Vascular endothelial growth factor (VEGF) and VEGF-receptor expression in placenta of hyperglycemic pregnant women☆

Article Outline

• Abstract
• 1. Introduction
• 2. Materials and methods
  • 2.1. Subjects
  • 2.2. Tissue samples
  • 2.3. Immunohistochemistry
  • 2.4. Congo Red
  • 2.5. Picrosirius red
  • 2.6. Western blotting
• 3. Results
  • 3.1. Morphology of chorionic villi in the placenta at term
  • 3.2. Immunolocalization of VEGF
  • 3.3. Immunolocalization of VEGFR-1 (Flt-1)
  • 3.4. Immunolocalization of VEGFR-2 (KDR)
  • 3.5. Western blotting
• 4. Discussion
  • 4.1. Placental morphology in glycemic disturbances
  • 4.2. VEGF and receptors immunolocalization
• Competing interests
• Acknowledgments
• References
• Copyright

Abstract 

Hyperglycemia occurs in a variety of conditions such as overt diabetes, gestational diabetes and mild hyperglycemia, all of which are generally defined based on the oral glucose tolerance test and glucose profiles. Whereas diabetes has received considerable attention in recent decades, few studies have examined the mechanisms of mild hyperglycemia and its associated disturbances. Mild gestational hyperglycemia is associated with macrosomia and a high risk of perinatal mortality. Morphologically, the placenta of these women is characterized by an increase in the number of terminal villi and capillaries, presumably as part of a compensatory mechanism to maintain homeostasis at the maternal-fetal interface. In this study, we analised the expression of VEGF and its receptors VEGFR-1 (Flt-1) and VEGFR-2 (KDR) in placentas from mildly hyperglycemic women. This expression was compared with that of normoglycemic women and women with gestational and overt diabetes. Immunohistochemistry revealed strong staining for VEGF and VEGFR-2 in vascular and trophoblastic cells of mildly hyperglycemic women, whereas the staining for VEGFR-1 was discrete and limited to the trophoblast. The pattern of VEGF and VEGF-receptor reactivity in placentas from women with overt diabetes was similar to that of normoglycemic women. In women with gestational diabetes, strong staining for VEGFR-1 was observed in vascular and trophoblastic cells whereas VEGF and VEGFR-2 were detected only in the trophoblast. The expression of these proteins was confirmed by western blotting, which revealed the presence of an additional band of 75 kDa. In the decidual compartment, only extravillous trophoblast reacted with all antibodies. Morphological analysis revealed collagen deposition around large arteries in all groups with altered glycemia. These findings indicate a placental response to altered glycemia that could have important consequences for the fetus. The change in the placental VEGF/VEGFR expression ratio in mild hyperglycemia may favor angiogenesis in placental tissue and could explain the hypercapillarization of villi seen in this gestational disturbance.

Keywords: Angiogenesis, Gestational diabetes, Hyperglycemia, Overt diabetes, Trophoblast, VEGF

Back to Article Outline

1. Introduction 

Vasculogenesis, the de novo formation of new blood vessels, and angiogenesis, the formation of blood vessels by sprouting from pre-existing ones, are important processes in many physiological and pathological conditions, including embryonic development [1], [2]. During vasculogenesis, formation of the earliest primitive capillaries is achieved by in situ differentiation of hemangiogenic stem cells that are derived from the pluripotent mesenchyme. The resulting angioblastic cells give rise to endothelial precursor cells, and only thereafter, during angiogenesis, are new blood vessels derived from already existing vessels [2], [3], [4].

Angiogenesis is closely related to gestational success. Maternal and fetal tissues interact during the entire gestation in a relationship that is totally dependent on the establishment of appropriate vascularization in the placental decidua and chorion. Decidual angiogenesis provides the uterine vascular elongation required for adaptation to fetal metabolic and gas exchange, whereas the chorion rapidly grows to form the chorionic villi that are characterized by constant blood vessel growth and differentiation. This process involves proliferation, migration and maturation of maternal and fetal endothelial cells and is continuous until the last trimester of the human pregnancy [5], [6], [7]. In the chorionic villi, the capillary bed is located in terminal branches of the villous tree, the terminal villi. Expression of vascular endothelial growth factor (VEGF), its receptors VEGFR-1 (Flt-1) and VEGFR-2 (KDR) and placenta growth factor (PlGF) have important roles in promoting normal placental angiogenesis.

Vascular disorders can change placental function and compromise fetal development [8], [9]. This condition has long been studied in diabetes. An increase in circulating glucose concentrations in diabetic mothers alters placental carbohydrate, protein and lipid metabolism, and a decrease in circulating insulin levels resulting from reduced pancreatic function can adversely affect fetal metabolism [10], [11]. An increase in glucose promotes atherosclerosis that may impair uteroplacental blood flow and lead to hypoxia/ischemia, in addition to creating excessive syncytial knots and accelerating perivillous fibrin deposition that can cause placental hypercapillarization [12]. A correlation has been observed between hypercapillarization and the paracrine stimulation of endothelial cells by VEGF via VEGFR-2 in response to low levels of oxygen [13], whereas villous degeneration without angiogenesis is generally associated with severe hypoxia/ischemia, despite stimulation by VEGF [14]. Maternal hyperglycemia has also been implicated in elevated birth weight [15], [16] and a causal factor in a variety of severe problems involving the fetus and neonate [10].

Gestational dysfunctions related to hyperglycemia are not restricted to women with overt diabetes diagnosed before gestation, but are also observed in gestational diabetes (occurring during gestation) and mild hyperglycemia. This classification is based on changes in the oral glucose tolerance test (OGTT) and glucose profiles in which women with gestational diabetes and overt diabetes have an abnormal OGTT and glucose profile whereas those with mild hyperglycemia show only an altered glucose profile [18], [19], [20].

In contrast to diabetes that has received considerable attention in recent decades, few studies have investigated the mechanisms involved in mild hyperglycemia and its associated disturbances [9], [17], despite the fact that this condition carries with it a high risk of macrosomia and perinatal mortality. Morphological analysis of placentas from women with mild hyperglycemia has revealed an increase in the number of villi and vessels, a finding that has been interpreted as a compensatory mechanism for maintaining homeostasis at the maternal-fetal interface [9].

In view of the physiological importance of VEGF during pregnancy, in this study we investigated the expression and localization of VEGF, VEGFR-1 and VEGFR-2 in placentas from women with mild hyperglycemia and compared the results with those for normoglycemic women and for women with gestational and overt diabetics.

Back to Article Outline

2. Materials and methods 

2.1. Subjects 

The mothers provided informed consent and the collection of placental tissue were approved by the Research Ethics Committee of the University of São Paulo and by the Department of Obstetrics and Gynecology at the University Hospital of Botucatu (UNESP), São Paulo State, Brazil.

Pregnancy was determined based on menstrual history and physical examination, and was confirmed by early ultrasound evaluation. Only term placentas from gestational age > 37 weeks from no smoking pregnant women aged 30–40 years, were used (Table 1). The pregnant women were classified into four groups based on their response to the OGTT (oral administration of 100 g of glucose and blood analysis 3 h later) according to ADA criteria [18], [19], [20] and their glucose profile, defined using Gillmer’s threshold values [21]. The four groups were: (a) normoglycemic (control) with a normal 100 g-OGTT and GP, (b) mildly hyperglycemic, with a normal 100 g-OGTT and altered GP, (c) gestational diabetes, with an abnormal 100 g-OGTT and normal GP, and (d) overt diabetes type 1 and 2, with an abnormal pre-pregnancy 100 g-OGGT. Normal values for the glucose profile were fixed at less than 100 mg/dL. Glucose levels were determined by the glucose oxidase method (Glucose-analyzer II, Beckman, Fullerton, CA, USA).

Table 1. Table with the patients’ datas.

Hyperglycemic pregnant women were treated as gestational diabetes, with diabetic diet, individualized and appropriate to their needs, considering even the pregnancy status. In the case of over diabetes, type 1 or type 2, continued in their routine treatment, i.e, with individualizes diet and insulin human (NPH, preferably, linked to adjust if necessary). No oral antihiperglicemiantes or insulin analogues were used.

2.2. Tissue samples 

Tissues samples were collected at cesarean delivery, based on obstetric indications. A tissue biopsy was dissected from the central part of the placental bed and prepared for histochemistry, immunohistochemistry and western blotting, as described below. The tissue samples were dissected into small fragments containing either only the villous region or decidua, although the latter sometimes contained small portions of the villous region. The fragments were then either immediately embedded in Trizol, frozen in liquid nitrogen and stored at − 80 °C for western blotting or fixed in 4% paraformaldehyde in 0.1 M phosphate buffered-saline (PBS), pH 7.4 for subsequent histochemical and immunohistochemical analyses.

2.3. Immunohistochemistry 

After fixation for 4–24 h, the placental samples were washed in PBS and processed for embedding in Histosec (Merck, Darmstadt, Germany). Deparaffinized sections 5 μm thick were stained with hematoxylin-eosin to identify the villous and decidual placental regions, with semi-serial sections then being obtained for immunohistochemistry. Samples from at least five different paraffin blocks of each placenta were analyzed.

The sections were initially incubated in 3% hydrogen peroxide in distilled water for 30 min to block endogenous peroxidase activity. The samples were then incubated sequentially with (a) blocking buffer (SuperBlock^®, Rockford, Illinois, USA) containing 1.5% donkey serum (Jackson Lab, Boston, MA, USA) to block non-specific antigen binding-sites (1 h, room temperature), (b) primary polyclonal goat anti-VEGF antibody (diluted 1:15; Santa Cruz Biotechnology, Santa Cruz, CA, USA), polyclonal rabbit anti-Flt-1 antibody (1:50; Santa Cruz Biotechnology) or polyclonal rabbit anti-Flk-1 antibody (1:50; Santa Cruz Biotechonology) diluted in blocking buffer (overnight, 4 °C), and (c) secondary rabbit anti-goat antibody or goat anti-rabbit IgG (both from KPL, Gaithersburg, Maryland, USA) in TBS (Tris-buffered saline) for 1 h at 37 °C. After washing in TBS, peroxidase activity of the secondary antibodies was detected by incubating the sections in TBS, pH 7.6, containing 3,3-diaminobenzedine tetrahydrochloride (DAB, 0.66 mg/mL; Sigma Chemical Co., St Louis, MO, USA) and 0.003% hydrogen peroxide (v/v). The slides were counter-stained with Mayer’s hematoxylin and mounted with Entellan^® (Merck, Darmstadt, Germany). Negative controls were incubated with either buffer or rabbit non-immune serum instead of the primary antibody.

The immunohistochemical reaction was also carried out to identify the trophoblast cell using the monoclonal antibody IgG mouse anti-human keratin in a 1:50 dilution in TBS-BSA. Interpretation of immunostaining was analyzed according to a subjective evaluation of intensity of reaction: a) negative or weak intensity (0 and -); b) moderate intensity (1 and ±) and c) strong intensity (2 and +).

2.4. Congo Red 

Amyloid is an association of proteins not found in the body under normal conditions but is common in immune disturbances. To assess whether glycemic disorders can lead to the deposition of amyloid in placental tissue, 5-μm-thick sections were stained with Congo Red dye (Sigma–Aldrich) which stains amyloid pink or red [22]. Deparaffinized sections were stained with Harris’ hematoxylin for 1 min, washed and counter-stained with an alkaline solution of 0.2% Congo Red in 80% ethanol for 1 h prior to mounting in Entellan^®. Reactions were analyzed using a scale of positive (+) to intense reaction, negative (−) for no markup and more or lessa (±) to fews reactive cells.

2.5. Picrosirius red 

Hyperglycemia can lead to the exacerbated production of collagen by vascular endothelium. To assess collagen production by placental tissue, 5-μm-thick sections were stained with Picrosirius Red, which allows the visualization of collagen fibers by light microscopy (red staining) or polarized light microscopy (red, green or yellow, depending on fiber thickness and organization) [23]. After deparaffinization, the sections were stained with 0.1% Picrosirius Red in a saturated solution of picric acid for 20 min. The sections were then washed, counter-stained with Mayer’s hematoxylin for 2 min and mounted in Entellan^®. The reading of the reactions was made using a scale of positive (+) to intense reaction, negative (−) for no markup and more or lessa (±) to fews reactive cells.

2.6. Western blotting 

Total protein extracts of placental tissue were obtained by using lysing solution containing RIPA buffer and a protease inhibitor mix (1 mM formil metil sulfonil fluoreto [PMSF], 3% benzamidine, 3% aprotinin and 3% leupeptin). Samples were run on 12% polyacrylamide gels containing SDS, where 30 μg of protein was applied to each lane. After electrophoresis, the proteins were transferred to nitrocellulose membranes (Hybond-ECL, Amersham, UK) and non-specific binding sites were then blocked by incubating the membranes with TBS containing 0.05% Tween 20 (TBST) and 5% dry milk for 4 h, at room temperature. The membranes were subsequently rinsed, incubated overnight at 4 °C with primary antibody (goat polyclonal anti-VEGF antibody diluted 1:500, rabbit polyclonal anti-Flt-1 antibody diluted 1:100 or rabbit polyclonal anti-Flk-1 antibody diluted 1:250, all from Santa Cruz Biotechnology; a monoclonal anti-β-actin antibody from Sigma diluted 1:10,000 was used as an internal control to detect β-actin), washed in TBS and then incubated with secondary antibody (rabbit anti-goat IgG diluted 1:5000 or goat anti-rabbit IgG diluted 1:750, both from KPL, or anti-mouse IgG peroxidase conjugate from Sigma, diluted 1:500). After a final wash with TBS, the bound enzyme was detected by chemiluminescence, according to the manufacturer’s instructions (Amersham). Qualitative analysis was performed using NIH image analysis software (Image J 1.42q, National Institutes of Health (NIH), Bethesda, Maryland, USA) and expressed in relation to β-actin.

Back to Article Outline

3. Results 

3.1. Morphology of chorionic villi in the placenta at term 

Placentas of normoglycemic pregnant women at term presented a large caliber chorionic villi emerged from the chorionic plate and gradually branched into smaller intermediate and terminal villi. Transverse and oblique sections showed to be surrounded by maternal blood and a fibrinoid substance (composed of fibrin and other glycoproteins that stained intensely with eosin). Some of the large villi were attached to the deciduous or basal plate and served to anchor the villi (Fig. 1A, D). The maternal blood circulated in the intervillous space. The center of the villi contained loose mesenchymal tissue with fibroblasts, macrophages and blood vessels. The villi were surrounded mainly by a syncytiotrophoblast in which the nuclei were often grouped together to form syncytial knots. Cytotrophoblastic cells were rare (Fig. 1C).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  Term placenta, normoglycemic women. Note the immunolabeling for cytokeratin (red color) in the chorionic villi (v) in the boundary between the maternal and fetal regions indicated by the dashed line in A. In the decidual layer (D) is observed the intensely cytokeratin reactive extravillous tissue (arrowhead, A; evCt, B) and the highlight of the coating of the syncytiotrophoblast (Sc; C) in the villi chorionic. (D) Negative reaction control in which the primary antibody was replaced by non-immune serum. (fc) fetal capillary, (m) mesenchyme, (arrow) in the syncytiotrophoblast cell. The bar in the picture D indicates 400 μm in A and D; and 20 μm in B and C.

The endometrial tissue of the basal plate contained decidual cells (processed from fibroblasts), leukocytes, cytokeratin-positive cytotrophoblastic cells and arteries and veins of different sizes (Fig. 1B). The structure and size of the blood vessels observed within the villi were directly proportional to the thickness of the villi, i.e., large villi contained well-defined arteries and veins and small villi contained only capillaries (Fig. 2, Fig. 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2 

  Term placentas. Portions of intermediate villi (V) in placentas of pregnant normoglycemic – N (A–D), mildly hyperglycemic – H (E–H), gestational diabetic – GD (I–L) and clinically diabetic – CD (M–P) women. Collagen fibers were stained with picrosirius red (B, F, J and N, conventional light microscopy; C, G, K and O, polarization microscopy) in the stroma of intermediate villi, mainly around vessels. Congo Red staining (D, H, L and P) was observed in vascular cells of capillaries and stromal villi (arrows) as a more intense red color. A, E, I and M – hematoxylin and eosin. The bar in A indicates 500 μm in A–E, G, M–N, P and 800 μm in F, H–L, O.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3 

  Term placentas. Portions of terminal villi in placentas of pregnant normoglycemic – N (A–C), mildly hyperglycemic – H (D–F), gestational diabetic – GD (G–I) and clinically diabetic – CD (J–L) women. Staining with picrosirius red (B, E, H and K) shows the presence of collagen in villous stroma (red) whereas staining with Congo Red highlights the presence of reactive cells (arrowhead) in placenta from women with hyperglycemia (F) and clinical diabetes (L). Reactivity in the vascular wall (arrows) was very rare (F). (arrows) villous capillaries. (*) syncytial knots. Note the large number of fetal capillaries in pregnant women with mild hyperglycemia (E). The bar in A indicates 500 μm in A, D, G and J (hematoxylin and eosin) and 250 μm in B–C, E–F, H–I and K–L.

In both groups of women with gestational diabetes and mild hyperglycemia, the mesenchymal tissue of intermediate villi was denser and stained more intensely with hematoxylin and eosin than in normoglycemic women (Fig. 2E, I). The mesenchyma around the blood vessels was more organized and homogeneous, and resembled a thick capsule with a glassy appearance in some specimens. Picrosirius-red staining revealed a greater condensation of collagen material organized around the vessels and villous stroma in general (Fig. 2F–G, J–K), mainly in placentas of women with gestational diabetes (Fig. 2 J–K). The presence of sub-syncytial fibrinoid material was also more common in these women than in villi of the other studied groups; the organization and condensation of this material in the villi of placentas from women with clinical diabetes (Fig. 2 M–P) was similar to that of most pregnant normoglycemic women (Fig. 2 N–O vs. 2B–C).

In all of the groups except normoglycemic women, the mesenchymal cells of the villous stroma showed discrete staining with Congo red (Fig. 2H, L, P); the staining around some arteries probably corresponded to elastic material (Fig. 2P). No amyloid was detected in the mesenchymal interstitium or vasculature in any of the samples.

The mesenchyme generally had a loose arrangement, contained numerous capillaries and was closely associated with the syncytiotrophoblast (Fig. 3A, D, J), except in small terminal villi of women with gestational diabetes in which the mesenchyme was very dense (Fig. 3 G). Staining with picrosirius red indicated a homogenous distribution of collagen in the villous stroma (Fig. 3B, E, K). Women with gestational diabetes had fewer villous capillaries (Fig. 3H) whereas these vessels were abundant in villi of women with mild hyperglycemia (Fig. 3E). Syncytial knots were more evident in placental villi of women with clinical diabetes (Fig. 3J–K). No amyloid deposits were observed, although Congo red did stain the cytoplasm of vascular cells and the mesenchymal stroma (Fig. 3F, L).

3.2. Immunolocalization of VEGF 

VEGF-positive cells were observed in different cellular components of the maternal and fetal placenta. In chorionic villi, VEGF was detected in the cytoplasm of the syncytiotrophoblast, endothelial cells of fetal capillaries and vessels, vessel smooth muscle cells and mesenchymal cells. In the decidua, the extravillous cytotrophoblast and occasionally vascular endothelial cells and muscle also contained VEGF. The extravillous cytotrophoblast was identified by the immunolocalization of cytokeratin in semi-serial sections. The cell type involved and the intensity of staining varied among the groups of women.

VEGF was generally detected in the syncytiotrophoblast cytoplasm (Fig. 4A, D, G, J), in muscle cells and endothelial vessels of intermediate villi (Fig. 4B, E, K), in capillary endothelial cells (Fig. 4A, D, J) and in mesenchymal and cytotrophoblastic cells in the basal decidua proximity to the maternal vessels (Fig. 4C, F, L). This pattern of staining was somewhat different in women with gestational diabetes; in villi of women with clinical diabetes, VEGF was detected in vascular smooth muscle cells, but not in endothelial cells (Fig. 4E, K) (Fig. 5).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4 

  Term placentas. VEGF expression in placentas of pregnant normoglycemic – N (A–C), mildly hyperglycemic – H (D–F), gestational diabetic – GD (G–I) and clinically diabetic – CD (J–L) women. Placental reactive components are indicated by the arrowheads. A, D, G, J and insets show regions of terminal villi whereas B, E, H and K show regions of intermediate villi. In C, F, I, L and inset show the decidua (D) with extravillous cytotrophoblast cells (EVct). (fc) fetal capillaries, (m) mesenchyme, (mv) maternal vessels, (Sc) syncytiotrophoblast, (vv) villous vessels. The bar in F indicates 200 μm in A–C and inset in C; 150 μm in D–F, J–K and inset in A and D; 400 μm in G and H, 40 μm in I and 30 μm in L.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 5 

  VEGF term placental expression scores according to glycemic status of pregnant women. Strong reactions were observed in the syncytiotrophoblast cytoplasm, in muscle cells and endothelial vessels of intermediate villi, in capillary endothelial cells and in mesenchymal and cytotrophoblastic cells in all groups with exception the women with gestational diabetes.

In contrast to the other groups of women, no VEGF was detected in the vascular endothelial and smooth muscle cells of pregnant women with gestational diabetes (Fig. 4G, H). Staining in the in extravillous cytotrophoblast was rather weak (Fig. 4I), but was observed in the mesenchymal cells and cytoplasm of the syncytiotrophoblast (Fig. 4G) Table 2.

Table 2. Reactivity of different placental cell types to VEGF antibody in the normoglicemic, mild hyperglycemic, gestational diabetes and clinical diabetes women.

± few reactive cells.

3.3. Immunolocalization of VEGFR-1 (Flt-1) 

Immunoreactivity for VEGFR-1 was intense in the normoglycemic (Fig. 6A–C), gestational diabetes (Fig. 6G–I) and clinical diabetes (Fig. 6J–L) groups but less marked in the hyperglycemic group (Fig. 6D–F) (Fig. 7). In the latter women, staining was observed in the vasculature of terminal (Fig. 6D) and intermediate (Fig. 6E) villi. The trophoblastic components also contained this receptor (Fig. 6D, F) Table 3.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 6 

  Term placentas. VEGFR-1 expression in placentas of pregnant normoglycemic – N (A–C), mildly hyperglycemic – H (D–F), gestational diabetic – GD (G–I) and clinically diabetic – CD (J–L) women. Placental reactive components are indicated by arrows and arrowheads and are stained brown. A, D, G, J and inset show regions of terminal villi, whereas B, E, H and K represent regions of intermediate villi. In C, F, I and L note the prominent staining of extravillous cytotrophoblast cells (EVct) in the decidua (D). Staining for VEGFR-1 is present in the syncytiotrophoblast (Sc) in all groups. (EVct, arrowhead) extravillous cytotrophoblast, (m) mesenchyme, (mv) maternal vessels, (v) terminal villi, (vv) villous vessels, (*) unstained vessels. The bar in A indicates 600 μm in A, C, I–J and H; 450 μm in B, G, I and K; 120 μm in F and 150 μm in D–E and L.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 7 

  VEGFR-1 term placental expression scores according to glycemic status of pregnant women. Intense reactions were observed in the normoglycemic, gestational diabetes and clinical diabetes groups but less intense in the hyperglycemic group.

Table 3. Reactivity of different placental cell types to VEGFR-1 antibody in the normoglicemic, mild hyperglycemic, gestational diabetes and clinical diabetes women.

3.4. Immunolocalization of VEGFR-2 (KDR) 

vReactivity to VEGFR-2 was observed in all of the groups and the controls, with intense staining in the syncytiotrophoblast cytoplasm (Fig. 8A, D, G and J) (Fig. 9). In endothelial cells, staining was observed in the capillaries, larger vessels and mesenchymal cells of normoglycemic and hyperglycemic women (Fig. 8A–B, D–F). A similar pattern was observed in the villi of women with clinical diabetes except that there was no significant reactivity in capillary endothelium (Fig. 8K). In pregnant women with gestational diabetes, staining was restricted to the syncytiotrophoblast cytoplasm (Fig. 8H–J). In the decidua, many cells of the extravillous trophoblast were stained, especially in placentas from pregnant women with hyperglycemia (Fig. 8C, G, M) Table 4.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 8 

  Term placentas. VEGFR-2 expression in placentas of pregnant normoglycemic – N (A–C), mildly hyperglycemic – H (D–G), gestational diabetic – GD (H–J) and clinically diabetic – CD (K–M) women. Immunolabeling is seen as brown staining. Reactive cells of the extravillous trophoblast are indicated by the arrowheads. (D) basal decidua, (fc) fetal capillaries, (m) mesenchyme, (Sc) syncytiotrophoblast, (v) intermediate villus, (vv) villous capillaries. The bar in A indicates 500 μm in A–B, G–I and K; 300 μm in D–F; 250 μm in C, J, L and the inset in H and 150 μm in M.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 9 

  VEGFR-2 term placental expression scores according to glycemic status of pregnant women. Intense staining was observed in the syncytiotrophoblast, capillaries, larger vessels and mesenchymal cells in of normoglycemic and hyperglycemic women. No significant reactivity was observed in clinical diabetes and gestational diabetes that was restricted to the syncytiotrophoblast cytoplasm.

Table 4. Reactivity of different placental cell types to VEGFR-2 antibody in the normoglicemic, mild hyperglycemic, gestational diabetes and clinical diabetes women.

± few reactive cells.

3.5. Western blotting 

Immunoreactive bands of approximately 50, 180 and 150 kDa were detected by VEGF, VEGFR-1 and VEGFR-2 antibodies, respectively, in extracts of villous and extravillous placental tissue of pregnant women with normal glycemia, mild hyperglycemia, gestational diabetes and clinical diabetes but with diffent intensities (Fig. 10). An additional band (∼75 kDa) that reacted with VEGFR-1 antibody was also detected in extravillous tissue of women with normal glycemia, gestational diabetes and clinical diabetes. The major differences in intensity of expression between the villous and extravillous regions were:

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 10 

  Qualitative analyses of Western blotting for VEGF (A), VEGFR-1 (B) and VEGFR-2 (C) in placenta of pregnant normoglycemic (1), gestational diabetic (2), clinically diabetic (3) and mildly hyperglycemic (4) women. The membrane strips shown at the bottom of each blot in A–C correspond to membranes stained with Ponceau red after each transfer.

Back to Article Outline

4. Discussion 

VEGF, a pivotal inducer of endothelial cell proliferation and migration, is a critical factor in angiogenesis. Although the extent of its role in placental development is not fully understood, VEGF is widely expressed by endothelial cells and by cytotrophoblast and, maternal and fetal macrophages [24], [25]. In this work, we studied the protein expression and distribution of VEGF, VEGFR-1 (Flt-1) and VEGFR-2 (KDR) in placentas from mildly hyperglycemic women; the results were compared with those for placentas from women with normal glycemia, clinical diabetes and gestational diabetes.

4.1. Placental morphology in glycemic disturbances 

The vascular changes associated with diabetes mellitus have been extensively studied because of the close relationship between this pathology and cardiovascular diseases [26], [27]. Diabetes is caused by defective insulin secretion and/or action that leads to abnormal protein, fat and carbohydrate metabolism, resulting in hyperglycemia [28], [29]. Diabetes and insulin resistance lead to endothelial dysfunction that may diminish the anti-atherogenic function of vascular endothelium [26]. The associated hyperglycemia produces marked changes in arterial vessels: in arteries there is a gradual decrease in the elasticity of the vessel wall that leads to hardening and increased susceptibility to atherosclerosis [27] whereas in the microcirculation alterations in the basal membrane of arterioles leads to weakening and dilation of the capillary walls with a tendency to rupture (microangiopathy) [30], [31]. These changes are common in diabetic retinopathy and nephropathy and can cause blindness and kidney failure, respectively [31].

As shown here, term placentas had intermediate and terminal villi with arterial and venous vessels of different sizes and complexities. The capillaries in the terminal villi of all groups studied had a normal morphology and were structurally similar.

The vasculature of intermediate villi showed alterations that varied according to the clinical pathology. In placentas from diabetic pregnant women the blood vessel walls were apparently thicker and the surrounding mesenchyme was denser than in pregnant normoglycemic women. This change was very subtle in women with clinical diabetes but very marked in women with gestational diabetes.

Hyperglycemia has been associated with the extracellular deposition of vascular amyloid. These deposits consist of a heterogeneous mixture of fibrillar proteins that develop in a variety of tissues in association with numerous diseases [32], [33]. Microscopically the amyloid is amorphous, acellular, hyaline, eosinophilic and homogeneous. Although the appearance of these deposits varied among pregnant women with varying degrees of hyperglycemia, a discrete production of these proteins could explain the appearance of more condensed mesenchyme around blood vessels. However, congo-red staining showed that there was no significant amyloid deposition in any of the cases analyzed.

In addition to amyloid deposition, hyperglycemia also results in exacerbated collagen production by the vascular endothelium. This enhanced collagen formation probably reflects an imbalance in the physical and chemical stimuli that normally regulate this production in the vasculature [34].Our results indicate that the accumulation of collagen in the tunica media of vessels in intermediate villi and in stromal villi in general varied with the extent of the glucose alterations. The normal collagen organization seen around blood vessels in normoglycemic pregnant women based on staining with Picrosirius Red was altered primarily in the placentas of women with gestational diabetes and, to a lesser extent, in women with hyperglycemia and clinical diabetes. Li et al. [35] also reported a close association between the hyperglycemia of type II diabetes (non-insulin-dependent) and hypertrophy of the extracellular matrix and collagen production. Vranes et al. [36] observed that the loss of blood vessel elasticity in diabetes was possibly related to the deposition of collagen in the tunica media.

Our findings raise two fundamental questions, namely, (1) why are fetal vessels so susceptible to change? and (2) why is the collagen deposition in fetal vessels so marked in women with gestational diabetes? We believe that part of the answer to these questions is that the alterations are related to the treatment used to control blood glucose levels. Women with clinical diabetes are normally submitted on insulin therapy, which means that they probably have more balanced blood glucose levels than women with gestational diabetes. On the other hand, the changes in fetal blood vessels may reflect the fetal blood glucose levels and these may not be significantly affected by the treatment given to the mother.

The terminal villi of placentas from pregnant women with mild hyperglycemia contained more capillaries than in the normoglycemic, gestational diabetes and clinical diabetes women. This finding corroborated the morphometric analysis of Calderon et al. [9] who reported increased capillarization of these villi in a similar group of pregnant women. These results were interpreted as a response to placental hypoxia induced by hyperglycemia. High circulating blood glucose concentrations and/or an absolute or relative deficiency in insulin production leads to changes in the structure of hemoglobin, with increased glycosylation of this protein that can lead to erythrocyte deformation and increased blood viscosity. These factors contribute significantly to reduce the affinity of hemoglobin for oxygen, leading to poor oxygen transport and local hypoxia [37], [38].

According to Calderon et al. and Rudge et al. [9], [17], the greatest number of villi and corresponding vessels in pregnant women with mild hyperglycemia provides a greater surface for maternal-fetal exchange. This functional adaptation would facilitate the passage of glucose to the developing fetus, thereby explaining fetal macrosomia. In addition, these authors also observed that pregnant women with gestational diabetes or clinical diabetes had the same villous surface area as normoglycemic pregnant women, but a smaller vascular surface area and lower degree of capillarization; these morphological differences could explain the restricted intrauterine growth of the fetus in these women [9]. Our analysis also revealed a reduction in the capillary network in pregnant women with gestational diabetes, although this assessment was not based on a quantitative evaluation.

4.2. VEGF and receptors immunolocalization 

VEGF and its receptors are essential for vascular development, with VEGF being a potent inducer of endothelial cell proliferation, activation and migration. VEGF was detected in placental extracts and in different cell types of the intermediate and terminal villi in normoglycemic pregnant women. In these women, VEGF was detected in endothelial cells, vascular smooth muscle cells, the mesenchyma, and syncytial and extravillous trophoblast. These findings corroborate those of other studies [5], [24], [25]. Ferrara et al. [39] emphasized the role of VEGF in promoting vascular endothelial cell growth and in increasing the number of vessels and capillaries, thereby ensuring the adequate supply of nutrients to the fetus. The production of VEGF by so many different cell types in the placenta reinforces its relevance to the success of pregnancy.

Small but important changes in the distribution of VEGF were observed in the placentas of women with hyperglycemia, e.g., no VEGF was detected in the placental cellular compartments, which suggested a decrease in VEGF production. In contrast, VEGF was always detected in the trophoblast, independently of the group analyzed. Among the various groups examined, mildly hyperglycemic women showed the greatest similarity in VEGF expression profile when compared to normoglycemic women. In the former group, only the vascular smooth muscle cells of intermediate villi did not react with the antibody. In women with gestational diabetes there was a significant decrease in the reactivity among placental components and VEGF was generally detected only in the trophoblast. In women with clinical diabetes, the staining was intermediate, with VEGF being detected in the endothelium of large vessels but not in capillaries of terminal villi. In addition to these villi, marked staining was also observed in the mesenchymal compartment. Other studies have also emphasized the production of VEGF by stromal cells, including Hofbauer cells, especially in intermediate villi [5], [24], [25]. This VEGF production may ensure the neoformation of placental vessels and their organization to meet the needs of the fetus and placenta, thus playing an important role in the adaptive capacity of the body.

The target cells for VEGF in the chorionic villi were determined based on the immunolocalization of VEGFR-1 and VEGFR-2. There was a significant decrease in the reactivity of the cellular compartment to anti-VEGFR-1 in the villi of women with mild hyperglycemia compared to the other groups. These data were confirmed by western blots. In women with gestational diabetes and clinical diabetes VEGFR-1 expression was very similar to that observed in the normoglycemic group. In general, trophoblast and mesenchymal cells were reactive, except for the smooth muscle cells of villi vessels. The endothelial cells of capillaries and large vessels were particularly reactive. In women with mild hyperglycemia, only the trophoblast was reactive. The binding between VEGF and VEGFR-1 triggers a signaling pathway that negatively regulates angiogenesis. In addition, VEGFR-1 acts as a receptor to sequester VEGF [40], [41] and can modulate the reorganization of actin filaments by p38 MAPK (mitogenic-activated protein kinase), but not the proliferative pathway in endothelial cells. By using antibodies to block VEGFR-1, Kanno et al. [42] inhibited the migration of endothelial cells in human umbilical vein in the presence of VEGF, and completely prevented angiogenesis. Over expression of this receptor may therefore provide a mechanism to limit placental angiogenesis. The high expression of VEGFR-1 seen in women with normoglycemia, gestational diabetes and clinical diabetes contrasted with the low levels detected in pregnant women with mild hyperglycemia.

The expression of VEGFR-2 is associated with the stimulation of vascular endothelial cells, survival/growth and angiogenesis, and this receptor has an essential role in the differentiation of endothelial cell progenitors [4], [43], [44]. As shown here, VEGFR-2 expression was particularly high in the placental capillary and large vessel endothelial cells of pregnant normoglycemic, mildly hyperglycemic and clinically diabetic women. Unlike VEGFR-1, in women with gestational diabetes VEGFR-2 was found only in reactive cells in the extravillous cytotrophoblast in decidua.

Kumazaki et al. [14] and Zhao et al. [13] observed strong staining for VEGF in villous blood vessels during hypoxia associated with a marked increase in VEGFR-2 expression in endothelial cells. These authors suggested that an association between VEGF/VEGFR-2 in intermediate villi is essential for the induction of hypercapillarization under mild and persistent hypoxia/ischemia, through classical feedback mechanism. Thus, the condition of hyperglycemia induces a state of mild and persistent ischemia and hypoxia with subsequent increase of hypercapillarization. In contrast, high blood glucose levels trigger severe hypoxia/ischemia, with inhibition of binding VEGF/VEGFR-2 and consequent reduction of hypercapillarization. In trophoblastic cells, VEGF acting via VEGFR-2 apparently stimulates the production of placental growth factor (PlGF), which then acts synergistically with VEGF in ischemia or inflammation to increase tissue angiogenesis [45].

In conclusion, the results described here reveal that hypercapillarization previously described in the placenta of women with mild hyperglycemia may result from the downregulation of VEGFR-1, thereby enhancing VEGF/VEGFR-2 interaction and the activation of angiogenesis. These changes may be part of a mechanism designed to compensate for changes in blood flow caused by the hemodynamic alterations associated with hyperglycemic during pregnancy.

Back to Article Outline

Competing interests 

None.

Back to Article Outline

Acknowledgments 

The authors thank Rosangela Oliveira for her excellent technical assistance, Mara Sandra Hoshida and Sara Maria Zago Gomes for advice on technical procedures, Profa. Dra. Fátima Luiz Böttcher and Profa. Liliana Andrade for editorial assistance and Dr. Stephen Hyslop for the version to the English language.

Back to Article Outline

References 

1. Vuorela P. Vascular endothelial growth factor, its receptors, and the Tie receptor in normal and complicated pregnancy. University of Helsinki, Finland. Academic Dissertation; 2000.
   • View In Article
2. Demir R, Kaysli UA, Cayli S, Huppertz B. Sequential steps during vasculogenesis and angiogenesis in the very early human placenta. Placenta. 2006;27:535–539
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (163 KB)
   • CrossRef
3. Demir S, Kayisli UA, Seval Y, Celik-Ozenci C, Korgun ET, Demir-Weusten AY, et al. Sequential expression of VEGF and its receptors in human placental villi during very early pregnancy: differences between placental vasculogenesis and angiogenesis. Placenta. 2004;25:560–572
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (1123 KB)
   • CrossRef
4. Breier G. Angiogenesis in embryonic development – a review. Placenta. 2000;21(A):S11–S15
   • View In Article
   • Abstract
   • Full-Text PDF (140 KB)
   • CrossRef
5. Cooper JC, Sharkey AM, Mclaren J, Charnock-Jones DS, Smith SK. Localization of vascular endothelial growth factor and its receptor, flt, in human placenta and decidua by imunohistochemistry. J Reprod Fertil. 1995;105:205–213
   • View In Article
   • MEDLINE
   • CrossRef
6. Charnock-Jones DS, Kaufmann P, Mayhew TM. Aspects of human fetoplacental vasculogenesis and angiogenesis. Placenta. 2004;25:103–113
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (599 KB)
   • CrossRef
7. Kingdom J, Huppertz B, Seaward G, Kaufmann P. Development of the placental villous tree and its consequences for fetal growth. Eur J Obstet Gynecol Reprod Biol. 2000;92:35–43
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (247 KB)
   • CrossRef
8. Jirkovska M, Kubinova L, Jancek J, Moravcova M, Krejci V, Karen P. Topological properties and spatial organization of villous capillaries in normal and diabetic placentas. J Vasc Res. 2002;39:268–278
   • View In Article
   • MEDLINE
   • CrossRef
9. Calderon IMP, Damasceno DC, Amorin RL, Costa RAA, Brasil MAM, Rudge MVC. Morphometric study of placental villi and vessels in women with mild hyperglycemia or gestational or overt diabetes. Diabetes Res Clin Pract. 2007;78:65–71
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (505 KB)
   • CrossRef
10. Rudge MVC, Calderon IMP, Ramos MD, Rodrigues MAM. Diabetes and experimental pregnacy in rats: course of maternal blood glucose levels and its repercussions on the blood glucose levels and pancreas of newborn pups. Braz J Med Biol Res. 1995;28:219–225
    • View In Article
    • MEDLINE
11. Rudge MVC, Gomes CMM, Calderon IMP, Ramos MD, Abbade JF, Oliveira MGM, et al. Study of the evolution of the placenta and fetal pancreas in the pathophysiology of growth retardation intrauterine due to restricted maternal diet. Sao Paulo Med J. 1999;117:49–56
    • View In Article
    • MEDLINE
12. Eriksson J, Hakan Borg LA, Forsberg H, Styrud J. Diabetic embryopathy. Studies with animal and in vitro models. Diabetes. 1991;40(2):94–98
    • View In Article
13. Zhao B, Cai J, Boulton M. Expression of placenta growth factor is regulated by both VEGF and hyperglycemia via VEGFR-2. Microvasc Res. 2004;68:239–246
    • View In Article
    • MEDLINE
    • CrossRef
14. Kumazaki K, Nakayama M, Suoriyuki N, Wada Y. Expression of vascular endothelial growth factor, placental growth factor and their receptors Flt-1 and KDR in human placenta under pathologic conditions. Hum Pathol. 2002;33:1069–1076
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1161 KB)
    • CrossRef
15. Kerche LTRL, Abbada JF, Costa RAA, Rudge MVC, Calderon IMP. Fetal macrosomia risk in pregnancies complicated by diabetes or daily hyperglycemia. Rev Bras Ginecol Obstet. 2005;27:580–587
    • View In Article
16. Jansson T, Ekstrand Y, Wennwegren M, Powell T. Placental glucose transport in gestational diabetes mellitus. Am J Obstet Gynecol. 2001;184:111–116
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (79 KB)
    • CrossRef
17. Rudge MVC, Calderon IMP, Ramos MD, Brasil MAP, Rugolo LMSS, Bossolan G, et al. Hiperglicemia maternal diária diagnosticada pelo perfil glicêmico: um problema de saúde pública materno perinatal. Rev Bras Ginecol Obstet. 2005;27:691–697
    • View In Article
18. Rudge MVC. Perfil glicêmico e teste de tolerância oral à glicose (GGT) no diagnóstico do diabetes na gestação (Livre-Docência). Botucatu, Faculdade de Medicina de Botucatu, UNESP; 1983;p83
    • View In Article
19. Rudge MVC, Peraçoli JC, Berezowski AT, Calderon IMP, Brasil MAM. The oral glucose tolerance test is a poor predictor of hyperglycemia during the pregnancy2. Braz J Med Biol Res. 1990;23:1079–1089
    • View In Article
    • MEDLINE
20. Americam Diabetes Association Gestational Diabetes Mellitus . Position statements. Diabetes Care. 2004;27(1):S88–S90
    • View In Article
21. Gillmer MDG, Beard RW, Brooke FM, Oakley NW. Carbohydrate metabolism in pregnancy. PartI. Diurnal plasma glucose profile in normal and diabetic women. Br Med J. 1975;16:399–404
    • View In Article
22. Meloan SN, Puchtler H. Demonstration of amyloid with Mesitol WLS-Congo Red: application of a textile auxiliary to histochemistry. Histochemistry. 1978;58:163–166
    • View In Article
    • MEDLINE
    • CrossRef
23. Junqueira LC, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Hist Jahrb. 1979;11:447–455
    • View In Article
24. Helske S, Vuorella P, Carpen O, Horning C, Weich H, Halmesmaki E. Expression of vascular endothelial growth factor receptors 1, 2 and 3 in placentas from normal and complicated pregnancies. Mol Hum Reprod. 2001;7:205–210
    • View In Article
    • MEDLINE
    • CrossRef
25. Vuorela P, Halmesmaki E. Vascular endothelial growth factor, its receptors, and the tie receptors in the placental bed of women with preeclampsia, diabetes, and intrauterine growth retardation. Am J Perinatol. 2006;23:255–263
    • View In Article
    • MEDLINE
    • CrossRef
26. Hadi HA, Suwaidi JA. Endothelial dysfunction in diabetes mellitus. Vasc Health Risk Mang. 2007;3:853–876
    • View In Article
27. Jin SM, Noh CI, Yang SW, Bae EJ, Shin CH, Chung HR, et al. Endothelial dysfunction and microvascular complications in type 1 diabetes mellitus. J Korean Med Sci. 2008;23:77–82
    • View In Article
    • CrossRef
28. Klip A, Marette A, Dimitrakoudis D, Ramlat T, Giacca A, Shi ZQ, et al. Effect of diabetes on glucoregulation. From glucose transporters to glucose metabolism in vivo. Diabetes Care. 1992;15:1747–1766
    • View In Article
    • MEDLINE
29. Taskinen MR. Criteria for metabolic control and intervention in diabetes. Diabetes. 1996;45(3):S120–S122
    • View In Article
30. Vlassara H, Brownlee M, Cerami A. Accumulation of diabetic rat peripheral nerve myelin by macrophages increases with the presence of advanced glycosylation endproducts. J Exp Med. 1984;160:197–207
    • View In Article
    • MEDLINE
    • CrossRef
31. Browlee M. Biochemistry and molecular cell biology of diabetic complications. Nature. 2001;13:813–820
    • View In Article
32. Hull RL, Westermark GT, Westermark P, Kahn SE. Islet amyloid: a critical entity in the pathogenesis of type 2 diabetes. J Clin Endocrinol Metab. 2004;89:3629–3643
    • View In Article
    • MEDLINE
    • CrossRef
33. Butler AE, Jang J, Gurlo T, Carty MD, Soeller WC, Bulter PC. Diabetes due to a progressive defect in beta-cell mass in rats transgenic for human islet amyloid polypeptide (HIP Rat): a new model for type 2 diabetes. Diabetes. 2004;53:1509–1516
    • View In Article
    • MEDLINE
    • CrossRef
34. Esper RJ, Vilariño JO, Machado RA, Paragano A. Endothelial dysfunction in normal and abnormal glucose metabolism. Adv Cardiol. 2008;45:17–43
    • View In Article
    • CrossRef
35. Li Q, Wu SH, Pan JM, Tang JL, Zhang Y, Lu HJ, et al. Glycated albumin level and its influential factors in type 2 diabetes mellitus. Zhonghua Yi Xue Za Zhi. 2007;87:2994–2996
    • View In Article
36. Vranes D, Cooper ME, Dilley RJ. Cellular mechanisms of diabetic vascular hypertrophy. Microvasc Res. 1999;57:8–18
    • View In Article
    • MEDLINE
    • CrossRef
37. Iino K, Yoshinari M, Doi Y, Shinohara N, Iwase M, Fujishima M. Reduced tissue oxygenation and its reversibility by glycemic control in diabetic patients. Diabetes Res Clin Pract. 1997;34:163–168
    • View In Article
    • Abstract
    • Full-Text PDF (470 KB)
    • CrossRef
38. Watala C, Golaski J, Witas H, Gurbiel R, Gwoidzifiski K, Trojanowski Z. The effects of in vivo and in vitro non-enzymatic glycosylation and glycoxidation on physico-chemical properties of haemoglobin in control and diabetic patients. Int J Biochem Cell Biol. 1996;28:1343–1403
    • View In Article
39. Ferrara N, Gerber HP, Lecouter J. The biology of VEGF and its receptors. Nat Med. 2003;9:669–676
    • View In Article
    • MEDLINE
    • CrossRef
40. Liekens S, Clercq ED, Neyts J. Angiogenesis: regulators and clinical applications. Biochem Pharmacol. 2001;61:253–270
    • View In Article
    • MEDLINE
    • CrossRef
41. Hicklin D, Ellis LM. Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol. 2005;23:1011–1027
    • View In Article
    • MEDLINE
    • CrossRef
42. Kanno S, Oda N, Abe M, Terai Y, Ito M, Shitara K, et al. Roles of two VEGF receptors, Flt-1 and KDR, in the signal transduction of VEGF effects in human vascular endothelial cells. Oncogene. 2000;19:2138–2146
    • View In Article
    • MEDLINE
    • CrossRef
43. Takashi H, Shibuya M. The vascular endothelial growth factor (VEGF)/VEGF receptor system and its role under physiological and pathological conditions. Clin Sci. 2005;109:227–241
    • View In Article
44. Shibuya M, Claesson-Welsh L. Signal transduction by VEGF receptors in regulation of angiogenesis and lymphagiesis. Exp Cell Res. 2006;312:549–560
    • View In Article
    • MEDLINE
    • CrossRef
45. Carmeliet P, Moons L, Lutun A. Synergism between vascular endothelial growth factor and placental growth factor contributes to angiogenesis and extravasation in pathological coditions. Nat Med. 2001;7:575–583
    • View In Article
    • MEDLINE
    • CrossRef