The Extent to which Relaxin Promotes Proliferation and Inhibits Apoptosis of Cervical Epithelial and Stromal Cells Is Greatest during Late P
http://www.100md.com
内分泌学杂志 2005年第1期
Department of Molecular and Integrative Physiology (H.-Y.L., S.Z., O.D.S.) and College of Medicine (O.D.S.), University of Illinois at Urbana-Champaign, Urbana, Illinois 61801; and Department of Cell Biology and Neuroscience, University of South Alabama College of Medicine (P.A.F.), Mobile, Alabama 36688
Address all correspondence and requests for reprints to: Dr. O. D. Sherwood, Department of Molecular and Integrative Physiology, University of Illinois at Urbana-Champaign, 524 Burrill Hall, 407 South Goodwin Avenue, Urbana, Illinois 61801. E-mail: od-sherw@uiuc.edu.
Abstract
Relaxin promotes marked growth of the cervix during the second half of rat pregnancy, and this growth is accompanied by an increase in both epithelial and stromal cells. The objective of this study was to test the hypothesis that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical cells is greatest during late pregnancy in rats. The influence of neutralization of circulating relaxin by iv injection of 5 mg monoclonal antibody against rat relaxin (MCA1) was examined at 3-d intervals throughout the second half of pregnancy. Controls were injected with either 5 mg monoclonal antibody against fluorescein or 0.5 ml PBS vehicle. To evaluate cell proliferation, 5'-bromo-2-deoxyuridine was injected sc 8 h before cervixes were collected. Terminal deoxynucleotidyl transferase-mediated deoxyuridine 5'-triphosphate nick end-labeling and electron microscopy were used to detect apoptotic cells. Neutralization of relaxin with MCA1 decreased the rate of proliferation and increased the rate of apoptosis of cervical cells by d 13. However, the extent to which relaxin influenced these processes was greatest and dramatic by late pregnancy. In MCA1-treated rats on d 22 of pregnancy, the rates of proliferation of both epithelial and stromal cells were less than 20% those in controls, and the rates of apoptosis in epithelial cells and stromal cells were more than 10- and 3-fold, respectively, greater than those in controls. In conclusion, this study provides evidence that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical epithelial and stromal cells is greatest during late pregnancy.
Introduction
RELAXIN IS SECRETED by the corpora lutea throughout the second half of the 23-d rat pregnancy. The hormone is first detected in the peripheral serum on d 10 of pregnancy, and levels increase to approximately 80 ng/ml by d 20. Serum relaxin surges to levels ranging from 120–200 ng/ml on d 21 of pregnancy, then rapidly declines to about 2 ng/ml by postpartum d 2 (1, 2).
Relaxin is indispensable during pregnancy in rats. When rats were made relaxin deficient by means of either ovariectomy or neutralization of circulating relaxin with monoclonal antibody specific for rat relaxin, the duration of delivery was markedly prolonged, and about 50% of the fetuses were born dead or retained in utero (3, 4, 5). Relaxin enables rapid and safe delivery of the fetuses by promoting growth and softening of the cervix during the second half of pregnancy (6, 7).
In the rat, cervical wet weight increases about 3-fold during the second half of pregnancy (8, 9, 10, 11). Our laboratory provided evidence that approximately 40–50% of the increase in the cervical wet weight that occurs during this period is relaxin dependent (5, 7). Early reports indicated that relaxin contributes to cervical growth by increasing the extracellular content of water, collagen, hyaluronic acid, and proteoglycan (12, 13). More recent studies demonstrated that the increase in cervical wet weight is also partially attributed to relaxin-induced accumulation of new cells (5, 14). Relaxin is responsible for more than 50% of the new cells that accumulate in the cervical epithelium and stroma throughout the entire second half of pregnancy (14).
The cellular mechanism(s) by which endogenous relaxin brings about the increase in cervical cell content has received little attention. Homeostatic control of cell number is thought to result from the dynamic balance between cell proliferation and apoptosis (15). Data we obtained only on d 22 of pregnancy indicate that relaxin may influence both processes. The finding that the number of cells that contained the thymidine analog 5-bromo-2'-deoxyuridine (BrdU) was far lower in cervixes from ovariectomized rats that were devoid of circulating relaxin throughout the last 14 d of pregnancy than in cervixes from controls can be interpreted to indicate that relaxin promotes cell proliferation (14). The observation that the rate of apoptosis in cervixes obtained from rats in which endogenous relaxin was neutralized on d 19–21 was greater than that in cervixes from controls provided evidence that relaxin inhibits apoptosis during the antepartum period (16). To date, it has not been determined whether endogenous relaxin influences the rates of either cervical cell proliferation or apoptosis throughout the second half of pregnancy. Moreover, it is not known whether relaxin’s effects on the rates of these two dynamic processes for regulating cell number change as pregnancy progresses.
There are at least three reasons to think that relaxin’s effects on cervical cell proliferation and apoptosis rates may be greatest during late pregnancy. First, the rising serum levels of relaxin during this period may exert increasingly pronounced affects on both processes. Second, there is a progressive shift toward increasing ratios of 6,500 mol wt and 13,000 mol wt forms of relaxin relative to a 60,000 mol wt form in the serum, and the smaller form(s) of relaxin may be most active (17). Third, because relaxin’s effects on the cervix are estrogen dependent (18, 19), a steady increase in the serum levels of estrogen during the second half of pregnancy (20) may enhance the efficacy of relaxin’s effects on cervical cells.
Based on this information, it was hypothesized that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical cells is greatest during late pregnancy in rats. This hypothesis was examined by determining the rates of proliferation and apoptosis of cervical cells after the passive neutralization of circulating endogenous relaxin at 3-d intervals throughout the second half of pregnancy. Relaxin was neutralized with a monoclonal antibody for rat relaxin (MCA1) (21), which effectively blocks the biological actions of relaxin in pregnant rats (4, 7, 8, 22, 23). The effects of relaxin on proliferation and apoptosis rates of cervical epithelial and stromal cells were independently determined because distinct patterns of proliferation and apoptosis for each of these cellular compartments during rat pregnancy were reported (24).
Materials and Methods
Animals
Primiparous Sprague Dawley-derived rats, bred at approximately 75 d of age, were obtained from Harlan (Indianapolis, IN) on d 3 of pregnancy. The day that sperm was found in the vagina was designated d 1 of pregnancy. The animals were housed individually and maintained in a light-controlled room (lights on between 0700 and 2100 h) at a temperature of 23–25 C. Beginning on d 8, the photoperiod was advanced 8 h so that lights were on from 2100–1100 h for the remainder of pregnancy. This shift in the light schedule was previously demonstrated to synchronize the time of delivery to the early morning hours on d 23 of pregnancy, which is also designated d 1 postpartum (25). Rats were provided with Teklad 6% mouse/rat diet 7002 (Harlan/Teklad, Madison, WI) and water ad libitum. The animal experimentation described in this study was approved by the University of Illinois laboratory animal advisory committee.
On d 8 of pregnancy, rats were anesthetized with ether, and ventral laparotomy was performed to determine the number of implantation sites. Because serum relaxin levels are directly proportional to litter size in rats with five or fewer conceptuses (26), rats with less than eight implantation sites were excluded from this study.
Animal treatment for examination of relaxin’s effects on cell proliferation
Figure 1 shows the experimental design. There were three treatments: PBS control, a monoclonal antibody for fluorescein (MCAF) (4) control, and MCA1. Six animals were used for each group. Cervical tissues were collected at 3-d intervals beginning on d 7 of pregnancy and continuing until d 3 postpartum. Preliminary experiments demonstrated that neutralization of endogenous relaxin with MCA1 for 48 h had maximal inhibiting effects on cellular proliferation in both cervical epithelial and stromal cells and that administration of BrdU sc 8 h before tissue collection enabled the demonstration of treatment effects (data not shown). Accordingly, for 2 consecutive days beginning at 0900 h on d 5, 8, 11, 14, 17, and 20 of pregnancy and on d 1 postpartum, rats were injected via the tail vein with 0.5 ml PBS vehicle, 5 mg MCAF, or 5 mg MCA1. Forty hours after treatment began (8 h before tissue collection), 800 μg BrdU (Sigma-Aldrich Corp., St. Louis, MO) dissolved in 0.5 ml sterile saline (0.9% NaCl) were injected sc.
FIG. 1. Diagram of the experimental design for examining endogenous relaxin’s effects on cervical cell proliferation. See Materials and Methods for details.
Tissue collection and processing for evaluation of cell proliferation
Animals were killed, and cervical tissues were collected at 0900 h on d 7, 10, 13, 16, 19, and 22 of pregnancy and on d 3 postpartum. Cervixes were removed, trimmed of extraneous tissues, and prefixed in 10% neutral buffered formalin for 2 h. Each cervix was then bisected in cross-section to obtain cephalic and caudal halves. Fixation was continued in fresh neutral buffered formalin for a total of 24 h. After fixation, cervixes were dehydrated in an ascending series of ethanol, cleared in xylene, and embedded in paraffin (27). The cervical tissue blocks were sectioned at 5 μm thickness, mounted on positively charged slides (SuperFrost Plus, Fisher Scientific, Pittsburgh, PA), and air-dried overnight. Tissue sections from each tissue block were at least 50 μm apart to ensure that different cervical cells were analyzed. Two sections from each block were analyzed.
BrdU immunohistochemistry for detection of cell proliferation
BrdU immunostaining was performed as previously described with some modification (14, 28). Briefly, BrdU incorporation into proliferating cells was determined immunohistochemically using a mouse monoclonal antibody (clone NCL-BrdU, Vector Laboratories, Inc., Burlingame, CA) and a Vectastain Elite ABC kit (Vector Laboratories, Inc.). All incubations, unless otherwise noted, were performed at 25 C. Slides were cleared in xylene and rehydrated in a descending series of ethanol. DNA was denatured by incubation in aqueous 2 N HCl for 30 min at 37 C. After this and all other incubations, slides were rinsed in three changes of PBS (pH 7.5; 5 min/rinse). Antibody penetration was improved by incubation in 0.01% trypsin (Sigma-Aldrich Corp.) in PBS for 15 min at 37 C. Tissue sections were immersed in 3% hydrogen peroxide for 15 min to quench endogenous peroxidase activity. After incubation with blocking buffer (3% normal horse serum in PBS; Vector Laboratories, Inc.) for 30 min, anti-BrdU antibody (1:300 dilution in blocking buffer) was applied, and slides were incubated overnight (16 h) in humidified chambers at 4 C. Biotinylated horse antimouse IgG and avidin-biotin-peroxidase complex were prepared as directed (Vectastain Elite ABC kit, Vector Laboratories, Inc.). Biotinylated horse antimouse IgG (1:65 dilution in blocking buffer) was applied for 30 min. The avidin-biotin-peroxidase complex was then applied for 30 min. Antibody-binding sites were visualized using a 3,3'-diaminobenzidine peroxidase substrate (Peroxidase Substrate Kit, Vector Laboratories, Inc.) that was prepared as directed and applied to sections for 2–3 min. Slides were rinsed with distilled water for 10 min and counterstained with Ehrlich’s hematoxylin for 10 min. Slides were washed with tap water for 10 min, then dehydrated in ethanol, cleared, and coverslipped with Permount (Fisher Scientific). For negative control slides, either anti-BrdU antibody was omitted (blocking buffer only) or nonspecific mouse IgGs were substituted (1:300 in blocking buffer; Sigma-Aldrich Corp.).
Morphometric analysis of relaxin’s effects on epithelial cell and stromal cell proliferation
To determine the cellular proliferation rate of the cervical cells, the labeling index (LI) of BrdU-positive cells was determined. The LI was obtained by dividing the number of BrdU-positive cells by the total number of cells analyzed per section and multiplying by 100. Sections were examined morphometrically at x400 magnification with a BH-2 light microscope (Olympus Corp., Melville, NY) equipped with a video camera and connected to a personal computer running a Stereo Investigator program (MicroBrightField, Inc., Colchester, VT). The Stereo Investigator program automatically controls the movement of the microscope stage to permit unbiased selection of fields of analysis. Epithelial cells and stromal cells were analyzed independently. Within the stroma, nonsmooth muscle cells (primarily fibroblasts) and smooth muscle cells (both circular and longitudinal) were analyzed independently. Data were obtained from four sections (two sections from each cervical half/rat), and at least 300 epithelial cells and 500 stromal cells were analyzed per section. Thus, at least 1200 epithelial cells and 2000 stromal cells were analyzed for each of the six rats per group.
Animal treatment for examination of relaxin’s effects on apoptosis
Figure 2 shows the experimental design. There were three treatments: PBS control (n = 6), MCAF control (n = 4), and MCA1 (n = 6). Cervical tissues were collected on d 5 and at 3-d intervals beginning on d 10 of pregnancy and continuing until d 3 postpartum. Previous experiments demonstrated that neutralization of endogenous relaxin with MCA1 for 24 h had a maximal effect on apoptosis in both cervical epithelial and stromal cells (16). Accordingly, at 0900 h on d 4, 9, 12, 15, 18, and 21 of pregnancy and on d 2 postpartum, rats were injected via the tail vein with 0.5 ml PBS vehicle, 5 mg MCA1, or 5 mg MCAF.
FIG. 2. Diagram of the experimental design for examining endogenous relaxin’s effects on cervical cell apoptosis. See Materials and Methods for details.
Tissue collection and processing for evaluation of apoptosis
Animals were killed, and cervical tissues were collected at 0900 h on d 5, 10, 13, 16, 19, and 22 of pregnancy and on d 3 postpartum. Cervixes were trimmed of extraneous tissue, fixed, and sectioned as described above for the evaluation of cell proliferation.
Terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine 5'-triphosphate nick end-labeling (TUNEL) immunohistochemistry for detection of apoptosis
TUNEL in conjunction with morphometric analysis was employed as previously described (16) to detect and quantify cells undergoing apoptosis. In brief, sections were stained immunocytochemically by TUNEL using the method described by Gavrieli et al. (29). A commercial kit (ApopTag In Situ Apoptosis Detection, Serologicals Corp., Norcross, GA), which links digoxigenin-nucleotide to DNA by TdT, was used. Sections were deparaffined with xylene, rehydrated with a descending series of ethanol, incubated with proteinase K, immersed in 3% aqueous hydrogen peroxide, and then pretreated with equilibration buffer. DNA was labeled at the 3' end by incubating sections with a mixture of digoxigenin deoxynucleotide triphosphate, unlabeled deoxynucleotide triphosphate, and TdT enzyme at 37 C for 1 h. Slides were washed with PBS and incubated with antidigoxigenin antibody conjugated to peroxidase at room temperature for 30 min. Slides were washed again in PBS, incubated with 3,3'-diaminobenzidine peroxidase substrate, counterstained with methyl green, mounted, and sealed. Positive control slides were treated with deoxyribonuclease I before the labeling reaction. Negative control slides were incubated with labeling reaction solution devoid of TdT enzyme.
Morphometric analysis of relaxin’s effects on epithelial and stromal cell apoptosis
Sections were examined morphometrically at x400 magnification with a BH-2 light microscope (Olympus Corp.) equipped with a video camera and connected to a personal computer running a Stereo Investigator program (MicroBrightField, Inc.). Epithelial cells and stromal cells were analyzed independently. As with cell proliferation, nonsmooth muscle cells and smooth muscle cells within the stroma were analyzed independently. The percentage of TUNEL-labeled cells (LI) was determined. Data were obtained from four sections (50 μm apart)/rat, and at least 500 epithelial cells and 500 stromal cells/section were analyzed per section. Thus, at least 2000 cells were analyzed per rat for the epithelium, and 2000 cells were analyzed per rat for the stroma.
Ultrastructural analysis of apoptosis in cervical cells
To confirm that TUNEL labeling of cervical tissue was indeed attributable to apoptosis, transmission electron microscopy was used to determine whether comparable percentages of cells demonstrated morphological features characteristic of apoptotic cells. Cervixes were collected from two PBS-, MCAF-, and MCA1-treated rats on d 5, 13, and 22 of pregnancy. Cervixes were cut into cephalic and caudal halves, placed in Karnovsky’s fixative (30) for 24 h, washed three times with 0.1 M cacodylate buffer, and processed for electron microscopic evaluation as previously described (16). Each of the cephalic and caudal tissues was cut in half for embedding and sectioning. Sections were taken from three levels of each half. One hundred epithelial, 100 stromal, and 100 smooth muscle cells were evaluated from each section for apoptosis, i.e. condensed chromatin or shrunken nucleus as indicated by the separation of the nuclear envelope. Therefore, from two rats on each of the 3 d of the study, 2400 of each cell type were evaluated. A cell had to have a nucleus in the section to be evaluated.
Evaluation of infiltrating leukocytes and mast cells
To determine the percentage of leukocytes within the cervical stroma, tissue sections were stained with hematoxylin and eosin and examined morphometrically with a BH-2 light microscope (Olympus Corp.). Leukocytes were identified by their multilobed nucleus and distinct cytoplasm (31). No attempt was made to differentiate the leukocytes into various subtypes. At least 250 stromal cells were evaluated per section, and four sections were evaluated per rat. Thus at least 1000 stromal cells were evaluated for each of the six rats per group.
To determine the extent to which mast cells comprise the cervical stromal cell population, they were identified through the naphthol AS-D chloroacetate (3-hydroxy-2-naphthoic-o-toluidide chloroacetate) esterase reaction. Mast cell granules contain proteases, including esterases that rapidly hydrolyze the -chloroacyl esters of the naphthol AS-D chloroacetate. Chloroacetate esterase staining was carried out according to the protocol reported by Sanderson et al. with some modification (32). Briefly, deparaffinized and rehydrated sections were incubated with substrate solution at room temperature for 30 min. The substrate solution was prepared in multisteps just before incubation. First, 0.1 ml pararosaniline (4% dissolved in 2 M HCl) was mixed with 0.1 ml 4% aqueous sodium nitrite for 30–60 sec, followed by addition of 30 ml 70 mM PBS. Finally, 0.01 g naphthol AS-D chloroacetate (Sigma-Aldrich Corp.) dissolved in 1 ml N-dimethylformamide was added, and the combined solution was filtered through Whatman filter paper (Clifton, NJ) before use in the incubation as described above. After incubation was completed, sections were counterstained with hematoxylin and mounted with a coverslip as described above. For determination of the total number of mast cells per cervical cross-section, the entire cross-sectional area was examined using the Axioskop 2 microscope at x400 magnification. Four sections (two from each cervical half) were analyzed per animal. The average total number of mast cell per cervical cross-sectional area was determined from six rats per group.
Statistics
Data were expressed as the mean ± SEM. To determine whether there were significant differences in the BrdU LI and the TUNEL LI of PBS- and MCAF-treated controls on different days, they were tested using one-way ANOVA, followed by Tukey’s test. Statistical analysis of MCA1 treatment effects was assessed using two-factor ANOVA, followed by preplanned, least squares means comparison. The level of significance was set at P < 0.05.
Results
After morphometric analysis of both cell proliferation and apoptosis, no difference between PBS- and MCAF-treated controls was found on any day of treatment.
Morphometric analysis of relaxin’s effects on epithelial and stromal cell proliferation
Figure 3 contains representative photomicrographs that demonstrate the influence of neutralization of endogenous relaxin on proliferation of cervical cells during 2 d of rat pregnancy. The administration of MCA1 did not influence the rate of cell proliferation in the epithelium (Fig. 3A) or in the stroma (Fig. 3B) after its administration during the first half of pregnancy (d 7) when relaxin is not detected in the serum. In contrast, after the administration of MCA1 during late pregnancy (d 22) when relaxin levels are maximal, there was a decline in the rate of cell proliferation in both cellular compartments. The results of morphometric analysis of the cervical epithelial cell proliferation rates on 6 days of pregnancy and d 3 postpartum are shown in Fig. 4A. On d 7, 10, and 13 of pregnancy, the mean BrdU LI in controls was 11.3%. It dropped about 50% on d 16 and remained lower than on d 7, 10, and 13 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 7 and 10 of pregnancy and on d 3 postpartum, when serum relaxin is not elevated, did not affect the BrdU LI. Neutralization of endogenous relaxin with MCA1 did not influence the BrdU LI in the epithelium until d 19 of pregnancy, when there was an approximately 50% reduction in the BrdU LI relative to the control value. Neutralization of relaxin during d 20–22 caused the greatest reduction in the BrdU LI (P 0.01) and did so by nearly 90%.
FIG. 3. Representative photomicrographs that illustrate the influence of neutralization of endogenous relaxin during early and late pregnancy on proliferation of epithelial cells (A) and stromal cells (B). Arrows indicate representative labeled cells. Bar in the lower right photomicrograph, 10 μm. All figures are the same magnification.
FIG. 4. Mean BrdU labeling index (+SEM) in cervical epithelial cells (A) and stromal cells (B) during pregnancy and early postpartum. Each bar represents the mean for six animals. Asterisks indicate differences in group MCA1 from PBS and MCAF controls on the same day (**, P < 0.01).
The results of morphometric analysis of the cervical nonmuscle stromal cell proliferation rates during pregnancy are shown in Fig. 4B. On d 7 and 10 of pregnancy, the mean LI in the stroma of controls was about 0.3%. It increased dramatically by d 13 to approximately 5.5% and remained greater than on d 7 and 10 throughout the remainder of pregnancy (P < 0.01). Administration of MCA1 on d 7 and 10 of pregnancy and on d 3 postpartum, when serum relaxin is not elevated, did not affect the BrdU LI. However, after neutralization of endogenous relaxin with MCA1, the BrdU LI was reduced on d 13, 16, 19, and 22 of pregnancy (P 0.01). Moreover, regression analysis showed that the effect of neutralization of relaxin on stromal cell proliferation was increasingly pronounced as pregnancy progressed (P 0.01). The BrdU LI for cervical smooth cells was less than 1.5% throughout pregnancy, and neutralization of relaxin had no effect on cell proliferation (data not shown).
Morphometric analysis of relaxin’s effects on epithelial cell and stromal cell apoptosis
Figure 5 contains representative photomicrographs that demonstrate the influence of neutralization of endogenous relaxin on apoptosis of cervical cells during 2 d of rat pregnancy. The administration of MCA1 did not influence the rate of apoptosis in the epithelium (Fig. 5A) or in the stroma (Fig. 5B) after its administration during the first half of pregnancy (d 5). In contrast, after the administration of MCA1 during late pregnancy (d 22), there was an increase in the rate of apoptosis in both cellular compartments. The results of morphometric analysis of cervical epithelial cell apoptosis rates during pregnancy are shown in Fig. 6A. On d 5 and 10 of pregnancy, mean TUNEL LI in controls was about 2%. It dropped about 90% on d 13, and it remained lower than on d 5 and 10 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 5 and 10 of pregnancy and on d 3 postpartum did not affect the TUNEL LI. After neutralization of endogenous relaxin, the TUNEL LI increased markedly on d 13, 16, 19, and 22 of pregnancy (P 0.01), and it increased most dramatically on d 19 and 22 (P 0.05).
FIG. 5. Representative photomicrographs that illustrate the influence of neutralization of endogenous relaxin during early and late pregnancy on apoptosis of epithelial cells (A) and stromal cells (B). Arrows indicate representative labeled cells. Bar in the lower right photomicrograph, 10 μm. All figures are the same magnification.
FIG. 6. Mean TUNEL labeling index (+SEM) in cervical epithelial cells (A) and stromal cells (B) during pregnancy and early postpartum. Each bar represents the mean for six animals (groups PBS and MCA1) or four animals (group MCAF). Asterisks indicate differences in group MCA1 from PBS and MCAF controls on the same day (**, P < 0.01).
The results of morphometric analysis of cervical nonmuscle stromal cell apoptosis rates during pregnancy are shown in Fig. 6B. On d 5 and 10 of pregnancy, the mean TUNEL LI for controls was about 0.4%. It decreased about 75% by d 13 and remained lower than on d 5 and 10 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 5 and 10 of pregnancy and on d 3 postpartum did not affect the TUNEL LI. After neutralization of endogenous relaxin with MCA1 the TUNEL LI increased markedly on d 13, 16, 19, and 22 of pregnancy, and the effects were maximal on d 22 (P 0.05). There was no evidence of apoptosis in smooth muscle cells.
Ultrastructural analysis of apoptosis in cervical cells
A cell was classified as apoptotic if it contained a nucleus with condensed chromatin and/or a shrunken nucleus, as indicated by separation from the nuclear envelope. The results of electron microscopic analysis of cervical epithelial and stromal cells in the three treatment groups on d 5, 13, and 22 of pregnancy are shown in Table 1. Findings were consistent with those obtained with the TUNEL method. After neutralization of relaxin, the percentage of apoptotic epithelial and stromal cells increased on d 13 and 22 of pregnancy, but not on d 5 of pregnancy. Moreover, the influence of neutralization of endogenous relaxin appeared to be greater on d 22 than on d 13 of pregnancy. No apoptotic smooth muscle cells were observed with any of the three treatments.
TABLE 1. Electron microscopy: influence of endogenous relaxin on rate of apoptosis in cervical epithelial and stromal cells
Leukocytes and mast cells
The percentage of infiltrating leukocytes within the cervical stroma remained less than 0.5% through d 19 and increased to about 2% on d 22 of pregnancy. The percentage of mast cells was less than 0.2% of the total cervical stromal cells throughout pregnancy. MCA1 treatment had no effect on the leukocyte or mast cell populations (data not shown).
Discussion
There are two novel findings in the present study. The first new finding is that relaxin both promotes proliferation and inhibits apoptosis of cervical cells throughout the second half of pregnancy. Neutralization of relaxin reduces the rate of proliferation of stromal cells and increases the rate of apoptosis of both epithelial and stromal cells by d 13 of pregnancy. The second finding is that relaxin’s effects on these processes are far more pronounced during late pregnancy. By d 22 of pregnancy, the great majority of the proliferation of both epithelial and stromal cells is relaxin dependent. Moreover, after neutralization of relaxin, the rate of apoptosis in both epithelial and stromal cells on d 22 is as great as it is on d 5 and 10 before relaxin is secreted.
We do not know the reason(s) why relaxin plays a far greater role in promoting proliferation and inhibiting apoptosis of cervical cells during late pregnancy than on d 13 and 16. Three possibilities are suggested. First, serum relaxin levels rise progressively during the second half of pregnancy (1, 2), and the effectiveness of the hormone may be proportional to its concentration. Second, multiple forms of immunoactive relaxin, designated C1, C2, and C3, with mol wt of 60,000, 13,000, and 6,500, respectively, are detected in the serum during pregnancy (17). Sherwood and co-workers (17) demonstrated that nearly all of the relaxin immunoactivity is associated with the 60,000 mol wt C1 form on d 15 and that increasing percentages of relaxin immunoactivity are associated with the two smaller forms as pregnancy nears term. The physiological significance of that finding is unknown, but it is possible that the two smaller forms of immunoactive relaxin are the most effective. Third, relaxin’s effects on the cervix are estrogen dependent (1, 18, 19). Secretion by developing ovarian follicles increases blood levels of estrogen about 2-fold during the last 6 days of pregnancy (20), and the elevation in estrogen may increase relaxin’s effectiveness.
It is interesting to note that whereas relaxin increases the proliferation rate of stromal cells throughout the second half of rat pregnancy, it increases the proliferation rate of epithelial cells only during the last few days of pregnancy. A possible explanation of why stromal cells proliferate in response to relaxin several days before cervical epithelial cells exhibit a similar response may be that proliferation of epithelial cells is dependent upon a paracrine factor(s) derived from the stromal compartment. Other investigators have shown that cervical stroma cells are responsive to relaxin in vitro (33). Perhaps relaxin acts first on stromal cells to increase their proliferation and differentiation, and these cells then produce and secrete a paracrine factor(s) that promotes the proliferation of epithelial cells. There is a precedent for such a mechanism. Estrogen-induced proliferation of epithelial cells in the mouse uterus (34) and vagina (35) has been demonstrated to be dependent upon the stromal action of estrogen. An alternative explanation, that cervical epithelium becomes responsive to relaxin during late pregnancy as a result of up-regulation of relaxin receptor expression, cannot be ruled out. Presently, neither the identification of cervical cells that contain relaxin receptors nor the levels of relaxin receptors in rat cervical tissue during rat pregnancy are well established. Whereas an immunohistochemical procedure identified binding sites for relaxin in epithelial cells, circular smooth muscle cells, and cells associated with blood vessels in the rat cervix (36), a single study to date has identified relaxin receptor LGR7 immunoactivity in smooth muscle cells, but not in epithelium (37). Preliminary studies of gene expression employing real-time PCR of rat cervical tissue obtained at 3-d intervals from d 7 of pregnancy until d 3 postpartum found that the expression of relaxin receptor LGR7 did not differ from d 7 on any subsequent day of pregnancy or on d 3 postpartum (Kumar, J. S., and O. D. Sherwood, unpublished observations).
This study and previous reports (5, 14) indicate that it is both epithelial and nonsmooth muscle stromal cells such as fibroblasts and cells associated with blood vessels that increase within the cervix as pregnancy progresses. The relaxin-induced increase in these cells is of physiological importance, because they contribute to enlargement of the cervix (14). Burger and Sherwood (14) reported that the nearly 2-fold relaxin-dependent increase in the circumference of the cervical lumen is accompanied by a similar increase in the total number of epithelial cells per luminal circumference. Relaxin’s actions, which bring about both enlargement of the luminal circumference and increased extensibility of the cervix (6, 7, 8, 9, 12, 16), are vital for successful expulsion of fetuses at term (3, 4).
Because white blood cells do not proliferate in the cervix (38), the infiltration of these cells in sufficiently large number would lower the BrdU and TUNEL LIs of the stromal cells. This does not happen. Leukocytes comprise less than 0.5% of total cervical stromal cells throughout nearly all of pregnancy and increase to only about 2% on d 22 of pregnancy. This result is consistent with an earlier finding by Ramos et al. (39), who demonstrated a similar increase in the percentage of eosinophils within the cervix during late pregnancy. The physiological significance of the increase in leukocyte infiltration is not known, but it may protect against bacterial infection during labor and/or be involved in the process of returning the softened cervix to the pregestational state. Mast cells are also found within the cervical stroma, but their numbers are extremely low and remain unchanged throughout pregnancy.
Relaxin is not the only ovarian hormone to promote cervical growth. Besides its permissive role in allowing relaxin to promote cervical growth, estrogen by itself promotes considerable growth of the cervix in ovariectomized nonpregnant rats (19). Efforts are underway to determine the extent to which estrogen promotes cervical growth throughout the second half of rat pregnancy.
The finding in this study that endogenous relaxin has a predominant influence on the rates of cervical cell proliferation and apoptosis during late rat pregnancy is consistent with a recent finding that exogenous relaxin promotes marked growth and softening of the cervix during late rat pregnancy. In that study (40), in which rats were made relaxin deficient by administering MCA1 daily throughout the second half of pregnancy, the durations of labor and delivery were prolonged, the incidence of live pups was reduced, and delivery was incomplete. However, when the relaxin-deficient rats were administered porcine relaxin during just the antepartum period (d 20–22), birth parameters were restored to values that did not differ from those in untreated controls.
One must be mindful that marked differences exist in the actions of relaxin among species. Nevertheless, the finding that the cervix is highly responsive to relaxin during late pregnancy in rats may have clinical implications. Perhaps exogenous relaxin will prove useful at delivery in species other than rats, where circulating relaxin either is not present or is present at levels that are too low to modify the cervix effectively at term.
In conclusion, this study provides evidence that relaxin plays a major role in promoting proliferation and inhibiting apoptosis of both epithelial and stromal cells in the rat cervix during the second half of pregnancy. Moreover, this study demonstrates that relaxin’s influence on these processes is greatest during late pregnancy. The relaxin-induced increase in cervical cells during the second half of pregnancy probably plays an important role in promoting cervical growth and remodeling, thereby ensuring rapid and safe delivery at term.
Acknowledgments
We thank B. Sylavong for supervision of animal care, and the College of Medicine Document Management Center for assistance with the preparation of the manuscript. We also acknowledge Dr. Roger Shanks for his help with statistical analysis.
References
Sherwood OD 2004 Relaxin’s physiological roles and other diverse actions. Endocr Rev 25:205–234
Sherwood OD, Crnekovic VE, Gordon WL, Rutherford JE 1980 Radioimmunoassay of relaxin throughout pregnancy and during parturition in the rat. Endocrinology 107:691–698
Downing SJ, Sherwood OD 1985 The physiological role of relaxin in the pregnant rat. I. The influence of relaxin on parturition. Endocrinology 116:1200–1205
Lao Guico-Lamm M, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. II. Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts birth in intact rats. Endocrinology 123:2479–2485
Burger LL, Sherwood OD 1995 Evidence that cellular proliferation contributes to relaxin-induced growth of both the vagina and the cervix in the pregnant rat. Endocrinology 136:4820–4826
Downing SJ, Sherwood OD 1985 The physiological role of relaxin in the pregnant rat. III. The influence of relaxin on cervical extensibility. Endocrinology 116:1215–1220
Hwang J-J, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. III. Passive immunization with monoclonal antibodies throughout the second half of pregnancy reduces cervical growth and extensibility in intact rats. Endocrinology 123:2486–2490
Hwang J-J, Shanks RD, Sherwood OD 1989 Monoclonal antibodies specific for rat relaxin. IV. Passive immunization with monoclonal antibodies during the antepartum period reduces cervical growth and extensibility, disrupts birth, and reduces pup survival in intact rats. Endocrinology 125:260–266
Harkness MLR, Harkness RD 1959 Changes in the physical properties of the uterine cervix of the rat during pregnancy. J Physiol 148:524–547
Zarrow MX, Yochim J 1961 Dilation of the uterine cervix of the rat and accompanying changes during the estrous cycle, pregnancy and following treatment with estradiol, progesterone, and relaxin. Endocrinology 69:292–304
Hollingsworth M 1981 Softening of the rat cervix during pregnancy. In: Ellwood DA, Anderson ABM, eds. The cervix in pregnancy and labor. New York: Churchill Livingstone; 13–33
Downing SJ, Sherwood OD 1986 The physiological role of relaxin in the pregnant rat. IV. The influence of relaxin on cervical collagen and glycosaminoglycans. Endocrinology 118:471–479
Vasilenko P, Mead JP 1987 Growth-promoting effects of relaxin and related compositional changes in the uterus, cervix, and vagina of the rat. Endocrinology 120:1370–1376
Burger LL, Sherwood OD 1998 Relaxin increases the accumulation of new epithelial and stromal cells in the rat cervix during the second half of pregnancy. Endocrinology 139:3984–3995
Wyllie AH 1992 Apoptosis and the regulation of cell numbers in normal and neoplastic tissues: an overview. Cancer Metastasis Rev 11:95–103
Zhao S, Fields PA, Sherwood OD 2001 Evidence that relaxin inhibits apoptosis in the cervix and the vagina during the second half of pregnancy in the rat. Endocrinology 142:2221–2229
Sherwood OD, Key RH, Tarbell MK, Downing SJ 1984 Dynamic changes of multiple forms of serum immunoactive relaxin during pregnancy in the rat. Endocrinology 114:806–813
Cullen BM, Harkness RD 1960 The effect of hormones on the physical properties and collagen content of the rat’s uterine cervix. J Physiol 152:419–436
Kroc RL, Steinetz BG, Beach VL 1959 The effects of estrogens, progestagens, and relaxin in pregnant and nonpregnant laboratory rodents. Ann NY Acad Sci 75:942–980
Taya K, Greenwald GS 1981 In vivo and in vitro ovarian steroidogenesis in the pregnant rat. Biol Reprod 25:683–691
Lao Guico-Lamm M, Voss Jr EW, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. I. Production and characterization of monoclonal antibodies that neutralize rat relaxin’s bioactivity in vivo. Endocrinology 123:2472–2478
Kuenzi MJ, Sherwood OD 1992 Monoclonal antibodies specific for rat relaxin. VII. Passive immunization with monoclonal antibodies throughout the second half of pregnancy prevents development of normal mammary nipple morphology and function in rats. Endocrinology 131:1841–1847
Lee AB, Hwang J-J, Haab LM, Fields PA, Sherwood OD 1992 Monoclonal antibodies specific for rat relaxin. VI. Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts histological changes associated with cervical softening at parturition in rats. Endocrinology 130:2386–2391
Ramos JG, Varayoud J, Bosquiazzo VL, Luque EH, Mu?oz-de-Toro M 2002 Cellular turnover in the rat uterine cervix and its relationship to estrogen and progesterone receptor dynamics. Biol Reprod 67:735–742
Sherwood OD, Downing SJ, Golos TG, Gordon WL, Tarbell MK 1983 Influence of light-dark cycle on antepartum serum relaxin and progesterone immunoactivity levels and on birth in the rat. Endocrinology 113:997–1103
Golos TG, Sherwood OD 1982 Control of corpus luteum function during the second half of pregnancy in the rat: a direct relationship between conceptus number and both serum and ovarian relaxin levels. Endocrinology 111:872–878
Sheehan DC, Hrapchak BB 1980 Fixation. In: Theory and practice of histotechnology. 2nd ed. Columbus, OH: Battelle Press; 40–58
Schutte B, Reynders MMJ, Bosman FT, Blijham GH 1987 Effect of tissue fixation on anti-bromodeoxyuridine immunohistochemistry. J Histochem Cytochem 35:1343–1345
Gavrieli Y, Sherman Y, Ben-Sasson SA 1992 Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493–501
Dawes CJ 1971 Biological techniques in electron microscopy. New York: Barnes and Noble; 42
Burkett HG, Young B, Heath JW 1993 Blood and supporting/connective tissue. In: Wheater’s functional histology: a text and color atlas. New York: Churchill Livingstone; 42–75
Sanderson IR, Slavin G, Walker-Smith JA 1985 Density of mucosal mast cells in the lamina propria of the colon and terminal ileum of children. J Clin Pathol 38:771–773
Palejwala S, Stein DE, Weiss G, Monia BP, Tortoriello D, Goldsmith LT 2001 Relaxin positively regulates matrix metalloproteinase expression in human lower uterine segment fibroblasts using tyrosine kinase signaling pathway. Endocrinology 142:3405–3413
Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach KS, Taylor J, Lubahn DB, Cunha GR 1997 Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc Natl Acad Sci USA 94:6535–6540
Buchanan DL, Kurita T, Taylor JA, Lubahn DB, Cunha GR, Cooke PS 1998 Role of stromal and epithelial estrogen receptors in vaginal epithelial proliferation, stratification, and cornification. Endocrinology 139:4345–4352
Kuenzi MJ, Sherwood OD 1995 Immunohistochemical localization of specific relaxin-binding cells in the cervix, mammary glands, and nipples of pregnant rats. Endocrinology 136:1367–1373
Hsu Y-H, Nakabayashi K, Nishi S, Kumagai J, Kudo M, Sherwood OD, Hsueh AJW 2002 Activation of orphan receptors by the hormone relaxin. Science 295:671–674
Janeway CA, Travers P 1994 Basic concepts in immunology. In: Immunobiology, the immune system in health and disease. 2nd ed. New York: Current Biology; 1:1–1:43
Ramos JG, Varayoud J, Kass L, Rodriguez H, Munoz de Toro M, Montes GS, Luque EH 2000 Estrogen and progesterone modulation of eosinophilic infiltration of the rat uterine cervix. Steroids 65:409–414
Zhao S, Sherwood OD 2004 Induction of labor with RU 486 (mifepristone) in relaxin-deficient rats: antepartum administration of relaxin facilitates delivery and increases pup survival. Am J Obstet Gynecol 190:229–238(Hyung-Yul Lee1, Shuangpin)
Address all correspondence and requests for reprints to: Dr. O. D. Sherwood, Department of Molecular and Integrative Physiology, University of Illinois at Urbana-Champaign, 524 Burrill Hall, 407 South Goodwin Avenue, Urbana, Illinois 61801. E-mail: od-sherw@uiuc.edu.
Abstract
Relaxin promotes marked growth of the cervix during the second half of rat pregnancy, and this growth is accompanied by an increase in both epithelial and stromal cells. The objective of this study was to test the hypothesis that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical cells is greatest during late pregnancy in rats. The influence of neutralization of circulating relaxin by iv injection of 5 mg monoclonal antibody against rat relaxin (MCA1) was examined at 3-d intervals throughout the second half of pregnancy. Controls were injected with either 5 mg monoclonal antibody against fluorescein or 0.5 ml PBS vehicle. To evaluate cell proliferation, 5'-bromo-2-deoxyuridine was injected sc 8 h before cervixes were collected. Terminal deoxynucleotidyl transferase-mediated deoxyuridine 5'-triphosphate nick end-labeling and electron microscopy were used to detect apoptotic cells. Neutralization of relaxin with MCA1 decreased the rate of proliferation and increased the rate of apoptosis of cervical cells by d 13. However, the extent to which relaxin influenced these processes was greatest and dramatic by late pregnancy. In MCA1-treated rats on d 22 of pregnancy, the rates of proliferation of both epithelial and stromal cells were less than 20% those in controls, and the rates of apoptosis in epithelial cells and stromal cells were more than 10- and 3-fold, respectively, greater than those in controls. In conclusion, this study provides evidence that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical epithelial and stromal cells is greatest during late pregnancy.
Introduction
RELAXIN IS SECRETED by the corpora lutea throughout the second half of the 23-d rat pregnancy. The hormone is first detected in the peripheral serum on d 10 of pregnancy, and levels increase to approximately 80 ng/ml by d 20. Serum relaxin surges to levels ranging from 120–200 ng/ml on d 21 of pregnancy, then rapidly declines to about 2 ng/ml by postpartum d 2 (1, 2).
Relaxin is indispensable during pregnancy in rats. When rats were made relaxin deficient by means of either ovariectomy or neutralization of circulating relaxin with monoclonal antibody specific for rat relaxin, the duration of delivery was markedly prolonged, and about 50% of the fetuses were born dead or retained in utero (3, 4, 5). Relaxin enables rapid and safe delivery of the fetuses by promoting growth and softening of the cervix during the second half of pregnancy (6, 7).
In the rat, cervical wet weight increases about 3-fold during the second half of pregnancy (8, 9, 10, 11). Our laboratory provided evidence that approximately 40–50% of the increase in the cervical wet weight that occurs during this period is relaxin dependent (5, 7). Early reports indicated that relaxin contributes to cervical growth by increasing the extracellular content of water, collagen, hyaluronic acid, and proteoglycan (12, 13). More recent studies demonstrated that the increase in cervical wet weight is also partially attributed to relaxin-induced accumulation of new cells (5, 14). Relaxin is responsible for more than 50% of the new cells that accumulate in the cervical epithelium and stroma throughout the entire second half of pregnancy (14).
The cellular mechanism(s) by which endogenous relaxin brings about the increase in cervical cell content has received little attention. Homeostatic control of cell number is thought to result from the dynamic balance between cell proliferation and apoptosis (15). Data we obtained only on d 22 of pregnancy indicate that relaxin may influence both processes. The finding that the number of cells that contained the thymidine analog 5-bromo-2'-deoxyuridine (BrdU) was far lower in cervixes from ovariectomized rats that were devoid of circulating relaxin throughout the last 14 d of pregnancy than in cervixes from controls can be interpreted to indicate that relaxin promotes cell proliferation (14). The observation that the rate of apoptosis in cervixes obtained from rats in which endogenous relaxin was neutralized on d 19–21 was greater than that in cervixes from controls provided evidence that relaxin inhibits apoptosis during the antepartum period (16). To date, it has not been determined whether endogenous relaxin influences the rates of either cervical cell proliferation or apoptosis throughout the second half of pregnancy. Moreover, it is not known whether relaxin’s effects on the rates of these two dynamic processes for regulating cell number change as pregnancy progresses.
There are at least three reasons to think that relaxin’s effects on cervical cell proliferation and apoptosis rates may be greatest during late pregnancy. First, the rising serum levels of relaxin during this period may exert increasingly pronounced affects on both processes. Second, there is a progressive shift toward increasing ratios of 6,500 mol wt and 13,000 mol wt forms of relaxin relative to a 60,000 mol wt form in the serum, and the smaller form(s) of relaxin may be most active (17). Third, because relaxin’s effects on the cervix are estrogen dependent (18, 19), a steady increase in the serum levels of estrogen during the second half of pregnancy (20) may enhance the efficacy of relaxin’s effects on cervical cells.
Based on this information, it was hypothesized that the extent to which relaxin promotes proliferation and inhibits apoptosis of cervical cells is greatest during late pregnancy in rats. This hypothesis was examined by determining the rates of proliferation and apoptosis of cervical cells after the passive neutralization of circulating endogenous relaxin at 3-d intervals throughout the second half of pregnancy. Relaxin was neutralized with a monoclonal antibody for rat relaxin (MCA1) (21), which effectively blocks the biological actions of relaxin in pregnant rats (4, 7, 8, 22, 23). The effects of relaxin on proliferation and apoptosis rates of cervical epithelial and stromal cells were independently determined because distinct patterns of proliferation and apoptosis for each of these cellular compartments during rat pregnancy were reported (24).
Materials and Methods
Animals
Primiparous Sprague Dawley-derived rats, bred at approximately 75 d of age, were obtained from Harlan (Indianapolis, IN) on d 3 of pregnancy. The day that sperm was found in the vagina was designated d 1 of pregnancy. The animals were housed individually and maintained in a light-controlled room (lights on between 0700 and 2100 h) at a temperature of 23–25 C. Beginning on d 8, the photoperiod was advanced 8 h so that lights were on from 2100–1100 h for the remainder of pregnancy. This shift in the light schedule was previously demonstrated to synchronize the time of delivery to the early morning hours on d 23 of pregnancy, which is also designated d 1 postpartum (25). Rats were provided with Teklad 6% mouse/rat diet 7002 (Harlan/Teklad, Madison, WI) and water ad libitum. The animal experimentation described in this study was approved by the University of Illinois laboratory animal advisory committee.
On d 8 of pregnancy, rats were anesthetized with ether, and ventral laparotomy was performed to determine the number of implantation sites. Because serum relaxin levels are directly proportional to litter size in rats with five or fewer conceptuses (26), rats with less than eight implantation sites were excluded from this study.
Animal treatment for examination of relaxin’s effects on cell proliferation
Figure 1 shows the experimental design. There were three treatments: PBS control, a monoclonal antibody for fluorescein (MCAF) (4) control, and MCA1. Six animals were used for each group. Cervical tissues were collected at 3-d intervals beginning on d 7 of pregnancy and continuing until d 3 postpartum. Preliminary experiments demonstrated that neutralization of endogenous relaxin with MCA1 for 48 h had maximal inhibiting effects on cellular proliferation in both cervical epithelial and stromal cells and that administration of BrdU sc 8 h before tissue collection enabled the demonstration of treatment effects (data not shown). Accordingly, for 2 consecutive days beginning at 0900 h on d 5, 8, 11, 14, 17, and 20 of pregnancy and on d 1 postpartum, rats were injected via the tail vein with 0.5 ml PBS vehicle, 5 mg MCAF, or 5 mg MCA1. Forty hours after treatment began (8 h before tissue collection), 800 μg BrdU (Sigma-Aldrich Corp., St. Louis, MO) dissolved in 0.5 ml sterile saline (0.9% NaCl) were injected sc.
FIG. 1. Diagram of the experimental design for examining endogenous relaxin’s effects on cervical cell proliferation. See Materials and Methods for details.
Tissue collection and processing for evaluation of cell proliferation
Animals were killed, and cervical tissues were collected at 0900 h on d 7, 10, 13, 16, 19, and 22 of pregnancy and on d 3 postpartum. Cervixes were removed, trimmed of extraneous tissues, and prefixed in 10% neutral buffered formalin for 2 h. Each cervix was then bisected in cross-section to obtain cephalic and caudal halves. Fixation was continued in fresh neutral buffered formalin for a total of 24 h. After fixation, cervixes were dehydrated in an ascending series of ethanol, cleared in xylene, and embedded in paraffin (27). The cervical tissue blocks were sectioned at 5 μm thickness, mounted on positively charged slides (SuperFrost Plus, Fisher Scientific, Pittsburgh, PA), and air-dried overnight. Tissue sections from each tissue block were at least 50 μm apart to ensure that different cervical cells were analyzed. Two sections from each block were analyzed.
BrdU immunohistochemistry for detection of cell proliferation
BrdU immunostaining was performed as previously described with some modification (14, 28). Briefly, BrdU incorporation into proliferating cells was determined immunohistochemically using a mouse monoclonal antibody (clone NCL-BrdU, Vector Laboratories, Inc., Burlingame, CA) and a Vectastain Elite ABC kit (Vector Laboratories, Inc.). All incubations, unless otherwise noted, were performed at 25 C. Slides were cleared in xylene and rehydrated in a descending series of ethanol. DNA was denatured by incubation in aqueous 2 N HCl for 30 min at 37 C. After this and all other incubations, slides were rinsed in three changes of PBS (pH 7.5; 5 min/rinse). Antibody penetration was improved by incubation in 0.01% trypsin (Sigma-Aldrich Corp.) in PBS for 15 min at 37 C. Tissue sections were immersed in 3% hydrogen peroxide for 15 min to quench endogenous peroxidase activity. After incubation with blocking buffer (3% normal horse serum in PBS; Vector Laboratories, Inc.) for 30 min, anti-BrdU antibody (1:300 dilution in blocking buffer) was applied, and slides were incubated overnight (16 h) in humidified chambers at 4 C. Biotinylated horse antimouse IgG and avidin-biotin-peroxidase complex were prepared as directed (Vectastain Elite ABC kit, Vector Laboratories, Inc.). Biotinylated horse antimouse IgG (1:65 dilution in blocking buffer) was applied for 30 min. The avidin-biotin-peroxidase complex was then applied for 30 min. Antibody-binding sites were visualized using a 3,3'-diaminobenzidine peroxidase substrate (Peroxidase Substrate Kit, Vector Laboratories, Inc.) that was prepared as directed and applied to sections for 2–3 min. Slides were rinsed with distilled water for 10 min and counterstained with Ehrlich’s hematoxylin for 10 min. Slides were washed with tap water for 10 min, then dehydrated in ethanol, cleared, and coverslipped with Permount (Fisher Scientific). For negative control slides, either anti-BrdU antibody was omitted (blocking buffer only) or nonspecific mouse IgGs were substituted (1:300 in blocking buffer; Sigma-Aldrich Corp.).
Morphometric analysis of relaxin’s effects on epithelial cell and stromal cell proliferation
To determine the cellular proliferation rate of the cervical cells, the labeling index (LI) of BrdU-positive cells was determined. The LI was obtained by dividing the number of BrdU-positive cells by the total number of cells analyzed per section and multiplying by 100. Sections were examined morphometrically at x400 magnification with a BH-2 light microscope (Olympus Corp., Melville, NY) equipped with a video camera and connected to a personal computer running a Stereo Investigator program (MicroBrightField, Inc., Colchester, VT). The Stereo Investigator program automatically controls the movement of the microscope stage to permit unbiased selection of fields of analysis. Epithelial cells and stromal cells were analyzed independently. Within the stroma, nonsmooth muscle cells (primarily fibroblasts) and smooth muscle cells (both circular and longitudinal) were analyzed independently. Data were obtained from four sections (two sections from each cervical half/rat), and at least 300 epithelial cells and 500 stromal cells were analyzed per section. Thus, at least 1200 epithelial cells and 2000 stromal cells were analyzed for each of the six rats per group.
Animal treatment for examination of relaxin’s effects on apoptosis
Figure 2 shows the experimental design. There were three treatments: PBS control (n = 6), MCAF control (n = 4), and MCA1 (n = 6). Cervical tissues were collected on d 5 and at 3-d intervals beginning on d 10 of pregnancy and continuing until d 3 postpartum. Previous experiments demonstrated that neutralization of endogenous relaxin with MCA1 for 24 h had a maximal effect on apoptosis in both cervical epithelial and stromal cells (16). Accordingly, at 0900 h on d 4, 9, 12, 15, 18, and 21 of pregnancy and on d 2 postpartum, rats were injected via the tail vein with 0.5 ml PBS vehicle, 5 mg MCA1, or 5 mg MCAF.
FIG. 2. Diagram of the experimental design for examining endogenous relaxin’s effects on cervical cell apoptosis. See Materials and Methods for details.
Tissue collection and processing for evaluation of apoptosis
Animals were killed, and cervical tissues were collected at 0900 h on d 5, 10, 13, 16, 19, and 22 of pregnancy and on d 3 postpartum. Cervixes were trimmed of extraneous tissue, fixed, and sectioned as described above for the evaluation of cell proliferation.
Terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine 5'-triphosphate nick end-labeling (TUNEL) immunohistochemistry for detection of apoptosis
TUNEL in conjunction with morphometric analysis was employed as previously described (16) to detect and quantify cells undergoing apoptosis. In brief, sections were stained immunocytochemically by TUNEL using the method described by Gavrieli et al. (29). A commercial kit (ApopTag In Situ Apoptosis Detection, Serologicals Corp., Norcross, GA), which links digoxigenin-nucleotide to DNA by TdT, was used. Sections were deparaffined with xylene, rehydrated with a descending series of ethanol, incubated with proteinase K, immersed in 3% aqueous hydrogen peroxide, and then pretreated with equilibration buffer. DNA was labeled at the 3' end by incubating sections with a mixture of digoxigenin deoxynucleotide triphosphate, unlabeled deoxynucleotide triphosphate, and TdT enzyme at 37 C for 1 h. Slides were washed with PBS and incubated with antidigoxigenin antibody conjugated to peroxidase at room temperature for 30 min. Slides were washed again in PBS, incubated with 3,3'-diaminobenzidine peroxidase substrate, counterstained with methyl green, mounted, and sealed. Positive control slides were treated with deoxyribonuclease I before the labeling reaction. Negative control slides were incubated with labeling reaction solution devoid of TdT enzyme.
Morphometric analysis of relaxin’s effects on epithelial and stromal cell apoptosis
Sections were examined morphometrically at x400 magnification with a BH-2 light microscope (Olympus Corp.) equipped with a video camera and connected to a personal computer running a Stereo Investigator program (MicroBrightField, Inc.). Epithelial cells and stromal cells were analyzed independently. As with cell proliferation, nonsmooth muscle cells and smooth muscle cells within the stroma were analyzed independently. The percentage of TUNEL-labeled cells (LI) was determined. Data were obtained from four sections (50 μm apart)/rat, and at least 500 epithelial cells and 500 stromal cells/section were analyzed per section. Thus, at least 2000 cells were analyzed per rat for the epithelium, and 2000 cells were analyzed per rat for the stroma.
Ultrastructural analysis of apoptosis in cervical cells
To confirm that TUNEL labeling of cervical tissue was indeed attributable to apoptosis, transmission electron microscopy was used to determine whether comparable percentages of cells demonstrated morphological features characteristic of apoptotic cells. Cervixes were collected from two PBS-, MCAF-, and MCA1-treated rats on d 5, 13, and 22 of pregnancy. Cervixes were cut into cephalic and caudal halves, placed in Karnovsky’s fixative (30) for 24 h, washed three times with 0.1 M cacodylate buffer, and processed for electron microscopic evaluation as previously described (16). Each of the cephalic and caudal tissues was cut in half for embedding and sectioning. Sections were taken from three levels of each half. One hundred epithelial, 100 stromal, and 100 smooth muscle cells were evaluated from each section for apoptosis, i.e. condensed chromatin or shrunken nucleus as indicated by the separation of the nuclear envelope. Therefore, from two rats on each of the 3 d of the study, 2400 of each cell type were evaluated. A cell had to have a nucleus in the section to be evaluated.
Evaluation of infiltrating leukocytes and mast cells
To determine the percentage of leukocytes within the cervical stroma, tissue sections were stained with hematoxylin and eosin and examined morphometrically with a BH-2 light microscope (Olympus Corp.). Leukocytes were identified by their multilobed nucleus and distinct cytoplasm (31). No attempt was made to differentiate the leukocytes into various subtypes. At least 250 stromal cells were evaluated per section, and four sections were evaluated per rat. Thus at least 1000 stromal cells were evaluated for each of the six rats per group.
To determine the extent to which mast cells comprise the cervical stromal cell population, they were identified through the naphthol AS-D chloroacetate (3-hydroxy-2-naphthoic-o-toluidide chloroacetate) esterase reaction. Mast cell granules contain proteases, including esterases that rapidly hydrolyze the -chloroacyl esters of the naphthol AS-D chloroacetate. Chloroacetate esterase staining was carried out according to the protocol reported by Sanderson et al. with some modification (32). Briefly, deparaffinized and rehydrated sections were incubated with substrate solution at room temperature for 30 min. The substrate solution was prepared in multisteps just before incubation. First, 0.1 ml pararosaniline (4% dissolved in 2 M HCl) was mixed with 0.1 ml 4% aqueous sodium nitrite for 30–60 sec, followed by addition of 30 ml 70 mM PBS. Finally, 0.01 g naphthol AS-D chloroacetate (Sigma-Aldrich Corp.) dissolved in 1 ml N-dimethylformamide was added, and the combined solution was filtered through Whatman filter paper (Clifton, NJ) before use in the incubation as described above. After incubation was completed, sections were counterstained with hematoxylin and mounted with a coverslip as described above. For determination of the total number of mast cells per cervical cross-section, the entire cross-sectional area was examined using the Axioskop 2 microscope at x400 magnification. Four sections (two from each cervical half) were analyzed per animal. The average total number of mast cell per cervical cross-sectional area was determined from six rats per group.
Statistics
Data were expressed as the mean ± SEM. To determine whether there were significant differences in the BrdU LI and the TUNEL LI of PBS- and MCAF-treated controls on different days, they were tested using one-way ANOVA, followed by Tukey’s test. Statistical analysis of MCA1 treatment effects was assessed using two-factor ANOVA, followed by preplanned, least squares means comparison. The level of significance was set at P < 0.05.
Results
After morphometric analysis of both cell proliferation and apoptosis, no difference between PBS- and MCAF-treated controls was found on any day of treatment.
Morphometric analysis of relaxin’s effects on epithelial and stromal cell proliferation
Figure 3 contains representative photomicrographs that demonstrate the influence of neutralization of endogenous relaxin on proliferation of cervical cells during 2 d of rat pregnancy. The administration of MCA1 did not influence the rate of cell proliferation in the epithelium (Fig. 3A) or in the stroma (Fig. 3B) after its administration during the first half of pregnancy (d 7) when relaxin is not detected in the serum. In contrast, after the administration of MCA1 during late pregnancy (d 22) when relaxin levels are maximal, there was a decline in the rate of cell proliferation in both cellular compartments. The results of morphometric analysis of the cervical epithelial cell proliferation rates on 6 days of pregnancy and d 3 postpartum are shown in Fig. 4A. On d 7, 10, and 13 of pregnancy, the mean BrdU LI in controls was 11.3%. It dropped about 50% on d 16 and remained lower than on d 7, 10, and 13 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 7 and 10 of pregnancy and on d 3 postpartum, when serum relaxin is not elevated, did not affect the BrdU LI. Neutralization of endogenous relaxin with MCA1 did not influence the BrdU LI in the epithelium until d 19 of pregnancy, when there was an approximately 50% reduction in the BrdU LI relative to the control value. Neutralization of relaxin during d 20–22 caused the greatest reduction in the BrdU LI (P 0.01) and did so by nearly 90%.
FIG. 3. Representative photomicrographs that illustrate the influence of neutralization of endogenous relaxin during early and late pregnancy on proliferation of epithelial cells (A) and stromal cells (B). Arrows indicate representative labeled cells. Bar in the lower right photomicrograph, 10 μm. All figures are the same magnification.
FIG. 4. Mean BrdU labeling index (+SEM) in cervical epithelial cells (A) and stromal cells (B) during pregnancy and early postpartum. Each bar represents the mean for six animals. Asterisks indicate differences in group MCA1 from PBS and MCAF controls on the same day (**, P < 0.01).
The results of morphometric analysis of the cervical nonmuscle stromal cell proliferation rates during pregnancy are shown in Fig. 4B. On d 7 and 10 of pregnancy, the mean LI in the stroma of controls was about 0.3%. It increased dramatically by d 13 to approximately 5.5% and remained greater than on d 7 and 10 throughout the remainder of pregnancy (P < 0.01). Administration of MCA1 on d 7 and 10 of pregnancy and on d 3 postpartum, when serum relaxin is not elevated, did not affect the BrdU LI. However, after neutralization of endogenous relaxin with MCA1, the BrdU LI was reduced on d 13, 16, 19, and 22 of pregnancy (P 0.01). Moreover, regression analysis showed that the effect of neutralization of relaxin on stromal cell proliferation was increasingly pronounced as pregnancy progressed (P 0.01). The BrdU LI for cervical smooth cells was less than 1.5% throughout pregnancy, and neutralization of relaxin had no effect on cell proliferation (data not shown).
Morphometric analysis of relaxin’s effects on epithelial cell and stromal cell apoptosis
Figure 5 contains representative photomicrographs that demonstrate the influence of neutralization of endogenous relaxin on apoptosis of cervical cells during 2 d of rat pregnancy. The administration of MCA1 did not influence the rate of apoptosis in the epithelium (Fig. 5A) or in the stroma (Fig. 5B) after its administration during the first half of pregnancy (d 5). In contrast, after the administration of MCA1 during late pregnancy (d 22), there was an increase in the rate of apoptosis in both cellular compartments. The results of morphometric analysis of cervical epithelial cell apoptosis rates during pregnancy are shown in Fig. 6A. On d 5 and 10 of pregnancy, mean TUNEL LI in controls was about 2%. It dropped about 90% on d 13, and it remained lower than on d 5 and 10 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 5 and 10 of pregnancy and on d 3 postpartum did not affect the TUNEL LI. After neutralization of endogenous relaxin, the TUNEL LI increased markedly on d 13, 16, 19, and 22 of pregnancy (P 0.01), and it increased most dramatically on d 19 and 22 (P 0.05).
FIG. 5. Representative photomicrographs that illustrate the influence of neutralization of endogenous relaxin during early and late pregnancy on apoptosis of epithelial cells (A) and stromal cells (B). Arrows indicate representative labeled cells. Bar in the lower right photomicrograph, 10 μm. All figures are the same magnification.
FIG. 6. Mean TUNEL labeling index (+SEM) in cervical epithelial cells (A) and stromal cells (B) during pregnancy and early postpartum. Each bar represents the mean for six animals (groups PBS and MCA1) or four animals (group MCAF). Asterisks indicate differences in group MCA1 from PBS and MCAF controls on the same day (**, P < 0.01).
The results of morphometric analysis of cervical nonmuscle stromal cell apoptosis rates during pregnancy are shown in Fig. 6B. On d 5 and 10 of pregnancy, the mean TUNEL LI for controls was about 0.4%. It decreased about 75% by d 13 and remained lower than on d 5 and 10 throughout the remainder of pregnancy (P 0.01). Administration of MCA1 on d 5 and 10 of pregnancy and on d 3 postpartum did not affect the TUNEL LI. After neutralization of endogenous relaxin with MCA1 the TUNEL LI increased markedly on d 13, 16, 19, and 22 of pregnancy, and the effects were maximal on d 22 (P 0.05). There was no evidence of apoptosis in smooth muscle cells.
Ultrastructural analysis of apoptosis in cervical cells
A cell was classified as apoptotic if it contained a nucleus with condensed chromatin and/or a shrunken nucleus, as indicated by separation from the nuclear envelope. The results of electron microscopic analysis of cervical epithelial and stromal cells in the three treatment groups on d 5, 13, and 22 of pregnancy are shown in Table 1. Findings were consistent with those obtained with the TUNEL method. After neutralization of relaxin, the percentage of apoptotic epithelial and stromal cells increased on d 13 and 22 of pregnancy, but not on d 5 of pregnancy. Moreover, the influence of neutralization of endogenous relaxin appeared to be greater on d 22 than on d 13 of pregnancy. No apoptotic smooth muscle cells were observed with any of the three treatments.
TABLE 1. Electron microscopy: influence of endogenous relaxin on rate of apoptosis in cervical epithelial and stromal cells
Leukocytes and mast cells
The percentage of infiltrating leukocytes within the cervical stroma remained less than 0.5% through d 19 and increased to about 2% on d 22 of pregnancy. The percentage of mast cells was less than 0.2% of the total cervical stromal cells throughout pregnancy. MCA1 treatment had no effect on the leukocyte or mast cell populations (data not shown).
Discussion
There are two novel findings in the present study. The first new finding is that relaxin both promotes proliferation and inhibits apoptosis of cervical cells throughout the second half of pregnancy. Neutralization of relaxin reduces the rate of proliferation of stromal cells and increases the rate of apoptosis of both epithelial and stromal cells by d 13 of pregnancy. The second finding is that relaxin’s effects on these processes are far more pronounced during late pregnancy. By d 22 of pregnancy, the great majority of the proliferation of both epithelial and stromal cells is relaxin dependent. Moreover, after neutralization of relaxin, the rate of apoptosis in both epithelial and stromal cells on d 22 is as great as it is on d 5 and 10 before relaxin is secreted.
We do not know the reason(s) why relaxin plays a far greater role in promoting proliferation and inhibiting apoptosis of cervical cells during late pregnancy than on d 13 and 16. Three possibilities are suggested. First, serum relaxin levels rise progressively during the second half of pregnancy (1, 2), and the effectiveness of the hormone may be proportional to its concentration. Second, multiple forms of immunoactive relaxin, designated C1, C2, and C3, with mol wt of 60,000, 13,000, and 6,500, respectively, are detected in the serum during pregnancy (17). Sherwood and co-workers (17) demonstrated that nearly all of the relaxin immunoactivity is associated with the 60,000 mol wt C1 form on d 15 and that increasing percentages of relaxin immunoactivity are associated with the two smaller forms as pregnancy nears term. The physiological significance of that finding is unknown, but it is possible that the two smaller forms of immunoactive relaxin are the most effective. Third, relaxin’s effects on the cervix are estrogen dependent (1, 18, 19). Secretion by developing ovarian follicles increases blood levels of estrogen about 2-fold during the last 6 days of pregnancy (20), and the elevation in estrogen may increase relaxin’s effectiveness.
It is interesting to note that whereas relaxin increases the proliferation rate of stromal cells throughout the second half of rat pregnancy, it increases the proliferation rate of epithelial cells only during the last few days of pregnancy. A possible explanation of why stromal cells proliferate in response to relaxin several days before cervical epithelial cells exhibit a similar response may be that proliferation of epithelial cells is dependent upon a paracrine factor(s) derived from the stromal compartment. Other investigators have shown that cervical stroma cells are responsive to relaxin in vitro (33). Perhaps relaxin acts first on stromal cells to increase their proliferation and differentiation, and these cells then produce and secrete a paracrine factor(s) that promotes the proliferation of epithelial cells. There is a precedent for such a mechanism. Estrogen-induced proliferation of epithelial cells in the mouse uterus (34) and vagina (35) has been demonstrated to be dependent upon the stromal action of estrogen. An alternative explanation, that cervical epithelium becomes responsive to relaxin during late pregnancy as a result of up-regulation of relaxin receptor expression, cannot be ruled out. Presently, neither the identification of cervical cells that contain relaxin receptors nor the levels of relaxin receptors in rat cervical tissue during rat pregnancy are well established. Whereas an immunohistochemical procedure identified binding sites for relaxin in epithelial cells, circular smooth muscle cells, and cells associated with blood vessels in the rat cervix (36), a single study to date has identified relaxin receptor LGR7 immunoactivity in smooth muscle cells, but not in epithelium (37). Preliminary studies of gene expression employing real-time PCR of rat cervical tissue obtained at 3-d intervals from d 7 of pregnancy until d 3 postpartum found that the expression of relaxin receptor LGR7 did not differ from d 7 on any subsequent day of pregnancy or on d 3 postpartum (Kumar, J. S., and O. D. Sherwood, unpublished observations).
This study and previous reports (5, 14) indicate that it is both epithelial and nonsmooth muscle stromal cells such as fibroblasts and cells associated with blood vessels that increase within the cervix as pregnancy progresses. The relaxin-induced increase in these cells is of physiological importance, because they contribute to enlargement of the cervix (14). Burger and Sherwood (14) reported that the nearly 2-fold relaxin-dependent increase in the circumference of the cervical lumen is accompanied by a similar increase in the total number of epithelial cells per luminal circumference. Relaxin’s actions, which bring about both enlargement of the luminal circumference and increased extensibility of the cervix (6, 7, 8, 9, 12, 16), are vital for successful expulsion of fetuses at term (3, 4).
Because white blood cells do not proliferate in the cervix (38), the infiltration of these cells in sufficiently large number would lower the BrdU and TUNEL LIs of the stromal cells. This does not happen. Leukocytes comprise less than 0.5% of total cervical stromal cells throughout nearly all of pregnancy and increase to only about 2% on d 22 of pregnancy. This result is consistent with an earlier finding by Ramos et al. (39), who demonstrated a similar increase in the percentage of eosinophils within the cervix during late pregnancy. The physiological significance of the increase in leukocyte infiltration is not known, but it may protect against bacterial infection during labor and/or be involved in the process of returning the softened cervix to the pregestational state. Mast cells are also found within the cervical stroma, but their numbers are extremely low and remain unchanged throughout pregnancy.
Relaxin is not the only ovarian hormone to promote cervical growth. Besides its permissive role in allowing relaxin to promote cervical growth, estrogen by itself promotes considerable growth of the cervix in ovariectomized nonpregnant rats (19). Efforts are underway to determine the extent to which estrogen promotes cervical growth throughout the second half of rat pregnancy.
The finding in this study that endogenous relaxin has a predominant influence on the rates of cervical cell proliferation and apoptosis during late rat pregnancy is consistent with a recent finding that exogenous relaxin promotes marked growth and softening of the cervix during late rat pregnancy. In that study (40), in which rats were made relaxin deficient by administering MCA1 daily throughout the second half of pregnancy, the durations of labor and delivery were prolonged, the incidence of live pups was reduced, and delivery was incomplete. However, when the relaxin-deficient rats were administered porcine relaxin during just the antepartum period (d 20–22), birth parameters were restored to values that did not differ from those in untreated controls.
One must be mindful that marked differences exist in the actions of relaxin among species. Nevertheless, the finding that the cervix is highly responsive to relaxin during late pregnancy in rats may have clinical implications. Perhaps exogenous relaxin will prove useful at delivery in species other than rats, where circulating relaxin either is not present or is present at levels that are too low to modify the cervix effectively at term.
In conclusion, this study provides evidence that relaxin plays a major role in promoting proliferation and inhibiting apoptosis of both epithelial and stromal cells in the rat cervix during the second half of pregnancy. Moreover, this study demonstrates that relaxin’s influence on these processes is greatest during late pregnancy. The relaxin-induced increase in cervical cells during the second half of pregnancy probably plays an important role in promoting cervical growth and remodeling, thereby ensuring rapid and safe delivery at term.
Acknowledgments
We thank B. Sylavong for supervision of animal care, and the College of Medicine Document Management Center for assistance with the preparation of the manuscript. We also acknowledge Dr. Roger Shanks for his help with statistical analysis.
References
Sherwood OD 2004 Relaxin’s physiological roles and other diverse actions. Endocr Rev 25:205–234
Sherwood OD, Crnekovic VE, Gordon WL, Rutherford JE 1980 Radioimmunoassay of relaxin throughout pregnancy and during parturition in the rat. Endocrinology 107:691–698
Downing SJ, Sherwood OD 1985 The physiological role of relaxin in the pregnant rat. I. The influence of relaxin on parturition. Endocrinology 116:1200–1205
Lao Guico-Lamm M, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. II. Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts birth in intact rats. Endocrinology 123:2479–2485
Burger LL, Sherwood OD 1995 Evidence that cellular proliferation contributes to relaxin-induced growth of both the vagina and the cervix in the pregnant rat. Endocrinology 136:4820–4826
Downing SJ, Sherwood OD 1985 The physiological role of relaxin in the pregnant rat. III. The influence of relaxin on cervical extensibility. Endocrinology 116:1215–1220
Hwang J-J, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. III. Passive immunization with monoclonal antibodies throughout the second half of pregnancy reduces cervical growth and extensibility in intact rats. Endocrinology 123:2486–2490
Hwang J-J, Shanks RD, Sherwood OD 1989 Monoclonal antibodies specific for rat relaxin. IV. Passive immunization with monoclonal antibodies during the antepartum period reduces cervical growth and extensibility, disrupts birth, and reduces pup survival in intact rats. Endocrinology 125:260–266
Harkness MLR, Harkness RD 1959 Changes in the physical properties of the uterine cervix of the rat during pregnancy. J Physiol 148:524–547
Zarrow MX, Yochim J 1961 Dilation of the uterine cervix of the rat and accompanying changes during the estrous cycle, pregnancy and following treatment with estradiol, progesterone, and relaxin. Endocrinology 69:292–304
Hollingsworth M 1981 Softening of the rat cervix during pregnancy. In: Ellwood DA, Anderson ABM, eds. The cervix in pregnancy and labor. New York: Churchill Livingstone; 13–33
Downing SJ, Sherwood OD 1986 The physiological role of relaxin in the pregnant rat. IV. The influence of relaxin on cervical collagen and glycosaminoglycans. Endocrinology 118:471–479
Vasilenko P, Mead JP 1987 Growth-promoting effects of relaxin and related compositional changes in the uterus, cervix, and vagina of the rat. Endocrinology 120:1370–1376
Burger LL, Sherwood OD 1998 Relaxin increases the accumulation of new epithelial and stromal cells in the rat cervix during the second half of pregnancy. Endocrinology 139:3984–3995
Wyllie AH 1992 Apoptosis and the regulation of cell numbers in normal and neoplastic tissues: an overview. Cancer Metastasis Rev 11:95–103
Zhao S, Fields PA, Sherwood OD 2001 Evidence that relaxin inhibits apoptosis in the cervix and the vagina during the second half of pregnancy in the rat. Endocrinology 142:2221–2229
Sherwood OD, Key RH, Tarbell MK, Downing SJ 1984 Dynamic changes of multiple forms of serum immunoactive relaxin during pregnancy in the rat. Endocrinology 114:806–813
Cullen BM, Harkness RD 1960 The effect of hormones on the physical properties and collagen content of the rat’s uterine cervix. J Physiol 152:419–436
Kroc RL, Steinetz BG, Beach VL 1959 The effects of estrogens, progestagens, and relaxin in pregnant and nonpregnant laboratory rodents. Ann NY Acad Sci 75:942–980
Taya K, Greenwald GS 1981 In vivo and in vitro ovarian steroidogenesis in the pregnant rat. Biol Reprod 25:683–691
Lao Guico-Lamm M, Voss Jr EW, Sherwood OD 1988 Monoclonal antibodies specific for rat relaxin. I. Production and characterization of monoclonal antibodies that neutralize rat relaxin’s bioactivity in vivo. Endocrinology 123:2472–2478
Kuenzi MJ, Sherwood OD 1992 Monoclonal antibodies specific for rat relaxin. VII. Passive immunization with monoclonal antibodies throughout the second half of pregnancy prevents development of normal mammary nipple morphology and function in rats. Endocrinology 131:1841–1847
Lee AB, Hwang J-J, Haab LM, Fields PA, Sherwood OD 1992 Monoclonal antibodies specific for rat relaxin. VI. Passive immunization with monoclonal antibodies throughout the second half of pregnancy disrupts histological changes associated with cervical softening at parturition in rats. Endocrinology 130:2386–2391
Ramos JG, Varayoud J, Bosquiazzo VL, Luque EH, Mu?oz-de-Toro M 2002 Cellular turnover in the rat uterine cervix and its relationship to estrogen and progesterone receptor dynamics. Biol Reprod 67:735–742
Sherwood OD, Downing SJ, Golos TG, Gordon WL, Tarbell MK 1983 Influence of light-dark cycle on antepartum serum relaxin and progesterone immunoactivity levels and on birth in the rat. Endocrinology 113:997–1103
Golos TG, Sherwood OD 1982 Control of corpus luteum function during the second half of pregnancy in the rat: a direct relationship between conceptus number and both serum and ovarian relaxin levels. Endocrinology 111:872–878
Sheehan DC, Hrapchak BB 1980 Fixation. In: Theory and practice of histotechnology. 2nd ed. Columbus, OH: Battelle Press; 40–58
Schutte B, Reynders MMJ, Bosman FT, Blijham GH 1987 Effect of tissue fixation on anti-bromodeoxyuridine immunohistochemistry. J Histochem Cytochem 35:1343–1345
Gavrieli Y, Sherman Y, Ben-Sasson SA 1992 Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493–501
Dawes CJ 1971 Biological techniques in electron microscopy. New York: Barnes and Noble; 42
Burkett HG, Young B, Heath JW 1993 Blood and supporting/connective tissue. In: Wheater’s functional histology: a text and color atlas. New York: Churchill Livingstone; 42–75
Sanderson IR, Slavin G, Walker-Smith JA 1985 Density of mucosal mast cells in the lamina propria of the colon and terminal ileum of children. J Clin Pathol 38:771–773
Palejwala S, Stein DE, Weiss G, Monia BP, Tortoriello D, Goldsmith LT 2001 Relaxin positively regulates matrix metalloproteinase expression in human lower uterine segment fibroblasts using tyrosine kinase signaling pathway. Endocrinology 142:3405–3413
Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach KS, Taylor J, Lubahn DB, Cunha GR 1997 Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc Natl Acad Sci USA 94:6535–6540
Buchanan DL, Kurita T, Taylor JA, Lubahn DB, Cunha GR, Cooke PS 1998 Role of stromal and epithelial estrogen receptors in vaginal epithelial proliferation, stratification, and cornification. Endocrinology 139:4345–4352
Kuenzi MJ, Sherwood OD 1995 Immunohistochemical localization of specific relaxin-binding cells in the cervix, mammary glands, and nipples of pregnant rats. Endocrinology 136:1367–1373
Hsu Y-H, Nakabayashi K, Nishi S, Kumagai J, Kudo M, Sherwood OD, Hsueh AJW 2002 Activation of orphan receptors by the hormone relaxin. Science 295:671–674
Janeway CA, Travers P 1994 Basic concepts in immunology. In: Immunobiology, the immune system in health and disease. 2nd ed. New York: Current Biology; 1:1–1:43
Ramos JG, Varayoud J, Kass L, Rodriguez H, Munoz de Toro M, Montes GS, Luque EH 2000 Estrogen and progesterone modulation of eosinophilic infiltration of the rat uterine cervix. Steroids 65:409–414
Zhao S, Sherwood OD 2004 Induction of labor with RU 486 (mifepristone) in relaxin-deficient rats: antepartum administration of relaxin facilitates delivery and increases pup survival. Am J Obstet Gynecol 190:229–238(Hyung-Yul Lee1, Shuangpin)