Insulin-like growth factor-1 receptor expression in the placentae of diabetic and normal pregnancies☆

1. Introduction 

Gestational diabetes mellitus (GDM) affects approximately 7% of all pregnancies, resulting in more than 200,000 cases per year [1]. GDM is associated with significant fetal complications and morbidity, which include macrosomia, birth trauma, neonatal hypoglycemia, respiratory distress syndrome and septal hypertrophic cardiomyopathy (sHCM) [1], [2], [3], [4], [5], [6], [7], [8], [9]. Infants of diabetic mothers (IDM) with hypertrophic cardiomyopathy may develop cardiac arrhythmias and heart failure shortly after birth [10], [11]. Despite tight control of maternal blood glucose levels in gestational diabetics, asymmetric septal hypertrophic cardiomyopathy is still encountered in up to 30% of the infants born [12]. As such there could be other factors that affect this outcome, with various growth factors known to exert endocrine, autocrine and paracrine effects on the fetal heart. Growth factors such as insulin-like growth factor 1 (IGF-1) are recognized as a major regulator of fetal growth. Elevated levels of maternal IGF-1 have been shown to be associated with infants who are large for gestational age [13]. In contrast, low maternal serum levels were detected when there was fetal growth restriction [14].

IGF-1 is synthesized primarily by the liver and largely bound to one of the six IGF-binding proteins [15], [16]. We have previously reported that maternal serum IGF-1 levels were significantly higher in GDM pregnancies than healthy matched controls. Of note, the six diabetic mothers whose infants were diagnosed to have asymmetric septal hypertrophic cardiomyopathy had even higher mean serum IGF-1 levels [17]. Cooper et al. suggested that this cardiac anomaly was associated with poor maternal metabolic control [10]. Interaction between neuroendocrine factors and mechanical forces is thought to lead to the cardiac hypertrophy [18] and in vitro studies have demonstrated that IGF-1 is a potential stimulus for myocyte hypertrophy [19].

Earlier studies have shown that maternal IGF-1 does not cross the placenta to the fetus [20], [21], [22], IGF-1 exerts its function by acting as a ligand to the receptor, IGF-1R [15]. In humans, IGF-1R is primarily expressed on the microvillous membranes on the maternal side of the placenta [23], suggesting that maternal IGF-1 may influence placental function. The IGF-1R is a member of the large family of protein-tyrosine kinases and consists of two [alpha] and two [beta] subunits. The [alpha] subunits are entirely extracellular while the [beta] subunits display a highly hydrophobic transmembrane domain [24], [25], [26], [27]. There is, however, a paucity of data describing the distribution of IGF-1R and the expression of this receptor in the human placenta, particularly that of diabetic pregnancies. In this study, we aimed to characterize the distribution and expression of IGF-1R in the placenta by immunohistochemistry and compare results from normal and diabetic pregnancies with and without neonatal septal hypertrophy. Furthermore, our previous study showing significantly higher maternal IGF-1 levels among IDMs with septal hypertrophy [17] prompted us to investigate whether there is any associated alteration in the expression of placental IGF-1R for this group of infants.

2. Materials and methods 

3. Results 

3.1. Clinical characteristics 

Of the 18 samples obtained, 12 were from women who had diabetes in pregnancy. Ten of these had gestational diabetes mellitus (GDM) and two had impaired glucose tolerance (IGT) (Table 1). All the 10 gestational diabetics were on insulin therapy with well-controlled blood sugar profiles and HBA1c levels. A total of six (five GDM and one IGT) women with diabetes in pregnancy had infants who developed septal hypertrophic cardiomyopathy (Table 1). The septal thickness ranged from 6 to 11.5 mm. Two of these infants were admitted into the neonatal intensive care unit because of mild breathing difficulties, which resolved promptly with a brief period of respiratory support. Their respiratory complication was not directly related to heart failure or arrhythmias. None of the infants of diabetic mothers (IDMs) had episodes of hypoglycaemia (defined as plasma glucose level of <2.6 mmol/L) within the first 6 h of life. Infants of diabetic mothers were born at a significantly lower gestation than normal infants. There was also a trend for IDMs to have a higher birth-weight, particularly in the group who had septal hypertrophy (Table 1). The average placental weight was highest in the group with septal hypertrophy but the difference was not statistically significant.

Table 1. Characteristics of the newborn infants and placentae of diabetic and normal pregnancies

	Pregnancy	Diabetica (n=12) GDM (n=10), IGT (n=2)		Normal (n=6)
	Neonatal sHCMb	Positive (n=6)	Negative (n=6)	Negative (n=6)	p-value
	Gestation (weeks)	37.7 (0.8)	37.8 (0.8)	39.5 (0.8)	0.002
	Birth weight (g)	3437 (554)	3038 (485)	3162 (362)	0.354
	Placental weight (g)	755 (83)	635 (156)	662 (86)	0.192

Values are expressed as mean (S.D.).

a From a pool of 50 women with well controlled diabetes (mean HbA1c of 6.3 S.D. 1.1%). b Two of 6 (33%) in the group of positive neonatal sHCM infants were admitted into the neonatal intensive care unit.

3.2. Placental immunohistochemistry 

IGF-IR staining was present in the deciduas, cytotrophoblasts and syncytiotrophoblasts of placental specimens from both diabetic and non-diabetic pregnancies (“2+” or “3+”). The staining was localised mainly in the cell membranes, but occasionally some cytoplasmic staining were seen, which may be the result of internalization of the receptor after activation [30]. Staining was, however, consistently absent (“0”) in the villous endothelium (Fig. 1A–C). In comparing the degree of expression of IGF-1R by staining intensity, all six placental specimens of diabetic pregnancies with neonatal septal hypertrophic cardiomyopathy showed “2+” staining in the deciduas, cytotrophoblasts and syncytiotrophoblasts (Fig. 1D and G). This was in contrast to strong staining (“3+”) of similar cells in the placentae from diabetic and normal pregnancies with infants who had no septal hypertrophy (Fig. 1E and H; F and I).

View Large Image Download to PowerPoint

Figure 1. Immunohistochemical staining for IGF-1R in the placentae from diabetic and normal pregnancies. Magnification bar on each of the figure=50 μm. From top to bottom: The left panel (A, D and G) are from placentae of diabetic pregnancies with neonatal sHCM. The middle panel (B, E and H) are from placentae of diabetic pregnancies without neonatal sHCM. The right panel (C, F and I) are from placentae of normal pregnancies without neonatal sHCM. From left to right: The top panel (A, B and C) illustrate staining of the villous endothelium (arrow head). The middle panel (D, E and F) illustrate staining of the decidua cells. The bottom panel (G, H and I) illustrate staining of the cytotrophoblasts (arrow) and syncytiotrophoblasts (arrow head). The presence of IGF-1R expression is represented by staining in brown. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4. Discussion 

In the early 1940s, Miller and Wilson described cardiac hypertrophy and hyperplasia of pancreatic islands of Langerhans in macrosomic infants of diabetic mothers [31], [32]. Later in the mid-1970s, with the emergence of echocardiography, the cardiac anomaly was re-defined as “transient hypertrophic subaortic stenosis” and more specifically, asymmetric septal hypertrophic cardiomyopathy (sHCM) [29]. Asymmetrical septal hypertrophy has been reported in 30–40% of infants of diabetic mothers [10], [11]. Although often transient, with most cases resolved within the first 6 months of life [29], some may be complicated by early heart failure and arrythmias [33]. Cooper et al. also reported a positive relationship between maternal hyperglycaemia and the development of sHCM [10]. Generally thought to be a “benchmark” indicator of maternal glycaemic control, the study by Sheehan et al. [12], however, did not show a significant relationship between neonatal hypertrophic cardiomyopathy and maternal HbA1c levels in 20 infants of diabetic mothers. Our previous study of well-controlled diabetic pregnancies [17] revealed that 12% of the infants developed sHCM. Rizzo et al. [34] and Vela-Huerta et al. [11] suggested that strict maternal diabetes control did not exclude accelerated fetal cardiac growth. These suggest that other factors such as maternal IGF-1 may contribute to the development of sHCM.

IGF-1 binds to IGF-1R to activate a series of responses in the cell that lead to DNA synthesis, cell proliferation and differentiation [35]. Most of the information on the localization and characterization of IGF-1R in the placenta has been derived from studies on animals of early gestations. In humans, placental IGF-1R has been mostly studied in the area of pre-eclampsia and fetal growth restriction. Holmes et al. [30] studied IGR-1R in the placentae of pregnancies with appropriately grown or growth restricted fetuses using an immunohistochemistry approach. They showed the presence of IGF-1R in deciduas, trophoblasts, villous stroma and endothelium with considerable cytoplasmic cell staining. Our study revealed that IGF-1R staining is predominantly membranous and localized in the deciduas, cytotrophoblasts and syncytiotrophoblasts, but not in the villous endothelium and stroma. The differences may be attributed to the mechanism by which antigens were retrieved during tissue processing and the anti-IGF-1 alpha subunit (extracellular) receptor monoclonal antibody that produced membranous and less non-specific staining.

As IGF-1 does not cross the placenta, the effects of this factor on fetal growth are likely to be indirect and modified by maternal serum concentrations of this peptide [21], [22], [36]. Karl [37] showed that IGF-1 acts on its receptor (IGF-1R) in the placenta to stimulate the uptake of amino acid, an effect which was earlier shown to be also dose-dependent [38]. Lauszus et al. [13] further showed that macrosomic IDMs were associated with high levels of maternal IGF-1. The anabolic effects of IGF-1 may be reflected in our six infants who developed septal hypertrophy and had a higher mean birth-weight, despite the absence of hypoglycaemic episodes to suggest poor maternal glycaemic control. These infants were also associated with a higher maternal mean serum IGF-1 than infants without septal hypertrophy [17], suggesting that other than maternal glycaemic control and fetal hyperinsulinism, the interaction between maternal IGF-1 and IGF-1R may be another factor that could affect neonatal outcome.

In this study, we examined all six of 50 (12%) placentae of diabetic pregnancies that were associated with neonatal sHCM. A lower incidence of neonatal sHCM compared to other studies may be a result of different echocardiographic classification used, and a more heterogeneous population of diabetic pregnancies, which included women with impaired glucose tolerance on diet control. A review of the 12 placental samples (stained with haematoxylin and eosin) from diabetic pregnancies that were obtained for routine histological examination did not show any characteristic changes of vacuole formation and acute atherosis. There was a trend towards a higher mean placental weight in sHCM babies although this was not statistically significant. However, placental staining for IGF-1R in the group of infants with sHCM was all weaker than the groups without sHCM, suggesting reduced expression of this receptor. We speculate that IGF-1R may be downregulated in response to elevated maternal IGF-1 levels as a protective mechanism to ensure fetal well-being and as a regulatory measure to decrease the extent and severity of fetal complications such as sHCM. Furthermore, changes in expression of IGF-1R may precede overt histological features in the placenta that are characteristic of macrosomic infants with organ complications. That IGF-1 could downregulate IGF-1R has been reviewed by Nissley et al. [39]. On the other hand, the study by Holmes et al. [30] did not show any increased expression of placental IGF-1R as a possible up-regulatory response to the very low levels of maternal IGF-1 associated with fetal growth restriction (FGR). Clearly, the regulation and interaction of maternal IGF-1 and placental IGF-1R are complex and require further elucidation.

In conclusion, this pilot observational study has revealed some important insights into the localization and expression of IGF-1R pertaining to diabetic pregnancies, elevated maternal IGF-1 levels and neonatal septal hypertrophic cardiomyopathy. Further studies involving larger numbers are required to confirm the postulated downregulatory expression of IGF-1R, so that modulation of maternal IGF-1 levels may be a target for therapeutic intervention in the control of fetal complications associated with diabetes in pregnancy.