| Sign In to gain access to subscriptions and/or personal tools. |
Hypermethylation of CpGs in the Promoter of the COL1A1 Gene in the Aged Periodontal Ligament
1 Division of Aging and Geriatric Dentistry, Department of Oral Function and Morphology, Tohoku University Graduate School of Dentistry, 4-1, Seiryo-machi, Aobaku, Sendai 980-8575, Japan; Correspondence: * corresponding author, tono{at}mail.tains.tohoku.ac.jp
Although the human periodontal ligament shows age-associated histological alterations, the molecular mechanisms are not yet understood. We previously found that COL1A1 gene expression declines with age. In this study, we asked whether DNA methylation in the regulatory region of the gene alters in the aging process, as a possible cause of the decline. The method used was a bisulfite modification of cytosine and nucleotide sequencing of DNA. While the 1st intron region was kept demethylated at young and old ages, the levels of methylation at most CpG sites in the proximal and distal regions of the promoter showed elevation at older ages. Analysis of the data indicates the possible importance of DNA hypermethylation in the promoter region for the age-associated decrease of COL1A1 gene expression.
Key Words: COL1A1 • DNA methylation • CpG islands • aging • periodontal ligament • human
One of the major causes of frequent tooth loss in older individuals is attributed to periodontal disease arising from multiple factors such as bacterial infection, immune dysfunction, tissue deterioration, etc. (Haffajee et al., 1991; van der Velden, 1991). An age-dependent decline in the function of periodontal tissues could also be one of the factors contributing to these problems. In fact, cells from the periodontal ligaments of healthy older individuals have less proliferative activity than do cells from younger people (Nishimura et al., 1997). Histological studies also report age-dependent alterations, like structural irregularity, decreases in fiber and cell content, etc. (Severson et al., 1978; van der Velden, 1984; Moxham and Evans, 1995). Thus, it is likely that the status of the periodontal ligament changes during the aging process. However, little is known about periodontal ligament changes at the molecular level. It has been previously reported that the mRNA levels of collagen  1(I) decrease with age in the human periodontal ligament (Takatsu et al., 1999). An age-dependent decline of collagen 1(I) has also been observed during the process of in vitro aging of cells derived from human periodontal ligament (Goseki et al., 1996). Since collagen 1(I) is one of the major proteins in the periodontal ligament, a decrease in the expression of its mRNA could affect the integrity of the tissue. What could cause the age-associated decline in mRNA levels? Among many possibilities, a change in DNA methylation is one candidate, because it has been shown to occur in the aging process for several genes (Ono et al., 1993; Issa, 1999; Zhang et al., 2002), and it has also proved to be associated with a change in gene expression (Komura et al., 1995; Issa, 1999). Many lines of evidence indicate that methylation of cytosine at the CpG sequence suppresses gene expression, and that demethylation activates gene expression. The relationship is clear, especially when the change in DNA methylation takes place in the promoter region, where CpG is present at high frequency, and is called the ’CpG island’ (Komura et al., 1995; Issa, 1999; Jaenisch and Bird, 2003). The relationship is further supported by the studies of cellular responses to 5-azacytidine, a potent demethylating agent. It induces the hypomethylation of many genes at CpG islands, as well as the alteration of gene expression and cellular differentiation (Makino et al., 1999). Expression of the COL1A1 gene is now known to be controlled by many factors, including a change of its DNA methylation status (Ghosh, 2002). Suppression of COL1A1 gene expression after the transformation of normal human fibroblasts by SV40 was associated with an increase of DNA methylation (Parker et al., 1982). Methylation in the promoter region as well as in exon 1 has been shown to depress COL1A1 gene expression in cultured 3T3 and F9 cells (Rhodes et al., 1994). In the case of the COL1A2 gene, which is coordinately regulated with COL1A1, methylation at a specific site in the gene (+7) was shown to reduce transcriptional activity of the gene through binding of the methylated DNA-binding protein/regulatory factor X (MDBP/RFX) at that site (Sengupta et al., 1999). Expression of the COL1A1 gene is also affected by the DNA sequence upstream from the first exon to the first intron of the gene (–3600 to +1440) (Bornstein, 1996; Rossert et al., 2000; Ponticos et al., 2004). Previous analysis of the sensitivity of the COL1A1 gene to methylation-sensitive restriction enzymes found an age-dependent increase in methylation at two CpG sites at the distal (–1705) and proximal (–80) promoter regions of the gene (Takatsu et al., 1999). However, the COL1A1 gene is known to be affected by and associated with many CpG sites (CpG islands) that can be methylated and associated with the regulation of gene expression. In this study, we focused on the three regions within the CpG island which have been shown to contain regulatory sequences for gene expression, and examined the degree of DNA methylation in all of the CpG sites in these regions in an effort to understand the nature of age-dependent alteration of DNA methylation in detail.
Tissue and DNA From four young (ages 9, 10, 12, 16 yrs) and four old (ages 63, 63, 64, 74 yrs) individuals, we obtained periodontal ligament tissues from teeth extracted because of dental caries, orthodontic therapy, or the presence of a third molar. Teeth extracted because of periodontitis were not used. Immediately after extraction, teeth were frozen and kept at –70°C until use. After thawing the tooth on ice, we removed tissues at the cervical and apical portions using a scalpel to avoid contamination from gingival and pulp tissue. The remaining periodontal ligament tissue in the central section of the root was removed by being scraped with a scalpel (Takatsu et al., 1999). Genomic DNA was extracted from the excised tissue with proteinase K treatment and a phenol-extraction procedure (Ono et al., 1985). The study was carried out after permission was obtained from the Ethics Committee, Tohoku University Graduate School of Dentistry, and with informed consent from the individuals.
Bisulfite Modification of DNA
The three DNA regions in and around the COL1A1 gene (Fig. 1
All of the primers were located in the DNA regions where there are no CpG sequences. All of the cytosines in the primer regions were assumed to be converted to uracil. For polymerization, Ex-Taq polymerase (Takara, Kyoto, Japan) was used.
Sequencing Analysis
Statistical Analysis
DNA Methylation in the Proximal Promoter Region The degree of DNA methylation in the –452 to +116 region was determined first. In this region, 18 CpG sites are present (Fig. 2A  ). After sodium bisulfite treatment of the genomic DNA from the periodontal ligament of a 9-year-old male, 40 COL1A1 gene clones were sequenced. The cytosines at all of the CpG sites were thymines, indicating that they were not methylated. One exception was the C in the second CpG on the 5' side of the region (–192) in clone 27 (Fig. 2B). It was cytosine, indicating that the C was methylated. In the same way, from 32 to 40 DNA clones were isolated from different individuals and analyzed for their nucleotide sequences (Fig. 2B). From these data, the frequency of methylation at each CpG site was calculated. All of the CpG sites showed very low levels of DNA methylation in younger tissue, whereas the levels were elevated in older tissue at all sites (Fig. 2C). The levels of methylation in older tissues varied from 3.7% at +70 and –88 to 39.5% at +27. It should be noted that many of the clones from tissues from older donors showed no methylation at any of the CpG sites—e.g., clones 4, 5, 10 in the sample labeled as age 63a (Fig. 2B).
DNA Methylation in the Distal Promoter Region A similar analysis was conducted for the upstream region from –2001 to –1501. This area had 16 CpG sites (Fig. 3A ). For each sample, from 22 to 34 DNA clones were analyzed (data not shown), and the levels of methylation were calculated. In this region, the levels of DNA methylation were higher than those seen in the proximal promoter region in tissues from younger donors (Fig. 3B). The extent of methylation ranged from 16.0% at –1650 to 77.0% at –1890 and –1562. In tissues from older donors, the level of methylation increased at all CpG sites examined (Fig. 3B).
DNA Methylation in the 1st Intron Although the evidence is controversial, the 1st intron of the COL1A1 gene is reported to have an important role in the regulation of gene expression (Bornstein, 1996; Ghosh, 2002), suggesting that DNA methylation in this region might be of importance. The levels of methylation at 14 CpG sites from +874 to +1132 were examined (Fig. 3C  ). For each individual, from 18 to 22 DNA clones were examined. All of the clones studied showed very low levels of methylation at all CpG sites, regardless of the age of the donor (Fig. 3D). Thus, this region was hypomethylated in both younger and older donors.
The present study indicates that the promoter region of the COL1A1 gene in the human periodontal ligament becomes hypermethylated during the aging process. This alteration is observed at most CpG sites in the proximal (–273 to +70) and distal (–1903 to –1533) promoter regions, but not in the intron 1 region (+908 to +1102), where practically all CpG sites remained unmethylated in both younger and older human samples. Since an increase in DNA methylation in the promoter and/or first exon regions has been shown to result in the repression of COL1A1 gene transcription (Thompson et al., 1991; Rhodes et al., 1994), the observed age-related hypermethylation in the promoter is in good agreement with the age-dependent decrease in mRNA levels previously reported (Takatsu et al., 1999). Although some genes are known to show age-related alterations in DNA methylation (Ono et al., 1993), the regions of the genes in which the alterations take place are remote from the promoters, and only a very limited number of genes—such as estrogen receptor (ER), insulin-like growth factor 2 (IGF2), and integrin alpha L (ITGAL)—show alterations in their promoter regions (Issa, 1999; Zhang et al., 2002). The COL1A1 gene in the human periodontal ligament can now be classified as one of those genes which show age-related alterations in their promoter regions. Further studies on the molecular mechanism responsible for the correlation between the promoter methylation and the decrease of mRNA level might provide a way to prevent age-associated changes in collagen level in the human periodontal ligament. This age-related increase in methylation was observed in most of the CpG sites in the distal and proximal promoter regions. Thus, it is impossible to speculate if there are any specific CpG sites where methylation plays a critical role in transcriptional regulation, or if hypermethylation at many CpG sites in the region contributes to the repression of mRNA levels. In the case of the COL1A2 gene, methylation of the CpG site located at +7 in exon 1 has been shown to enhance the binding of a protein named MDBP/RFX, suggesting that it could be a part of methylation-mediated suppression of transcription (Sengupta et al., 1999). In the case of COL1A1, however, such binding activity was not observed, although the nucleotide sequence in the corresponding region showed some similarity to that seen in COL1A2 (Sengupta et al., 1999). It is likely that hypermethylation at many CpG sites, rather than a specific CpG, in the promoter region is involved in the alteration of COL1A1 gene expression, probably through alteration in chromatin structure (Jaenisch and Bird, 2003).
One interesting characteristic of DNA methylation, revealed by the present study, is its randomness. The locations at which CpG sites were methylated were random in young samples in the distal promoter region. When the degree of methylation increased with increasing age, it was not focused at any specific CpG site in the region, nor on any particular DNA molecule. Instead, the alteration seemed to occur at random. A similar tendency was observed in the proximal promoter region in samples from older donors (Fig. 2B Although the levels of methylation were elevated in samples from older donors in most of the CpG sites in the distal and proximal promoter regions, methylation in the first intron region remained undetected, even in older samples. Rhodes and Breindl previously reported (1992) that the first intron is not methylated in various tissues examined, regardless of wide differences in COL1A1 gene expression. Thus, it is likely that the intron 1 region is strongly protected from hypermethylation, probably through a unique chromatin structure (Jaenisch and Bird, 2003), but the mechanism of this protection is not yet known.
This work was supported by a grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan. We thank Dr. L.N. Kapp for editorial help. Received for publication April 15, 2005. Revision received October 12, 2005. Accepted for publication October 26, 2005.
Journal of Dental Research, Vol. 85, No. 3,
245-250 (2006)
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
