Molecular and Ceihdar Endocrinology, 90 (1992) I- 15 0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/92/$05.00

MOLCEL 02867

Molecular cloning and cDNA sequence analysis of coho salmon stanniocalcin Graham F. Wagner a, Gabriel E. Dimattia b, James R. Davie ‘, D. Harold Copp d and Henry G. Friesen b aDepartment of Physiology, Unicersity of Western Ontario, London, Ontario, Canada, ’ Department of Physiology and ’ Department of Biochemistry and Molecular Biology, Unicersity of Manitoba, Winnipeg, Manitoba, Canada, and d Department of Physiology, Unicersity of British Columbia, Vancoucer, British Columbia, Canada

(Received 24 June 1992; accepted 8 August 1992)

Key words: Stanniocalcin; Molecular cloning; cDNA sequencing; (Coho salmon)

Summary Stanniocalcin (ST0 (formerly known as both teleocalcin and hypocalcin) is an anti-hypercalcemic, glycoprotein hormone that is produced by the corpuscles of Stannius (CS), endocrine glands that are confined to bony fishes. The hormone has a unique amino acid sequence and exists as a disulfide-linked homodimer in the native state. In previous studies, we have described the purification and characterization of two salmon STCs, and examined the regulation of hormone secretion in response to calcium using both in vitro and in vivo model systems. This report describes the molecular cloning and cDNA sequence analysis of a coho salmon STC messenger RNA (mRNA) from a salmon CS lambda gtl0 cDNA library. The STC mRNA in salmon is approximately 2 kilobases in length and encodes a primary translation product of 256 amino acids. The first 33 residues comprise the prepro region of the hormone, whereas the remaining 223 residues make up the mature form of the hormone. One N-linked, glycosylation consensus sequence was identified in the protein coding region as well as an odd number of half cysteine residues, the latter of which would allow for interchain bonding or dimerization of monomeric subunits. In addition, three sites were identified in the mature protein core of STC (two dibasic, one tribasicl that may be acted upon by endopeptidases to produce truncated forms of the hormone. In support of this latter possibility, Western blot analysis revealed multiple molecular weight forms of STC within salmon glands.

Introduction Stanniocalcin (STC) is a glycoprotein hormone that is synthesized and secreted by the corpuscles of Stannius (CS), endocrine glands that are unique to teleostean and holostean fishes (Stannius, 1839). The cells comprising the CS are derived embryologically from the kidney tubules and coalesce during development to form two or more glands that lie on or within the kidneys Krishnamurthy, 1976). In the broadest sense STC performs a calcitonin-like function in fishes, as its main role is the prevention of hypercalcemia. As in the case of calcitonin, STC secretion is positively regulated by extracellular levels of

Correspondence to: Dr. Graham F. Wagner, Department of Physiology, Faculty of Medicine, University of Western Ontario, London, Ontario N6A SCl, Canada.

ionic calcium both in vitro and in vivo (Wagner et al., 1989, 1991; Glowacki et al., 19901. However, the similarity between the two hormones diverges sharply from this point onward. For instance, calcitonin is a small polypeptide comprising 32 amino acids (Breimer et al., 19881, whereas STC is a glycoprotein with a molecular weight of approximately 50 kDa and shares no similarities with calcitonin in amino acid sequence (Wagner et al., 1986, 1988a; Butkus et al., 1987; Lafeber et al., 1988a). The two hormones also act through different target tissues. Calcitonin functions via the inhibition of osteoclastic bone resorption (Friedman et al., 1965; Milhaud et al., 1965; Aliapoulios et al., 19661, whereas STC acts on the fish gill to reduce the inward transport of calcium from the aquatic environment (So and Fenwick, 1977, 1979). The different target tissues affected by the two hormones underscore the different mechanisms employed by fishes and higher vertebrates in the maintenance of calcium homeostasis. In contrast to

x

higher vertebrates that rely upon bone as a calcium reservoir, the environment is the principal source of calcium in fishes and the gill transport system is used to draw from this reservoir for the regulation of plasma calcium levels. Our studies on STC commenced with its isolation and characterization, first from sockeye salmon and subsequently from coho salmon. We found that the purified hormone had a unique N-terminal sequence (STC ,_4r,) and that STC appeared to have the structure of a disulfide-linked homodimer. The hormone was also glycosylated, with the carbohydrate moiety consisting principally of mannose, glucosamine and sialic acid. Furthermore, the salmon STCs inhibited gill calcium transport when administered to salmon and were capable of reversing the effects of stanniectomy (accelerated gill calcium transport) in the North American eel (Wagner et al., 1986, 1988a). In a series of follow-up studies, we have since examined the STC cells by immunocytochemistry (Wagner et al., 1988b), characterized the stored and secreted forms of the hormone (Gellersen et al., 1988) and studied the regulation of STC secretion by calcium in vitro and in vivo (Wagner et al., 1989, 1991; Glowacki et al., 1990). As a first step towards studies on the STC gene, in this report we have determined the complete primary structure of a coho salmon STC mRNA through molecular cloning and cDNA sequence analysis. Material

and methods

supplied by Kockwood Experimental Hatchery, Department of Federal Fisheries, Balmoral, Manitoba. North American eel (Anguilla rostrata) CS were kindly provided by Dr. J.C. Fenwick, Department of Biology, University of Ottawa, Ottawa, Ontario. Corpuscles of Stannius and testes from a representative holostean fish, the bowfin (Amia calr,a), were kindly supplied by Dr. J.H. Youson, Department of Zoology, Life Sciences Division, University of Toronto, Toronto, Ontario. All tissues were frozen immediately on dry ice and maintained at - 70°C until use. Library construction and cDNA sequence analysis

Total RNA was extracted from coho salmon CS using the guanidinium isothiocyanate/cesium chloride method (Chirgwin et al., 1979). Poly(A)+ RNA was isolated by two passages through a column of oligo dT cellulose. The cDNA library was constructed using a commercial kit based on the method of Gubler and Hoffman (1983) and primed with both oligo dT and random oligonucleotide primers. The oligo dT and random primed cDNAs were pooled, blunt ended, ligated to EcoRI linkers and inserted into the cloning vector, lambda gtl0. The library was screened with a “‘P-labelled, oligonucleotide probe (5lmer), synthesized according to the established N-terminal sequence of coho salmon STC (STC,-,,,; Wagner et al., 1988a) based on the preferred codon usage for eel STC (Butkus et al., 1987). The sequence of the oligomer was as follows:

Reagents, biochemicals and supplies

Fine chemicals and reagents were purchased from BDH, restriction enzymes, modifying enzymes and DNA size markers were from Boehringer-Mannheim, radioisotopes and cloning kits were from Amersham, plasmids and the Erase-a-Base kit for preparing nested deletions were from Promega, chromatography supplies and random prime labelling kits were from Pharmacia, hybridization membranes were from Micron Separations, electrophoresis reagents and Western blotting supplies were from BioRad and microbiological supplies were obtained from Difco. Source of tissue

Tissues from teleost fishes were obtained from the following sources. Coho salmon (Oncorhynchus kisutch) CS were collected from freshwater spawning adult males and females at the Capilano Hatchery, North Vancouver, British Columbia. Chinook salmon (Oncorhynchus tschawytcha) testes for DNA isolation and CS from saltwater-adapted coho salmon were kindly supplied by Dr. E.M. Donaldson, Pacific Environment Institute, West Vancouver, British Columbia. Tissues and CS from rainbow trout (Oncorhynchus mykiss) and arctic char (Salvelinus alpinus) were kindly

ccc

CTG

AL.4

LEU

Prior to screening the library with the oligomeric probe, Northern blot analysis (Maniatis et al., 1982) was used to establish that it hybridized to a single 2 kilobase (kb) message that was confined to the CS in coho salmon (not shown). All cDNA clones were plaque purified and subcloned into plasmids [pGEM-SZf( + > or PGEM 7Zf(+)] prior to cDNA sequence analysis using the dideoxy chain termination method @anger et al., 1977). Nested deletions were prepared in PGEM 7Zf(+ > using the Erase-a-Base system to facilitate cDNA sequence analysis. The GCG Program (Genetics Computer Group, Version 7, 1991) was used to compare the derived nucleotide and deduced protein sequences of salmon STC with those reported for Australian eel STC (Butkus et al., 1987) and to calculate the isoelectric point of the salmon hormone. Northern blot analysis

Northern blot analysis was performed using established methods (Maniatis et al., 1982). Using the guani-

9

dinium/cesium chloride method (Chirgwin et al., 19791, total RNA was isolated from the CS of two additional representative teleost fishes, rainbow trout and arctic char, and a representative holostean fish, the bowfin. The bowfin was included as it is considered to be one of the most primitive fishes with CS glands and predates the teleost fishes in evolutionary history. The species is restricted to the temperate freshwaters of eastern North America and does not migrate to sea like the salmon. As we have already established that bowfin STC crossreacts strongly with salmon STC antiserum (Marra et al., 19921, whether or not the mRNA encoding the bowfin hormone hybridized to salmon cDNA probes was of particular interest. Total RNA was also isolated from the heart, brain, gill, kidney and liver of rainbow trout. All RNA was subjected to electrophoresis in 1% agarose/formaldehyde gels at concentrations of lo-50 Fg RNA/lane and blotted onto nitrocellulose. Following a 2 h pre-hybridization period (under the same conditions used for hybridization with probes), blots were probed with random primed, 32P-labelled cDNAs (pure insert; 1.5 X lo6 dpm/ml) under conditions of high stringency (50% formamide, 6 x standard saline citrate @SC), 1.25 X Denhardt’s solution, 100 pg/ml salmon sperm DNA, 0.1% sodium dodecyl sulfate (SDS) at 42°C) or low stringency (same hybridization solution minus formamide at temperatures from 37650. Blots probed at lower conditions of stringency were washed for 4 x 15 min in 2 x SSC/O.l% SDS at room temperature followed by 2 x 30 min in 2 x SSC/O.l% SDS at the same temperature used in the initial hybridization. When probing membranes that contained RNA from bowfin CS, salmon CS RNA was included as a positive control. Southern blot analysis

For Southern blotting (Maniatis et al., 19821, genomic DNA was isolated from the testes of chinook salmon (a species of pacific salmon closely related to and used in lieu of coho salmon), rainbow trout, arctic char and the bowfin. Tissue was pulverized in liquid nitrogen and digested overnight (37°C) in a solution of proteinase K (100 pg enzyme/ml in 0.1 M Tris, 0.1 M NaCl, 5 mM EDTA, 1% SDS, pH 8.0) at a concentration of 0.5 g tissue/ml. Following a treatment with RNase A (200 pg/ml for 15 min at 65”C), the digest was extracted 2 times each with phenol, phenol/ chloroform and chloroform/ isoamyl alcohol (25 : 1). Aliquots of DNA (15 pg) were digested overnight with BamHI, Hind111 and PstI. The restriction digests were then subjected to electrophoresis in 0.8% agarose gels and blotted onto nitrocellulose. Prior to hybridization with radiolabelled probes, genomic blots were subjected to a 2 h period of prehybridization at 65°C in 6.6 x SCP, 1% sarkosyl, 100 pg/ml salmon sperm DNA

and 1 x Denhardt’s solution, where 1 x SCP consisted of 0.1 M NaCl, 0.03 M dibasic sodium phosphate and 1 mM EDTA, pH 6.2. Blots were then hybridized overnight at 65°C in hybridization solution of the same formulation containing 10% dextran sulfate and a random primed, 32P-labelled cDNA probe (pure insert, 2 x lo6 dpm/ml). Blots were then washed 2 X 15 min in 6.6 x SCP containing 1% sarkosyl at room temperature, 1 x 1.5 h in 1 x SCP, 1% sarkosyl at 65°C dried and exposed to X-ray film. Western blot analysis

Western blot analysis was carried out on coho salmon and North American eel CS extracts using a polyclonal antiserum generated against sockeye salmon STC and a previously established Western blotting protocol (Wagner et al., 1988b). The STC antiserum has been characterized by Western blot analysis as being highly specific for pro and mature STC in numerous fish species (Gellersen et al., 1988; Wagner et al., 1988b, 1991; Marra et al., 1992). In preparation for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), frozen CS from coho salmon and North American eel were homogenized directly in SDS-PAGE sample buffer containing 5% P-mercaptoethanol at a concentration of 0.1 mg/ml. The homogenates were centrifuged at 1000 Xg and supernatants were subjected to SDS-PAGE in 12% acrylamide gels (LaemmIi, 1971) under reducing conditions. Marker proteins labelled with 14C were included for estimation of molecular weight. Western blot analysis was carried out as described previously using a 1: 400 dilution of STC antiserum, followed by a 1: 3000 dilution of peroxidase-conjugated, goat anti-rabbit immunoglobulin (Wagner et al., 1988b). After development, the nitrocellulose membranes were dried and exposed to X-ray film to visualize the protein markers. Results Characterization of coho salmon STC cDNAs

The yield of recombinants was 5.2 x 105/pg of CS poly(A)+ RNA. In an initial screening of 1.3 x lo6 recombinants, 33 positive cDNA clones were identified. Of the cDNAs that were isolated and characterized, one was a near full-length clone encoding part of the pro sequence, the complete protein coding region of mature STC and the entire 3’ untranslated region 07B2; 1761 bp in length containing nucleotides 1741934). This clone was RE digested with EcoRI and PstI into three smaller fragments, designated 2.9 (nucleo~i;spts7n4-274; Arg_ ,-Asn + 291,2A.3 (nucleotides 269675-1934; +2FGln + 165 1 and 3A.l (nucleotides and the complete 3’ untranslated Gin+ 165-G1U+223 region), each of which was subcloned. An additional 274 bp cDNA, clone 4 (nucleotides l-2741, was isolated that contained sequence which

commenced in the 5’ untranslated region, continued into the protein coding region and overlapped with subclone 2.9. Clone 4 encoded 86 bp of the 5’ untranslated region, the prepro region of STC and 29 residues of the mature hormone, encompassing residues Met p33 to AsP+~ Six additional cDNA clones which encoded sections of the 3’ untranslated region were isolated and sequenced. Northern blot analysis confirmed that all of these cDNA clones hybridized to the same sized message in total CS RNA when employed as probes (not shown). The sequencing strategy employed and a composite of the salmon STC mRNA are shown schematically in Fig. 1.

salmon STC to lower molecular weight forms. Moreover, in corroboration of our earlier evidence suggesting that salmon STC exists as a disulfide-linked homodimer in the native state, there were 11 half cysteine residues in the mature protein core, leaving one unpaired half cysteine in the monomer to participate in interchain bonding with a second monomeric subunit. There was also an additional half cysteine in the presequence of salmon STC that may play a role in stabilizing the monomeric primary translation product. Using the GCG ISOELECTRIC program, the isoelectric point calculated for the non-glycosylated, mature protein core (STC,_,,,) was 7.47. Sequence similarities between salmon and eel STC

Structural analysis of coho salmon STC mRNA

Comparisons were made at the nucleotide and protein level between salmon and Australian eel STC (Butkus et al., 19871, the only other species in which STC has been cloned and sequenced. The entire protein coding region of salmon STC (prepro STC) shared 66.8% identity with the eel hormone at the nucleotide level (Fig. 3). This was based on a 774 bp overlap between the two species and taking into account one gap inserted in the salmon sequence and two in the eel for maximum alignment. When individual regions of the salmon and eel hormones were compared separately, there was considerably less similarity in the nucleotide sequence encoding the prepro (bp 87-185 in salmon; 51% based on a 100 bp overlap) and Cterminal regions (bp 6966854 in salmon; 59.4% based on a 159 bp overlap) than that encoding the N-terminal and mid-regions of the molecule (bp 186-695 in salmon; 72% based on a 510 bp overlap). This particular point of division between the mid- and C-terminal regions (bp 695-696) of salmon STC was selected on account of the extreme divergence in deduced amino acid sequence that occurred thereafter, between bp 696-854, when compared with the eel (see below). At the protein level, mature salmon STC was eight residues shorter than mature eel STC (Fig. 2). In total, there was 62% amino acid sequence identity between

The cDNA and deduced amino acid sequence is shown in Fig. 2 and reveals that salmon STC mRNA contains all the salient features of a eukaryotic mRNA, such as start (bp 87-89) and stop codons (bp 855-8571, a polyadenylation signal (bp 1868-1873) and a poly(A) tail. The mRNA encodes a primary translation product of 256 amino acids. As the deduced amino acid sequence from Phe,, to Ile+,, is identical to that derived by N-terminal sequence analysis of purified coho STC (Wagner et al., 1988a), it appears that an endopeptidase cleavage at Arg_ ,-Phe,, yields the mature salmon hormone (STC,_,,,). Hence, the first 33 residues from Met _33 to Argg, comprise the prepro region of the hormone. The cleavage site between pre and pro STC remains to be established. Some notable features within the mature protein core of salmon STC included a single glycosylation consensus sequence at residues 29 through 31 (AsnSer-Thr), the same position as suggested by amino acid sequence analysis of purified sockeye and coho salmon STC (Wagner et al., 1986, 1988a). In addition, there were two dibasic pairs (Arg,,-Arg,, and Lys,,,His,,,) as well as a single tribasic site (Arg,,,-Lys,,,Arg,,,) within the mature protein core that could be employed for the post-translational processing of

500

1000

I

1500

I

I

1 preproSTC

5’ 4 4-F

< )

t 4

3’

I

) k

-+

)

100bp

Fig. 1. The sequencing strategy employed for cDNAs encoding salmon stanniocalcin. The protein coding region is represented by the open box (prepro STC) and the 5’ and 3’ untranslated regions by the heavy solid lines. The direction of sequencing is indicated by the arrows. Lines with arrowheads on both ends represent individual cDNA clones or subcloned fragments that were sequenced entirely in both directions.

11

the two species in the mature protein core based on a 223 amino acid overlap (the length of mature salmon STC). As mentioned above, a large proportion of the sequence dissimilari~ was due to the near complete divergence in amino acid sequence that occurred after proline,,,,. From Phe, to Pro,,,, the two species shared 78% sequence identity, whereas thereafter between

180 CGT GCA CGC TTC A20 ALA Ana PHp THR

TIC

mu

Am Ama

cTG

mu

TTC Pm

SER

210 200 MC AGC CCC TCG OAT GTG ‘XT Aa” 613 PRO SE2 Am VAL ALA SER

240 230 GGC GCT CTA GCC Gs ALA Lm h3.A VAL

axa

cw

270 GAG AAT

190 TCA TCC mu Llm ALA

TCT ACC

260 250 GGA T&T GGT ACG m GCC TGC CT0 %LY am! PEE ALA CYS I.m +*7

%rx cm

smt

ALA 300

290 280 TGT GAC ACT GAT GGC ATG CAT CY8 MD p%p ASP %bY YFI 81%

310

GAT ATC TGT A%gP IX.11 CYB GUI

CAA Ott, +42 ARG

340 350 330 GCA GCT ACC TlT hhc ACh CA0 GGT AAG ACA ALA a&A TBO par Ma TEX am %LY LYE TER +57

320 l-~-l’ CAC KC PIIS *Ia m

LYS

GLY

LN

220 AGO ‘PDT AR% CY8 +1*

020 410 AAA GTC TTT CAD ACC ATC AGG CGC TGT

430 GGA OTC Tl!C

440 CA0 AGA ATC

+67

Glnm-Glum (bp 696-854) the level of identity dropped to 7.5%. In the prepro region, there was roughly the same sequence identity between species at the amino acid level (48.5%) as was observed at the nucleotide level (51%). Salmon STC also contained one additional residue in the prepro region of the molecule (33 in salmon, 32 in eel). The mRNA encoding STC was the same size (- 2 kb) in freshwater and saltwater-adapted salmon (not shown) and similar in size among the representative salmonids that were examined (Fig. 48; coho salmon, rainbow trout, arctic char). Expression of the STC gene also appeared to be unique to the CS as no hybridization was observed in heart, liver, brain, kidney or gill (not shown). In addition, trace amounts of a larger message were observed in the CS of arctic char after a prolonged film exposure and may correspond to the primary transcript or heteronuclear STC mRNA in this species (Fig. 4A). No additional mRNAs hybridized to the probes at lower stringency conditions. In contrast, we were unable to detect the STC message in bowfin CS RNA even at the lowest conditions of stringency

TATI’AATGGA

1094 AGTTTOOGCC

TTTXJGGGhAG CATCTCTCS

ThhATGCTGT

1154 TGAAGGTATT

TTGAThCF,‘G

ATGAhGGGGh

GhTGAAACcT

GTPA‘XI’AGA

GmAATGT

t%AGGAWAT

ATCTCCTCAG

AATAGACTCG

ACTBJLWA~

AGhGCThlTG

I.274 AAAAGTCFGA

ACATTTAATA

PlBACAAGW3

AAACAWPCA

AATGCCACCT

AAGAAAACDA

3.334 ACCATCACTG

TAGTECATT

GGATfTCAhC

GTGGCCACTA

CGGCCATAAC

ATCCCCGTTT

1394 GGACCAGTCA

TTAAhGACCG

ATGGGTATAT

TATTATAATA

ATA~A~A

TATS-A’PTTT

1454 CTTACAGAAT

GTTTAlWtAT

GASQPI’GTGT

TG’ZATCTAGT

TGTAACTCGG

TTPZAGTTTC

CCCCAGTGGG

TGGGGTTTGA

CATCCTGATA

TGACGTCACT

GGCTGATGTA

TTGCTCTATG

1574 AGGATGTCAC

CACCTCAGAG

GGACAGTGTG ATCTCACCAT

GCTGGTTGTG

GGACTCACCG

1634 CGTCCCTCTC

TGTCTTCATC

TGTGTTAA~

TAAGAKCTG

TAGTGTGTAA

AGACAW’ATA

1694 GAGTGATCCT

WGCTGTGTT

CCTCAGATGT

GGTTATGTGG

STAFF

GAGAWXXQT

1754 GTGAGAA’IGT

GTGCTAGTCA

GGTACI’ATAC

ACCTTGGGGC

TFXGGTCAT

TCTCTCIGCT

I.614 ATGAATAChT

TTTGTGACW

TCATAATTTT

GWTGAGGTG

ATTTGTGTCG

AGAffCACTG

1874 GTGEE&&C

GCTCACAGTT

TCM

ACACACACAT

AGCTTTACCC

TCUATGATT

TGT’ETAATG

TAGTAWGh

TCTGTGTGGG

AAGGCTAAAT

LE”

ATT TLS

450 TCT s8R

490 480 470 460 GAG OTC CAG GAG GAG TGT ThC AGT AGA CT0 GAC ATC TGT (il.0 PAL 0I.R BLD am me TYB. SLR Apa I.80 ASP ILZC CY8 +x01 LYS

500 GGT GT-3 GCT CGC TCT AAC am VAL ALA A%% Elm hali 0x24 SER

540 GTC C‘X’ VAL Pm

Tee Em

550

GCA

CAC TIC ALA SIB Pm! SER GLN

590 CTG CIA Gee lyiT LSP LSV ALA CY8 TRR

630 GGG CPT am ~2%

CTG ‘ZAG AAC AAA Lsu %LR Aa% u THR 720

AAC TCh WC a81

CCC AAC Pno As*

560 AGG TAC AR% TYR

600 OAT GAG %A0 AU P&w awl aw Tsh ASP

570 TAC AGC ACT TY% BBR Tmt

ssmrm

GLY PRO “AL

660 ACT em “AL

700 690 CAC TGC CCC CAG GGT TCT AAC cm PI&D %Ln %LY EmI A8x hLA PRO SER ALA ALA

+-ix,

580 CTG CTC mo I,lm

CAG

ow

610 620 GTG DC-T GTG OTC AGG GCA PAL ALA VA& VAT” AR0 ALA GL” GLN

640 650 GT’C GCT hGG CTG GGG CCA GAC ATG GA.A vAt UL *Q I*O bltx ~2% MP m or.0 SER %LU GLU GLY

680 TTG LBU

LB” 510 520 530 CCT GAG GCC ATT CGA %AG GTG GTG CAG PRO OLtl ALh ILE %LY %LV VhL VAL aw RGP ALA

CTC TX rm mm

+132

,147

670 CA% GtR +162

710

ChG GGT Ccp %&I am PRO ,I,, GLY l!R%

740 730 790 760 CCC GCT GGC TGG CGC TGG CCA A%‘G GGG TCG CCT CCT PPO aa axa 1110 ana =P PRO OLY 8m PRO PRO +x92 GLY FLY sml ARG CYS PRO TRP GLY

me

770 TCC TTC AAG ATC CAG CCC 622 mm LYll ILR am m% PRO CYS SER ARG SER SSR

790 780 800 AGC hTG AGh GGA AGh GAC CCC ACC CAC 8sl m AR% O&I Aal% Am *no Tm e.8 +207 PRO TRR CYS ALA PRO GLY T%R PRO PRO

%A%

atu +*23 PRO ARG LEU

Ar.h

mu

MET ASP

as

PRO

+231

864 874 884 404 914 894 TAGAFPGGAG AAGAGCAGGC AGhCAThCAC ACCACTTATA CCTTAAGCAT ACATYCACAT

GTACACRCAC

ACATACACAC

ACCACAGCTA

CCTTAAACAC

AAAhCAChCT

974 CATGATAGCP

TTGCTCACAC

ACACACTGAC

TCACACGCAC

ACTGACACAC

ACACATTWC

1034 ACACACATGC

1934 V

-

AM&AAAM

-

Fig. 2. cDNA and deduced amino acid sequence of coho salmon STC mRNA. The mRNA encodes 256 residues, 223 of which comprise mature STC. The residues that differ in Australian eel STC, or which are missing entirely (-1, are shown under the salmon sequence. The initiator methionine, N-terminal phenylalanine of mature salmon STC, asparagine-linked gIycosylation consensus sequence, potential dibasic and tribasic cleavage sites and the ~iyadenylation signal are all underlined. The cleavage site between pre and pro STC, which occurs between Ala_,,-Tyr_,s in the Australian eel, has not been determined in salmon.

(6 x SSC, 37°C) using RNA concentrations as high as 100 pg/lane (results not shown). Southern blots were performed on genomic DNA isolated from three representative salmonids (chinook salmon, rainbow trout and arctic char) following digestion with BamHI, Hind111 and PstI. The blots were first hybridized with subclone 2.9 (101 bp; nucleotides 174-274) that encoded residues Argg, to Asn,,, and washed at high stringency. Following exposure to film, the blots were stripped and re-hybridized under the same conditions with a 1666 bp EcoRI fragment of clone 17B2 (nucleotides 269-19341 encoding 194 residues of mature STC and the complete 3’ untranslated region. The results of this study revealed in the case of BamHI and Hind111 digested DNA, that at

-28s

-18s

Fig. 4. (A) kidney

Northern

tolal

illustrating

RNA

both the mature

that are confined jected

encoding

salmonid,

and heteronuclear

to CS tissue. Total

to Northern

stringency

blot analysis of corpuscle of stannius (CS) and from a representative

RNA

blot analysis, probed

with a random the N-terminal

Mature

STC

mRNA

behind

1%

ribosomal

prime-labelled region) RNA.

Heteronuclear

fR)

STC

representative @g/lane)

mRNA

salmonids

encoding (2

kh).

from three representative

char and rainbow

trout

to film for

2 kb in length ahead

mature

was sub-

conditions

of high

STC subclone (2.9; 101 bp

second, less intense band migrating The

(30 Kg/lane)

under

and exposed

is approximately

the arctic char,

forms of STC mRNA

Samples

STC of 2%

is the

were subjected

mRNA

is the

ribosomal

RNA.

same

of CS

salmonids

IO days.

and migrates

size among

total

RNA

(30

coho salmon, arctic

to Northern

blot analysis

and probed under the same conditions as described in (A).

Fig. 3. Comparative

nucleotide

tralian

eel prepro

prepro

STC is derived

numbered

STC

overlap

(minus

The

is numbered

article.

the stop codon

sequence

two species. The

initiator

sequence

methionine.

and the termination

encoding

eel

and nucleotides

are

nucleotide

sequence

a 6 bp gap in the

similarity

N-terminal

between

the

phenylalanine

of

codons have been highlighted

hoth species.

of

the basis of a 774 bp

and including

there was 66.X%

STC

The

as in Fig. 2. On

salmon sequence) mature

DNA

from Butkus et al. (1987)

as in the original

salmon

sequences of coho salmon and Aus-

stanniocalcin.

in

least two fragments of genomic DNA hybridized to both the larger and the smaller cDNA probes (not shown). In the case of PstI digestion, in all species there were multiple DNA fragments that hybridized to both probes (Fig. SA and B). In addition, several fragments hybridized uniquely to one probe or the other (see arrows and arrowheads on figures). There are no BamHI sites and only one Hind111 site (bp 163-168) in the entire sequence of coho STC, the latter of which lies outside the regions encoded by the cDNAs used as probes. However, given that there is only one PstI site tbp 675-680) in the entire sequence and the unlikelihood of there being multiple introns, all containing PstI sites, within the 101 bp stretch encoded by subclone 2.9, the results obtained with PstI suggest that there is more than one copy of the STC gene in salmonids. Alternatively, the multiple PstI fragments may be indicative of one or more pseudogenes,

13

. .