European Journal of Obstetrics & Gynecology and Reproductive Biology 168 (2013) 26–29
Contents lists available at SciVerse ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Adipokines and their relation to maternal energy substrate production, insulin resistance and fetal size Fredrik Ahlsson a,*, Barbro Diderholm a, Uwe Ewald a, Bjo¨rn Jonsson a, Anders Forslund a, Mats Stridsberg b, Jan Gustafsson a a b
Department of Women’s and Children’s Health, Uppsala University, Uppsala, Sweden Department of Medical Sciences, Uppsala University, Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 July 2012 Received in revised form 24 November 2012 Accepted 6 December 2012
Objective: The role of adipokines in the regulation of energy substrate production in non-diabetic pregnant women has not been elucidated. We hypothesize that serum concentrations of adiponectin are related to fetal growth via maternal fat mass, insulin resistance and glucose production, and further, that serum levels of leptin are associated with lipolysis and that this also influences fetal growth. Hence, we investigated the relationship between adipokines, energy substrate production, insulin resistance, body composition and fetal weight in non-diabetic pregnant women in late gestation. Study design: Twenty pregnant women with normal glucose tolerance were investigated at 36 weeks of gestation at Uppsala University Hospital. Levels of adipokines were related to rates of glucose production and lipolysis, maternal body composition, insulin resistance, resting energy expenditure and estimated fetal weights. Rates of glucose production and lipolysis were estimated by stable isotope dilution technique. Results: Median (range) rate of glucose production was 805 (653–1337) mmol/min and that of glycerol production, reflecting lipolysis, was 214 (110–576) mmol/min. HOMA insulin resistance averaged 1.5 0.75 and estimated fetal weights ranged between 2670 and 4175 g ( 0.2 to 2.7 SDS). Mean concentration of adiponectin was 7.2 2.5 mg/L and median level of leptin was 47.1 (9.9–58.0) mg/L. Adiponectin concentrations (7.2 2.5 mg/L) correlated inversely with maternal fat mass, insulin resistance, glucose production and fetal weight, r = 0.50, p < 0.035, r = 0.77, p < 0.001, r = 0.67, p < 0.002, and r = 0.51, p < 0.032, respectively. Leptin concentrations correlated with maternal fat mass and insulin resistance, r = 0.76, p < 0.001 and r = 0.73, p < 0.001, respectively. There was no correlation between maternal levels of leptin and rate of glucose production or fetal weight. Neither were any correlations found between levels of leptin or adiponectin and maternal lipolysis or resting energy expenditure. Conclusion: The inverse correlations between levels of maternal adiponectin and insulin resistance as well as endogenous glucose production rates indicate that low levels of adiponectin in obese pregnant women may represent one mechanism behind increased fetal size. Maternal levels of leptin are linked to maternal fat mass and its metabolic consequences, but the data indicate that leptin lacks a regulatory role with regard to maternal lipolysis in late pregnancy. ß 2012 Elsevier Ireland Ltd. All rights reserved.
Keywords: Adiponectin Fetal weight Glucose production rate Leptin Lipolysis
1. Introduction The number of large for gestational age (LGA) infants, born of non-diabetic women, has increased during the last decades [1]. Infants born LGA are at increased risk for perinatal complications as well as metabolic disease later in life [2–4]. Maternal overweight
* Corresponding author at: Department of Women’s and Children’s Health, Uppsala University, SE-751 85 Uppsala, Sweden. Tel.: +46 18 6115885; fax: +46 18 6115853. E-mail address:
[email protected] (F. Ahlsson). 0301-2115/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejogrb.2012.12.009
and obesity are related to increased infant size [3]. During pregnancy several physiological alterations occur, one of which is a metabolic adaptation to ensure an optimal supply of glucose, amino acids and triacylglycerols to the fetus. The mechanisms behind this process, however, are yet to be elucidated. The adipokines, leptin and adiponectin, are of interest in this respect, given their involvement in satiety control, regulation of body fatness, insulin resistance and energy expenditure [5]. Several studies on the relationship between adipokines and body composition, energy expenditure, and insulin resistance during pregnancy have been performed, but so far no reports on the relationship between adipokines and maternal energy substrate
F. Ahlsson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 168 (2013) 26–29
production, glucose production rate and lipolysis have been published. During pregnancy serum concentrations of leptin increase by the action of pregnancy-specific hormones and as a consequence of central leptin resistance [6]. Highman et al. [7] reported a strong correlation between maternal fat mass and leptin concentrations in early, mid- and late pregnancy. Since leptin is considered to be a lipolytic hormone, a relationship between the levels of leptin and the increased lipolysis in pregnancy might be anticipated. It is well established that serum levels of adiponectin relates inversely with insulin resistance in both the pregnant and the non-pregnant state [8]. As far as associations between adiponectin and fat mass or BMI are concerned, diverging results have been reported [9–12]. Rates of glucose production as well as lipolysis are related to fetal growth [13], but the association between adipokines and energy substrate production and fetal size in late pregnancy has not yet been established. Our hypothesis was that serum concentrations of adiponectin are related to fetal growth via maternal fat mass, insulin resistance and glucose production rate. Further, we hypothesized that serum levels of leptin are associated with lipolysis and that this also influences fetal growth. The aims of the study were to investigate the relationship between adipokines and maternal energy substrate production rate in late pregnancy and to study the association between adipokines and maternal body composition, insulin resistance and fetal weight. 2. Materials and methods 2.1. Subjects The investigation comprised twenty healthy, non-smoking and non-diabetic pregnant women. Data on the relationships between maternal fat mass and energy substrate production in this cohort have been reported recently [13]. Three of the women were treated with thyroxine for hypothyroidism. All women in the study had normal serum levels of TSH, f-T4 and T3 at the time of investigation. The recruitment process has been described in detail previously [13]. We recruited twenty pregnant women for this study; ten of the women had previously given birth to an LGA infant. Large for gestational age was defined as having a birth weight > 2 standard deviations above the mean birth weight (i.e. close to the 97th percentile) for gestational age according to the Swedish birth standard [14]. Of the 20 infants, six were born LGA of women who had previously given birth to LGA infants. This approach was chosen in order to create a study population of women carrying fetuses with a wide range of sizes, from normal to large. The estimated fetal weights ranged from 2670 to 4175 g ( 0.2 to 2.7 SDS), (mean 3193 454 g). Consent was obtained after oral and written information. The study was approved by the Human Ethics Committee of the Medical Faculty of the University of Uppsala. The pregnancies were dated by ultrasound examination at 16– 18 weeks of gestation. The maternal pre-pregnancy characteristics are presented in Table 1. Five women were overweight (BMI > 25) and three women were obese (BMI > 30). The pregnant women were screened for gestational diabetes by random, non-fasting blood glucose levels four times during Table 1 Pre-pregnancy characteristics of the women in the study (n = 20). cf. [13]. Mean SD Age (years) Height (cm) Parity Pre-pregnancy weight (kg) Pre-pregnancy BMI (kg/m2)
Median
Range
1 66.5 23.9
0–3 55–112 20.0–39.7
33.0 4.8 167.7 4.6
27
pregnancy (weeks 10–14, 20–24, 28–32, and 32–36). An abnormal glucose screening result was defined as a random plasma glucose above 9 mmol/L [15]. Two of the women were submitted to a 2 h oral glucose tolerance test (OGTT) with 75 g of glucose because of high screening blood glucose values. The results of the tolerance tests were normal. The women’s HbA1c levels were normal, 4.3 0.4% (mean SD). Eighteen of the infants were delivered vaginally and two by cesarean sections, performed at the mother’s request. All pregnancies were without complications. 2.2. Methods Glucose production rate and lipolysis were estimated by a stable isotope dilution technique. The following isotopic tracers were used to estimate rates of glucose production (GPR) and lipolysis: [6,6-2H2]-glucose (isotopic purity 98 atom%) and [1,1,2,3,3-2H5]-glycerol (isotopic purity 98 atom%), purchased from Cambridge Isotope Laboratories, Woburn, MA, USA. Glycerol and glucose production rates were calculated from isotopic enrichments of [1,1,2,3,3-2H5]-glycerol and [6,6-2H2]-glucose obtained during periods of approximate steady state [13]. The mean coefficients of variation for enrichments of glycerol and glucose were 10 5% and 2 1%, respectively. Standard curves obtained by gradually increasing the amounts of labeled glycerol and glucose in relation to the corresponding unlabelled compounds were used. Glycerol production rates and GPR were calculated as follows: production rate = i 100/IR; where i is the infusion rate of the tracer and IR is the isotopic ratio of the tracer in plasma [i.e. labeled (tracer)/ unlabelled substrate in %] [13]. 2.3. Chemical procedures Biochemical analyses were performed by established routine methods at the certified laboratory of the Department of Clinical Chemistry at the University Hospital in Uppsala. The samples were frozen at 20 8C until analyzed. Measurements of routine clinical chemistry analytes and hormones were performed on an Architect Ci82001 analyser (Abbott, Abbot Park, IL, USA) or on an automated immunoassay system (Modular E170, Roche Diagnostics GmbH, Mannheim, Germany). HbA1c was measured on a Variant II HPLC system (Bio-Rad Laboratories, Hercules, CA, USA). Concentrations of serum adiponectin and serum leptin were measured by commercial ELISA-kits (Linco Research Inc., St. Charles, MI, USA). The isotopic enrichments of [1,1,2,3,3-2H5]-glycerol and [6,6-2H2]-glucose were determined as described previously [13]. Insulin resistance was assessed by using the HOMA Calculator 2.2 program (Diabetes Research Laboratory, Oxford, United Kingdom) [16]. Resting energy expenditure was measured with an ergospirometer (Sensormedics 2900Z, Anaheim, CA, USA). The women were considered to be at rest during the last 15 minutes of the measuring period [17], and the mean values during this period were used to calculate resting energy expenditure [18]. Body composition was assessed with a three-compartment model combining measurements of multi-frequency bio-impedance (Xitron Hydra, San Diego, USA) and skinfold thickness (John Bull, British Indicators, St. Albans, England, UK). 2.4. Study design The women were studied at a mean length of gestation of 36 1.0 weeks. They were fasted and rested in bed from 10 pm in the evening of admission and through the whole investigation [13]. Two peripheral vein catheters were inserted at 5 am on the morning after admission, one for collection of blood samples and one for infusion of the tracers. The maternal body composition was assessed at 8 am
28
F. Ahlsson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 168 (2013) 26–29
Table 2 Maternal and fetal characteristics. cf. [13]. N = 20 Maternal Post menstrual weeks at study Weight at study (kg) BMI at study (kg/m2) Fat-free mass (kg) Fat mass (kg) Resting energy expenditure (kcal/day) S-adiponectin (mg/L) (N = 18) S-leptin (mg/L) (N = 18) Offspring Estimated fetal weight (g) Estimated fetal weight SDS Birth weight (g) Birth weight SDS
Mean SD
Median
Range
83.5 29.4
73–117 26–41
47.1
9.9–58.0
36.4 1
55.8 6.3 30.5 7.5 1908.3 307.2 7.2 2.5
3193.3 454.4
2670–4175 0.2 to 2.7 3180–5040 0.3 to 2.8
4092 406
SDS, standard deviation score.
followed by indirect calorimetry at 9 am measuring resting energy expenditure, after approximate 11 h of fasting. Subsequently, fetal ultrasound was performed to estimate fetal weight [19]. After 6 h of tracer infusion, five blood samples were taken with 10-min intervals. Considering the long fasting period before the examination enteral contributions of glycerol and glucose were considered to be negligible. 2.5. Statistical analysis The results are presented as mean SD, and as median and range if the variable could not be considered normally distributed. Deviation from normality was tested using the Kolmogorov–Smirnov test. Correlation analyses were performed using Spearman’s rho. The statistical package SPSS, version 15 (LEAD Technologies, Chicago, IL, USA) was used in the data analyses. Results were considered significant at p values less than 0.05.
3. Results Maternal and fetal characteristics are shown in Table 2. The median (range) rate of glucose production was 805 (653– 1337) mmol/min and that of glycerol production, reflecting lipolysis, was 214 (110–576) mmol/min [13]. HOMA insulin resistance averaged 1.5 0.75. Maternal fat mass at the time of the study averaged 30.5 5.3 kg and mean REE/day was 1908 307 kcal (Table 2). In 18 of the women levels of adiponectin and leptin were measured. The mean concentration of adiponectin was 7.2 2.5 mg/L (n = 18). Maternal serum levels of adiponectin at the time of investigation correlated inversely with fat mass (r = 0.50, p < 0.035). Insulin resistance, glucose production (Fig. 1), and fetal weight correlated inversely with serum levels of adiponectin, (r = 0.77, Table 3 Bivariate Spearman correlations for S-adiponectin and S-leptin (n = 18).
Total GPR (mmol/min) Total lipolysis (mmol/min) IR with HOMA Calculator Fat mass (kg) BMI at study (kg/m2) Resting energy expenditure (kcal/day) Estimated fetal weight (g) GPR, glucose production rate.
S-adiponectin (mg/L)
S-leptin (mg/L)
r= ns r= r= r= ns
ns ns r = 0.73, p < 0.001 r = 0.76, p < 0.000 r = 0.72, p < 0.001 ns
0.67, p < 0.002 0.77, p < 0.000 0.50, p < 0.035 0.69, p < 0.002
r = 0.51, p < 0.032
ns
Fig. 1. The correlation between maternal S-adiponectin and total glucose production (r = 0.67, p < 0.002) (n = 18).
p < 0.001), (r = 0.67, p < 0.002) and (r = 0.51, p < 0.032), respectively. The median level of leptin was 47.1 (9.9–58.0) mg/L (n = 18). Levels of leptin correlated with maternal fat mass and insulin resistance (r = 0.76, p < 0.001) and (r = 0.73, p < 0.001), respectively. There were no correlations between maternal levels of leptin and glucose production rate or fetal size. Nor were any correlations found between levels of leptin or adiponectin and maternal lipolysis (glycerol production) or maternal resting energy expenditure (Table 3).
4. Comments In the present study we investigated the association between adipokines and maternal fat mass, insulin resistance, energy substrate production, resting energy expenditure and fetal weight in a cohort of non-diabetic pregnant women with a wide range of estimated fetal weights ( 0.2 to 2.7 SDS). There was a strong inverse relationship between rates of maternal glucose production assessed with stable isotope dilution technique and levels of maternal adiponectin. This is in agreement with results from earlier studies on non-pregnant women with fatty liver disease and patients with type 2 diabetes [20,21]. The data are also in line with results on rodents demonstrating that adiponectin has a direct inhibitory effect on endogenous glucose production [22,23]. The present results indicate that the inhibitory effect on glucose production is preserved even during pregnancy. The data further show that maternal levels of adiponectin are not related to lipolysis in late pregnancy, which is in agreement with results by Catalano et al. [9] who reported that levels of NEFA in pregnant women are unrelated to those of adiponectin. Levels of adiponectin in the non-pregnant state are inversely related to body weight and body fat mass. The present data confirm that this is valid also during pregnancy. In addition, maternal insulin sensitivity is closely related to levels of adiponectin during late pregnancy, indicating a possible role for adiponectin in the decrease of insulin sensitivity with advancing gestation. On the hepatic level, reduced insulin sensitivity will result in a higher rate of glucose production, in turn contributing to an increased fetal growth [13]. Consequently, reduced levels of adiponectin may represent one mechanism by which a high maternal fat mass leads to an increased fetal weight.
F. Ahlsson et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 168 (2013) 26–29
In line with previous investigations [11,24], maternal levels of leptin were associated with fat mass and BMI. There was, however, no association between rate of glycerol production (lipolysis) and levels of leptin in our group of pregnant women. This is in some contrast to previous data indicating that leptin has a role in regulation of b-oxidation during late pregnancy [25] and suggests that the lipolytic effect of leptin seen in the non-pregnant state is suppressed by peripheral leptin resistance during pregnancy. Despite the fact that leptin is considered to be a key hormone in the control of energy balance [5], we could not demonstrate any association between maternal leptin levels and resting energy expenditure. This is in agreement with data of Eriksson et al. [11] in their study of pregnant women and strengthens the hypothesis that central leptin resistance occurs during late pregnancy. We did not find any association between maternal leptin levels and fetal size. The reason behind this could be that maternal leptin levels increase until mid pregnancy due to placental-derived leptin and then remain at a steady level until parturition [26]. The association between leptin levels and insulin resistance can be explained by the close relation between maternal fat mass and insulin resistance [13]. Our data should be interpreted with some caution. The sample size is relatively small as a result of the laborious and timeconsuming methods used in the study. In the interpretation of the data the difficulties in estimating fetal size with ultrasound must be considered [19], even though there was a good correlation between estimated fetal weight and birth weight. Measurement of body composition in pregnant women is associated with several confounding factors. In order to measure maternal body composition as accurately as possible we used a three-compartment model which has been evaluated against underwater weighing combined with dual energy X-ray absorptiometry [27]. In this model multifrequency bio-impedance was used, which has been shown to be as reliable as deuterium dilution technique when measuring total body water in pregnant women [28]. Glucose tolerance was not assessed by OGTT. Instead, random plasma glucose was used as screening method for GDM [15]. Two of the women had elevated random levels of plasma glucose on one occasion and were subjected to an OGTT, which yielded normal results. Since OGTT was not performed in all cases there is possible uncertainty on whether GDM can be completely excluded in all the women, but the fact that levels of HbA1c were normal in all women also speaks in favor of a normal glucose tolerance. In conclusion, the strong inverse correlations shown between levels of maternal adiponectin and insulin resistance as well as rates of endogenous glucose production indicate that a decrease of levels of adiponectin may represent one mechanism behind increased fetal size in obese pregnant women. Our data confirm that levels of leptin during pregnancy are linked to maternal fat mass and insulin resistance, but indicate that there is no regulatory role of leptin with regard to lipolysis in late pregnancy. Acknowledgments The authors are grateful to Elisabeth So¨derberg for excellent technical assistance. We also thank Cecilia Ewald and Roger Olsson, Uppsala University Children’s Hospital, for skillful assistance. This work was supported by the Gillberg Foundation, the Solstickan Foundation, the Swedish Medical Society, the Wera Ekstrom Foundation and Swedish Society for Medical Research. References [1] Surkan PJ, Hsieh CC, Johansson AL, Dickman PW, Cnattingius S. Reasons for increasing trends in large for gestational age births. Obstetrics and Gynecology 2004;104:720–6.
29
[2] Jolly MC, Sebire NJ, Harris JP, Regan L, Robinson S. Risk factors for macrosomia and its clinical consequences: a study of 350,311 pregnancies. European Journal of Obstetrics Gynecology and Reproductive Biology 2003;111: 9–14. [3] Ahlsson F, Gustafsson J, Tuvemo T, Lundgren M. Females born large for gestational age have a doubled risk of giving birth to large for gestational age infants. Acta Paediatrica 2007;96:358–62. [4] Wei JN, Sung FC, Li CY, et al. Low birth weight and high birth weight infants are both at an increased risk to have type 2 diabetes among schoolchildren in taiwan. Diabetes Care 2003;26:343–8. [5] Henry BA, Clarke IJ. Adipose tissue hormones and the regulation of food intake. Journal of Neuroendocrinology 2008;20:842–9. [6] Grattan DR, Ladyman SR, Augustine RA. Hormonal induction of leptin resistance during pregnancy. Physiology and Behavior 2007;91:366–74. [7] Highman TJ, Friedman JE, Huston LP, Wong WW, Catalano PM. Longitudinal changes in maternal serum leptin concentrations, body composition, and resting metabolic rate in pregnancy. American Journal of Obstetrics and Gynecology 1998;178:1010–5. [8] Silha JV, Nyomba BL, Leslie WD, Murphy LJ. Ethnicity, insulin resistance, and inflammatory adipokines in women at high and low risk for vascular disease. Diabetes Care 2007;30:286–91. [9] Catalano PM, Hoegh M, Minium J, et al. Adiponectin in human pregnancy: implications for regulation of glucose and lipid metabolism. Diabetologia 2006;49:1677–85. [10] Fuglsang J, Skjaerbaek C, Frystyk J, Flyvbjerg A, Ovesen P. A longitudinal study of serum adiponectin during normal pregnancy. British Journal of Obstetrics and Gynaecology 2006;113:110–3. [11] Eriksson B, Lof M, Olausson H, Forsum E. Body fat, insulin resistance, energy expenditure and serum concentrations of leptin, adiponectin and resistin before, during and after pregnancy in healthy Swedish women. British Journal of Nutrition 2010;103:50–7. [12] Mazaki-Tovi S, Kanety H, Pariente C, et al. Maternal serum adiponectin levels during human pregnancy. Journal of Perinatology 2007;27:77–81. [13] Ahlsson F, Diderholm B, Jonsson B, et al. Insulin resistance, a link between maternal overweight and fetal macrosomia in nondiabetic pregnancies. Hormone Research in Paediatrics 2010;74:267–74. [14] Niklasson A, Ericson A, Fryer JG, Karlberg J, Lawrence C, Karlberg P. An update of the Swedish reference standards for weight, length and head circumference at birth for given gestational age (1977–1981). Acta Paediatrica Scandinavica 1991;80:756–62. [15] Ostlund I, Hanson U. Repeated random blood glucose measurements as universal screening test for gestational diabetes mellitus. Acta Obstetricia et Gynecologica Scandinavica 2004;83:46–51. [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] Sjodin AM, Forslund AH, Westerterp KR, Andersson AB, Forslund JM, Hambraeus LM. The influence of physical activity on BMR. Medicine and Science in Sports and Exercise 1996;28:85–91. [18] Weir JB. New methods for calculating metabolic rate with special reference to protein metabolism. Journal of Physiology 1949;109:1–9. [19] Persson PH, Weldner BM. Intra-uterine weight curves obtained by ultrasound. Acta Obstetricia et Gynecologica Scandinavica 1986;65: 169–73. [20] Bugianesi E, Pagotto U, Manini R, et al. Plasma adiponectin in nonalcoholic fatty liver is related to hepatic insulin resistance and hepatic fat content, not to liver disease severity. Journal of Clinical Endocrinology and Metabolism 2005;90:3498–504. [21] Bajaj M, Suraamornkul S, Piper P, et al. Decreased plasma adiponectin concentrations are closely related to hepatic fat content and hepatic insulin resistance in pioglitazone-treated type 2 diabetic patients. Journal of Clinical Endocrinology and Metabolism 2004;89:200–6. [22] Viollet B, Guigas B, Leclerc J, et al. AMP-activated protein kinase in the regulation of hepatic energy metabolism: from physiology to therapeutic perspectives. Acta Physiologica (Oxford) 2009;196:81–98. [23] Miller RA, Chu Q, Le Lay J, et al. Adiponectin suppresses gluconeogenic gene expression in mouse hepatocytes independent of LKB1-AMPK signaling. Journal of Clinical Investigation 2011;121:2518–28. [24] Butte NF, Hopkinson JM, Nicolson MA. Leptin in human reproduction: serum leptin levels in pregnant and lactating women. Journal of Clinical Endocrinology and Metabolism 1997;82:585–9. [25] Okereke NC, Huston-Presley L, Amini SB, Kalhan S, Catalano PM. Longitudinal changes in energy expenditure and body composition in obese women with normal and impaired glucose tolerance. American Journal of Physiology Endocrinology and Metabolism 2004;287:E472–9. [26] Zavalza-Gomez AB, Anaya-Prado R, Rincon-Sanchez AR, Mora-Martinez JM. Adipokines and insulin resistance during pregnancy. Diabetes Research and Clinical Practice 2008;80:8–15. [27] Forslund AH, Johansson AG, Sjodin A, Bryding G, Ljunghall S, Hambraeus L. Evaluation of modified multicompartment models to calculate body composition in healthy males. American Journal of Clinical Nutrition 1996;63: 856–62. [28] Butte NF, King JC. Energy requirements during pregnancy and lactation. Public Health Nutrition 2005;8:1010–27.