| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Departments of Cell Biology, Physiology and Immunology,
b Pathology, Faculty of Medicine, University of Córdoba, 14004 Córdoba, Spain
ABSTRACT
We studied the morphological and quantitative changes in cyclic corpora lutea (CCL) and in CL of pregnancy (CLP) during structural luteolysis. Elimination of CCL takes several cycles, and cell death occurs as successive apoptotic bursts, from 2100 h in proestrus to 1300 h in estrus. Each apoptotic burst determined a 60% decrease in the CL volume and an 80% decrease in the number of steroidogenic cells (SC). All these changes were inhibited by blocking the preovulatory prolactin (PRL) surge with bromocryptine (CB154). Neither apoptotic cells nor changes in the number of SC were found in regressing CLP from Day 21 of pregnancy to Day 2 postpartum, although there was a 50% decrease in the CLP volume and a 30% decrease in the mean cross-sectional area of SC. Treatment with CB154 on the day of parturition did not modify these regressive changes. On Day 5 postpartum, the volume of the CLP and the number of SC were equivalent in lactating rats (showing high PRL concentrations induced by pup suckling) and nonlactating noncycling rats (in which cyclicity and, therefore PRL surges, were blocked by treatment with LHRH antagonist). However, on Day 10 postpartum, the CLP volume and the number of SC were significantly decreased in lactating rats, and apoptotic cells were frequent. In postpartum cycling rats, the CLP did not show apoptotic cells on the day of the second postpartum estrus (on Day 5 postpartum), whereas on the day of the third postpartum estrus (on Day 9 postpartum), apoptotic cells were abundant. These results indicate that PRL does not induce apoptosis in the CLP before Day 5 postpartum and strongly suggest that the proapoptotic effect of PRL is dependent on the degree of differentiation of luteal cells.
apoptosis, corpus luteum, lactation, ovary, prolactin
INTRODUCTION
Prolactin (PRL) is the main luteotrophic hormone in the rat [1]. However, when the corpus luteum (CL) ceases of producing progesterone, PRL induces structural luteolysis [1], which is responsible for the elimination of luteal cells in the nonfunctional CL. In cycling rats, the first morphological signs of structural luteolysis are associated with the following ovulation during the transition from proestrus to estrus [2–6], although complete elimination of the CL takes several cycles. The preovulatory PRL surge on the evening of proestrus induces luteolysis in either the CL of the current or previous cycles [2]. Structural luteolysis can be also induced in hypophysectomized [7–11] or cycling [2, 4, 6] rats by exogenous PRL administration. Conversely, luteolytic changes can be inhibited by blocking the preovulatory PRL surge with the dopaminergic agonist CB154 [2, 4, 12].
Although it has been demonstrated conclusively that PRL is responsible for the decrease in luteal tissue weight during luteolysis, the mechanisms by which PRL exerts this effect are not well known. It is widely assumed that luteal cell death occurs through apoptosis, and several studies have explored the role of PRL in triggering programmed cell death [2, 4, 6], as well as the expression of apoptosis-related genes [13–15] or enzymes [16, 17] involved in the apoptotic pathway. However, whether the luteolytic effects of PRL are mainly due to the induction of apoptosis or to other pathways and to what extent apoptosis is responsible for the decrease in luteal tissue mass is not clear, and literature data are rather controversial.
For instance, highly variable amounts of apoptosis have been reported during CL regression. In cycling rats, we have previously found [4] that the preovulatory PRL surge on the evening of proestrus induces a large (100-fold) increase in the number of apoptotic cells on the morning of estrus. In contrast, other authors [5, 6] reported considerably lower (three- to fourfold) increases in the number of apoptotic cells during a similar period. In studies focused on the regression of the CL of pregnancy (CLP) [14], it has been reported that more than 50% of luteal cells undergo apoptosis on Day 3 postpartum. Nevertheless, such a high rate of apoptosis would eliminate the totality of luteal cells shortly after parturition, which is in contrast to previous studies [18] indicating the presence of regressing CLP at advanced stages of lactancy, several weeks after parturition.
In addition, although it has been clearly shown that PRL induces cell death in the regressing CL, the dependence of luteal cell apoptosis on the preovulatory PRL surge is also unclear. In some studies, apoptosis was completely inhibited by blocking the preovulatory PRL surge [2, 4]. This suggests that PRL is absolutely necessary for the induction of apoptosis, although additional factors could also be needed [4, 19]. On the contrary, a recent study [5] reported that blocking the PRL surge decreased apoptosis by only 30% and concluded that much of the increase in apoptosis from proestrus to estrus was independent of the preovulatory PRL surge, and that the loss of luteal cell mass during the same period may be due to other PRL-mediated pathways.
These discrepancies could be due, at least in part, to the use of different experimental models involving different endocrine environments (i.e., cycling versus hypophysectomized rats), different types of CL (i.e., cyclic CL versus CLP), different criteria to identify dying cells (i.e., morphological evaluation versus in situ-labeling techniques), or the study of different time points during the estrous cycle (i.e., early versus late estrus).
The timing of the occurrence of apoptosis, the impact of programmed cell death on luteolysis, and the possible existence of differences in the response to the proapoptotic effect of PRL in different types of CL are crucial for studies focused on the mechanisms underlying PRL-induced CL regression. In this work, we have studied the luteolytic effects of PRL in two different types of CL (cyclic CL and CLP), involving a different degree of luteal cell differentiation. Luteal cell death was evaluated by two different approaches. First, by analyzing the amount of apoptosis and second, by quantifying changes in the number of steroidogenic luteal cells during luteolysis.
MATERIALS AND METHODS
Animals and Drugs
Female Wistar rats (about 200 g body weight) were purchased from Panlab (Barcelona, Spain), maintained under controlled light (14L:10D, lights-on at 0500 h) and temperature (21–22°C) and had free access to rat chow and tap water. Vaginal smears were taken daily; only animals displaying at least two consecutive 4-day cycles were used. All experiments were carried out in accordance to the Guide for the Care and Use of Laboratory Animals and were approved by the Ethical Committee of the University of Córdoba.
The dopaminergic agonist 2-Br-
-ergocryptine (CB154) that specifically inhibits PRL secretion was purchased from Sandoz (Basel, Switzerland). The LHRH antagonist (LHRHa) used was ORG.30276 (Ac-D-p-Cl-Phe-D-p-Cl-Phe-D-Trp-Ser-Tyr-D-Arg-Leu-Arg-Pro-D-Ala-NH2·CH3·COOH) (Organon, Oss, The Netherlands).
Experimental Designs
Experiment 1 This experiment was conducted to analyze the timing of apoptosis in regressing CL of the estrous cycle, during the transition from proestrus to estrus, in relation to the preovulatory PRL surge. In our colony, PRL concentrations are maximal at 1830 h on the evening of proestrus [20]. Three intact cycling rats per time point were killed on proestrus (at 1000 h, 1600 h, and 2100 h) or estrus (at 1000 h, 1300 h, and 1800 h). The right ovaries were fixed in Bouin-Hollande fluid for 24 h and processed for paraffin embedding. Five-micron-thick sections were cut and stained with hematoxylin and eosin. Two left ovaries per time point were fixed in 4% paraformaldehyde in Sorensen buffer (pH 7.2) for 24 h and processed for paraffin embedding. Five-micron-thick sections were placed in poly-L-lysine-coated slides and stained with the TUNEL method for in situ demonstration of DNA fragmentation, following previously described methods [4]. The number of apoptotic cells was determined in CL of the current cycle on proestrus and in the same generation of CL on estrus.
Experiment 2 The objective of this experiment was to analyze the morphological and quantitative changes that happen in cycling rats during the transition from proestrus to estrus in regressing CL of the current and of the previous cycle, and the effect of blocking the preovulatory PRL surge by administration of CB154. Three cycling rats were killed on the morning of proestrus (1000 h). Additional rats were injected s.c. on the morning of proestrus (1000 h) with 1 mg/rat of CB154 or vehicle (70% ethanol in saline). Previous studies have shown that treatment with 1 mg of CB154 on the morning of proestrus effectively blocks the preovulatory PRL surge [21] and inhibits structural luteolysis on the transition from proestrus to estrus [3, 4]. Three animals per group were killed on the morning (1000 h) of estrus or metestrus. The ovaries were dissected and fixed for 24 h in Bouin-Hollande fluid at room temperature, processed for paraffin embedding, serially sectioned at 5 µm, and stained with hematoxylin and eosin.
Apoptotic cells were counted on proestrus, estrus, and metestrus CL. The volume of the CL, the cross-sectional area, the number of steroidogenic cells per CL, and the ratio of nonsteroidogenic to steroidogenic cells were determined on proestrus and metestrus (before and after the apoptotic burst on the transition from proestrus to estrus) in the last two generations of regressing CL.
Experiment 3 The objective of this experiment was to analyze whether the preovulatory PRL surge associated with postpartum ovulation on the day of parturition [22, 23] induces luteolytic changes in the CLP. Three pregnant rats were killed on Day 21 of pregnancy. On the morning of Day 22 of pregnancy (day of parturition), pregnant rats were injected s.c. with CB154 (1 mg/rat at 1000 h to block the PRL surge associated to the postpartum ovulation) or vehicle, pups were removed at delivery, and the animals were killed on Days 1 and 2 postpartum (three rats per day). Previous data indicate that the inhibitory effects of CB154 on PRL secretion last for at least 48 h [21]. The ovaries were fixed in Bouin-Hollande fluid, embedded in paraffin, and the right ovaries were serially cut at 5 µm thick. The number of apoptotic cells, the volume of the CL, the cross-sectional area and the number of steroidogenic cells per CL were determined in the CLP.
Experiment 4 The objective of this experiment was to analyze the CLP during postpartum regression in different situations with respect to the patterns of PRL secretion. Pregnant rats were divided into three groups at parturition. In a first group (lactating rats), dams were left with pups (10 pups/dam) and killed on Day 5 or 10 of lactation (five animals per group). In a second group (nonlactating, noncycling rats), pups were removed at delivery, and cyclicity (and, therefore, PRL surges) was blocked by treatment with LHRH antagonist (1 mg/rat, s.c., on Day 2, 4, 6, and 8 postpartum). These animals were killed on Day 5 or 10 postpartum (five animals per group). The ovaries were processed as in previous experiments, serially sectioned at 5 µm, and stained with hematoxylin and eosin. The number of apoptotic cells, the volume of the CL, the cross-sectional area, and the number of steroidogenic cells were determined in the CLP. In a third group (nonlactating, cycling rats), pups were removed at delivery. Daily vaginal smears were taken. The day of parturition, at which the first postpartum ovulation occurs [23], was considered as the first postpartum estrus. Three rats per group were killed on the second postpartum estrus (on Day 5 after parturition) or on the third postpartum estrus (on Day 9 after parturition). The ovaries were fixed in Bouin-Hollande fluid and processed as in previous experiments. The number of apoptotic cells were counted in the CLP, as well as in the cyclic CL from the first and second postpartum ovulations.
Cell Counting and Stereological Study
Identification of the different types of CL In cycling rats, the last two generations of regressing CL were easily recognized in hematoxylin- and eosin-stained sections, based on their different size and ratio of nonsteroidogenic to steroidogenic cells (Fig. 3). Corpus lutea more than two cycles old were considerably smaller and were not considered in this study. The CLP was easily recognized either on Day 21 of pregnancy or during postpartum regression by its large size and the presence of arterioles and fully luteinized steroidogenic cells, displaying cytoplasmic vacuolation during postpartum regression. In all types of CL, steroidogenic cells were clearly distinguished from nonsteroidogenic cells by their large cytoplasm and round nucleus containing a prominent nucleolus.
|
Counting of apoptotic cells Apoptotic cells were counted in hematoxylin- and eosin-stained sections. They were recognized by specific morphological features [24–26], such as shrunken eosinophilic cytoplasm, chromatin condensation, and nuclear segmentation into apoptotic bodies at advanced stages. Previous studies showed that luteal cells showing the morphological characteristics of apoptosis contained fragmented DNA when stained with the TUNEL method and counts of luteal apoptotic cells based on morphological criteria were equivalent to those based on in situ 3' end-labeling [4]. The presence of fragmented DNA during early PRL-induced luteolysis was also confirmed in this study (first experiment) by staining with the TUNEL method. Apoptotic cells were counted with a 100x objective in three different sections per CL and three different CL per rat, in each CL generation. Each CL section was systematically scored and the number of apoptotic nuclei expressed per area unit of luteal tissue.
Determination of CL volume, steroidogenic cell size, and number
The volume of the CL was obtained from serial sections, by considering it as an ellipsoid. For this, two diameters at right angles were measured under a 4x objective in the largest CL section, with a micrometer eyepiece incorporated into the microscope, in at least five different CL per ovary and CL generation. The cross-sectional area of steroidogenic cells was obtained by point counting with the aid of a 121 test-point reticle (16 000 µm2) incorporated into the microscope. Only cell sections showing the nucleolus were measured. The number of test points that hit on the cell section were counted with the 100x objective, in at least 100 steroidogenic cells per animal and CL generation, and the mean cross-sectional area was Pi (a/p), where Pi is the number of points hitting on the cell section, and a/p the area corresponding to each test point. The number of steroidogenic cells per CL was obtained by the nucleator method [27, 28]. The volume density of steroidogenic cell nuclei (Vvn) was obtained by point counting with a 121 test-point reticle. The volume density was Vvn = Pi/Pt, where Pi corresponds to the numbers of points hitting on cell nuclei and Pt the number of test points. The product of Vvn by the CL volume gives the absolute volume occupied by steroidogenic cell nuclei per CL (Vn). In each steroidogenic cell nuclear section showing the nucleolus, four separate radii (li) were measured from the nucleolus to the nuclear membrane, with a micrometer eyepiece incorporated to the microscope under the 100x objective, in at least 100 steroidogenic cells per rat and CL generation. The unbiased mean volume of the steroidogenic cell nuclei was vn = (4
/3)[(l13 + l23 + l33 + l43)/4] = (4/3)li3. By dividing Vn by vn, we obtained the number of steroidogenic cells per CL. The ratio of nonsteroidogenic to steroidogenic cells was obtained by counting the number nuclei of each cell type per area unit.
Data are presented as the mean ± SEM. Statistical analyses were performed by ANOVA and Tukey test for multiple comparison among means.
RESULTS
Experiment 1. Timing of Apoptosis During the Transition from Proestrus to Estrus in Cycling Rats
In the CL of the current cycle of proestrus, apoptotic cells were very scarce on the morning (1000 h) and early evening (1600 h) of proestrus, before the preovulatory PRL surge. However, apoptosis was extremely abundant shortly after the preovulatory PRL surge, at 2100 h on the evening of proestrus (Fig. 1A). In situ 3' end-labeling with the TUNEL method confirmed that these early apoptotic cells contained fragmented DNA (Fig. 1B). On the morning of estrus (1000 h), apoptotic cells were still abundant, decreasing at 1300 h, and returning to basal values at 1800 h on the evening of estrus. Quantitative data are presented in Figure 2. The number of apoptotic cells was significantly increased from 2100 h on the evening of proestrus to 1000 h on the morning of estrus, significantly decreased at 1300 h on estrus, whereas at 1800 h on the evening of estrus the number was equivalent to that found before the preovulatory PRL surge.
|
|
Experiment 2. Structural and Quantitative Changes During the Transition from Proestrus to Estrus in Cycling Rats. Role of PRL
On the morning of proestrus, CL of the current cycle showed well-preserved tissue structure (Fig. 3A), and apoptotic cells were only occasionally observed. Morphological changes that happen on the transition from proestrus to estrus have been previously described [4]. Briefly, on the morning of estrus there was a general disorganization of the luteal tissue, and apoptotic cells were very abundant (Fig. 3B). On the morning of metestrus, apoptotic cells were only occasionally found, and the proportion of nonsteroidogenic cells was increased (Fig. 3C). The CL of the previous cycle were easily identified by their smaller size and a higher proportion of nonsteroidogenic cells (Fig. 3D) than those of the current cycle. Similarly to that found for the youngest generation of regressing CL, apoptotic cells were very scarce on the morning of proestrus (Fig. 3D), abundant on the morning of estrus (Fig. 3E), and scarce again on the morning of metestrus (Fig. 3F), when the proportion of nonsteroidogenic cells was clearly increased. In rats lacking a PRL surge by treatment with CB154, all the above-described changes were absent in both CL generations.
Quantitative data are presented in Figure 4. In the youngest generation of regressing CL (during the first transition from proestrus to estrus) there was a 56-fold increase in the number of apoptotic cells from proestrus to estrus (first apoptotic burst). In these CL, there was a 57% fall in the volume of the CL and an 81% fall in the number of steroidogenic cells per CL from proestrus to metestrus. The ratio of nonsteroidogenic to steroidogenic cells increased from 0.88 ± 0.09 in proestrus to 2.17 ± 0.22 in metestrus (mean ± SEM for n = 3; P < 0.01). In two-cycle-old CL, the number of apoptotic cells increased 8.2-fold from proestrus to estrus (second apoptotic burst), and there was a 65% fall in CL volume and a 83% fall in the number of steroidogenic cells from proestrus to metestrus. The ratio of nonsteroidogenic to steroidogenic cells increased from 2.19 ± 0.31 in proestrus to 3.30 ± 0.08 in metestrus (mean ± SEM for n = 3; P < 0.05). However, the mean cross-sectional area of steroidogenic cells did not change significantly from proestrus to metestrus in any CL generation. All these quantitative changes were also blocked in animals injected with CB154 on the morning of proestrus.
|
Experiment 3. Luteolytic Changes in the CL of Pregnancy During Early Postpartum. Role of the Postpartum Ovulation PRL Surge
On Days 1 and 2 postpartum, newly formed CL (Fig. 5) destined to develop into CL of lactation, as well as oocyte-containing cumulus in the oviduct, were present as a consequence of the ovulatory event that happened on the day of parturition. Morphological changes in CLP, associated with postpartum ovulation, were different from those observed in the cyclic regressing CL during the transition from proestrus to estrus (Fig. 5 compares cyclic CL and CLP, following ovulation). On Day 21 of pregnancy and Day 1 postpartum, isolated steroidogenic cells showed highly vacuolated cytoplasm (Fig. 6, A and B). On Day 2 postpartum, all steroidogenic cells showed cytoplasmic vacuolation and were considerably smaller (Fig. 6C). However, apoptotic cells were not observed (Figs. 5, B and D and 6, B and C). The same morphological features were observed in rats injected on the morning of the day of parturition with CB154. Quantitative data for CLP during early postpartum regression are presented in Figure 7. From Day 21 of pregnancy to Day 2 postpartum there was a 57% decrease in the volume of the CL and a 36% decrease in the mean cross-sectional area of steroidogenic cells, whereas no changes were found for the number of steroidogenic cells per CL. These changes were not affected by blocking PRL secretion on the day of parturition with CB154.
|
|
|
Experiment 4. Luteolytic Process of the CL of Pregnancy During Postpartum. Role of PRL
In postpartum rats that were not cycling, due to either pup suckling (lactating rats) or by treatment with LHRHa (nonlactating, noncycling rats), two sets of CL were present: regressing CLP, and CL derived from the postpartum ovulation. These later CL were similar to regressing cyclic CL on proestrus, without signs of structural luteolysis, in LHRHa-treated rats, whereas they were large and highly luteinized in lactating rats (CL of lactation). Nonlactating, noncycling rats showed, on Day 5 postpartum, large CLP containing steroidogenic cells with variable cytoplasmic vacuolation, whereas on Day 10 postpartum vacuolation was not evident. The tissue structure was well preserved, and apoptotic cells were not found at any time. In lactating rats, and on Day 5 postpartum, the morphology of the CLP was similar, except for the presence of scattered apoptotic cells, whereas on Day 10 postpartum the CLP was clearly smaller, and apoptotic cells were more abundant. Quantitative data are shown in Figure 8. On Day 5 postpartum there were no significant differences between nonlactating noncycling and lactating rats for any variable, except for the number of apoptotic cells, that was significantly higher in lactating rats. On Day 10 postpartum, the number of apoptotic cells was also significantly higher in lactating rats, and the volume of the CLP and the number of steroidogenic cells were significantly (P < 0.01) decreased. The mean cross-sectional areas of steroidogenic cells were not significantly different.
|
In postpartum cycling rats, the day of estrus was confirmed by the presence of newly formed CL and oocyte-containing cumulus in the oviduct. On the day of the second postpartum estrus, two sets of regressing CL were present: the cyclic CL from the postpartum ovulation (CL1) and the CLP. On the day of the third postpartum estrus, postpartum cycling rats showed three sets of regressing CL: those from the first postpartum ovulation (CL1), those from the second postpartum ovulation (CL2), and the CLP. Quantitative data are presented in Figure 9. Apoptotic cells were practically absent in CLP on the day of the second postpartum estrus (on Day 5 postpartum) and were abundant on the day of the third postpartum estrus (on Day 9 postpartum). Cyclic CL of the first (CL1) and second (CL2) postpartum generations showed the expected number of apoptotic cells corresponding to the first and second cyclic apoptotic bursts (see experiment 2).
|
DISCUSSION
The data of this study confirmed previous reports indicating that apoptosis in regressing cyclic CL occurs during the transition from proestrus to estrus [2, 4] and affected different CL generations [2]. This is the first study quantifying changes in the number and size of steroidogenic cells during structural luteolysis in the rat. Changes in cell numbers, together with the amount of apoptosis, provides information about the magnitude of cell death during each transition from proestrus to estrus. About 80% of steroidogenic cells were eliminated in each apoptotic burst. Because the size of steroidogenic cells did not significantly decrease through CL regression, cell loss by apoptosis seemed to be the main mechanism for the PRL-induced decrease in luteal tissue mass during structural luteolysis in cycling rats, although a decrease in the volume occupied by blood vessels is likely also involved. The large decrease found in the number of steroidogenic cells after each apoptotic burst indicates that apoptosis preferentially affects this luteal cell subpopulation. This was also supported by the increase in the ratio of nonsteroidogenic to steroidogenic cells after each apoptotic round. Previous studies have found that proliferative activity was very low in regressing CL [3] and, therefore, the increase in the proportion of nonsteroidogenic cells seemed to be due mainly to steroidogenic cell loss. Macrophages, however, that invade the CL following the PRL surge [3, 10] also contribute to the number of nonsteroidogenic cells.
The large decrease in the number of steroidogenic cells in both CL generations was in keeping with the abundance of apoptosis induced by the preovulatory PRL surge. Our results are not in agreement with previous studies [5, 6] that reported very low rates of apoptosis during the transition from proestrus to estrus. This discrepancy may be due to several reasons. In some experiments [5, 6], these authors scored apoptosis on the evening of estrus. However, PRL-induced apoptotic bursts are transitory. In the present study, the apoptotic burst started shortly after the preovulatory PRL surge (at 1830 h in our colony [20]), and apoptotic cells were extremely abundant from the evening (2100 h) of proestrus to the morning (1000 h) of estrus but were already scarce on the evening (1800 h) of estrus. This agrees with a previous study [2] indicating that DNA fragmentation was decreased on the morning of estrus. Additionally, differences could also be due to different criteria to identify apoptotic cells. In the present study, apoptotic cells were recognized by their characteristic morphological features [24–26], that constitute a relatively unequivocal hallmark of this type of cell death [25, 26]. In a recent review [26] on the methods to identify cell death in tissues, morphological changes are considered to be the most accurate indicator of apoptosis. In situ-3' end-labeling methods have several pitfalls, false-positive and -negative results are frequent and results should be carefully evaluated and contrasted with morphological data [26]. In our experience, in situ-3' end-labeling methods are highly variable, likely due to sensitivity to tissue processing and proteinase treatment. Moreover, the number of apoptotic cells detected by stringent interpretation of TUNEL-stained sections, was not significantly different to that detected by morphological interpretation [4].
The increase in apoptosis found in the present study in regressing cyclic CL from proestrus to estrus and the consequent decrease in CL volume and number of steroidogenic cells were dependent on the preovulatory PRL surge, as indicated by the absence of apoptosis and changes in stereological variables in CB154-treated animals. This agrees with previous studies reporting inhibition of luteal cell apoptosis [4] or DNA fragmentation [2] after blocking PRL secretion. Although previous studies [4, 19] support that additional factors may be necessary for luteal cell death, the PRL surge is an absolute requirement for the induction of apoptosis. In this study, treatment with CB154 also inhibited apoptosis in corpora lutea of the previous cycle, indicating that PRL surge was responsible for the different apoptotic bursts along CL regression.
We also studied regressive changes in the CLP during postpartum involution, and the role of PRL on the regression of fully differentiated luteal cells. Surprisingly, during the first 2 days postpartum, CLP did not show apoptotic cells in response to the preovulatory PRL surge on the day of parturition [22, 23]. This agrees with the stereological data indicating that the number of steroidogenic cells did not change from Day 21 of pregnancy to Day 2 postpartum. Furthermore, aside from absence of apoptosis, steroidogenic cells showed distinctive morphological features such as cytoplasmic vacuolation and cellular atrophy (i.e., decrease in cell size). Therefore, and in contrast to what happens in cyclic CL, the decrease in cell size seemed to be the main mechanism for the decrease in the volume of the CLP during early postpartum. This strongly suggests that during early postpartum, the rat CLP follows a distinctive luteolytic pattern that is independent of PRL, as indicated by the lack of effects of treatment with CB154 on the day of parturition. Probably, early changes during postpartum involution are associated with the reported luteolytic effects of LH during late pregnancy [29, 30].
Comparison of lactating rats (showing high PRL concentrations induced by pup suckling [31, 32]) and nonlactating, noncycling rats (showing absence of PRL surges) indicates that PRL does not induce significant amounts of apoptosis before Day 5 postpartum. Although apoptotic cells were present in low numbers in lactating rats on Day 5 postpartum, the lack of significant changes in the CL volume or in the number of steroidogenic cells suggested that apoptosis had just started at this time. Otherwise, a large decrease in both the number of steroidogenic cells and the CL volume was found in lactating rats on Day 10. This agrees with previous studies indicating that long-term regression of the CLP during lactation is dependent on PRL concentrations [18, 33]. The absence of PRL surges in LHRHa-treated rats was also evidenced by the lack of structural luteolysis in the CL of the postpartum ovulation on Days 5 and 10 postpartum. Conversely, in lactating rats, the existence of increased PRL concentrations induced by pup suckling was evidenced by the development of CL of the postpartum ovulation into fully luteinized CL of lactation.
The lack of PRL responsiveness of the CLP during the first days postpartum is also in accordance with data from postpartum cycling rats. On the second postpartum estrus (on Day 5 postpartum), apoptosis was absent in the regressing CLP, in spite of the abundance of apoptotic cells that were present in the regressing cyclic CL from the postpartum ovulation. However, apoptosis was abundant on the third postpartum estrus (on Day 9 postpartum), both in the regressing CLP and in the two sets of regressing cyclic CL from the first and second postpartum ovulations. The data from this study suggest that the loss of the morphological and functional characteristics of fully differentiated luteal steroidogenic cells (i.e., dedifferentiation) is required before becoming sensitive to the luteolytic effects of PRL. Because PRL is also the main luteotrophic hormone in the rat [8, 34], the lack of responsiveness of fully differentiated luteal cells to the proapoptotic effects of PRL could provide a safety mechanism to prevent luteolysis in the CLP. Additional studies should explore whether PRL itself, or its effects on CL, protect against PRL-induced luteolysis, as well as the mechanisms underlying differences in the sensitivity to PRL of fully versus nonfully differentiated luteal cells.
ACKNOWLEDGMENTS
The authors are very grateful to J. Molina, T. Recio, and E. Tarradas for their technical assistance.
FOOTNOTES
First decision: 6 February 2001.
1 This work has been subsidized by grants 1FD97-1065-CO3-03 and PM98-0167 from the Dirección General de Enseñanza Superior e Investigación Cientifica, Spain. 
2 Correspondence. FAX: 34 57 218288; fi1sacrj{at}lucano.uco.es
Accepted: March 14, 2001.
Received: January 10, 2001.
REFERENCES
-ergocryptine (CB154). Experientia 1971; 27:464-465[CrossRef][Medline]-hydroxysteroid dehydrogenase luteal activation in late pregnant rat. J Endocrinol 1996; 150:423-429 in LH-induced luteolysis in pregnant rats. J Endocrinol 1998; 156:253-259[Abstract]This article has been cited by other articles:
|
|
 |
|
  R. Gonzalez-Fernandez, E. Martinez-Galisteo, F. Gaytan, J. A. Barcena, and J. E. Sanchez-Criado Changes in the Proteome of Functional and Regressing Corpus Luteum During Pregnancy and Lactation in the Rat Biol Reprod, July 1, 2008; 79(1): 100 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ben-Jonathan, C. R. LaPensee, and E. W. LaPensee What Can We Learn from Rodents about Prolactin in Humans? Endocr. Rev., February 1, 2008; 29(1): 1 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A Goyeneche, J. M Harmon, and C. M Telleria Cell death induced by serum deprivation in luteal cells involves the intrinsic pathway of apoptosis Reproduction, January 1, 2006; 131(1): 103 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Gonzalez-Fernandez, F Gaytan, E Martinez-Galisteo, P Porras, C A Padilla, J E Sanchez Criado, and J A Barcena Expression of glutaredoxin (thioltransferase) in the rat ovary during the oestrous cycle and postnatal development J. Mol. Endocrinol., June 1, 2005; 34(3): 625 - 635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Takiguchi, N. Sugino, K. Esato, A. Karube-Harada, A. Sakata, Y. Nakamura, H. Ishikawa, and H. Kato Differential Regulation of Apoptosis in the Corpus Luteum of Pregnancy and Newly Formed Corpus Luteum after Parturition in Rats Biol Reprod, February 1, 2004; 70(2): 313 - 318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-P. Sun, T. J. Lee, and G. V. Shah Calcitonin Expression in Rat Anterior Pituitary Gland Is Regulated by Ovarian Steroid Hormones Endocrinology, October 1, 2002; 143(10): 4056 - 4064. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Gaytan, C. Morales, C. Bellido, and J. E. Sanchez-Criado Selective Apoptosis of Luteal Endothelial Cells in Dexamethasone-Treated Rats Leads to Ischemic Necrosis of Luteal Tissue Biol Reprod, January 1, 2002; 66(1): 232 - 240. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
