|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
School of Biomedical Sciences,2 University of Nottingham, Nottingham, NG7 2UH United Kingdom
School of Human Development,3 University of Nottingham, Nottingham, NG5 1PB United Kingdom
School of Biomedical Sciences,4 University of Ulster, Coleraine, BT52 1SA United Kingdom
ABSTRACT
The placenta is a specialized vascular interface between the maternal and fetal circulations that increases in size to accommodate the nutritional and metabolic demands of the growing fetus. Vascular proliferation and expansion are critical components of placental development and, consequently, interference with vascular growth has the potential to severely restrict concurrent development of both the placenta and fetus. In this study, we describe the effects of an antiangiogenic agent, TNP-470, on placental vascular development and the induction of a form of intrauterine growth restriction (IUGR) in mice. Administration of TNP-470 to dams in the second half of pregnancy resulted in a smaller maternal weight gain accompanied by decreased placental and fetal sizes in comparison with control animals. Total numbers of fetuses per litter were not affected significantly. Stereological analysis of placentas revealed no changes in the combined lengths of vessels. However, the mean cross-sectional areas of maternal and fetal vessels in the labyrinth of TNP-470-treated mice were reduced at Embryonic Day 13.5 (E13.5) but not at E18.5. Further analysis showed reduced placental endothelial proliferation at E13.5 and E18.5 in TNP-470-treated animals. No other structural or morphometric differences in placentas were detected between TNP-470-treated and control mice at E18.5. This study provides conclusive evidence that administration of TNP-470 interferes with placental vascular proliferation and vessel caliber and results in a reproducible model of IUGR.
blood vessel, developmental biology, embryo, endothelium, intrauterine growth restriction, mouse, placenta, pregnancy, proliferation, TNP-470, trophoblast
INTRODUCTION
Intrauterine growth restriction (IUGR) is a serious cause of fetal morbidity and mortality [1]. Most cases of IUGR are associated with placental insufficiency, reflecting an underlying pathology resulting in an inability of the placenta to supply the metabolic demands of the rapidly growing fetus [2]. By juxtaposition of the maternal and fetal vasculatures, the placenta allows close communication between the two circulations. In human IUGR, the placenta may be smaller and display abnormal vascular development [3, 4] as a result of defective trophoblast invasion of the decidua or defective fetoplacental perfusion or morphogenesis. Both processes may influence vascular impedances. In this study, we test the hypothesis that inhibition of placental endothelial cell proliferation (by TNP-470) affects vascular development and provides an experimental model of IUGR.
TNP-470, an endothelium-specific proliferation inhibitor, was discovered as a fungally derived contaminant of primary endothelial cell cultures [5]. Although it was shown to covalently and irreversibly bind to methionine aminopeptidase 2, this mechanism is not responsible for its ability to inhibit endothelial cell proliferation [6]. Rather, the interference of TNP-470 with endothelial proliferation occurs within the G1 phase of the cell cycle [7] and is mediated via activation of the TRP53 pathway, leading to accumulation of the cyclin-dependent kinase inhibitor CDKN1A (also known as WAF1 or CIP1) [8, 9].
When TNP-470 is administered during murine estrus, it results in a postmenopausal appearance of the uterus and failure to conceive. Single doses of TNP-470 administered on Embryonic Day 0 (E0, the day of conception) or E7 lead to spontaneous abortion of all conceptuses due to placental failure [10]. However, when administered on E14, no effect of TNP-470 on subsequent embryonic growth was reported. It is clear from previous studies [10, 11] that interference with placental development before or during chorioallantoic fusion at around E9.0E10.0 is incompatible with embryonic survival. It is less clear what pathological sequelae may result from interference with endothelial growth after E10, when the placenta is functionally mature [12].
In humans, various factors are associated with development of IUGR, including high maternal blood pressure, nutrient restriction, smoking, genetic predisposition, and maternal infection [13]. Most current studies on IUGR in animal models focus on the consequences for the neonate [1416]. One reason for this focus is the wide acceptance that IUGR can lead to increased incidence of serious diseases in adult life (such as atherosclerosis and diabetes), otherwise known as the Barker hypothesis [13]. As IUGR is invariably accompanied by placental insufficiency, it is possible that at least part of the mechanism of IUGR is mediated via inappropriate vascular growth during placental maturation [17].
In the mouse, fetoplacental blood vessels are characterized by a lining of endothelial cells whereas maternal blood spaces are lined by trilaminar trophoblast [18, 19]. This allows the differentiation of fetal blood vessels from maternal blood spaces and facilitates stereological analysis of murine placentas. Consequently, we examined the effect of therapeutic doses of TNP-470, delivered every other day from E10.5 to E16.5 and examined the effects on maternal, placental, and fetal-growth parameters. We employed stereological methods to investigate the effects of TNP-470 on the major placental tissue compartments and, more specifically, the vasculature. In addition, several measures of maternal and fetal growth were used to assess the effects of TNP-470 on fetal development.
MATERIALS AND METHODS
Determination of Bioactivity and Specificity of TNP-470 In Vitro
To assess the bioactivity and relative endothelial specificity of TNP-470 (Takeda Chemical Industries, Ltd., Tokyo; molecular weight 401.89), in vitro proliferation and apoptosis assays were performed on Human Umbilical Vein Endothelial Cells (HUVEC, passage 38), HUman Lung Endothelial Cells (HULEC, passage 8) and HUman Adipose Microvascular Endothelial Cells (HUAMEC, passage 8), a St. Georges Hospital Placental trophoblast cell Line (SGHPL4, passage 8), human Asian endometrial adenocarcinoma (Ishikawa, passage 4; ECACC, Salisbury, U.K.) and human Caucasian endometrial adenocarcinoma (MFE-296, passage 10; ECACC). HUVEC and SGHPL4 cells were plated at either 5 x 103 or 5 x 104 (subconfluent and confluent, respectively) in 96-well microtiter plates (Costar, Cambridge, MA). Ishikawa and MFE-296 cells were plated at 5 x 103 and the HULEC and HUAMEC at 1 x 104 (confluent only).
All endothelial cells were maintained on gelatinized plastic in M199 medium with Earles salts, 20% iron-supplemented calf serum, 0.15% sodium bicarbonate, 15 U/ml Heparin, 0.014 M HEPES, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM L-glutamine, 10 ng/ml fibroblast growth factor, and epidermal growth factor. SGHPL4 cells [20] were passaged on Ham F-10 nutrient mixture supplemented with 10% fetal bovine serum (FBS), 2.5 µg/ml amphotericin B, 200 U/ml penicillin, and 200 µg/ml streptomycin. The Ishikawa cells were grown in minimum essential medium (MEM) containing 5% FBS, 1% nonessential amino acids, 0.5% Pen/Strep, and 0.1% Amphortericin-B. MFE-296 cells were grown in MEM, with 10% FBS, 200 mM L-Glutamine, 0.5% Pen/Strep, and 0.1% Amphotericin-B.
Assessment of the bioactivity of TNP-470 was based on a previously published method [21]. Briefly, cells were seeded until confluent and subconfluent in 96-well plates (Costar) and incubated at 37°C with 5% CO2. After 24 h of culture, TNP-470 diluted in the respective media, was added at doses ranging from 10 ng/ml to 100 µg/ml. Cells were also incubated with media alone, media with 0.1% dimethyl sulfoxide (DMSO), or media containing 3 µg/ml Staurosporine (a concentration at which the proliferation and apoptosis of HUVECs was known to be affected) as a positive control. Identical plates were incubated for 48 h at 37°C with 5% CO2, after which time fresh media containing TNP-470 was added. After a further 48 h, and according to the relevant manufacturers' instructions, the proliferative activity of the cells was determined using a 5-bromo-2'-deoxy-uridine (BrdU) labeling and detection kit (Roche Diagnostics Ltd.) and apoptosis was investigated using an APOPercentage apoptosis assay kit (Biocolor Ltd.). All assays were performed in quadruplicate with the values expressed as means and standard error of the means (SEM).
Animal model
The animal experimentation in this study was approved and conducted according to both institutional and national (Home Office) ethics committee guidelines. Adult C57Bl6J mice were obtained from Charles River Breeding Laboratories U.K. and bred in-house under standard Home Office Procedures. Timed matings were arranged between male and female C57Bl6J mice, the presence of a vaginal plug being used to define E0.5. Either 0, 3, 30 or 100 mg/kg bodyweight of O-(chloroacetylcarbamoyl) fumagillol (TNP-470) was injected subcutaneously into the dorsum of pregnant dams at E10.5, E12.5, E14.5, and E16.5. On day E18.5, pregnant mice were killed by cervical dislocation. Subgroups of mice were used in autoradiographic experiments to identify endothelial cells in S-phase of the mitotic cell cycle (see below).
Gravid uteri were carefully dissected free and, after amnionectomy, fetuses were delivered and killed and placentas were dissected free from the uterus and fetus. Fetuses and placentas were fixed overnight in either 4% paraformaldehyde in PBS (pH 7.4) for immunohistochemistry, 10% formalin for histological examination, or 2.5% glutaraldehyde in cacodylate buffer for ultrastructural analysis. Following fixation in 10% formalin, the brains, livers, and hearts were dissected free, blotted dry, and the weights recorded from a total of 10 randomly chosen E18.5 vehicle-only-treated control and nine 30-mg/kg TNP-470-treated fetuses.
Alizarin Red and Alcian Blue Staining
Fetuses from vehicle-only, 3 mg/kg, and 30 mg/kg TNP-470 treatment groups were fixed in formalin overnight, the skin was removed, and the samples were stained in Alcian Blue for 3 days, dehydrated through an ethanol series, immersed in 1% potassium hydroxide (KOH) for 2 days, and stained with Alizarin Red S Solution until the bones were purple. Following Alizarin Red staining, the fetuses were placed in 1% KOH for 3 h, a minimum of three times, and finally taken through increasing concentrations of glycerol/KOH. Fetuses stained were of a similar size to enable direct comparisons of control and treated fetuses.
Assessment of Cell Cycle S-Phase Labeling by 3H-TdR Autoradiography
A dose of 3 µCi/gm bodyweight 3H-thymidine was injected intraperitoneally to label cells in S-phase of the cell cycle, 1 h before euthanasia. Samples of placenta from control and 30 mg/kg TNP-470-treated groups were embedded in poly-hydroxyethyl methacrylate (pHEMA). Sections 1.5 µm thick were cut on a microtome (Reichert-Jung 1140 Autocut) and placed on glass slides. In a dark room, slides were coated with a solution of NTB2 (1:1 with aqua dest; Kodak) photographic emulsion and dried overnight in a box containing desiccant. Following 3 wk of exposure at 4°C, slides were developed in D19 (Kodak), washed for 1 min in a stop bath (1% acetic acid in water), and fixed in acid-hardener fixer for 4 min (Kodak) followed by a 10-min washing in running tap water. Following Toluidine blue staining, slides were viewed on a microscope (Olympus) at 250x magnification and systematic random samples of labyrinthine tissue were captured as electronic images on an Olympus T4040 digital camera. Only endothelial cell profiles with three or more autoradiographic grains overlying a nuclear profile were considered to be positive [22]. Resulting profile counts were analyzed by one-way ANOVA.
Transmission Electron Microscopy
Placental samples were embedded in Epon Araldite (TAAB, Holland) and semithin sections (~ 0.8 µm thickness) were cut from blocks on a Reichert-Jung Ultracut-E microtome (Germany) and stained with Toluidine blue (5%) to determine location for further ultrastructural analysis. Ultrathin sections of gold interference color (approximate 80 nm thickness) were cut and contrasted with uranyl acetate and lead citrate. These sections were viewed using a JEOL JEM-1010 transmission electron microscope (Japan) at an accelerating voltage of 80 kV. Electronic images were captured on a Kodak Megaplus camera model 1.6i (Kodak).
Stereological Analysis
Systematic random sampling [23] of placental sections was carried out as described previously [24] using the QProdit (v3.1) image analysis program (Leica Microsystems Imaging Solutions Ltd.) on a Leitz dialux 20EB (Leitz, Germany). Images were captured on a JVC TK-C1380 color video camera. Overlays of test points were randomly superimposed on images, and volume densities of selected components were estimated by point-counting procedures [25]. For analysis of blood vessel length densities, transections of fetal blood vessels and maternal blood spaces from within the labyrinthine layer were counted [25]. From volume and length densities, the mean cross-sectional areas of vessels were calculated. Relative volumes and lengths were converted into total volumes and lengths per placenta by taking into account organ weights. Results were analyzed using two-way ANOVA to test for the presence of main (treatment and gestational age) effects and interaction (treatment x age) effects. Where necessary, post hoc testing (Scheffé method) was applied. Analyses were undertaken using SPSS v11.0 or Unistat v5.5 software. A probability level of P < 0.05 was considered significant.
Immunohistochemistry
Freshly dissected samples of placenta from control and 30 mg/kg TNP-470-treated mice were fixed in 4% paraformaldehyde in PBS (pH 7.4) for 24 h and transferred to PBS before paraffin embedding. Sections (thickness 5 µm) were cut and placed on glass slides, deparaffinized, and rehydrated in a descending series of ethanol concentrations. Following rehydration, sections were immersed in PBS for 5 min and endogenous peroxidase activity quenched by addition of 1% H2O2 in absolute methanol. Subsequently, slides were washed in three changes of PBS (5 min each) and antigen retrieval performed by microwaving (800 W) for 20 min in 0.1 M citrate buffer. Nonspecific antibody binding was blocked by incubating the sections in 20% normal goat serum for 30 min. An anti-TRP53 (transformation-related protein 53) mouse monoclonal antibody (PAB246) and an anti-CDKN1A (cyclin-dependent kinase inhibitor 1A) antibody (F-5; both from Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a dilution of 1:100 in PBS were applied for 4 h at room temperature. Sections were washed in PBS and a Vectastain Elite ABC mouse kit (PK-6102; Vector Laboratories Ltd., Peterborough, U.K.) used according to manufacturer's instructions. Following immunostaining, samples were dehydrated through ascending grades of ethanol, cleared in xylene, and mounted in DePeX (BDH Laboratory Supplies; Poole, U.K.).
RESULTS
Confirmation of Bioactivity and Specificity of TNP-470 In Vitro
Incubation with TNP-470 for different periods (24, 48, or 72 h) was examined, but there were no significant differences in absorbance (in either the bromodeoxyuridine [BrdU] or ApoPercentage assay) in any cell line (data not shown). After 96 h of incubation with TNP-470, many of the cell lines showed a distinct response to the drug. Therefore, in subsequent assays, all cell lines were incubated for 96 h with either a dose of TNP-470 or vehicle-only control medium. Spectrophotometric determination of BrdU incorporation at a wavelength of 405 nm revealed a dose-dependent decrease in fluorescence intensity in HUVEC cultures exposed to 10 ng to 100 µg/ml TNP-470 for 96 h (Fig. 1a). However, there was no significant effect on apoptosis (Fig. 1b) under the same experimental conditions, indicating that TNP-470 primarily affects cell proliferation in this cell type. The measured BrdU incorporation in SGHPL4 cells decreased, but only following treatment with the maximal dose of 100 µg/ml TNP-470 (Fig. 1c). TNP-470 had no significant effect on the levels of apoptosis in SGHPL4 cells (Fig. 1d). Proliferation assays were also conducted on two cancer cell lines (Ishikawa and MFE-296) and two more primary EC cultures (HULEC and HUAMEC) to ascertain the cell specificity of action of TNP-470. The cancer cell lines showed similar levels of BrdU incorporation, indicating that the cells were not prevented from proliferating (Fig. 1, e and f) despite the much higher doses of 10 µg, 100 µg, and 1 mg of TNP-470 being applied. In response to treatment with TNP-470, the HULEC and HUAMEC showed statistically significant reductions in absorbance levels, indicating decreasing numbers of proliferating cells (Fig. 1, g and h, respectively). In contrast with the HUVECs (which showed a dose-dependent response to TNP-470), HULEC and HUAMEC appeared to show a decreased absorbance at all doses (10 µg to 1 mg). However, the apparent differences between doses were not statistically significant.
|
Maternal and Fetal Growth Characteristics
The viability and numbers of embryos per litter were not compromised by administration of TNP-470 to pregnant dams from E10.5E16.5 (Table 1). At a dose of 30 mg/kg TNP-470, maternal weight gain was significantly decreased at both E17.5 and E18.5 (P < 0.05). When TNP-470 was administered at 3 or 30 mg/kg, several indicators of placental and embryonic size were significantly reduced (Table 1). Plotting data for pup weights as a percentile graph (Fig. 2a) showed that 83% of the 30 mg/kg TNP-470 group fell below the 10th centile for weight of the control group. A total of 10 TNP-470-treated (30 mg/kg) and 10 vehicle-only E18.5 fetuses were randomly sampled and, following careful dissection, the wet weights of brain, heart, and liver were recorded. The livers and hearts were significantly smaller than those in control animals, but the average size of brains was similar in both groups. Alizarin Red and Alcian Blue staining revealed poor ossification of ribs, vertebrae, and digital bones (Fig. 2, bd).
|
|
Stereological and Ultrastructural Analysis of Placentas
Although placentas from TNP-470-treated mice were significantly smaller (Table 1), we detected no significant differences in the volume densities of structural compartments within the placenta (Table 2). The implication is that the absolute volumes of all compartments declined equiproportionally after treatment. Except for the decidual layer, giant trophoblast cells, yolk sac, and chorion, significant changes with gestational age (between E13.5 and E18.5) were found for all volume densities (Table 3). A significant group x age effect on the volume density of all labyrinthine blood vessels was detected (P < 0.05) and this was attributable to a decrease induced by TNP-470 at E13.5. A similar decrease in volume density of Reichert membrane was observed in placentas from TNP-740-treated mice at E13.5.
|
|
Apart from significant age differences in the caliber (mean cross-sectional area) of labyrinthine fetal vessels and the ratio of labyrinthine fetal blood vessel:labyrinthine maternal blood space volumes, two-way ANOVA failed to demonstrate significant treatment, age, or interaction effects accompanying the volumetric changes (Table 3). There were no significant effects of TNP-740 on the absolute lengths of fetal or maternal vessels. However, post hoc testing revealed significant decreases in the calibers and volumes of fetal blood vessels and maternal blood spaces at E13.5 (Table 3).
Quantitation of endothelial cell proliferation within the labyrinth of placentas from mice treated with TNP-470 at a dose of 30 mg/kg revealed an approximately 40% drop at E13.5 and E18.5 in comparison with vehicle-only mice. Transmission electron microscopic studies (data not shown) revealed that fetal blood vessel and maternal blood space architecture, endothelial cell morphology, basement membrane thickness, tight junction morphology, tissue boundaries, and trophoblast ultrastructure were similar in the labyrinthine layers from both control and TNP-470-treated placentas.
Immunohistochemical Detection of TRP53 and CDKN1A in Placentas
Conspicuous staining for TRP53 was noted in the nuclei of giant trophoblast cells and labyrinthine layer fetal blood vessel endothelial cell nuclei in placentas from mice treated with TNP-470 (Fig. 3, e and f , respectively) but was not present in vehicle-only placentas (Fig. 3, c and d). Immunostaining for CDKN1A was noted also in the cytoplasm of giant cell trophoblast and in fetal endothelial cells in the labyrinthine layer of TNP-470-treated animals (Fig. 3, i and j) but not in vehicle-only placentas (Fig. 3, g and h). Experimental controls (primary antibody excluded from the protocol) are also presented in Figure 3.
|
DISCUSSION
In the present study, TNP-470 did not induce endothelial cell apoptosis in vitro, rather, its mechanism of action was mediated via a dose-dependent reduction in cell proliferation, similar to that reported in previous studies [26, 27]. At much higher doses, there was also reduced proliferation of placental cytotrophoblasts, similar to that described for tumor cell lines [26, 27]. The results of these in vitro studies confirmed both the bioactivity of TNP-470 and its endothelium-specific effects at low concentrations. Although the effects of TNP-470 are generally considered to be endothelium-specific, recent studies have shown that it reduces cell growth and proliferation in a variety of cell types, including fibroblasts from mouse embryos and human lung, rat smooth muscle cells [8, 28], and in several tumor cell lines [27, 29, 30]. Thus, not all of the activity of TNP-470 may be related to in vivo effects on endothelium, as discussed previously [31, 32]. This raises the possibility that TNP-470 may have a direct effect on fetal growth in addition to an indirect action via placental insufficiency mediated by endothelial proliferation.
The weights of pregnant dams did not increase in line with the vehicle-only group following injections of TNP-470. Loss of body-weight is a common complication of treatment with therapeutic doses of TNP-470 [33, 34] and, consequently, a proportion of the weight loss in pregnant mice treated with TNP-470 is likely to be attributable to both maternal and fetal weight losses. Weight loss resulting from TNP-470 administration [35, 36] was previously attributed to decreased food uptake, although it can also lead to a reduction in adipose tissue mass (independently of food intake) by inhibiting angiogenesis [36]. Adipose tissue requires a rich vascular supply [37]. Despite this documented side-effect of TNP-470, it is likely that the majority of the maternal weight loss described here was due to decreased weights of placentas and fetuses in utero, as the majority of weight loss (62%) can be accounted for by the combined weight decrease in fetuses and placentas. The reduction in body weight was not due to a loss of embryos, as neither the viability nor the number of fetuses was compromised.
To date, only one other study has investigated the effects of TNP-470 during pregnancy, when mice injected with TNP-470 spontaneously aborted due to lack of implantation (when injected on E1) or placental failure (after injection on E7) [10]. The experimental protocol employed in this study did not reveal an increase in spontaneous abortions and all embryos were alive in utero. Therefore, the effects of TNP-470 on fetal development are clearly specific to the stage of development in which this antiangiogenic agent is administered. Furthermore, our findings support the hypothesis that TNP-470 does not cause fetal abortion or death in utero, provided that TNP-470 is administered after implantation, and that yolk sac and chorion development have occurred [10].
Maternal organ weights in TNP-470-treated dams were similar in size to controls. This finding contrasts with studies showing effects on splenic and renal weight in TNP-470-treated animals [38, 39], but these effects were after prolonged use of the drug. No other effects on maternal condition or behavior were observed and the area around the site of injection did not show any signs of bleeding or tissue damage (data not shown). The most likely mechanism of action of TNP-470 in this model is by a direct effect on the fetoplacental unit. Administration of TNP-470 to pregnant mice led to a smaller placenta. However, there was no significant change in the volume density of vascular compartments. Similarly, xenografted tumors grown in nude mice and subsequently treated with TNP-470 showed evidence of stabilized tumor growth without any change in vessel density [40, 41].
The 10th percentile birth weight for gestational age is commonly used to diagnose growth restriction in human IUGR [42]. In this study, 83% of the fetuses from TNP-470-treated dams had weights that were within or below the 10th centile of the control group. A reduced abdominal circumference is also believed to represent poor weight gain [42] and fetuses from TNP-470-treated dams had significantly lowered abdominal transverse and antero-posterior diameters, in addition to decreased crown-rump lengths. These observations are consistent with maternal administration of TNP-470 inducing a murine model of IUGR. Lack of weight gain in fetuses is of great importance in human obstetric practice as small-for-gestational-age fetuses (commonly termed IUGR) is the primary cause of infant morbidity and mortality [1]. In this study, the fetal weights (at E18.5) in TNP-470-treated groups (3 mg/kg, 30 mg/kg, and 100 mg/kg bodyweight) were significantly reduced in comparison with control animals. Human IUGR has been linked to poor maternal and fetal nutrition [43, 44], infections [45], exposures to toxicants [46, 47], and lack of oxygen [43] and may have long-term consequences, such as adult-onset diseases, including cardiovascular disorders, hypertension, and non-insulin-dependent diabetes mellitus [1416]. Animal models for IUGR include mouse undernourishment [48], superovulation models [49], chronic infection models [50, 51], an ethanol-induced model [52], a spontaneous genetic mouse model [53], and chemical induction model [54]. The present study displays a type of IUGR characterized by reduced proliferation of vascular endothelial cells and smaller mean cross-sectional areas of blood vessels in the placenta during fetal development.
Fractional analyses of E13.5 placentas from 30 mg/kg TNP-470-treated dams revealed smaller volume densities of labyrinthine blood vessels and Reichert membrane than observed in control placentas. The Reichert membrane is thought to be derived from trophoblast cells with endoderm cells and basement membrane material attaching to these cells [55]. The loss, breakdown, or disturbance of Reichert membrane (if caused early enough during gestation) can lead to embryonic lethality by exposing the fetus directly to the maternal circulation or uterine environment [56, 57]. Alternatively, the membrane may play an important role in materno-embryonic exchange of oxygen, nutrients, and waste products [56, 57]. Therefore, it is possible that a 50% reduction in the Reichert membrane at E13.5 could contribute directly to the growth restriction observed in fetuses from dams exposed to TNP-470. Fractional analysis also showed that the total labyrinthine blood vessel volume density (which included all fetal blood vessels and maternal blood spaces) was significantly reduced at E13.5 but not at E18.5. It is interesting to note that, as the labyrinthine blood vessel size was not decreased at E18.5, either TNP-470 only has an effect on early blood vessel formation, or a compensatory mechanism enables angiogenesis to occur at the later stages of development. In mice injected with a single dose of TNP-470 at either E1 or E7, pathological labyrinthine disorganization occurred (in contrast with this study), whereas injection on E14 had no adverse effects on fetal or placental growth [10]. Thus, early interference with vascular development in the murine placenta is incompatible with embryonic survival [43, 58], whereas in those insults that affect mature placental vascular growth, compensatory mechanisms exist to facilitate fetal survival [10]. The spongiotrophoblast layer was not reduced in size at either time point in this study. Placental trophoblast proliferation was not measured in this study, however, as trophoblast growth and differentiation depend on an appropriate vascular supply [10], any reduction in tissue fraction is likely to be linked to the decreased EC proliferation observed following in vivo administration of TNP-470. Other tissue compartments of the placenta demonstrated a uniform reduction after TNP-470, consistent with the hypothesis that endothelial growth and overall placental growth are closely matched.
Analysis of labyrinthine blood vessel/space dimensions in E13.5 placentas from TNP-470-treated mice revealed significant decreases in both fetal blood vessel and maternal blood space cross-sectional areas, but by E18.5, no such differences in vessel calibers were observed. Despite the similar vessel dimensions observed in all groups at E18.5, TNP-470 may still act as an angiogenic inhibitor in the placenta, as similar vascular density does not exclude a partial inhibitory effect on angiogenesis [59]. Significant decreases in the total capillary volumes, surfaces [60], and lengths have also been observed in human IUGR groups [17].
The number of proliferating endothelial cells (in the fetal blood vessels of the labyrinthine layer of the placenta) was higher at E13.5 than at E18.5 in both the control and TNP-470-treated animals. This finding is consistent with other studies, which have shown that proliferation of placental cells decreases throughout gestation [6163]. Following injection of TNP-470, the number of proliferating ECs at E13.5 and E18.5 decreased significantly (by about 40%) in comparison with the control group. Such reductions may influence the ability of the placenta to deliver blood to the developing fetus by affecting placental blood vessel physiology. Reduced proliferation rates have been observed in some cases of human IUGR [64], but not in other studies [63]. TNP-470 decreases the growth of both primary and metastatic lesions in a wide variety of tumors [35, 60], whereas in this study, it affects physiological angiogenesis during development of the placenta and fetus in vivo.
While the effect of TNP-470 on EC proliferation is well established, no difference was observed in the number of cells undergoing apoptosis within TNP-470-treated placentas (data not shown). This is in contrast with studies that have shown that placentas from IUGR-complicated pregnancies have increased levels of apoptosis [65]. The apparent discrepancy may be explained by species-specific differences, as murine trophoblast apoptosis is low, whereas cell turnover in placentas from IUGR or preeclamptic models can be relatively high.
Many pregnancy losses and complications are due to problems in human placental formation and function and the main function of the placenta is to act as an interface for exchange of gases, nutrients, and waste products between the fetal and maternal environments [18]. It has been suggested that abnormal placental vasculature causes inadequate maternal-fetal circulation, leading to intrauterine fetal compromise [66]. This is thought to result in the release of fetoplacental factors that damage the maternal vascular endothelium and then contribute to the various systemic manifestations of preeclampsia [66], the leading cause of maternal morbidity in Western countries [53]. It would useful to examine the blood pressure and proteinuria levels (necessary for a clinical diagnosis of preeclampsia) in this model to determine if these animals recapitulate the clinical correlates.
Although the growth-restricted fetuses were grossly indistinguishable from control fetuses, the question as to their internal development is also of interest. Brain weights were equivalent between in both vehicle-only and TNP-470-treated groups, whereas the liver and heart weights were significantly reduced, providing an animal model of human asymmetric IUGR.
The mechanism of action of TNP-470 is much disputed, but it has been shown to inhibit cyclin-dependent kinase (CDK) activity by upregulating the CDK inhibitor CDKN1A, which in turn is activated by TRP53 [8]. Immunohistochemical staining for CDKN1A and TRP53 was observed in fetal labyrinthine blood vessel ECs and in giant trophoblast cells (GTCs) in TNP-470-treated but not control placentas. GTCs contain up to 1000 times the haploid DNA content of normal cells due to a process termed endoreduplication, a naturally occurring disruption of the mitotic cycle in which rounds of DNA synthesis occur in the absence of mitosis [67, 68]. Research has shown that the cyclins and CDKs play a very important role in the endoreduplication observed in GTCs [69]. GTCs differentiate from proliferative trophoblasts and reduced levels of TRP53 are present as the cell exits the cell cycle and enters a genome-amplifying endocycle (soloveva). TRP53 is known to induce the CDK inhibitor CDKN1A, which in turn suggests that the increased levels of the cyclins and CDKs within the GTCs (consistent with the immunohistochemical results presented in this study) could prevent transition to an endocycle, which may lead to malfunction of the GTCs. Previously, GTCs have been associated with blastocyst implantation, invasion into the uterus, maternal blood vessel endothelial cell displacement [11], and with the release of angiogenic, antiangiogenic, vasoactive compounds, cytokines, and growth hormones [70].
In conclusion, this study supports that notion that TNP-470 directly affects placental development, as the proportion of Reichert membrane is decreased (possibly as a result of its endoderm and trophoblast cell lineage) in the placenta in vivo. The proportion of spongiotrophoblast layer in the E13.5 and E18.5 placentas was unaffected by the drug regime, suggesting that at the highest dose of 30 mg/kg TNP-470 does not affect trophoblast cells in vivo. It has also been shown that injecting TNP-470 every other day from E10.5 to E18.5 results in poor growth of both placentas and fetuses. This murine IUGR is associated with decreased proliferation of fetal vascular ECs and reduced vessel calibers in vivo but had no effect on the number of cells undergoing apoptosis in vitro (in HUVEC or SGHPL4) or in vivo. This study provides a possible mechanism of action for decreased placental dimensions as increased levels of CDKN1A and TRP53 are observed within the placenta, specifically in the GTCs and fetal labyrinthine ECs, which are critical to placental development. We conclude also that injecting 3 or 30 mg/kg doses of TNP-470 every other day from E10.5 to 18.5 results in a highly reproducible murine model of asymmetric intrauterine growth restriction, which offers a potentially useful tool to dissect the pathophysiology of placental insufficiency and test potential treatments.
ACKNOWLEDGMENTS
The authors would like to thank Mr. Trevor Gray and Mr. Philip Hinson (Electron Microscopy Unit, Histopathology, Queens' Medical Center, Nottingham, U.K.) for their assistance with sectioning placentas and for allowing us to use their laboratory equipment. The authors would also like to thank Professor Robert Weiss (UCLA) for testing and providing the immunohistochemistry protocols and Dr. Cliff Murray (The Cancer Research UK Tumor Cytokine Biology Group, University of Nottingham, U.K.) for providing the cell lines used in this publication.
FOOTNOTES
1 Correspondence: Christopher A. Mitchell, Centre for Molecular Biosciences, School of Biomedical Science, University of Ulster at Coleraine, Cromore Road, Coleraine, Co. Londonderry, BT52 1SA, United Kingdom. FAX: 44 0 28 7032 4965; ca.mitchell{at}ulster.ac.uk ![]()
Received: 17 May 2005.
First decision: 9 June 2005.
Accepted: 3 August 2005.
REFERENCES
This article has been cited by other articles:
![]() |
M. M. Veras, N. R. Damaceno-Rodrigues, E. G. Caldini, A. A.C. M. Ribeiro, T. M. Mayhew, P. H.N. Saldiva, and M. Dolhnikoff Particulate Urban Air Pollution Affects the Functional Morphology of Mouse Placenta Biol Reprod, September 1, 2008; 79(3): 578 - 584. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. P. Hewitt, P. J. Mark, and B. J. Waddell Glucocorticoids Prevent the Normal Increase in Placental Vascular Endothelial Growth Factor Expression and Placental Vascularity during Late Pregnancy in the Rat Endocrinology, December 1, 2006; 147(12): 5568 - 5574. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |