Enzyme-linked immunosorbent assay of relaxin-like gonad-stimulating peptide in the starfish Patiria (Asterina) pectinifera.

A relaxin-like gonad-stimulating peptide (RGP) from starfish Patiria (Asterina) pectinifera is the first identified invertebrate gonadotropin for final gamete maturation. Recently, we succeeded in obtaining specific antibodies against P. pectinifera RGP (PpeRGP). In this study, the antibodies were used for the development of a specific and sensitive enzyme-linked immunosorbent assay (ELISA) for the measurement of PpeRGP. A biotin-conjugated peptide that binds to peroxidase-conjugated streptavidin is specifically detectable using 3,3',5,5'-tetramethylbenzidine (TMB)/hydrogen peroxide as a substrate; therefore, biotin-conjugated RGP (biotin-PpeRGP) was synthesized chemically. Similarly to PpeRGP, synthetic biotin-PpeRGP bound to the antibody against PpeRGP. In binding experiments with biotin-PpeRGP using wells coated with the antibody, a displacement curve was obtained using serial concentrations of PpeRGP. The ELISA system showed that PpeRGP could be measured in the range 0.01-10pmol per 50µl assay buffer. On the contrary, the B-chains of PpeRGP, Asterias amurensis RGP, Aphelasterias japonica RGP, and human relaxin showed minimal cross-reactivity in the ELISA, except that the A-chain of PpeRGP affected it slightly. These results strongly suggest that this ELISA system is highly specific and sensitive with respect to PpeRGP.

Relaxin 2/RXFP1 Signaling Induces Cell Invasion via the ß-Catenin Pathway in Endometrial Cancer.

Relaxin is known to play an important role in animal pregnancies, including those of humans. It is suggested that relaxin induces aggressive cell growth and invasiveness in several types of cancer, including endometrial cancer. However, the mechanisms of relaxin remain largely unclear. In this study, we examined the effects of relaxin 2 (RLN2), the major circulating relaxin in humans, on human endometrial carcinoma cell lines. RLN2 treatment induced invasion in HEC-1B and Ishikawa cells. RLN2-induced cell invasion was significantly decreased by transfection of relaxin receptor 1 (RXFP1) siRNAs. The ß-catenin inhibitor, XAV939, also significantly inhibited the RLN2-induced cell invasions. Both a decrease of cadherin expression and an increase of ß-catenin phosphorylation were observed in response to the RLN2 treatment in HEC-1B and Ishikawa cells. We then examined RLN2 and RXFP1 expression in 80 human endometrioid endometrial carcinoma tissues. RLN2 immunoreactivity was detected in the human endometrial carcinoma cells and had a correlative tendency with histological grade and RXFP1. These results suggest that adherens junctions in cancer cells are weakened by the breakdown of the cadherin/catenin complex, which is induced by ß-catenin phosphorylation via RLN2/RXFP1 signaling.

Genetic associations of relaxin: preterm birth and premature rupture of fetal membranes.

OBJECTIVE: Relaxin H2 (RLN2) is a systemic hormone (sRLN) that is produced by the corpus luteum, whereas decidual RLN (dRLN) acts only locally. Elevated sRLN is associated with spontaneous preterm birth (sPTB) and elevated dRLN with preterm premature rupture of membranes (PPROM). Associations were sought between single nucleotide polymorphisms (SNPs) in the RLN2 promoter with levels of dRLN and sRLN in Filipino patients with sPTB, PPROM, or normal term delivery. STUDY DESIGN: Stringent selection of women with sPTB (n = 20) or PPROM (n = 20) and term control subjects (n = 20) was made from >8000 samples from Filipino patients who delivered at 34-36 weeks' gestation. Twelve SNPs were genotyped on maternal blood, with 9 excluded based on the high linkage disequilibrium or being the same as in the control population. Quantitative immunocytochemistry on parietal decidual tissue was performed (n = 60); sRLN was measured by enzyme-linked immunosorbent assay in a subset of patients (n = 21). RESULTS: SNP rs4742076 was associated significantly with PPROM (P < .001) and increased expression of dRLN (P < .001). The genotype TT had increased dRLN in PPROM (P < .05). SNP rs3758239 was associated significantly with both PPROM and sPTB (P < .01), and genotype AA had increased dRLN expression (P < .05). The sRLN showed a trend of higher levels in PPROM and sPTB, but was not significant. CONCLUSION: SNP rs4742076 in the RLN2 promoter was associated with increased dRLN expression and PPROM; SNP rs3758239 was associated with both PPROM and sPTB in these Filipino patients. Specific homozygous genotypes were identified for both SNPs and were shown to be associated with increased dRLN tissue expression.

Significance of relaxin-2 expression in hepatocellular carcinoma: relation with clinicopathological parameters.

BACKGROUND: A number of putative roles, including the modulation of tumor growth, neovascularization, metastasis and oncogenic progression, have been correlated to relaxin-2 overexpression. However, the clinical significance of relaxin-2 expression in hepatocellular carcinoma (HCC) remains unclear. The aim of this study was to investigate the expression of relaxin-2 in HCC and determine its correlation with tumor progression and prognosis. PATIENTS AND METHODS: 180 HCC patients who had undergone curative liver resection were selected and immunohistochemistry was performed to analyze relaxin-2 expression in the respective tumors. RESULTS: Immunohistochemistry confirmed relaxin-2 overexpression in HCC tissues compared with their adjacent nonneoplastic tissues (p < 0.01). Additionally, immunostaining showed more relaxin-2 positive cells in the higher tumor grade (III) than in the lower tumor stage (I, II; p = 0.026). Moreover, HCC patients with high relaxin-2 expression were significantly associated with lower 5-year overall survival (p < 0.01) and lower 5-year disease-free survival (p < 0.01), respectively. Furthermore, immunostaining showed more relaxin-2 positive cells in the tumor recurrence (ETR) patients than non-ETR patients (p = 0.001). The Cox proportional hazards model further showed that relaxin-2 was an independent poor prognostic factor for both 5-year disease-free survival (hazards ratio [HR] = 1.872, 95% confidence interval [CI] = 1.18-5.146, p = 0.023) and 5-year overall survival (HR = 3.637, CI = 1.443-7.15, p = 0.001) in HCC. CONCLUSIONS: Our data suggest for the first time that the overexpression of relaxin-2 protein in HCC tissues is of predictive value on tumor progression and poor prognosis.

The molecular detection of relaxin and its receptor RXFP1 in reproductive tissue of Felis catus and Lynx pardinus during pregnancy.

Relaxin acts as a pregnancy-specific signal in feline species, but specific information about protein structure and binding is essential for the improvement of pregnancy diagnosis in endangered feline species, like the Iberian lynx. To generate a felid-specific relaxin antibody, the DNA and protein sequences of lynx and cat were determined and peptides were chosen for antibody generation. In addition, relaxin and relaxin receptor (RXFP1) mRNA expressions were measured in uteri and ovaries of pregnant domestic cats and lynx placentae. Using real-time PCR and immunohistochemistry, it was established that feline placenta is the main source of relaxin during pregnancy. In other tested tissues, relaxin mRNA expression was weak. The RXFP1 mRNA expression was found mainly in cat uterine tissue and feline placentae. It was assumed that these tissues were main targets for relaxin. In the ovary, relaxin immunostaining was associated with blood vessels, signifying its role in vascularization.

Characterization and biological activity of relaxin in porcine milk.

A lactocrine mechanism for delivery of maternally derived relaxin (RLX) into the neonatal circulation as a consequence of nursing was proposed for the pig. Immunoreactive RLX was detected in colostrum and in the serum of newborn pigs only if they were allowed to nurse. Milk-borne RLX concentrations are highest during early lactation (9-19  ng/ml), declining to <2  ng/ml by postnatal day 14. Whether milk-borne RLX is bioactive is unknown. Evidence that RLX concentrations in milk are higher than in maternal circulation in several species suggests the mammary gland as a site of local RLX production. It is unknown whether the porcine mammary gland is a source of RLX. Therefore, objectives were to evaluate RLX bioactivity in porcine milk during the first 2 weeks of lactation, identify the form of RLX in porcine milk, and determine whether mammary tissue from early lactation is a source of milk-borne RLX. Milk RLX bioactivity was determined using an in vitro bioassay in which cAMP production by human embryonic kidney (HEK293T) cells transfected with the human RLX receptor (RXFP1) was measured. RLX bioactivity was highest at lactation day (LD) 0, decreasing to undetectable levels by LD 4. Immunoblot analysis of milk proteins revealed an 18  kDa band, indicating proRLX as the primary form of RLX in porcine milk. ProRLX protein and transcripts were detected in porcine mammary tissue on LD 0 and 7. Results support the lactocrine hypothesis by defining the nature and a potential source for bioactive proRLX in porcine colostrum/milk.

Commercial immunoassays for human relaxin-2.

Several different immunoassays have been used in the commercial pharmaceutical development of serelaxin. These assays have been well validated for submission of GLP preclinical and clinical studies to the FDA and EU regulatory bodies. The requirements for these assays exceed that of most research assays commonly developed in academic research but have been and are currently available to academic researchers. Additionally, many human relaxin immunoassays are commercially available from a variety of vendors. Validation procedures for immunoassays are well understood and documented, however validation of these assays is often lacking or completely absent. The data derived from these assays must be questioned if the investigator does not supply information on the validation of the assay used, either from the supplier or through their own efforts. Many recent papers on determination of serum relaxin in clinical settings have recently been published. The assay used for this determination varies but generally is one of two commercially available. These manuscripts and the assay used is discussed. Direct comparisons of assays are lacking but some general conclusions can be drawn by comparing results from similar studies using different assays. There is disagreement among the results of the concentrations of serum relaxin from the use of different assays that raise questions on assay reliability. The differences in the quality of immunoassays used for detection of serum relaxin should be part of the decisions making process in choosing an assay. While the end user bears the ultimate responsibility to demonstrate the assay is valid for the stated claims, reviewers and editors also share responsibility for quality of published results.

Identification of porcine relaxin in plasma by liquid chromatography-high resolution mass spectrometry.

Relaxin (RLX) has demonstrated diverse pharmacological effects in various scientific and clinical studies. The emergence of porcine relaxin (pRLX) has raised concerns on the doping potential of pRLX. There have also been speculations in the horseracing industry on its covert use. To control the abuse of pRLX in equine sports, a method to detect pRLX effectively and to provide its unequivocal identification in equine biological samples is required. This paper reports on the detection and confirmation of pRLX in equine plasma by liquid chromatography-high resolution mass spectrometry. pRLX was isolated from equine plasma by immunoaffinity purification using anti-pRLX antibody-coated magnetic beads. Anti-pRLX antibody was generated in-house by purifying antisera from rabbits immunized with pRLX. The isolated pRLX was subjected to reduction of their disulfide bonds to obtain their respective A- and B-chains. The extracts were then further purified and concentrated prior to reversed-phase LC separation and high resolution accurate mass measurement. As detection of the A-chains was far more sensitive than that of the B-chains, the A-chain of pRLX was set as the targets for detection and confirmation. The limit of detection for pRLX was around 0.005 ng/mL (~ 0.86 fM) and the limit of confirmation was around 0.02 ng/mL (~ 3.4 fM). It was observed that method sensitivity was improved at least 5-fold by using an EASY-spray column and emitter in place of the conventional ESI source. The applicability of this method was demonstrated by the identification of pRLX and its metabolites in equine plasma obtained after subcutaneous administration. To our knowledge, this is the first report of a validated mass spectrometry method for the unequivocal confirmation of pRLX in any biological fluid. Copyright © 2016 John Wiley & Sons, Ltd.

The role of relaxin in endometrial cancer.

Relaxin (RLN) is a naturally occurring hormone that is known to modulate connective tissue remodeling in the uterus and cervix. Our goal was to investigate the role of RLN in endometrial cancer. RLN expression was evaluated using immunohistochemistry in 57 samples of invasive endometrial carcinoma (EC) and ten benign endometrial tissues. 67% of high-stage (III/IV) tumors demonstrated strong RLN expression compared to 37% of low-stage (I/II) cases. Strong RLN expression associated significantly with high-grade and depth of myometrial invasion. Notably, strong RLN expression was associated with a significantly shorter overall survival (p < 0.005) compared to weak or moderate expression. Using RT-PCR, the expression of RLN and its receptor (LGR7) was detected in EC cell lines (HEC-1B and KLE); in addition, LGR7 was expressed in 86% of 15 primary EC tissue samples. Exogenous RLN stimulation caused a significant increase in migration and invasion in both cell lines, but did not stimulate proliferation in vitro. Addition of the MMP inhibitor, FN439 abolished the stimulatory effect of RLN on invasion in both HEC-1B and KLE cells. RLN stimulation caused a significant increase in levels of activated MMP-2 in KLE cells and activated MMP-9 in HEC-1B cells compared to unstimulated cells. Inhibition of endogenous RLN signaling via siRNA targeted to LGR7 caused a significant reduction of EC cell invasiveness. Our results indicate that RLN overexpression is significantly asso- ciated with aggressive features such as high-grade and deep myometrial invasion. We provide the first evidence that overexpression of RLN is associated with poor clinical outcome in women with EC. RLN stimulation enhances the invasive potential of endometrial cancer cells by upregulating MMPs. In turn, downregulation of endogenous RLN signaling decreases invasiveness of endometrial cancer cells. These novel findings may have therapeutic implications in the management of patients with endometrial carcinoma.

Expression and functional role of hepatocyte growth factor and its receptor (c-met) during fetal mouse testis development.

The hepatocyte growth factor (HGF) is a pleiotropic cytokine able to regulate different cellular functions. HGF action is mediated by its receptor, c-met, a glycoprotein with tyrosine kinase activity. We previously demonstrated that c-met is expressed in the newly formed seminiferous cords of the mice embryonic testes and that HGF acts as a morphogenetic factor. In this paper, we report that at 15.5 days post-coitum (dpc) c-met is expressed in the testicular cords, whereas at 18.5 dpc c-met expression is almost exclusively localized in the interstitial tissue of the testis in particular in the fetal Leydig cells. In addition, we demonstrate that HGF gene is expressed during the fetal period of testis development, heavily detectable in the interstitial compartment of 18.5 dpc testes. Interestingly, HGF is not expressed in the Leydig cells that, as above reported, express the HGF receptor. Looking for the functional role of HGF on Leydig cells, we evaluated the amount of testosterone secreted by testes isolated from 18.5 dpc embryos and cultured in the presence of HGF. The results of the in vitro organ culture show that, at this age, HGF increases the amount of testosterone secreted in the culture medium. On the contrary, HGF does not modulate the amount of testosterone secreted by testes isolated from 15.5 dpc embryos. In conclusion, we report that HGF is produced in the interstitial compartment of the developing testis but not by the Leydig cells. Conversely, the HGF receptor c-met is expressed in the Leydig cells and HGF modulates Leydig cell function during the late period of prenatal development.

Quantitation of estrogen receptors and relaxin binding in human anterior cruciate ligament fibroblasts.

The significantly higher incidence of anterior cruciate ligament (ACL) injuries in collegiate women compared with men may result from relative ligament laxity. Differences in estrogen and relaxin activity, similar to that seen in pregnancy, may account for this. We quantified estrogen receptors by flow cytometry and relaxin receptors by radioligand binding assay in human ACL cells and compared the presence of these receptors in males and females. ACL stumps were harvested from seven males and eight females with acute ACL injuries. The tissue was placed in M199 cell culture medium. Outgrowth cultures were obtained, and passage 2 cells were used for all studies. Estrogen receptor determination was performed using flow cytometry. Relaxin binding was performed in ACL cells derived from five female and male patients using I(125)-labeled relaxin. Estrogen receptors were identified by flow cytometry in 4 to 10% of ACL cells. Mean fluorescence of cells expressing estrogen receptors was approximately twice that of controls, with no significant differences between males and females. Relaxin studies showed low-level binding of I(125)-relaxin-labeled ACL cells. Relaxin binding was present in four out of five female ACL cells versus one out of five male ACL cells.

Altered relaxin family receptors RXFP1 and RXFP3 in the neocortex of depressed Alzheimer's disease patients.

RATIONALE: The G-protein-coupled relaxin family receptors RXFP1 and RXFP3 are widely expressed in the cortex and are involved in stress responses and memory and emotional processing. However, the identification of these receptors in human cortex and their status in Alzheimer's disease (AD), which is characterized by both cognitive impairments and neuropsychiatric behaviours, have not been reported. OBJECTIVES: In this study, we characterized RXFP receptors for immunoblotting and measured RXFP1 and RXFP3 immunoreactivities in the postmortem neocortex of AD patients longitudinally assessed for depressive symptoms. METHODS: RXFP1 and RXFP3 antibodies were characterized by immunoblotting with lysates from transfected HEK cells and preadsorption with RXFP3 peptides. Also, postmortem neocortical tissues from behaviourally assessed AD and age-matched controls were processed for immunoblotting with RXFP1 and RXFP3 antibodies. RESULTS: Compared to controls, putative RXFP1 immunoreactivity was reduced in parietal cortex of non-depressed AD patients but unchanged in depressed patients. Furthermore, putative RXFP3 immunoreactivity was increased only in depressed AD patients. RXFP1 levels in the parietal cortex also correlated with severity of depression symptoms. In contrast, RXFP1 and RXFP3 levels did not correlate with dementia severity or ß-amyloid burden. CONCLUSION: Alterations of RXFP1 and RXFP3 may be neurochemical markers of depression in AD, and relaxin family receptors warrant further preclinical investigations as possible therapeutic targets for neuropsychiatric symptoms in dementia.

Early human preantral follicles have relaxin and relaxin receptor (LGR7), and relaxin promotes their development.

The regulatory mechanisms of early follicle development are not clearly understood. Although relaxin is a peptide that controls cell proliferation and differentiation in many tissues, its role in human follicular development is unclear. In this study we cultured slices of human ovarian cortical tissue in the presence and absence of recombinant human relaxin. Ovarian tissue was obtained by biopsy during gynecological laparotomy or laparoscopy (14 women; mean age +/- sem, 29.0 +/- 6.1 yr; range, 17-37 yr). A significantly higher proportion of secondary follicles (14.5% vs. 5.0% in the control group; P < 0.01) and a significantly decreased proportion of primordial follicles (30.1% vs. 47.4% in the control group; P < 0.05) were found in tissues cultured with relaxin for 7 d. Immunocytochemical studies with the anti-C-peptide of prorelaxin and antirelaxin antibodies revealed the localization of relaxin in the oocyte and in flat pregranulosa and granulosa cells of primordial, primary, and secondary follicles. The presence of the relaxin receptor LGR7 was observed in flat pregranulosa and granulosa cells of primordial, primary, and secondary follicles by immunocytochemical and in situ hybridization analyses. These results suggest that relaxin plays a role through its receptor during the early stage of follicle development.

The association between follicular fluid levels of cathepsin B, relaxin or AMH with clinical pregnancy rates in infertile patients.

OBJECTIVE: The aim of this study was to investigate the relationship of cathepsin B, relaxin and anti-Mullerian hormone (AMH) in follicular fluid (FF) with pregnancy rates in infertility patients. STUDY DESIGN: Seventy-nine infertile couples who underwent ICSI were included in the study. The FF levels of cathepsin B, relaxin and AMH were measured using ELISA kits. RESULTS: The FF cathepsin B levels were statistically higher in pregnant patients compared with non-pregnant patients (0.20±0.13 versus 0.13±0.03; P<0.001). There were statistically significant differences in the total number of oocytes (10.00±6.85 versus 5.96±3.94); the mean number of M2 oocytes (8.65±5.63 versus 4.58±3.36) between the pregnant and non-pregnant patients (P<0.05). There were no significant correlations between pregnancy rates and relaxin and AMH (P>0.05). The area under the curve of cathepsin B for prediction of pregnancy was 0.662 (p=0.024, 95% Confidence Interval 0.528-0.797). CONCLUSIONS: This study demonstrated that increased level of cathepsin B in FF significantly correlates with a better chance of clinical pregnancy. Further studies are needed to clarify the role of cathepsin B in the reproductive process of humans.

Development of a homologous equine relaxin radioimmunoassay.

Equine relaxin (eRlx) immunoactivity has previously been measured in the mare during pregnancy using the porcine relaxin (pRlx) RIA (pRlx-RIA). This was not the optimal system for measurement of eRlx because the dose-response curve obtained with equine plasma was not parallel to the pRlx standard curve. A homologous eRlx-RIA has been developed and used to measure relaxin immunoactivity during pregnancy and parturition in the mare. Highly purified eRlx was used for the generation of antiserum in rabbits, preparation of tracer, and as assay standards. A double antibody eRlx RIA (eRlx-RIA) was developed which was highly specific and sensitive (0.023 ng relaxin/tube). The dose-response curve obtained with eRlx plasma was parallel to the equine standard curve while there was no parallelism noted between the pRlx and eRlx standard curves. This assay was utilized to measure eRlx in samples collected during pregnancy which were previously measured for relaxin content in the pRlx-RIA. It was found that the two assay systems gave almost identical patterns of secretion throughout pregnancy. The two assays differed in the amount of relaxin measured, with the eRlx-RIA measuring considerably higher amounts during gestation than the pRlx assay. In oxytocin-induced foalings, the differences became greater with the equine assay measuring as much as 10-fold greater concentrations.

Human seminal relaxin is a product of the same gene as human luteal relaxin.

Unlike that of other species, which have only one gene encoding relaxin, the human genome contains two nonallelic genes for relaxin, designated H1 and H2, which encode markedly different relaxin peptides. Whereas human relaxin gene H2 is selectively expressed in the ovary, no ovarian expression of gene H1 has been detected. Since relaxin is actively produced in the human male, it is possible to postulate divergent gene expression of relaxin in the male and female. We examined this question directly through the structural determination of human seminal relaxin and its comparison with the structure of human luteal relaxin. Partially purified relaxin, prepared from pooled human seminal plasma which had been delipidated by extraction with acid acetone and hexane, subjected to two cycles of HPLC and an additional purification step by ion-exchange chromatography, was further purified by immunoaffinity chromatography, using a monoclonal antibody to the H2 relaxin A chain which cross-reacts with synthetic H1 relaxin, followed by an additional HPLC step performed on a C4 reverse-phase column. The recovered, purified relaxin was then analyzed by N-terminal gas-phase sequencing and fast atom bombardment mass spectroscopy for determination of the amino acid sequence and molecular ions of the A and B chains, respectively. The results demonstrate that the structure of the predominant relaxin in human semen plasma is derived from the product of the H2 gene, consisting of a N-terminal pyroglutamic acid A-24 A chain and a mixture of B-26 and B-27 B chains. With the exception of degradation of the seminal relaxin B chain C-terminus, this structure is identical to the structure of human luteal relaxin. Therefore, both human seminal and luteal relaxin are products of the H2 gene.

N-alpha-formyl-tyrosyl-relaxin - a reliable tracer for relaxin radioimmunoassay.

Porcine relaxin has no radio-iodinatable amino acid and must therefore be modified prior to iodination. In this paper a synthesis of an iodinatable relaxin derivative is described which leads to a fairly uniform product. The nature of the modifying group (formyltyrosine) allows for easy and accurate estimation of the degree of substitution attained as well as determination of the position where substitution has occurred.

Ultrastructural localization of relaxin immunoreactivity in corpora lutea of pregnant rats.

Relaxin was localized in cells of corpora lutea of pregnant rats at the ultrastructural level using a highly specific antirat relaxin serum and the unlabeled peroxidase-antiperoxidase technique. Electron-dense, membrane-bound granules (maximum diameter, 270 nm), which are present in luteal cells during the last third of gestation, were the only inclusions that were immunochemically stained. The number of granules observed in the luteal cell cytoplasm varied from cell to cell within a particular section. Furthermore, in the granule-rich luteal cells, the granules appeared in clusters. This study establishes that these electron-dense granules represent the subcellular sites of relaxin localization within luteal cells of pregnant rats.

Immunocytochemical localization of relaxin in endometrial glands of the pregnant guinea pig.

Relaxin was localized in the cytoplasm of endometrial gland cells in the uterus of day 60 pregnant guinea pigs with the peroxidase-antiperoxidase method. The immunoprecipitate was located over the apical portions of the glandular cells. Stain was not found over non-endometrial portions of the uterus, endometrial surface epithelium or cells of the uterine cervical glands. Agar double-immunodiffusion analyses demonstrated a reaction of identity when extracts of uterus from day 60 pregnant guinea pigs and a porcine relaxin standard were tested with an antiserum produced against highly purified porcine relaxin. An extract of uteri from day 60 pregnant guinea pigs gave a positive response in the mouse uterus bioassay for relaxin.

Development of a homologous radioimmunoassay for rat relaxin.

Highly purified rat relaxin has been radioiodinated to specific activities of approximately 100 micro Ci/microgram with the Bolton and Hunter reagent [N-succinimidyl 3-(4-hydroxy-5-[125I]iodophenyl) propionate]. A rabbit antirat relaxin serum, applicable in a final dilution of 1:100,000, was developed in a rabbit using unconjugated highly purified rat relaxin. A specific and precise double antibody RIA for rat relaxin sufficiently sensitive to routinely measure from 32--2000 pg rat relaxin was developed. Using this RIA, relaxin immunoactivity levels in extracts of pregnant rat ovaries were found to rise from 0.8 microgram/geq ovarian fresh tissue on day 8 of pregnancy to 723 microgram/geq ovarian fresh tissue on day 20 of pregnancy and then to drop precipitously to 6 microgram/geq ovarian fresh tissue on day 1 of lactation. Consistent with the occurrence and relative levels of relaxin in the ovarian extracts, levels of relaxin in pregnant rat serum were less than 2 ng/ml on day 10 of pregnancy, approximately 150 ng/ml on days 20 and 22 of pregnancy, and 12 ng/ml on day 2 of lactation.