| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Howard Florey Institute
b Department of Zoology, University of Melbourne, Victoria 3010, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
corpus luteum, parturition, pregnancy, relaxin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The tammar wallaby is a monovular, seasonally breeding marsupial with a prolonged phase of lactation. A single corpus luteum (CL) is present in 1 ovary for the length of gestation, and a new CL forms in the contralateral ovary after ovulation between 1 and 2 days postpartum. In the presence of a pouch young, the conceptus develops to a unilaminar blastocyst of approximately 100 cells and then enters lactational diapause. The new CL also becomes quiescent. Loss or removal of the pouch young terminates lactational diapause and both the quiescent blastocyst and CL resume development, with births expected 26 or 27 days later [7]. Unlike most mammals, the lifespan of the CL is not extended by the presence of a conceptus, and the luteal phase in the nonpregnant cycle of unmated animals is similar in duration to the length of gestation. Circulating progesterone and estrogen concentrations do not differ between the 2 reproductive cycles [7], but no study has compared ovarian peptide synthesis in pregnant and nonpregnant marsupials.
Early studies demonstrated that crude extracts of corpora lutea from the pregnant tammar wallaby and brushtail possum, Trichosurus vulpecula, contain bioactive relaxin [8, 9]. Maximum bioactivity was observed in the second half of pregnancy, with a marked decrease postpartum [9]. Both the CL and placenta are sites of relaxin mRNA expression during pregnancy in the tammar but transcripts are considerably more abundant in the CL [3, 10]. Within the tammar CL, relaxin gene expression is highest in early and mid pregnancy, it is reduced at term, and is absent postpartum [3]. Antisera to porcine relaxin were used to show immunoreactive relaxin localized to membrane-bound, electron-dense granules in the luteal cell cytoplasm [11]. A large increase in immunoreactive relaxin in the tammar CL is observed from Day 22 of the 26-day gestation period, but does not coincide with the increase in relaxin gene expression [3]. It was suggested that prorelaxin is stored in cytoplasmic granules in the CL during the early stages of gestation and that further processing of the prohormone to the mature relaxin peptide occurs toward the end of gestation when the hormone is released. Relaxin production by the CL of unmated tammars at equivalent stages of the luteal phase in the nonpregnant cycle has not been examined.
The objective of this study was to isolate and purify prorelaxin or mature relaxin from the tammar wallaby CL. Analysis of the molecular weights, amino acid composition, and biological properties was then completed for comparison with other species. To test the hypothesis that ovarian prorelaxin is converted to mature relaxin in the later stages of gestation, the presence of 18-kDa prorelaxin and 6-kDa relaxin peptides was examined in CL extracts throughout gestation. Finally, we investigated whether or not relaxin synthesis in the tammar wallaby is specific to pregnancy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experiments were conducted with approval from the University of Melbourne Animal Experimentation and Ethics Committees according to the National Health and Medical Research Council's Guidelines for the Use of Animals in Biomedical Research and Guidelines for the Use of Native Fauna in Biomedical Research. Pregnant tammar wallabies at different stages of the 26-day gestation period were shot in the wild on Kangaroo Island (South Australia) with approval from the South Australia National Parks and Wildlife Services (permit A23892-2). Corpora lutea were dissected from the ovaries and stages of pregnancy were determined from blastocyst and vesicle diameters, number of somites in early embryos, or fetal crown rump and head lengths [12]. Corpora lutea were also collected from 3 unmated tammars between Days 21 and 25 of the 27-day nonpregnant cycle. Tissues were snap-frozen in liquid nitrogen and stored at -20°C until used. Immunohistochemistry was performed on whole ovaries obtained between Days 21 and 25 of gestation and 1 day postpartum. Ovaries were fixed in 4% paraformaldehyde, rinsed twice in 0.1 M PBS, and embedded in paraffin.
Isolation of Tammar Relaxin
Peptide extraction was performed according to the method of Layden and Tregear [13] with the following modifications. Approximately 1.7 g of CL from Days 20–26 of pregnancy were homogenized in 15 ml of ice-cold extraction buffer consisting of 15% (v/v) trifluoroacetic acid (TFA), 5% (v/v) formic acid, 1% (v/v) NaCl, 1 M HCl, 5 mM EDTA, 10 mM benzamidine, 1 mM 1,10-phenanthroline, and 0.5 µM pepstatin to inhibit proteinase activity. The homogenate was centrifuged at 15 000 x g at 4°C for 20 min (Sorval RC-5C plus; DuPont, Newtown, CT) and the supernatant was filtered through Millipore (Bedford, MA) 0.45-µm filters before being applied to Waters 500 mg Sep-Pak C-18 cartridges (5 ml per cartridge) on a vacuum manifold. The Sep-Pak columns were prewashed with 10 ml of 50% acetonitrile (ACN) and 0.1% TFA, followed by 30 ml of distilled water. Filtered homogenates were collected, reapplied, and then columns were washed with 10 ml of distilled water followed by protein elution with successive washes of 5 ml of 10%, 20%, 30%, 40%, 50%, and 80% ACN and 0.1% TFA. Each eluent was collected, freeze-dried, and then resuspended in 100 µl of double-distilled water for further analysis and purification.
High Performance Liquid Chromatography
The 30% fraction collected from C-18 minicolumn extraction, which was identified to contain tammar relaxin, was purified to homogeneity by HPLC. Proteins were first separated across a 60-min linear gradient of 18%–35% ACN and 0.1% TFA on a Waters HPLC system equipped with a 10x 250-mm Vydac 218 TP column packed with C-18 silica gel (330 Å pore size, 10 µm particle size; Edwards, Hesperia, CA). Fractions were collected at 1-min intervals, and peaks identified at 214 nm were analyzed by matrix-assisted laser desorption ionization time of flight (MALDI-TOF) mass spectrometry (MS). Multiple relaxin peaks were identified to elute within a 5-min period, which by MS analysis, were shown to differ in the lengths of their B chains. Individual fractions that were shown by MS to contain single relaxin forms were further purified to homogeneity using size exclusion, followed by ion-exchange chromatography on the Waters HPLC system. Size exclusion separation was performed at 1 ml/min with 0.1 M sodium phosphate buffer pH 3.4 on a 300x 7.8-mm Phenomenex Biosep SEC-2000 column (145 Å pore size, 5 µm particle size; Phenomenex, Torrance, CA). In all cases, a peak was collected that eluted in the correct molecular weight range and was identified by MS as tammar relaxin. Ion-exchange separation was then performed on these single collected peaks at 1 ml/min with a 0%–50% linear gradient of 0.5 M KCl in 50 mM potassium phosphate buffer pH 7 on a 100x 4.6-mm PolyCAT A column (300 Å pore size, 5 µm particle size; Activon, Gladesville, NSW, Australia). Single peaks were collected, which eluted at similar retention times to H2 relaxin, and which were confirmed to be tammar relaxin peptides by MS. The homogeneous tammar relaxin peptide fractions were desalted across a C-18 HPLC column as described above and freeze-dried for further analysis.
Mass Spectrometry and Amino Acid Analysis
MALDI-TOF MS was performed in the linear mode at 19.5 kV on a Bruker BIFLEX instrument (Bruker, Bremen, Germany) equipped with delayed ion extraction. Because numerous putative relaxin peptides were identified, which were postulated to be forms with different B chain lengths, each peptide was reduced in the presence of ß-mercaptoethanol to determine the mass differences in the individual chains. The reduced peptides were purified using C-18 Zip tips (Millipore, Bedford, MA) before being applied to the MS targets. Peptide amino acid analysis was performed in duplicate 24-h hydrolysates of a single homogeneous tammar relaxin peptide (B chain, Z(1)–W(28); molecular weight, 5917.9 daltons) on a GBC automatic amino acid analyzer (Melbourne, Australia).
Relaxin Receptor Assay
The ability of the purified tammar relaxin to bind to rat brain cortical relaxin receptors was tested in a relaxin radio-receptor assay. Recombinant human gene 2 relaxin (H2, B33, and B29) was a gift of Connetics Corp. (Palo Alto, CA). Native rat relaxin was purified previously [14]. H2 relaxin (B33) was labeled with 33P using the catalytic subunit of cyclic AMP-dependent protein kinase as previously described [15]. Rat cerebral cortex was collected from Sprague-Dawley rats that had been killed, it was immediately frozen in liquid nitrogen, stored at -80°C, and used within 1 wk for assay. The tissue was homogenized in 20x (v:w) ice-cold binding buffer (20 mM Hepes pH 7.5, 0.1 mg/ml lysine, 1.5 mM CaCl2, 50 mM NaCl, and 0.01% NaN3) and the membrane fractions were isolated by centrifugation at 1000 x g for 15 min followed by centrifugation of the supernatant at 50 000 x g for 60 min (Sorvall RC5C plus). The resulting pellet containing the membrane fraction was resuspended in binding buffer for assay.
Competitive binding to relaxin receptors was performed as described in Parsell et al. [16] with the following modifications. Increasing concentrations of H2 relaxin (B29), native rat relaxin, tammar relaxin B32, and B29 in 50 µl of binding buffer plus 1% (w:v) BSA were mixed with 100 pM 33P-labeled H2 relaxin (B33) in 50 µl of binding buffer plus 1% (w:v) BSA and 100 µl of membrane solution. Nonspecific binding was determined by adding 500 nM H2 relaxin (B29). Reactions were incubated for 90 min at room temperature, and then terminated by the addition of ice-cold PBS and centrifugation at 15 000 x g for 5 min in an Eppendorf centrifuge. The pellets were washed once with 1 ml of PBS followed by centrifugation, and then resuspended in 500 µl of 1 M NaOH. They were then transferred to scintillation vials, mixed with 3 ml of liquid scintillation cocktail (Ultima Gold, Packard, Meriden, CT), and counted on a liquid scintillation analyzer (Packard 1900 TR). Data are expressed as a percentage of specific binding ± SEM of triplicate determinations from 2 independent experiments, and are plotted using the single site competition function in PRISM (Graphpad Inc., San Diego, CA). Binding affinities (Ki) were calculated using the built-in Cheng and Prusoff [17] equation, and binding curves were compared using two-way ANOVA.
THP-1 Cell Relaxin Bioassay
Tammar relaxin was assayed for its ability to induce cAMP production in a relaxin-expressing cell line (THP-1) following the procedure described by Parsell et al. [16] with the following modifications. THP-1 cells, which had been viability-tested using Trypan blue, were resuspended in media and transferred to a 96-well plate at a density of 40 000 cells/well. Peptides were added to the wells together with 1 µM forskolin and 50 µM isobutylmethylxanthine (IBMX) in media and incubated at 37°C for 30 min. The plate was then briefly centrifuged, the medium was removed, and the cells were resuspended in lysis buffer. Complementary AMP levels were measured in the lysates using the cAMP Biotrak enzyme immunoassay system (Amersham Pharmacia, Rydalmere, NSW, Australia). Data are expressed as a percentage of maximum relaxin response ± SEM of triplicate determinations from three independent experiments, and the EC50 values were calculated using the PRISM program.
Analysis of Relaxin Content of the Corpus Luteum Through Pregnancy
Whole corpora lutea from specific stages of pregnancy or the nonpregnant cycle (n = 3 for both) were homogenized separately in 5 ml of extraction buffer. Relaxin was extracted from the tissue as described above using individual C-18 minicolumns. The 30% ACN fractions were collected, freeze-dried, and resuspended in 40 µl of double-distilled water for SDS-PAGE analysis. Representative samples from each stage of pregnancy and a representative, nonpregnant CL were analyzed by MS.
Gels and Protein Staining
Protein extracts (1 µl of pooled fractions, 5 µl of individual CLs) were mixed with sample loading buffer (reducing + ß-mercaptoethanol), incubated at 100°C for 3 min, and then electrophoresed on 15% Tris-Tricine SDS-PAGE gels using a standard protocol [18]. The resolved proteins were either fixed for staining or transferred by electroblotting to polyvinylidene fluoride microporous membranes (PVDF, 0.45 µm; Millipore) for Western blot analysis. Proteins were visualized using Coomassie brilliant blue or silver staining. Relaxin band density was measured using a densitometer (Bio-Rad GS-710, Quantity One quantitation software, Bio-Rad, Hercules, CA), calculated as a percentage of the relaxin band with the maximum staining intensity, which was used as a control on every gel, and expressed as relative optical density (mean ± SEM). Initial quantification (data not shown) showed that relative optical density levels were not different between Coomassie-stained and silver-stained gels, and hence, only silver staining was used for the rest of the study. Data were plotted using the PRISM program and analyzed by one-way ANOVA followed by the Newman-Keuls multiple comparison test.
Antibodies
A number of relaxin-specific antibodies were screened for cross-reactivity with tammar relaxin 30% fractions. The only antisera that gave significant cross-reactivity in dot blots was an anti-H2 relaxin antisera that had been raised in rabbits against the B29 H2 relaxin injected subcutaneously with Freund complete adjuvant.
Western Blot Analysis
Membranes were allowed to dry for 30 min and were then rewet using 100% methanol followed by washing with 20 mM Tris-HCl (pH 7.5)/0.5 M NaCl (TBS) for 5 min. The membranes were incubated for 1 h in 10% skim milk powder in TBS, then washed briefly with TBS, followed by incubation with an anti-H2 relaxin antibody (1:1000 dilution) in 3% (w/v) skim milk powder in TTBS (TBS + 0.05% Tween 20) for 1 h at room temperature on an orbital shaker. The membranes were then washed 3 times for 10 min with TTBS, followed by incubation with anti-rabbit horseradish peroxidase (HRP)-linked secondary antibody (1:2500; Bio-Rad) in 3% skim milk powder in TTBS for 1 h. Membranes were washed 3 times for 10 min in TTBS followed by visualization of the antibody-peptide complexes with enhanced chemiluminescence (ECL Western blotting detection kit, Amersham). A low molecular weight rainbow marker (Amersham) was used as the molecular weight size standard, and 100 ng of H2 relaxin was used as a positive control.
Immunohistochemistry
Relaxin immunoreactivity was localized on 5-µm sections from paraffin-embedded tammar ovaries mounted on silane-coated slides. Sections were dewaxed before being treated with target unmasking fluid (Signet Laboratories, Dedham, MA) to maximize antibody penetration into the section. The slides were then washed with PBS before blocking endogenous peroxidase activity for 30 min using 0.3% H2O2 in methanol. Slides were again washed before nonspecific binding blocking for 30 min with 10% normal horse serum. The incubation with primary antibody (rabbit anti-H2 relaxin, 1:1000 dilution) was performed overnight at 4°C. Visualization with a biotinylated secondary and an avidin-HRP tertiary antibody was performed using a Vectorstain kit (Vector Laboratories, Burlingame, CA), and the sections were counterstained with hematoxylin. For controls, the primary antiserum was preabsorbed with 1 µM H2 relaxin for 1 h before use.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A well-defined acid extraction and C-18 minicolumn protocol was used to separate tammar wallaby CL proteins. As seen in Figure 1, Western blotting with an anti-H2 relaxin antibody of the elution fractions from the C-18 columns shows an immunoreactive band in the 30% ACN elution fraction. Although the molecular weight of this band was slightly larger than that of the H2 relaxin control (5962.1 daltons), further SDS-PAGE analysis indicated that this was due to differences between the highly purified H2 relaxin and molecular weight marker, compared with the complex protein mixture in the 30% elution fraction. The tammar relaxin band in the 30% mixture was slightly retarded on the gel due to the large amount of protein in the lane, and therefore ran at a higher relative molecular weight.
|
This 30% fraction was then further separated using HPLC, and numerous peaks, eluting within a 5-min period, were collected and analyzed by MS. The HPLC peaks collected contained numerous peptides with molecular masses corresponding to the predicted amino acid sequence of tammar relaxin (Table 1, [3]). The differences in molecular masses of these peptides can be attributed to differences in the C-terminal lengths of the B chain (Table 1) as well as the cyclization of the glutamine residue at the N-terminus of the B chain to the pyroglutamyl form in some samples (Table 1, [19]). These theoretical structures were tested by reducing the peptides and then analyzing them by MS to determine the molecular masses of the individual A and B chains. As can be seen in Table 2, the molecular masses obtained correlate with the predicted A chain structure and subsequent differences in B chain C-terminal length. Peaks collected from reverse-phase HPLC, which by MS showed the presence of individual relaxin peptides, were then further purified by size exclusion followed by ion-exchange chromatography. Three peptides, B32, B29, and B28 tammar relaxin, were purified to homogeneity as indicated by MS and SDS-PAGE analysis. The homogeneous B28 tammar relaxin peptide corresponding to B chain Z(1) to W(28) (molecular weight, 5918.99 daltons) is demonstrated in Figure 2. Reduction of the peptide gave the predicted A and B chain molecular masses (Fig. 2, Table 2). The peptide was subjected to amino acid analysis to confirm the peptide sequence and subsequently resulted in the expected ratios of amino acids corresponding to the predicted amino acid sequence. The B32 and B29 variants were used in relaxin receptor assays, and B29 was used in the THP-1 relaxin bioassay.
|
|
|
Relaxin Content of Corpus Luteum During Pregnancy and the Nonpregnant Cycle
Protein extracts of corpora lutea from different stages of pregnancy were separated on SDS-PAGE gels to compare the relaxin content of the CL through pregnancy. Three corpora lutea from the nonpregnant cycle were also extracted to determine whether relaxin is produced in the CL of unmated animals. Acid extracts of whole corpora lutea were first separated across C-18 minicolumns and the 30% ACN elution fractions, which clearly have been shown above to contain the relaxin peptide, were freeze-dried and resuspended in a fixed volume for all samples (40 µl). Representative samples from each stage of pregnancy and one nonpregnant sample were analyzed by MS. All samples showed the presence of multiple forms of tammar relaxin corresponding to the variants shown in Table 1. Five microliters of each individual fraction was separated by SDS-PAGE, and the proteins were visualized by Coomassie blue or silver staining. All 30% fractions showed prominent relaxin bands running at slightly larger than 6 kDa (Fig. 1), and the bands running at a higher relative molecular weight due to the large amounts of protein in the mixture. Confirmation of the identity of the bands is represented in Figure 3, where two representative corpora lutea extracts have been separated under reducing and nonreducing conditions in comparison to H2 relaxin. They clearly show that the prominent band is lost under reducing conditions, generating a smear containing the separate A and B chains of relaxin. The separate A and B chains are not resolved, as is the case with the H2 relaxin sample, due to the chains being very similar in size and beyond the resolving capacity of the SDS-PAGE gel. Figure 4 shows a representative SDS-PAGE gel run in duplicate and compares detection by silver staining and Western blot analysis. The H2 antisera clearly cross-reacts with the prominent band on the silver-stained gels running at slightly larger than 6 kDa (as in Fig. 1). Furthermore, a further band is present running at approximately 14 kDa, which corresponds to the larger band at about 12 kDa present in the H2 relaxin lane. These bands in all likelihood represent relaxin dimers that are not visible on the silver-stained gels. Although the H2 relaxin antisera recognizes the tammar relaxin bands, the results were not as sensitive or reproducible as the silver staining results. Hence, because Figure 3 clearly shows that the relaxin band does not mask the presence of other proteins, we quantified the relaxin bands directly from the silver-stained gels using a densitometer. The intensities were normalized to the intensity of the highest expressing fraction, which was used as a control on all gels. Hence, we were able to compare the relaxin contents in individual corpora lutea by comparing the intensity of the relaxin bands on the SDS-PAGE gels. The results of this comparison are shown in Figure 5. The results indicate that the relaxin content of CL does not significantly differ from early (Days 2–10) to late (Day 25) pregnancy (ANOVA). Three corpora lutea from unmated animals were also extracted and compared with pregnant levels. Two of these samples had lower levels of relaxin (Day 23 estrous, 5.79%; Day 25 estrous, 8.66%), whereas the other was higher (Day 21, 62.86%) and, taken together, the data were not significantly different from levels during pregnancy (ANOVA).
|
|
|
Relaxin Receptor Binding and Bioactivity
The ability of the purified tammar relaxin to displace 33P-labeled H2 relaxin (B33) from rat cortical relaxin receptors was tested in a relaxin radio-receptor assay. Competition binding to rat cortex (Fig. 6A) was best fitted by a one-site competition function for all relaxin peptides tested, as previously described [15]. Unlabeled H2 relaxin (B29) displaced [33P]H2 relaxin (B33) binding with high affinity (Ki = 0.49 nM), whereas native rat relaxin competed for the binding sites with slightly lower affinity (Ki = 4.94 nM). Both B32 (Ki = 3.2 nM) and B29 (Ki = 4.9 nM) tammar relaxin showed high affinity for the rat relaxin receptor. Binding curves for rat, B32, and B29 relaxin were not significantly different, but were all significantly different from H2 relaxin (P < 0.001, two-way ANOVA).
|
Bioassay
Tammar relaxin (B29) was tested for its ability to stimulate cAMP production in the human monocytic cell line, THP-1. H2 (B29) relaxin stimulated cAMP production from THP-1 cells in a dose-responsive manner, with maximum cAMP production achieved at 2.5 nM and an EC50 of 0.201 nM (Fig. 6B). B29 tammar relaxin was also able to stimulate cAMP production in THP-1 cells, albeit with lower activity (EC50 = 14.8 nM) than H2 relaxin. The specificity of the cAMP response is demonstrated by the inability of bovine insulin to stimulate significant cAMP production at a concentration of 1 µM. These data suggest that tammar relaxin has slightly lower affinity for the human relaxin receptor compared to the rat relaxin receptor.
Immunohistochemistry
Relaxin-containing cells in the tammar ovary were localized using an anti-H2 relaxin antibody (Fig. 7). This antibody recognized tammar relaxin in Western blots (Fig. 1), albeit with a much lower affinity than H2 relaxin. As expected from the protein extraction data, relaxin was localized to CL cells, and the staining intensity was not different between Days 21, 22, 23, 24, and 25 of pregnancy (only Days 21, 23, and 25 are shown in Fig. 7). The majority of CL cells contained relaxin immunoreactivity, and no staining was seen in the ovarian stroma or associated with follicles. As can be seen under higher magnification in Figure 7G, the staining is localized to the cytoplasm of the cells, suggesting that the peptide is stored before being secreted. In ovaries from animals 1 day after birth (postpartum; Fig. 7H), relaxin immunoreactivity was present, however, the staining was more diffuse, reflecting the decaying status of the CL at this stage. In all stages, preincubation of the antisera with 1 µM H2 relaxin resulted in the loss of immunoreactivity.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tammar relaxin peptides were extracted from pooled corpora lutea of late pregnancy tammars using a combination of reverse phase, size exclusion, and ion-exchange HPLC, and identified using Western blotting with a H2 relaxin antisera and MALDI-TOF MS. Although no prorelaxin was identified, multiple 6-kDa peptides were detected, which corresponded to the predicted mature tammar relaxin amino acid sequence with an A chain of 24 amino acids, but with different B chain lengths of 28, 29, 30, and 32 amino acids. The B chain was shown to begin from Gln (27) of the prorelaxin sequence, as was predicted from the putative signal peptide cleavage site (26) Ala
Gln (27) [20]. Furthermore, some peptides demonstrated cyclization of this glutamine residue to pyroglutamyl, as has been demonstrated previously in relaxin peptides from other species [19].
Furin is a proprotein convertase that is generally localized in all cell types and cleaves precursor proteins of various types at the consensus sequence (R-X-K/R-R) [21]. The putative tammar prorelaxin molecule contains consensus sequences for furin at both the B-domain/C-peptide junction (R-W-K-R) and the C-peptide/A-domain junction (R-K-K-R). Based on these consensus cleavage sites for furin and the subsequent cleavage after the final arginine residue (R-X-K/R-RX), mature tammar relaxin would be expected to contain an A chain of 24 amino acids and a B chain to contain the B30 form (Q–R). The A chain therefore conforms exactly to that which would be expected from furin cleavage. It is interesting that the predominant form of relaxin found was the B29 variant Q–WK with multiple other forms isolated, including a B32 variant. This conforms to the heterogeneity in B chain sequences of relaxin isolated from the CL of pigs [22, 23] and rats [24]. The pig sequence also contains variants from B28 through to B31, although the most common variant is B29 [23], which relates to B30 in tammar relaxin.
It is quite possible that the longer form of relaxin is due to inefficient cleavage, whereas the shorter forms are due to proteolytic "trimming." The extraction protocol used in this study was developed to reduce proteolysis in the extraction procedure [13, 25], hence it is likely that the trimmed variants are true endogenous products rather than the result of proteolysis during the extraction protocol. It is interesting that, unlike in pigs [13] and rats [6] prorelaxin was not detected in any of the CL extracts. It is unlikely that the prorelaxin was lost or degraded during the extraction procedure because the extraction protocol was optimized for prorelaxin isolation [13]. However, it is also possible that prorelaxin is colocalized with mature relaxin in luteal cells but is present in lower amounts and was, hence, not visible on silver-stained protein gels and did not cross-react with the H2 relaxin antisera. Regardless of the endogenous length of the B chain, both MALDI-TOF MS data as well as amino acid analysis confirmed that the sequence of the tammar relaxin peptide corresponded exactly to that demonstrated by cDNA cloning [3].
The activity of tammar relaxin was tested in a binding assay [26] as well as in a well-characterized bioassay using the human cell line THP-1 [16]. Both B29 and B32 tammar relaxin showed high affinity for the rat relaxin receptor, and B29 relaxin stimulated a cAMP response in THP-1 cells. These data indicate that tammar relaxin binds to and activates relaxin receptors with high affinity. The two B chain variants showed no difference in their affinity for the relaxin receptor, which correlates with data using pig relaxin, in which B chain C-terminal shortened variants were demonstrated to have equal biological activity [23, 27].
The tammar CL contains large quantities of mature relaxin peptides, which seem to be stored in the CL. The putative yield of tammar relaxin peptide is 120 µg/g of tissue, and because this material is highly purified, the yield is probably an underestimate of the amount of peptide actually stored in the CL. Previous studies in pigs (
500 µg/g of tissue [13] and 200 µg/g of tissue [24]) and rats (280 µg/g of tissue [24]) have demonstrated similar levels of mature relaxin peptides in the CL of pregnant animals.
Analysis of individual tammar CL indicated that mature relaxin peptides were present in similar quantities from Day 2 of the 26-day gestation period and also in unmated animals at equivalent stages of the luteal phase in the nonpregnant cycle. Although a comprehensive analysis of the proportions of the different B chain tammar relaxin variants was not performed in all individual CL samples, MS analysis of representative samples indicated that all the B chain variants found in the pooled CL sample were present throughout gestation and in the nonpregnant cycle. There is, therefore, no evidence for differential processing of relaxin peptides at different stages of gestation or in the nonpregnant CL. Immunoreactive relaxin was localized specifically to the luteal cells of the CL at all stages of gestation examined. Intensity of immunostaining did not vary between stages, except in the CL of postpartum animals, where there was a clear decrease in immunoreactive relaxin in association with luteolysis. These data correlate well with relaxin gene expression, which is highest in early and mid pregnancy, and does not decrease until term [3]. However, in comparison with other species, the CL of both pregnant and unmated tammar wallabies is capable of producing mature relaxin. This highlights a difference between marsupials and eutherians in terms of relaxin synthesis because relaxin peptide concentrations in nonpregnant animals are generally undetectable [5]. One explanation is that ovarian relaxin synthesis in the tammar wallaby is not influenced by the conceptus. This is in contrast to rats, in which removal of conceptuses leads to a decline in plasma relaxin concentrations within 24 h [28, 29].
Estrogen and progesterone production by the ovaries is similar in pregnant and nonpregnant tammars, suggesting that regulation of ovarian endocrinology involves conceptus-independent mechanisms. Another interesting aspect of ovarian function in this marsupial is that the CL of the tammar wallaby enters a period of quiescence along with the unilaminar blastocyst at the 100-cell stage [7]. Once embryonic diapause is terminated, both the diapausing blastocyst and quiescent CL resume development. The CL enters quiescence whether or not there is an embryo in utero, and is presumably reactivated by the same mechanisms in both pregnant and unmated tammars. Thus, the onset of relaxin synthesis by the CL after reactivation is triggered by a stimulus other than the conceptus. The presence of high concentrations of prolactin binding sites on tammar luteal cells after ovulation [30] suggests that prolactin may be one factor involved.
Although large amounts of relaxin are clearly present in the tammar CL throughout gestation, the temporal pattern of relaxin release has not been established. However, recent data suggest that a systemic endocrine factor acts on the cervices of both the gravid and nongravid uterus and stimulates a reorganization of the collagen fibers in late gestation. The strongest evidence to date that relaxin may be this factor comes from preliminary autoradiography studies in which specific 33P-labeled tammar relaxin binding sites were localized to the stromal tissue surrounding the lumen of both cervices [10]. The function of the cervix is particularly important in marsupials because it remains tightly closed throughout pregnancy to prevent premature loss of the relatively small marsupial fetus. However, at the end of pregnancy, it has to soften and dilate to enable the fetus to pass into the birth canal. Removal of the CL on Day 23 of gestation interferes with parturition in 90% of tammars [31], suggesting that relaxin from the CL is essential for successful birth in marsupials.
In summary, this study has shown that tammar wallabies synthesize a true relaxin molecule in the context of a physiology that is present in most mammalian species investigated. The expression of significant amounts of relaxin in tammar CL during gestation implies a conservation of essential functions, particularly in regard to reproduction. It is interesting that unmated tammars also produce large amounts of relaxin, suggesting that relaxin expression is not under the control of the feto-placental unit. Therefore, the marsupial model provides unique opportunities for investigating the potential regulators of both relaxin and relaxin receptor synthesis, independent of the conceptus.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This research was supported by an Australian Research Council grant (ARC A19930117) and by an institute block grant to the Howard Florey Institute (983001) from the National Health and Medical Research Council of Australia (NHMRC). R.A.D.B. was a recipient of an NHMRC Howard Florey Centennial Fellowship, and L.J.P. is a recipient of an ARC QEII fellowship (F10020028). 
2 Correspondence. FAX: 61 3 9348 1707; r.bathgate{at}hfi.unimelb.edu.au
Accepted: February 5, 2002.
Received: November 22, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. A. Bathgate, R. Ivell, B. M. Sanborn, O. D. Sherwood, and R. J. Summers International Union of Pharmacology LVII: Recommendations for the Nomenclature of Receptors for Relaxin Family Peptides. Pharmacol. Rev., March 1, 2006; 58(1): 7 - 31. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
