Keywords: MYH gene; 8-OHdG; Hypermethylation; Inflammation; Oxidative DNA damage; IL-1; Free radicals

Abbreviations: 8-OHdG, 8-hydroxy-2′-deoxyguanosine; BER, base excision repair; IL-1, interleukin-1; MYH, human MutY glycosylase homolog

Article Outline

Materials and methods

Subjects

Variation screening for the MYH gene

Methylation analysis

Measurements of 8-OHdG contents in genomic DNA of blood cells

Quantitative assay of IL-1 in plasma

Statistical analysis

Results

A novel MYH variation identified in a Chinese population

The genotype distribution of this novel AluYb8 insertion in a Chinese population

Methylation of the AluYb8 insertion fragment

Impact of the AluYb8MYH variation on the repair of DNA oxidative damage

The plasma levels of the proinflammation marker IL-1 with AluYb8MYH variation

Discussion

Acknowledgements

References

Oxidative DNA injury is a common consequence of exposure to a variety of environmental and disease-generated free radicals, with potentially deleterious effects including mutagenesis and carcinogenesis. For example, oxidation of guanosine to 8-hydroxy-2′-deoxyguanosine (8-OHdG) can lead to mismatching with adenine. The human body has developed a fairly efficient base excision repair (BER) system, which is critical for repairing oxidized DNA and maintaining the fidelity of DNA replication [1]. Several DNA glycosylases are associated with the BER system. Human MutY glycosylase homolog (MYH) is specifically involved in the removal of adenines mismatched with 8-OHdG that arise through DNA replication errors and DNA recombination [2]. Together with 8-oxoG glycosylase (hOGG1) and human MutT homolog, the MYH protein protects the cell from the mutagenic effects of 8-OHdG, a stable product caused by oxidative damage to DNA, because 8-OHdG in DNA can lead to misincorporation of adenines opposite the 8-OHdG lesions, leading to C:G to A:T transversions [3]. Hence, the level of 8-OHdG in DNA is widely used as a diagnostic marker for loss of base excision repair activity [4].

Furthermore, the level of 8-OHdG in DNA is linked to aging [5]. Indeed, age-related diseases, such as diabetes, atherosclerosis, and cancers, are associated with high levels of 8-OHdG in DNA, suggesting a direct or indirect causal relationship [6]. Variations detected in BER genes that lead to increased levels of 8-OHdG in DNA have been described. Thus, it has been proposed that genetic variation in oxidative DNA repair may lead to altered susceptibility to age-related diseases [7].

Aging is also associated with chronic and low-grade systemic inflammation. This is not surprising because many age-related processes such as generalized atherosclerosis must cause low-grade inflammation [8]. Indeed, measurement of C-reactive protein and myeloperoxidase in plasma is predictive of coronary artery disease. Measurement of interleukin-1 (IL-1) as a key proinflammatory cytokine has also been extensively investigated in the pathogenesis of age-related diseases [9] and has been used as a predictor for the severity of age-related diseases [10].

In this study, we report a novel variant of the MYH gene in the Chinese population. This variant is caused by an Alu insertion in the MYH gene and we describe for the first time that it is a common mutation in the Chinese population. It is associated with increased concentration of leukocyte 8-OHdG in the genomic DNA and increased circulating levels of IL-1. The data from this study indicate that this variant in the Chinese population is likely to be a significant risk factor for the common age-related diseases and probably mortality, because the frequency of this mutation decreased with age in both the “healthy” population studied and an unselected population. By implication these results may also extend to the Western world.

Materials and methods

Subjects

Cohort A: a cross-sectional study was carried out in Jiangsu Province, Eastern China, from 2006 to 2008. We randomly recruited subjects 20–59 years of age from people who attended hospital for a routine health examination. Medical examination included a detailed interview by a physician, chest X-ray, B ultrasound of abdomen, and cardiogram. Various laboratory tests were done by standard methods, including the test for hepatic function, fasting glucose, rheumatoid factor, α-fetoprotein, and carcinoembryonic antigen. According to the clinical and laboratory characteristics, persons suffering from certain diseases, such as acute inflammation, tuberculosis, autoimmune diseases, diabetes, cancer, and cardiovascular diseases, were excluded. As a result, 73 persons diagnosed with diabetes, 67 with cardiovascular diseases, and 12 with other forms of chronic diseases were not included in this cohort.

A total of 1146 subjects (631 males; 515 females) classified as healthy were thus included in this investigation. The average age was 36.6  ±  11.1 years (range 20–59 years). The cohort was divided into four 10-year groups, and there were 426, 254, 249, and 217 cases in the age groups 20–29, 30–39, 40–49, and 50–59 years, respectively.

Cohort B: a total of 562 subjects ages 20–59 years (280 males; 282 females), who attended hospital for a routine health examination and were unselected on the basis of health, were randomly recruited for another cohort analysis. The average age was 39.3  ±  10.3 years. This cohort was also subdivided into four 10-year groups, and there were 121, 174, 147, and 120 cases in the age groups 20–29, 30–39, 40–49, and 50–59 years, respectively.

Informed consent was obtained from all individuals and the study was approved by the local authorities.

Variation screening for the MYH gene

Blood samples were obtained from healthy subjects included in cohort A and every individual in cohort B. Peripheral lymphocytes were separated for genomic DNA isolation, and the plasma was stored at − 80°C for other assays. Genomic DNA was extracted using an UltraPure blood kit (SBS Genetech Co. Ltd., Shanghai, China). The primers were designed to amplify the following sequence containing the variation: forward, 5′-TCTTGACCTGGAGACCTTCC-3′, and reverse, 5′-AGCTGCTTCCTCCAAACAGC-3′. The PCR products were separated on 1% agarose gels (Invitrogen, Carlsbad, CA, USA) for assessing the pattern of variation in the MYH gene. PCR products presenting a variant band(s) were further cloned into the PMD-18-T vector (TaKaRa, Shiga, Japan), transformed in Escherichia coli Top10, and then sequenced using BigDye Terminator (Applied Biosystems, Foster City, CA, USA) on an ABI Prism 3100 genetic analyzer (Applied Biosystems).

According to the absence or presence of the variant fragment, the MYH genotypes were classified as homozygous absence of this variation (absence/absence, A/A), homozygous presence of this variation (presence/presence, P/P), and heterozygote (absence/presence, A/P).

Methylation analysis

To detect the methylation status of the variation fragment, a methylation analysis was performed on DNA samples from 30 P allele carriers. These subjects were randomly picked from the healthy population age group 20–29 years, including 16 individuals with the P/P genotype (9 males and 7 females, mean age 26.1  ±  2.5 years) and 14 individuals with the A/P genotype (8 males and 6 females, mean age 25.2  ±  2.7 years).

Genomic DNA (1 μg) was bisulfate treated with the CpGenome DNA modification kit (Chemicon, Temecula, CA, USA) and amplified by methylation-specific PCR (MSP) and bisulfate sequencing PCR (BSP). For MSP, 1 μl bisulfite-modified DNA was amplified using MSP primers that specifically recognized either unmethylated (forward, 5′-TGAGTAGTTGGGATTATAGGTGTTTG-3′, and reverse, 5′-ACTACAATTAATTCCCCTCCCAAAC-3′) or methylated (forward, 5′-CGAGTAGTTGGGATTATAGGCGTTC-3′, and reverse, 5′-GCTACGATTAATTCCCCTCCCA-3′) sequence. Amplification products were visualized on 1.0% agarose gels. For BSP, the primers without any CpG dinucleotides were specifically designed for the modified templates: forward, 5′-GAGTTTTGTGGGATATGAATTGTGG-3′; reverse, 5′-CTTCCTCCAAACAACCTTTCCT-3′, product size 721 bp. The amplified products were gel-purified and ligated into the PMD-18-T vector. Ten clones from each isolate were sequenced separately.

Measurements of 8-OHdG contents in genomic DNA of blood cells

For measurement of 8-OHdG, 100 subjects with the three genotypes 20–29 years of age were randomly recruited from the healthy population being investigated. DNA extraction from fasting venous whole blood was performed within 1 h of collection, using the salting out method [11]. The DNA was stored at − 80°C until all samples were assayed as a single batch.

Initially the 100 frozen DNA samples were thawed, and the absorbance of DNA was measured at 230, 260, and 280 nm using an Eppendorf BioPhotometer Plus (Eppendorf North America). An OD (optical density, the unit of absorbance) reading at 260 nm of 1 corresponds to approximately 50 μg/ml for double-stranded DNA; the DNA concentration was calculated as follows: DNA concentration (μg/ml) =  (OD value at 260 nm) × (50 μg/ml) × dilution factor. The purity of the DNA sample was checked by OD_260 nm/OD_280 nm, which should be 1.80–1.85, and OD_260 nm/OD_230 nm, which should be 2.20–2.25. Of the DNA samples extracted, 98 were qualified as being sufficiently pure for 8-OHdG quantification (55 males and 43 females, mean age 24.3  ±  3.2 years) and the remaining 2 samples were excluded. Of the 98 samples for 8-OHdG measurement, 36 were from homozygous A/A carriers (20 males and 16 females, mean age 25.4  ±  3.0 years), 29 homozygous P/P (18 males and 11 females, mean age 24.0  ±  3.2 years), and 33 heterozygous (17 males and 16 females, mean age 23.4  ±  3.3 years).

According to the concentration of DNA detected above, DNA samples (200 μg) were dissolved in 135 μl water. Sodium acetate (15 μl, 200 mM) and nuclease P1 (15 μl, 6 units; Sigma, St. Louis, MO, USA) were added to the DNA solution and incubated for 30 min at 37°C. Tris–HCl buffer (15 μl, 1 M, pH 7.4) and alkaline phosphatase (7 μl, 2 units; TaKaRa) were added and incubated for another 30 min at 37°C. The hydrolysate was filtered through Millipore Microcon at 14,000 rpm for 10 min and 50 μl of DNA digested as stated above was drawn and applied to one well of an ELISA kit (Highly Sensitive 8-OHdG Check; JaICA, Fukuroi, Shizuoka, Japan). According to the kit instructions, a typical layout for loading in triplicate for each sample was used in our study and 18 samples consisting of three genotypes were included in every plate.

According to the kit instructions, results are normally expressed in nanograms per milliliter, but we have converted all results to 8-OHdG/10⁶ dG using the following calculation: 200 μg DNA was dissolved in a total of 187 μl solution (135 μl water, 15 μl acetate, 15 μl nuclease P, 15 μl Tris–HCl, 7 μl alkaline phosphatase), yielding a concentration of 1.07 μg DNA/μl, or 1.07 mg DNA/ml. Thus, the resulting unit of 1 ng/ml is equivalent to 1 ng 8-OHdG/1.07 mg DNA, that is, 0.93 ng 8-OHdG/mg DNA. Because the molecular weight of 8-OHdG is 283.24 Da, 0.93 ng 8-OHdG/mg DNA is equal to 0.0033 nmol 8-OHdG/mg DNA. According to Halliwell [12], 1 nmol of 8-OHdG/mg DNA is 318 8-OHdG/10⁶ DNA bases; thus 0.0033 nmol 8-OHdG/mg DNA can be converted to 1.05 8-OHdG/10⁶ DNA bases. As 1 8-OHdG/10⁵ dG in DNA is 2.2 8-OHdG/10⁶ DNA bases [12], that means 1 8-OHdG/10⁶ DNA bases is 4.54 8-OHdG/10⁶ dG. Therefore, 1.05 8-OHdG/10⁶ DNA bases gives the result 4.8 8-OHdG/10⁶ dG. Briefly, 1 ng/ml is converted to 4.8 8-OHdG/10⁶ dG based on Halliwell.

Based on the principle of the salting out method for genomic DNA isolation and ELISA for 8-OHdG measurement, the process has a limited potential to generate artificial ex vivo oxidative damage. The genomic DNAs were extracted within 1 h and stored frozen until they were assayed together. Then, the DNA samples were quantified and detected in parallel.

Quantitative assay of IL-1 in plasma

To evaluate the level of IL-1 in plasma, 149 subjects (79 males and 70 females) were randomly recruited from the healthy population in age range of 20–29 years (mean age 25.3  ±  2.1 years). In detail, there were 41 subjects with the P/P genotype (18 males and 23 females, mean age 26.0  ±  1.7 years), 52 with the A/A genotype (29 males and 23 females, mean age 25.4  ±  3.0 years), and 56 with the A/P genotype (32 males and 24 females, mean age 24.8  ±  2.2 years). Plasma samples from these subjects were stored at − 80°C and assays were performed as a single batch to quantify the concentration of plasma IL-1 with a commercial ELISA kit (rat anti-human IL-1; Adilitteram Diagnostic Laboratories, Inc., USA). The minimum level detected with this kit was 1 pg/ml.

Statistical analysis

All statistical analysis was carried out using the statistical program SPSS version 11.0. Descriptive statistical values included mean  ±  SD values for continuous data and percentages for categorical data. Comparisons of the genotypes and allelic frequencies for individuals were performed with chi-square tests. Separate comparisons of 8-OHdG contents in genomic DNA and IL-1 in plasma among subjects were conducted with ANOVA and followed by post hoc analysis. In all cases, a p value of less than 0.05 was considered statistically significant.

Results

A novel MYH variation identified in a Chinese population

When we initially screened the genomic DNA variation of the MYH gene in Chinese, an unexpected fragment (about 300 bp) was detected within intron 15. It was a novel insertion variation and presented a polymorphism distribution in a Chinese population, which classified the MYH gene as genotype A/A, A/P, and P/P (Fig. 1A).

The genotype distribution of this novel AluYb8 insertion in a Chinese population

The incidence of the AluYb8MYH variation was observed in 1146 healthy Chinese. Overall, it was a common variation and the allele frequencies of P and A were 43.2 and 56.8%, respectively. For the proportions of the three kinds of genotype, A/A was detected in 400 (34.9%), A/P in 501 (43.7%), and P/P in 245 (21.4%) of the 1146 subjects investigated. The Hardy–Weinberg law fails to apply in this population because the healthy individuals were chosen from the general population (P =  0.031).

Interestingly, the prevalence of the AluYb8MYH variation was different among the four age groups (Table 1). The frequency of P/P genotype declined from 25.8 to 15.7% with age (p =  0.007). Compared with the age group of 20–29 years, the frequency of the P/P genotype was significantly lower in the 40–49 and 50–59 years groups (p =  0.010 and p =  0.004, respectively). Meanwhile, the incidence of the A/A or A/P genotype showed little fluctuation among these groups.

The methylation study was performed on 30 P allele carriers (16 with P/P and 14 with A/P genotype). Hypermethylation was detected in all the P alleles among subjects with P/P or A/P genotype (Fig. 3B). An unmethylated band was not observed in any target sequence. The bisulfate sequencing PCR confirmed the results above. All 24 of the CpG dinucleotides in the inserted AluYb8 sequence and 6 flanking ones were sequenced. The results showed that each CpG site was methylated in all cases. No different pattern was detected (Fig. 3C). We also investigated the methylation modification of the CpG island that was reported in the promoter region of the MYH gene. The CpG island in the MYH promoter region was hypomethylated, and non-CpG methylation was detected in the 30 samples with P/P or A/P genotype and in 7 with A/A genotype (data not shown).

Impact of the AluYb8MYH variation on the repair of DNA oxidative damage

To evaluate the potential functional effect of the AluYb8MYH variation, the 8-OHdG levels in the genomic DNA of blood cells were tested in 98 randomly picked healthy Chinese subjects (Table 3). They included 36 subjects with the A/A genotype, 33 with A/P, and 29 with P/P.

Discussion

One of the consequences of mutation in the BER system is a functional defect in the exons of genes, leading to the generation of missense and nonsense mutations [14]. The Alu element is the most abundant family of repetitive DNA sequences in the human genome and is often present in noncoding regions, such as introns, 3′ untranslated regions of genes, and intergenic regions [15]. An increasing number of studies show that Alu's inserted in the human genome are associated with human disease [16] and AluYb8 is one of the most active Alu subfamilies [17]. In this work, we described a novel MYH polymorphism regarding Alu, an AluYb8 insertion in intron 15 of the MYH gene among the Chinese population. Our data suggest that this mutation may contribute significantly to age-related diseases by leading to an accumulation of 8-OHdG in genomic DNA and increased circulating concentrations of IL-1, although the latter (IL-1) is a marker of chronic disease rather than the cause.

Alu retrotransposons also have a high content of GC, and CpG dinucleotides are frequently hypermethylated in their insertion sites [18]. In this study it was demonstrated that the AluYb8 retrotransposon inserted in the MYH gene was hypermethylated in genomic DNA by MSP (Fig. 3). This result was confirmed by BSP analysis. Hypermethylation status and high GC content of the Alu sequence have been reported to affect the dynamics of the surrounding nucleotide sequence, such as the cis-regulatory element, whereas many large introns of eukaryotes cover cis-regulatory elements, including enhancers or silencers [19] and [20]. Because intron 15 is the largest intron in the MYH gene, consisting of 1078 bp, exploring the possible effects of the AluYb8 insertion on gene regulation is an attractive option.

To evaluate the potential function of this variation occurring in the MYH gene, a series of experiments was performed in this study. Thus, we showed that subjects with the AluYb8MYH P/P genotype exhibited higher levels of leukocyte 8-OHdG, compared to the A/A and A/P genotypes. It is well recognized that the MYH gene initiates the BER pathway by removing adenines paired with 8-OHdG, and this offers the hOGG1 gene another chance to repair the 8-OHdG lesion [2]. Thus, a mutation in these genes is likely to impair the ability of cells to repair damaged DNA. Dherin et al. [21] reported that the wild-type hOGG1-Ser³²⁶ protein was more active than the mutant hOGG1-Cys³²⁶ protein. Because homozygous insertion of the AluYb8 sequence is associated with increased accumulation of 8-OHdG, we conclude that the AluYb8MYH mutation attenuates the ability of MYH to effect DNA repair, leading to increased levels of 8-OHdG in leukocytes and presumably other cells.

Oxidative DNA damage may be a risk factor for or a secondary consequence of age-related degenerative diseases such as diabetes mellitus, hypertension, coronary artery disease, and neurological diseases [22], [23] and [24]. The development of cancer as well as aging is thought to be caused in part by oxidative DNA injury [12], [25] and [26].

Importantly we also observed an increase in circulating IL-1 as well as increased 8-OHdG in leukocyte DNA among the subjects with the homozygous AluYb8MYH P/P genotype, suggesting that the three processes are linked, i.e., Alu mutation, oxidative DNA injury, and increased plasma IL-1 levels suggesting chronic inflammation. IL-1 is a proinflammatory cytokine released by many cells as a key mediator of the peripheral immune response during infection and inflammation. Increased levels of IL-1 may increase the potential for low-grade inflammation and be involved in the pathogenesis of insulin resistance, cardiovascular disease, and degenerative neurological diseases [10]. Thus, other studies have suggested that oxidative damage and IL-1 are involved in the pathogenesis of several age-associated degenerative diseases [27].

Significantly, chronic inflammation also leads to the formation of highly reactive oxygen and nitrogen species, which react with DNA, causing oxidative damage, and this may be important in the pathophysiology of cancer, atherosclerosis, and Alzheimer diseases, as well as aging [28], [29] and [30].

Based on our observation that the AluYb8 mutation decreased from 25.8 to 15.7% among healthy Chinese, we can conclude only that such subjects either die younger or develop chronic diseases that resulted in exclusion from cohort A. To glean some insight into this we also studied a smaller unselected cohort of 562 subjects and observed a nonsignificant decrease in P/P homozygosity with age, suggesting that maybe there is increased mortality in the P/P cohort. However, our study was not powered sufficiently to show this. When the cohorts were simply divided into 20–39 and 40–59 years age groups, the data reanalysis also showed that the P/P frequency declined with age. In addition, a small sample screening in healthy Germans also showed that the polymorphism distribution of the AluYb8 insertion was similar to that in Chinese, and the varying trend with age was maintained in the Caucasian population as well (S. Aretz, data not shown).

We recognize that the specificity of the ELISA 8-OHdG kit that we used in this study is under debate and that it might overestimate the true 8-OHdG levels. The results of the ELISA measurements represent 8-OHdG like immunoreactivity, and the use of the term 8-OHdG in our studies should be interpreted with this in mind, even though we use the term 8-OHdG for the expression of results. On the other hand, although an Alu insertion with hypermethylation was detected in the MYH gene, a direct mechanism underlying the regulation of gene expression, transcription, translation, and protein structure was not described. Our studies in healthy subjects indicate that the presence of the wild-type MYH gene is likely to be important in maintaining normal 8-OHdG and IL-1 levels and that the mutant homozygote leads to impaired gene function and increased disease susceptibility. However, the pathogenic basis of these diseases may be sought earlier in life [31]. Our small cohort study detected a consistent increase in 8-OHdG and IL-1 associated with AluYb8MYH, which played a role in the development of type 2 diabetes mellitus (data not shown). A large-scale investigation should be done to demonstrate the relationship between the AluYb8MYH allele and age-related diseases.

In summary, we first describe a common variation, the AluYb8 insertion in intron 15 of the MYH gene. The Alu retrotransposon insertion in the MYH gene might alter the gene function, possibly through DNA methylation modification. The homozygous genotype of this variation showed an age-related distribution and was associated with increased levels of 8-OHdG and IL-1 among a healthy Chinese population, suggesting that the AluYb8MYH P/P polymorphism might be a candidate modifier for common age-related diseases. We conclude that the process of aging and accelerated oxidative DNA damage are associated with the AluYb8 mutation and that screening for this genetic variation might have an important role in predicting potential longevity of subjects in China, as well as in the Western world.

Acknowledgments

We are especially grateful to Dr. Stefan Aretz and to Siegfried Uhlhaas from the Institute of Human Genetics, University of Bonn, Germany, for providing the information on screening the Alu insertion in German. This work was partly supported by the National Natural Science Foundation of China (Grant 30800546), the Doctoral Foundation of Education, Ministry of China (Grant 20070284015), and the Jiangsu Science and Technology Foundation (Grant BZ2008055).