Introduction

Circadian rhythms regulate multiple aspects of physiology and metabolism1. These rhythms are generated by an endogenous mechanism involved in endocrine signaling2, which is comprised of circadian clocks3. This circadian clock system has been related to metabolic processes4 and metabolic and cardiovascular disease5,6,7,8. In this regard, the role of the circadian clock system regulating adipose tissue physiology9 and therefore energy balance and glucose and lipid metabolism has been established10,11,12,13.

The control of the circadian rhythms in mammals is carried out by the Clock genes that encode proteins implicated in positive and negative regulatory pathways14. Among them, CLOCK is a key component of the molecular circadian clock, regulating the expression of an important number of transcription factors15. CLOCK has been linked to metabolism homeostasis, including lipid metabolism16. Several studies in different racial populations have associated single nucleotide polymorphisms (SNPs) in the CLOCK gene with obesity and body mass index (BMI)17,18,19,20,21 as well as with metabolic syndrome18,22 and type-2 diabetes23,24.

Other of the components of the circadian clock that has a key role in the machinery of circadian rhythms is REV-ERBα, that regulates the circadian rhythms in a negative way. REV-ERBα suppresses the expression of the main components of this machinery ARTNL/BMAL1, CLOCK and CRY114. REV-ERBα is considered to be essential regulating circadian behavior and metabolism25,26. Thus, SNPs in the REV-ERBα gene have been investigated in relationship with metabolic alterations and have been associated with obesity and body mass index in adults27,28,29 as well as in children population27,30. However, a sex-dependent association of the SNPs in the REV-ERBα gene with anthropometric variables has been described28,30.

The association of polymorphisms in CLOCK and REV-ERBα with blood lipid levels has been less studied. Studies analyzing CLOCK SNPs failed to find any association24,31,32. A study analyzing the implication of CLOCK SNPs in weight reduction reported their association with differences in plasma cholesterol in response to dietary treatment19. Studies analyzing the association of plasma lipid levels with REV-ERBα SNPs failed to find any significant association27. However, no differences between sexes were examined.

Our study aimed to evaluate the relationship of several SNPs in CLOCK (rs180126, rs4580704, rs3749474, rs3736544, rs4864548) and in REV-ERBα (rs2071427, rs2071570, rs2314339) with BMI and plasma lipid levels in 6–8-year-old boys and girls.

Results

Table 1 shows characteristics of the study participants according to sex. 637 were males and 631 were females. The mean age was 7.2 ± 0.6 years, without differences between sexes. The prevalence of obesity was similar in boys and girls (8.3% and 9.8%, respectively). Compared to males, females presented significantly higher LDL-cholesterol (110.2 ± 26.9 mg/ml vs. 107.3 ± 25.4 mg/ml, p < 0.05) and Apo B (71.4 ± 14.9 mg/ml vs. 68.8 ± 14.1 mg/ml, p < 0.01) levels, and lower Apo A-I (135.6 ± 18.9 mg/ml vs. 138.3 ± 19.1 mg/ml, p < 0.05) levels.

Table 1 Characteristics (means ± SD) of the study participants by sex.

Table 2 shows the location, minor allele frequencies and HWE p values for the studied SNPs. The minor allele frequencies for the analyzed SNPs ranged from 27.1 to 38.5% for CLOCK SNPs and from 12.3 to 28.7% for REV-ERBα SNPS. Frequencies for the studied SNPs were consistent with Hardy–Weinberg equilibrium (p > 0.05).

Table 2 Characteristics of studied SNPs in the CLOCK and REV-ERBα genes.

Associations of the studied SNPs with BMI

The analysis of the association between the studied CLOCK gene SNPs and BMI is shown in Table 3. The variants rs3749474 and rs4864548 were found to be significantly associated with BMI, in girls but no in boys. Carriers of the minor allele T of the SNP rs3749474 and the minor allele A of the SNP rs4864548 presented significantly lower BMI compared to non-carriers (p value 0.032 and 0.022, respectively). No association between BMI and the CLOCK SNPs rs1801260 and rs4580704 was found in any sex.

Table 3 BMI (kg/m2), expressed by means and standard deviation, according to the genotypes of the studied CLOCK SNPs by sex.

The analysis of the association between the analyzed REV-ERBα SNPs and BMI revealed no significant associations.

Associations of the studied SNPs with plasma lipid levels

When analyzing the association of these SNPs in CLOCK and in REV-ERBα with plasma lipid levels, we observed, in males, a significant association of the SNP in REV-ERBα rs2071570 with total cholesterol (Fig. 1a), LDL-cholesterol (Fig. 1b) and Apo B (Fig. 1c) that is not observed in females (Fig. 1d,e,f). Boys homozygous for the minor allele A showed lower total cholesterol levels compared to carriers of the major allele C (AA vs. CA p = 0.053) (Fig. 1a). Significantly lower plasma levels of LDL-cholesterol were observed in male carriers of the AA allele when comparing with carriers of the C allele (AA vs. CC p = 0.021; AA vs. CA p = 0.027) (Fig. 1b). Similar results were found for Apo B levels (Fig. 1c). AA males showed lower plasma levels of Apo B as compared with homozygous for the C allele (p = 0.009) and as compared with CA carriers (p = 0.061). No associations with plasma lipid levels were found for the rest of SNPs analyzed in CLOCK and REV-ERBα genes in any sex.

Figure 1
figure 1

Plasma lipid levels according to genotypes of the REV-ERBα rs2071570 SNP in boys (ac) and girls (df). *p value < 0.05; **p value < 0.01.

Discussion

In this study we have analyzed the potential associations of SNPs of the CLOCK and REV-ERBα genes with body mass index (BMI) and lipid levels in a cohort of boys and girls aged 6–8 years. We found sex-dependent associations of the SNPs rs3749474 and rs4864548 of CLOCK with BMI and of the SNP rs20711570 of REV-ERBα with plasma total cholesterol, LDL-cholesterol and Apo-B.

The association between SNPs of the CLOCK gene and anthropometric variables has been widely investigated. In our cohort, we found that carriers of the less common alleles for both, the rs3749474 and rs4864548 CLOCK SNPs, presented lower BMI compared to non-carriers. These SNPs, alone or combined in haplotypes, have been linked to the individual susceptibility to obesity in adults17,18,19,20,21. However, we found no association with rs1801260, one of the CLOCK SNPs most frequently studied in relationship with anthropometric variables in adult populations. Studies conducted in Caucasian populations have been able to find consistent associations of the SNP rs1801260 with obesity and BMI19,20,22,24,33,34 that are not evident in our population. Furthermore, the associations between the CLOCK SNPs and BMI found in our study are presented in girls but no in boys. A sex-dependent association between CLOCK SNPs and overweight and obesity has been previously described in adults35. We hypothesized that age-related factors may be affecting the relationship of these SNPs with anthropometric variables. It has been described that aging may result in gain or loss of rhythmic circadian expression36. In this sense, a different regulation of circadian gene expression between middle/older aged individuals and younger adults has been reported37 which may contribute to explain the described discrepancies in findings depending on age.

Regarding the relationship of CLOCK SNPs with plasma lipid levels, no consistent associations have been found in our cohort. The relationship of CLOCK SNPs with plasma lipid levels remains imprecise. No association was found in the PREDIMED trial analyzing the association of the SNP of CLOCK rs4580704 with total cholesterol, LDL-cholesterol or HDL-cholesterol24. However, a study analyzing the implication of Clock genes in weight reduction in obese patients found that CLOCK SNPs were associated with serum cholesterol changes in response to a dietary intervention19. Unlike this lack of association with lipid levels observed for CLOCK SNPs, we found an association of plasma total cholesterol, LDL-cholesterol and Apo B levels with the SNP rs20711570 in the promoter region of the REV-ERBα, which is another important regulator of metabolism. Again, we described a sex-dependent association of this SNP with lipid parameters, as we observed that male carriers of the AA genotype showed significantly lower plasma levels of total cholesterol, LDL-cholesterol and Apo B concentrations as compared with C allele carriers. However no significant differences in lipid levels among genotypes were observed in females. REV-ERBα polymorphisms have been studied in relationship with body weight regulation, but its association with lipid levels has been less explored. An association of genetic variants in the REV-ERBα gene, including the REV-ERBα SNP rs2071570, with obesity has been reported27,28,30. However, as happened with CLOCK SNPs, this association appears to be sex-dependent28. In a study including adolescents, Nascimento Ferreira et al. described a significant association between REV-ERBα SNPs and BMI only in boys30. In our study, although we failed to find any association of the REV-ERBα SNPs with BMI, we observed a sex-dependent association of the SNP rs20711570 of REV-ERBα with lipid levels, which reinforces the hypothesis of the existence of sex-related factor interacting with these Clock genes polymorphisms affecting metabolic alterations. In this sense, a sexual dimorphism in Clock genes expression in human adipose tissue has been described38. The role of sex in the different influence of circadian rhythms on cardiovascular disease in males and females has been extensively discussed39. In this regard, the sex-related differences in Clock controlled processes have been associated with suprachiasmatic nucleus (SCN) morphology and signaling40, as sex differences in SCN morphology and neuropeptide expression are known41. SCN receives both estrogenic and androgenic inputs42. These aspects may be related to the different association of the studied SNPs with anthropometric and lipid parameters observed in males and females.

More studies are needed taking into account how the possible interactions between the Clock genes polymorphisms and sex, as a biological variable, influence the development of obesity and metabolic alterations, including variations in lipid levels43.

In conclusion, in our study we report sex-dependent associations between the CLOCK SNPs rs3749474 and rs4864548 and BMI as well as between the REV-ERBα SNP rs20711570 and plasma LDL-cholesterol and Apo-B concentrations. Our findings suggest the role of the interaction between sex-related factors and Clock genes polymorphisms in obesity and plasma lipid levels.

Materials and methods

Study participants

Our population-based sample comprises a total of 1268 prepubertal children (6–8-year-old), participants in a cross-sectional study examining cardiovascular risk factors in Spanish schoolchildren44 in whom information on anthropometric variables and lipid levels was available.

The study complies with Helsinki Declaration guidelines and was approved by the Clinical Research Ethics Committee of the IIS-Fundación Jiménez Díaz (PIC016-2019 FJD). All parents provided written informed consent for their children to participate in the study. Children reported by their parents to be suffering from chronic diseases were excluded of the study.

Anthropometric measurements

Weight and height were measured with children wearing light clothing and barefoot. Height was measured to the millimeter using a portable stadiometer. Weight was recorded to the nearest 0.1 kg using a standard electronic digital scale. The classification by obesity category was carried out according to the cut-off points established for children by Cole et al.45.

Biochemical determinations

Fasting (12 h) blood samples were obtained by venipuncture. Serum and plasma samples were separated by centrifugation and stored at − 70 °C. Biochemical determinations were performed as previously described44. Cholesterol and triglyceride (TG) concentrations were measured enzymatically (Menarini Diagnostics, Florence, Italy) in a RA-1000 Autoanalyzer (Technicon Ltd., Dublin, Ireland). High-density lipoprotein cholesterol (HDL-cholesterol) was measured after precipitation of apolipoprotein B-containing lipoproteins with phosphotungstic acid and Mg (Roche Diagnostics, Madrid, Spain). Plasma apolipoprotein A-I (Apo A-I) and apolipoprotein B (Apo B) concentrations were measured by immunonephelometry (Dade Berhing, Frankfurt, Germany). Low-density lipoprotein cholesterol (LDL-cholesterol) was calculated according to the Friedewald formula.

SNP genotyping

The SNPs of CLOCK rs1801260, rs4580704, rs3749474, rs3736544 andrs4864548 and the SNPs of REV-ERBα rs2017427, rs20711570 and rs2314339 SNPs were selected to be studied in our population based on the results of previous studies reporting association of these SNPs with anthropometric variables. Genotyping was carried by RT-PCR using a 7500 Fast Real-Time System (Applied Biosystems) and performed with 10 ng of genomic DNA. Samples were cycled under the following recommended conditions: 95 °C for 10 min, 95 °C for 15 s, and 60 °C for 1 min, repeated over 40 cycles.

The SNPs of CLOCK rs1801260, rs4580704, rs3749474, rs3736544 and rs4864548 were genotyped using the C_8746719_20, C_28028791_10, C_26405955_10, C_22273263_10 and C_11821276_10 Applied Biosystems predesigned allelic discrimination TaqMan® assays. The SNPs of REV-ERBα rs2017427, rs20711570 and rs2314339 using the TaqMan® assays C_7479334_10, C_7479332_10, and C_177490_10 respectively (Applied Biosystems).

Statistical analyses

Participant characteristics were described by means and standard deviations for continuous variables and counts and percentages for categorical variables. Comparisons between boys and girls were performed using the Student’s t test. Hardy–Weinberg deviations for each of the studied SNPs were tested by a chi-squared goodness-of-fit test. The normality of the distribution of the variables under study was examined using the Kolmogorov–Smirnov test. Differences in mean values for the variables under study according to the different genotypes of each of the SNPs studied were tested individually using the linear regression models taking the more common genotype for each polymorphism as reference. Means and standard deviations of the analyzed variables corresponding to each genotype are shown in Table 3 and means and standard error are shown in Fig. 1. Analyses were performed separately in males and females. Statistical analysis was performed using the R software, 4.1.2 (R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria) and GraphPad Prism statistical software (San Diego, CA, USA, Version 8).