Relationship between fbroblast growth factor 21 and thyroid stimulating hormone in healthy subjects without components of metabolic syndrome

Tianxu Fu, Shanshan Yuan, Ying Gao, Chenxue Qu, Lan Chen, Yan Hui, Guizhi Lu, Yanming Gao, Xiaohui Guo

Relationship between fbroblast growth factor 21 and thyroid stimulating hormone in healthy subjects without components of metabolic syndrome

Tianxu Fu1, Shanshan Yuan1, Ying Gao1, Chenxue Qu2, Lan Chen1, Yan Hui1, Guizhi Lu1, Yanming Gao1, Xiaohui Guo1

Objective:To determine the relationship between human fbroblast growth factor 21 (FGF21) and thyroid stimulating hormone (TSH) by testing the level of FGF21, lipid metabolism, and carbohydrate metabolism-related indices, as well as the level of TSH, among metabolic syndromefree patients with normal physical examination results.

Methods:An enzyme-linked immunosorbent assay (ELISA) was used to test the levels of serum FGF21 and free fatty acids (FFA) in metabolic syndrome-free patients with normal physical examination results, and electrochemiluminescence (ECLIA) was used to measure TSH, thyroglobulin antibodies (TGAbs), and thyroid peroxidase antibody (TPOAb) levels.

Results:Three hundred ffty-six metabolic syndrome-free patients (116 males and 240 females; average age, 43±13 years) with normal physical examination results were enrolled. Among the patients with normal physical examination results, FGF21 had a weak relationship with obesity indices, such as the waist circumference (r=0.110,P=0.038), the waist-to-hip ratio (r=0.119,P=0.025), and the triglycerides level (TG; r=0.302,P=0.000), and a weak relationship with blood lipid levels, such as total cholesterol (TCHO; r=0.113,P=0.012) and low-density lipoprotein-cholesterol (LDL-C; r=0.175,P=0.001), but no relationship with TSH (r=—0.023,P=0.666). In addition, the FGF21 levels in thyroid autoantibody-positive and -negative groups were not signifcantly different.

Conclusion:Among the metabolic syndrome-free patients with normal physical examination results, FGF21 has no apparent relationship with TSH or thyroid autoimmunity.

Human fbroblast growth factor 21 (FGF21), Thyroid stimulating hormone (TSH), Autoimmunity, Free fatty acids (FFA)

Introduction

Human fbroblast growth factor 21 (FGF21), a newly discovered cell factor, is an important participant in regulating carbohydrate and lipid metabolism, and energy balance [1]. Some studies have shown that FGF21 has a close relationship with metabolic syndromerelated indices, such as the content of body fat, body mass index (BMI), waist-to-hip ratio, triglycerides (TG), free fatty acids (FFA), fasting blood glucose (FBG), fasting insulin (FIns), and insulin resistance [2, 3]. Different thyroid function states and thyroid stimulating hormone (TSH) levels also have an effect on carbohydrate and lipid metabolism [4—8], and an increase in TSH and FGF21 levels is closely associated with metabolic syndrome [9—11]. Thereare no relevant reports focusing on whether TSH or another pathway plays a role in the relationship between FGF21 and metabolic syndrome, i.e., whether or not there is a relationship between TSH and FGF21. In addition, Xiao et al. [12] reported that FGF21 has a relationship with autoantibodies, such as glutamic acid decarboxylase antibody (GADA) and insulinoma-associated protein 2 autoantibodies (IA2As), thus it was suggested that FGF21 may also have a relationship with autoimmune diseases.

The purpose of the current study was to collect data from metabolic syndrome-free patients with normal physical examination results to determine the difference in FGF21 levels among patients with different TSH levels through comparison and to detect the relationship between FGF21 and lipid and carbohydrate metabolism, and to further determine the relationship between FGF21 and thyroid autoimmunity by testing thyroid-related antibodies.

Subjects and methods

Subjects

Nine hundred three patients (438 males and 465 females; average age, 48±15 years) who had physical examinations between October 2012 and April 2013 at Peking University First Hospital were selected, and 356 patients (116 males and 240 females; average age, 43±13 years) who were free from any signs of metabolic syndrome and had no history of diabetes, hypertension, chronic kidney disease, polycystic ovary syndrome, autoimmune diseases, or pituitary and hypothalamic disease were enrolled. The metabolic syndrome was diagnosed as per the Chinese Society of Diabetes2004 Recommendation on Metabolic Syndrome[13]. Specifcally, metabolic syndrome was diagnosed in any person who satisfes three or all four of the following features: (1) overweight and/or obesity, BMI ≥25 kg/m2; (2) hyperglycemia, FBG ≥6.1 mmol/L and/or 2-h post-challenge plasma glucose (2-h PPG)≥7.8 mmol/L and/or patients with a confrmed diagnosis of diabetes under treatment; (3) hypertension, blood pressure (BP) ≥140/90 mmHg and/or patients with a confrmed diagnosis of hypertension under treatment; (4) dyslipidemia, fasting plasma TG ≥1.7 mmol/L and/or high-density lipoproteincholesterol (HDL-C) ＜0.9 mmol/L (male) or ＜1.0 mmol/L (female).

Methods

The age and gender were recorded, and the height, weight, waist circumference, hip circumference, and blood pressure were measured. The thyroid volume was estimated by palpation and the percentage body fat was determined using the biological impedance method (Tanita Company, Shanghai, China). Five millilitre fasting blood was drawn from the elbow vein, and examined using an automatic biochemical analyzer (Hitachi 7600/110, Beijing, China) to determine the alanine aminotransferase (ALT), aspartate aminotransferase (AST), serum creatinine (Scr), uric acid (UA), FBG, TG, total cholesterol (TCHO), HDL-C, and low-density lipoprotein-cholesterol (LDL-C) levels. Another 2 mL of serum was stored in a —80°C freezer for uniform preservation to test TSH, thyroglobulin antibody (TgAb), thyroid peroxidase antibody (TPOAb), and Fins from the same batch. The ECLIA method (Siemens ADVIA Centaur?XP detector, Beijing, China) was used to test TSH (normal reference range, 0.55—4.78 mIU/L; between-run coeffcient of variance, 2.857%—3.755%). The ECLIA method (Roche Cobas 601 detector, Shanghai, China) was used to test TPOAb, TgAb, and Fins (TPOAb ＞34 μIU/mL and TgAb＞115 μIU/mL were judged to be positive). The ELISA method was used to determine the levels of serum FGF21 (betweenrun coeffcient of variance, 13.64%; Li Ka Shing Faculty of Medicine, the University of Hong Kong, China) and FFA (between-run coeffcient of variance, 13.22%; BlueGene Company, Shanghai, China). The BMI (kg/m2) was calculated as the weight in kg/height in m2, the estimated glomerular fltration rate (eGFR) (mL/[min*1.73 m2]) was calculated as 175×(Scr in mg/dL)—1.234×(age in years)—0.179×(0.79 for females) [14], and the homeostasis model assessment for insulin resistance (HOMA-IR) was calculated as fasting insulin (mIU/L)×fasting blood glucose (mmol/L)÷22.5 [15].

Statistical analysis

Results

TSH and thyroid autoantibody test results

Among 356 patients, 8 were in the decreased TSH group, 320 were in the normal TSH group, 28 were in the increased TSH group, 75 were TgAb- or TPOAb-positive, and 280 were TgAb- and TPOAb-negative; 1 patient was not tested for thyroid autoantibodies due to an insuffcient blood sample.

The relationship between FGF21 and various

metabolic indices

As shown in Table 1, among 356 metabolic syndrome-free patients, FGF21 had a weak relationship with obesity indices, such as waist circumference (r=0.110,P=0.038) and waist-tohip ratio (r=0.119,P=0.025); FGF21 had no relationship with BMI, hip circumference, and body fat percentage. FGF21 had a positive relationship with TG (r=0.302,P=0.000) and had a weak relationship with blood lipid levels, including TCHO (r=0.113,P=0.012) and LDL-C (r=0.175,P=0.001), but no relationship with HDL-C. In addition, FGF21 had no relationship with TSH (r=—0.023,P=0.666) and even no relationship with FBG, FIns, HOMA-IR, UA, FFA, and systolic and diastolic blood pressures. With FGF21 designated as the dependent variable and indices correlated with FGF21 as independent variables (waist circumference, TG, TCHO, and LDL-C), the results from multiple linear stepwise regression analysis after age and weight were entered showed that TG had an independent relationship with FGF21 (P=0.000), and the standardized regression equation was as follows: YFGF21=230.84×TG+5.26×age.

The relationship between TSH and various metabolic indices

The serum TSH level had a positive relationship with FFA (r=0.840,P=0.000) and no relationship with BMI, waist circumference, hip circumference, waist-to-hip ratio, body fat percentage, TG, TCHO, HDL-C, LDL-C, FBG, FIns, HOMA IR, UA, and systolic or diastolic blood pressure.

Table 1. The relationship between FGF21 and various metabolic indices

Comparison of various metabolic indices of patients with different thyroid functions

As shown in Table 2, there was a signifcant increase in the serum FFA level of the increased TSH group when compared with the normal and decreased TSH groups; the medians were 338.8, 112.4, and 27.3 μg/mL (P=0.000), respectively. As shownin Table 2, there was no signifcant difference in the FGF21 level among the three groups; the medians were 392.0, 317.0, and 327.0 pg/mL (P=0.441), respectively. As shown in Table 2, there was no signifcant difference in age, BMI, waist circumference, hip circumference, waist-to-hip ratio, body fat percentage, blood lipid levels, FBG, FIns, HOMA-IR, UA, and systolic or diastolic blood pressure level among the three groups. After matching age, gender, and BMI at a 1:3 ratio, the UA (mean values=315.0 and 114.0 μmmol/L, respectively;P=0.000) and FFA levels (averages=315.0 and 114.0 μmmol/L, respectively;P=0.000) were signifcantly higher in the increased TSH group than the normal TSH group; there was no signifcant difference in the FGF21 level, BMI, waist circumference, hip circumference, waist-to-hip ratio, body fat percentage, blood lipid profle, FBG, FIns, HOMA-IR, UA, eGFR, and systolic or diastolic blood pressure levels between the two groups.

Comparison of FGF21 levels among different thyroid antibody groups

The TgAb- or TPOAb-positive and double antibody groups were compared with the seropositive group and both seronegative groups; no signifcant difference in FGF21 levels was noted (median=331 and 318 pg/mL, respectively;P=0.755).

Reference range of the FGF21 level

The reference range for the FGF21 level was 49.47—1123.6 pg/mL (368.51±281.545 pg/mL) based on the FGF21 levels of 193 patients whose BMI, FBG, blood lipid, and TSH levels were within the normal ranges, the physical examination was normal, and who had no history of diabetes, hypertension, chronic kidney disease, polycystic ovary syndrome, and autoimmune diseases.

Table 2. Data of patients with different thyroid functions

Discussion

A number of studies have suggested that the increase in FGF21 and TSH levels is directly related to metabolic syndrome, but the relationship between FGF21 and TSH is unclear [9—11]. To rule out the impact of metabolic syndrome on the relationship between FGF21 and TSH, the current study selected patients without metabolic syndrome and who had no history of diseases known to affect the FGF21 level, including polycystic ovary syndrome, chronic kidney disease, and autoimmune diseases [12, 16, 17].

The current study grouped patients on the basis of different TSH levels; no relationship existed between FGF21 and TSH. FGF21 had a positive relationship with TG, with the exception of patients with an elevated TG level, suggesting that FGF21 can still participate in TG metabolism when TG and TSH levels are within the normal range. Lee et al. [18] showed that the FGF21 level of patients with elevated TSH (hypothyroidism) was increased when compared with euthyroid patients. FGF21 has a positive relationship with TSH, and the difference is signifcant, even after TG correction [18]. A relationship between FGF21 and TSH was not observed in this study, and possible reasons were as follows: (1) Small sample size; there were only 8 patients in the decreased TSH group and 28 patients in the increased TSH group. (2) Patients with known diseases which have an impact on FGF21 were not selected as subjects in this study to rule out the impact of abnormal glucose metabolism, hypertriglyceridemia, and obesity on FGF21, and to rule out pathologic states induced by the effect of polycystic ovary syndrome, chronic kidney disease, and autoimmune diseases on the FGF21 level. Lee et al. [18] did not screen out known diseases. To summarize, the relationship between FGF21 and TSH was not observed in the current study, and further studies on the relationship and mechanism of action are needed for clarifcation.

The current study indicated that among the metabolic syndrome-free patients, FGF21 had a relationship with TG, a weak relationship with blood lipid levels (TCHO and LDL-C), and a weak relationship with obesity indices (waist circumference and waist-to-hip ratio); however, a relationship between FGF21 and carbohydrate metabolism was not observed. Li and others [19] reported that FGF21 has a relationship with TG and TCHO, but no relationship with the secretion and sensitivity of insulin when patients with abnormal glucose tolerance and healthy patients were compared. Li proposed that FGF21 has a much closer relationship with abnormal lipid metabolism and obesity, and the difference in expression among people with different glucose metabolism levels may be secondary to obesity. The relationship between FGF21 and glucose metabolism was also not observed in the current study, which indicated that FGF21 has a much closer relationship with TG metabolism.

The study also suggested that the serum FFA level of the increased TSH group was higher than the normal TSH group, and the serum FFA level of the normal TSH group still had a positive relationship with TSH among the normal TSH patients, which is quite similar to the results reported by Yu et al. [20]. It has been confrmed that there is a TSH receptor on the surface of fat cells [21], and TSH accelerates intracellular lipolysis to increase the generation of FFA in the human body through downstream protein kinase C [22, 23].

Xiao et al. [12] reported that the serum FGF21 level increased in patients with type 1 diabetes and latent autoimmune diabetes in adults (LADA) and had a positive relationship with GADA and IA2A. Xiao et al. [12] proposed that FGF21 may be involved in the autoimmune process. A relationship between the FGF21 level and thyroid autoantibody titers was not observed in the current study, thus additional studies are needed to verify the relationship between the FGF21 level and other autoimmune diseases.

A number of studies have suggested that FGF21 has a relationship with a variety of metabolic diseases, but there are no domestic or international studies involving the normal reference range of FGF21. Li et al. [19] studied 353 Chinese Han people (including patients with normal carbohydrate metabolism, impaired fasting glucose, impaired glucose tolerance, and diabetes), and demonstrated that the fuctuation range of FGF21 was 11.0—930.0 pg/mL. In the current study, the reference range for FGF21 level was 49.47—1123.6 pg/mL using a typical ELISA method based on the FGF21 level of 193 patients with BMI, FBG, blood lipids and TSH were within normal ranges, normal physical examinations, and no history of polycystic ovary syndrome, diabetes, hypertension, fatty liver, autoimmune diseases, or chronic kidney disease. The current study suggested that there is a large difference in FDG21 levels, even amongst the metabolic syndrome-free patients, which is consistent with the FGF21 fuctuation range reported by Li et al. [19]. Therefore, the above-mentioned facts might need to be considered when the relationship between FGF21 and diseases is studied.

In conclusion, among metabolic syndrome-free patients, we observed that FGF21 has no signifcant relationship with the TSH level, and thus may be unrelated to thyroid autoimmunity. FGF21 may participate in maintaining metabolic balance in humans through a mechanism different from TSH.

Confict of interest

The authors declare no confict of interest.

1. Itoh N. Hormone-like (endocrine) Fgfs: their evolutionary history and roles in development, metabolism, and disease. Cell Tissue Res 2010;342:1—11.

2. Semba RD, Sun K, Egan JM, Crasto C, Carlson OD, Ferrucci L. Relationship of serum fbroblast growth factor 21 with abnormal glucose metabolism and insulin resistance: the Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab 2012;97:1375—82.

3. Reinehr T, Woelfe J, Wunsch R, Roth CL. Fibroblast growth factor 21 (FGF-21) and its relation to obesity, metabolic syndrome, and nonalcoholic fatty liver in children: a longitudinal analysis. J Clin Endocrinol Metab 2012;97:2143—50.

4. Duntas LH, Brenta G. The effect of thyroid disorders on lipid levels and metabolism. Med Clin North Am 2012;96:269—81.

5. Canaris GJ, Manowitz NR, Mayor G, Ridgway EC. The Colorado thyroid disease prevalence study. Arch Intern Med 2000;160:526—34.

6. Hueston WJ, Pearson WS. Subclinical hypothyroidism and the risk of hypercholesterolemia. Ann Fam Med 2004;2:351—5.

7. Iqbal A, Jorde R, Figenschau Y. Serum lipid levels in relation to serum thyroid-stimulating hormone and the effect of thyroxine treatment on serum lipid levels in subjects with subclinical hypothyroidism: the Tromso Study. J Intern Med 2006;260:53—61.

8. Asvold BO, Vatten LJ, Nilsen TI, Bj?ro T. The association between TSH within the reference range and serum lipid concentrations in a population-based study. The HUNT Study. Eur J Endocrinol 2007;156:181—6.

9. Zhang X, Yeung DC, Karpisek M, Stejskal D, Zhou ZG, Liu F, et al. Serum FGF21 levels are increased in obesity and are independently associated with the metabolic syndrome in humans. Diabetes 2008;57:1246—53.

10. Erdogan M, Canataroglu A, Ganidagli S, Kulaks?zoglu M. Metabolic syndrome prevalence in subclinic and overt hypothyroid patients and the relation among metabolic syndrome parameters. J Endocrinol Invest 2011;34:488—92.

11. Gardu?o-Garcia Jde J, Alvirde-Garcia U, López-Carrasco G, Padilla Mendoza ME, Mehta R, Arellano-Campos O, et al. TSH and free thyroxine concentrations are associated with differing metabolic markers in euthyroid subjects. Eur J Endocrinol 2010;163:273—8.

12. Xiao Y, Xu A, Law LS, Chen C, Li H, Li X, et al. Distinct changes in serum fbroblast growth factor 21 levels in different subtypes of diabetes. J Clin Endocrinol Metab 2012;97:E54—8.

13. Chinese Society of Diabetes Research Group. Chinese Society of Diabetes-Recommendation on Metabolic Syndrome. Chin Journal of Diabetes 2004;12(3):156—61.

14. National Cooperative Group on eGFR Subject. Improvement and assessment of MDRD equation in Chinese patients with chronic kidney. Chinese J Nephrol 2006;22:589—95.

15. Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191—2.

16. Lin Z, Zhou Z, Liu Y, Gong Q, Yan X, Xiao J, et al. Circulating FGF21 levels are progressively increased from the early to end stages of chronic kidney diseases and are associated with renal function in Chinese. PLoS One 2011;6:e18398.

17. Gorar S, Culha C, Uc ZA, Dellal FD, Serter R, Aral S, et al. Serum fbroblast growth factor 21 levels in polycystic ovary syndrome. Gynecol Endocrinol 2010;26:819—26.

18. Lee Y, Park YJ, Ahn HY, Lim JA, Park KU, Choi SH, et al. Plasma FGF21 levels are increased in patients with hypothyroidism independently of lipid profle. Endocr J 2013;60:977—83.

19. Li H, Bao Y, Xu A, Pan X, Lu J, Wu H, et al. Serum fbroblast growth factor 21 is associated with adverse lipid profles and gamma-glutamyltransferase but not insulin sensitivity in Chinese subjects. J Clin Endocrinol Metab 2009;94:2151—6.

20. Yu HY, Zhang MX, Zhang JH, Yang Y. The relationship between free fatty acid and insulin resistance under different thyroid function status. Chinese J Diab Mellitus 2005;13:328—9.

21. Endo T, Kobayashi T. Expression of functional TSH receptor in white adipose tissues of hyt/hyt mice induces lipolysis in vivo. Am J Physiol Endocrinol Metab 2012;302:E1569—75.

22. Gagnon A, Antunes TT, Ly T, Pongsuwan P, Gavin C, Lochnan HA, et al. Thyroid-stimulating hormone stimulates lipolysis in adipocytes in culture and raises serum free fatty acid levels in vivo. Metabolism 2010;59:547—53.

23. Thrush AB, Gagnon A, Sorisky A. PKC activation is required for TSH-mediated lipolysis via perilipin activation. Horm Metab Res 2012;44:825—31.

1. Department of Endocrinology, Peking University First Hospital, Beijing, 100034, China

2. Department of Clinical Laboratory, Peking University First Hospital, Beijing, 100034, China

Ying Gao

Department of Endocrinology, Peking University First Hospital, Beijing, 100034, China

E-mail: bjgaoying@yahoo.com

We acknowledge Professor Aimin Xu, Department of Medicine, Li Ka Shing Faculty of Medicine, University of Hong Kong, who provided the FGF21 reagent kits in this experiment as a gift.

20 June 2014;

Accepted 15 August 2014