| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Medical Research Council Human Reproductive Sciences Unit, Centre for Reproductive Biology, Edinburgh EH3 9ET, United Kingdom
b Molecular Angiogenesis Laboratories, ICRF, Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DS, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-smooth muscle actin, respectively. Vascular endothelial growth factor (VEGF) and its receptors, kinase insert domain-containing region (KDR) and fms-like tyrosine kinase (Flt) mRNA, were localized and quantified in in situ hybridization. In addition, the effects of immunoneutralization of VEGF on establishment and maintenance of pregnancy were investigated by administering a VEGF neutralizing antibody on Days 0–10 of the luteal phase during potentially fertile cycles (n = 10) and compared with fertile controls (n = 6). No differences in the cellular or morphological parameters were found between pregnant and structurally intact nonpregnant corpora lutea. No major differences were found in expression of VEGF, Flt, or KDR in these CL. VEGF immunoneutralization markedly suppressed plasma progesterone secretion during treatment, but pregnancy rate was not significantly reduced. Thus, a role for VEGF in early pregnancy in the marmoset remains to be established. These results show that, by the late luteal phase in the marmoset, the corpus luteum has established a mature vascular system and the molecular capacity to synthesize VEGF and its receptors. A pregnancy-induced spurt of angiogenesis or gene expression does not appear to take place; rather, maintenance of the existing vasculature is all that is required for the establishment of pregnancy.
corpus luteum, corpus luteum function, ovary, ovulation, progesterone
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the corpus luteum of early pregnancy, the changes in angiogenesis and the role of the luteal vasculature in the maintenance of progesterone secretion remain to be elucidated. When pharmacologic "rescue" of the CL by human chorionic gonadotrophin (hCG) treatment in vivo was carried out in rhesus monkeys, no stimulation of angiogenesis, as monitored by Ki-67 incorporation, was observed despite maintaining plasma progesterone concentrations [3]. In apparent contrast, in women similarly treated with hCG, a second wave of angiogenesis, accompanied by vascular stabilization by pericytes and increased VEGF expression, was associated with luteal rescue [7, 8]. These findings suggest that luteal angiogenesis could be a component of luteal rescue, which is important for prolonging the lifespan of the corpus luteum.
Because failure of the CL due to a malfunction of the vasculature might lead to miscarriage or infertility, it was considered important to establish a nonhuman primate model of early pregnancy in which this issue could be addressed. The marmoset monkey offers several potential advantages for such studies. First, it has an exceptionally high rate of fertility (>70%) when compared with other primates, so physiological pregnancy, rather than pharmacologic hCG rescue, may be studied [9]. Second, we have a detailed description of the cellular and molecular regulation of angiogenesis during the normal cycle [4]. Angiogenesis was therefore determined by comparing bromodeoxyuridine (BrdU) incorporation into corpora lutea of the late luteal phase versus early pregnancy. Vascular stability was monitored by determining pericyte area. Localization of expression of VEGF mRNA and its receptors, KDR and Flt, was determined and changes associated with pregnancy quantified using in situ hybridization. Finally, the effect of neutralization of VEGF on luteal function and pregnancy rate in mated marmosets was studied.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experiments were carried out under the Animals (Scientific Procedures) Act (1986) and approved by the Ethical Review Committee. Blood samples were collected from adult female marmosets three times per week and assayed for presence of ovulatory rises in plasma progesterone [10]. During the mid to late luteal phase of the second recorded cycle, prostaglandin (PG) analogue, 1 µg PGF2 (cloprostenol, Planate, Coopers Animal Health Ltd., Crewe, UK) i.m., was administered to induce luteolysis so that timing of the subsequent ovulation could be identified accurately. At the time of PG administration, a fertile male was introduced. Marmosets were humanely killed on Day 21 or 28 postovulation after receiving BrdU as described previously [5]. Of 17 animals mated, 12 were confirmed pregnant by the presence of a trophoblast in serial sections of the uterus and plasma levels of CG of >20 ng/ml. CG was measured using the assay described by Saltzman et al. [11] and had a detection limit of 12 ng/ml. According to these criteria, six animals on Day 21 and six on Day 28 were pregnant. The nonpregnant animals were studied on Day 21 together with three additional animals, recruited by carrying out the above procedure in the absence of a male. Day 21 is the mean time of luteolysis in the marmoset, but this is variable. Of the eight nonpregnant animals, three had structurally regressed CL and the remainder had structurally intact CL. Uteri were fixed in 4% paraformaldehyde (PFA). Ovaries were bisected and fixed, one half of each ovary in 4% paraformaldehyde and the other in 4% neutral buffered formalin (NBF). Experiments were initially carried out on both NBF- and PFA-fixed tissue, but on analysis, no significant difference was found between the two fixatives; thus, only results from NBF-fixed tissues are presented here.
Immunocytochemistry
Cellular responses were studied by 1) measuring cell size in sections stained with hematoxylin and eosin, 2) quantifying the number of mitotic cells stained for BrdU, 3) examining the establishment of the microvascular network using CD31 staining to identify endothelial cells, 4) studying the recruitment of pericytes using immunocytochemical staining of smooth muscle -actin, and 5) dual labeling to record the incidence of colocalization of BrdU and CD31.
Tissue sections (5 µm) were cut onto Super-frost plus slides (Sigma, Poole, Dorset, UK) for immunocytochemistry. Sections were dewaxed in xylene, rehydrated in descending concentrations of ethanol, and washed in distilled water. Antigen retrieval was performed by boiling sections in a Tefal Clypso pressure cooker (Tefal, Essex, UK) in 0.01 M citrate buffer, pH 6, for 6 min at high pressure setting 2. Slides were then left for 20 min in hot buffer and washed in Tris buffered saline (TBS; 0.05 mol/L Tris, pH 7.4, NaCl2 9 g/L). To reduce nonspecific binding, sections were blocked in normal rabbit serum (diluted 1:5 in TBS) for 30 min. The primary antibodies used were CD31 (monoclonal; Dako, Denmark; diluted 1:20 in TBS), BrdU (monoclonal; Boehringer, Mannheim, Germany; diluted 1:30 in TBS), and -smooth muscle actin (monoclonal; Dako, Denmark; diluted 1:20 in TBS). Incubation was carried out overnight at 4°C. Slides were then washed three times in TBS. Incubation with the secondary antibody (rabbit anti-mouse Ig; Dako, 1:60 diluted in TBS) was for 40 min at room temperature. This was followed by two TBS washes and incubation of the APAAP complex (Dako; 1:100 dilution in normal rabbit serum and TBS) for 40 min at room temperature. Visualization was performed using nitroblue tetrazolium (NBT) solution containing 45 µl NBT substrate (Boehringer Mannheim), 10 ml NBT buffer, 35 µl Xphosphate (Boehringer Mannheim), and 10 µl levamisole (Sigma). Sections for BrdU were counterstained with hematoxylin (Richard-Allan, Richland, MI), whereas sections for CD31 and smooth muscle actin were not counterstained so that quantitative image analysis could be performed. For dual labeling, slides were incubated first with CD31 and visualized with fast red (Sigma), followed by incubation with BrdU and visualized with NBT as described above. Fast red solution contained 1 mg fast red/ml buffer (20 mg naphthol AS-MX phosphate, 2 ml dimethyl formamide, 98 ml 0.1 M Tris, pH 8.2; Sigma).
Localization of VEGF, Flt, and KDR mRNA in the Corpus Luteum
To investigate possible changes in expression patterns of the mRNA for angiogenic factors and their receptors in the corpus luteum of pregnancy, in situ hybridization was performed as described previously using complementary RNA probes for human VEGF A, Flt, and KDR [8]. Sense and antisense probes were prepared using an RNA transcription kit (Ambion, Austin, TX) and labeled with 35S uridine 5'-triphosphate (NEN, Boston, MA). Deparaffinized sections were treated with 0.1 N HCl for 20 min and then digested in proteinase K (5 µg/ml; Sigma) for 30 min at 37°C. After prehybridization for 2 h at 55°C, subsequent hybridization was performed in a moist chamber overnight at 55°C. High stringency posthybridization washings and ribonuclease A treatment were used to remove excess probe. Slides were then dehydrated, dried, and dipped in Ilford G5 liquid emulsion (H.A. West, Edinburgh, UK). Exposure times for VEGF, Flt, and KDR were 5, 9 and 9 wk, respectively. Slides were subsequently developed (Kodak D19 developer, Kodak, Rochester, NY) and fixed (Kodak GBS, Kodak). All slides were counterstained with hematoxylin, dehydrated, and mounted.
Quantification
Quantification of lutein cell area Hematoxylin and eosin-stained sections were examined under a 40x objective lens, images captured, and lutein cells identified according to morphologic appearance. Cell perimeter was accurately determined and cross-sectional area measured. Ten lutein cells from each of 10 fields were randomly selected from ovarian cross-sections from each animal.
Quantification of BrdU staining Sections were examined using a 20x objective lens. Image Pro-Plus was used to measure the number of dark-stained nuclei only (those with incorporated BrdU) and the number of total cells per field (dark- and light-stained nuclei). These parameters were measured across the whole cross-section of the CL. To obtain a proliferation index, the number of BrdU-positive cells was expressed as a percentage of total cell nuclei per field of view. Mean values per animal were recorded.
Quantification of CD31 and smooth muscle -actin
Sections were examined using a 20x objective lens and the area of CD31 and smooth muscle -actin staining recorded across the whole corpus luteum. Because immunostaining was localized either to the cytoplasm or plasma membrane, visualization of individual cells was difficult; thus, it was not possible to measure absolute numbers of endothelial cells or pericytes. The image analysis system was therefore set to measure the area of staining within a given field of view. Mean values were calculated per animal.
Quantification of dual labeling (BrdU and CD31) Dual labeling for BrdU and CD31 enabled the percentage of proliferating cells that were endothelial to be determined. Total incidences of dual staining were recorded and expressed as a percentage of total number of proliferating cells throughout the CL. Three animals from each group were assessed and a mean percentage calculated.
Quantification of VEGF, Flt, and KDR in situ Quantification of VEGF, Flt, and KDR mRNA was carried out using an image-analysis macro designed to distinguish silver grains in dark field at 20x objective magnification against the black background of the tissue. Both grain area per field of view (µm2), giving a relative estimate of number of cells per field (3.2 x 105 µm2) and grain density/µm2, representative of amount of expression per cell, were recorded.
Effects of Immunoneutralization of VEGF> on Establishment of Pregnancy
In a second experiment, we sought to investigate the role of luteal VEGF in the establishment and maintenance of early pregnancy by inhibiting its action with a neutralizing antibody in vivo. The experiment was designed so that effects could be monitored by measuring plasma progesterone concentrations during the first third of pregnancy and by recording number of live births. Females used in this experiment had all previously given birth to live young on at least two occasions. Each female had been continually housed with the same male for over a year in stable family groups consisting of the dominant female and male together with their offspring. In these conditions, the marmoset is a highly fecund species, commonly ovulating within 2 wk postpartum and becoming pregnant during the first postpartum cycle [9]. This was confirmed in the animals studied; in the pregnancy prior to treatment, 13 out of 16 (81%) had had an interbirth interval of 150–164 days (normal gestation length being 144 days). These animals therefore represented a potentially highly efficient group in which to study the effect of anti-VEGF treatment on incidence of pregnancy. Animals in which vaginal lavage had shown the presence of sperm on Day 7 or 9 postpartum were treated starting at Day 10 postpartum, i.e., around the time of anticipated ovulation. Ten marmosets received VEGF antibody in the regime described previously [5], i.e., 2 mg i.v. on Day 10, followed by 1 mg on Days 11, 12, 13, 15, 17, and 19, while six controls received the same dose of mouse gamma globulin (Sigma). Blood samples were collected daily from 2 days prior to onset of treatment (Day 0) to treatment Day 10, then three times per week until Day 40, and plasma assayed for progesterone concentration. This treatment had previously been shown to inhibit the development of the luteal microvascular tree and suppress plasma progesterone in normal cycles [5]. In the current study, suppression of progesterone was taken as an index of effectiveness of inhibition of luteal angiogenesis by VEGF antibody. Thus, the end point of the study was whether pregnancy rate was affected. The day of delivery, total number of offspring, and surviving offspring were recorded for each animal.
Statistical Analyses
Separate one-way analysis of variance (ANOVA) tests were conducted for each set of data and a Bonferroni post hoc test performed. P < 0.05 was taken as the level of significance for each test. Tests were performed using SPSS 10 for Macintosh. For determining differences in plasma progesterone between groups, repeated ANOVAs were used to compare study groups on each day of blood sampling.
To compare the number of successful pregnancies after anti-VEGF or mouse gamma globulin treatment, the Fisher exact test was used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Pregnancy was confirmed by the presence of a trophoblast and CG levels greater than 20 ng/ml, while nonpregnant animals had a uterus compatible with an infertile cycle and nondetectable levels of CG. Plasma progesterone profiles of pregnant and nonpregnant marmosets from ovulation (Day 0) to Day 21 are shown in Figure 1. The late luteal nonpregnant animals were subdivided into two groups according to whether functional and structural regression had taken place by Day 21. In five animals, plasma progesterone levels were maintained and CL were structurally intact despite being anticipated to shortly undergo functional and structural regression. The nonregressed state of the CL was not dependent on a male being present, as one of this group included an animal that was housed in the absence of a male and we have commonly observed this phenomenon in other females in our colony. Three of the late luteal control animals had undergone functional and structural regression accompanied by low progesterone levels by Day 21. In these animals, there was a significant decline in plasma progesterone compared with both the pregnant and control animals.
|
Following the initial rise in progesterone indicative of ovulation, progesterone levels remained elevated (>150 nmol/L) in all four groups until Day 10–11 postovulation, which corresponds to the time of implantation in marmosets [12]. After Day 14, the regressed late luteal control animals showed declining progesterone for the remainder of the recorded cycle, whereas a continued elevation of progesterone was observed in both groups of pregnant animals and the structurally intact control animals. Progesterone remained elevated between Days 21 and 28 in the Day 28 pregnant marmosets (data not shown). Statistical analysis revealed that progesterone levels in the regressed control group were significantly different from the structurally intact controls and both groups of pregnant animals (P < 0.05) from Day 19.
Lutein Cell Area
Cross-sectional area of individual lutein cells was not significantly different (P > 0.05) between structurally intact nonpregnant CL and the pregnant animals at Days 21 and 28 postovulation (492 ± 26, 530 ± 21, and 584 ± 34, respectively), while those in the regressed CL showed variable degrees of involution and degenerative change, which precluded meaningful measurements (Fig. 2).
|
BrdU Immunostaining
BrdU immunostaining indicating proliferating cells was readily detected in all corpora lutea. Previously published data [4] from the early, mid and late luteal phase are included in the graph to show the changes in endothelial cell proliferation throughout the luteal phase of the cycle (Fig. 3, A and B). An intense burst of angiogenesis with high endothelial cell proliferation is observed in the early luteal phase, decreasing in the mid luteal phase, and further still toward the end of the luteal phase. No significant difference (P > 0.05) was observed between structurally intact late luteal and pregnant CL. However, significantly higher (P < 0.05) proliferation indices were observed in regressed CL. The insert (Fig. 3a) illustrates the colocalization of CD31 and BrdU immunostaining, i.e., proliferating endothelial cells. Quantification of intact CL showed that more than 90% of proliferating cells were endothelial cells. In structurally regressed CL, it was not possible to determine whether or not these cells were endothelial due to the diffuse nature of the CD31 staining.
|
CD31 Immunostaining
Measurement of CD31 was used to quantify endothelial cell area. There was no significant difference between groups (P > 0.05) (Fig. 3, C and D). By the late luteal phase, the microvascular tree of the corpus luteum is extensive, and from the photomicrograph, it can be seen that every lutein cell is in contact with at least one endothelial cell. Structurally regressed CL were not included in the quantitative analysis because of poor structural integrity, leading to diffuse staining.
Pericyte Immunostaining
Immunostaining for -smooth muscle actin was evident in all stages, predominantly in large luminal vessels but also to a lesser extent in smaller vessels and capillaries (Fig. 3, E and F). No significant difference in endothelium coverage by pericytes (P > 0.05) was seen between study groups. Again, structurally regressed CL were not quantified because of diffuse staining.
Vascular Endothelial Growth Factor
Intense hybridization to VEGF mRNA was localized to lutein cells, with absence of hybridization to endothelial cells. Distinct, punctate hybridization in lutein cells was clearly seen in structurally intact control and pregnant CL (Fig. 4, A and B). There was no significant difference (P > 0.05) in either mean grain area or mean grain density between the groups (Fig. 4C). In the structurally regressed CL, most of the lutein cells were devoid of punctate VEGF expression and there was a marked reduction in grain area and density (Fig. 5, A and B).
|
|
Kinase Insert Domain-Containing Region and Flt
Hybridization to KDR mRNA was localized to endothelial cells (Fig. 6, A and B). No significant difference in grain area or density was observed between groups (P > 0.05) as illustrated by Figure 6E. Structurally intact and regressed nonpregnant CL were analyzed independently but, in contrast with VEGF, KDR expression was maintained in the regressed CL. Because grain area and density were the same in each group, the results were combined.
|
Hybridization to Flt mRNA was localized to endothelial cells (Fig. 6, D and E). A significantly lower mean grain area was observed in Day 28 pregnant corpora lutea than in Day 21 luteal animals (P < 0.05), as illustrated by Figure 6F. However, no significant difference in grain area was seen between late luteal and pregnant Day 21 animals nor was there any observable difference in grain density between groups. As above, late luteal regressed and intact controls were combined because there was no difference in Flt expression between them.
Effects of Immunoneutralization of VEGF on Progesterone and Pregnancy Rate
Plasma progesterone profiles for pregnant control and anti-VEGF-treated marmosets from the time of postpartum ovulation are shown in Figure 7. In control marmosets, progesterone rises indicative of ovulation and establishment of early pregnancy were observed in five of the six animals (83%), with ovulation occurring, as expected, on Days 8–11 postpartum. In the treated animals, all 10 ovulated 9–11 days postpartum, as indicated by a rise in plasma progesterone, but during the treatment period, this level was significantly lower than in control animals (P < 0.05). In five marmosets in this group, the posttreatment period was associated with a variable degree of apparent partial recovery of progesterone, which was followed by functional luteolysis around Day 20–25 postovulation. These animals were classified as the delayed-pregnancy group (because they did not become pregnant during the treatment cycle but conception occurred during the subsequent cycle). In the remaining five treated animals, a more robust recovery of plasma progesterone was evident, followed by a sustained rise, reaching the same levels as attained in the pregnant control animals by Day 20–25. Because gestational length subsequently confirmed that these animals had become pregnant within the treatment cycle, they were thus classified as having no delayed pregnancy. A Fisher exact test to compare number of pregnancies achieved between the 6 control and 10 treated animals revealed no significant difference (P > 0.05). Plasma progesterone in the delayed pregnancy group continued to be significantly lower than controls until Day 27, when levels began to rise again, indicating ovulation, and all these animals became pregnant in the subsequent cycle.
|
Interbirth interval was recorded for all animals. Using the Student t-test, the 1-mo delay in conception observed in 50% of treated animals (188 days ± 1.5 SEM) was significantly different from the interbirth interval seen in controls (157 days ± 0.89 SEM) (P < 0.05). In the treated group in which pregnancy was not delayed, the interbirth interval was not significantly different from controls (159 days ± 1.8 SEM). The number of offspring was also compared for each group. The delayed pregnancy group had a significantly higher number of live births than the control group (P < 0.05) (Fig. 8). No abnormalities were observed in any of the offspring.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the human corpus luteum, the pattern of endothelial cell proliferation is similar to that of the marmoset in the early, mid, and late luteal phase, with maximal proliferation occurring in the early luteal phase and declining in the mid and late luteal phase [2, 13]. This is also true in the rhesus monkey [3], sheep [14, 15], and cattle [1]. During pharmacologically induced luteal rescue in women by hCG, an increase in endothelial cell proliferation was observed [7], accompanied by an increase in VEGF expression [7, 8]. In addition, when changes in luteal cell volume were taken into account, an increase in endothelial cell and pericyte area on luteal sections was found [7]. Taken together, these findings suggest that, in the human CL, hCG treatment results in increased angiogenesis accompanied by vessel stabilization [7]. This appears to mirror the situation observed in rats, where endothelial cell proliferation is also seen to increase in early pregnancy [16, 17]. In contrast, the apparent absence of an increase in endothelial cell proliferation in early pregnancy in the marmoset is similar to findings in the sheep [14] and, after hCG-induced rescue of the CL, in the rhesus monkey [3].
The rescue of the corpus luteum of pregnancy allows it to persist functionally and structurally for longer than in the nonpregnant cycle. Presumably, this survival requires a stable vasculature, with increased recruitment of pericytes and prolonged endothelial cell survival in addition to prolongation of the lifespan of the luteal cells. Because endothelial cell proliferation did not increase after administration of hCG in the rhesus monkey [3], it was suggested that early pregnancy may be associated with an increase in vessel stability, and there is recent evidence for an hCG-induced increase in pericyte number in the human CL [7]. However, we found no evidence for this in the marmoset.
The increase in pericyte and endothelial cell area after hCG in the human was dependent on taking account of the increase in lutein cell volume after hCG administration [7]. However, in the pregnant marmoset, no such increase in lutein cell volume was observed. This is consistent with the findings of Webley et al. [18], who reported that size distribution of luteal cells from pregnant and nonpregnant marmosets revealed an increase in mean cell diameter from luteal Day 6 to 14 in both pregnant and nonpregnant animals but no further increase on Day 20 [18].
While in the structurally regressed CL, VEGF mRNA expression was markedly reduced, there was no apparent difference in VEGF expression between structurally intact late luteal controls and pregnant corpora lutea in the marmoset. VEGF expression is maintained in the CL of pregnancy in a number of species, implying a role in either angiogenesis or endothelial cell survival [8, 19, 20]. A significant increase in VEGF expression, compared with the late luteal phase, was found in hCG rescued human CL using the same method as in the present study [7]. Cell culture studies using human luteinized granulosa cells have also revealed that hCG treatment leads to increased VEGF mRNA levels [21]. This reported up-regulation of VEGF by hCG rescue supports the view that, in the human, pregnancy is associated with further angiogenesis. The absence of up-regulation of VEGF mRNA during early pregnancy in the marmoset is similar to observations in the early pregnant CL of the cow [20]. It is also of interest that there is no evidence for a decline in VEGF mRNA preceding anticipated luteolysis; rather, the decrease of VEGF mRNA was only observed at the time of structural regression.
Levels of KDR mRNA also did not change in the pregnant animals. Flt mRNA was unchanged in the Day 21 pregnant group but significantly decreased in the Day 28 animals compared with the nonpregnant controls. This could be explained by the fact that Flt expression is associated with proliferating endothelial cells, of which there are more in the earlier luteal phases of the cycle. Therefore, by 28 days postovulation, the endothelial cells from these earlier phases may have passed the stage of maximal Flt expression. In the human CL, RT-PCR demonstrated that there was no increase in KDR or Flt mRNA levels in the CL of pregnancy compared with the nonpregnant cycle [22]. Thus, luteal rescue is not associated with an increase in VEGF receptors. These results lead us to propose that the structurally intact late luteal CL of the marmoset retains its molecular integrity and is similar to the rescued CL with regard to VEGF mRNA and its receptors until such time as it does undergo structural regression.
In view of these results, it was considered that a period of intense angiogenesis may be occurring earlier than Day 21, e.g., at around Day 14 of early pregnancy, soon after implantation of the blastocyst on Day 11. However, a pilot study, examining endothelial cell proliferation in the CL from animals at Day 14 of pregnancy, failed to demonstrate increased angiogenesis at this time (unpublished data).
While we have focused on VEGF and its receptors in this study, it is possible that other angiogenic factors may be involved in regulation of the vasculature of early pregnancy. Because the angiopoietins are thought to participate in stabilization of blood vessels, we examined mRNA levels for Angiopoietin-1 and -2 and their receptor, Tie-2, as described for the human CL [7]. However, low expression of the angiopoietins was observed and, as with VEGF, there was no difference between rescued and regressing CL (unpublished data).
The length of the luteal phase in the nonpregnant marmosets varied. This was advantageous here because it allowed for the study of both structurally intact nonpregnant and structurally regressed CL, but the cause of this variation is unclear. Some animals have functionally and structurally regressed CL by Day 21, whereas other nonpregnant animals are still functionally and structurally active although anticipated to shortly undergo regression. Recent estimations from our colony have shown luteal length to vary between 16 and 24 days (unpublished data), which is consistent with previous reports on variability in luteal phase length [23–25]. It is well established that progesterone levels are maintained due to the continued responsiveness of luteal cells to gonadotropins via CG in pregnancy [26]. It would appear that, in the marmoset, despite the absence of CG, the late luteal CL of a nonpregnant cycle is maintained until it undergoes a rapid regression and thus decline in progesterone production. Luteal regression in the marmoset must therefore be a rapidly triggered event because, until that point, the CL of a nonfertile cycle retains its morphology and functional capacity to the same extent as in a pregnant cycle. As reported previously, increased BrdU incorporation was observed in the structurally regressed CL. It was not possible to determine whether or not these cells were endothelial due to the diffuse nature of the CD31 staining in these animals. It may be that these cells are fibroblasts, but this remains to be elucidated [27].
In view of the fact that the period of intense luteal angiogenesis in the fertile cycle of the marmoset is restricted to the early luteal phase, the effects of inhibiting VEGF during this period were examined in animals in stable family groups containing a fertile male. A previous study in marmosets not exposed to pregnancy had shown that such treatment inhibited angiogenesis, prevented formation of the microvascular tree, and suppressed plasma progesterone by more than 60% [5]. The present results confirmed that the anti-VEGF treatment suppressed plasma progesterone in all animals. However, after cessation of treatment, a degree of recovery of progesterone secretion was observed. While in 5 of the 10 treated marmosets, luteal regression subsequently occurred, in the remainder, the corpus luteum function recovered and these animals went on to deliver offspring after a similar gestation period to the controls. The fact that pregnancy was successfully established in the face of a marked reduction in plasma progesterone concentrations suggests that, in the marmoset, the normal quota of luteal progesterone is in excess of requirement. Furthermore, the rise in plasma progesterone posttreatment in half of the animals implies that angiogenesis is capable of reinitiation after cessation of VEGF inhibition. It is of particular interest that, despite systemic administration of the antibody to VEGF, implantation must having taken place during the treatment cycle, gestation length was normal, and no abnormalities were observed in the offspring. Because VEGF is likely to be involved in implantation, it may be that the large molecular weight of the antibody prevented access to the fetal-placental unit.
Where delayed pregnancy occurred, the resulting number of live births in the subsequent cycle was significantly higher than in controls. This may be the result of an overcompensation in VEGF production during the treatment period, resulting in the development of more antral follicles and thus higher numbers of offspring per animal in the following cycle.
This study has demonstrated that, by the late luteal phase in the marmoset, the corpus luteum has already established a mature vascular system and the molecular capacity to synthesize VEGF and its receptors, which are not further enhanced in early pregnancy. A pregnancy-induced spurt of angiogenesis does not take place; rather, a maintenance of the existing vasculature is all that is required for establishment of pregnancy in the marmoset. Therefore, it has to be concluded that the changing morphological and cellular events thought to be associated with early pregnancy in the human after hCG rescue do not occur in the marmoset.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received: 20 November 2001.
First decision: 26 December 2001.
Accepted: 6 May 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. C. Duncan, S. van den Driesche, and H. M. Fraser Inhibition of Vascular Endothelial Growth Factor in the Primate Ovary Up-Regulates Hypoxia-Inducible Factor-1{alpha} in the Follicle and Corpus Luteum Endocrinology, July 1, 2008; 149(7): 3313 - 3320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Silvestri and H. M Fraser Oestrogen and progesterone receptors in the marmoset endometrium: changes during the ovulatory cycle, early pregnancy and after inhibition of vascular endothelial growth factor, GnRH or ovariectomy Reproduction, August 1, 2007; 134(2): 341 - 353. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J Sengupta, P G L Lalitkumar, A R Najwa, D S Charnock-Jones, A L Evans, A M Sharkey, S K Smith, and D Ghosh Immunoneutralization of vascular endothelial growth factor inhibits pregnancy establishment in the rhesus monkey (Macaca mulatta) Reproduction, June 1, 2007; 133(6): 1199 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. M Fraser, H. Wilson, C. Wulff, J. S Rudge, and S. J Wiegand Administration of vascular endothelial growth factor Trap during the 'post-angiogenic' period of the luteal phase causes rapid functional luteolysis and selective endothelial cell death in the marmoset. Reproduction, October 1, 2006; 132(4): 589 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Fraser, H. Wilson, K. D. Morris, I. Swanston, and S. J. Wiegand Vascular Endothelial Growth Factor Trap Suppresses Ovarian Function at All Stages of the Luteal Phase in the Macaque J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5811 - 5818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Nakhuda, R. C. Zimmermann, P. Bohlen, F. Liao, M. V. Sauer, and J. Kitajewski Inhibition of the Vascular Endothelial Cell (VE)-Specific Adhesion Molecule VE-Cadherin Blocks Gonadotropin-Dependent Folliculogenesis and Corpus Luteum Formation and Angiogenesis Endocrinology, March 1, 2005; 146(3): 1053 - 1059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Pauli, H. Tang, J. Wang, P. Bohlen, R. Posser, T. Hartman, M. V. Sauer, J. Kitajewski, and R. C. Zimmermann The Vascular Endothelial Growth Factor (VEGF)/VEGF Receptor 2 Pathway Is Critical for Blood Vessel Survival in Corpora Lutea of Pregnancy in the Rodent Endocrinology, March 1, 2005; 146(3): 1301 - 1311. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ferrara Vascular Endothelial Growth Factor: Basic Science and Clinical Progress Endocr. Rev., August 1, 2004; 25(4): 581 - 611. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J Rowe, C. Wulff, and H. M Fraser Angiogenesis and microvascular development in the marmoset (Callithrix jacchus) endometrium during early pregnancy Reproduction, July 1, 2004; 128(1): 107 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.A. Molskness, R.L. Stouffer, K.A. Burry, M.J. Gorrill, D.M. Lee, and P.E. Patton Circulating levels of free and total vascular endothelial growth factor (VEGF)-A, soluble VEGF receptors-1 and -2, and angiogenin during ovarian stimulation in non-human primates and women Hum. Reprod., April 1, 2004; 19(4): 822 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-y. Liu, Y.-q. Wei, L. Yang, X. Zhao, L. Tian, J.-m. Hou, T. Niu, F. Liu, Y. Jiang, B. Hu, et al. Immunotherapy of tumors with vaccine based on quail homologous vascular endothelial growth factor receptor-2 Blood, September 1, 2003; 102(5): 1815 - 1823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kisliouk, N. Levy, A. Hurwitz, and R. Meidan Presence and Regulation of Endocrine Gland Vascular Endothelial Growth Factor/Prokineticin-1 and Its Receptors in Ovarian Cells J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3700 - 3707. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
—
- - - 
—