Abstract
Context
Selenoprotein P is a hepatokine associated with several metabolic processes. Rs7579 (C > T) is a SeP-related functional single nucleotide polymorphism.
Objective
In this study, we aimed to identify the environmental factors affecting the relationship between rs7579 and metabolic diseases, such as metabolic dysfunction-associated steatotic liver disease, in the general population.
Methods
This cross-sectional study was based on the Shika Study, a survey of residents in the Noto Peninsula of Ishikawa Prefecture. We analyzed a total of 900 adults, measuring full-length selenoprotein P (FL-SeP) serum levels using a sol-particle homogeneous immunoassay.
Results
We observed that selenium and FL-SeP serum levels were associated with dyslipidemia. In males, serum selenium was associated with dyslipidemia and hepatic steatosis. However, in females, FL-SeP tended to be associated with diabetes. Participants carrying the TT genotype and hepatic steatosis exhibited higher levels of liver enzymes, insulin, the homeostatic model assessment of insulin resistance (HOMA-IR), and the homeostasis model assessment of β-cell function than those without hepatic steatosis or with other genotypes. In females carrying the TT genotype of rs7579, hepatic steatosis, hypertension, diabetes, obesity, and metabolic syndrome were associated with higher HOMA-IR levels.
Conclusion
In this study, we revealed that the association between metabolic diseases and HOMA-IR differed single nucleotide polymorphism genotype and sex dependently. In females carrying the TT genotype of rs7579, hepatic steatosis-associated metabolic disorders (diabetes, hypertension, obesity, and metabolic syndrome) were associated with higher HOMA-IR. The results of this study open the way to genetic signatures-based personalized preventive medicines.
Keywords: selenoprotein P, rs7579, metabolic syndrome, MASLD, SNPs, CCDC152
Selenoprotein P (SeP; encoded by SELENOP in humans) is a hepatokine produced mainly in the liver and is associated with several metabolic processes as a major selenium transporter in the body [1]. SeP protein is upregulated by selenium supply, followed by incorporation into selenocysteine. In addition, SeP is regulated at gene expression levels by nutrients, insulin, metformin, and virus infection through the transcription factors, such as FoxOs, SREBP-1c, HNF-4a, and C/EBPa [2-7]. SeP scavenges reactive oxygen species, thereby inducing diverse types of signal transduction resistance such as insulin resistance in the liver and skeletal muscle [8] as well as impaired insulin secretion [9], angiogenesis resistance [10], exercise resistance [11], and thermogenesis resistance [12] through reductive stress [11, 13], leading to diabetes pathology. SeP serum levels increase during aging [14], type 2 diabetes [8], prediabetes [15], and metabolic dysfunction-associated steatotic liver disease (MASLD) [16, 17].
We developed a sol-particle homogeneous immunoassay method that selectively measures full-length selenoprotein P (FL-SeP) in the human serum [17, 18]. Circulating FL-SeP is positively associated with exercise resistance [11], impaired thermogenesis [12], and the future onset of glucose intolerance [14].
A functional single nucleotide polymorphism (SNP) rs7579 (C > T) in the 3′untranslated region of SELENOP is reportedly associated with increased plasma SeP [19] and selenium [20], SELENOP mRNA [19, 21], and insulin [22] levels as well as with metabolic syndrome (MetS) risk [23]. However, the metabolic dysfunction-related gene-environment interactions remain underrecognized. In this study, we aimed to reveal the environmental factors affecting the relationship between rs7579 and metabolic diseases (eg, MASLD) in the general population.
Materials and Methods
Study Population
This cross-sectional study was based on the Shika Study, a survey of residents of Shika, a town with a population of approximately 20 000 people, located in a rural area in the Noto Peninsula of Ishikawa Prefecture, Japan. The Shika Study is an ongoing population-based survey that began in 2011 with the goal of developing advanced preventive methods for lifestyle-related diseases. In the present study, the target population involved adult residents aged at least 40 years. We used data from self-administered questionnaires and comprehensive health examinations between 2013 and 2019. All participants provided written informed consent to participate in the study. We conducted our study in accordance with the principles of the Declaration of Helsinki. Our study protocol was approved by the Ethics Committee for Human Research of Kanazawa University Hospital (1491, 326). This study was registered in the University Hospital Medical Information Network (UMIN) Clinical Trials Registry as UMIN 000024915.
In this study, we enrolled a total of 1335 adult participants aged ≥40 years living in the model districts. Of them, 434 disapproved of genome analysis or failed to pass the quality control (QC) mainly due to cryptic relatedness. We exclude 1 participant due to a lack of physical and blood sampling data. Finally, we analyzed 900 individuals (Fig. 1).
Figure 1.
Study design and study participant flowcharts.
Measurements
We assessed the age, sex, height, weight, waist circumference, and systolic and diastolic blood pressures in the health checkups of all participants. We calculated the body mass index (BMI) as weight (kg) divided by height (m2). We defined hypertension as a mean systolic blood pressure of ≥140 mmHg or a mean diastolic blood pressure of ≥90 mmHg at health checkups or a history or previous treatment of hypertension in the questionnaire [24]. We defined participants with a BMI of ≥25 as obese [25].
Blood Sample Collection and Assays
We collected fasting blood samples from each participant between 0800 and 1200 from the forearm vein after an overnight fast. We delivered the serum samples to Kanazawa University via a commercial laboratory (SRL Kanazawa Laboratory, Kanazawa, Japan). The sera were frozen and stored at −30 °C until the assay. We specifically measured the serum FL-SeP concentrations using a sol-particle homogeneous immunoassay with 2 types of SeP monoclonal antibodies (catalog # AA3, RRID: AB_3712444, catalog # AH5, RRID: AB_3712445, and catalog # Human Selenoprotein P ELISA (FL-SeP SPIA), AA3 and AH5, RRID: AB_3712623), as described previously [17, 18, 26]. We measured serum selenium concentrations using atomic absorption spectrophotometry [27].
We applied the homeostatic model assessment of insulin resistance (HOMA-IR) and the homeostasis model assessment of β-cell function (HOMA-β) to evaluate insulin resistance and secretory capacity, respectively [28-30]. We performed a detailed calculation using the following formula: HOMA-IR = [fasting insulin (μU/mL) × fasting glucose (mg/dL)]/405, and HOMA-β = [fasting insulin (μU/mL) × 360]/[fasting glucose (mg/dL) − 63].
We defined diabetes mellitus as hemoglobin A1c (HbA1c) of ≥6.5% or fasting plasma glucose ≥126 mg/dL based on blood sampling or a history of diabetes mellitus under treatment in the questionnaire.
We defined dyslipidemia as a fasting triglyceride concentration of ≥150 mg/dL, a high-density lipoprotein cholesterol (HDL-C) concentration of <40 mg/dL, a Friedewald-estimated low-density lipoprotein cholesterol (LDL-C) concentration of ≥140 mg/dL [31], or a history or previous treatment of dyslipidemia in the questionnaire.
We calculated the fibrosis-4 index (Fib-4) using the following formula: age(years) × aspartate aminotransferase [U/L]/(platelets [109/L] × (alanine aminotransferase [U/L])1/2) [32].
MetS
Although multiple MetS definitions have been used worldwide, we defined MetS as visceral obesity (waist circumference ≥85 cm in males and ≥90 cm in females) and at least 2 of the following 3 metabolic abnormalities: hypertension (systolic blood pressure ≥130 mmHg and/or diastolic blood pressure ≥85 mmHg and/or receiving hypertension treatment), dyslipidemia (fasting triglycerides ≥150 mg/dL, HDL-C <40 mg/dL, and/or receiving dyslipidemia treatment), and impaired glucose tolerance (fasting plasma glucose ≥110 mg/dL and/or receiving diabetes treatment) [33-35].
Hepatic Steatosis Assessment
We determined hepatic steatosis using B-mode ultrasonography performed by experienced hepatologists. We defined hepatic steatosis based on the presence of at least 1 of the following symptoms: increased hepatorenal contrast, liver brightness, deep attenuation, or vascular blurring [36].
MASLD Definition
We defined MASLD based on the diagnostic criteria proposed in the multisociety Delphi consensus statement [37]. Participants with no history of moderate or greater alcohol consumption (>30 g/day in males and >20 g/day in females), hepatic steatosis on abdominal ultrasound, and at least 1 of the following 5 cardiovascular metabolic risk factors are classified as having MASLD: (1) BMI ≥23 kg/m2 or waist circumference >94 cm in males or >80 cm in females; (2) fasting serum glucose ≥100 mg/dL or 2-hour postload plasma glucose levels ≥140 mg/dL or HbA1c ≥5.7% or prevalence of type 2 diabetes or treatment for type 2 diabetes; (3) blood pressure ≥130/85 mmHg or antihypertensive drug treatment; (4) triglycerides ≥150 mg/dL or lipid-lowering treatment; or (5) HDL-C ≤40 mg/dL in males or ≤50 mg/dL in females or lipid-lowering treatment.
Nutrient Assessment
We used a simplified self-administered dietary history questionnaire (BDHQ), developed in Japan, to assess nutritional intake. The BDHQ is a dietary history questionnaire that evaluates the intake frequency of 58 foods and beverages consumed by an average Japanese person in the preceding month. BDHQ reproducibility and validity have been described in previous studies [38]. We excluded from the analysis daily energy intakes considered either extremely low or extremely high, ie, below 600 kcal (corresponding to 50% of the required energy intake for the lowest physical activity category) or ≥4000 kcal (1.5 times higher than the energy intake required for the highest physical activity category). We also assessed alcohol intake from the BDHQ.
Smoking
We stratified the smoking status as nonsmokers, past smokers, and current smokers.
Genetic Analyses
We extracted genomic DNA from the blood samples using the QIAamp DNA Blood Maxi Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions or commissioned a company specialized in clinical laboratory testing to perform the necessary tasks (SRL Inc., Tokyo, Japan). We performed genome-wide SNP genotyping using Japonica Array v2 [39] (TOSHIBA Co., Ltd., Tokyo, Japan). The details of the QC and genotype imputation procedures for the array data obtained have been described in a previous study [40]. Briefly, the QC filtering of SNPs and participants was based on sex identities between karyotypes and questionnaire findings, call rate, Hardy–Weinberg equilibrium test, inbreeding coefficient, cryptic relatedness, and population structure, using PLINK 1.09 [41] and EIGENSOFT 7.2.1 [42]. We performed the genotype imputation using Beagle 4.1 [43] and 1000 Genomes Phase 3 [44] as a reference panel. In this study, we extracted the genotypes of rs7579 for the 900 unrelated participants who passed the QC (based on genome-wide π ^ values) from the postimputation data. The estimated allelic squared correlation of rs7579 was 1.00, suggesting the very high accuracy of the genotype imputation for this SNP.
Statistical Analysis
We presented the continuous variables as the means ± SD and the discrete variables as frequencies and percentages. We examined the baseline characteristics-related differences between males and females using Student's or Welch's t-test. We compared obesity, MetS, hypertension, dyslipidemia, diabetes, and hepatic steatosis prevalences as well as smoking rates between males and females using Pearson’s chi-square test. We analyzed the association between selenium or FL-SeP serum levels and the previously described metabolic diseases using multiple logistic regression with the presence of disease as the dependent variable and selenium and FL-SeP as independent variables. Multiple logistic regression was adjusted for age, sex, BMI, alcohol intake, and smoking. We analyzed the association between rs7579 and selenium and SeP concentrations using a 1-way analysis of covariance (ANCOVA) adjusted for age, sex, BMI, alcohol intake, and smoking. We assessed the association between metabolic diseases and the rs7579 genotype by multiple logistic regression adjusted for age, sex, and BMI. We performed 2-way ANCOVA with physiological and biochemical data as well as nutrient intake as dependent variables and hepatic steatosis and the rs7579 genotype as fixed factors, adjusted for age, sex, and BMI. If the interaction was significant in the 2-way ANCOVA, we performed a post hoc simple main effect test with Bonferroni correction. We used an additive model for rs7579 in this study.
We considered P-values of P < .05 statistically significant. We performed all the statistical analyses using the Statistical Package for the Social Sciences software version 25.0 (IBM).
Results
Participant Characteristics
We enrolled a total of 1335 participants in the study, of whom we included 900 persons with available SNP analyses (Fig. 1). In total, 45.3% of the participants were males with a mean age of 61.8 ± 10.8 years and a mean BMI of 23.3 ± 3.3. Of the total participants, 58.1%, 34.9%, 53.9%, 12.9%, and 17% exhibited hypertension, hepatic steatosis, dyslipidemia, diabetes, and MetS, respectively. We observed sex-related differences in lipid, liver enzyme, blood glucose, HbA1c, HOMA-β, and serum FL-SeP levels; energy, protein, fat, and alcohol intake; and age, BMI, waist circumference, blood pressure, and smoking rate (Table 1).
Table 1.
Baseline characteristics of the participants
| Alla | Females | Males | |||||
|---|---|---|---|---|---|---|---|
| n | Mean ± SD | n | Mean ± SD | n | Mean ± SD | P-valueb | |
| Age (years) | 900 | 61.8 ± 10.8 | 492 | 60.8 ± 10.7 | 408 | 62.9 ± 10.7 | .004 |
| BMI (kg/m2) | 900 | 23.3 ± 3.3 | 492 | 22.7 ± 3.3 | 408 | 24.1 ± 3.0 | <.001 |
| WC (cm) | 899 | 84.0 ± 9.1 | 491 | 82.2 ± 9.0 | 408 | 86.2 ± 8.7 | <.001 |
| Obesity (n, %) | 900 | 249, 27.7 | 492 | 111, 22.6 | 408 | 138, 33.8 | <.001 |
| MetS (n, %) | 898 | 153, 17.0 | 490 | 42, 8.6 | 408 | 111, 27.2 | <.001 |
| Systolic BP (mmHg) | 900 | 138.0 ± 19.3 | 492 | 135.0 ± 19.4 | 408 | 141.5 ± 18.5 | <.001 |
| Diastolic BP (mmHg) | 900 | 80.1 ± 11.3 | 492 | 78.3 ± 10.4 | 408 | 82.2 ± 11.9 | <.001 |
| Hypertension (n, %) | 900 | 523, 58.1 | 492 | 239, 48.6 | 408 | 284, 69.6 | <.001 |
| HDL-C (mg/dL) | 821 | 65.3 ± 17.3 | 449 | 69.9 ± 17.7 | 372 | 59.8 ± 15.0 | <.001 |
| Triglycerides (mg/dL) | 821 | 116.7 ± 74.2 | 449 | 104.3 ± 59.7 | 372 | 131.7 ± 86.3 | <.001 |
| LDL-C (mg/dL) | 821 | 127.8 ± 33.8 | 449 | 130.8 ± 35.1 | 372 | 124.1 ± 31.8 | .005 |
| Dyslipidemia (n, %) | 900 | 485, 53.9 | 492 | 271, 55.1 | 408 | 214, 52.5 | .431 |
| GOT (IU/L) | 821 | 24.3 ± 9.6 | 449 | 22.6 ± 7.0 | 372 | 26.5 ± 11.6 | <.001 |
| GPT (IU/L) | 821 | 22.4 ± 13.0 | 449 | 19.0 ± 8.8 | 372 | 26.5 ± 15.8 | <.001 |
| γ-GTP (IU/L) | 821 | 42.1 ± 53.2 | 449 | 27.9 ± 26.3 | 372 | 59.1 ± 69.9 | <.001 |
| Plt (104/μL) | 821 | 24.5 ± 6.1 | 449 | 25.2 ± 6.4 | 372 | 23.6 ± 5.6 | <.001 |
| Fib-4 index | 820 | 1.49 ± 0.92 | 448 | 1.44 ± 1.03 | 372 | 1.55 ± 0.78 | .089 |
| Hepatic steatosis (n, %) | 816 | 285, 34.9 | 443 | 140, 31.6 | 373 | 145, 38.9 | .032 |
| MASLD (n, %) | 503 | 173, 34.4 | 312 | 214, 31.4 | 191 | 75, 39.3 | .072 |
| FPG (mg/dL) | 822 | 96.9 ± 18.0 | 450 | 93.4 ± 14.4 | 372 | 101.1 ± 20.8 | <.001 |
| HbA1c (%) | 822 | 5.91 ± 0.64 | 450 | 5.85 ± 0.59 | 372 | 5.97 ± 0.69 | .001 |
| Diabetes mellitus (n, %) | 900 | 116, 12.9 | 492 | 39, 7.9 | 408 | 77, 18.9 | <.001 |
| HOMA-IR | 816 | 1.32 ± 1.02 | 447 | 1.28 ± 0.96 | 369 | 1.38 ± 1.08 | .156 |
| HOMA-β (%) | 816 | 63.5 ± 42.1 | 447 | 67.9 ± 41.1 | 369 | 58.1 ± 42.7 | <.001 |
| Insulin (μU/mL) | 816 | 5.4 ± 3.5 | 447 | 5.4 ± 3.3 | 369 | 5.4 ± 3.7 | .868 |
| Smoking status | 900 | 492 | 408 | <.001 | |||
| Nonsmoker (n, %) | 480, 53.3 | 398, 80.9 | 82, 20.1 | ||||
| Past smoker (n, %) | 259, 28.8 | 51, 10.4 | 208, 51.0 | ||||
| Current smoker (n, %) | 161, 17.9 | 43, 8.7 | 118, 28.9 | ||||
| Serum selenium (μg/L) | 514 | 159.0 ± 28.7 | 340 | 156.7 ± 32.0 | 267 | 161.4 ± 24.4 | .061 |
| FL-SeP (μg/mL) | 571 | 3.97 ± 0.96 | 340 | 3.85 ± 1.03 | 267 | 4.11 ± 0.84 | <.001 |
| Energy (kcal) | 696 | 1827.0 ± 605.0 | 380 | 1620.2 ± 511.7 | 316 | 2075.8 ± 615.4 | <.001 |
| Protein (% of energy) | 696 | 15.3 ± 3.3 | 380 | 16.0 ± 3.3 | 316 | 14.5 ± 3.0 | <.001 |
| Fat (% of energy) | 696 | 25.0 ± 6.1 | 380 | 26.7 ± 5.7 | 316 | 22.9 ± 6.0 | <.001 |
| Carbohydrate (% of energy) | 696 | 53.8 ± 8.5 | 380 | 54.2 ± 8.2 | 316 | 53.2 ± 8.9 | .136 |
| Alcohol (g/day) | 696 | 12.7 ± 21.3 | 380 | 4.4 ± 10.8 | 316 | 22.7 ± 26.0 | <.001 |
| Alcohol (% of energy) | 696 | 4.5 ± 7.2 | 380 | 2.0 ± 4.7 | 316 | 7.6 ± 8.3 | <.001 |
Abbreviations: BMI, body mass index; BP, blood pressure; Fib-4 index, Fibrosis-4 index; FL-SeP, full-length selenoprotein P; FPG, fasting plasma glucose; GOT, glutamic oxaloacetic transaminase; γ-GTP, gamma-glutamyl transpeptidase; HbA1c, hemoglobin A1c; HDL-C, high density lipoprotein cholesterol; HOMA-β, homeostasis model assessment of β-cell function; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-C, low density lipoprotein cholesterol; MetS, metabolic syndrome; Plt, platelet; WC, waist circumference.
aAll values are means ± SDs.
bStudent's t-test or Welch's t-test were used to compare all continuous variables. Pearson's chi-square test was used to compare the prevalence of obesity, MetS, hypertension, hepatic steatosis, dyslipidemia, diabetes mellitus, and smoking status.
Relationship Between Serum Selenium and FL-SeP Levels and Metabolic Diseases
We observed a positive correlation between the selenium and FL-SeP serum levels (R = 0.725, P < .001). Serum levels of SeP showed a positive correlation with age in all participants (R = 0.156, P < .001). Our multiple logistic regression analyses (adjusted for age, sex, and BMI) demonstrated that selenium [odds ratio (OR) = 1.011, 95% confidence interval (CI) = 1.003-1.019, P = .005] and FL-SeP (OR = 1.328, 95% CI = 1.060-1.663, P = .014) serum levels were associated with dyslipidemia (model 1, Table 2). We obtained similar results when adjusted for age, sex, BMI, alcohol intake, and smoking (model 2, Table 2). In males, the serum selenium concentration was associated with dyslipidemia (OR = 1.013, 95% CI = 1.001-1.024, P = .027) and hepatic steatosis in both model 1 (OR = 1.014, 95% CI = 1.001-1.027, P = .038) and model 2. However, in females, FL-SeP tended to be associated with diabetes in model 1 (OR = 1.343, 95% CI = 0.997-1.808, P = .052) and model 2 (OR = 1.329, 95% CI = 0.985-1.825, P = .062) (Table 2).
Table 2.
Association between the serum FL-SeP level and selenium concentrations and metabolic diseases
| Selenium | FL-SeP | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 1 | Model 2 | ||||||||||
| Dependent variable | OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | |
| All | Hypertension | 1.007 | 0.999-1.015 | .080 | 1.004 | 0.996-1.012 | .315 | 1.244 | 0.984-1.573 | .068 | 1.149 | 0.913-1.446 | .238 |
| Dyslipidemia | 1.011 | 1.003-1.019 | .005 | 1.012 | 1.004-1.020 | .005 | 1.328 | 1.060-1.663 | .014 | 1.349 | 1.068-1.703 | .012 | |
| Diabetes mellitus | 1.005 | 0.997-1.013 | .246 | 1.006 | 0.998-1.014 | .168 | 1.136 | 0.906-1.424 | .270 | 1.140 | 0.902-1.440 | .273 | |
| Obesity | 0.993 | 0.985-1.002 | .118 | 0.994 | 0.985-1.002 | .152 | 1.046 | 0.858-1.276 | .656 | 1.071 | 0.877-1.307 | .501 | |
| Hepatic steatosis | 1.007 | 0.999-1.014 | .081 | 1.007 | 0.999-1.015 | .073 | 1.141 | 0.915-1.422 | .241 | 1.170 | 0.937-1.462 | .167 | |
| MetS | 1.005 | 0.993-1.017 | .396 | 1.001 | 0.987-1.014 | .927 | 1.253 | 0.915-1.717 | .160 | 1.089 | 0.765-1.552 | .636 | |
| Males | Hypertension | 1.006 | 0.994-1.018 | .310 | 0.998 | 0.985-1.011 | .746 | 1.481 | 0.972-2.257 | .068 | 1.228 | 0.798-1.890 | .351 |
| Dyslipidemia | 1.013 | 1.001-1.024 | .027 | 1.013 | 1.001-1.026 | .028 | 1.285 | 0.924-1.787 | .136 | 1.263 | 0.900-1.771 | .176 | |
| Diabetes mellitus | 1.002 | 0.988-1.015 | .813 | 1.003 | 0.989-1.017 | .668 | 0.903 | 0.624-1.307 | .588 | 0.904 | 0.618-1.332 | .601 | |
| Obesity | 0.995 | 0.984-1.006 | .375 | 0.994 | 0.983-1.006 | .361 | 1.281 | 0.950-1.728 | .104 | 1.356 | 0.993-1.853 | .056 | |
| Hepatic steatosis | 1.014 | 1.001-1.027 | .038 | 1.016 | 1.002-1.030 | .029 | 1.106 | 0.791-1.546 | .556 | 1.176 | 0.826-1.674 | .369 | |
| MetS | 1.008 | 0.994-1.022 | .251 | 1.001 | 0.986-1.016 | .890 | 1.362 | 0.911-2.034 | .132 | 1.105 | 0.727-1.678 | .641 | |
| Females | Hypertension | 1.007 | 0.996-1.018 | .220 | 1.007 | 0.996-1.018 | .227 | 1.120 | 0.853-1.471 | .415 | 1.094 | 0.836-1.432 | .512 |
| Dyslipidemia | 1.007 | 0.997-1.018 | .175 | 1.008 | 0.997-1.019 | .161 | 1.263 | 0.930-1.715 | .135 | 1.329 | 0.957-1.847 | .090 | |
| Diabetes mellitus | 1.006 | 0.996-1.016 | .257 | 1.006 | 0.996-1.017 | .231 | 1.343 | 0.997-1.808 | .052 | 1.341 | 0.985-1.825 | .062 | |
| Obesity | 0.989 | 0.976-1.003 | .110 | 0.992 | 0.978-1.005 | .227 | 0.826 | 0.583-1.170 | .281 | 0.832 | 0.582-1.190 | .313 | |
| Hepatic steatosis | 1.003 | 0.991-1.014 | .662 | 1.002 | 0.991-1.014 | .713 | 1.168 | 0.871-1.565 | .299 | 1.166 | 0.867-1.569 | .310 | |
| MetS | 0.995 | 0.966-1.025 | .725 | 0.995 | 0.966-1.026 | .757 | 1.092 | 0.566-2.108 | .793 | 0.976 | 0.460-2.123 | .976 | |
P-values < .05 are in bold.
Model 1: Multiple logistic regression analysis adjusted by age, sex, and body mass index.
Model 2: Multiple logistic regression analysis adjusted by age, sex, body mass index, alcohol intake, and smoking.
Abbreviations: CI, confidence interval; FL-SeP, full-length selenoprotein P; MetS, metabolic syndrome; OR, odds ratio.
SELENOP SNP rs7579 Analysis
The T allele frequency of rs7579 was 0.35, similar to the data of the SNP database dbSNP and the Japanese Multi-Omics Reference Panel 54KJPN (jMorp, https://jmorp.megabank.tohoku.ac.jp/). The rs7579 genotypes were not associated with either serum selenium and FL-SeP concentrations (Table S1) [45] or metabolic diseases (hypertension, dyslipidemia, diabetes, obesity, hepatic steatosis, or MetS) (Table S2) [45].
Next, we investigated whether any environmental factors would interact with the relationships between rs7579 and metabolic diseases. As environmental factors, we examined biochemical [glutamic oxaloacetic transaminase (GOT), glutamic pyruvic transaminase (GPT), gamma-glutamyl transpeptidase(γ-GTP), fasting blood glucose, HbA1c, insulin, HOMA-IR, HOMA-β, triglycerides, HDL-C, LDL-C, and serum selenium and FL-SeP levels] and physiological (BMI, waist circumference, and systolic and diastolic blood pressure) data as well as BDHQ-based nutritional (energy, lipid, carbohydrate, and protein) intake.
First, we examined the environmental factors interacting with hepatic steatosis and rs7579 (Table 3). We observed that the rs7579 genotype mainly affected the GOT, GPT, insulin, HOMA-IR, and HOMA-β levels as well as the energy intake. Selenium, triglyceride, LDL-C, GOT, GPT, γ-GTP, insulin, HOMA-IR, and HOMA-β levels displayed an interaction between rs7579 and hepatic steatosis. In the case of TT but not other genotypes, serum selenium was associated with hepatic steatosis. Participants carrying the TT genotype with hepatic steatosis exhibited higher liver enzyme, insulin, HOMA-IR, and HOMA-β levels than those without hepatic steatosis or those with other genotypes (Table 3). Upon sex-based stratification, TT genotype-carrier participants with hepatic steatosis were associated with increased liver enzyme levels in males and elevated insulin and HOMA-IR levels in females (Table S3) [45].
Table 3.
Clinical parameters of participants according to the presence of hepatic steatosis and SELENOP gene polymorphism (rs7579) genotype
| CC (n = 337) | CT (n = 378) | TT (n = 101) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Clinical parameter | Hepatic steatosis | Mean | SD | Mean | SD | Mean | SD | P-value for hepatic steatosisa | P-value for rs7579a | P-value for interactiona | P-value for CCb | P-value for CTb | P-value for TTb | |||
| BMI (kg/m2) | − | 22.4 | ± | 2.8 | 22.3 | ± | 3.0 | 22.5 | ± | 2.8 | <.001 | .811 | .881 | |||
| + | 25.3 | ± | 3.3 | 25.1 | ± | 3.2 | 25.0 | ± | 3.8 | |||||||
| WC (cm) | − | 81.7 | ± | 8.3 | 81.5 | ± | 8.5 | 81.7 | ± | 7.9 | <.001 | .655 | .634 | |||
| + | 88.5 | ± | 8.8 | 89.2 | ± | 8.9 | 87.6 | ± | 10.0 | |||||||
| Systolic BP (mmHg) | − | 137.4 | ± | 19.5 | 137.7 | ± | 20.1 | 134.3 | ± | 15.7 | .247 | .388 | .094 | |||
| + | 140.6 | ± | 18.9 | 137.2 | ± | 17.4 | 141.2 | ± | 17.9 | |||||||
| Diastolic BP (mmHg) | − | 79.5 | ± | 11.4 | 78.7 | ± | 11.6 | 78.7 | ± | 10.1 | .656 | .576 | .708 | |||
| + | 83.0 | ± | 11.4 | 81.2 | ± | 9.9 | 83.2 | ± | 11.8 | |||||||
| Triglycerides (mg/dL) | − | 101.2 | ± | 53.2 | 107.7 | ± | 71.4 | 102.6 | ± | 47.5 | <.001 | .136 | .035 | .003 | .108 | <.001 |
| + | 140.4 | ± | 101.8 | 134.4 | ± | 76.1 | 171.4 | ± | 92.7 | |||||||
| HDL-C (mg/dL) | − | 67.2 | ± | 15.9 | 68.6 | ± | 17.8 | 67.7 | ± | 17.5 | .030 | .614 | .687 | |||
| + | 59.6 | ± | 14.3 | 60.9 | ± | 15.2 | 57.7 | ± | 21.1 | |||||||
| LDL-C (mg/dL) | − | 127.5 | ± | 32.2 | 121.2 | ± | 30.8 | 127.7 | ± | 29.3 | .572 | .867 | .038 | .912 | .005 | .482 |
| + | 131.1 | ± | 43.1 | 136.4 | ± | 34.1 | 125.8 | ± | 36.5 | |||||||
| GOT (IU/L) | − | 23.4 | ± | 7.2 | 23.7 | ± | 6.8 | 22.7 | ± | 4.9 | <.001 | .001 | <.001 | .004 | .099 | <.001 |
| + | 25.8 | ± | 10.8 | 24.7 | ± | 9.2 | 33.9 | ± | 26.7 | |||||||
| GPT (IU/L) | − | 19.0 | ± | 8.0 | 19.6 | ± | 8.7 | 19.4 | ± | 7.4 | <.001 | .022 | .002 | <.001 | .002 | <.001 |
| + | 30.0 | ± | 19.4 | 25.9 | ± | 13.9 | 34.7 | ± | 22.5 | |||||||
| γ-GTP (IU/L) | − | 41.0 | ± | 53.5 | 35.9 | ± | 31.0 | 32.2 | ± | 30.4 | <.001 | .079 | .008 | .156 | .221 | <.001 |
| + | 45.0 | ± | 57.0 | 39.2 | ± | 34.4 | 47.3 | ± | 91.2 | |||||||
| Fib-4 index | − | 1.58 | ± | 0.91 | 1.58 | ± | 0.64 | 1.49 | ± | 0.61 | .625 | .473 | .117 | |||
| + | 1.27 | ± | 0.69 | 1.29 | ± | 0.62 | 1.43 | ± | 1.05 | |||||||
| FPG (mg/dL) | − | 94.8 | ± | 14.8 | 95.1 | ± | 17.6 | 94.2 | ± | 12.2 | .003 | .794 | .686 | |||
| + | 101.2 | ± | 19.3 | 99.2 | ± | 22.7 | 100.6 | ± | 18.6 | |||||||
| HbA1c (%) | − | 5.85 | ± | 0.61 | 5.85 | ± | 0.52 | 5.79 | ± | 0.44 | .004 | .761 | .926 | |||
| + | 6.02 | ± | 0.65 | 6.03 | ± | 0.80 | 5.98 | ± | 0.60 | |||||||
| Insulin (μU/mL) | − | 4.3 | ± | 2.4 | 4.5 | ± | 2.7 | 4.8 | ± | 2.8 | <.001 | <.001 | .001 | <.001 | .255 | <.001 |
| + | 7.1 | ± | 3.9 | 6.3 | ± | 3.5 | 8.8 | ± | 6.7 | |||||||
| HOMA-IR | − | 1.02 | ± | 0.59 | 1.08 | ± | 0.71 | 1.12 | ± | 0.72 | <.001 | .002 | .003 | <.001 | .109 | <.001 |
| + | 1.81 | ± | 1.16 | 1.59 | ± | 1.13 | 2.28 | ± | 2.20 | |||||||
| HOMA-β (%) | − | 55.4 | ± | 33.8 | 56.5 | ± | 35.6 | 59.4 | ± | 34.2 | .016 | .004 | .027 | .341 | .773 | .005 |
| + | 74.8 | ± | 47.3 | 70.4 | ± | 39.3 | 96.0 | ± | 84.5 | |||||||
| Serum selenium (μg/L) | − | 158.5 | ± | 39.0 | 156.2 | ± | 22.7 | 154.2 | ± | 19.6 | .009 | .288 | .044 | .878 | .115 | .008 |
| + | 154.4 | ± | 25.5 | 159.7 | ± | 23.1 | 173.2 | ± | 26.2 | |||||||
| FL-SeP (μg/mL) | − | 3.93 | ± | 1.20 | 3.92 | ± | 0.85 | 3.95 | ± | 0.48 | .180 | .819 | .817 | |||
| + | 3.93 | ± | 0.90 | 3.99 | ± | 0.91 | 4.12 | ± | 0.78 | |||||||
| Energy (kcal) | − | 1836.4 | ± | 594.2 | 1780.7 | ± | 576.1 | 1686.4 | ± | 482.4 | .375 | .012 | .209 | |||
| + | 1857.7 | ± | 689.5 | 1943.3 | ± | 596.9 | 1690.6 | ± | 588.4 | |||||||
| Protein (% of energy) | − | 15.4 | ± | 3.4 | 15.4 | ± | 3.2 | 14.4 | ± | 3.2 | .461 | .132 | .832 | |||
| + | 15.2 | ± | 3.1 | 15.5 | ± | 3.1 | 14.5 | ± | 3.2 | |||||||
| Fat (% of energy) | − | 24.7 | ± | 6.1 | 24.6 | ± | 6.0 | 23.3 | ± | 6.2 | .117 | .473 | .900 | |||
| + | 25.8 | ± | 6.4 | 25.5 | ± | 5.9 | 24.8 | ± | 7.1 | |||||||
| Carbohydrate (% of energy) | − | 54.1 | ± | 8.5 | 54.5 | ± | 8.1 | 56.1 | ± | 8.4 | .047 | .786 | .477 | |||
| + | 53.2 | ± | 8.5 | 52.6 | ± | 8.8 | 52.3 | ± | 8.7 | |||||||
P-values < .05 are in bold.
Abbreviations: BMI, body mass index; BP, blood pressure; Fib-4 index, Fibrosis-4 index; FL-SeP, full-length selenoprotein P; FPG, fasting plasma glucose; GOT, glutamic oxaloacetic transaminase; γ-GTP, gamma-glutamyl transpeptidase; HbA1c, hemoglobin A1c; HDL-C, high density lipoprotein cholesterol; HOMA-β, homeostasis model assessment of β-cell function; HOMA-IR, homeostasis model assessment of insulin resistance; LDL-C, low density lipoprotein cholesterol; WC, waist circumference.
aTwo-way analysis of covariance between hepatic steatosis and rs7579 genotypes adjusted by age, sex, and BMI.
bBonferroni post hoc test.
The Fib-4 index, a marker of liver fibrosis, is recommended as the first step in evaluating MASLD in primary care [46]. To determine whether liver fibrosis was associated with elevated liver enzyme levels only in males carrying the TT genotype, we analyzed whether the Fib-4 index interacted with hepatic steatosis and rs7579. In males with hepatic steatosis carrying the TT genotype, the Fib-4 index was higher than in those carrying other genotypes (Table S3) [45]. As the median Fib-4 index was 1.32, we stratified patients with Fib-4 indices ≤1.3 and >1.3. Only in the group with a Fib-4 index >1.3, GOT, GPT, and γ-GPT levels were higher in TT genotype-carrier males with hepatic steatosis (Table S4) [45].
Finally, we analyzed hypertension, dyslipidemia, diabetes, obesity, MetS, and hepatic steatosis. Among females, TT genotype-carrier participants displayed higher HOMA-IR than those carrying other genotypes with hypertension (P = .006), diabetes (P = .003), obesity (P < .001), or MetS (P < .001) (Tables S5-S8) [45]. HOMA-IR did not interact with rs7579 and dyslipidemia (Table S9) [45].
MASLD Analysis
We also investigated the relationship between rs7579 and environmental factors for MASLD. Of the 900 subjects, we included 503 in the analysis and excluded 397 as follows: 204 with no or inadequate BDHQ data (<600 kcal/day or ≥4000 kcal/day), 128 with moderate or higher alcohol intake (>30 and 20 g/day for males and females, respectively), and 65 with no abdominal echo data. Of the 503 participants, 173 exhibited MASLD. Selenium was associated with MASLD when adjusted for age, sex, BMI, alcohol intake, and smoking (model 2, Table S10) [45]. In the 2-way ANCOVA, we observed an interaction between MASLD and rs7579 by LDL-C, insulin, HOMA-IR, and HOMA-β overall and in females (Table S11) [45]. Participants carrying the TT genotype with MASLD were associated with high insulin, HOMA-IR, and HOMA-β levels.
Discussion
In females carrying the TT genotype of rs7579, we observed that hepatic steatosis, hypertension, diabetes, obesity, and MetS were associated with high HOMA-IR levels. This is the first study that describes an interaction between the SNP rs7579 of SELENOP and HOMA-IR in hepatic steatosis-associated metabolic disorders.
The rs7579 (C > T) is a functional variant located in the 3′ untranslated region of SELENOP. The T allele carriers have reportedly been associated with increased plasma SeP [19], plasma selenium [20], SELENOP mRNA [19, 21], and insulin [22] levels as well as MetS risk [23]. However, in this study, we detected no association between rs7579 and serum selenium or FL-SeP levels. We estimate the nutrient intake based on the BDHQ questionnaire, and, unfortunately, BDHQ does not assess selenium intake. Therefore, it is impossible to compare selenium intake between our study and previous studies. However, selenium is abundant in seafood and eggs and in North American cereals (depending on the selenium content of the soil), meat, and dairy products [47, 48]. Selenium intake is high in Japan, North America, and Venezuela while it is low in eastern Europe. China displays both selenium-deficient and selenium-rich regions [49]. The plasma selenium concentration in the previous study was approximately 90 μg/L [19, 20], which was lower than the 159 μg/L measured in the present study. In addition, unlike the assays in previous studies assessing total SeP including truncated fragments, we specifically measured FL-SeP in this study [18]. These reasons could be potentially responsible for the inconsistency between our current study and previous reports.
The rs7579 in the 3′ untranslated region overlaps with the CCDC152 gene, a long noncoding RNA that inhibits SELENOP translation. Rs7579 is an expression quantitative trait locus for both SELENOP and CCDC152. The GTEx database indicates that the TT genotype of rs7579 is associated with reduced SELENOP and increased CCDC152 expression in several organs, such as the pancreas and thyroid, which differs from previous reports describing upregulated SELENOP in the TT genotype. However, in the GTEx database, no significant SELENOP or CCDC152 expression-related differences were observed between the rs7579 genotypes in the liver, the main SeP-producing organ [50]. Indeed, we detected no difference in the serum SeP protein levels among the genotypes of rs7579 in this study (GTEx Analysis Release V8 on 10/24/2024, dbGaP accession number phs000424.v8.p2).
One reason for this phenomenon might be the sex-related differences in the expression levels by genotype. Indeed, in the present study, we detected an interaction between HOMA-IR and metabolic diseases (hepatic steatosis, obesity, diabetes, hypertension, and MetS) in the TT genotype of rs7579 only in females. Although we could not prove a causal relationship due to the cross-sectional nature of the study, we hypothesize that females carrying the TT genotype could be more susceptible to metabolic diseases if they display a high HOMA-IR level.
The sex difference in the results might be due to sexual dimorphism in SeP [51, 52]. In rodents, hepatic SELENOP mRNA levels are higher in females than in males, regardless of age [53]. However, in humans, there are reports of no sex difference [54] and higher hepatic SELENOP mRNA levels in males than in females [14]. Moreover, we previously reported that alcohol intake was associated with increased selenium and SeP serum levels, and this effect was more strongly pronounced in males than in females. These sexually dimorphic findings might be attributed to the fact that males consume more alcohol than females, and alcohol intake is more frequently associated with the consumption of selenium-rich foods such as seafood [48]. In our previous study, serum SeP levels and hepatic Selenop expression are associated with hyperglycemia and insulin resistance. SeP causes hepatic and skeletal muscle insulin resistance, and insulin downregulates the hepatic SeP expression [8]. However, in the present study, the genotype of rs7579 was not associated with the blood concentration of SeP. Therefore, the present finding of HOMA-IR interaction between rs7579 and hepatic steatosis in females is unexpected but may involve a novel biological phenomenon. Additional experimental studies are needed to clarify the interaction of SeP with sex hormone actions in the future.
We observed an interaction between hepatic steatosis and higher liver enzyme levels and the Fib-4 index in males carrying the TT genotype: TT genotype-carrier males with hepatic steatosis displayed higher liver enzyme levels and Fib-4 index. As the present general population cohort did not include people with severe liver fibrosis, rs7579 involvement in liver fibrosis should be tested in hospital cohorts.
Previous studies describe a positive correlation between serum selenium concentrations and LDL-C and triglyceride levels [55]. Our study proved that increased SeP levels are associated with dyslipidemia. Concerning sex differences, elevated SeP was a risk of dyslipidemia and hepatic steatosis in males and diabetes in females. SeP levels were negatively correlated to adiponectin levels in people with type 2 diabetes [56]. Therefore, SeP and selenium could potentially affect lipid metabolism.
The present study has certain limitations. This was a cross-sectional study and analyzed the prevalence, not the incidence, of metabolic diseases. Therefore, it is not possible to prove a causal relationship between HOMA-IR and metabolic diseases among genotypes. Moreover, in the questionnaire survey conducted in this study, it was not possible to eliminate residual confounding variables such as physical activity and history of treatment. In addition, there is a bias in the subject population: Japanese aged over 40 years living in rural areas.
In conclusion, in this study, we revealed that the association between metabolic diseases and HOMA-IR differed based on the SNP genotype and sex. In females carrying the TT genotype of SELENOP (rs7579), hepatic steatosis-related metabolic disorders (diabetes, hypertension, obesity, and MetS) were associated with high HOMA-IR levels. These discoveries will potentially open the way to genetic signatures-based personalized preventive medicine.
Acknowledgments
We would like to thank the study participants who responded to the survey and all staff for their cooperation in the Shika Study. We would like to thank Enago (https://www.enago.jp/) for proofreading our paper.
Contributor Information
Reina Yamamoto, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan; Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan.
Yumie Takeshita, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan; Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan.
Hiroaki Takayama, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Hiromasa Tsujiguchi, Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan; Department of Hygiene and Public Health, Graduate School of Medical Sciences, Kanazawa University, Kanazawa 920-8640, Japan.
Takayuki Kannon, Department of Biomedical Data Science, School of Medicine, Fujita Health University, Toyoake, Aichi 470-1192, Japan.
Takehiro Sato, Department of Human Biology and Anatomy, Graduate School of Medicine, University of the Ryukyus, Nishihara, Okinawa 903-0215, Japan.
Kazuyoshi Hosomichi, Laboratory of Computational Genomics, School of Life Science, Tokyo University of Pharmacy and Life Sciences, Hachioji, Tokyo 192-0392, Japan.
Keita Suzuki, Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan.
Takeo Tanaka, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Hisanori Goto, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Yujiro Nakano, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Tatsuya Yamashita, Department of Gastroenterology, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Shuichi Kaneko, Department of Gastroenterology, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Masao Honda, Department of Gastroenterology, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan.
Yoshiro Saito, Laboratory of Molecular Biology and Metabolism, Graduate School of Pharmaceutical Sciences, Tohoku University, Sendai 980-8578, Japan.
Atsushi Tajima, Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan; Department of Bioinformatics and Genomics, Graduate School of Advanced Preventive Medical Sciences, Kanazawa University, Kanazawa 920-8640, Japan.
Hiroyuki Nakamura, Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan; Department of Hygiene and Public Health, Graduate School of Medical Sciences, Kanazawa University, Kanazawa 920-8640, Japan.
Toshinari Takamura, Email: ttakamura@med.kanazawa-u.ac.jp, Department of Endocrinology and Metabolism, Kanazawa University Graduate School of Medical Sciences, Kanazawa 920-8640, Japan; Environmental Stress Research Center (eSRC), Kanazawa University, Kanazawa, 920-8640, Japan.
Funding
This study was supported by the Japan Diabetes Society Junior Scientist Development Grant supported by Novo Nordisk Pharma Ltd. and The Japan Diabetes Society (R.Y.), Japan Society for the Promotion of Science KAKENHI grant 21K19503 (To.T.), Japan Agency for Medical Research and Development grant JP20fk0210073 (To.T. and M.H.), 22ek0210144h0003 (Y.S. and To.T.), and the Takeda Science Foundation (To.T.).
Author Contributions
The authors’ responsibilities were as follows: To.T. designed the study; R.Y., Y.T., Ta.T., H.G., and Y.N. conducted the research; T.K. built the database for the cohort; T.S., K.H., and A.T. performed genetic analyses; T.Y., S.K., and M.H. evaluated the hepatic steatosis; Y.S. and H.T. assayed selenium and FL-SeP; R.Y. analyzed the data with support by H.T., A.T., and H.N.; R.Y. wrote the paper; To.T. edited the paper and had primary responsibility for final content. All authors read and approved the final manuscript.
Disclosures
The authors have nothing to disclose.
Data Availability
The datasets presented in this article are not readily available due to several ongoing studies using the datasets in this study. Requests to access the datasets should be directed to ttakamura@med.kanazawa-u.ac.jp.
References
- 1. Burk RF, Hill KE. Selenoprotein P: an extracellular protein with unique physical characteristics and a role in selenium homeostasis. Annu Rev Nutr. 2005;25(1):215‐235. [DOI] [PubMed] [Google Scholar]
- 2. Takayama H, Misu H, Iwama H, et al. Metformin suppresses expression of the selenoprotein P gene via an AMP-activated kinase (AMPK)/FoxO3a pathway in H4IIEC3 hepatocytes. J Biol Chem. 2014;289(1):335‐345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Yao X, Takayama H, Kamoshita K, et al. Cyclosporine A downregulates selenoprotein P expression via a signal transducer and activator of transcription 3–Forkhead box protein O1 pathway in hepatocytes in vitro. J Pharmacol Exp Ther. 2022;382(2):199‐207. [DOI] [PubMed] [Google Scholar]
- 4. Tajima-Shirasaki N, Ishii KA, Takayama H, et al. Eicosapentaenoic acid down-regulates expression of the selenoprotein P gene by inhibiting SREBP-1c protein independently of the AMP-activated protein kinase pathway in H4IIEC3 hepatocytes. J Biol Chem. 2017;292(26):10791‐10800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kamoshita K, Tajima-Shirasaki N, Ishii KA, et al. Forkhead box protein O1 (FoxO1) knockdown accelerates the eicosapentaenoic acid (EPA)-mediated Selenop downregulation independently of sterol regulatory element-binding protein-1c (SREBP-1c) in H4IIEC3 hepatocytes. Endocr J. 2022;69(8):907‐918. [DOI] [PubMed] [Google Scholar]
- 6. Kamoshita K, Tsugane H, Ishii KA, et al. Lauric acid impairs insulin-induced Akt phosphorylation by upregulating SELENOP expression via HNF4α induction. Am J Physiol Endocrinol Metab. 2022;322(6):E556‐E568. [DOI] [PubMed] [Google Scholar]
- 7. Murai K, Honda M, Shirasaki T, et al. Induction of selenoprotein P mRNA during hepatitis C virus infection inhibits RIG-I-mediated antiviral immunity. Cell Host Microbe. 2019;25(4):588‐601.e7. [DOI] [PubMed] [Google Scholar]
- 8. Misu H, Takamura T, Takayama H, et al. A liver-derived secretory protein, selenoprotein P, causes insulin resistance. Cell Metab. 2010;12(5):483‐495. [DOI] [PubMed] [Google Scholar]
- 9. Mita Y, Nakayama K, Inari S, et al. Selenoprotein P-neutralizing antibodies improve insulin secretion and glucose sensitivity in type 2 diabetes mouse models. Nat Commun. 2017;8(1):1658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ishikura K, Misu H, Kumazaki M, et al. Selenoprotein P as a diabetes-associated hepatokine that impairs angiogenesis by inducing VEGF resistance in vascular endothelial cells. Diabetologia. 2014;57(9):1968‐1976. [DOI] [PubMed] [Google Scholar]
- 11. Misu H, Takayama H, Saito Y, et al. Deficiency of the hepatokine selenoprotein P increases responsiveness to exercise in mice through upregulation of reactive oxygen species and AMP-activated protein kinase in muscle. Nat Med. 2017;23(4):508‐516. [DOI] [PubMed] [Google Scholar]
- 12. Oo SM, Oo HK, Takayama H, et al. Selenoprotein P-mediated reductive stress impairs cold-induced thermogenesis in brown fat. Cell Rep. 2022;38(13):110566. [DOI] [PubMed] [Google Scholar]
- 13. Takamura T. Hepatokine selenoprotein P-mediated reductive stress causes resistance to intracellular signal transduction. Antioxid Redox Signal. 2020;33(7):517‐524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Oo SM, Misu H, Saito Y, et al. Serum selenoprotein P, but not selenium, predicts future hyperglycemia in a general Japanese population. Sci Rep. 2018;8(1):16727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Yang SJ, Hwang SY, Choi HY, et al. Serum selenoprotein P levels in patients with type 2 diabetes and prediabetes: implications for insulin resistance, inflammation, and atherosclerosis. J Clin Endocrinol Metab. 2011;96(8):E1325‐E1329. [DOI] [PubMed] [Google Scholar]
- 16. Choi HY, Hwang SY, Lee CH, et al. Increased selenoprotein p levels in subjects with visceral obesity and nonalcoholic fatty liver disease. Diabetes Metab J. 2013;37(1):63‐71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tanaka M, Saito Y, Misu H, et al. Development of a sol particle homogeneous immunoassay for measuring full-length selenoprotein P in human serum. J Clin Lab Anal. 2016;30(2):114‐122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Saito Y, Misu H, Takayama H, et al. Comparison of human selenoprotein P determinants in serum between our original methods and commercially available kits. Biol Pharm Bull. 2018;41(5):828‐832. [DOI] [PubMed] [Google Scholar]
- 19. Donadio JLS, Rogero MM, Guerra-Shinohara EM, et al. Genetic variants in selenoprotein genes modulate biomarkers of selenium status in response to Brazil nut supplementation (the SU.BRA.NUT study). Clin Nutr. 2019;38(2):539‐548. [DOI] [PubMed] [Google Scholar]
- 20. Méplan C, Crosley LK, Nicol F, et al. Genetic polymorphisms in the human selenoprotein P gene determine the response of selenoprotein markers to selenium supplementation in a gender-specific manner (the SELGEN study). FASEB J. 2007;21(12):3063‐3074. [DOI] [PubMed] [Google Scholar]
- 21. Donadio JLS, Rogero MM, Cockell S, Hesketh J, Cozzolino SMF. Influence of genetic variations in selenoprotein genes on the pattern of gene expression after supplementation with Brazil nuts. Nutrients. 2017;9(7):739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hellwege JN, Palmer ND, Ziegler JT, et al. Genetic variants in selenoprotein P plasma 1 gene (SEPP1) are associated with fasting insulin and first phase insulin response in Hispanics. Gene. 2014;534(1):33‐39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Zhou L, Luo C, Yin J, et al. Diverse associations of plasma selenium concentrations and SELENOP gene polymorphism with metabolic syndrome and its components. Oxid Med Cell Longev. 2020;2020:5343014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Umemura S, Arima H, Arima S, et al. The Japanese Society of hypertension guidelines for the management of hypertension (JSH 2019). Hypertens Res. 2019;42(9):1235‐1481. [DOI] [PubMed] [Google Scholar]
- 25. Ogawa W, Hirota Y, Miyazaki S, et al. Definition, criteria, and core concepts of guidelines for the management of obesity disease in Japan. Endocr J. 2024;71(3):223‐231. [DOI] [PubMed] [Google Scholar]
- 26. Saito Y, Watanabe Y, Saito E, Honjoh T, Takahashi K. Production and application of monoclonal antibodies to human selenoprotein P. J Health Sci. 2001;47(4):346‐352. [Google Scholar]
- 27. Gardiner PH, Littlejohn D, Halls DJ, Fell GS. Direct determination of selenium in human blood serum and plasma by electrothermal atomic absorption spectrometry. J Trace Elem Med Biol. 1995;9(2):74‐81. [DOI] [PubMed] [Google Scholar]
- 28. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412‐419. [DOI] [PubMed] [Google Scholar]
- 29. Haffner SM, Miettinen H, Stern MP. The homeostasis model in the San Antonio heart study. Diabetes Care. 1997;20(7):1087‐1092. [DOI] [PubMed] [Google Scholar]
- 30. Turner RC, Holman RR, Matthews D, Hockaday TD, Peto J. Insulin deficiency and insulin resistance interaction in diabetes: estimation of their relative contribution by feedback analysis from basal plasma insulin and glucose concentrations. Metabolism. 1979;28(11):1086‐1096. [DOI] [PubMed] [Google Scholar]
- 31. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499‐502. [PubMed] [Google Scholar]
- 32. Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43(6):1317‐1325. [DOI] [PubMed] [Google Scholar]
- 33. Yamagishi K, Iso H. The criteria for metabolic syndrome and the national health screening and education system in Japan. Epidemiol Health. 2017;39:e2017003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Matsuzawa Y. Metabolic syndrome—definition and diagnostic criteria in Japan. J Atheroscler Thromb. 2005;12(6):301‐301. [DOI] [PubMed] [Google Scholar]
- 35. Alberti KGMM, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the international diabetes federation task force on epidemiology and prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; AND International Association for the study of obesity. Circulation. 2009;120(16):1640‐1645. [DOI] [PubMed] [Google Scholar]
- 36. Wang JH, Hung CH, Kuo FY, et al. Ultrasonographic quantification of hepatic–renal echogenicity difference in hepatic steatosis diagnosis. Dig Dis Sci. 2013;58(10):2993‐3000. [DOI] [PubMed] [Google Scholar]
- 37. Rinella ME, Lazarus JV, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol. 2023;79(6):1542‐1556. [DOI] [PubMed] [Google Scholar]
- 38. Kobayashi S, Honda S, Murakami K, et al. Both comprehensive and brief self-administered diet history questionnaires satisfactorily rank nutrient intakes in Japanese adults. J Epidemiol. 2012;22(2):151‐159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Kawai Y, Mimori T, Kojima K, et al. Japonica array: improved genotype imputation by designing a population-specific SNP array with 1070 Japanese individuals. J Hum Genet. 2015;60(10):581‐587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Nomura A, Sato T, Tada H, et al. Polygenic risk scores for low-density lipoprotein cholesterol and familial hypercholesterolemia. J Hum Genet. 2021;66:1079‐1087. [DOI] [PubMed] [Google Scholar]
- 41. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015;4(1):7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Patterson N, Price AL, Reich D. Population structure and eigenanalysis. PLoS Genet. 2006;2(12):e190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Browning BL, Browning SR. Genotype imputation with millions of reference samples. Am J Hum Genet. 2016;98(1):116‐126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Auton A, Abecasis GR, Altshuler DM, et al. A global reference for human genetic variation. Nature. 2015;526(7571):68‐74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Yamamoto R. The SELENOP polymorphism rs7579 predicts hepatic steatosis in females with insulin resistance in the general population. Data (Basel). Accessed July 14, 2025. Available at: http://hdl.handle.net/2297/0002003076 [DOI] [PMC free article] [PubMed]
- 46. Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77(5):1797‐1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Yoshida M. Distribution and nutritional availability of selenium contained in foods. Biomed Res Trace Elem. 2008;19(4):290‐296. [Google Scholar]
- 48. Isobe Y, Asakura H, Tsujiguchi H, et al. Alcohol intake is associated with elevated serum levels of selenium and selenoprotein P in humans. Front Nutr. 2021;8:633703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Rayman MP. Food-chain selenium and human health: emphasis on intake. Br J Nutr. 2008;100(2):254‐268. [DOI] [PubMed] [Google Scholar]
- 50. Hill KE, Wu S, Motley AK, et al. Production of selenoprotein P (Sepp1) by hepatocytes is central to selenium homeostasis. J Biol Chem. 2012;287(48):40414‐40424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Demircan K, Hybsier S, Chillon TS, et al. Sex-specific associations of serum selenium and selenoprotein P with type 2 diabetes mellitus and hypertension in the Berlin Aging Study II. Redox Biol. 2023;65:102823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Schomburg L, Schweizer U. Hierarchical regulation of selenoprotein expression and sex-specific effects of selenium. Biochim Biophys Acta. 2009;1790(11):1453‐1462. [DOI] [PubMed] [Google Scholar]
- 53. Schomburg L, Riese C, Renko K, Schweizer U. Effect of age on sexually dimorphic selenoprotein expression in mice. Biol Chem. 2007;388(10):1035‐1041. [DOI] [PubMed] [Google Scholar]
- 54. Hybsier S, Schulz T, Wu Z, et al. Sex-specific and inter-individual differences in biomarkers of selenium status identified by a calibrated ELISA for selenoprotein P. Redox Biol. 2017;11:403‐414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Zhou L, Liang X, Xie M, et al. A functional variant in SEPP1 interacts with plasma selenium concentrations on 3-year lipid changes: a prospective cohort study. Front Nutr. 2021;8:789577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Misu H, Ishikura K, Kurita S, et al. Inverse correlation between serum levels of selenoprotein P and adiponectin in patients with type 2 diabetes. PLoS One. 2012;7(4):e34952. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets presented in this article are not readily available due to several ongoing studies using the datasets in this study. Requests to access the datasets should be directed to ttakamura@med.kanazawa-u.ac.jp.