1. Introduction
Coenzyme Q
10 (CoQ
10), which can be synthesized de novo, is a fat-soluble molecule involved in energy production and modulation of the redox state of lipid components in cells and body fluids [
1,
2,
3]. A decrease in bodily CoQ
10 levels, owing to aging or age-related neurodegenerative diseases, may lead to mitochondrial dysfunction and increased lipid peroxidation [
3]. Therefore, supplementation with CoQ
10 can be beneficial for overall health. In animal models, CoQ
10 supplementation slowed aging; reduced oxidative damage to proteins, lipids, and DNA [
4,
5,
6]; and improved oxidative stress response to exercise [
7,
8,
9], cognitive function [
10], and cognitive performance [
11,
12]. Consequently, some human interventional studies have indicated that CoQ
10 may exert beneficial effects in issues related to aging, age-related deterioration of quality of life (QOL), and degenerative disorders affecting longevity [
13,
14,
15]. However, the benefits of CoQ
10 supplementation are still under investigation, probably because of inconsistencies seen among the results of previous studies. Indeed, CoQ
10 intervention studies involving patients with Parkinson’s disease, statin-associated myalgia, and obesity, have failed to show any benefits [
16,
17,
18].
Serum CoQ
10 levels, which were increased by continuous CoQ
10 supplementation, have shown significant variance [
14,
19]. Such variance, resulting from genetic and dietary factors, may have led to the inconsistencies observed in the studies investigating the beneficial effects of CoQ
10 supplementation. To identify factors affecting serum CoQ
10 levels, we investigated dietary habits and single nucleotide polymorphisms (SNPs) in CoQ
10 and cholesterol metabolism-related genes in the participants of the Ubiquinol Health Examination [
14,
20]. Participants with higher serum CoQ
10 levels tended to consume more eggs and dairy products, although the results failed to indicate a significant difference [
21]. The SNPs found to be associated with increased serum CoQ
10 levels following 1 year of CoQ
10 supplementation in women [
22] were rs3808607 in
CYP7A1 [
23], rs2072183 in
NPC1L1 [
24,
25], rs2032582 in
ABCB1 [
26], and rs1761667 in
CD36 [
27]. Furthermore, grouping based on the above-mentioned SNPs helped identify individuals with higher CoQ
10 bioavailability following supplementation, who were also likely to exhibit the beneficial effects of CoQ
10 supplementation.
As a first step toward studying the role of SNPs in the beneficial effects of CoQ
10 supplementation, we reanalyzed their relationships using the findings of the Medical Outcome Study 36-Item Short-Form Health Survey (SF-36, subjective QOL score) [
28,
29] and the four SNPs identified in the participants of the Ubiquinol Health Examination, held from November 2013 to November 2016 at Kamijima-Cho, Ehime Prefecture, Japan. A previous study had revealed that supplementation with ubiquinol, a reduced form of CoQ
10, significantly increased certain SF-36 subscores in women, although not in men [
14]. Therefore, in the present study, we performed a two-way repeated-measures analysis of variance (ANOVA) to investigate the possible effects of any interaction between genotypes and changes in SF-36 scores following long-term supplementation with the reduced form of CoQ
10.
3. Results
Participant characteristics, including age, body mass index (BMI), serum TC levels, serum CoQ
10 levels, and SF-36 scores, at baseline and 1 year after supplementation, are shown in
Table 1. After supplementation, serum CoQ
10 levels were increased to 5.35 ± 2.12 and 5.80 ± 1.99 µmol/L in men and women, respectively. With reference to SF-36 scores in women, VT increased significantly (
p = 0.007) 1 year after supplementation. By contrast, none of the subscale or summary scores of men differed significantly (
Table 1). Therefore, we primarily analyzed the data of female participants. We noted that the results mentioned above were similar to those obtained in the original report [
14]. However, the participants of the present study and those of the previous study did not correspond completely, as some patients who participated in the current study did not participate in the previous study, and some who participated in the previous study did not participate in the current study. Approximately 50% of the male and approximately 60% of the female participants in the current study also participated in the previous study.
To assess whether each SNP affected an increase in serum CoQ
10 and changes in SF-36 scores in women supplemented with CoQ
10, we analyzed the interaction effects between the SNPs, serum CoQ
10 levels, and SF-36 scores using two-way repeated-measures ANOVA (
Table 2,
Table 3,
Table 4 and
Table 5). Female participants were divided into major homozygote and non-major homozygote (heterozygote and minor homozygote) categories for each SNP, following which serum CoQ
10 levels and SF-36 scores before and after supplementation were recorded for each group (
Table 2,
Table 3,
Table 4 and
Table 5). Because rs3808607 T (minor), rs2072183 C (major), rs2032582 G (major), and rs1761667 A (minor) are associated with the high responder (HR) of the increased CoQ
10 after 1 year of supplementation [
22], we regarded rs3808607 GT/TT, rs2072183 CC, rs2032582 GG, and rs1761667 GA/AA as the HR-associated genotype groups. The left and right columns in each table indicate HR and the low-responder (LR) of the bioavailability of CoQ
10 supplementation-associated genotypes. We also compared serum CoQ
10 levels and SF-36 scores before and after supplementation for each group, using a paired
t-test, to interpret the effect of each SNP on these parameters when interaction was indicated by the results of the two-way repeated-measures ANOVA. The three SNPs of rs3808607 (G > T) of
CYP7A1, rs2072183 (C > G) of
NPC1L1, and rs2032582 (G > T) of
ABCB1 did not interact with any analytical values upon supplementation (
Table 2,
Table 3 and
Table 4). The SNP rs1761667 (G > A) in
CD36 interacted with RP (
p = 0.016) and MH (
p = 0.017) subscale scores (
Table 5). In the HR rs1761667 GA/AA group, the RP (paired
t-test,
p = 0.003) and MH (paired
t-test,
p = 0.015) subscale scores were significantly increased following supplementation, but such a change was not observed in the LR rs1761667 GG group. Pearson’s chi-square test (
Table S2) did not detect a bias between the CoQ
10 supplement form and rs1761667 genotypes. By contrast, the four SNPs did not show any interaction with analytical values upon supplementation in men (
Tables S3–S6).
Next, we divided the women into two groups based on the four SNPs described previously [
22]. The participants belonging to group 1 carried four or more of rs3808607 T, rs2072183 C, rs2032582 G, and rs1761667 A alleles, whereas the participants belonging to group 2 carried three or fewer of these alleles. We confirmed no bias between the CoQ
10 supplement form and the grouping mentioned above via Pearson’s chi-square test (
Table S2E). Next, we investigated whether the grouping interacted with the SF-36 scores upon supplementation (
Table 6). Although interaction between the subgrouping and GH, RE, and MH scores were significant (
p = 0.045,
p = 0.008, and
p = 0.019, respectively;
Table 6). The subgrouping also interacted with serum CoQ
10 levels upon supplementation (
p = 0.008,
Table 6), demonstrating that CoQ
10 bioavailability in group 1 was higher following supplementation. Although the GH, RE, and MH subscale scores of group 1 increased significantly following the 1-year supplementation period (paired
t-test,
p = 0.042,
p = 0.016, and
p = 0.009, respectively;
Table 6), those of group 2 did not. By contrast, in men, subgrouping based on the above-mentioned alleles did not reveal interactions with any SF-36 subscale or summary score or increased serum CoQ
10 levels (
Table S7). These results suggested that the grouping based on the four SNPs may be useful for predicting higher CoQ
10 bioavailability and certain SF-36 scores, especially the subscales related to psychological parameters, in women. However, interactions between individual SNPs and SF-36 scores were minimal.
4. Discussion
In the current study, we reanalyzed the subjective QOL SF-36 scores of the Ubiquinol Health Examination to investigate whether the four SNPs, rs3808607, rs2072183, rs2032582, and rs1761667, found, respectively, in the genes
CYP7A1,
NPC1L1,
ABCB1, and
CD36, which regulate CoQ
10 bioavailability, would affect the beneficial effects of CoQ
10 supplementation. Acquisition of the SF-36 scores continued after our previous study, in which we showed that psychological QOL had increased following CoQ
10 supplementation in women [
14]. However, some participants dropped out, and others took part in the study after the results were reported. Therefore, the means and standard deviations of SF-36 scores at baseline, as well as following supplementation, in the current study are not the same as those in the previous study. In addition, although the trend in score changes upon supplementation in women in the present study was similar, only the increase in VT score was statistically significant (
Table 1). Consistent with those of the previous report, the scores did not change significantly in men.
In women, the interactions between individual SNPs and serum CoQ
10 levels, as well as SF-36 scores, were not strong (
Table 2,
Table 3,
Table 4 and
Table 5). The present study found that the four SNPs did not interact with increased serum CoQ
10 levels following supplementation (
Table 2,
Table 3,
Table 4 and
Table 5), while the interaction between rs1761667 and the RP and MH subscales was significant.
Classification based on combining the four SNPs to predict the HR/LR of the bioavailability of CoQ
10 supplementation, which we previously reported, may help find women who may easily benefit from CoQ
10 supplementation. The grouping revealed interactions with not only increases in serum CoQ
10 levels but also increases in GH, RE, and ME subscales (
Table 6). Following supplementation, significant increases in the four above-mentioned SF-36 subscales were observed in the HR-associated allele-rich group 1, but not in those of the HR-associated allele-poor group 2. Furthermore, a significant increase in the RP and VT subscale and the RCS summary score was observed following supplementation only in group 1, although no interaction was observed upon grouping. These results indicate the potential of this classification as a tool that may be used to select women who would benefit from CoQ
10 supplementation for health maintenance and promotion. These results suggested that the above-mentioned SNPs may enhance the beneficial effects of CoQ
10 supplementation, as indicated by increased SF-36 scores.
CYP7A1 and
NPC1L1 are involved in cholesterol metabolism. Both the T-allele of rs3808607 and C-allele of rs2072183, associated with the HR phenotype, were found to be involved in the insensitivity of phytosterol-dependent decrease in serum TC levels [
30,
31]. In addition, the A-allele of rs1761667, associated with the HR phenotype, increased the ratio of people exhibiting high serum cholesterol levels [
32]. Although we did not detect the differences in basal serum TC levels between each SNP subgroup, the risk of increased serum total and LDL cholesterol levels may be higher in individuals carrying the above-mentioned HR-associated alleles. This indicates that the beneficial effect of CoQ
10 supplementation may be propagated via changes in cholesterol metabolism rather than via the activation of mitochondrial respiratory chains.
As described above, rs3808607, rs2072183, and rs1761667 were associated with increased serum cholesterol levels [
30,
33,
34]. In addition, the mean age of the women included in this study was 64.3 years, suggesting that most were menopausal or postmenopausal. A previous study showed that serum cholesterol and inflammatory substances were higher, and antioxidant activities were lower, in postmenopausal women than in premenopausal women [
35]. Furthermore, compared to premenopausal women, postmenopausal women had higher stress and lower QOL scores with more plasma lipid peroxide levels [
36]. Considering these reports, it is conceivable that women in group 1, who were prone to increases in serum cholesterol levels, may be more susceptible to oxidative stress, with lower QOL scores related to the psychological parameters, than group 2 women. In addition, excess cholesterol plays a role in the pathogenesis of chronic non-communicable diseases (CNCDs), such as atherosclerosis, diabetes, chronic kidney disease, hepatic disease, Alzheimer’s disease, osteoporosis, and osteoarthritis, at least in part, via mitochondrial dysfunction-induced reactive oxygen species [
37]. Such CNCDs and their risk factors may also exert an impact on QOL [
38]. Because CoQ
10 acts not only to promote ATP synthesis in mitochondria but also to regulate mitochondrial and extramitochondrial redox homeostasis [
39], the beneficial effect of CoQ
10 supplementation on the SF-36 scores of women may materialize through the buffering of cholesterol-induced oxidative stress. Under these circumstances, the reason for the increase in SF-36 score being more preferentially observed in group 1 than in group 2 women may be attributed to differences in the bioavailability of supplemental CoQ
10 and the risk of excess serum cholesterol between these two groups. Baseline serum TC levels of the two groups were not different (
Table 6), probably because approximately 40% of female participants had been treated for dyslipidemia.
In contrast to women, the four SNPs, as well as the classification representing the combination of all four SNPs, did not interact with any SF-36 subscales or summary scores in men (
Tables S3–S7). None of the SF-36 subscales or summary scores increased upon supplementation in men. One possible reason for us being unable to observe an interaction between the four SNPs and changes in the SF-36 scores in men may be the small sample size; the number of male participants (
n = 31) was approximately 50% of that of the female participants (
n = 61) in the study. Another reason for the women-specific beneficial effects of CoQ
10 supplementation may stem from reduced estrogen level changes associated with menopause, leading to a high risk of excess serum cholesterol accumulation, resulting in CNCDs, such as cardiovascular disease [
37,
40,
41]. Serum cholesterol levels plateau in men >50 years of age [
41]. In either case, an allied study with a large sample consisting of a significant number of men may help clarify whether the four SNPs or their classification interact with the changes in SF-36 scores upon CoQ
10 supplementation.
This study has some limitations. Firstly, we evaluated only the four SNPs associated with CoQ
10 bioavailability found in our previous study [
22]. Additional SNPs that are highly suited to interact with increased serum CoQ
10 and SF-36 scores may exist. A genome-wide association study of the human SNP array may provide further information regarding other related SNPs. Secondly, this study was a single-arm and open-label study. Follow-up randomized clinical trials focusing on CoQ
10 supplementation would be required to enhance confidence in the results. Thirdly, the present study only evaluated a subjective QOL, SF-36. Further studies aimed at determining objective assessments for physical and psychological QOL, as well as at measuring oxidative stress markers and inflammatory markers, may be required to confirm the association between SNPs and the beneficial effects of CoQ
10 supplementation.
In addition, whether the classification of participants using a combination of these four SNPs would lead to substantial evidence supporting other beneficial health effects of CoQ10 remains to be investigated. Additional interventional studies combining CoQ10 and SNP genotyping may provide an answer to these issues.