- Email: [email protected]
The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus
G Model
DIABET-901; No. of Pages 5 Diabetes & Metabolism xxx (2017) xxx–xxx
Available online at
ScienceDirect www.sciencedirect.com
Original article
The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus S. Kralisch a,b,1, A. Hoffmann a,1, J. Kratzsch c, M. Blu¨her a, M. Stumvoll a, M. Fasshauer a,b,1, T. Ebert a,b,*,1 a b c
Department of Endocrinology and Nephrology, University of Leipzig, 04103 Leipzig, Germany Leipzig University Medical Center, IFB Adiposity Diseases, 04103 Leipzig, Germany Institute of Laboratory Medicine, University of Leipzig, 04103 Leipzig, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 March 2017 Received in revised form 26 May 2017 Accepted 2 June 2017 Available online xxx
Aims. – Neuregulin 4 has recently been recognized as a novel adipokine secreted by brown adipose tissue (BAT), with beneficial effects on murine insulin resistance and hepatic steatosis. Yet, thus far, neither regulation of neuregulin 4 in gestational diabetes mellitus (GDM) nor its longitudinal changes in the peripartum period have been elucidated. Methods. – Circulating neuregulin 4 levels were measured by ELISA in 74 women with GDM and 74 healthy, gestational-age-matched controls. Also, neuregulin 4 was quantified during pregnancy and compared with postpartum levels in a follow-up study of 25 women with previous GDM and 25 healthy control women. Results. – Women with GDM had lower median serum levels of the novel BAT-secreted adipokine neuregulin 4 (3.0 mg/L) compared with healthy (non-GDM) pregnant controls (3.5 mg/L; P = 0.020), and the area under the glucose curve (AUCGlucose) was an independent and negative predictor of circulating neuregulin 4 (P = 0.033). Also, median postpartum serum concentrations of neuregulin 4 (3.2 mg/L) were not significantly different from prepartum levels (2.8 mg/L; P = 0.328). In addition, neuregulin 4 was positively and independently associated with irisin (P = 0.009), but not other BAT-secreted adipokines. Conclusion/interpretation. – Women with GDM have significantly lower circulating neuregulin 4 levels compared with healthy pregnant controls, and the AUCGlucose is negatively and independently associated with neuregulin 4 during pregnancy. Neuregulin 4 is positively correlated with irisin during pregnancy, as well as in a longitudinal fashion. Future studies are now needed to better elucidate the precise pathomechanisms of the regulation of BAT-secreted adipokines during pregnancy.
C 2017 Elsevier Masson SAS. All rights reserved.
Keywords: Adipokine Brown adipose tissue Gestational diabetes mellitus Insulin resistance Neuregulin 4 Pregnancy
Introduction During the last few decades, the prevalence of gestational diabetes mellitus (GDM) has increased significantly [1]. As GDM contributes to an increased risk of acute and chronic complications Abbreviations: ADA, American Diabetes Association; ANGPTL8, Angiopoietin-like protein 8; AUCGlucose, Area under the glucose curve; AFABP, Adipocyte fatty acidbinding protein; BAT, Brown adipose tissue; BMI, Body mass index; ELISA, Enzymelinked immunosorbent assay; FFA, Free fatty acids; FGF21, Fibroblast growth factor 21; FI, Fasting insulin; GDM, Gestational diabetes mellitus; HDL, High-density lipoprotein; HOMA-IR, Homoeostasis model assessment of insulin resistance; LDL, Low-density lipoprotein; OGTT, Oral glucose tolerance test; TG, Triglyceride. * Corresponding author at: Liebigstr. 20, 04103 Leipzig, Germany. Fax: +49 341 9713389. E-mail address: [email protected] (T. Ebert). 1 These authors equally contributed to this work.
in both the mother and newborn [2], the pathophysiological mechanisms behind this metabolic disorder during pregnancy are of great interest. Thus, over the past several years, a variety of cytokines derived from adipose tissue, i.e. ‘adipokines’, have been identified as influencing both type 2 diabetes mellitus (T2DM) and GDM including adiponectin, leptin and adipocyte fatty acidbinding protein (AFABP)[3–6]. However, most adipokines are secreted predominantly by white adipose tissue and mediate adverse metabolic systemic effects. In contrast, several adipokines derived from brown adipose tissue (BAT) have recently been suggested to beneficially influence systemic metabolism, as reviewed by Villarroya et al. [7]. Neuregulin 4 has recently been recognized as a novel and predominantly BAT-secreted adipokine that protects against dietinduced insulin resistance and hepatic steatosis in mice [8]. Wang and co-workers [8] elegantly demonstrated that mice deficient in
http://dx.doi.org/10.1016/j.diabet.2017.06.001 C 2017 Elsevier Masson SAS. All rights reserved. 1262-3636/
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 2
S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
neuregulin 4 have increased insulin resistance and hepatic steatosis after high-fat feeding. Conversely, when mice with transgenic overexpression of neuregulin 4 were exposed to a highfat diet, they showed less body weight gain, and improved dyslipidaemia, as well as improved insulin sensitivity [8]. Mechanistically, the authors suggested that hepatic expression of lipogenic genes is lower in transgenic mice overexpressing neuregulin 4 compared with their littermate controls [8]. Taking these data together, neuregulin 4 appears to be a novel BATderived adipokine with beneficial effects on both glucose homoeostasis and hepatic steatosis. However, there are no studies investigating neuregulin 4 in GDM and in a longitudinal fashion during and after pregnancy, so far. In addition, the associations of neuregulin 4 with other BAT-secreted adipokines and markers of BAT function have not been tested in patients with GDM. Therefore, our present study has quantified circulating neuregulin 4 concentrations in 74 well-phenotyped women with GDM during pregnancy and compared them with 74 gestational-agematched, healthy pregnant controls. Neuregulin 4 was also correlated with clinical and biochemical measures of obesity, hypertension, indices of glucose metabolism, lipid metabolism, renal function and inflammation, as well as other markers of BAT function. Furthermore, neuregulin 4 was assessed longitudinally in 25 healthy (non-GDM) and 25 GDM women during pregnancy and postpartum in a follow-up study. Our hypothesis was that: women with GDM have decreased neuregulin 4 levels during pregnancy compared with non-diabetic, healthy pregnant women; circulating postpartum neuregulin 4 is increased because of reduced insulin resistance postpartum; and neuregulin 4 correlates with other markers of BAT function, including irisin, adiponectin, fibroblast growth factor (FGF)-21 and angiopoietin-like protein 8 (ANGPTL8) [7].
The present study was approved by the local ethics committee, and all subjects gave their written informed consent before taking part. Assays All blood samples were obtained after a fasting period of at least 8 h. Serum levels of neuregulin 4 (Phoenix Pharmaceuticals, Burlingame, CA, USA) were quantified using enzyme-linked immunosorbent assays (ELISAs) as per the manufacturer’s instructions. All other adipokines were quantified by ELISA kits as previously reported [9,11,19]. Fasting insulin (FI) was determined by two-site chemiluminescent enzyme immunometric assay, using a LIAISON automated analyzer (DiaSorin, Saluggia, Italy). All other parameters were measured by standard laboratory methods at a certified laboratory, as previously described [9–14]. Statistical analysis SPSS version 24.0 software (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Differences between women with GDM and the controls were assessed by non-parametric Mann– Whitney U test. Univariate correlations were performed using nonparametric Spearman’s rank-correlation method. Multivariate linear-regression analysis was subsequently performed to identify independent relationships. Before the multivariate correlation analyses were calculated, distribution was tested for normality, using the Shapiro–Wilk W test, and non-normally distributed parameters were logarithmically transformed. Longitudinal changes in neuregulin 4 during pregnancy compared with postpartum levels were analyzed by non-parametric Wilcoxon signed-rank test. A P-value < 0.05 was considered statistically significant in all analyses.
Results
Methods
Baseline characteristics
Study participants
The median [interquartile range] serum level of neuregulin 4 for the total sample during pregnancy was 3.3 [1.7] mg/L. Clinical characteristics of both subgroups (controls and GDM) are shown in Table 1. Median circulating neuregulin 4 levels were significantly lower in women with GDM (3.0 [1.2] mg/L) compared with the non-GDM pregnant controls (3.5 [1.9] mg/L; P = 0.020; Table 1). At the first time point during pregnancy, markers of glucose homoeostasis, including plasma glucose levels during OGTT, AUCGlucose, FI and HOMA-IR, as well as free fatty acids (FFAs), were significantly higher in women with GDM compared with the controls (P < 0.05; Table 1). In contrast, there were no significant differences in age, gestational age at blood sampling, gestational age at delivery, birth weight or markers of obesity, hypertension, dyslipidaemia, renal function or inflammation.
The design of the present study has been described previously [9–14]. In brief, 148 pregnant women were consecutively recruited from the outpatient care unit of the Department of Endocrinology and Nephrology, University of Leipzig, between 2006 and 2011. Based on the 2012 guidelines of the American Diabetes Association (ADA) [15], a 75-g 2-h oral glucose tolerance test (OGTT) was performed in all participants. As per ADA criteria, GDM was defined as having one or more elevated plasma glucose levels during OGTT, using the following thresholds: fasting plasma glucose (FPG) 5.1 mmol/L; 1-h FPG 10.0 mmol/L; and 2-h FPG 8.5 mmol/L. Based on these levels, 74 women were classified as patients with GDM, whereas 74 gestational-agematched pregnant women with normal glucose tolerance served as controls. Anthropometric measures included body mass index (BMI), determined by weight before pregnancy divided by squared height (kg/m2). In addition, the homoeostasis model assessment of insulin resistance (HOMA-IR) score and area under the glucose curve (AUCGlucose) were calculated, as previously described [16,17]. To investigate postpartum regulation of the novel BATsecreted factor, follow-up examination was carried out in 2012. For this postpartum subcohort, 25 women with previous GDM and 25 healthy former control women were included in the analysis. Relative changes (postpartum-to-prepartum ratios) were calculated for all adipokines and BAT markers as follows: parameter (ratio) = parameterpostpartum/parameterprepartum [18].
Univariate correlations On univariate correlation analysis of the entire cohort during pregnancy, neuregulin 4 positively correlated with systolic and diastolic blood pressure (P < 0.05) (Table 2). Also, the novel adipokine was negatively correlated with glycated haemoglobin (HbA1c), 1-h and 2-h glucose during OGTT, and AUCGlucose (P < 0.05; Table 2). In contrast, no significant correlations were established between circulating neuregulin 4 and markers of gestational outcome, obesity, dyslipidaemia, renal function or inflammation in the total study population (Table 2).
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx Table 1 Baseline characteristics of the study population during pregnancy.
n Neuregulin 4 (mg/L) Age (years) Gestational age at blood sampling (days) Gestational age at delivery (days) Infant birth weight (g) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) HbA1c (%) HbA1c (mmol/mol) Glucose 0 h(OGTT) (mmol/L) Glucose 1 h(OGTT) (mmol/L) Glucose 2 h(OGTT) (mmol/L) AUCGlucose (mmol/L) Fasting insulin (pmol/L) HOMA-IR Cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) Free fatty acids (mmol/L) Creatinine (mmol/L) hsCRP (mg/L) Leptin (mg/L)
Controls
GDM
P
74 3.5 (1.9) 28.9 (4.5) 199 (40) 275 (15) 3360 (790) 22.4 (6.7) 125 (17) 75 (13) 5.3 (0.6) 34.4 (6.6) 4.3 (0.5) 7.5 (1.6) 6.4 (1.8) 12.7 (2.4) 57.9 (38.4) 1.56 (0.96) 6.31 (1.84) 1.93 (0.51) 3.73 (1.57) 2.02 (1.43) 0.47 (0.30) 49.0 (11.0) 4.20 (4.30) 23.0 (11.9)
74 3.0 (1.2) 31.0 (7.5) 202 (33) 273 (14) 3400 (805) 24.5 (6.6) 120 (20) 73 (15) 5.4 (0.6) 35.5 (6.6) 4.5 (0.9) 10.1 (1.7) 8.7 (2.3) 16.6 (2.1) 70.6 (66.7) 1.98 (1.89) 6.71 (1.74) 1.82 (0.80) 4.05 (1.91) 2.14 (1.31) 0.55 (0.31) 46.0 (11.3) 3.99 (6.09) 26.5 (21.6)
0.020* 0.087 0.568 0.312 0.471 0.117 0.338 0.349 0.729 0.729 < 0.001* < 0.001* < 0.001* < 0.001* 0.003* < 0.001* 0.199 0.419 0.569 0.453 0.047* 0.086 0.903 0.131
Values are medians (interquartile range). GDM: gestational diabetes mellitus; HbA1c: glycated haemoglobin; OGTT: oral glucose tolerance test; AUCGlucose: area under the glucose curve; HOMA-IR: homoeostasis model assessment of insulin resistance; HDL/LDL: high-density/low-density lipoprotein; hsCRP: high-sensitivity C-reactive protein. * P < 0.05 (Mann–Whitney U test).
Multivariate regression analysis To identify independent associations, multivariate linearregression analysis was carried out. Here, AUCGlucose remained a
3
negative predictor of circulating neuregulin 4 after adjusting for age, gestational age, BMI, blood pressure and irisin (P = 0.033; Table 2). Also, diastolic blood pressure and irisin were significantly and independently associated with neuregulin 4 during pregnancy (P < 0.05; Table 2). Longitudinal changes in neuregulin 4 before and after delivery For all follow-up samples available for the present study (n = 50: 25 former controls; 25 women with previous GDM), the median circulating neuregulin 4 was 2.8 [1.7] mg/L during pregnancy and 3.2 [0.8] mg/L at the postpartum time point (P = 0.328, longitudinal analysis). When women with previous GDM and the former controls were analyzed separately, there was again no statistical difference in pre- and postpartum neuregulin 4 levels (Fig. 1). Interestingly, women with previous GDM had significantly higher postpartum neuregulin 4 levels (3.4 [0.8] mg/L) compared with the former controls (2.8 [0.9] mg/L; P = 0.032). However, there was no significant correlation with other metabolic markers postpartum (data not shown). Correlation of baseline neuregulin 4 and its relative changes with other BAT-secreted adipokines and markers of BAT function In all women during pregnancy (n = 148), prepartum neuregulin 4 serum levels were positively correlated to irisin levels (P = 0.047; Table 2) (Table 3). For the matched prepartum and postpartum samples of all women (n = 50), relative neuregulin 4 changes (ratio of neuregulin 4 = neuregulin 4postpartum/neuregulin 4prepartum) also correlated with relative changes of irisin (P = 0.005) (Table 4). Interestingly and unlike irisin, circulating neuregulin 4 levels, as well as its relative changes, were not associated with other BAT-secreted adipokines, such as adiponectin, FGF-21 and ANGPTL8 (Tables 3 and 4).
Table 2 Univariate correlations of serum neuregulin 4 in all women during pregnancy and multivariate regression analysis of neuregulin 4 (lg) as a dependent variable. Multivariate regression analysisa
Univariate correlations
Age (years) Gestational age at blood sampling (days) Gestational age at delivery (days) Infant birth weight (g) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) HbA1c (%) HbA1c (mmol/mol) Glucose 0 h(OGTT) (mmol/L) Glucose 1 h(OGTT) (mmol/L) Glucose 2 h(OGTT) (mmol/L) AUCGlucose (mmol/L) Fasting insulin (pmol/L) HOMA-IR Cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) Free fatty acids (mmol/L) Creatinine (mmol/L) hsCRP (mg/lL) Leptin (mg/L) Irisin (mg/L)
r
P
0.137 0.139 0.037 0.085 0.072 0.168 0.182 0.190 0.190 0.100 0.184 0.198 0.190 0.035 0.041 0.003 0.107 0.001 0.136 0.007 0.019 0.068 0.126 0.163
0.097 0.093 0.668 0.325 0.396 0.046b 0.031b 0.023b 0.023b 0.224 0.029b 0.020b 0.025b 0.674 0.621 0.973 0.196 0.991 0.099 0.935 0.817 0.411 0.127 0.047b
Standardized b coefficient 0.102 0.126 – – 0.144 – 0.190 – – – – – 0.180 – – – – – – – – – – 0.226
P 0.228 0.132 – – 0.102 – 0.027c – – – – – 0.033c – – – – – – – – – – 0.009c
HbA1c: glycated haemoglobin; OGTT: oral glucose tolerance test; AUCGlucose: area under the glucose curve; HOMA-IR: homoeostasis model assessment of insulin resistance; HDL/LDL: high-density/low-density lipoprotein; hsCRP: high-sensitivity C-reactive protein. a Non-normally distributed variables assessed by Shapiro–Wilk test were logarithmically transformed prior to multivariate testing. b Significant by Spearman’s correlation method. c Significant by multivariate analysis.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
4
Fig. 1. Longitudinal analysis of neuregulin 4 serum levels in a study subset (n = 50: 25 women with gestational diabetes mellitus (GDM); 25 controls). The overall P-value was assessed by non-parametric Wilcoxon signed-rank test.
Discussion In the present study, we demonstrate that women with GDM have lower serum levels of the novel BAT-secreted adipokine neuregulin 4 compared with the non-GDM pregnant controls. The AUCGlucose remains negatively and independently associated with circulating neuregulin 4 levels. Moreover, postpartum serum concentrations of neuregulin 4 do not significantly differ from those prepartum. However, women with previous GDM have increased neuregulin 4 compared with the former controls. Finally, the BAT-derived adipokine is positively associated with irisin, but not with other BAT-secreted adipokines [7].
Neuregulin 4 is a novel adipokine predominantly expressed in BAT [8]. Interestingly, both mice and humans with obesity have decreased Nrg4 mRNA expression, and mice deficient in Nrg4 display an adverse metabolic phenotype, including increased insulin resistance, dyslipidaemia (increased plasma TG) and hepatic steatosis [8]. More specifically, Nrg4-knockout mice fed a high-fat diet show elevated glucose concentrations during glucose tolerance tests [8]. Conversely, mice with transgenic overexpression of Nrg4 have lower glucose concentrations than wild-type control mice during glucose tolerance tests [8]. Taking these murine results into consideration, our human results in vivo are in almost perfect agreement demonstrating that the AUCGlucose during a 75-g OGTT is negatively associated with serum levels of neuregulin 4 even after adjusting for age, gestational age, markers of fat mass, blood pressure and irisin. These data for women with GDM are also in accordance with the findings of other cohorts investigating neuregulin 4 in metabolic disease states, including T2DM [20], the metabolic syndrome [21] and non-alcoholic fatty liver disease [22]. However, it should be noted that a recent study by Kang and co-workers [23] found increased neuregulin 4 serum levels in 57 South-Korean patients with newly diagnosed T2DM compared with controls. Hypothetically, different patient characteristics and ELISA systems might well explain the observed variations. Interestingly, several novel adipokines are not only adiposetissue-derived, but also co-secreted by the placenta during pregnancy, including irisin, fetuin B and preadipocyte factor 1 [9,14,24,25]. To determine whether the placenta significantly influences serum neuregulin 4 levels, we have performed a longitudinal analysis in 50 subjects (25 previous GDM patients and
Table 3 Univariate correlations of serum levels of adipokines with each other in all women during pregnancy (n = 148). During pregnancy
During pregnancy Neuregulin 4 Irisin ANGPTL8 Adiponectin FGF-21
r P r P r P r P r P
Neuregulin 4
Irisin
ANGPTL8
Adiponectin
– – 0.163 0.047a 0.090 0.274 0.084 0.311 0.044 0.592
– – 0.091 0.272 0.068 0.409 0.177 0.031a
– – 0.057 0.495 0.053 0.524
– – 0.267 0.001a
FGF-21
– –
ANGPTL8: angiopoietin-like protein 8; FGF-21: fibroblast growth factor 21. a Significant by Spearman’s correlation method.
Table 4 Univariate correlations of relative changes (postpartum/prepartum ratio) in adipokines with each other in all women included in the present analysis (n = 50). Relative changes (postpartum/prepartum)
Relative changes (postpartum/prepartum) Neuregulin 4 Irisin ANGPTL8 Adiponectin FGF-21
r P r P r P r P r P
Neuregulin 4
Irisin
– – 0.394 0.005a 0.144 0.317 0.187 0.306 0.371 0.052
– – 0.162 0.261 0.178 0.330 0.198 0.312
ANGPTL8
Adiponectin
FGF-21
– – 0.368 0.038a 0.165 0.401
– – 0.102 0.606
– –
ANGPTL8: angiopoietin-like protein 8; FGF-21: fibroblast growth factor 21. a Significant by Spearman’s correlation method.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
25 former controls). However, as there is no longitudinal change in circulating neuregulin 4 in our cohort, it would appear that placental co-secretion of neuregulin 4 does not significantly contribute to the observed results. In our follow-up cohort, women with previous GDM have increased neuregulin 4 levels in the postpartum state compared with the former controls. Further studies are now needed to validate this counterintuitive finding, and to assess whether neuregulin 4 can serve as a marker of adverse metabolic status in women. Besides markers of glucose tolerance, neuregulin 4 is also independently and positively associated with blood pressure. These data are in contrast to findings by Cai and co-workers [21], demonstrating lower neuregulin 4 levels in patients with hypertension than in normotensive subjects. However, the subjects included in that study had significantly higher median blood pressure than the women included in our present analysis. Thus, different patient characteristics may well explain these observed differences. Interestingly, both prepartum neuregulin 4 and its relative changes are positively correlated with irisin, another metabolically beneficial [26] adipokine/myokine influencing BAT function [27,28]. Yet, despite its low expression in BAT [28], irisin exerts beneficial metabolic effects, including increased energy expenditures and improved glucose homoeostasis, through BAT activation [27]. Based on our present study, it is tempting to speculate that both neuregulin 4 and irisin could act as cytokines improving metabolic status in GDM. In conclusion, women with GDM have significantly lower serum concentrations of neuregulin 4 compared with healthy (non-GDM) pregnant control women, and the AUCGlucose is negatively and independently associated with neuregulin 4 during pregnancy. Neuregulin 4 is also positively correlated with irisin during pregnancy, as well as in a longitudinal fashion. Future studies are now needed to better elucidate the role of BAT-secreted adipokines during pregnancy. Funding This work was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1501 (IFB AdiposityDiseases, project K6a-87) and by the Deutsche Forschungsgemeinschaft (DFG; SFB 1052/2, C06 to MF, B01 to MB, and A01 to MS). TE was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1001 (IFB AdiposityDiseases, MetaRot program). Author contributions T.E., S.K., and M.F. wrote the manuscript and researched data. A.H., and J.K. researched data and reviewed/edited the manuscript. M.B. and M.S. contributed to the discussion and reviewed/edited the manuscript. Dr Thomas Ebert is the guarantor of this work and, as such, had full access to all the data in the study, and takes responsibility for the integrity of the data and accuracy of the data analysis. Disclosure of interest The authors declare that they have no competing interest.
5
References [1] Lavery JA, Friedman AM, Keyes KM, Wright JD, Ananth CV. Gestational diabetes in the United States: temporal changes in prevalence rates between 1979 and 2010. BJOG 2017;124:804–13. [2] Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol 2012;8:639–49. [3] Fasshauer M, Paschke R. Regulation of adipocytokines and insulin resistance. Diabetologia 2003;46:1594–603. [4] Fasshauer M, Blu¨her M. Adipokines in health and disease. Trends Pharmacol Sci 2015;36:461–70. [5] Fasshauer M, Blu¨her M, Stumvoll M. Adipokines in gestational diabetes. Lancet Diabetes Endocrinol 2014;2:488–99. [6] Miehle K, Stepan H, Fasshauer M. Leptin, adiponectin and other adipokines in gestational diabetes mellitus and pre-eclampsia. Clin Endocrinol 2012;76:2–11. [7] Villarroya F, Cereijo R, Villarroya J, Giralt M. Brown adipose tissue as a secretory organ. Nat Rev Endocrinol 2017;13:26–35. [8] Wang G-X, Zhao X-Y, Meng Z-X, Kern M, Dietrich A, Chen Z, et al. The brown fat-enriched secreted factor Nrg4 preserves metabolic homeostasis through attenuation of hepatic lipogenesis. Nat Med 2014;20:1436–43. [9] Ebert T, Stepan H, Schrey S, Kralisch S, Hindricks J, Hopf L, et al. Serum levels of irisin in gestational diabetes mellitus during pregnancy and after delivery. Cytokine 2014;65:153–8. [10] Ebert T, Hindricks J, Kralisch S, Lossner U, Jessnitzer B, Richter J, et al. Serum levels of fractalkine are associated with markers of insulin resistance in gestational diabetes. Diabet Med 2014;31:1014–7. [11] Ebert T, Kralisch S, Wurst U, Lo¨ssner U, Kratzsch J, Blu¨her M, et al. Betatrophin levels are increased in women with gestational diabetes mellitus compared to healthy pregnant controls. Eur J Endocrinol 2015;173:1–7. [12] Platz M, Stepan H, Schrey S, Kralisch S, Wurst U, Lossner U, et al. Serum levels of sclerostin in cardiometabolic disorders during pregnancy. Cytokine 2015;76:591–3. [13] Ebert T, Platz M, Kralisch S, Lossner U, Jessnitzer B, Richter J, et al. Serum levels of copeptin are decreased in gestational diabetes mellitus. Exp Clin Endocrinol Diabetes 2016;124:257–60. [14] Kralisch S, Hoffmann A, Lo¨ssner U, Kratzsch J, Blu¨her M, Stumvoll M, et al. Regulation of the novel adipokines/hepatokines fetuin A and fetuin B in gestational diabetes mellitus. Metabolism 2017;68:88–94. [15] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012;36:S67–74. [16] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9. [17] Enigk U, Breitfeld J, Schleinitz D, Dietrich K, Halbritter J, Fischer-Rosinsky A, et al. Role of genetic variation in the human sodium–glucose cotransporter 2 gene (SGLT2) in glucose homeostasis. Pharmacogenomics 2011;12:1119–26. [18] Ebert T, Hopf LM, Wurst U, Bachmann A, Kralisch S, Lo¨ssner U, et al. Circulating adipocyte fatty acid binding protein is increased in chronic and acute renal dysfunction. Nutr Metab Cardiovasc Dis 2014;24:1027–34. [19] Stein S, Stepan H, Kratzsch J, Verlohren M, Verlohren H-J, Drynda K, et al. Serum fibroblast growth factor 21 levels in gestational diabetes mellitus in relation to insulin resistance and dyslipidemia. Metabolism 2010;59:33–7. [20] Zhang L, Fu Y, Zhou N, Cheng X, Chen C. Circulating neuregulin 4 concentrations in patients with newly diagnosed type 2 diabetes: a cross-sectional study. Endocrine 2017. http://dx.doi.org/10.1007/s12020-017-1324-3 [Epub ahead of print]. [21] Cai C, Lin M, Xu Y, Li X, Yang S, Zhang H. Association of circulating neuregulin 4 with metabolic syndrome in obese adults: a cross-sectional study. BMC Med 2016;14:165. [22] Dai Y-N, Zhu J-Z, Fang Z-Y, Zhao D-J, Wan X-Y, Zhu H-T, et al. A case–control study: association between serum neuregulin 4 level and non-alcoholic fatty liver disease. Metabolism 2015;64:1667–73. [23] Kang YE, Kim JM, Choung S, Joung KH, Lee JH, Kim HJ, et al. Comparison of serum Neuregulin 4 (Nrg4) levels in adults with newly diagnosed type 2 diabetes mellitus and controls without diabetes. Diabetes Res Clin Pract 2016;117:1–3. [24] Wurst U, Ebert T, Kralisch S, Stumvoll M, Fasshauer M. Serum levels of the adipokine Pref-1 in gestational diabetes mellitus. Cytokine 2015;71:161–4. [25] Schrey S, Wurst U, Ebert T, Kralisch S, Drewlo S, Stepan H, et al. The adipokine preadipocyte factor-1 is downregulated in preeclampsia and expressed in placenta. Cytokine 2015;75:338–43. [26] Ebert T, Kralisch S, Wurst U, Scholz M, Stumvoll M, Kovacs P, et al. Association of metabolic parameters and rs726344 in FNDC5 with serum irisin concentrations. Int J Obes 2016;40:260–5. [27] Bostro¨m P, Wu J, Jedrychowski MP, Korde A, Ye L, Lo JC, et al. A PGC1-adependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature 2012;481:463–8. [28] Roca-Rivada A, Castelao C, Senin LL, Landrove MO, Baltar J, Crujeiras AB, et al. FNDC5/Irisin is not only a myokine but also an adipokine. PLoS One 2013;8:e60563.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
DIABET-901; No. of Pages 5 Diabetes & Metabolism xxx (2017) xxx–xxx
Available online at
ScienceDirect www.sciencedirect.com
Original article
The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus S. Kralisch a,b,1, A. Hoffmann a,1, J. Kratzsch c, M. Blu¨her a, M. Stumvoll a, M. Fasshauer a,b,1, T. Ebert a,b,*,1 a b c
Department of Endocrinology and Nephrology, University of Leipzig, 04103 Leipzig, Germany Leipzig University Medical Center, IFB Adiposity Diseases, 04103 Leipzig, Germany Institute of Laboratory Medicine, University of Leipzig, 04103 Leipzig, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 March 2017 Received in revised form 26 May 2017 Accepted 2 June 2017 Available online xxx
Aims. – Neuregulin 4 has recently been recognized as a novel adipokine secreted by brown adipose tissue (BAT), with beneficial effects on murine insulin resistance and hepatic steatosis. Yet, thus far, neither regulation of neuregulin 4 in gestational diabetes mellitus (GDM) nor its longitudinal changes in the peripartum period have been elucidated. Methods. – Circulating neuregulin 4 levels were measured by ELISA in 74 women with GDM and 74 healthy, gestational-age-matched controls. Also, neuregulin 4 was quantified during pregnancy and compared with postpartum levels in a follow-up study of 25 women with previous GDM and 25 healthy control women. Results. – Women with GDM had lower median serum levels of the novel BAT-secreted adipokine neuregulin 4 (3.0 mg/L) compared with healthy (non-GDM) pregnant controls (3.5 mg/L; P = 0.020), and the area under the glucose curve (AUCGlucose) was an independent and negative predictor of circulating neuregulin 4 (P = 0.033). Also, median postpartum serum concentrations of neuregulin 4 (3.2 mg/L) were not significantly different from prepartum levels (2.8 mg/L; P = 0.328). In addition, neuregulin 4 was positively and independently associated with irisin (P = 0.009), but not other BAT-secreted adipokines. Conclusion/interpretation. – Women with GDM have significantly lower circulating neuregulin 4 levels compared with healthy pregnant controls, and the AUCGlucose is negatively and independently associated with neuregulin 4 during pregnancy. Neuregulin 4 is positively correlated with irisin during pregnancy, as well as in a longitudinal fashion. Future studies are now needed to better elucidate the precise pathomechanisms of the regulation of BAT-secreted adipokines during pregnancy.
C 2017 Elsevier Masson SAS. All rights reserved.
Keywords: Adipokine Brown adipose tissue Gestational diabetes mellitus Insulin resistance Neuregulin 4 Pregnancy
Introduction During the last few decades, the prevalence of gestational diabetes mellitus (GDM) has increased significantly [1]. As GDM contributes to an increased risk of acute and chronic complications Abbreviations: ADA, American Diabetes Association; ANGPTL8, Angiopoietin-like protein 8; AUCGlucose, Area under the glucose curve; AFABP, Adipocyte fatty acidbinding protein; BAT, Brown adipose tissue; BMI, Body mass index; ELISA, Enzymelinked immunosorbent assay; FFA, Free fatty acids; FGF21, Fibroblast growth factor 21; FI, Fasting insulin; GDM, Gestational diabetes mellitus; HDL, High-density lipoprotein; HOMA-IR, Homoeostasis model assessment of insulin resistance; LDL, Low-density lipoprotein; OGTT, Oral glucose tolerance test; TG, Triglyceride. * Corresponding author at: Liebigstr. 20, 04103 Leipzig, Germany. Fax: +49 341 9713389. E-mail address: [email protected] (T. Ebert). 1 These authors equally contributed to this work.
in both the mother and newborn [2], the pathophysiological mechanisms behind this metabolic disorder during pregnancy are of great interest. Thus, over the past several years, a variety of cytokines derived from adipose tissue, i.e. ‘adipokines’, have been identified as influencing both type 2 diabetes mellitus (T2DM) and GDM including adiponectin, leptin and adipocyte fatty acidbinding protein (AFABP)[3–6]. However, most adipokines are secreted predominantly by white adipose tissue and mediate adverse metabolic systemic effects. In contrast, several adipokines derived from brown adipose tissue (BAT) have recently been suggested to beneficially influence systemic metabolism, as reviewed by Villarroya et al. [7]. Neuregulin 4 has recently been recognized as a novel and predominantly BAT-secreted adipokine that protects against dietinduced insulin resistance and hepatic steatosis in mice [8]. Wang and co-workers [8] elegantly demonstrated that mice deficient in
http://dx.doi.org/10.1016/j.diabet.2017.06.001 C 2017 Elsevier Masson SAS. All rights reserved. 1262-3636/
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 2
S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
neuregulin 4 have increased insulin resistance and hepatic steatosis after high-fat feeding. Conversely, when mice with transgenic overexpression of neuregulin 4 were exposed to a highfat diet, they showed less body weight gain, and improved dyslipidaemia, as well as improved insulin sensitivity [8]. Mechanistically, the authors suggested that hepatic expression of lipogenic genes is lower in transgenic mice overexpressing neuregulin 4 compared with their littermate controls [8]. Taking these data together, neuregulin 4 appears to be a novel BATderived adipokine with beneficial effects on both glucose homoeostasis and hepatic steatosis. However, there are no studies investigating neuregulin 4 in GDM and in a longitudinal fashion during and after pregnancy, so far. In addition, the associations of neuregulin 4 with other BAT-secreted adipokines and markers of BAT function have not been tested in patients with GDM. Therefore, our present study has quantified circulating neuregulin 4 concentrations in 74 well-phenotyped women with GDM during pregnancy and compared them with 74 gestational-agematched, healthy pregnant controls. Neuregulin 4 was also correlated with clinical and biochemical measures of obesity, hypertension, indices of glucose metabolism, lipid metabolism, renal function and inflammation, as well as other markers of BAT function. Furthermore, neuregulin 4 was assessed longitudinally in 25 healthy (non-GDM) and 25 GDM women during pregnancy and postpartum in a follow-up study. Our hypothesis was that: women with GDM have decreased neuregulin 4 levels during pregnancy compared with non-diabetic, healthy pregnant women; circulating postpartum neuregulin 4 is increased because of reduced insulin resistance postpartum; and neuregulin 4 correlates with other markers of BAT function, including irisin, adiponectin, fibroblast growth factor (FGF)-21 and angiopoietin-like protein 8 (ANGPTL8) [7].
The present study was approved by the local ethics committee, and all subjects gave their written informed consent before taking part. Assays All blood samples were obtained after a fasting period of at least 8 h. Serum levels of neuregulin 4 (Phoenix Pharmaceuticals, Burlingame, CA, USA) were quantified using enzyme-linked immunosorbent assays (ELISAs) as per the manufacturer’s instructions. All other adipokines were quantified by ELISA kits as previously reported [9,11,19]. Fasting insulin (FI) was determined by two-site chemiluminescent enzyme immunometric assay, using a LIAISON automated analyzer (DiaSorin, Saluggia, Italy). All other parameters were measured by standard laboratory methods at a certified laboratory, as previously described [9–14]. Statistical analysis SPSS version 24.0 software (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Differences between women with GDM and the controls were assessed by non-parametric Mann– Whitney U test. Univariate correlations were performed using nonparametric Spearman’s rank-correlation method. Multivariate linear-regression analysis was subsequently performed to identify independent relationships. Before the multivariate correlation analyses were calculated, distribution was tested for normality, using the Shapiro–Wilk W test, and non-normally distributed parameters were logarithmically transformed. Longitudinal changes in neuregulin 4 during pregnancy compared with postpartum levels were analyzed by non-parametric Wilcoxon signed-rank test. A P-value < 0.05 was considered statistically significant in all analyses.
Results
Methods
Baseline characteristics
Study participants
The median [interquartile range] serum level of neuregulin 4 for the total sample during pregnancy was 3.3 [1.7] mg/L. Clinical characteristics of both subgroups (controls and GDM) are shown in Table 1. Median circulating neuregulin 4 levels were significantly lower in women with GDM (3.0 [1.2] mg/L) compared with the non-GDM pregnant controls (3.5 [1.9] mg/L; P = 0.020; Table 1). At the first time point during pregnancy, markers of glucose homoeostasis, including plasma glucose levels during OGTT, AUCGlucose, FI and HOMA-IR, as well as free fatty acids (FFAs), were significantly higher in women with GDM compared with the controls (P < 0.05; Table 1). In contrast, there were no significant differences in age, gestational age at blood sampling, gestational age at delivery, birth weight or markers of obesity, hypertension, dyslipidaemia, renal function or inflammation.
The design of the present study has been described previously [9–14]. In brief, 148 pregnant women were consecutively recruited from the outpatient care unit of the Department of Endocrinology and Nephrology, University of Leipzig, between 2006 and 2011. Based on the 2012 guidelines of the American Diabetes Association (ADA) [15], a 75-g 2-h oral glucose tolerance test (OGTT) was performed in all participants. As per ADA criteria, GDM was defined as having one or more elevated plasma glucose levels during OGTT, using the following thresholds: fasting plasma glucose (FPG) 5.1 mmol/L; 1-h FPG 10.0 mmol/L; and 2-h FPG 8.5 mmol/L. Based on these levels, 74 women were classified as patients with GDM, whereas 74 gestational-agematched pregnant women with normal glucose tolerance served as controls. Anthropometric measures included body mass index (BMI), determined by weight before pregnancy divided by squared height (kg/m2). In addition, the homoeostasis model assessment of insulin resistance (HOMA-IR) score and area under the glucose curve (AUCGlucose) were calculated, as previously described [16,17]. To investigate postpartum regulation of the novel BATsecreted factor, follow-up examination was carried out in 2012. For this postpartum subcohort, 25 women with previous GDM and 25 healthy former control women were included in the analysis. Relative changes (postpartum-to-prepartum ratios) were calculated for all adipokines and BAT markers as follows: parameter (ratio) = parameterpostpartum/parameterprepartum [18].
Univariate correlations On univariate correlation analysis of the entire cohort during pregnancy, neuregulin 4 positively correlated with systolic and diastolic blood pressure (P < 0.05) (Table 2). Also, the novel adipokine was negatively correlated with glycated haemoglobin (HbA1c), 1-h and 2-h glucose during OGTT, and AUCGlucose (P < 0.05; Table 2). In contrast, no significant correlations were established between circulating neuregulin 4 and markers of gestational outcome, obesity, dyslipidaemia, renal function or inflammation in the total study population (Table 2).
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx Table 1 Baseline characteristics of the study population during pregnancy.
n Neuregulin 4 (mg/L) Age (years) Gestational age at blood sampling (days) Gestational age at delivery (days) Infant birth weight (g) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) HbA1c (%) HbA1c (mmol/mol) Glucose 0 h(OGTT) (mmol/L) Glucose 1 h(OGTT) (mmol/L) Glucose 2 h(OGTT) (mmol/L) AUCGlucose (mmol/L) Fasting insulin (pmol/L) HOMA-IR Cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) Free fatty acids (mmol/L) Creatinine (mmol/L) hsCRP (mg/L) Leptin (mg/L)
Controls
GDM
P
74 3.5 (1.9) 28.9 (4.5) 199 (40) 275 (15) 3360 (790) 22.4 (6.7) 125 (17) 75 (13) 5.3 (0.6) 34.4 (6.6) 4.3 (0.5) 7.5 (1.6) 6.4 (1.8) 12.7 (2.4) 57.9 (38.4) 1.56 (0.96) 6.31 (1.84) 1.93 (0.51) 3.73 (1.57) 2.02 (1.43) 0.47 (0.30) 49.0 (11.0) 4.20 (4.30) 23.0 (11.9)
74 3.0 (1.2) 31.0 (7.5) 202 (33) 273 (14) 3400 (805) 24.5 (6.6) 120 (20) 73 (15) 5.4 (0.6) 35.5 (6.6) 4.5 (0.9) 10.1 (1.7) 8.7 (2.3) 16.6 (2.1) 70.6 (66.7) 1.98 (1.89) 6.71 (1.74) 1.82 (0.80) 4.05 (1.91) 2.14 (1.31) 0.55 (0.31) 46.0 (11.3) 3.99 (6.09) 26.5 (21.6)
0.020* 0.087 0.568 0.312 0.471 0.117 0.338 0.349 0.729 0.729 < 0.001* < 0.001* < 0.001* < 0.001* 0.003* < 0.001* 0.199 0.419 0.569 0.453 0.047* 0.086 0.903 0.131
Values are medians (interquartile range). GDM: gestational diabetes mellitus; HbA1c: glycated haemoglobin; OGTT: oral glucose tolerance test; AUCGlucose: area under the glucose curve; HOMA-IR: homoeostasis model assessment of insulin resistance; HDL/LDL: high-density/low-density lipoprotein; hsCRP: high-sensitivity C-reactive protein. * P < 0.05 (Mann–Whitney U test).
Multivariate regression analysis To identify independent associations, multivariate linearregression analysis was carried out. Here, AUCGlucose remained a
3
negative predictor of circulating neuregulin 4 after adjusting for age, gestational age, BMI, blood pressure and irisin (P = 0.033; Table 2). Also, diastolic blood pressure and irisin were significantly and independently associated with neuregulin 4 during pregnancy (P < 0.05; Table 2). Longitudinal changes in neuregulin 4 before and after delivery For all follow-up samples available for the present study (n = 50: 25 former controls; 25 women with previous GDM), the median circulating neuregulin 4 was 2.8 [1.7] mg/L during pregnancy and 3.2 [0.8] mg/L at the postpartum time point (P = 0.328, longitudinal analysis). When women with previous GDM and the former controls were analyzed separately, there was again no statistical difference in pre- and postpartum neuregulin 4 levels (Fig. 1). Interestingly, women with previous GDM had significantly higher postpartum neuregulin 4 levels (3.4 [0.8] mg/L) compared with the former controls (2.8 [0.9] mg/L; P = 0.032). However, there was no significant correlation with other metabolic markers postpartum (data not shown). Correlation of baseline neuregulin 4 and its relative changes with other BAT-secreted adipokines and markers of BAT function In all women during pregnancy (n = 148), prepartum neuregulin 4 serum levels were positively correlated to irisin levels (P = 0.047; Table 2) (Table 3). For the matched prepartum and postpartum samples of all women (n = 50), relative neuregulin 4 changes (ratio of neuregulin 4 = neuregulin 4postpartum/neuregulin 4prepartum) also correlated with relative changes of irisin (P = 0.005) (Table 4). Interestingly and unlike irisin, circulating neuregulin 4 levels, as well as its relative changes, were not associated with other BAT-secreted adipokines, such as adiponectin, FGF-21 and ANGPTL8 (Tables 3 and 4).
Table 2 Univariate correlations of serum neuregulin 4 in all women during pregnancy and multivariate regression analysis of neuregulin 4 (lg) as a dependent variable. Multivariate regression analysisa
Univariate correlations
Age (years) Gestational age at blood sampling (days) Gestational age at delivery (days) Infant birth weight (g) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) HbA1c (%) HbA1c (mmol/mol) Glucose 0 h(OGTT) (mmol/L) Glucose 1 h(OGTT) (mmol/L) Glucose 2 h(OGTT) (mmol/L) AUCGlucose (mmol/L) Fasting insulin (pmol/L) HOMA-IR Cholesterol (mmol/L) HDL cholesterol (mmol/L) LDL cholesterol (mmol/L) Triglycerides (mmol/L) Free fatty acids (mmol/L) Creatinine (mmol/L) hsCRP (mg/lL) Leptin (mg/L) Irisin (mg/L)
r
P
0.137 0.139 0.037 0.085 0.072 0.168 0.182 0.190 0.190 0.100 0.184 0.198 0.190 0.035 0.041 0.003 0.107 0.001 0.136 0.007 0.019 0.068 0.126 0.163
0.097 0.093 0.668 0.325 0.396 0.046b 0.031b 0.023b 0.023b 0.224 0.029b 0.020b 0.025b 0.674 0.621 0.973 0.196 0.991 0.099 0.935 0.817 0.411 0.127 0.047b
Standardized b coefficient 0.102 0.126 – – 0.144 – 0.190 – – – – – 0.180 – – – – – – – – – – 0.226
P 0.228 0.132 – – 0.102 – 0.027c – – – – – 0.033c – – – – – – – – – – 0.009c
HbA1c: glycated haemoglobin; OGTT: oral glucose tolerance test; AUCGlucose: area under the glucose curve; HOMA-IR: homoeostasis model assessment of insulin resistance; HDL/LDL: high-density/low-density lipoprotein; hsCRP: high-sensitivity C-reactive protein. a Non-normally distributed variables assessed by Shapiro–Wilk test were logarithmically transformed prior to multivariate testing. b Significant by Spearman’s correlation method. c Significant by multivariate analysis.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
4
Fig. 1. Longitudinal analysis of neuregulin 4 serum levels in a study subset (n = 50: 25 women with gestational diabetes mellitus (GDM); 25 controls). The overall P-value was assessed by non-parametric Wilcoxon signed-rank test.
Discussion In the present study, we demonstrate that women with GDM have lower serum levels of the novel BAT-secreted adipokine neuregulin 4 compared with the non-GDM pregnant controls. The AUCGlucose remains negatively and independently associated with circulating neuregulin 4 levels. Moreover, postpartum serum concentrations of neuregulin 4 do not significantly differ from those prepartum. However, women with previous GDM have increased neuregulin 4 compared with the former controls. Finally, the BAT-derived adipokine is positively associated with irisin, but not with other BAT-secreted adipokines [7].
Neuregulin 4 is a novel adipokine predominantly expressed in BAT [8]. Interestingly, both mice and humans with obesity have decreased Nrg4 mRNA expression, and mice deficient in Nrg4 display an adverse metabolic phenotype, including increased insulin resistance, dyslipidaemia (increased plasma TG) and hepatic steatosis [8]. More specifically, Nrg4-knockout mice fed a high-fat diet show elevated glucose concentrations during glucose tolerance tests [8]. Conversely, mice with transgenic overexpression of Nrg4 have lower glucose concentrations than wild-type control mice during glucose tolerance tests [8]. Taking these murine results into consideration, our human results in vivo are in almost perfect agreement demonstrating that the AUCGlucose during a 75-g OGTT is negatively associated with serum levels of neuregulin 4 even after adjusting for age, gestational age, markers of fat mass, blood pressure and irisin. These data for women with GDM are also in accordance with the findings of other cohorts investigating neuregulin 4 in metabolic disease states, including T2DM [20], the metabolic syndrome [21] and non-alcoholic fatty liver disease [22]. However, it should be noted that a recent study by Kang and co-workers [23] found increased neuregulin 4 serum levels in 57 South-Korean patients with newly diagnosed T2DM compared with controls. Hypothetically, different patient characteristics and ELISA systems might well explain the observed variations. Interestingly, several novel adipokines are not only adiposetissue-derived, but also co-secreted by the placenta during pregnancy, including irisin, fetuin B and preadipocyte factor 1 [9,14,24,25]. To determine whether the placenta significantly influences serum neuregulin 4 levels, we have performed a longitudinal analysis in 50 subjects (25 previous GDM patients and
Table 3 Univariate correlations of serum levels of adipokines with each other in all women during pregnancy (n = 148). During pregnancy
During pregnancy Neuregulin 4 Irisin ANGPTL8 Adiponectin FGF-21
r P r P r P r P r P
Neuregulin 4
Irisin
ANGPTL8
Adiponectin
– – 0.163 0.047a 0.090 0.274 0.084 0.311 0.044 0.592
– – 0.091 0.272 0.068 0.409 0.177 0.031a
– – 0.057 0.495 0.053 0.524
– – 0.267 0.001a
FGF-21
– –
ANGPTL8: angiopoietin-like protein 8; FGF-21: fibroblast growth factor 21. a Significant by Spearman’s correlation method.
Table 4 Univariate correlations of relative changes (postpartum/prepartum ratio) in adipokines with each other in all women included in the present analysis (n = 50). Relative changes (postpartum/prepartum)
Relative changes (postpartum/prepartum) Neuregulin 4 Irisin ANGPTL8 Adiponectin FGF-21
r P r P r P r P r P
Neuregulin 4
Irisin
– – 0.394 0.005a 0.144 0.317 0.187 0.306 0.371 0.052
– – 0.162 0.261 0.178 0.330 0.198 0.312
ANGPTL8
Adiponectin
FGF-21
– – 0.368 0.038a 0.165 0.401
– – 0.102 0.606
– –
ANGPTL8: angiopoietin-like protein 8; FGF-21: fibroblast growth factor 21. a Significant by Spearman’s correlation method.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001
G Model
DIABET-901; No. of Pages 5 S. Kralisch et al. / Diabetes & Metabolism xxx (2017) xxx–xxx
25 former controls). However, as there is no longitudinal change in circulating neuregulin 4 in our cohort, it would appear that placental co-secretion of neuregulin 4 does not significantly contribute to the observed results. In our follow-up cohort, women with previous GDM have increased neuregulin 4 levels in the postpartum state compared with the former controls. Further studies are now needed to validate this counterintuitive finding, and to assess whether neuregulin 4 can serve as a marker of adverse metabolic status in women. Besides markers of glucose tolerance, neuregulin 4 is also independently and positively associated with blood pressure. These data are in contrast to findings by Cai and co-workers [21], demonstrating lower neuregulin 4 levels in patients with hypertension than in normotensive subjects. However, the subjects included in that study had significantly higher median blood pressure than the women included in our present analysis. Thus, different patient characteristics may well explain these observed differences. Interestingly, both prepartum neuregulin 4 and its relative changes are positively correlated with irisin, another metabolically beneficial [26] adipokine/myokine influencing BAT function [27,28]. Yet, despite its low expression in BAT [28], irisin exerts beneficial metabolic effects, including increased energy expenditures and improved glucose homoeostasis, through BAT activation [27]. Based on our present study, it is tempting to speculate that both neuregulin 4 and irisin could act as cytokines improving metabolic status in GDM. In conclusion, women with GDM have significantly lower serum concentrations of neuregulin 4 compared with healthy (non-GDM) pregnant control women, and the AUCGlucose is negatively and independently associated with neuregulin 4 during pregnancy. Neuregulin 4 is also positively correlated with irisin during pregnancy, as well as in a longitudinal fashion. Future studies are now needed to better elucidate the role of BAT-secreted adipokines during pregnancy. Funding This work was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1501 (IFB AdiposityDiseases, project K6a-87) and by the Deutsche Forschungsgemeinschaft (DFG; SFB 1052/2, C06 to MF, B01 to MB, and A01 to MS). TE was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1001 (IFB AdiposityDiseases, MetaRot program). Author contributions T.E., S.K., and M.F. wrote the manuscript and researched data. A.H., and J.K. researched data and reviewed/edited the manuscript. M.B. and M.S. contributed to the discussion and reviewed/edited the manuscript. Dr Thomas Ebert is the guarantor of this work and, as such, had full access to all the data in the study, and takes responsibility for the integrity of the data and accuracy of the data analysis. Disclosure of interest The authors declare that they have no competing interest.
5
References [1] Lavery JA, Friedman AM, Keyes KM, Wright JD, Ananth CV. Gestational diabetes in the United States: temporal changes in prevalence rates between 1979 and 2010. BJOG 2017;124:804–13. [2] Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol 2012;8:639–49. [3] Fasshauer M, Paschke R. Regulation of adipocytokines and insulin resistance. Diabetologia 2003;46:1594–603. [4] Fasshauer M, Blu¨her M. Adipokines in health and disease. Trends Pharmacol Sci 2015;36:461–70. [5] Fasshauer M, Blu¨her M, Stumvoll M. Adipokines in gestational diabetes. Lancet Diabetes Endocrinol 2014;2:488–99. [6] Miehle K, Stepan H, Fasshauer M. Leptin, adiponectin and other adipokines in gestational diabetes mellitus and pre-eclampsia. Clin Endocrinol 2012;76:2–11. [7] Villarroya F, Cereijo R, Villarroya J, Giralt M. Brown adipose tissue as a secretory organ. Nat Rev Endocrinol 2017;13:26–35. [8] Wang G-X, Zhao X-Y, Meng Z-X, Kern M, Dietrich A, Chen Z, et al. The brown fat-enriched secreted factor Nrg4 preserves metabolic homeostasis through attenuation of hepatic lipogenesis. Nat Med 2014;20:1436–43. [9] Ebert T, Stepan H, Schrey S, Kralisch S, Hindricks J, Hopf L, et al. Serum levels of irisin in gestational diabetes mellitus during pregnancy and after delivery. Cytokine 2014;65:153–8. [10] Ebert T, Hindricks J, Kralisch S, Lossner U, Jessnitzer B, Richter J, et al. Serum levels of fractalkine are associated with markers of insulin resistance in gestational diabetes. Diabet Med 2014;31:1014–7. [11] Ebert T, Kralisch S, Wurst U, Lo¨ssner U, Kratzsch J, Blu¨her M, et al. Betatrophin levels are increased in women with gestational diabetes mellitus compared to healthy pregnant controls. Eur J Endocrinol 2015;173:1–7. [12] Platz M, Stepan H, Schrey S, Kralisch S, Wurst U, Lossner U, et al. Serum levels of sclerostin in cardiometabolic disorders during pregnancy. Cytokine 2015;76:591–3. [13] Ebert T, Platz M, Kralisch S, Lossner U, Jessnitzer B, Richter J, et al. Serum levels of copeptin are decreased in gestational diabetes mellitus. Exp Clin Endocrinol Diabetes 2016;124:257–60. [14] Kralisch S, Hoffmann A, Lo¨ssner U, Kratzsch J, Blu¨her M, Stumvoll M, et al. Regulation of the novel adipokines/hepatokines fetuin A and fetuin B in gestational diabetes mellitus. Metabolism 2017;68:88–94. [15] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2012;36:S67–74. [16] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9. [17] Enigk U, Breitfeld J, Schleinitz D, Dietrich K, Halbritter J, Fischer-Rosinsky A, et al. Role of genetic variation in the human sodium–glucose cotransporter 2 gene (SGLT2) in glucose homeostasis. Pharmacogenomics 2011;12:1119–26. [18] Ebert T, Hopf LM, Wurst U, Bachmann A, Kralisch S, Lo¨ssner U, et al. Circulating adipocyte fatty acid binding protein is increased in chronic and acute renal dysfunction. Nutr Metab Cardiovasc Dis 2014;24:1027–34. [19] Stein S, Stepan H, Kratzsch J, Verlohren M, Verlohren H-J, Drynda K, et al. Serum fibroblast growth factor 21 levels in gestational diabetes mellitus in relation to insulin resistance and dyslipidemia. Metabolism 2010;59:33–7. [20] Zhang L, Fu Y, Zhou N, Cheng X, Chen C. Circulating neuregulin 4 concentrations in patients with newly diagnosed type 2 diabetes: a cross-sectional study. Endocrine 2017. http://dx.doi.org/10.1007/s12020-017-1324-3 [Epub ahead of print]. [21] Cai C, Lin M, Xu Y, Li X, Yang S, Zhang H. Association of circulating neuregulin 4 with metabolic syndrome in obese adults: a cross-sectional study. BMC Med 2016;14:165. [22] Dai Y-N, Zhu J-Z, Fang Z-Y, Zhao D-J, Wan X-Y, Zhu H-T, et al. A case–control study: association between serum neuregulin 4 level and non-alcoholic fatty liver disease. Metabolism 2015;64:1667–73. [23] Kang YE, Kim JM, Choung S, Joung KH, Lee JH, Kim HJ, et al. Comparison of serum Neuregulin 4 (Nrg4) levels in adults with newly diagnosed type 2 diabetes mellitus and controls without diabetes. Diabetes Res Clin Pract 2016;117:1–3. [24] Wurst U, Ebert T, Kralisch S, Stumvoll M, Fasshauer M. Serum levels of the adipokine Pref-1 in gestational diabetes mellitus. Cytokine 2015;71:161–4. [25] Schrey S, Wurst U, Ebert T, Kralisch S, Drewlo S, Stepan H, et al. The adipokine preadipocyte factor-1 is downregulated in preeclampsia and expressed in placenta. Cytokine 2015;75:338–43. [26] Ebert T, Kralisch S, Wurst U, Scholz M, Stumvoll M, Kovacs P, et al. Association of metabolic parameters and rs726344 in FNDC5 with serum irisin concentrations. Int J Obes 2016;40:260–5. [27] Bostro¨m P, Wu J, Jedrychowski MP, Korde A, Ye L, Lo JC, et al. A PGC1-adependent myokine that drives brown-fat-like development of white fat and thermogenesis. Nature 2012;481:463–8. [28] Roca-Rivada A, Castelao C, Senin LL, Landrove MO, Baltar J, Crujeiras AB, et al. FNDC5/Irisin is not only a myokine but also an adipokine. PLoS One 2013;8:e60563.
Please cite this article in press as: Kralisch S, et al. The brown-fat-secreted adipokine neuregulin 4 is decreased in gestational diabetes mellitus. Diabetes Metab (2017), http://dx.doi.org/10.1016/j.diabet.2017.06.001