DIABETICMedicine DOI: 10.1111/dme.12436
Short Report: Pathophysiology Decreased plasma chemerin levels in women with gestational diabetes mellitus K. J. Hare1,2, L. Bonde1,3, J. A. Svare4, H. S. Randeva5, M. Asmar6, S. Larsen1, T. Vilsbøll1 and F. K. Knop1,7 1 Diabetes Research Division, Department of Medicine, Gentofte Hospital, University of Copenhagen, Hellerup, Denmark, 2Department of Obstetrics and Gynaecology, Hvidovre Hospital, University of Copenhagen, Hvidovre, Denmark, 3Department of Obstetrics and Gynaecology, Roskilde Hospital, University of Copenhagen, Roskilde, Denmark, 4Department of Obstetrics and Gynaecology, Herlev Hospital, 5Warwick Medical School, University of Warwick, Coventry, UK, 6 Department of Endocrinology, Hvidovre Hospital, University of Copenhagen, Hvidovre, Denmark and 7Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
Accepted 10 March 2014
Abstract Aims To evaluate fasting and post-prandial serum chemerin levels in pregnant women with and without gestational diabetes, and again following delivery when normal glucose homeostasis is re-established. Methods Chemerin levels were measured in serum from nine women with gestational diabetes, and from eight age- and BMI-matched pregnant women with normal glucose tolerance during two meal tests: in the third trimester and 3–4 months post partum. All women with gestational diabetes re-established normal glucose tolerance after delivery.
Meal intake did not affect serum chemerin levels. The group with gestational diabetes had lower mean serum chemerin levels during the third trimester compared with the group with normal glucose tolerance (28 1.3 vs. 88 3.5 ng/ml, P < 0.0001). In the group with normal glucose tolerance, mean serum chemerin levels decreased significantly post partum to 57 2.8 ng/ml (P = 0.0001), but remained significantly (P = 0.0003) higher than postpartum levels in the group with gestational diabetes (31 1.9 ng/ml), which did not differ significantly from third trimester levels (P = 0.31). Results
Conclusions Normal pregnancy is associated with increased circulating chemerin levels, which may act to reduce pregnancy-induced insulin resistance and prevent glucose intolerance. Women with gestational diabetes, however, have severely reduced chemerin levels that remain low after delivery, which may contribute to the insulin resistance, glucose intolerance and high Type 2 diabetes risk associated with gestational diabetes.
Diabet. Med. 00, 000–000 (2014)
Introduction Gestational diabetes mellitus is characterized by insulin resistance and glucose intolerance that usually resolves post partum. Nevertheless, gestational diabetes represents an independent risk factor for the later development of Type 2 diabetes, with a hazard ratio of up to 17 . The mechanisms responsible for the insulin resistance associated with gestational diabetes are poorly understood. The adipokine chemerin has been suggested to constitute an insulin-sensitizing factor , which may counteract insulin resistance. In line with this, circulating chemerin levels are elevated in obesity and Type 2 diabetes  and
have been shown to correlate with insulin resistance independently of BMI and fasting insulin . During pregnancy, serum chemerin has been reported to increase with gestational age (as insulin sensitivity attenuates) in lean pregnant women with normal glucose tolerance , whereas no differences in serum levels were reported between lean and overweight pregnant women at term, with or without gestational diabetes [5–7]. In the present study, we evaluated serum chemerin levels in pregnant women with normal glucose tolerance and in women with gestational diabetes, using 4-h liquid meal tests both during the third trimester and in the post-partum state, 3–4 months after delivery, when normal glucose tolerance had been re-established.
Correspondence to: Filip K. Knop. E-mail: [email protected]
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Plasma chemerin levels in gestational diabetes K. J. Hare et al. Table 1 Characteristics of the two study groups
What’s new? Study group
• The mechanisms responsible for the severe insulin resistance associated with gestational diabetes mellitus are poorly understood. • We investigated levels of the novel adipokine chemerin (which has been suggested to constitute an insulinsensitizing factor) in pregnant women with and without gestational diabetes during the third trimester, and in the post-partum state when normal glucose homeostasis had been re-established. • We show that normal pregnancy is associated with increased circulating chemerin levels that decrease post partum, whereas gestational diabetes seems to be characterized by reduced chemerin levels that do not change with normalization of glucose tolerance following delivery. • Our findings may help to explain the severe insulin resistance observed in gestational diabetes and the high risk of Type 2 diabetes associated with it.
Subjects and Methods Subjects, experimental design and trial registration
Serum, collected during the third trimester of pregnancy and 3–4 months post partum, from 17 white women participating in a previously published study  was analysed. Detailed subject characteristics have been published previously  and are shown in Table 1. Briefly, nine of the women were diagnosed with gestational diabetes according to Danish guidelines, i.e. 2-h plasma glucose concentration after 75 g oral glucose tolerance test ≥ 9.0 mM. The remaining women, with 2-h plasma glucose < 9 mM were enrolled as control subjects with normal glucose tolerance. None of the women with gestational diabetes received antidiabetic treatment. Patients with gestational diabetes returned to normal glucose homeostasis 3–4 months after delivery, which was verified by 75 g oral glucose tolerance test (fasting plasma glucose < 7.0 mM and 2-h plasma glucose < 7.8 mM). Detailed descriptions of the study design and experimental procedures have been published previously . Briefly, after screening, the women in their third trimester of pregnancy underwent a standardized 4-h liquid meal test [100 g NAN 1 (Nestle Nutrition, Nunspet, Netherlands): 2170 kJ, 9.5% protein, 58% carbohydrate and 28% fat] after an overnight fast. All participants underwent a similar meal test 3– 4 months after delivery. The original protocol and the present investigation were approved by the Scientific Ethical Committee of the Capital Region of Denmark (KA-060015 and H-2-2011-041). The
Characteristic Age, years BMITT, kg/m2 BMIPP, kg/m2 P† Gestational age at third trimester examination, days Birth weight, g HbA1cTT (%, mmol/mol) HbA1cPP (%, mmol/mol) P† FPGTT, mM FPGPP, mM P† FPITT, pM FPIPP, pM P†
Gestational diabetes (n = 9)
Normal glucose tolerance (n = 8)
31.0 26.9 29.1 0.006 222
1.9 2.3 2.5
27.0 25.6 27.4 0.012 212
1.1 2.1 1.8
0.190 0.717 0.568
3299 5.6 5.5 0.42 4.9 5.6 0.022 111 77 0.001
163 0.2 0.1
3624 5.4 5.3 0.18 4.6 5.2 0.001 79 66 0.360
268 0.2 0.1
0.326 0.434 0.232
0.2 0.2 27 25
*Gestational diabetes vs normal glucose tolerance. †Third trimester vs post partum. Data are mean values SEM with P values from paired two-tailed Student t-tests within the groups and from unpaired analyses between the two groups. TT, third trimester; PP, post partum; FPG, fasting plasma glucose; FPI, fasting plasma insulin.
original study was registered with ClinicalTrials.gov (NCT 01274052) and conducted according to the principles of the Helsinki Declaration II.
Chemerin levels in sera were measured using a commercially available enzyme-linked immunosorbent assay kit (R & D Systems, Abingdon, UK) according to the manufacturer’s protocol, with an intra-assay coefficient of variation of < 9%. Calculations and statistical analyses
The data were compared using paired two-tailed Student ttests within the groups and comparisons between the two groups were made using unpaired analyses. ANOVA with post hoc analysis was applied. P values < 0.05 were considered to indicate statistical significance.
Results Chemerin data are shown in Fig. 1. Mean SEM fasting serum chemerin levels were 28 3.0 and 88 4.2 ng/ml (P < 0.0001) in the group with gestational diabetes and normal glucose tolerance, respectively, during the third trimester, and 30 3.8 and 57 5.9 ng/ml (P = 0.001) post partum. Meal intake did not affect serum chemerin levels during the third trimester or post partum in either of
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
FIGURE 1 Serum chemerin levels during a 4-h liquid meal test in pregnant women with gestational diabetes mellitus (triangles) and normal glucose tolerance (circles) during third trimester of pregnancy (closed symbols) and 3–4 months post partum (open symbols), when normal glucose homeostasis was re-established in all women. Each symbol represents the mean values SEM in the nine women with gestational diabetes, and in the eight women with normal glucose tolerance. Meal intake did not affect the serum chemerin levels significantly (P > 0.05) in any of the groups; neither during the third trimester meal test or the post partum meal test. Asterisks indicate significantly higher (P < 0.001) third trimester levels in women with normal glucose tolerance compared with both third trimester and postpartum levels in the gestational diabetes group (between which no significant differences were found; P > 0.05). Hash tags indicate significant differences between third trimester and post-partum levels in the group with normal glucose tolerance.
the groups. The group with gestational diabetes was observed to have much lower mean serum chemerin levels during the third trimester compared with the group with normal glucose tolerance (28 1.3 vs. 88 6.2 ng/ml, P < 0.0001). In the group with normal glucose tolerance, chemerin levels decreased significantly after delivery to 57 5.9 ng/ml (P = 0.0001), but remained significantly higher (P = 0.0003) than post-partum levels in the group with gestational diabetes (30 3.8 ng/ml), which did not differ from third trimester levels (P = 0.31).
Discussion In the present study we showed that normal pregnancy was associated with increased circulating chemerin levels that decrease post partum, whereas gestational diabetes seems to be characterized by reduced chemerin levels that do not change with normalization of glucose tolerance following delivery. This may help to explain the severe insulin resistance in and the high Type 2 diabetes risk associated with gestational diabetes. Furthermore, we show that serum chemerin levels in the present cohort of white women did not change in response to ingestion of a mixed liquid meal. An increase in circulating levels of chemerin with gestational age has been described earlier , but serum chemerin ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
levels during pregnancy have, to our knowledge, not been compared with corresponding levels in the same individuals following delivery, either in women with gestational diabetes or in women with normal glucose tolerance. Neither have chemerin levels in the post-prandial state, to our knowledge, been investigated previously. We found postprandial chemerin levels of particular interest in women with gestational diabetes because of the impact of pregnancy-induced insulin resistance on post-absorptive glucose excursions and often unaffected fasting blood glucose levels. We describe a significant difference in third trimester chemerin levels between gestational diabetes and normal glucose tolerance pregnancies. The two groups were matched according to gestational age and BMI, the majority of women having a BMI of 25–30 kg/m2 during pregnancy . As chemerin has been suggested to be an insulinsensitizing adipokine and its secretion has been shown to increase (perhaps as a compensatory mechanism) in insulinresistant states (in obese subjects and in subjects with Type 2 diabetes) [3,7,9,10], we expected to find increased serum chemerin levels (especially during the third trimester) in gestational diabetes; however, we found the exact opposite. During the third trimester, the women with gestational diabetes were found to have significantly lower circulating chemerin levels compared with the pregnant women with normal glucose tolerance. When examining the same women after delivery, when all participants had regained normal glucose tolerance, this paradoxical and unexpected difference persisted; however, after delivery the difference was diminished as the serum chemerin levels in women with normal glucose tolerance decreased significantly, and the corresponding levels in women with gestational diabetes did not change. We speculate, therefore, that the inability to increase serum chemerin levels during pregnancy among women with gestational diabetes may contribute to their insulin resistance and thereby to the development of the disease. Furthermore, the significantly lower serum chemerin levels observed in these women after delivery, when normal glucose tolerance was re-established, could constitute an important factor in the dramatically increased risk of Type 2 diabetes among women with previous gestational diabetes. Both Barker et al.  and Pfau et al.  have published data comparing circulating chemerin levels during pregnancy (at time of admission for elective caesarean section at term  and at gestational age ~200 days ) in women with normal glucose tolerance and gestational diabetes, and recently, van Poppel et al.  reported circulating chemerin levels in women with and without gestational diabetes during gestational week 32 and at term. In neither of these studies were differences in maternal chemerin levels between women with and without gestational diabetes observed. This discrepancy between the present findings and earlier reports may be attributable to several factors. Pfau et al.  report an independent association between chemerin levels
Plasma chemerin levels in gestational diabetes K. J. Hare et al.
and insulin resistance (according to the homeostatic model assessment). Importantly, the participants with gestational diabetes and those with normal glucose tolerance studied by Pfau et al. were matched for fasting insulin levels, indicating that both groups may have been insulin-resistant. Consistent with this, homeostatic model assessment of insulin resistance was similar in the two groups. Furthermore, the group with normal glucose tolerance represented individuals at high risk (> 60% had a family history of Type 2 diabetes, 25% had been overweight/obese before pregnancy, and ~75% had had a pathological 50 g oral glucose tolerance test with 1-h plasma glucose value > 7.8 mM during pregnancy); thus, the lack of difference in chemerin levels in the groups may be explained by similar insulin resistance and the high-risk profile of the control group. Likewise, both groups of women (with gestational diabetes and normal glucose tolerance, respectively) studied by Barker et al.  were characterized by high BMIs; 85% of participants had a BMI > 30 kg/m2 at the time of delivery. In the study by van Poppel et al. , almost 30% of the women were obese before pregnancy (the distribution of obesity across the gestational diabetes and non-gestational diabetes groups was not reported) and, interestingly, these women tended to be more glucose-tolerant compared with the non-obese women. Also, the women without gestational diabetes tended to exhibit higher chemerin levels than those with gestational diabetes (although this difference did not reach statistical significance) . Thus, the discrepancies between previously published chemerin data in pregnant women with and without gestational diabetes [5–7] and the present results may be explained by our ‘clean’ non-obese control group with normal glucose tolerance and without any family history of diabetes (and negative glutamic acid decarboxylase-65 and islet autoantibodies) compared with the high-risk control groups in the studies by Pfau et al., Barker et al. and van Poppel et al. Interestingly, Barker et al.  suggest that the good glycaemic control among their women with gestational diabetes (some of whom were treated with insulin) may explain the similar plasma chemerin levels observed in their groups of women with gestational diabetes and normal glucose tolerance, respectively. The present study shows that circulating chemerin levels remain unchanged in the post-prandial state. This supports the notion that serum chemerin mainly originates from adipose tissue, which does not seem to respond acutely (with detectable changes in serum concentrations)  to increases in, for example, circulating glucose and insulin after meal ingestion. The present data are limited by the relatively low number of women participating in the study. Despite this, clear and significant differences were observed. One of the strengths of the present experimental design is that the women acted as their own controls; i.e. the disturbed glucose homeostasis and its impact on chemerin levels during pregnancy are
compared with chemerin levels during a normo-glycaemic state (following delivery) in the same individual. Confounding effects attributable to subject inter-variability are thereby avoided. In conclusion, our results suggest, firstly, that normal pregnancy is associated with increased circulating chemerin levels, perhaps important for adaption to pregnancy-induced insulin resistance and prevention of glucose intolerance, and secondly, that gestational diabetes is characterized by reduced chemerin levels during the third trimester and post partum; perhaps contributing to the insulin resistance, glucose intolerance, and high Type 2 diabetes risk associated with gestational diabetes. Clearly, mechanistic studies are needed to confirm this hypothesis.
The present study was supported by unrestricted grants from the Danish Diabetes Association, Aase and Ejnar Danielsen’s Foundation and the Novo Nordisk Foundation.
We thank all participants for spending time on this project and we are grateful for expert experimental assistance from Jette Blach, Tove Laursen, Helle H. Lethmar and Trine Nielsen.
References 1 Feig DS, Shah BR, Lipscombe LL, Wu CF, Ray JG, Lowe J et al. Preeclampsia as a risk factor for diabetes: a population-based cohort study. PLoS Med 2013; 10: e1001425. 2 Bergmann K, Sypniewska G. Diabetes as a complication of adipose tissue dysfunction. Is there a role for potential new biomarkers? Clin Chem Lab Med 2013; 51: 177–185. 3 Zabel BA, Allen SJ, Kulig P, Allen JA, Cichy J, Handel TM et al. Chemerin activation by serine proteases of the coagulation, fibrinolytic, and inflammatory cascades. J Biol Chem 2005; 280: 34661–34666. 4 Garces MF, Sanchez E, Ruiz-Parra AI, Rubio-Romero JA, AngelMuller E, Suarez MA et al. Serum chemerin levels during normal human pregnancy. Peptides 2013; 42: 138–143. 5 Barker G, Lim R, Rice GE, Lappas M. Increased chemerin concentrations in fetuses of obese mothers and correlation with maternal insulin sensitivity. J Matern Fetal Neonatal Med 2012; 25: 2274–2280. 6 Pfau D, Stepan H, Kratzsch J, Verlohren M, Verlohren HJ, Drynda K et al. Circulating levels of the adipokine chemerin in gestational diabetes mellitus. Horm Res Paediatr 2010; 74: 56– 61. 7 van Poppel MN, Zeck W, Ulrich D, Schest EC, Hirschmugl B, Lang U et al. Cord blood chemerin: differential effects of gestational diabetes mellitus and maternal obesity. Clin Endocrinol 2014; 80: 65–72.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
8 Bonde L, Vilsbøll T, Nielsen T, Bagger JI, Svare JA, Holst JJ et al. Reduced postprandial GLP-1 responses in women with gestational diabetes mellitus. Diabetes Obes Metab 2013; 15: 713–720. 9 El-Mesallamy HO, El-Derany MO, Hamdy NM. Serum omentin-1 and chemerin levels are interrelated in patients with Type 2 diabetes mellitus with or without ischaemic heart disease. Diabet Med 2011; 28: 1194–1200.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
10 Fatima SS, Bozaoglu K, Rehman R, Alam F, Memon AS. Elevated chemerin levels in Pakistani men: an interrelation with metabolic syndrome phenotypes. PLoS One 2013; 8: e57113. 11 Bozaoglu K, Bolton K, McMillan J, Zimmet P, Jowett J, Collier G et al. Chemerin is a novel adipokine associated with obesity and metabolic syndrome. Endocrinology 2007; 148: 4687– 4694.