| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Anatomy and Cell Biology, East Carolina University School of Medicine, Greenville, North Carolina 27858
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
growth factors, implantation, placenta, relaxin, trophoblast
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
O'Byrne et al. [11] first reported serum relaxin concentrations in the hamster during the latter half of pregnancy using a homologous porcine relaxin RIA. Subsequently, Steinetz et al. [12] used this assay to determine the primary source of relaxin in the hamster and to estimate the amount of hormone present in the placenta during late pregnancy. These investigators noted that use of the porcine relaxin RIA provided relative but not accurate measurement of relaxin concentrations in blood and tissue extracts from this species. The hamster is the only rodent in which a fetal (placental) source of relaxin has been identified (reviewed in [1]). Given the high reproductive capacity and short gestation length in the hamster, this species may be useful in evaluating mechanisms that control relaxin expression by the placenta provided a highly specific and sensitive RIA is available.
We have purified hamster relaxin from chorioallantoic placental extracts but recovery from that tissue was extremely low [13]. Given the low recovery of relaxin from the hamster placenta, chemical or biological synthesis (expression in transformed bacteria or yeast) were the only options available for procurement of adequate hormone to develop a homologous RIA for the hamster. Human [9], gorilla [14], rhesus monkey [14], dog [10], and rat [15] relaxins have been chemically synthesized, and highly specific and sensitive RIAs have been developed for the human [9] and dog [10] hormones using these reagents. In a previous study [16], the complete nucleotide and amino acid sequence of the coding region of hamster relaxin was determined. Subsequently, as reported herein, these data were utilized to chemically synthesize hamster relaxin using F-MOC (N-(9-fluorenyl)methoxycarbonyl) methodology.
Here, we describe the biochemical evaluation of chemically synthesized hamster relaxin and the development of a highly specific and sensitive RIA using the synthetic hormone. Maternal blood and tissue hormone concentrations throughout the latter half of pregnancy and fetal blood hormone concentrations during late pregnancy were measured using the described assay. Cellular localization and synthesis of hamster relaxin was evaluated using a specific antibody and cDNA probe.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female golden (Syrian) hamsters (Charles River, Kingston, NY) were maintained in the Department of Comparative Medicine on a 14L:10D (lights-on at 0600 h) schedule. Animals were checked daily between 1900 and 2000 h for estrus, as determined by expression of lordosis in the presence of a male. To obtain pregnant females, animals in estrus were housed overnight with a male, and the following day was designated Day 1 of gestation. Pregnant hamsters were anesthetized with Metofane (Pitman-Moore, Mudelein, IL), and blood samples (0.5 ml) were collected from the retro-orbital plexus using heparinized capillary tubes. Blood samples were collected in the morning (0800–1000 h) on Days 6, 8, 10, 12, 14, 15, and 16 of gestation and on Days 1 and 5 postpartum (n = 8 animals/day). Additional samples were collected in the evening on Day 15 of gestation (15 pm; 2000–2200 h) from another group of animals (n = 8). Plasma was stored at -20°C until assayed for relaxin content. For determining relaxin tissue concentrations, implantation sites (Days 6, 8, and 10) and chorioallantoic placentae (Days 12, 14, 15 am, and 15 pm) were recovered from animals (n = 3 animals/day) that were anesthetized and killed by cervical dislocation. Oviductal tissues were recovered on each day of the estrous cycle, and ovaries were recovered on Days 12 and 15 of gestation. Tissues were quick frozen in liquid nitrogen and stored at -70°C or fixed in Bouin solution and processed for paraffin embedment.
Fetal Samples
On Day 15 of pregnancy, conceptuses (n = 3 litters) were removed from the uterus, and individual fetuses were dissected free of the chorioallantoic placenta and placental membranes and rinsed in sterile saline. Immediately following decapitation with clean scissors, fetal trunk blood was recovered using a 100-µl glass capillary tube rinsed in heparin (100 units/ml of saline). Because of the small volume recovered, blood samples from several fetuses (two or three) were pooled. All animal maintenance and handling was in accordance with university guidelines, and procedures were conducted with the approval of the university's Animal Care and Use Committee.
Preparation of Crude Extract
All implantation sites from each litter for Days 6, 8, and 10 of pregnancy were pooled, and multiple chorioallantoic placentas from each animal were pooled for the remaining days. The aqueous extraction solution (10 mM Tris-HCl, 0.15 M NaCl, 0.1% Triton-X, pH 9.0) contained PMSF (30 µg/ml), sodium EDTA (10 mM), leupeptin (0.5 µg/ml), pepstatin (0.7 µg/ml), and sodium azide (0.02%) to inhibit proteolysis. All extraction steps were conducted at 4°C. Frozen tissues were homogenized (Polytron; Brinkmann Instruments, Westbury, NY) in extraction solution (10 ml/g tissue), and the extract was centrifuged (8000 x g) for 1 h. The supernatant was decanted and stored at -20°C until assayed for relaxin. Prior unpublished results indicated that this extraction method yielded approximately 10-fold greater recovery of relaxin immunoactivity from small quantities of placental tissue than the method of Griss et al. [17].
Synthetic Relaxin
Relaxin was synthesized (Dr. C. Schwabe, Department of Biochemistry, Medical University of South Carolina, SC) using F-MOC chemistry as previously described [10]. Prior to synthesis, the probable structure of hamster relaxin was determined by examining the deduced amino acid sequence of the prohormone [16] for likely prohormone convertase cleavage sites, which would remove the c-peptide and yield the sequence of the individual chains. The synthesized hormone was constructed according to the following sequences: a-chain, YTSIYMSHQC CFRGCSRRSL TAAC; b-chain, RVTKEWLDEV IHVCGREYVR ATLDICAATV GLEAPPL.
Evaluation of Synthetic Hamster Relaxin
The molecular weight and isoelectric point of synthetic hamster relaxin were determined by SDS-PAGE and isoelectric focusing (IEF), respectively, as previously described [13]. Biological activity of the synthetic hormone was measured with the mouse pubic ligament assay [18] using five mice/dose. Mice (CD-1) weighing 14–17 g were primed (s.c.) with estradiol cypionate (5 µg; Upjohn, Kalamazoo, MI), and 7 days later synthetic hamster relaxin (1, 5, 10 µg) or purified porcine relaxin (0.25, 0.5, 1 µg; M.J. Fields, University of Florida, Gainesville, FL) was administered by s.c. injection. Twenty-four hours after vehicle (1% aqueous benzopurpurin 4B) or relaxin treatment, animals were killed and the interpubic ligament was exposed by blunt dissection and transilluminated, and the distance between the pubic bones was measured. Values were evaluated by ANOVA, and individual means were compared using the Neuman-Keuls procedure.
Hamster Relaxin RIA
Relaxin was measured in plasma and tissue extracts using a homologous hamster relaxin RIA. Synthetic hamster relaxin was iodinated with 125I-Na (Amersham Corp., Arlington Heights, IL) using the method of Jockenhovel et al. [19]. Specific activity of the iodinated hormone was 45–60 µCi/µg.
Antiserum to synthetic hamster relaxin was generated in rabbits by intradermal injection of hormone (100 µg) mixed with complete Freund adjuvant (diluted 1:5 with incomplete adjuvant; final concentration of 0.01 mg mycobacterium/ml) at multiple sites in the shoulder area. Rabbits were given two monthly booster injections (incomplete adjuvant) followed by additional booster injections 2 and 5 mo later. The antiserum used for these studies was collected after the final booster.
Plasma samples from eight animals were assayed at each gestational time point. In the RIA, 20 and 200 µl of each sample of maternal plasma (diluted 1:25 in assay buffer) were assayed in triplicate. Tissue extracts from three animals were assayed at each time point. In the RIA, 200, 50, and 10 µl of extract were assayed in triplicate on Days 6, 8, and 10, respectively. Twenty microliters of extract (diluted 1:25 in assay buffer) was used for the remaining gestational time points. Because of a significant depression in tracer binding when 25 µl of male plasma was included in the assay, a volume of male serum equal to the volume of fetal serum being measured (25 µl) was added to standard curve tubes to correct for the nonspecific serum effect. The assay buffer contained 20 mM barbital, 0.15 M NaCl, 0.029% (w/v) EDTA, and 0.25% (w/v) BSA and was adjusted to pH 8.0. For the standard curve, synthetic relaxin was dissolved in buffer and diluted sequentially, and 0.2 ml was added to each assay standard tube (19.5 pg to 10.0 ng). The antisera (N80) to synthetic relaxin (0.1 ml; 1:30 000 dilution) and 125I-labeled relaxin (0.1 ml; 10 000 cpm) were added, and the assay was held at 4°C for 24 h. This dilution of antiserum precipitated 35–40% of the added tracer. Rabbit IgG (0.05 ml; 500 µg/ml; Sigma, St. Louis, MO) and goat anti-rabbit immunoglobulin-G (1:7 dilution; Scantibodies Laboratories, Sante, CA) were added, and the assay was incubated for an additional 24 h. After incubation, tubes were centrifuged at 2000 x g for 30 min and the radiolabeled tracer in each pellet was counted.
The N80 antiserum did not show significant cross-reactivity with insulin or proinsulin (data not shown), and the minimal detectable concentration of hamster relaxin was 9.75 pg/tube. The working range of the assay was 39.5–2500 pg/tube, and intra- and interassay coefficients of variation were 12.7% and 11.3%, respectively. Increasing quantities of diluted (1:25) plasma and tissue extract produced binding plots that were parallel to the standard curve (Fig. 1). Tracer binding was not depressed when male (Fig. 1) and nonpregnant female (data not shown) hamster serum (diluted 1:25) were assayed, although there was significant depression when increasing volumes of undiluted male plasma were added to the assay (data not shown). Rat relaxin (O.D. Sherwood, University of Illinois, Urbana-Champaign, IL) was not detected when added to the assay in amounts up to 2500 pg (Fig. 1). Maternal plasma and tissue relaxin concentrations were log transformed because of heterogeneity of variance and analyzed by ANOVA for effect of day of gestation. Comparisons among days of pregnancy were tested by the Neuman-Keuls procedure.
|
Immunocytochemistry and In Situ Hybridization
Implantation sites (Days 8 and 10 of pregnancy), chorioallantoic placentas (Days 12, 14, and 15 of pregnancy), ovaries (Days 12 and 15 of pregnancy), and oviducts (each day of the 4-day estrous cycle) were collected for immunocytochemistry and in situ hybridization histochemistry (ISHH). Cellular localization of relaxin in placental, ovarian, and oviduct tissues was performed as previously described [20] using the N80 antibody at a dilution of 1:20 000. Cellular localization of relaxin mRNA in chorioallantoic placentas (Days 10, 12, and 15) was performed as previously described [21] using a 33-base oligonucleotide designed from the published sequence of the b-chain of hamster relaxin [16]. Hybridization temperature was 42°C.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The molecular weight and isoelectric point of synthetic hamster relaxin, using SDS-PAGE and IEF-PAGE, were identical to that of hamster relaxin purified from the placenta [13] (Fig. 2). The mouse interpubic ligament bioassay is an objective and specific method for evaluating a variety of relaxin preparations [12, 18]; however, the precision of the assay is low, with 20 animals/dose required to achieve estimates with limits of error <50% [22]. Accordingly, the biological activity data obtained in this study are an approximation. Interpubic ligament distance was greater than that of controls (P 
0.05) at all quantities of synthetic relaxin tested (Fig. 3).
|
|
Maternal Plasma Relaxin Concentrations
Plasma relaxin concentrations on Day 6 of pregnancy were at or slightly above the minimal detectable value for the assay (Fig. 4), and on Day 8 only four of eight animals had hormone levels within the working range of the assay (39.5–2500 pg/tube). By Day 10, relaxin was readily detectable in the plasma of all animals (mean ± SEM: 69.6 ± 13.4 ng/ml), and plasma concentrations increased steadily (P < 0.05) to Day 14 (373.1 ± 82.6 ng/ml). Thereafter, plasma relaxin concentrations rose dramatically (P < 0.05) to maximum concentrations in the evening of Day 15 (826.0 ± 124.0 ng/ml). On Day 16 (day of parturition), mean plasma concentrations decreased; however, values for individual animals were related to whether parturition and delivery of placentas was completed at the time of sampling. Low concentrations of hormone were detected in some animals on Day 1, but no relaxin was detected by Day 5 postpartum.
|
Implantation Site/Placental Tissue Relaxin Concentrations
Relaxin was detected in implantation site tissues recovered on Day 6 of pregnancy (17.4 ± 1.3 ng/g tissue), and concentrations increased steadily (P < 0.05) to Day 12 (Fig. 3). Chorioallantoic placental tissue concentrations remained elevated through the morning of Day 15 but were reduced (P < 0.05) by the afternoon of Day 15.
Fetal Plasma Relaxin Concentrations
Day 15 fetal plasma sample pools (n = 8) from three litters were assayed for relaxin immunoactivity (25 µl/assay tube). Mean plasma concentrations were 5.89 ng/ml, with a range of 3.9–9.5 ng/ml.
Cellular Localization of Relaxin Protein and mRNA
Relaxin localization was restricted to primary and secondary trophoblast giant cells in the trophospongium layer of the Day 15 placenta (Fig. 5, A and D) and in giant cells invading the decidua on Day 10 (Fig. 5, B and C). Contrary to a previous report from this laboratory, relaxin was not detected in endometrial granulocytes surrounding the decidual sheathed arteries on Day 10 (Fig. 5, C and F). This expression pattern was confirmed by ISHH. Relaxin was not detected by immunocytochemistry in oviducts of cycling hamsters or in ovaries of pregnant hamsters (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
O'Byrne et al. [11] were the first to report circulating concentrations of relaxin in the hamster using a homologous porcine RIA. Although their assay did not quantitatively measure hamster relaxin, it provided surprisingly accurate data concerning the pattern of circulating hormone in this species. These authors reported that relaxin is first detected on Day 8 of pregnancy and then increases dramatically to Day 14 in parallel with development of the chorioallantoic placenta. Concentrations increased 3-fold on Day 15 (peak value of 29 ng/ml) before rapidly falling to low levels on Day 1 postpartum. The homologous hamster relaxin RIA reported in the current study revealed that the pattern of circulating relaxin in pregnant hamsters was similar to that reported by O'Byrne et al. [11]; however, peak concentrations in the afternoon of Day 15 averaged 826 ng/ml, which far exceeds blood concentrations of this hormone in any rodent species reported to date.
In the current study, low quantities of relaxin (approximately 1% of maternal values) were detected in pooled samples of blood from Day 15 fetuses, raising the possibility of a role for this hormone in the fetus. The source of this hormone probably is isolated aggregates of cytotrophoblast cells and relaxin-containing secondary giant trophoblast cells (GTCs) that appear within the placental labyrinth beginning on Day 15 [23]. Ogren and Talamantes [24] reported that mouse placental lactogen-II, also synthesized by GTCs present in the labyrinth during late pregnancy, is present in fetal serum at 10% of maternal values.
The significance of high maternal plasma concentrations of relaxin for the physiology of pregnancy in the hamster is unknown. Unemori et al. [25] reported that relaxin stimulated expression of vascular endothelial growth factor (VEGF) by human endometrial tissues in vitro and that relaxin treatment intensified menstrual flow in women, suggesting that the relaxin-stimulated increase in VEGF caused neovascularization of the endometrial lining. Treatment of ischemic wounds in rats with relaxin has also stimulated angiogenesis, presumably because of the observed increase in VEGF and basic fibroblast growth factor [26]. These findings suggest that relaxin could be important in the neovascularization associated with implantation, decidualization, and placental development especially in animals with a fetal source (placenta) of relaxin. Relaxin inhibits apoptosis in epithelial and stromal tissues of the reproductive tract [27] and modulates matrix metalloprotease activity [28–30]. These data suggest an important role for relaxin in the reproductive tract tissue remodeling that accompanies pregnancy. The relatively short gestation period of the hamster (16 days) versus that for the rat (22 days) and mouse (19 days) requires completion of many of the events associated with successful pregnancy and parturition within a reduced time. Perhaps increased local and circulating concentrations of relaxin are required. The role of relaxin in the fetus during late pregnancy remains to be determined.
Concentrations of relaxin extracted from implantation site and chorioallantoic placental tissues mirrored changes in circulating levels of the hormone through the morning of Day 15. Subsequently, tissue relaxin concentrations decreased 40% by the evening of Day 15, whereas plasma concentrations increased a similar amount over the same period, suggesting release of stored hormone by the placenta. In the corpus luteum of the rat, relaxin is stored in membrane-bounded cytoplasmic secretory granules following synthesis [31]. Ultrastructural data are consistent with the hypothesis that the dramatic antepartum increases in circulating relaxin in this species are the consequence of luteolysis and the release of stored hormone [32, 33], although direct evidence for this process is lacking. In contrast, Johns and Renegar reported that GTCs (source of relaxin in the hamster) do not contain secretory granules and suggested that relaxin produced in the hamster placenta is released soon after synthesis, which is typical of peptides processed through the constitutive secretory pathway [34]. This hypothesis is supported by the fact that hamster placental trophoblast cells do not contain prohormone convertase 1/3 (unpublished data), a prohormone-processing enzyme associated with secretory granules containing relaxin [35] and insulin [36]. Thus, increased transcription of the relaxin gene and/or translation of the relaxin mRNA probably are responsible for the antepartum increase in serum relaxin in the hamster. Significance of the observed decrease in tissue relaxin during the antepartum period is not apparent.
Localization of mRNA to primary and secondary GTCs confirms synthesis of relaxin by these cells in the hamster. Failure to detect relaxin protein or mRNA in endometrial granulocytes surrounding the sheathed arteries of the Day 10 decidua was in contrast to results of a previous study [20] involving a polyclonal antibody to porcine relaxin. Similarly, in another study involving a different polyclonal antibody to porcine relaxin, relaxin was reported to be expressed in the oviduct of the cycling hamster [37]. We were unable to confirm that observation in the present study. These findings again demonstate the necessity of using species-specific reagents when studying relaxin synthesis and secretion in the hamster.
The hamster is the only rodent identified to date that has a fetal (placenta) rather than a maternal (corpus luteum, uterus) source of relaxin. Accordingly, this species, with its short gestation time and relatively high endogenous levels of relaxin during pregnancy, offers a unique opportunity to examine the structure, function, and expression of relaxin from a fetal tissue. In the current study, reliable information concerning the expression of relaxin in the hamster was gathered through the use of immunological and molecular tools. These tools will facilitate further study of the factors that control relaxin gene expression and the possible role of this hormone in fetal development.
|
FOOTNOTES |
|---|
1 This work was supported by grant HD-23481 from the National Institutes of Health. 
2 Correspondence: Randall H. Renegar, Department of Anatomy and Cell Biology, Brody School of Medicine, East Carolina University, 600 Moye Blvd., Greenville, NC 27858. FAX: 252 816 2850; renegarra{at}mail.ecu.edu
Accepted: February 14, 2002.
Received: October 3, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. J. de Haas van Dorsser, W. F. Swanson, S. Lasano, and B. G. Steinetz Development, Validation, and Application of a Urinary Relaxin Radioimmunoassay for the Diagnosis and Monitoring of Pregnancy in Felids Biol Reprod, June 1, 2006; 74(6): 1090 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. T. Nguyen and C. W. Dessauer Relaxin Stimulates Protein Kinase C {zeta} Translocation: Requirement for Cyclic Adenosine 3',5'-Monophosphate Production Mol. Endocrinol., April 1, 2005; 19(4): 1012 - 1023. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. D. Sherwood Relaxin's Physiological Roles and Other Diverse Actions Endocr. Rev., April 1, 2004; 25(2): 205 - 234. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, n = 8 animals/day) and implantation site or chorioallantoic placental tissue (
, n = 3 animals/day) of hamsters during the latter half of gestation