoter region of CALM1, which affects transcription of the gene through varying affinity for nuclear protein(s). Inhibition of CaM by specific antagonists decreases expression of the major cartilage matrix genes in chondroprogenitor cells. Our findings indicate that CALM1 plays a crucial role in chondrocyte differentiation and that decreased CALM1 expression might contribute to OA.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Large-scale association study
To identify susceptibility loci for hip OA, we performed a large-scale, case–control association study using 94 individuals with hip OA and 633 control subjects (first screen). We were able to genotype 75 253 of 81 398 gene-based SNPs selected from the JSNP database. Among these, we tested the association of 71 880 SNPs that were in Hardy–Weinberg equilibrium. In each of four models (allelic and genotypic frequencies in both recessive and dominant models), 2219 SNPs had P-values less than 0.01. The results were similar to those of a previous study that identified a susceptibility gene for myocardial infarction using the same approach (9). We then genotyped the qualifying SNPs from the first screen in a replication panel consisting of 334 individuals with hip OA and 375 control subjects and tested again for association. These independent tests identified several SNPs highly associated with hip OA (Mabuchi et al., manuscript in preparation). Among these was a SNP (IVS3–293C>T) in intron 3 of CALM1 on chromosome 14q24–q31. Association of the SNP was most significant in the recessive model (‘TT’ versus other genotypes) (Table 1). The association was significant in both male and female sub-populations (Supplementary Material, Table S1). When association was tested by stratifying the cases into sub-populations with/without acetabular dysplasia, each sub-population showed significant association. The significance was higher in the cases without acetabular dysplasia (P=0.000033). The frequency of ‘TT’ genotype differed between the populations, but the difference was not statistically significant (P=0.19). The association was not found in knee OA (data not shown).

View this table: [in this window] [in a new window]   Table 1. Association of the SNP in intron 3 of CALM1 between cases and controls

 
To localize the susceptibility gene present in this region, we evaluated the LD extension with IVS3–293C>T. We genotyped 334 individuals with hip OA for SNPs in the 400 kb region around IVS3–293C>T, then examined the LD index (D') between IVS3–293C>T and SNPs having allele frequencies >10%. All SNPs showing D' scores greater than 0.99 were localized within CALM1, indicating that the LD extension with IVS3–293C>T was limited to this gene (Fig. 1A).

View larger version (20K): [in this window] [in a new window]   Figure 1. D' block and genomic structure of the CALM1 gene. (A) Linkage disequilibrium around CALM1. D' scores between the SNP in the CALM1 locus (IVS3–293C>T) and other SNPs in the region are shown with a transcriptional map. Horizontal arrows indicate the orientation of each gene (5'3'). (B) Exon–intron structure and SNPs of the CALM1. Boxes indicate exons. White fields indicate the untranslated region, and black fields indicate the coding regions. SNPs in this region are shown as axial bars with ID number.

 
Expression analyses of CALM1
CALM1 encodes CaM, a ubiquitous eukaryotic calcium-binding protein, which is a principal mediator of the calcium signal (10). CaM participates in cartilage metabolism in response to mechanical stimulation (11). We found that CALM1 was expressed in human articular chondrocytes in alginate bead culture and in human articular cartilage from OA patients (Supplementary Material, Fig. S1A). Microarray analysis showed higher levels of CALM1 expression in hip and knee OA cartilage than in normal cartilage (Supplementary Material, Fig. S1B and C), further implicating CALM1 in OA pathogenesis.

Search for susceptibility SNP
To identify a susceptibility SNP in the genomic region of CALM1, we examined SNPs for all exons of CALM1, their 5' and 3' flanking sequences and a 1.4 kb promoter region using DNA from 16 hip OA patients. Within the CALM1 region, we confirmed a total of 21 SNPs (Fig. 1B). We genotyped and divided the SNPs that were in absolute LD relationship with each other (pairwise LD index, >0.97) into eight groups (Supplementary Material, Table S2). For each SNP group with minor allele frequencies >10%, we genotyped representative SNPs in 334 case and 375 control individuals and analyzed the haplotype structure of the region. This analysis identified four common haplotypes, which covered >90% of the population haplotypes in both groups (Table 2). For each haplotype, we examined the association with hip OA and found a significant association with haplotype II (Table 2). This haplotype contained only group D SNPs, represented by the original marker SNP, IVS3–293C>T (SNP 13 in Fig. 1B). No other haplotype had a P-value lower than haplotype II. Moreover, three SNPs not included in the haplotype analyses (SNPs 14, 16 and 17 in Fig. 1B) showed no significant association in the same case–control population (data not shown). Therefore, we concluded that the association with susceptibility to hip OA originated from the group D SNPs. We confirmed that all five SNPs in group D had significant association equivalent to IVS3–293C>T (Table 3), indicating that these SNPs are all candidate susceptibility SNPs for hip OA. The P-value of the haplotype association is less than those of respective SNPs. Therefore, the possibility that the haplotype II itself is implicated in OA susceptibility is unlikely.

View this table: [in this window] [in a new window]   Table 2. Haplotype association analysis using representative SNPs

 

View this table: [in this window] [in a new window]   Table 3. Association of the SNPs of CALM1 in group D

 
In vitro functional analyses of the susceptibility SNP located on the promoter region
One of the five candidate susceptibility SNPs, –16C>T (SNP 2 in Fig. 1B), was localized within a core promoter region of CALM1, which includes the TATA element. To investigate possible allelic differences in promoter activity generated by this SNP, we generated fusion constructs which express the luciferase gene under the control of a 1.4 kb segment of CALM1 (nucleotides –1231 to +202, shown in the left panel of Fig. 2B) that has previously been reported to have promoter activity (12). Reporter constructs containing the –16C or –16T allele were transfected in parallel into the OUMS-27 cells or Huh-7 cells. OUMS-27 cells are derived from a human chondrosarcoma and retain differentiated chondrocytic phenotypes (13); Huh-7 cells are derived from a hepatocellular carcinoma (14). In OUMS-27 cells, the construct containing the –16C allele showed >2-fold luciferase activity than that containing the –16T allele (Fig. 2A). Similar results were observed in Huh-7 cells (Fig. 2B, top).

View larger version (40K): [in this window] [in a new window]   Figure 2. Transcriptional regulatory activity affected by the –16C>T SNP. (A) Luciferase assay in OUMS-27. DNA fragments of the 1.4 kb promoter region (–1231 to +202) containing the –16C or –16T allele were ligated into the promoterless/enhancerless firefly luciferase expression vector pGL3-basic. The OA-susceptible –16 allele shows reduced activity. (B) Identification of the critical region for the difference in transcriptional activity. The 1.4 kb reporter constructs containing the –16C or –16T allele, and their deletion mutants are shown in the left panel. These construct were co-transfected with pRL-TK into Huh-7 cells. Data are mean±SEM of values from six transfections. Single asterisk and double asterisks indicate P<0.05 and 0.01 by Student's t-test, respectively. (C) Gel-shift assay. Nuclear extract from Huh-7 cells was incubated with double-stranded, DIG-labeled oligonucleotides (C or T). For competition studies, the nuclear extract was pre-incubated with unlabeled oligonucleotides (125-fold excess) before adding labeled oligonucleotide. Arrows indicate the bands that show tighter binding to the T allele. The experiments were repeated three times with similar results.

 
To investigate the critical region responsible for differences in transcriptional activity between the –16C and –16T alleles, we generated three deletion mutants of the CALM1 promoter construct. Approximately 2-fold differences in luciferase activity between the two alleles were observed with each construct, including the shortest 61 bp construct (nucleotides –53 to +8), in Huh-7 cells (Fig. 2B). These results indicate that the difference in transcriptional activity results from the sequence immediately surrounding the SNP.

Using a gel-shift assay, we examined whether a nuclear extract from Huh-7 cells could bind to the sequence surrounding –16C>T (Fig. 2C). The band corresponding to the –16T allele was more intense than that corresponding to the –16C allele, indicating that one or more nuclear factors bound to the –16T allele more tightly than to the –16C allele. Taken together with the reduced transcriptional activity of the –16T allele relative to the –16C allele, this result suggests the presence of suppressor(s) of CALM1 transcription, which bind more tightly to the –16T allele.

Allelic differences in the expression of CALM1 in vivo
To investigate the effects of –16C>T on CALM1 transcription in vivo, we measured the amount of CALM1 transcripts under the control of each allele using RNA difference plot analysis (15). Using fluorescence-labeled primers, we amplified cDNA and genomic DNA from the cells of an individual who was heterozygous with respect to –16C>T and evaluated the allelic difference of the linked marker in the transcribed region (+114G>A in the 5' untranslated region, SNP 3 in Fig. 1B). Molecular haplotyping confirmed the haplotype structures of –16 to +114 as C–G and T–A. Amplified products were separated by electrophoresis specialized for polymerase chain reaction (PCR)- single-strand conformation polymorphism analysis, and the signal intensity of each allele was measured. The expression ratio of the +114G to +114A allele corresponding to that of –16C and –16T, respectively, was 1.09 (99% CI=1.04–1.15). This result was replicated through a similar experiment using another marker (exon 7+1996 G>C, SNP 20 in Fig. 1B; the expression ratio: 1.12; 95% CI=1.05–1.20). Thus, the susceptible –16T allele showed decreased in vivo expression levels relative to the non-susceptible –16C allele.

Role of CaM in chondrogenesis
To investigate the role of CaM in chondrogenesis, we used an in vitro chondrogenesis model, the ATDC5 cell line, which mimics chondrocyte differentiation in the presence of insulin (16). During differentiation, the expression of Col2a1 and Agc1, which encode major cartilage extracellular matrix proteins, increased in a time-dependent manner (Fig. 3A and B). Addition of the W-7, a naphthalene sulfonamide derivative which directly binds to and inhibits CaM, suppressed the expression of these genes by half (Fig. 3A and B). This observation indicates that CaM transduces one or more signals that induce expression of cartilage matrix genes during chondrocyte differentiation. To further investigate the role of CaM in chondrogenesis, we examined the effect of W-7 on Ca²⁺-regulated cartilage matrix gene expression in the rat chondrogenic cell line RCJ3.1C5.18. In these cells, expression of Col2a1 and Agc1 increased in the presence of ionomycin, a Ca²⁺ ionophore that increases the intracellular concentration of Ca²⁺ (Fig. 3C and D). These increases were suppressed by W-7 in a dose-dependent manner (Fig. 3C and D). This observation indicated that the Ca²⁺ signal induces cartilage matrix gene expression and is regulated by CaM in chondroprogenitor cells.

View larger version (28K): [in this window] [in a new window]   Figure 3. Effect of the CaM antagonist W-7 on chondrogenic activity. (A and B) Effect of W-7 on ATDC5 differentiation. ATDC5 cells were untreated (open circles) and exposed to 10 µg/ml insulin alone (closed squares) or to 10 µg/ml insulin plus 20 µM W-7 (closed triangles) for 6, 10, 14, 18 and 22 days. At each time point, cells were harvested and mRNA levels of Col2a1 (A) and Agc1 (B) were quantified. (C and D) Effect of W-7 on transcription of cartilage matrix genes stimulated by ionomycin in RCJ3.1C5.18 cells. Cells were untreated (white columns) or exposed to 2 µM ionomycin (Io) (black columns) with various concentration of W-7. After 24 h, cells were harvested and levels of Col2a1 (C) and Agc1 (D) mRNA were quantified. Copy numbers of these genes were measured by quantitative real-time PCR analysis and normalized relative to the amount of total RNA. Expression levels are shown as fold change when compared with levels at day 0 (A and B) or with non-ionomycin stimulated conditions (C and D). Values are average ±SEM (n=3). Statistical tests were performed between insulin alone and insulin plus W-7 (A and B). Single asterisk and double asterisks indicate P<0.05 and 0.01 by Student's t-test, respectively. Similar results were obtained in three independent studies.

 
Analyses of combinatorial effect of CALM1 and other genes
To examine the relation of CALM1 to other genes in the etiology of hip OA, we estimated the combinatorial effect of CALM1 and other genes by combining genotype data on pair of loci in 323 cases and in 374 controls. We found significant combinatorial association between the CALM1 SNP and the asparatic acid repeat polymorphism (D14 allele) in the asporin gene (ASPN) (17), with regard to hip OA susceptibility. The two loci are not linked physically. The genotype that was homozygous with respect to the CALM1 allele and homozygous or heterozygous with respect to D14 allele in ASPN showed highest odds ratio (Table 4).

View this table: [in this window] [in a new window]   Table 4. Combinatorial effect between CALM1 and ASPN

 

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Through a large-scale association study using SNPs and subsequent LD mapping, we have identified CALM1 as a susceptibility gene for hip OA. Multiple testing that might produce false-positive results is a considerable problem in large-scale association study because it uses a huge number of markers as in this study. The final P-value for the association of the SNP was calculated as 9.8x10^–7 by multiplying the P-values of the two independent tests. When Bonferroni's correction was applied to this result, we obtained only marginally significant association (the corrected P=0.070). Therefore, the possibility remains that the association of CALM1 might be false positive. However, it is widely recognized that Bonferroni's correction is overly conservative. If we exclude the makers that are in almost complete LD (>0.97) with others (n=19 794), the corrected P-value becomes 0.051. In addition to the association data, we have shown that CALM1 is expressed in normal articular cartilage, with increased expression in OA cartilage. CALM1 encodes CaM, which plays a crucial role in facilitating chondrocyte differentiation and maintaining the cartilage matrix. We have also shown that a SNP in the core promoter region of CALM1 exerts an allele-specific effect on gene expression in vitro and in vivo. These functional significances in cartilage and OA support the candidacy of CALM1 as a susceptibility gene for OA.

CaM binds to intracellular Ca²⁺ and mediates various signals involved in cellular function including proliferation, motility, cell-cycle progression and transcription (10). In chondroprogenitor cells, the Ca²⁺ signal regulates chondrogenic differentiation. The elevation of intracellular concentration of Ca²⁺ by ionomycin induces expression of Agc1 in the rat chondrogenic cell line RCJ3.1C5.18 (18). Ionomycin also induces chondrogenesis in the limb bud mesenchyme of chicken (18). Our experiments using W-7, specific antagonist of CaM, indicate that this signal is mediated by CaM. The crucial role of CaM in chondrogenic differentiation was also confirmed in ATDC5. Chondroprogenitor cells are present in human articular cartilage (19) and might be involved in cartilage repair. Decompensation of the cartilage repair capacity in chondrocyte function has been suggested as potential pathogenic factor for OA (4). Therefore, aberration of the Ca²⁺–CaM signal regulating chondrogenesis over time is likely to promote the onset and progression of OA.

The Ca²⁺–CaM signal also has a crucial function in maintaining the cartilage phenotype in response to mechanical stimuli in mature chondrocytes. Mechanical stimuli can change intracellular concentration of Ca²⁺ in chondrocytes in seconds to minutes (20,21). Cyclic or dynamic loading stimulates the synthesis of cartilage extracellular matrix components (22,23). CaM is required for the transient elevation of Agc1 mRNA in bovine articular cartilage explants by low-level compressive loading (11). These findings suggest that daily mechanical stimuli might maintain healthy articular cartilage via the Ca²⁺–CaM signaling mechanism. Therefore, aberration of this mechanism in articular chondrocytes would also eventually spur the onset and progression of OA.

The identity of downstream regulators of CaM in chondrocytes remains unclear. CaM might bind directly to SOX9, a key regulator of chondrogenesis, and facilitate its nuclear translocation (24). CaM also could facilitate chondrogenesis through activating calcineurin, a Ca²⁺–CaM dependent phosphatase, which is involved in maintaining the chondrocyte phenotype (18). Calcineurin also plays a role in the pathway mediating the transient elevation of Agc1 mRNA in response to low-level compressive loading in bovine articular cartilage explants (11). The contribution of CaM-dependent protein kinase II in this pathway has also been established (11). Further study is necessary to characterize the signaling pathway downstream of CaM. Clarification of CaM-mediated pathways would present a new paradigm for the etiology and pathogenesis of OA, as well as a new target for treatment of OA.

With regard to complex genetic traits, each related gene contributes disease susceptibility to some extent, but such genes may also interact with each other. Most recently, we have found ASPN as a susceptibility gene for both knee and hip OA (17). Asporin is an extracellular matrix protein, which inhibits chondrocyte differentiation by binding TGF-ß1 and suppressing its signal. The combinatorial effect between CALM1 and ASPN found in this study suggests presence of the epistatic effect of the two loci. There may be a cross-talk between the TGF-ß and CaM signals in chondrocyte. The relationship between CaM and asporin should be clarified in future work.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Subjects
We recruited 428 individuals affected with hip OA (24 males and 404 females, mean age: 53.7 years) and 1008 controls (517 males and 491 females, mean age: 46.7 years) from several medical institutions in Japan. All subjects were Japanese. Most cases were female (94%). All individuals with hip OA were symptomatic and were treated in these medical institutions on a regular basis. Hip OA was diagnosed by clinical symptoms and radiological findings of narrowing of the joint space and/or osteophytes of the hip joint. Because acetabular dysplasia is the most common cause and predisposing factor of hip OA in Japan (25), 40% of the cases had acetabular dysplasia (center-edge angle <25° or acetabular angle >25° in the anteroposterior radiograph of the hip). We obtained informed consent from each subject as approved by the ethical committees of the participating medical institutions and the SNP Research Center of The Institute of Physical and Chemical Research (RIKEN). Genomic DNA was extracted from peripheral blood leukocytes of each subject, using a standard protocol.

Large-scale association study
For the association study, we used our own SNP database (JSNP database, http://snp.ims.u-tokyo.ac.jp/) and carried out screening of SNPs as described previously (9,26). First, we genotyped 81 398 SNPs in genomic DNA from 94 individuals with hip OA and 633 control individuals. For the successfully genotyped SNPs, we calculated P-values as previously described (the first screen) (9). SNPs that passed the first screen were further genotyped in a second replication panel consisting of individuals with hip OA (334) and control (375) subjects, as an independent test. For both screens, P-values less than 0.01 were considered significant.

SNP discovery and genotyping
For a fine-scale association study and LD mapping of the 400 kb region containing CALM1, we detected SNPs in the region by direct sequencing of genomic DNA from 16 affected individuals as described previously (9,26). We genotyped the SNPs using the Invader assay, TaqMan assay and direct sequencing of PCR products with capillary sequencers (ABI3700, Applied biosystems) and tested for the association.

Statistical analysis
We carried out statistical analysis for the association study, Hardy–Weinberg equilibrium, calculation of LD coefficient (D') and LD index () as described previously (27,28). We estimated haplotype frequencies using the expectation-maximization algorithm (29). We examined combinatorial effects of CALM1 and other loci with regard to OA susceptibility as previously described (27). Odds ratio was defined against the genotype consisted of homozygotes of non-susceptibility alleles from both loci. Luciferase assay data were analyzed by Student's t-test using Excel software (Microsoft).

Luciferase assay
DNA fragments corresponding to the promoter and 5'-untranslated region of CALM1 (left panel in Fig. 2B) were amplified by PCR using genomic DNA containing the –16T or –16C allele as template, then cloned into the NheI–XhoI sites of pGL3-basic (Promega) in the 5'–3' orientation. The primer sequences used for PCR are listed in Supplementary Material, Table S3. We grew OUMS-27 and Huh-7 cells (JCBR cell bank, Osaka, Japan) in Dulbecco's Modified Eagle's medium (DMEM) (Sigma) supplemented with 10% FBS containing antibiotics (100 U/ml penicillin-G and 100 µg/ml streptomycin). We then transfected cells (5x10⁴ cells/well) with 0.2 µg of the constructs and 4 ng of pRL-TK vector (Promega) as an internal control for transfection efficiency, using Fugene-6 (Roche). After 24 h, cells were solubilized and luciferase activity was measured using the Pikkagene dual luciferase assay system (Toyo Ink, Tokyo, Japan).

Gel-shift assay
Nuclear extract was prepared from Huh-7 cells as previously described (30). We incubated the nuclear extract with 34 bp double-strand oligonucleotide probes for the –16C and –16T alleles for 20 min at room temperature. Oligonucleotide sequences are listed in Supplementary Material, Table S4. Probes were labeled using the digoxigenin gel-shift kit (Roche). For competition studies, we pre-incubated the nuclear extract with unlabeled oligonucleotides (125-fold excess) before adding labeled oligonucleotide. Protein–DNA complexes were separated by electrophoresis on a 6% polyacrylamide gel in 0.5x TBE buffer, followed by transfer to nitrocellulose membrane and detection using a chemiluminescent signal detection system (Roche) according to the manufacturer's instructions.

RNA difference plot analysis
RNA difference plot analysis was performed as previously described (15). The cDNA and genomic DNA from normal human articular chondrocytes (nHAC, Cambrex Corporate, East Rutherford, NJ, USA) were used as templates for PCR. Fragments containing +114G or +114A were amplified using specific primers listed in Supplementary Material, Table S5. The 5' ends of the forward primers were labeled with Texas Red. PCR was carried out using AmpliTaq DNA polymerase (Applied biosystems) in 10 µl of reaction volume, followed by treatment with 0.5 U of Klenow fragment (Takara, Tokyo, Japan). For single-strand conformation polymorphism analysis, an aliquot was diluted with 10 volumes of a loading solution containing 90% deionized formamide and 0.01% fuchsin (Sigma), then denatured and applied to a 0.5x MDE gel solution (Cambrex bio science Rockland, Inc., Rockland, ME, USA) containing 1xTBE. Electrophoresis was carried out at 20°C and 30 W for 2 h using an SF5200 autosequencer (Hitachi Ltd, Tokyo, Japan). The signal intensity of the polymorphic fragments was analyzed using Allele Links (Hitachi Ltd). Experiments were repeated nine times independently. The expression ratio between two alleles was calculated by dividing the RNA ratio by the DNA ratio with confidence interval.

Chondrogenic differentiation of ATDC5 cells
A chondrogenic mouse embryonic carcinoma cell line, ATDC5 (RIKEN cell bank, Tsukuba, Japan), was maintained in a 1 : 1 mixture of DMEM and Ham's F12 medium (DMEM/F12, Invitrogen) supplemented with 5% FBS containing antibiotics (100 U/ml penicillin-G and 100 µg/ml streptomycin). Cells were plated at 3x10⁴ cells/well in 12-well plates. After cells reached confluence, the culture medium was replaced with differentiation medium (DMEM/F12 supplemented with 5% FBS, antibiotics, 10 µg/ml bovine insulin, 10 µg/ml human transferrin and 3x10^–8 M sodium selenite) in the presence (20 µM) or absence of the CaM inhibitor W-7 (Wako, Osaka, Japan) and cultured for several time periods (0, 6, 10, 14, 18 and 22 days). The culture medium was replaced every other day.

Ionomycin stimulation of RCJ3.1C5.18 cells
RCJ3.1C5.18 cells were generously provided by Jane Aubin (University of Toronto) (31). Cells were maintained in DMEM supplemented with 10% FBS, penicillin (100 U/ml) and streptomycin (100 µg/ml). Cells were plated at 1x10⁵ cells/well in 12-well plates. After cells reached confluence, the culture medium was replaced with medium containing 2 µM ionomycin (Sigma) with or without various concentrations of W-7 and was cultured for an additional 24 h.

Quantitative real-time PCR analysis
We extracted total RNA from ATDC5 and RCJ3.1C5.18 cells and synthesized first-strand cDNA. We carried out real-time PCR on an ABI PRISM 7700 sequence detection system (Applied biosystems) using QuantiTect SYBR Green PCR (Qiagen) according to the manufacturer's instructions. First-strand cDNA was amplified using primers specific for the mouse and rat 1 chain of the type II collagen (Col2a1) gene, the aggrecan 1 (Agc1) gene and rodent Gapdh (Applied biosystems). Copy numbers of Col2a1 and Agc1 were calculated by referring standard curves and normalized by the total RNA using Gapdh as an internal control. Specific primer sequences are listed in Supplementary Material, Table S6.

	SUPPLEMENTARY MATERIAL

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 
Supplementary Material is available at HMG Online.

	ACKNOWLEDGEMENTS

 
We thank patients for participating the study. We also thank Jane Aubin, Hideki Kizawa, Ryo Yamada, Ikuyo Kou, Eiji Nakashima and Koji Yoshimura for help in performing the study and Tomoko Kusadokoro and Mizuho Kosuge for excellent technical assistance. This work was supported by Japanese Millennium Project.

	FOOTNOTES

 
Present address: Department of Orthopaedic Surgery, Osaka Rosai Hospital, 1179-3 Nagasone-cho, Sakai, Osaka 591-8025, Japan

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTARY MATERIAL REFERENCES

 

1. Lawrence, R.C., Helmick, C.G., Arnett, F.C., Deyo, R.A., Felson, D.T., Giannini, E.H., Heyse, S.P., Hirsch, R., Hochberg, M.C., Hunder, G.G. et al. (1998) Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum., 41, 778–799.[CrossRef][Web of Science][Medline]

2. Inoue, K., Wicart, P., Kawasaki, T., Huang, J., Ushiyama, T., Hukuda, S. and Courpied, J.-P. (2000) Prevalence of hip osteoarthritis and acetabular dysplasia in French and Japanese adults. Rheumatology (Oxford), 39, 745–748.

3. Yelin, E. (1998) The economics of osteoarthritis. In Brandt, K.D., Doherty, M. and Lohmander, L.S. (eds), Osteoarthritis. Oxford University Press, New York, pp. 23–30.

4. Poole, A.R. and Howell, D.S. (2001) Etiopathogenesis of osteoarthritis. In Moskowits, R.W., Howell, D.S., Altman, R.D., Buckwalter, J.A. and Goldberg, V.M. (eds), Osteoarthritis. W.B. Saunders, Philadelphia, pp. 29–47.

5. Spector, T.D. and MacGregor, A.J. (2004) Risk factors for osteoarthritis: genetics. Osteoarthritis Cartilage, 12, S39–S44.

6. Loughlin, J. (2001) Genetic epidemiology of primary osteoarthritis. Curr. Opin. Rheumatol., 13, 111–116.[CrossRef][Web of Science][Medline]

7. Loughlin, J. (2003) Genetics of osteoarthritis and potential for drug development. Curr. Opin. Pharmacol., 3, 295–299.[CrossRef][Web of Science][Medline]

8. Loughlin, J., Dowling, B., Chapman, K., Marcelline, L., Mustafa, Z., Southam, L., Ferreira, A., Ciesielski, C., Carson, D.A. and Corr, M. (2004) Functional variants within the secreted frizzled-related protein 3 gene are associated with hip osteoarthritis in females. Proc. Natl Acad. Sci. USA, 101, 9757–9762.[Abstract/Free Full Text]

9. Ozaki, K., Ohnishi, Y., Iida, A., Sekine, A., Yamada, R., Tsunoda, T., Sato, H., Sato, H., Hori, M., Nakamura, Y. et al. (2002) Functional SNPs in the lymphotoxin- gene that are associated with susceptibility to myocardial infarction. Nat. Genet., 32, 650–654.[CrossRef][Web of Science][Medline]

10. Carafoli, E. (1987) Intracellular calcium homeostasis. Annu. Rev. Biochem., 56, 395–433.[CrossRef][Web of Science][Medline]

11. Valhmu, W.B. and Raia, F.J. (2002) myo-Inositol 1,4,5-trisphosphate and Ca²⁺/calmodulin-dependent factors mediate transduction of compression-induced signals in bovine articular chondrocytes. Biochem. J., 36, 1689–1696.

12. Toutenhoofd, S.L., Foletti, D., Wicki, R., Rhyner, J.A., Garcia, F., Tolon, R. and Strehler, E.E. (1998) Characterization of the human CALM2 calmodulin gene and comparison of the transcriptional activity of CALM1, CALM2 and CALM3. Cell Calcium, 23, 323–338.[CrossRef][Web of Science][Medline]

13. Kunisada, T., Miyazaki, M., Mihara, K., Gao, C., Kawai, A., Inoue, H. and Namba, M. (1998) A new human chondrosarcoma cell line (OUMS-27) that maintains chondrocytic differentiation. Int. J. Cancer, 77, 854–859.[CrossRef][Web of Science][Medline]

14. Nakabayashi, H., Taketa, K., Miyano, K., Yamane, T. and Sato, J. (1982) Growth of human hepatoma cells lines with differentiated functions in chemically defined medium. Cancer Res., 42, 3858–3863.[Abstract/Free Full Text]

15. Murakami, Y., Isogai, K., Tomita, H., Sakurai-Yageta, M., Maruyama, T., Hidaka, A., Nose, K., Sugano, K. and Kaneko, A. (2004) Detection of allelic imbalance in the gene expression of hMSH2 or RB1 in lymphocytes from pedigrees of hereditary, nonpolyposis, colorectal cancer and retinoblastoma by an RNA difference plot. J. Hum. Genet., 49, 635–641.[CrossRef][Web of Science][Medline]

16. Atsumi, T., Miwa, Y., Kimata, K. and Ikawa, Y. (1990) A chondrogenic cell line derived from a differentiating culture of AT805 teratocarcinoma cells. Cell Differ. Dev., 30, 109–116.[CrossRef][Web of Science][Medline]

17. Kizawa, H., Kou, I., Iida, A., Sudo, A., Miyamoto, Y., Fukuda, A., Mabuchi, A., Kotani, A., Kawakami, A., Yamamoto, S. et al. (2005) An aspartic acid repeat polymorphism in asporin inhibits chondrogenesis and increases susceptibility to osteoarthritis. Nat. Genet., 37, 138–144.[CrossRef][Web of Science][Medline]

18. Tomita, M., Reinhold, M.I., Molkentin, J.D. and Naski, M.C. (2002) Calcineurin and NFAT4 induce chondrogenesis. J. Biol. Chem., 277, 42214–42218.[Abstract/Free Full Text]

19. Alsalameh, S., Amin, R., Gemba, T. and Lotz, M. (2004) Identification of mesenchymal progenitor cells in normal and osteoarthritic human articular cartilage. Arthritis Rheum., 50, 1522–1532.[CrossRef][Web of Science][Medline]

20. Guilak, F., Zell, R.A., Erickson, G.R., Grande, D.A., Rubin, C.T., McLeod, K.J. and Donahue, H.J. (1999) Mechanically induced calcium waves in articular chondrocytes are inhibited by gadolinium and amiloride. J. Orthop. Res., 17, 421–429.[CrossRef][Web of Science][Medline]

21. Roberts, S.R., Knight, M.M., Lee, D.A. and Bader, D.L. (2001) Mechanical compression influences intracellular Ca²⁺ signaling in chondrocytes seeded in agarose constructs. J. Appl. Physiol., 90, 1385–1391.[Abstract/Free Full Text]

22. Parkkinen, J.J., Lammi, M.J., Helminen, H.J. and Tammi, M. (1992) Local stimulation of proteoglycan synthesis in articular cartilage explants by dynamic compression in vitro. J. Orthop. Res., 10, 610–620.[CrossRef][Web of Science][Medline]

23. Valhmu, W.B., Stazzone, E.J., Bachrach, N.M., Saed-Nejad, F., Fischer, S.G., Mow, V.C. and Ratcliffe, A. (1998) Load-controlled compression of articular cartilage induces a transient stimulation of aggrecan gene expression. Arch. Biochem. Biophys., 353, 29–36.[CrossRef][Web of Science][Medline]

24. Argentaro, A., Sim, H., Kelly, S., Preiss, S., Clayton, A., Jans, D.A. and Harley, V.R. (2003) A SOX9 defect of calmodulin-dependent nuclear import in campomelic dysplasia/autosomal sex reversal. J. Biol. Chem., 278, 33839–33847.[Abstract/Free Full Text]

25. Nakamura, S., Ninomiya, S. and Nakamura, T. (1989) Primary osteoarthritis of the hip joint in Japan. Clin. Orthop., 241, 190–196.

26. Iida, A., Saito, S., Sekine, A., Mishima, C., Kondo, K., Kitamura, Y., Harigae, S., Osawa, S. and Nakamura, Y. (2001) Catalog of 258 single-nucleotide polymorphisms (SNPs) in genes encoding three organic anion transporters, three organic anion-transporting polypeptides, and three NADH: ubiquinone oxidoreductase flavoproteins. J. Hum. Genet., 46, 668–683.[CrossRef][Web of Science][Medline]

27. Yamada, R., Tanaka, T., Unoki, M., Nagai, T., Sawada, T., Ohnishi, Y., Tsunoda, T., Yukioka, M., Maeda, A., Suzuki, K., et al. (2001) Association between a single-nucleotide polymorphism in the promoter of the human interleukin-3 gene and rheumatoid arthritis in Japanese patients, and maximum-likelihood estimation of combinatorial effect that two genetic loci have on susceptibility to the disease. Am. J. Hum. Genet., 68, 674–685.[CrossRef][Web of Science][Medline]

28. Devlin, B. and Risch, N. (1995) A comparison of linkage disequilibrium measures for fine-scale mapping. Genomics, 29, 311–322.[CrossRef][Web of Science][Medline]

29. Excoffier, L. and Slatkin, M. (1995) Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol. Biol. Evol., 12, 921–927.[Abstract]

30. Andrews, N.C. and Faller, D.V. (1991) A rapid micropreparation technique for extraction of DNA-binding proteins from limiting numbers of mammalian cells. Nucleic Acids Res., 19, 2499.[Free Full Text]

31. Grigoriadis, A.E., Heersche, J.N.M. and Aubin, J.E. (1990) Continuously growing bipotential and monopotential myogenic, adipogenic, and chondrogenic subclones isolated from the multipotential RCJ 3.1 clonal cell line. Dev. Biol., 142, 313–318.[CrossRef][Web of Science][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				A M Valdes, T D Spector, S Doherty, M Wheeler, D J Hart, and M Doherty Association of the DVWA and GDF5 polymorphisms with osteoarthritis in UK populations Ann Rheum Dis, December 1, 2009; 68(12): 1916 - 1920. [Abstract] [Full Text] [PDF]

				S.-C. Mu, H.-C. Liu, J.-Y. Wu, M.-T. M. Lee, H.-P. Chuang, L.-K. Chen, and Y.-T. Chen A large kindred of early-onset osteoarthritis of the knee and hip: excluding the link to COL2A1 gene Rheumatology, April 1, 2009; 48(4): 371 - 374. [Abstract] [Full Text] [PDF]

				H. Mototani, A. Iida, M. Nakajima, T. Furuichi, Y. Miyamoto, T. Tsunoda, A. Sudo, A. Kotani, A. Uchida, K. Ozaki, et al. A functional SNP in EDG2 increases susceptibility to knee osteoarthritis in Japanese Hum. Mol. Genet., June 15, 2008; 17(12): 1790 - 1797. [Abstract] [Full Text] [PDF]

				I. Meulenbelt, J. L. Min, S. Bos, N. Riyazi, J. J. Houwing-Duistermaat, H.-J. van der Wijk, H. M. Kroon, M. Nakajima, S. Ikegawa, A. G. Uitterlinden, et al. Identification of DIO2 as a new susceptibility locus for symptomatic osteoarthritis Hum. Mol. Genet., June 15, 2008; 17(12): 1867 - 1875. [Abstract] [Full Text] [PDF]

				K. Chapman, A. Takahashi, I. Meulenbelt, C. Watson, J. Rodriguez-Lopez, R. Egli, A. Tsezou, K. N. Malizos, M. Kloppenburg, D. Shi, et al. A meta-analysis of European and Asian cohorts reveals a global role of a functional SNP in the 5' UTR of GDF5 with osteoarthritis susceptibility Hum. Mol. Genet., May 15, 2008; 17(10): 1497 - 1504. [Abstract] [Full Text] [PDF]

				J J Ryder, K Garrison, F Song, L Hooper, J Skinner, Y Loke, J Loughlin, J P T Higgins, and A J MacGregor Genetic associations in peripheral joint osteoarthritis and spinal degenerative disease: a systematic review Ann Rheum Dis, May 1, 2008; 67(5): 584 - 591. [Abstract] [Full Text] [PDF]

				H. Masuya, K. Nishida, T. Furuichi, H. Toki, G. Nishimura, H. Kawabata, H. Yokoyama, A. Yoshida, S. Tominaga, J. Nagano, et al. A novel dominant-negative mutation in Gdf5 generated by ENU mutagenesis impairs joint formation and causes osteoarthritis in mice Hum. Mol. Genet., October 1, 2007; 16(19): 2366 - 2375. [Abstract] [Full Text] [PDF]

				L. Southam, J. Rodriguez-Lopez, J. M. Wilkins, M. Pombo-Suarez, S. Snelling, J. J. Gomez-Reino, K. Chapman, A. Gonzalez, and J. Loughlin An SNP in the 5'-UTR of GDF5 is associated with osteoarthritis susceptibility in Europeans and with in vivo differences in allelic expression in articular cartilage Hum. Mol. Genet., September 15, 2007; 16(18): 2226 - 2232. [Abstract] [Full Text] [PDF]

				J. Hata, K. Matsuda, T. Ninomiya, K. Yonemoto, T. Matsushita, Y. Ohnishi, S. Saito, T. Kitazono, S. Ibayashi, M. Iida, et al. Functional SNP in an Sp1-binding site of AGTRL1 gene is associated with susceptibility to brain infarction Hum. Mol. Genet., March 15, 2007; 16(6): 630 - 639. [Abstract] [Full Text] [PDF]

				S. E. Ali-Khan and B. F. Hales Novel Retinoid Targets in the Mouse Limb during Organogenesis Toxicol. Sci., November 1, 2006; 94(1): 139 - 152. [Abstract] [Full Text] [PDF]

				J Loughlin Osteoarthritis linkage scan: more loci for the geneticists to investigate. Ann Rheum Dis, October 1, 2006; 65(10): 1265 - 1266. [Full Text] [PDF]

				K. Yamazaki, D. McGovern, J. Ragoussis, M. Paolucci, H. Butler, D. Jewell, L. Cardon, M. Takazoe, T. Tanaka, T. Ichimori, et al. Single nucleotide polymorphisms in TNFSF15 confer susceptibility to Crohn's disease Hum. Mol. Genet., November 15, 2005; 14(22): 3499 - 3506. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Supplementary Material All Versions of this Article: 14/8/1009    most recent ddi093v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (26) Request Permissions Google Scholar Articles by Mototani, H. Articles by Ikegawa, S. Search for Related Content PubMed PubMed Citation Articles by Mototani, H. Articles by Ikegawa, S. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics