There have been many studies assessing the exact changes in inflammatory cytokines throughout normal gestation, but explorations of the relationships among these cytokines and metabolic indexes in Chinese pregnant women are still limited. Through our investigation, in addition to the positive associations of leptin, IL-8 and TNF-α with insulin resistance or secretion, NGF was found for the first time to be higher in GDM patients and positively associated with fasting and postprandial glucose metabolism, insulin resistance and pancreatic β cell function in the second trimester in Chinese.
NGF is an essential factor for the development of the nervous system in individuals and is necessary for healthy pregnancies[15, 16]. In addition, NGF is an inflammatory factor that can be induced by a variety of inflammatory diseases. In many articles, NGF expression was shown to be increased in diabetes mellitus and it has been proven to be an important protective factor in diabetic neuropathy and vasculopathy[18-21]. We suspected that the vascular and neuroprotective effects of NGF also played significant roles in the physiological processes of GDM. In our study, serum levels of NGF were significantly higher in GDM patients than in normal controls, suggesting the essential role of NGF in GDM. However, the levels of NGF in this study were different from previous reports[22-24]. The difference of sample types might be part of the reasons as most of those studies were conducted with the plasm while we used the serum. In addition, the method we used which could detect all of seven inflammatory factors in one panel is also different from the traditional ELISA method.
Currently, studies have found that NGF and its receptor Tyrosine kinase A (TrkA) are expressed in many tissues, including β cells of the pancreas[25-27], and that high glucose levels could dramatically enhance the secretion of NGF and subsequently activate β-cell TrkA receptors to acutely promote glucose-stimulated insulin secretion[28, 29]. These results were consistent with our finding that NGF was positively related to glucose levels and insulin secretion in the second trimester. Moreover, NGF could reduce insulin resistance both in vitro and in vivo by activating insulin receptor substrate 1 (IRS1)[30, 31]. Therefore, NGF levels are supposed to have a negative relationship with insulin resistance. However, in our study, we found NGF levels were positively associated with insulin resistance, which might be ascribed to the compensatory adjustment. These data suggest that NGF acts as a protective factor in glucose metabolism and insulin sensitivity and plays a significant role in glucose-induced insulin secretion. Although the correlations between NGF and these metabolism indexes were poor as all r < 0.3, given the significant role of NGF in insulin signal pathway, and the close relationship between NGF and GDM, we could not neglect the essential role of NGF in GDM.
Compared to the increasing NGF levels and decreasing IL-8 levels in GDM patients in the second trimester, in the third trimester, only leptin levels showed a significant difference between the GDM group and NGT group. Considering the close relationships between glucose levels and inflammatory responses, we speculated that the different results might be due to the diverse glucose metabolism states of the participants, as 16.7% of the GDM patients in the third trimester had been treated with insulin. Unfortunately, there is no possibility to study the rest patients who didn't receive insulin therapy on account of the lack of data on glucose and insulin levels in the third trimester. Additionally, participants in the second and third trimesters were different, although they shared the same origin and showed no significant differences in age and pregestational BMI, we could not exclude the possibility of population heterogeneity.
Correlations of IL-6, leptin and TNF-α with diabetes have been explored extensively, and most of these studies obtained results similar to ours that TNF-α, IL-6 and leptin were positively associated with insulin resistance or secretion[33, 34]. However, after adjustments, the relationship between TNF-α and insulin resistance disappeared. In addition, no relationships were found between MCP-1 or IL-8 and glucose metabolism or insulin resistance and secretion, which was contradictory to the consequences of some studies[33, 35]. In this study, serum levels of IL-8 were statistical significance between GDM patients and normal controls but showed no relationships with glucose metabolism indexes or insulin secretion/resistance. However, we could not rule out the possibility that there were associations between IL-8 and other metabolism indexes causing the differences, although we did not have other clinical metabolism indexes to perform the correlation analysis. In addition, the special physiological status of GDM and the combined action of diverse diets, lifestyles and gene-profiles might partly account for these differences.
Our limitations are as follows: First, the study was a case-control study that could not explore the changes in serum inflammatory cytokine levels throughout the course of pregnancy. Moreover, in the third trimester, due to missing data on glucose levels and insulin secretion, no investigation of the relationships of inflammatory cytokines and glucose metabolism was conducted. Last, our participants in the second and third trimesters were from two different populations, and we could not determine if the differences in the second and third trimesters resulted from GDM therapy or population heterogeneity. Further longitudinal cohort and large sample size studies with more comprehensive data are still needed to explore which inflammation-related factors are involved in the progression of GDM throughout pregnancy.