Expression of human relaxin genes: characterization of a novel alternatively-spliced human relaxin mRNA species

ELSEVIER

Molecular and Cellular Endocrinology

II8 (1996) 85-94

Expression of human relaxin genes: characterization of a novel alternatively-spliced human relaxin mRNA species Jenny M. Gunnersen, Ping Fu, Peter J. Roche*, Geoffrey W. Tregear Howard

Florey

Institute

qf’ E,xperimentul

Physiology

and Medicine,

Unicersit~*

of’ Melbourne.

Parkcille.

Melhournc.

Vic,foria

3052,

Australia

Received 6 November 1995: accepted 22 January 1996

Abstract Relaxin is a two-chain peptide hormone encoded by two non-allelic genes in humans and great apes. and by a single gene in all other species studied. We have characterized the expression of the human relaxin genes (Hl and H2) in placenta, decidua, prostate and ovary by reverse-transcription/polymerase chain reaction (RT/PCR). H2 relaxin mRNA was detected in the ovary, term placenta, decidua, and prostate gland. In contrast, HI gene expression was detected only in the prostate gland. In addition to the relaxin PCR product of the predicted size (486 bp), a larger relaxin-specific product (587 bp) was detected in both Hl and H2 amplifications and in amplifications of chimpanzee relaxin from placenta and corpus luteum. Sequencing of human and chimpanzee PCR products, and human relaxin genomic clones, revealed that the larger product arises from an alternativelyspliced relaxin mRNA species incorporating an extra exon. This is the first evidence that the structure of the human and chimpanzee relaxin genes differ from that of other characterized relaxin genes, such as pig and rat. The novel peptide arising from this alternate message would be identical to prorelaxin in the B-chain and part of the C-peptide (extending to the position of the i&on) but would differ from prorelaxin in the carboxy-terminal domain. Observation of a similar mRNA species in the chimpanzee suggests that this conserved relaxin-like peptide may have a significant biological role. Keywnrd~:

Relaxin; Human; Gene structure; mRNA; Chimpanzee

-

1. Introduction The peptide hormone relaxin is produced in, and acts upon, tissues of the reproductive tract in many mammalian species to facilitate birth. In the human genome, there are two non-allelic genes coding for relaxin, designated Hl and H2 [l]. The two genes, located on the short arm of chromosome 9, are thought to have arisen from a common precursor gene by a process of gene duplication [2]. The duplication event is likely to have been relatively recent, in evolutionary terms [2,3]. While Old World monkeys (rhesus monkey and baboon) have only one relaxin gene, all great apes (chimpanzee, pygmy chimpanzee, gorilla and orangutan) have two relaxin genes [3]. In orangutan and gorilla, the gene corresponding to Hl is probably non-functional [3], but in the case of the human Hl relaxin gene, all the

* Corresponding author. Fax.: + 61 3 9348 1707.

elements required for a functional gene appear to be present and there are no inactivating mutations [4]. Synthetic relaxin peptides corresponding to human genes Hl and H2 are both bioactive in the mouse interpubic ligament and rat uterine strip bioassays [1,4,5]. In addition, synthetic Hl and H2 peptides have similar potency on rat cardiac atria, increasing the rate and force of contraction in vitro [6]. A recent investigation into the ability of HI relaxin to inhibit myometrial contractions in vitro revealed equal potency of Hl and H2 relaxins on pig myometrium [7]. Interestingly, neither HI or H2 relaxins had substantial activity on human myometrium when compared with that of porcine relaxin on this tissue [7,8]. A prominent source of circulating relaxin in women is the corpus luteum of the ovary during pregnancy [l]. In men, relaxin is secreted from the prostate gland to become a component of seminal fluid [9]. Considering that the relaxin genes are functional and that the pep-

030%7207/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PII SO303-7207(96)03770-7

J.M.

86

Gunnersen et al.

//Molrculur und Ccllulur Endocrinology 118 (1996) 85-94

tides they encode have similar bioactivities, it is interesting that the relaxin expressed in the corpus luteum is solely the product of the H2 gene [l]. Similarly, relaxin purified from human seminal plasma is also the product of the H2 gene [lo]. These findings raise questions about the physiological roles and the tissue distribution of Hl relaxin. Previous reports of Hl gene expression in human decidua, trophoblast and prostate gland [ll], and human breast [ 121, have described somewhat inconsistent results, arising from the use of different primer sets in reverse transcription/polymerase chain reaction (RT/PCR) experiments. In this study, we have used RTjPCR to determine the expression pattern of Hl and H2 in several human tissues. These experiments also revealed the presence of an alternatively-spliced mRNA species for human relaxin which encodes a novel relaxin-like peptide. The novel mRNA species is also observed in the chimpanzee, suggesting that this conserved relaxin-like peptide may have a significant biological role.

2. Materials

dard diagnostic procedures from the Pathology Department, Royal Melbourne Hospital. Placental and decidual tissue was obtained from the Monash University Department of Obstetrics and Gynecology. Placentae were collected after spontaneous onset of labour and normal vaginal delivery. Three additional sets of paired decidual and placental samples were provided after histological examination to verify that the samples were free of contaminating tissue types. 2.2. Analysis of relaxin gene expression by RTjPCR Total RNA was prepared from individual human tissue samples (5 placental samples, 3 sets of paired decidual and placental samples, 4 prostate samples, 1 peripheral leukocyte preparation) by the method of Chirgwin et al. [ 131 taking extreme caie to avoid crosscontamination of samples. The homogenization probe was soaked in 0.2 M NaOH for 30 min between samples and PCR-grade dedicated solutions were used in all stages of the procedure. The integrity of RNA preparations was monitored on agarose-formaldehyde gels. Total RNA (5 pg) from the tissues listed above, and ovary polyA+ RNA (approximately 20 ng with 5 lug carrier E. coli tRNA) was pre-treated with RNase-free DNase (RNA, 25 U pancreatic RNase inhibitor (RNasin), 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, 10 mM MgCl,, 1 mM DTT, 1 U RNase-free DNase) followed by phenol/chloroform extraction and ethanol precipitation. The reverse transcription reaction was primed with an antisense oligonucleotide primer specific for either Hl or H2 relaxin (primer 6 for Hl and

and methods

2.1. Tissue samples

Human tissue samples were obtained and used according to the provisions of the appropriate hospital ethics committees, and those of the Howard Florey Institute Human Experimentation Ethics Committee. Excess prostate biopsy samples (benign prostatic hyperplasia) were obtained after surgical removal for stanY-UTR ~~ S

1

B

T’? ? 1-

(Hl and I-Q):

5’-

8

tt

9

%;f

TCTGTTTACTACTGAACCAATTT

-3'

2 - (HI)

5'- GCCAAATGGAAGGACGAT

-3'

3-W)

5'- GACTCATGGATGGAGGAA

-3'

4-(Hl)

5'- CTCAAACAGTGCCACGTAGGGTCG

-3'

5-o-Q) 6 - (HI)

5'- ATTAGCCAATGCACTGTAGAGTTG

-3'

5'- TTAGGCAACA'IYTCTCAAACAG

-3'

7-o-w

5'- CATWXAACATTTATTAGCCAA

-3'

8-W)

5'- AATGAATK!CAACATGATAATTATA

-3'

9-W)

5'- AACAAATTCTGACATCATATTTATG

-3'

Fig. 1. Positions and sequences of PCR primers and probes used in RTjPCR analysis of human and chimpanzee relaxin. S, signal peptide; B, B chain; C, C peptide; A, A chain; UTR, untranslated region. -+ denotes forward PCR primers, +- denotes reverse PCR primers, - denotes probes. PCR primer pairs spanned the position of the intron/exon junction in the C-peptide coding region. The intron ( - 3kb) is marked by a triangle. The gene-specificity of each oligonucleotide is indicated in parentheses.

87

Hl

H2

Chimp 62

Fig. 2. (a) Expression of human relaxin gene H2 in five term placentae. This Southern blot of PCR products (I I)IO of total) from reactions primed either with HI-specific primers (1 and 6; tracks labelled HI) or HZ-specific primers (1 and 7; tracks labelled H2) has been probed with an were observed. C, negative control HZ-specific oligonucleotide. Products of the expected size (486 bp) and - 100 bp larger (unlabelled arrowhead) lane. No hybridizing products were observed on a duplicate blot probed with an HI-specific probe (data not shown). (b) Southern blots of PCR products from chimpanzee placenta (lane I) and corpus luteum (lane 2) probed with a relaxin gene 2-specific oligonucleotide. The primers used were identical to the human H2 primers described in (a). Two specific relaxin products are evident,

primer 7 for H2; Fig. 1). Reverse transcription and polymerase chain reactions with Taq polymerase were carried out according to standard protocols [14] using a 5’ primer common to both genes (primer 1, Fig. 1) and gene-specific 3’-primers (primers 6 and 7, Fig. 1). Complementary DNA (cDNA) products generated by reverse transcription (l/IO of total RT reaction containing cDNA derived from 0.5 pug total RNA) were used as the template for PCR amplification. Negative control reactions contained no cDNA. PCR reactions were pre-digested with HueIII, eliminating the possibility of amplification from any contaminating doublestranded DNA. Pre-digestion (37°C) and amplification of the samples was carried out in a Perkin-Elmer-Cetus Thermal Cycler (Model 4800). Amplification proceeded for 40 cycles: denaturing at 94°C for 1 min, annealing

at 53°C for 1 min and extension at 72°C for 2 min, and was followed by a 10 min extension period at the conclusion of the cycles. Reaction products (l/l0 of total products) were analysed by agarose gel electrophoresis. Specific relaxin products were detected by autoradiography of duplicate Southern blots probed with radiolabelled oligonucleotide probes complementary to either Hl or H2 mRNA (oligonucleotides 8 and 9; Fig. I). Blots were hybridized in Aquahyb (5 x SSC. 1% SDS, 5 x Denhardt’s solution, 0.1 mM ATP, 100 /lg/ml denatured herring sperm DNA) at 42°C and washed in 1 x SSC, 1% SDS at 50°C. RTjPCR analysis of chimpanzee placenta and corpus luteum RNA was carried out using the human relaxinprimers and probes described in Fig. 1. Nested PCR of human decidual and placental samples was carried out

88

J.M.

Gunnersen et al. / Molecular and Cellular Endocrinology 118 (1996) 85-94

Dl

D2

D3

Pl

P2

P3

lid HZ Hl H2 Hl H2 Hl H2 HI H2 Hl H2 Hl

C

Fig. 3. Expression of H2 in paired samples of term decidua (D) and placenta (P). Nested PCR products (l/10) amplified using Hl- or H2-specific nested primers (as marked on lanes) were analyzed on a 2% agarose gel, stained with ethidium bromide. M, pUC/HpaII molecular weight markers; C, negative control lane. Products of the expected size (435 bp) and y 100 bp larger are marked by arrows. No Hl products were detected when a Southern blot of this gel was probed with an oligonucleotide specific for Hl.

as follows: 0.1% of the total PCR products from the first-round amplification was re-amplified using nested gene-specific primers (primers 2 and 4 for Hl and primers 3 and 5 for H2; Fig. 1). Amplification proceeded for 30 cycles, with annealing at 50°C. 2.3. Sequencing of PCR products and genomic clones Hl, H2 and chimpanzee gene 2 PCR products were subcloned into pBLUESCRIPT II KS (Stratagene, La Jolla, CA) for sequencing. Relaxin genomic clones for the Hl and H2 genes were isolated from a human genomic cosmid library [15]. Nucleotide sequence analysis of PCR products and the Hl and H2 gene introns (5’-end) was conducted by the chain termination method [16]. Sequence data analysis and database comparisons were carried out using Staden software inAustralian National Genomic cluded in the Information Services (ANGIS) suite of programs.

3. Results 3.1. Expression of relaxin genes in human placenta, decidua, prostate and ovary The very sensitive technique of RTjPCR was used to survey several human tissues for expression of the relaxin genes Hl and H2. The positions of the PCR

primers and the oligonucleotide probes used to detect Hl- or HZspecific PCR products on Southern blots are shown in Fig. 1. The PCR primers were selected to span the intron/exon junction. This feature of the experimental design ensured that PCR products of the correct size (486 bp) could only have been generated from a mRNA template and could not have arisen from genomic DNA due to the presence of the - 3.7 kb intron [4]. In addition, the potential for double stranded cDNA contamination was minimized by DNase digest of the RNA prior to reverse transcription and the HaeIII restriction enzyme pre-digest of the PCR reactions prior to amplification. For each amplification, a negative control (containing no products of the RT reaction) was included. Fig. 2a shows the results of RTjPCR amplification of five human placental samples. The Southern blot, probed with an H2-specific oligonucleotide probe, demonstrates expression of the H2 gene in all five placentae, detectable after 40 cycles of amplification (Fig. 2a). Surprisingly, in addition to the product of the expected size (486 bp), a larger H2-hybridizing band of approximately equal abundance was observed. This product was also observed in amplifications of chimpanzee relaxin from placenta and corpus luteum (Fig. 2b). No Hl-specific products were detected on a duplicate Southern blot probed for Hl (data not shown), nor was Hl detected in other experiments with increased numbers of cycles to increase sensitivity. H2 gene ex-

J.M.

Gunnersen et al. / Molecular and Cellular Endocrinologlj 118 (1996) 85-94

89

Hl

Fig. 4. Expression of Hl and H2 in human prostate gland. Duplicate Southern blots of PCR products (l/10) were probed with either HI-specific (upper panel) or H2-specific (lower panel) oligonucleotides. Lanes labelled Hl contain Hl-specific amplification products (reactions primed with oligonucleotides 1 and 6) while those labelled H2 contain H2-specific amplification products (reactions primed with oligonucleotides 1 and 7). C, negative control lanes. Filters were washed at 55°C in 0.2 x SSC, 1% SDS and exposed to X-ray film for 2 h at - 80°C with an intensifying screen. After a longer exposure (overnight at room temperature), Hl and H2 products were observed for all four prostate biopsy samples (data not shown). The unlabelled arrowheads indicate the larger relaxin-like PCR products in Hl and H2 amplifications. pression

was

detected

in both

decidua

and

placenta

in

PCR analysis, however no Hl gene expression was observed in either tissue under these conditions (Fig. 3). Expression of H2 was not detected in decidual RNA from the first set of paired samples (Fig. 3), however, this RNA was found to be of poor quality when analysed by Northern gel electrophoresis. In contrast to the situation in the placenta and decidua, expression of both Hl and H2 genes was readily detected in three human prostate biopsy samples (from benign hyperplasia of the prostate) (Fig. 4). Expression of both genes was also detected in a fourth sample after a longer autoradiographic exposure of the same filter (data not shown). These results from prostate gland amplifications, in which both Hl and H2 products were generated, clearly demonstrate the specificity of both the amplification and detection conditions. No H2 products were amplified using the Hl-specific primer pair (and vice versa), and no crossreactivity of the probes was observed under the hybridization and washing conditions used. Consistent with the results obtained for human and chimpanzee placenta, two PCR products (one of the expected size and one approximately 100 bp larger) were also detected in amplifications of prostate gland. Larger products which hybridized to the relaxin gene-specific probes, were observed in both Hl- and H2-specific amplifications (Fig. 4). However, the relative abunpaired

samples

using

nested

dance of the larger product was lower in prostate gland PCR products than in placental PCR products, being less than 10% of the total relaxin product. In the single sample of human whole ovary RNA which was available for testing, expression of H2, but not HI, was detected and only PCR products of the expected size (486 bp) were detected by Southern blot (data not shown). No relaxin gene expression was detected in peripheral leukocytes, which were included as a control to ensure that products amplified from tissues did not originate from blood. 3.2. Sequencing of relaxin PCR products and genomic clones Relaxin PCR products from the prostate gland were subcloned and multiple individual clones were sequenced to verify the nature of the PCR products. Hl and H2-hybridizing products were found to correspond to authentic Hl and H2 relaxin sequences, respectively. In addition, the sequences of clones containing the larger Hl- and HZspecific PCR products were determined. These clones contained inserts originating from Hl and H2 amplifications of prostate tissue, Hl and H2 amplifications of RNA from a human prostate cell line LNCaP.FGC [17], and amplifications of chimpanzee placenta and ovary using gene 2-specific primers. The longer PCR products were identical to the products of the predicted size (486 bp) except for the

J.M.

90

Gunnersen et cd. 1 Molecular and Cellular Endocrinology 118 (1996) k--94

PRLX

tgctcgtaacttgctttctagcgagtatttcatttcat~attt~accagaaaga~aa ::: :: ::::: :: :::z::::

El/HZ

ctttgc&TK TlEA&&& CTaOOMTCTCAC~ca . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . :::2:::: :t::::::::::::::::

c oa

Qm&l-rTm~

TATCAC-B

PRLX

gcatcaggaacagcaaatagcttcactt~a~tcc--~cgcccttcacctcct~c ::: :::::: :: ::::::: :::: ::: :: ::::::

alhi

GCACC~~ CCTYX&aggca :: :::s:::::::::::::::::::::::t:::::::::::::::::::: ~~ccGAGMTTc-ccrrTcc

c Q2

Fig. 5. Sequence alignment of the 101 bp exon of the human Hl/H2 and chimpanzee G2 (C G2) relaxin genes with the homologous region of the porcine relaxin (PRLX) gene intron (123 bp from the 5’-end). Splice acceptor and donor consensus sequences flanking the human relaxin sequences are boxed.

insertion of a 101 bp sequence exactly at the position of the intron/exon junction. The 101 bp sequence was absolutely conserved between HI and H2 and was identical for all tissue samples, while the chimpanzee G2 sequence differed from the human insert sequence in only three positions (Fig. 5). A search of the Genbank nucleic acid sequence database revealed a similar sequence 123 bp from the 5’-end of the porcine relaxin gene intron (Fig. 5). In the case of the porcine sequence, splice donor and acceptor sequences flanking the sequence were lacking. Since the sequence of the corresponding regions of the Hl and H2 genes had not been previously determined, genomic clones for Hl and H2 were sequenced. Splice donor and acceptor consensus sequences were found to be located in the intron sequences flanking the 101 bp sequence in both the Hl and H2 genes (Fig. 5) identifying the 101 bp sequence as an additional exon. Thus, the larger PCR products observed in amplifications of human relaxin have arisen from alternatively-spliced relaxin mRNA species incorporating an extra exon (Fig. 6b).

4. Discussion In these experiments, RTjPCR was used to characterize the expression patterns of the Hl and H2 human relaxin genes in several human tissues, namely the placenta, decidua, prostate gland and ovary. Expression of the Hl gene was detected in the prostate gland (using primers to the signal peptide and A-chain regions which span the position of the intron/exon junction) but not in the placenta, decidua or ovary under the same (or more sensitive) conditions. Expression of the H2 gene was detected in all four tissues. These results are similar to those described in a previous report by other workers [I 11, including the inability to amplify Hl products

from placenta and decidua using a primer set spanning the intron/exon junction position. Unlike the authors of the previous report, however, we have concluded that while the Hl gene is clearly expressed in the prostate gland, it is not expressed in the decidua or placenta at the time of delivery. Evidence for relaxin gene expression in the decidua and placenta has been provided by northern analysis and in situ hybridization [11,18,19] although it has not been possible to distinguish between Hl and H2 mRNA species using this technique ([l 11; J. Gunnersen, unpublished results). The studies presented here demonstrate that relaxin produced locally in the decidua and placental trophoblast is the product of the H2 gene. Relaxin in intrauterine tissues has been proposed to act in a paracrine or autocrine manner [20-221. Limited evidence suggests that relaxin may be involved in fetal membrane remodelling and rupture [20], and that relaxin can stimulate decidual cells in vitro [23,24]. Relaxin immunoreactivity [25] and mRNA [26] have been detected in the human prostate gland, the source of the relaxin in seminal fluid [9]. In this study, we have demonstrated that both the Hl and H2 genes are expressed in the human prostate, and that the levels of Hl gene expression are comparable to those of H2. In the light of reports that seminal fluid relaxin is predominantly the product of the H2 gene [lo] and that relaxin cannot be detected in male serum [27], it is possible to postulate a role for Hl and/or H2 relaxin within the prostate gland itself. Relaxin may regulate growth of the prostate gland via its collagen remodelling properties [28,29] as there is an emerging connection between rates of collagen turnover and prostatic growth [30,31]. Further elucidation of the roles of Hl and H2 relaxin peptides in the prostate will be difficult in the absence of antibodies which can distinguish between the two peptides.

J.M.

Gunnrrsen

rr al. ! Moleculur

und CeNulur

Endocrino1og.t~

118 (1996)

85-94

Fig. 6. (a) Structure of pig and rat relaxin mRNAs, and one of the human and chimpanzee relaxin mRNA isoforms. (b) Human and chimpanzee relaxin mRNA isoform generated by alternative splicing. Exons are numbered in roman numerals. Incorporation of the 101 bp exon generates a stop codon in the C-p&tide codin$ region (*).

The unexpected observation of an alternativelyspliced relaxin mRNA species in human and chimpanzee has provided new information regarding the organization of these relaxin genes. Although previously thought to be similar to the pig relaxin gene which contains two exons separated by a single intron (see Fig. 6a), sequence analysis of PCR products and human relaxin genomic clones has led to the realization that the human and chimpanzee relaxin genes are composed of three exons and two introns (Fig. 6b). Thus, two different mRNA isoforms are generated by the inclusion or exclusion of the small central exon by alternative splicing (Fig. 6). Splicing-in of the 101 bp exon introduces a reading-frame shift and results in the generation of a stop codon in the C-peptide coding region (marked as * in Fig. 6b). Although the sequence of the rat relaxin gene intron is not known, the gene structure appears to be similar to that of the pig relaxin gene since PCR amplification of rat relaxin mRNA yielded a single product of the predicted size [32]. The hypothetical translations of the alternativelyspliced Hl, H2 and chimp G2 mRNA species are shown in Fig. 7. The mRNA isoforms encode 93 amino acid peptides which are identical to the corresponding prorelaxins in the B-chain and the first 13 amino acids of the C-peptide. However, the 47 amino acids forming the carboxy-terminal of the putative 93 amino acid molecule would differ from those found in prorelaxin. The predicted H2- and chimp G2-like peptides would have identical carboxy-terminal regions, since the three nucleotide differences between these cDNA sequences are conservative substitutions. Interestingly, although these molecules would lack the A-chain and hence the two inter-chain disulfide bonds in native relaxin would

be unable to form, the putative H2-/chimp G2-like peptide contains four cysteine residues which could participate in disulfide bond formation (Fig. 7). The predicted Hl-like peptide would differ slightly from the H2/G2-like peptide, with two amino acid changes in the carboxy-terminal portion (including one of the cysteine residues). Production of the longer relaxin mRNA isoform appears to be tissue-specific. This isoform, while relatively abundant in the placenta and a number of human cell lines (J. Gunnersen, unpublished results), was found to be of much lower relative abundance in the prostate gland and was not detected in the ovary. Tissue specificity of exon inclusion and exclusion has been reported for other genes, including those for troponin T [33], calcitonin/calcitonin gene-related peptide [34], neural cell adhesion molecule [35], C-SK [36] and leukocyte-common antigen [37]. Suboptimal splice site and branchpoint sequences have been implicated in tissue-specific exon exclusion. In addition, positive and negative regulatory cis-elements governing tissue-specific splicing events have been identified which are thought to bind tissue-specific trans-acting protein factors (reviewed in [38]). Thus, in the case of the relaxin pre-mRNA, placental and prostate-specific factors may be involved in selecting for the inclusion of exon II (nomenclature of Fig. 6b) by enhancing the strength of the splice sites flanking the exon. Alternatively, luteal cells may contain an inhibitor that prevents recognition of exon II that is either not present or inactive in placenta and prostate. In these tissues, both alternative isoforms of relaxin mRNA were detected however it is not known whether the alternative splicing is cell-type specific. Generation of antisera which recognize the

J.M.

92

Gunnersen et al.

I Molecular

xrinology

118 (1996) 85-94

%2-like CZ-like m-like a2-like c2-lika El-lika

x2-like cl- lika Hl-likw 60 #1-like CZ-like U-like‘

icily

Asp

Phe Ile Qln Thr Val 8er Leu (31~11s Ser Pro Asp

75

Hl-like c27liks Hl-like

Gly Gly LylsAla Leu Arg Thr Gly Serm

Phe Thr Arg Qlu Phe

90

Ha-like CZ-like Hl-1ik.a

Leu Gly Ala Leu Ser LyarLau~Hie

H1 -like

Gin Lye Pro

c2-like El-like

Pro Set Ser Thr Ly#a 11s

Wr

Leu

iiiiiiil

B-chain

Fig. 7. Sequence alignment of the putative propeptides generated by translation of the longer human and chimpanzee relaxin mRNA isoforms (shown in Fig. 6b). The amino-terminal portions of the molecules, which are identical to those of relaxin, are shaded. The unshaded residues represent the novel portion of the molecule. Cysteine residues are boxed. For the C2- and Hl-like sequences, only residues which differ from those of the H2-like sequence are shown. The predicted molecular weight of the alternatively-spliced peptide is approximately 11000 daltons, compared to 6300 daltons for relaxin.

novel portion of the relaxin-like peptide is currently in progress, and these antisera should prove useful in determining the precise cellular distribution of this peptide in the placenta and prostate gland. In conclusion, this study has confirmed the expression of the H2 relaxin gene in human ovary, placenta, decidua and prostate and provided evidence that the Hl gene, while expressed in the prostate gland, is not expressed in the other tissues surveyed. Novel human and chimpanzee relaxin mRNA isoforms have been characterized, providing evidence for structural differences between the relaxin genes of human and chim-

panzee and those of more evolutionarily such as the pig and the rat.

distant species

Acknowledgements

This work was supported by an Institute Block Grant to the Howard Florey Institute from the National Health and Medical Research Council of Australia. The authors are grateful to: Dr B. Evans for the provision of chimpanzee placental and corpus luteum RNA, Dr R. Crawford for providing human ovary PolyA+

J.M. Gunnersen et al.

: Molecular and Cellular Endocrinology 118 (1996) 85-94

RNA, Dr. Robin Bell and M. Grehan for the collection of human placental and decidual samples, Dr K.H. Choo for the human genomic cosmid library, Dr L. Cussen for histological analysis and S. Khoury for oligonucleotide synthesis.

References [I] Hudson, P., John, M., Crawford, R., Haralambidis, J., Scanlon, D., Gorman. J., Tregear, C., Shine, J. and Niall, H. (1984) Relaxin gene expression in human ovaries and the predicted structure of a human preprorelaxin by analysis of cDNA clones. EMBO J. 3, 2333-2339. [2] Crawford, R.J., Hudson, P., Shine, J., Niall, H.D., Eddy, R.L. and Shows, T.B. (1984) Two human relaxin genes are on chromosome 9. EMBO J. 3, 2341-2345. [3] Evans, B.A., Fu, P. and Tregear, G.W. (1994) Characterization of primate relaxin genes. Endocr. J. 2, 81-86. [4] Hudson, P., Haley, J., John, M., Cronk, M., Crawford, R., Haralambidis, J., Tregear, G., Shine, J. and Niall, H. (1983) Structure of a genomic clone encoding biologically active human relaxin. Nature 301, 628-631. [5] Johnston, P.D., Burnier, J. and Chen, S. (1985) Structure/function studies on human relaxin. In: Peptides Structure and Function (Deber, C.M., Hruby, V.J. and Kopple, K.D., eds.), pp. 683 686, Pierce Publications, Illinois. [6] Kakouris, H., Eddie, L.W. and Summers, R.J. (1992) Relaxin analogues: comparative inotropic and chronotropic effects in isolated rat atria. Clin. Exp. Pharmacol. Physiol. Suppl. 21, 33. [7] MacLennan, A.H. (1995) The effect of relaxins on myometrial activity and cervical ripening. In: Progress in Relaxin Research (MacLennan, A.H., Tregear, G.W. and Bryant-Greenwood, G.D.. eds.), pp. 263. 288, Global Publications Service, Singapore. A.H. and Grant, P. (1991) Human relaxin: in vitro PI MacLennan, response of human and pig myometrium. J. Reprod. Med. 36, 630.-634. [91 Essig, M., Schoenfeld, C., D’Eletto, R., Amelar, R., Oubin, L. and Steinetz, B.G. (1982) Relaxin in human seminal plasma. Ann. NY Acad. Sci. 380, 224-230. [lOI Winslow, J.W., Shih, A., Bourell, J.H., Weiss, G., Reed, B., Stults, J.T. and Goldsmith, L.T. (1992) Human seminal relaxin is a product of the same gene as human luteal relaxin. Endocrinology 130, 2660 -2668. G.D. and Greenwood, F.C. u II Hansell. D.J., Bryant-Greenwood, (1991) Expression of the human relaxin HI gene in the decidua, trophoblast and prostate. J. Clin. Endocrinol. Metab. 72, 899% 904. L.S., Mazoujian, G. and Bryant-Greenwood, G.D. [I21 Tashima, (1994) Human relaxins in normal, benign and neoplastic breast tissue. J. Mol. Endocrinol. 12, 351-364. R. and Rutter, W. (1979) 1131Chirgwin, J., Pryzala, A., MacDonald, Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299. [I41 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning ~ A Laboratory Manual, 2nd edn., Cold Spring Harbour Laboratory Press, New York. [I51 Choo, K.H., Filby, G., Greco, S., Lan, Y-F. and Kan, Y.W. (1986) Cosmid vectors for high efficiency DNA-mediated transformation and gene amplification in mammalian cells: studies with the human growth hormone gene. Gene 46, 277-286. [I61 Sanger, F., Miklen, S. and Coulson, A.R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463 --5467.

u71 Gunnersen,

93

J.M., Roche, P.J.. Tregedr, G.W. and Crawford, R.J. (I 995) Characterization of human relaxin gene regulation in the relaxin-expressing human prostate adenocarcinoma cell line LNCaP.FGC. J. Mol. Endocrinol. (in press). F.C. and Bryant-Green[I81 Sakbun, V., Ali, S.M., Greenwood. wood, G.D. (1990) Human relaxin in the amnion, chorion, decidua parietalis, basal plate and placental trophoblast by immunocytochemistry and northern analysis. J. Clin. Endocrinol. Metab. 70, 508- 514. G.D. (1995) [I91 Bogic, L.V.. Mandel, M. and Bryant-Greenwood, Relaxin gene expression in human reproductive tissues by in situ hybridization. J. Clin. Endocrinol. Metab. 80. I30- 137. G.D.. Yamamoto, S.Y. and PO1 Koay, E.S.C., Bryant-Greenwood, Greenwood, F.C. (1986) The human fetal membranes: a target tissue for relaxin. J. Clin. Endocrinol. Metab. 62, 513 -521. G.D. (1991)The human relaxins: consensus PII Bryant-Greenwood, and dissent. Mol. Cell. Endocrinol. 79. Cl25%Cl32. A.H., Grant. P. and Borthwick, A. (1991) Relaxin WI MacLennan, and relaxin C-peptide levels in human reproductive tissues. Reprod. Fert. Dev. 3, 5777583. S. ~231Poisner. A.M., Thraillkill. K., Poisner, R. and Handwerger, (1990) Relaxin stimulates the release of prorenin from human decidual cells: evidence for autocrine:paracrine regulation. J. Clin. Endocrinol. Metab. 70, 1765 1767. ~241Thraillkill, K.M., Clemmons. D.R., Busby, W.H. and Handwerger, S. (1990) Differential regulation of insulin-like binding protein secretion from human decidual cells by IGF-I, insulin and relaxin. J. Clin. Invest. 86, 87%883. P.D. and ~251Sokol, R.Z.. Wang, X.S., Lechago. J.. Johnston, Swerdlotf. R.S. (1989) lmmunohistochemical localization of relaxin in human prostate. J. Histochem. Cytochem. 37, 1253 1255. F. and Dawood, M.Y. WI Ivell, R., Hunt, N.. Khan-Dawood. (1989) Expression of the human relaxin gene in the corpus luteum of the menstrual cycle and in the prostate. Mol. Cell. Endocrinol. 66, 25l- 255. v71 Lucas, C., Bald. L.N., Martin, M.C.. Jaffe, R.B., Drolet, D.W., P.D. Mora-Worms. M., Bennett, G., Chen, A.B. and Johnston. (1989) An enzyme-linked immunosorbent assay to study human relaxin in human pregnancy and in pregnant rhesus monkeys. J. Endocrinol. 120, 449-457. WI Unemori, E.N. and Amento, E.P. (1990) Relaxin modulates synthesis and secretion of procollagenase and collagen by human dermal fibroblasts. J. Biol. Chem. 265, 10681~10685. 1291 Unemori, E.N., Beck. L.S.. Lee, W.P., Xu, Y., Siegel, M., Keller, G.. Liggitt, H.D., Bauer, E.A. and Amento, E.P. (1993) Human relaxin decreases collagen accumulation in vivo in two rodent models of fibrosis. J. Invest. Dermatol. 101, 280-285. [301 Muntzing, J. (1980) Androgen and collagen as growth regulators of the rat ventral prostate. Prostate I. 71 78. M.O., Frederickson, R.. Matal. J. and Mawhinney, [311 Thornton, M. (1984) In: New Approaches to the Study of Benign Prostatic Hyperplasia (Kimball. F.A., Buhl. A.E. and Carter. D.A., eds.). pp. 143-158. Alan Liss. New York. J.M., Crawford, R.J. and Tregear, G.W. (1995) ~321Gunnersen, Expression of the rat relaxin gene in rat tissues. Mol. Cell. Endocrinol. I IO, 55-64. B. (1987) Developmentally [331 Breibart, R.E. and Nadal-Ginard. induced, muscle-specific trans factors control the differential splicing of alternative and constitutive troponin T exons. Cell 49, 793--803. [341 Crenshaw, E.B. III, Russo, A.F., Swanson, L.W. and Rosenfeld. M.G. (1987) Neuron-specific alternative RNA processing in transgenic mice expressing a metallothionein-calcitonin fusion gene. Cell 49, 389-398. [35] Tacke, R. and Goridis, C. (1991) Alternative splicing in the neural cell adhesion molecule pre-mRNA: regulation of exon I8 skipping depends on the 5’ splice site. Genes De\. 5. 1416~1429.

94

J.M.

Gunnersen et al. / Molecular and Cellular Endocrinology I18 (1996) 85-94

[36] Black, D.L. (1992) Activation of c-src neuron-specific splicing by an unusual RNA element in vivo and in vitro. Cell 69, 7955807. [37] Streuli, M. and Saito, H. (1989) Regulation of tissue specific alternative-splicing: exon-specific elements govern the splicing of

leukocyte common antigen pre-mRNA. EMBO J. 8, 787796. [38] Hodges, D. and Bernstein, S.L. (1994) Genetic and biochemical analysis of alternative RNA splicing. Adv. Genet. 31, 2077281.