Marzena Wojcik M.D., Department of Structural Biology, Faculty of Biomedical Sciences and Postgraduate Education, Medical University of Lodz, 7/9 Zeligowskiego St., 90–752 Lodz, Poland, phone: +48 42 639 32 38; fax: +48 42 639 32 21, e-mail: marzena.wojcik@umed.lodz.pl

Introduction

Gestational Diabetes Mellitus (GDM) is an increasingly common complication of pregnancy characterised by carbohydrate intolerance resulting in maternal hyperglycaemia with onset or first recognition during pregnancy. GDM is associated with adverse outcomes for the mother and her offspring [1]. In the long term, GDM women are at elevated risk of developing type 2 diabetes (T2DM), cardiovascular diseases, and metabolic syndrome at a later age [2–4]. Additionally, children born with pregnancy complicated by GDM are at increased risk for the development of obesity and abnormal glucose metabolism during childhood, adolescence, and adulthood [5].

The screening procedures and diagnostic criteria for GDM vary between countries and even within countries, leading to different estimates of the prevalence of GDM, among others. In 1999, the World Health Organisation (WHO) introduced the diagnostic criteria for GDM based on a two-hour 75 g oral glucose tolerance test (OGTT) with fasting plasma glucose (FPG) concentration of ≥ 126 mg/dL (7 mmol/L) or twohour plasma glucose concentration of ≥ 140 mg/dL (7.8 mmol/L) as the cutoff for diagnosis of GDM [6]. The Polish Diabetes Association (PDA) recommended the use of modified WHO criteria with lowered the FPG cutoff to 100 mg/dL (5.6 mmol/L) [7]. In an attempt to unify the GDM criteria throughout the world, the International Association of the Diabetes in Pregnancy Study Group (IADPSG) proposed in 2010 new diagnostic criteria for GDM based on the results of the Hyperglycaemic Adverse Pregnancy Outcome (HAPO) study, which estimated the risk of maternal and foetal outcomes related to maternal hyperglycaemia [8]. According to the IADPSG guidelines, GDM is diagnosed based on a two-hour 75 g OGTT with the following thresholds: fasting plasma glucose (FPG) ≥ 92 mg/dL (5.1 mmol/L), or one-hour glucose concentration ≥ 180 mg/dL (10.0 mmol/L), or two-hour glucose concentration ≥ 153 mg/dL (8.5 mmol/L) [9]. The cutoff values in the new criteria were set to reflect an odds ratio of at least 1.75 (compared with the population mean) for various adverse foetal outcomes. The IADPSG criteria were endorsed by the WHO in 2013 [10] and PDA in 2014 [11].

Inflammation is a complex process regulated by a cascade of cytokines and growth factors that has been recognised as one of the factors associated with GDM. Among cytokines, interleukin (IL)-6 has drawn much attention not only as an immune-modulating molecule with important functions in the pathology of several inflammation-related diseases such as rheumatoid arthritis (RA), but also as the cytokine linked to type 1 and type 2 diabetes. In this regard, the involvement of IL-6 in the regulation of glucose homeostasis and metabolism by action on skeletal muscle cells, adipocytes, hepatocytes, and pancreatic β-cells has been demonstrated [12]. Interestingly, although IL-6 has been largely seen as a pro-inflammatory cytokine, recent findings suggest its anti-inflammatory role during obesity-associated inflammation and metabolic disorders [12]. Despite the fact that the relationship of IL-6 with GDM has been increasingly investigated in recent years, conflicting findings exist regarding its levels in patients with GDM vs. healthy pregnant controls. In this respect, a recently performed metaanalysis of IL-6 levels in patients with and without GDM in the 2nd/3rd trimesters has revealed that among seven analyzed studies, four have shown comparable concentrations of this cytokine in patients with and without GDM, and three have reported its elevated levels in patients with GDM compared with healthy controls [13].

These inconsistencies may be attributed to differences in diagnostic GDM criteria applied (i.e. WHO, National Diabetes Data Group, Canadian Diabetes Association, Australian Diabetes in Pregnancy) and/or gestational age at sampling and/or type of sample (i.e. plasma, serum, culture supernatant).

Since IL-6 is produced by various cells and tissues during diabetic pregnancy, including leukocytes, adipocyte tissue, placenta, skeletal muscle, fibroblast, and endothelial cells, and it is difficult to segregate their significance in IL-6 production during GDM, we have undertaken studies on the contribution of IL-6 production at its transcriptional level in leukocytes obtained from GDM women who were diagnosed by either the PDA 2011 or the PDA 2014 criteria. We also evaluated the impact of the PDA 2011 and 2014 guidelines on correlation analyses done between leukocyte IL6 expression and the clinical characteristics of patients. In this study we used leukocytes because these cells are well-known to be engaged in modulating inflammatory processes during diabetes and its complications and, on the other hand, they are a good alternative for less accessible metabolic tissues that are difficult to obtain from pregnant women.

Material and methods

Study design

A total of 145 Caucasian pregnant women were enrolled and studied at the Outpatient Diabetological Clinic “OmniMed” in Lodz (Poland) from June 2011 to November 2013. All pregnant women were routinely screened for GDM by a 75-g, two-hour OGTT at 24–28 weeks’ gestation or later if it was not possible during this period, according to the PDA 2011 and 2014 guidelines [7, 11]. Out of all pregnant women recruited, those with family history of diabetes in first-degree relatives, GDM in a previous pregnancy, diabetes diagnosed prior to pregnancy, systemic infection, or taking any drugs known to affect carbohydrate metabolism were excluded from the study.

All clinical investigations were conducted in accordance with the guidelines of the Declaration of Helsinki and were approved by the Bioethics Committee for Research on Humans at the Medical University in Lodz (No. RNN/154/09/KB). All participants provided written, informed consent.

Anthropometric and biochemical measurements

The information on maternal age and pre-pregnancy weight were collected from medical records. Maternal height and weight were measured by standard methods, and body mass index (BMI) was calculated by dividing the weight in kilograms by the height in metres squared.

Plasma glucose was measured by the glucose oxidase method, glycated haemoglobin (HbA1C) was assayed by a latex enhanced turbidimetric immunoassay using specific monoclonal antibodies, and the C reactive protein (CRP) concentration was determined by turbidimetric assay with the use of the cassette COBAS INTEGRA C-Reactive Protein (Latex) according to the manufacturer’s instructions (Roche Diagnostics GmbH, Mannheim, Germany). Lipid profiles, including total cholesterol (TC), HDL-cholesterol (HDL-C), LDL-cholesterol (LDL-C), and triglyceride (TG) were determined by enzymatic colorimetric methods (Roche Diagnostics GmbH, Mannheim, Germany). The biochemical assays were carried out with a COBAS INTEGRA analyser (Roche, SA). Plasma insulin level was quantified using Elecsys insulin assay (Roche Diagnostics GmbH, Mannheim, Germany). The homeostasis model assessment (HOMA) index was used to calculate insulin resistance (HOMA-IR) and beta-cell function (HOMA-B) as follows [14]:

HOMA-IR = [fasting insulin (μU/mL) × fasting glucose (mg/dL)]/405 and,

HOMA-B = [360 × fasting insulin (μU/mL)]/[fasting glucose (mg/dL) – 63]

To assess insulin sensitivity, the quantitative insulin sensitivity check index (QUICKI-IS) was calculated as follows: QUICKI = 1/[log(I0) + log(G0)], where I0 is the fasting plasma insulin level (μU/mL) and G0 is the fasting blood glucose level (mg/dL) [15].

Leukocyte isolation and RNA extraction

Leukocytes were isolated from the heparinised venous blood of the subjects (10 mL) as previously described [16, 17]. Total RNA was extracted from leukocytes using Tri Reagent according to the manufacturer’s instructions (Ambion, United States). RNA concentration was quantified using a LAMBDA 25 UV spectrophotometer (PerkinElmer, UK), and RNA quality and integrity was assessed by the A260/A280 ratio. Samples were kept at -80°C until assayed.

cDNA synthesis and Real-Time Polymerase Chain Reaction

Four μg of high-quality total RNA was converted to cDNA using RevertAid™ H Minus M-MuLV reverse transcriptase kit (Fermentas, Lithuania) according to the manufacturer’s recommendations. The cDNA was diluted ten-fold, and 2 μL of cDNA was used to perform RT-PCR using Maxima™ SYBR Green/ROX qPCR Master Mix (2×) (Thermo Scientific, United States) and specific primers for IL6 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a housekeeping gene. Amplification was carried out on a 7500 Real Time PCR System (Applied Biosystems, United States) with initial denaturation at 95°C for 10 minutes, followed by 40 cycles of 95°C for 60 seconds, 60°C for 60 seconds. All samples were run in duplicate. Amplification of specific transcripts was confirmed by melting curve profiles at the end of each PCR. The specificity of the PCR was further verified by subjecting the amplification products to agarose gel electrophoresis. RNA without reverse transcriptase during cDNA synthesis as well as PCR reaction using water instead of a template showed no amplification.

The threshold cycle (Ct) of each target product was determined, and ΔCt between target and endogenous GAPDH control was calculated as: ΔCt = Ct(IL6) - Ct(GAPDH). The relative expression of the IL6 gene relative to invariant control GAPDH was determined using the 2-ΔCt formula [18].

Statistical analysis

The data are expressed as median values with 25–75% interquartile range. The distribution of analysed biochemical and expression data was checked by the Shapiro-Wilk test. Differences between the groups studied were compared by the nonparametric Mann- Whitney U (Wilcoxon) test. The non-parametric Spearman’s rank test was used for analysis of correlation between variables. Additionally, Pearson’s parametric correlation was calculated for leukocyte IL6 expression and two-hour plasma OGTT transformed by the root of the fourth degree to obtain a normal distribution. Statistical analyses were carried out using a commercially available statistical software package (Statistica version 12.5, StatSoft, Poland), and statistical significance was set at P < 0.05.

Results

Subject’s characteristics

A total of 145 Caucasian pregnant women underwent screening for GDM, and the number of GDM cases identified by the PDA 2011 and 2014 criteria was 113 (i.e. the GDM/PDA 2011 group) and 104 (i.e. the GDM/ PDA2014 group), respectively. Out of 113 GDM/PDA 2011 patients, 21 women had the two-hour OGTT values between 140 and 153 mg/dL and therefore their GDM status was changed to NGT when the PDA 2014 guidelines were used. Among 32 subjects classified as the NGT/PDA 2011 group with the PDA 2011 criteria, 11 had plasma glucose values higher than cutoff with the PDA 2014 criteria at one-hour OGTT, and only one case had fasting plasma glucose ≥ 92 mg/dL. Hence, 12 NGT/PDA2011 women were identified as having GDM by the PDA 2014 criteria.

Clinical characteristics of the studied GDM and NGT groups are summarised in Table I. The GDM and NGT groups were comparable for age and parameters of adiposity such as pre- and pregnancy BMI and body weight gain (P > 0.05), irrespective of any criteria applied. Compared to respective NGT control groups, the GDM/PDA 2014 women exhibited significantly higher glucose concentrations (i.e. fasting, one-hour and two-hour OGTT), HbA1C and HOMA-IR values, and lower QUICKI indices whereas the GDM/PDA 2011 women had markedly higher glucose concentrations (i.e. fasting and 2 h OGTT) and HbA1C, and lower plasma HDL-C concentrations (P < 0.05). When compared the two NGT groups in respect to clinical parameters, glucose plasma concentrations were significantly lower at 1h and higher at 2h OGTT in the NGT/PDA 2014 group than the NGT/PDA 2011 (P < 0.05). Of note, there were no significant differences in clinical parameters between the GDM/PDA 2011 and GDM/PDA 2014 groups (P > 0.05).

Leukocyte IL6 gene expression

To examine IL6 gene expression in peripheral leukocytes from GDM patients diagnosed by the PDA 2011 and 2014 vs. respective NGT controls, quantitative realtime PCR studies were performed. As can be seen in Figure 1, a significant increase in leukocyte IL6 mRNA expression with a 2.24-fold up-regulation was detected in the GDM/PDA 2011 group compared with the NGT/PDA 2011 group (P<0.05). Interestingly, there was no difference in the leukocyte IL6 mRNA level between the GDM/PDA 2014 and NGT/PDA 2014 groups (P > 0.05). We compared further leukocyte IL6 gene expression between the two NGT groups and the two GDM groups. As shown in Figure 1, leukocyte IL6 mRNA expression was significantly increased (2.08-fold up-regulation) in the NGT/PDA 2014 group compared with the NGT/PDA 2011 group (P < 0.05), whereas it did not significantly differ between the two GDM groups (P > 0.05).

Correlation studies

To establish whether leukocyte IL6 expression is associated with clinical characteristics of the patients given in Table I, correlation analyses were made in the NGT and GDM groups, classified according to the PDA 2011 and 2014 criteria, as well as in the entire study group. With the PDA 2014 guidelines, leukocyte IL6 mRNA positively correlated with two-hour post-load glucose concentration in the NGT/PDA 2014 and GDM/PDA 2014 groups (Spearman’s r = 0.483, P = 0.001 and Spearman’s r = 0.280, P = 0.005, respectively) (Table II). Moreover, IL6 gene expression associated positively with plasma LDL-C concentration (Spearman’s r = 0.355, P = 0.050) and negatively with plasma HDL-C concentration (Spearman’s r = –0.442, P = 0.013) in the NGT/PDA 2014 group. With the PDA 2011 criteria, no correlation was evident between leukocyte IL6 mRNA and any of clinical parameters of patients from the NGT/PDA 2011 and GDM/PDA 2011 groups. It is noteworthy that significant positive correlation of leukocyte IL6 gene expression with two-hour post-load glucose concentration (Spearman’s r = 0.293, P < 0.001 and Pearson’s r = 0.302, P < 0.001) was observed in the entire study group (Table II and Fig. 2).

Discussion

Despite GDM being a common metabolic disease of pregnancy that shares many features of T2DM, including glucose intolerance and insulin resistance, its screening and diagnosis have been controversial over the last three decades, resulting in many inconsistencies in findings obtained from research, clinical, and population studies. To standardise the diagnosis of GDM, the IADPSG recommended in 2010 universal screening of all pregnant women with the 75 g OGTT and proposed new blood glucose thresholds for GDM diagnosis, which were adopted in 2014 by the PDA [9]. Until the end of 2013, the PDA recommended the modified WHO criteria for GDM diagnosis [7].

The primary goal of this study was to compare leukocyte IL6 gene expression in Polish GDM pregnant women diagnosed using the PDA 2011 criteria (the modified WHO recommendation) and PDA 2014 criteria (the same as those the IADPSG guidelines). A total of 145 pregnant women were included in the present study. By applying the PDA 2011 and 2014 criteria, 113 and 104 pregnant women, respectively, were recognised as having GDM. Out of 113 GDM women, 21 had a two-hour cut-point between 140 and 153 mg/dL, and therefore these cases were classified as NGT by the PDA 2014 criteria. Thus, it is reasonable to assume that since the PDA, like the IADPSG, has increased the two-hour cut-point to 153 mg/dL, many cases of GDM might be missed. Interestingly, we also observed that out of 32 women classified as NGT with the PDA 2011 guidelines, 12 were picked up by the PDA 2014 criteria as GDM.

In the present study, we found that shifting from the PDA 2011 criteria to the PDA 2014 guidelines was accompanied by changes in metabolic phenotypes of GDM patients compared to their respective NGT controls. According to this, the GDM/PDA 2011 patients had a higher fasting and two-hour post-load glucose and HbA1c levels and lower HDL-C concentration whereas the GDM/PDA 2014 patients were characterised by a higher fasting, one-hour and two-hour post-load glucose, HbA1c, and insulin resistance assessed by the HOMA-IR method and lower insulin sensitivity calculated from the QUICKI equation. The high HOMA-IR values and low QUICKI-IS values observed in the GDM group are consistent with the findings obtained by Endo et al. [19]. Interestingly, no difference was found in plasma CRP concentrations between the two GDM groups and their respective NGT control groups. CRP is a non-specific acute-phase reactant primarily produced by the liver in response to inflammatory stimuli and its positive association with GDM was previously reported [20, 21]. On the other hand, the effect of maternal obesity on circulating CRP levels was also demonstrated, suggesting that obesity can be a major determinant of CRP concentration in pregnancy [22]. Therefore, it is more likely that the lack of differentiation in plasma CRP levels between the GDM and NGT groups found in our study, regardless of the diagnostic criteria applied, could reflect no difference in pre-pregnancy BMI between them.

To gain more information about the reasons for the above-described metabolic differences by using the PDA 2011 and 2014 criteria, a comparison of clinical parameters between the two GDM groups and the two NGT groups was performed. Although the results obtained showed a lack of differences in clinical parameters between the GDM/PDA 2011 and GDM/ PDA 2014 groups, lower one-hour and higher two-hour glucose values were observed in the NGT/PDA 2014 group vs. the NGT/PDA 2011 group. These findings are not entirely surprising given that 21 GDM cases with a two-hour cut-point of 140–153 mg/dL shifted from the GDM/PDA 2011 group to the NGT/PDA 2014 group and 11 NGT cases with one-hour cut-point ≥ 180 mg/dL shifted from the NGT/PDA 2011 group to the GDM/ PDA 2014 group.

It is now well accepted that chronic, low-grade, systemic inflammation, shown by alterations in the concentrations of circulating cytokines, is associated with diabetes [23]. Among a panel of diabetes-related cytokines, IL-6 has drawn much attention as a pleiotropic cytokine engaged in metabolic events during T2DM [24], obesity [25, 26], and impaired glucose tolerance [27], but its relevance to GDM has been unclear in previous studies [13]. Indeed, Kuzmicki et al. [28, 29] observed elevated IL-6 concentrations in GDM patients compared with normal pregnant women, whereas Georgiou et al. [30] did not find any difference. Furthermore, a lack of difference in IL-6 release from placenta, adipose tissue, and skeletal muscle between GDM and NGT women was also observed [31]. By contrast, increased IL6 gene expression was detected in subcutaneous fat, but not in placenta of GDM women vs. control subjects [32]. The conflicting results obtained can in part be attributed to differences in GDM diagnostic criteria. Therefore, we renew interest in the connection of IL-6 with GDM in the present study by investigating its expression in leukocytes of diabetic women diagnosed by either the PDA 2011 or 2014 guidelines. Our results revealed that compared to respective NGT control groups, leukocyte IL6 gene expression was significantly increased in the GDM/PDA 2011 group, but it remained unchanged in the GDM/PDA 2014 group. When leukocyte IL6 expression was compared between the two GDM groups and the two NGT groups, there was no change in its level between the GDM/PDA 2011 and GDM/PDA 2014 groups, while it was markedly increased in the NGT/PDA 2014 group vs. the NGT/PDA 2011 group. These observations suggest that higher two-hour OGTT glucose values can be related to increased leukocyte IL6 expression in pregnancy because there were more GDM women with a two-hour cut-point of 140–153 mg/dL, who shifted from the GDM/PDA 2011 group to the NGT/PDA 2014 group than NGT women with one-hour cut-point ≥ 180 mg/dL, who shifted from the NGT/PDA 2011 group to the GDM/PDA 2014 group.

In this study, although statistical analyses failed to identify any correlation of leukocyte IL6 mRNAs with the clinical parameters of the NGT/PDA 2011 and GDM/PDA 2011 patients, significant positive linear relationships were observed between leukocyte IL6 expression and two-hour post-load glucose concentration in the NGT/PDA 2014 and GDM/PDA 2014 groups. Moreover, the IL6 mRNA level positively correlated with plasma LDL-C concentration and negatively with plasma HDL-C concentration in the NGT/PDA 2014 group. Nevertheless, there was a significant positive correlation of leukocyte IL6 gene expression with two-hour postload glucose concentration in the entire study group, suggesting that a change of leukocyte IL6 expression can be related with a role of IL-6 in regulating blood glucose levels during pregnancy. In line with this, increased blood glucose after IL-6 infusion in healthy individuals [33], as well as a direct stimulatory effect of IL-6 on hepatic glucose release from glycogen pools by inhibiting glycogen synthase [34], was shown. On the other hand, we cannot exclude the possibility that high glucose conditions may induce the production of greater amounts of IL-6. In this regard, high glucose concentration was demonstrated in vitro to induce IL-6 expression at mRNA and protein levels in monocytes through activation of protein kinase C (PKC)-α/β, p38 mitogen-activated protein kinase (p38 MAPK), and nuclear factor-κB (NF-κB) [35]. In addition to glucose action, sugar-derived substances called advanced glycation end products (AGEs) were reported to stimulate IL-6 production in human monocytes, implying that hyperglycaemia-induced oxidative stress could participate in these events [36]. Taken together, considerable future research will be necessary to establish the cause-effect relationship between leukocyte IL-6 expression and glucose metabolism in the context of pregnancy.

Evidence supports IL-6 as an important inducer of CRP expression in hepatocytes by activating the transcription factors STAT3 and CCAAT/enhancer binding protein β (C/EBP β) [37–39]. Despite the fact that several studies have documented the existence of relationship of IL-6 with CRP in GDM women [20, 21], no correlation was seen between leukocyte IL6 gene expression and plasma CRP concentration among GDM patients in the present study. This observation is not surprising because many molecules, such as IL-1β, IL-4, IL-11, TNF-α, and TGF-β, have been shown to be involved in regulating CRP expression [40–44].

In summary, our comprehensive comparison of clinical characteristics of pregnant women and their leukocyte IL6 expression, along with its relationships with metabolic parameters of patients, by applying the PDA 2011 and 2014 criteria, points to differences in metabolic, genetic, and statistical data between the GDM groups, identified based on the PDA 2011 and 2014 criteria, and their respective NGT control groups. These differences seem to be highly related to changes in gestational glucose tolerance status resulting from using the PDA 2014 criteria. Importantly, our findings are in line with the hypothesis supporting the relationship of IL-6 with glucose metabolism during pregnancy, but further studies are needed in order to better understand the causal pathway that links IL-6 and plasma glucose levels in pregnant women.