| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Animal and Range Sciences, North Dakota State University, Fargo, North Dakota 58105
ABSTRACT
Because of rapid growth followed by spontaneous regression, the ovarian corpus luteum (CL) is an excellent model to study angiogenesis in vivo. To evaluate the expression of vascular endothelial growth factor (VEGF) protein during luteal development, ovaries were collected from FSH-stimulated ewes throughout the estrous cycle. VEGF was immunolocalized in tissue sections by using an affinity-purified antibody. VEGF protein localized exclusively to the thecal layer of preovulatory follicles, while the granulosa was devoid of staining. Associated with the periovulatory period was intense expression of VEGF by thecal cells at the basement membrane and subsequent invasion of the granulosa layers by these VEGF-positive cells immediately after ovulation. The early CL showed staining for VEGF in thecal-derived compartments, and strong staining for VEGF was also seen in cells within the granulosa-derived parenchymal lobules. Dual immunohistochemical localization of VEGF and smooth muscle cell
-actin indicated that the VEGF-positive cells were capillary pericytes or vascular smooth muscle cells. In another experiment, we quantified proliferation of endothelial cells and pericytes throughout luteal development. Pericytes represented a large proportion of the proliferating cells during the early luteal phase and then decreased dramatically. Perivascular cells, therefore, may play a critical role in angiogenesis that occurs during transformation of the follicle into the highly vascular CL of the sheep. As angiogenesis occurs only at the level of capillaries, and pericytes are integral members of these microvessels, regulation of pericytes may provide a novel mechanism for regulating luteal growth and tissue growth in general.
angiogenesis, corpus luteum, follicle, ovary
INTRODUCTION
Ovulation is the critical event that initiates the processes that transform the fluid-filled preovulatory follicle into the solid corpus luteum (CL). The preovulatory follicle is compartmentalized into a highly vascular thecal layer and a nonvascular granulosa layer that are separated by a basement membrane and independently regulated by the two major gonadotropins, LH and FSH, respectively [1]. Definitive structural and functional changes occur in these two compartments around the time of ovulation. Changes occurring after ovulation include dramatic growth and vascularization of the ovulated follicle, transforming it into a mature CL. The periovulatory period thus provides a paradigm, encompassing the follicle to luteal transition, to study the steps involved in the formation of the nascent CL. In this regard, in preliminary studies we previously have observed that vascularization of the recently ovulated follicle by specific components of the thecal vasculature is an initial event during luteinization [2].
Tissue growth of any kind requires the formation of the associated vascular system to supply the increasing metabolic demands of the proliferating cells [3–5]. The rate of growth of the early developing CL is extremely rapid and has been shown to be as high as that of extremely aggressive tumors [6–8]. Consistent with this high rate of growth of the CL is the rapid increase in luteal vascularity. The CL achieves this vascular supply by recruiting new blood vessels from the thecal-derived vascular beds through the process of angiogenesis [9–11].
The vascular development of the follicle and CL are probably controlled by angiogenic factors, including vascular endothelial growth factor (VEGF), which is a potent angiogenic factor that is selectively mitogenic for vascular endothelial cells [5, 12]. VEGF mRNA accompany follicular growth [16]. In addition, luteal expression of VEGF in the early, mid, and late stages of the estrous cycle follows the pattern of luteal vascularization, being greatest during early luteal development and least during regression [17, 18].
To date, there is little information about the expression of VEGF in the ovary, especially in vivo and in situ protein expression. The current studies indicate important species differences in VEGF protein expression that require further investigation. Herein, we focus primarily on characterizing the cell specificity of VEGF protein expression in the sheep throughout luteal growth beginning with ovulation. In addition, cellular proliferation rates in the vascular compartment were determined across luteal development.
MATERIALS AND METHODS
Collection of Periovulatory Follicles
Ewes were stimulated with FSH on Days 13–15 after estrus, as previously described [19, 20]. With this treatment regimen, estrus occurs at approximately 84 h and ovulation at approximately 108 h after the initiation of treatment [20, 21]. This treatment regimen results in the formation of an average of 13 CL that exhibit relatively normal development and function [8, 19, 20].
Ovaries were collected from ewes slaughtered at 60, 84, 108, 132, or 156 h after the first gonadotropin injection (n = 6 ewes/group). These times correspond to approximately -48, -24, 0, 24, and 48 h with respect to the predicted time of ovulation; i.e., the last 48 h of preovulatory follicular development and the first 48 h of luteal development. A sample of jugular venous blood was obtained from each ewe at 24-h (Days 13, 14, and 15) or 6-h (Day 16 and later) intervals, beginning with the first FSH injection, to determine plasma concentrations of LH.
Pieces of ovary containing large preovulatory follicles or developing CL were fixed in Carnoy and formalin solutions, as we have described before [7, 20]. Other pieces containing preovulatory follicles or CL were used for other experiments. A portion of each ovary that did not contain preovulatory follicular or luteal tissue also was fixed in Carnoy to serve as control tissue. Carnoy fixed tissues were used for immunohistochemistry.
Luteinizing Hormone RIA
Luteinizing hormone was measured in 100 µl of plasma by using a previously validated RIA [22]. All samples were run in a single assay, and the intra-assay coefficient of variation was 4.1%.
Collection of CL
Ovaries were obtained from ewes of mixed breeds that had exhibited at least one estrous cycle of normal duration (15–18 days). The ewes were superovulated as described previously [7]. Briefly, FSH was injected i.m. twice daily on Days 14, 15, and 16 after last estrus. Corpora lutea from Days 2, 4, 10, and 15 (n = 8 ewes/day) of the estrous cycle were fixed in Carnoy and formalin solutions and then used for immunohistochemistry.
Immunohistochemistry
In this study, VEGF was immunolocalized to characterize the expression of VEGF in the various follicular and luteal compartments. Lectin-binding, a specific marker of endothelial cells [23], smooth muscle cell -actin (SMCA), a specific marker for vascular smooth muscle and pericytes [24], and 3ß-hydroxysteroid dehydrogenase (3ß-HSD), a specific marker for steroidogenic cells [25] also were immunolocalized in the same tissues to identify positively the location of these cell types and to compare them to cells expressing VEGF. Immunostaining was done as described previously [7] by using a peptide affinity-purified anti-VEGF rabbit serum raised against a VEGF peptide [17]; biotin-labeled lectin (Bandeiraea simplicifolia, BS-1; Sigma, St. Louis, MO), anti-3ß-HSD (monoclonal antibody [mAB]; Oxygene, Dallas, TX) [7], and anti-SMCA (mAB; Boehringer Mannheim, Indianapolis, IN) antibodies to detect VEGF, lectin-binding, 3ß-HSD, and SMCA. The VEGF antibody should detect all isoforms because it was made against a peptide near the N-terminal region of ovine VEGF common to all isoforms. Incubation with specific primary antibodies was followed by biotin-labeled secondary antibody and detected using primarily avidin-peroxidase or avidin-alkaline phosphatase (for Vector Blue only) conjugates and Vector VIP, Vector SG, AEC, DAB, Vector Blue, or Vector NovaRED substrates. Except where noted, all secondary antibodies and reagents used for detection were purchased from Vector Labs (Burlingame, CA). For dual fluorescent staining of VEGF and lectin-binding, VEGF was detected as described above except by using a fluorescein-labeled secondary antibody. Lectin binding was visualized by using biotin-labeled lectin (BS-1; Sigma) followed by Texas red-conjugated streptavidin. SMCA was detected with a biotin-labeled secondary antibody followed by Texas red-conjugated streptavidin. To colocalize VEGF and lectin binding colorimetrically, tissue sections were stained with VEGF as described above. Then the same tissue sections were rinsed with distilled water, washed with blocking buffer, and incubated with biotin-labeled lectin (12.5 µg/ml) overnight at 4°C. The sections were then rinsed in PBS and incubated with the appropriate conjugate and substrate. To colocalize VEGF and 3ß-HSD colorimetrically, tissue sections were stained for VEGF as described above and for 3ß-HSD using dual immunohistochemical procedures previously described by our laboratory [7]. For controls, VEGF antibody was preabsorbed with the peptide it was made against [17]. Controls for 3ß-HSD and SMCA were purified normal mouse IgG. For all controls, staining was essentially nonexistent and is therefore not shown.
Cellular Proliferation in Vascular Elements of the CL
Luteal tissues collected from ewes treated with bromodeoxyuridine (BrdU; a thymidine analog that is incorporated into DNA during the S-phase of the cell cycle) immediately before slaughter were from a previously published study from our laboratory [7]. CL from that study were obtained on Days 2, 4, 8, 12, or 15 (n = 3–6/day) of the estrous cycle. To determine cellular proliferation in vascular elements for the study reported herein, tissues from Jablonka-Shariff et al. [7] were sectioned and immunohistochemically stained for dual localization of BrdU as we previously described [7], and lectin-binding or SMCA as described above. Number of proliferating (BrdU-positive) lectin-binding-positive and SMCA-positive cells was determined by procedures that we have previously published [7]. Briefly, BrdU-labeled nuclei in lectin- and SMCA-stained cells, as well as BrdU-labeled nuclei in all cells, were determined by counting in photomicrographs at a magnification of 400x. For each CL, a minimum of five randomly chosen fields (270 x 180 µm) on each photomicrograph was evaluated in one to four sections (n = 5–20 areas/CL).
Photomicroscopy
All photomicrographs (except fluorescent micrographs) were taken by using a Nikon Microphot microscope and a Nikon Cool-Pix 990 digital camera. All photomicrographs were taken by using either direct interference-contrast, bright-field, or fluorescence microscopy. A Nikon FA 35-mm camera was used for fluorescent micrographs.
Statistical Analysis
Data were analyzed using the general linear models procedure of SAS [26]. For serum LH concentrations, data for each ewe were normalized to the time of maximal LH concentration that was designated the LH peak and assigned Time 0, and the main effect for statistical analysis was hours before or after the LH peak. For cellular proliferation, total number and total proliferating endothelial cells and pericytes as well as proportions of proliferating vascular cells were analyzed for day of the estrous cycle. Differences between specific means were evaluated by using the Bonferroni t test [27].
RESULTS
Estrus and Serum LH Concentrations
For the majority of ewes (25 of 30), an LH peak (7.2 ± 0.2 ng/ml) was observed and occurred at or shortly before estrus (mean day of LH peak = 16.7 ± 0.1; mean day of estrus = 17.0 ± 0.1). As expected, an LH peak was not observed for ewes slaughtered before the predicted time of the LH surge (i.e., -48 h with respect to the predicted time of ovulation), and their serum LH concentrations averaged 1.36 ± 0.46 ng/ml.
Immunohistochemistry
Follicles Lectin binding was found exclusively in endothelial cells, and in follicles it was restricted to the blood vessels in the theca layer of the preovulatory follicle. Well-defined lectin binding in the theca was usually observed in the vascular wreath immediately surrounding the basement membrane (Fig. 1, preovulatory). Near ovulation, cells capable of binding lectin were seen in the junction between the thecal and granulosa layers (Fig. 1, postovulatory) and then appeared in the granulosa layer during the immediate postovulatory period, a time of rapid vascularization. In all of the preovulatory follicles examined, VEGF protein was localized exclusively to the theca, while the granulosa was devoid of staining (Fig. 1). During the preovulatory period, expression of VEGF was observed primarily in the theca externa and subsequently in thecal-derived cells that appeared to be invading the granulosa layer during the immediate postovulatory period (Fig. 1). As with VEGF, SMCA staining in ovine preovulatory follicles was localized in thecal tissue and not granulosa and primarily in cells associated with the vasculature; however, staining was consistently more abundant than VEGF and found in both theca interna and externa (Fig. 1). Also, as with VEGF, SMCA was localized in thecal-derived cells that appeared to be invading the granulosa layer during the immediate postovulatory period. Serial sections of a preovulatory and postovulatory follicle stained separately for VEGF, SMCA and lectin binding demonstrate the similarity in staining patterns for both VEGF and SMCA, and the distinctly different staining pattern of lectin binding (Fig. 1). Dual localization of VEGF and lectin binding during the immediate postovulatory period clearly demonstrates that VEGF-positive cells are among the first cells found in or adjacent to the antrum. In addition, dual localization of VEGF and lectin binding showed an abundance of both lectin-positive and VEGF-positive cells in the granulosa layer and branching from pre-existing capillaries in the thecal layer (Fig. 2) of postovulatory follicles. Dual SMCA and VEGF immunofluorescence on the same tissue sections revealed that VEGF-positive cells were always positive for SMCA, indicating that they are indeed perivascular cells that include vascular smooth muscle cells and pericytes (not shown for follicles but shown for CL below).
|
|
Corpora lutea The CL showed staining for VEGF in the theca-derived compartments, namely cells associated with the vasculature of connective tissue tracts between the developing parenchymal lobules early in luteal development and in some of the cells within the parenchyma throughout the early and mid luteal phase (Fig. 3). VEGF staining appeared to be associated primarily with the developing microvasculature (Fig. 4), and this pattern of VEGF localization persisted throughout early luteal development. There was a marked disappearance of VEGF-positive cells during the mid and late luteal phase where VEGF staining was confined to the perivascular areas of large blood vessels (Fig. 3) and the luteal capsule.
|
|
Simultaneous detection of VEGF and lectin binding on the same tissue section (Figs. 3 and 4) demonstrated that VEGF localized to specific cell types distinctly different from the endothelial cells. Overall, VEGF and lectin binding did not colocalize but stained different cell populations in the CL. In addition, VEGF-positive cells were almost exclusively located immediately adjacent to an endothelial cell that bound lectin (i.e., perivascular). The area and intensity of staining for lectin binding in the CL appeared to be constant throughout the estrous cycle except that intensity of lectin binding waned in regressing late-cycle CL (Fig. 3).
The 3ß-HSD-positive cells were distinct from the VEGF-positive cells (Fig. 5). Again, the VEGF-positive cells labeled perivascular cells as described above, whereas the 3ß-HSD antibody labeled exclusively the cytoplasm of the steroid-producing large and small luteal cells that are present in the parenchyma of the CL. The intensity of 3ß-HSD staining was highest during the mid-luteal stage and lowest during the late stage of the estrous cycle.
|
SMCA staining in the ovine CL was remarkably similar to the staining for VEGF in the early CL (compare Fig. 6 to Fig. 3), being present in capsule, connective tissue tracts, nonsteroidogenic cells within the parenchyma, and the perivascular areas of small and large blood vessels. However, while VEGF staining from early- to the mid- and late-stage CL changed dramatically (Fig. 3), SMCA staining remained fairly abundant and constant in these tissue compartments (Fig. 6).
|
Dual fluorescent localization of VEGF and lectin binding (Fig. 7) clearly demonstrates that VEGF is localized in other cells distinctly different from endothelial cells, and also as shown in Figures 3 and 4, that VEGF staining is perivascular. Dual immunofluorescent localization of VEGF and SMCA on the same tissue section revealed that all VEGF-positive cells were also SMCA positive (Fig. 8), as suggested previously in Figures 1 and 2 for follicles and by comparing staining in Figures 4 and 6 for CL. In the early developing CL a majority of SMCA-positive cells also expressed VEGF. By mid-cycle, while the number of SMCA-positive cells remained high, the fraction that expressed VEGF was greatly reduced and, by late cycle, was totally lost by most cells in the parenchymal areas (fluorescent micrographs not shown, but the same data can be derived from comparing VEGF and SMCA staining throughout the luteal phase in Figs. 3 and 6).
|
|
Proliferation of Luteal Endothelial Cells and Pericytes
Dual localization of BrdU and lectin binding or BrdU and SMCA is shown in Figures 9 and 10. Proliferating endothelial (BrdU- and lectin-positive) cells and proliferating perivascular (BrdU- and SMCA-positive) cells were readily distinguishable as were the remainder of the proliferating cells and thus were easily quantified by morphometry. Table 1 shows the total number of luteal cells as well as the number of endothelial cells and pericytes throughout the luteal phase of the estrous cycle. The number of luteal endothelial cells and pericytes increased (P < 0.02) from Day 2 through Day 8 after estrus. The number of luteal endothelial cells then remained constant through Day 15. In contrast, the number of luteal pericytes declined (P < 0.02) by 35% between Day 8 and Day 15. Together, the endothelial cells and pericytes comprised 44% of the cells in the CL on Day 8 after estrus. By Days 12 and 15, however, the endothelial cells and pericytes represented only 16%–26% of the total number of luteal cells. In terms of proliferating luteal cells, 22% of the total number of proliferating cells during the early luteal phase (i.e., Day 2 after estrus) was represented by pericytes (Table 1). The number of proliferating pericytes as well as the proportion of total proliferating cells represented by pericytes both decreased dramatically (P < 0.01) after Day 2. Interestingly, because the number of proliferating endothelial cells remained constant throughout the estrous cycle, they represented 74% of the total proliferating luteal cells by Day 15 (Table 1). As shown in Table 1, labeling index (LI, proliferating cells expressed as proportion of total number of cells) was extremely high for pericytes on Day 2 and high for both endothelial and pericytes on Day 4. The LI dropped dramatically by Day 12 and stayed low through Day 15 for both endothelial cells and pericytes.
|
|
DISCUSSION
Early in luteal development, shortly after ovulation, the thecal vasculature begins to invade the growing granulosa layer [5]. As a preexisting vascular bed is essential for angiogenesis, the thecal vasculature becomes the repository of vascular elements that the developing CL will draw upon to neovascularize the presumptive parenchymal tissue. The impetus for thecal blood vessel invasion of the granulosa layer is poorly understood, but the mechanism by which the CL achieves its high degree of vascularity may be speculated from the data collected in these experiments. Angiogenesis consists of at least three steps: 1) breakdown of the basement membrane of the existing blood vessel, 2) migration of the endothelial cells toward an angiogenic stimulus, and 3) proliferation of the endothelial cells in the new place to establish a new blood vessel sprout [28]. The key step in understanding the entire process of angiogenesis is incumbent upon discerning growth factors that can fulfill all these functional criteria.
A number of studies indicates that one such multipotent growth factor is VEGF. Roberts and Palade [29] have shown, using the rat, that topical or intradermal application of VEGF increases permeability of capillaries and postcapillary venules and induces formation of fenestrations in venular and capillary endothelia that are normally not fenestrated. We have previously shown, using migration/mitogenesis assays [5], that luteal explant conditioned media possess both migration-stimulating and mitogenic activity for endothelial cells and that the majority of this activity could be neutralized by using antibodies to VEGF [12]. VEGF is not only a potent migration-stimulating factor and can thus attract the endothelial cells from blood vessels but is also a mitogenic factor that can induce the endothelial cells to proliferate subsequent to migration. The early, middle, and late stages of luteal development correspond to periods of rapid growth and vascularization, maturity and differentiated function, and regression and loss of differentiated function, respectively. Cell-specific expression of VEGF protein in this study correlates with the changing physiological state of the CL and associated changes in vascularity.
Dual immunohistochemical localization of VEGF with cell-specific markers revealed that endothelial or steroidogenic luteal cells did not express VEGF. Because VEGF-containing cells appeared perivascular (larger microvessels) and were present only in single cells (smaller microvessels and capillaries), we tentatively identified these VEGF-positive cells as vascular pericytes based on their location and morphology.
Pericytes are perivascular cells that are typically associated with endothelial cells in microvessels; they are similar to vascular smooth muscle cells (VSMC) and have long been implicated as precursors of VSMC [30]. Both cell types express SMCA, a specific marker for cells of VSMC lineage [31, 32]. However, other than identification based on VSMC markers, pericytes can be identified primarily by location, being associated almost exclusively with capillaries and postcapillary venules, and secondarily by structure, as they have a cell body that is ensheathed by the capillary basement membrane, protrude above the capillary wall, and are generally single cells distributed along capillary vessels [33]. Vascular smooth muscle cells on the other hand are associated with arterioles and muscular venules and are generally multilayered and spindle-shaped. By using dual immunofluorescence procedures, we have shown that in the follicle and CL, VEGF-positive cells are always positive for SMCA. Thus, depending on their location, VEGF-positive cells appear to be pericytes when present in the granulosa layer during the immediate postovulatory period or in luteal parenchyma while those present in the larger blood vessels are probably VSMC.
Pericytes are regulators of endothelial function and have been shown to produce angiogenic factors in vitro under hypoxic conditions [34]. Numerous studies previously have shown that VSMC produce VEGF [35–41]; however, only two reports have shown that cells, unsubstantiated as pericytes, produce or express VEGF in vivo [38, 39]. To the best of our knowledge, this is one of the seminal reports to show conclusively that VEGF is produced by pericytes in vivo, and furthermore, that VEGF is localized exclusively to VSMC and pericytes. In addition, this report demonstrates clearly that luteal pericytes and luteal VSMC express VEGF protein in vivo and that expression of VEGF in luteal tissue is limited primarily to these cell types. This is in agreement with a recent study by Berisha et al. [18], where vascular smooth muscle of larger microvessels of bovine CL were also shown to express VEGF protein, especially in late-cycle CL. However, in contrast to the present study, these investigators showed that steroidogenic bovine luteal cells were also shown to express VEGF, which again suggests that there may be substantial species differences in the mechanisms surrounding new blood vessel growth. Interestingly, pericytes are probably better well known for releasing angiostatic factors such as transforming growth factor-ß under normoxic conditions [42].
The granulosa-derived parenchymal lobules of the early developing CL exhibit all of the characteristics of hypoxia, as this compartment is rapidly growing and there are no blood vessels supplying it before or immediately after ovulation [1]. As shown in the present study, pericytes appear to be among the first cells to migrate into the hypoxic granulosa layer after ovulation and produce VEGF, which subsequently may initiate angiogenesis. VEGF-positive cells derived from the well-vascularized theca of the preovulatory follicle were seen invading the avascular granulosa-derived regions of the early CL within a few hours after ovulation, and this process appeared to precede endothelial invasion of the granulosa-derived regions. We definitively identified these VEGF-containing cells as pericytes based on positive SMCA staining. This supports previous findings that pericytes serve as guiding structures, aiding outgrowth of endothelial cells from existing vessels [33]. In contrast, Goede et al. [43] suggested that endothelial cells lead the invasion of pericytes; however, this was documented subjectively in early CL rather than in recently ovulated follicles. A possibility exists that there are substantial species differences in the mechanisms of invasion that may account for the different observations. Nonetheless, this report presents a novel finding that pericytes are among the first cells to invade the luteinizing granulosa layer immediately after ovulation, while at the same time demonstrating that these cells produce VEGF, strongly linking these cells to a major role in the angiogenic process. For sheep, it is tempting to speculate that as a follicle matures and approaches ovulation, the granulosa cells produce a factor that stimulates the invasion of pericytes into the granulosa cell layer immediately after ovulation. The presence of pericytes and their production of VEGF would likely stimulate the migration and proliferation of endothelial cells and the formation of new capillary vessels.
In follicles from the present study, VEGF protein localized to the theca layer, the granulosa layer being devoid of VEGF. No VEGF- or SMCA-positive cells were ever observed in the granulosa layer of large follicles before the periovulatory period. This pattern of staining agreed with similar observations in human preovulatory follicles by several investigators [16, 38, 44]. However, others have shown VEGF expression by both theca and granulosa layers [14, 15, 45, 46], a number of in vitro studies have shown VEGF expression by granulosa cells [47–49]. These variations in VEGF expression could be due to the different species involved and also due to the culture conditions used in these studies, as oxygen tension strongly regulates VEGF expression [50].
From results of this study, we suggest that at least in sheep, pericytes begin to produce and secrete VEGF, which in turn has been shown to increase vascular permeability. VEGF also serves as a chemoattractant and mitogen for endothelial cells, two well-known components of the angiogenic process. Recently, it has been shown that VEGF is requisite for luteal formation and vascular development [11, 51]. As the vascular demand increases in association with continued luteal growth, pericytes may further differentiate into VSMC of the larger luteal blood vessels [24]. Given this model and the fact that pericytes are present in all tissues, we therefore postulate that in sheep pericytes are the primary cell types initiating microvascular growth and/or maintenance through their production of VEGF.
The expression of VEGF in the early developing CL, which is a site of rapid tissue growth, is consistent with the proposed role for VEGF in neovascularization. It is interesting to note from the present study that of all the proliferating cells during early luteal development, pericytes represented a large portion (nearly one-fourth of the total proliferating cells) and agrees with the notion that if pericytes play a leading role in the angiogenic process, they must be available in significant numbers early in the process. The persistence of VEGF expression after vascularization is completed has been interpreted to suggest that VEGF is important not only for vascular development and organization but also for vascular maintenance [40, 52]. The progressive association of pericytes with capillaries has been identified with the maturation of capillary vessels and marks the end of a plasticity window of vascular remodeling [53]. With this concept in mind, pericyte production of VEGF then likely contributes to vascular maintenance in mature vessels. In the present study, VEGF in the mature mid-cycle CL was localized primarily to the larger blood vessels but was conspicuously absent from the connective tissue tracts, and when present in the parenchyma was of lesser intensity and abundance than in the developing CL. This is different from SMCA staining (i.e., VSCM and pericytes) that was present in these luteal tissue compartments at fairly abundant levels throughout the estrous cycle. Based on these observations, it appears that pericytes remain abundant in the luteal connective tissue and parenchymal compartments throughout the life span of the CL even though VEGF-expressing cells disappear from these compartments. This leads us to believe that although VSMC and pericytes are still present at mid-cycle, many of them are no longer producing VEGF. This might be expected because by mid-cycle the CL reaches its maximum size, functionality, and vascular supply, and thus no longer requires angiogenesis. Hence, VEGF expression is lost in much of the connective tissue tracts and parenchymal compartments of the mature CL.
The late-stage regressing CL showed a dramatic loss of VEGF expression in all compartments except the large blood vessels. These data correspond exactly to VEGF mRNA expression that we have previously reported in ovine CL, where VEGF expression is high early, but then drops off significantly by mid-cycle [17]. Again, SMCA staining remained relatively constant in these luteal tissue compartments across all stages. Continued presence of VEGF in predominantly the larger vessels of late-cycle CL is consistent with previous studies in our laboratory and as reported for the bovine CL [18], showing that while a portion of the capillary bed regresses concomitant with luteal parenchymal regression, many of the larger microvessels are maintained presumably to aid in rapid resorption of the luteal tissue [5]. In agreement with this observation, the proportion of pericytes that are proliferating in the late-cycle CL is drastically reduced from the early cycle. Yet, the proportion of endothelial cells that are proliferating is nearly three-fourths of the total number of proliferating cells that would seemingly be suggestive of vascular maintenance rather than new blood vessel growth requiring the migration and presence of VEGF-secreting pericytes.
Taken together, these observations suggest a novel mechanism for tissue vascularization and support the hypothesis that mesodermally derived perivascular or connective tissue cells may direct vascular development and maintenance in the CL through production of VEGF.
|
ACKNOWLEDGMENTS
We thank Jim D. Kirsch and Kim C. Kraft for their expert technical assistance.
FOOTNOTES
1 This work was supported in part by USDA Competitive Grants 96-35203-3269 and 98-35203-6222 and Hatch Project ND01705 to D.A.R. and L.P.R. A portion of the work presented herein was part of V.D.'s dissertation research. 
2 Correspondence: Dale A. Redmer, Dept. Animal and Range Science, 187
Hultz Hall, North Dakota State University, Fargo, ND 58105-5727. FAX: 701 231 7590; dale_redmer{at}ndsu.nodak.edu
3 Current address: Vinayak Doraiswamy, Promega Corp., 2800 Woods Hollow
Rd., Madison, WI 53711.
4 Current address: Albina Jablonka-Shariff, Department of Molecular Biololgy
and Pharmacology, Washington University, School of Medicine, St.
Louis, MO 63110.
Accepted: April 25, 2001.
Received: February 5, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  F. Kuhnert, B. Y. Y. Tam, B. Sennino, J. T. Gray, J. Yuan, A. Jocson, N. R. Nayak, R. C. Mulligan, D. M. McDonald, and C. J. Kuo Soluble receptor-mediated selective inhibition of VEGFR and PDGFR{beta} signaling during physiologic and tumor angiogenesis PNAS, July 22, 2008; 105(29): 10185 - 10190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Tigges, E. G. Hyer, J. Scharf, and W. B. Stallcup FGF2-dependent neovascularization of subcutaneous Matrigel plugs is initiated by bone marrow-derived pericytes and macrophages Development, February 1, 2008; 135(3): 523 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Grazul-Bilska, C. Navanukraw, M. L. Johnson, K. A. Vonnahme, S. P. Ford, L. P. Reynolds, and D. A. Redmer Vascularity and expression of angiogenic factors in bovine dominant follicles of the first follicular wave J Anim Sci, August 1, 2007; 85(8): 1914 - 1922. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kisliouk, A. Friedman, E. Klipper, Q.-Y. Zhou, D. Schams, N. Alfaidy, and R. Meidan Expression Pattern of Prokineticin 1 and Its Receptors in Bovine Ovaries During the Estrous Cycle: Involvement in Corpus Luteum Regression and Follicular Atresia Biol Reprod, May 1, 2007; 76(5): 749 - 758. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. S. Sleer and C. C. Taylor Platelet-Derived Growth Factors and Receptors in the Rat Corpus Luteum: Localization and Identification of an Effect on Luteogenesis Biol Reprod, March 1, 2007; 76(3): 391 - 400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T Grazul-Bilska, C. Navanukraw, M. L. Johnson, D. A Arnold, L. P Reynolds, and D. A Redmer Expression of endothelial nitric oxide synthase in the ovine ovary throughout the estrous cycle. Reproduction, October 1, 2006; 132(4): 579 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A Vonnahme, D. A Redmer, E. Borowczyk, J. J Bilski, J. S Luther, M. L. Johnson, L. P Reynolds, and A. T Grazul-Bilska Vascular composition, apoptosis, and expression of angiogenic factors in the corpus luteum during prostaglandin F2{alpha}-induced regression in sheep. Reproduction, June 1, 2006; 131(6): 1115 - 1126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Martelli, P Berardinelli, V Russo, A Mauro, N Bernabo, L Gioia, M Mattioli, and B Barboni Spatio-temporal analysis of vascular endothelial growth factor expression and blood vessel remodelling in pig ovarian follicles during the periovulatory period J. Mol. Endocrinol., February 1, 2006; 36(1): 107 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Tesone, R. L. Stouffer, S. M. Borman, J. D. Hennebold, and T. A. Molskness Vascular Endothelial Growth Factor (VEGF) Production by the Monkey Corpus Luteum During the Menstrual Cycle: Isoform-Selective Messenger RNA Expression In Vivo and Hypoxia-Regulated Protein Secretion In Vitro Biol Reprod, November 1, 2005; 73(5): 927 - 934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Ozerdem, K. Alitalo, P. Salven, and A. Li Contribution of Bone Marrow-Derived Pericyte Precursor Cells to Corneal Vasculogenesis Invest. Ophthalmol. Vis. Sci., October 1, 2005; 46(10): 3502 - 3506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kisliouk, H. Podlovni, K. Spanel-Borowski, O. Ovadia, Q.-Y. Zhou, and R. Meidan Prokineticins (Endocrine Gland-Derived Vascular Endothelial Growth Factor and BV8) in the Bovine Ovary: Expression and Role as Mitogens and Survival Factors for Corpus Luteum-Derived Endothelial Cells Endocrinology, September 1, 2005; 146(9): 3950 - 3958. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Rovas, P. Sladkevicius, E. Strobel, and L. Valentin Three-Dimensional Power Doppler Ultrasound Assessment of the Cervix for the Prediction of Successful Induction of Labor With Prostaglandin in Prolonged Pregnancy J. Ultrasound Med., July 1, 2005; 24(7): 933 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Haimov-Kochman, D. Prus, E. Zcharia, D. S. Goldman-Wohl, S. Natanson-Yaron, C. Greenfield, E. Y. Anteby, R. Reich, J. Orly, A. Tsafriri, et al. Spatiotemporal Expression of Heparanase During Human and Rodent Ovarian Folliculogenesis Biol Reprod, July 1, 2005; 73(1): 20 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Zhang, D. Chaturvedi, L. Jaggar, D. Magnuson, J. M. Lee, and T. B. Patel Regulation of Vascular Smooth Muscle Cell Proliferation and Migration by Human Sprouty 2 Arterioscler. Thromb. Vasc. Biol., March 1, 2005; 25(3): 533 - 538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P D Taylor, S G Hillier, and H M Fraser Effects of GnRH antagonist treatment on follicular development and angiogenesis in the primate ovary J. Endocrinol., October 1, 2004; 183(1): 1 - 17. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. Clark, H. F. Irving-Rodgers, A. M. Dharmarajan, and R. J. Rodgers Theca Interna: The Other Side of Bovine Follicular Atresia Biol Reprod, October 1, 2004; 71(4): 1071 - 1078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-i. Fukushi, I. T. Makagiansar, and W. B. Stallcup NG2 Proteoglycan Promotes Endothelial Cell Motility and Angiogenesis via Engagement of Galectin-3 and {alpha}3{beta}1 Integrin Mol. Biol. Cell, August 1, 2004; 15(8): 3580 - 3590. [Abstract] [Full Text] |
