Several splicing variants of isl-1 like genes in the chinook salmon (Oncorhynchus tschawytscha) encode truncated transcription factors containing a complete LIM domain.

Several novel cDNA clones have been isolated from a chinook salmon pituitary cDNA library. Sequence analysis of these clones indicates that they are closely related to the rat LIM domain homeobox gene, isl-1. Due to differential splicing, several of the clones encode truncated polypeptides containing a complete copy of the LIM domain without the homeodomain and C-terminal activation domain. The roles of these truncated polypeptides are discussed.

Seasonal variation in the level of antifreeze protein mRNA from the winter flounder.

Livers from winter flounder collected at monthly intervals throughout the year were analyzed for their content of antifreeze protein mRNA. Putative mRNA was detected in liver RNAs from summer fish by liquid hybridization to kinetically purified antifreeze protein cDNA. These mRNA sequences were shown by Northern blot analysis to be of the same length as mature antifreeze protein mRNA isolated from winter fish, and were able to direct the incorporation of alanine into a translation product which comigrated with antifreeze preproprotein. The build-up of antifreeze protein mRNA levels in the autumn and their decline in the spring to summer levels were measured by liquid hybridization. These seasonal fluctuations match closely, but slightly precede, the rise and fall in plasma osmolality due to the presence of antifreeze protein. In mid-winter 0.5% of the total liver RNA is antifreeze protein mRNA, and although by late August the level of this mRNA has declined to 0.0007% of the total RNA, at no time during the summer is the mRNA undetectable. These results suggest that antifreeze protein production is more likely to be regulated by changing the rate of transcription of their genes than by switching them between active and inactive states.

Single crystals of a type III antifreeze polypeptide from ocean pout.

Antifreeze protein (HPLC-2) from ocean pout was purified from serum using column chromatography on Sephadex G75 and reverse-phase HPLC columns. Single crystals were grown by batch methods at 4 degrees C from a 1.5 M solution of ammonium sulphate (pH 7.1). The crystals diffracted to about 2.5 A resolution at 4 degrees C and belong to the monoclinic space group P2(1), with cell parameters: a = 39.77 A, b = 58.51 A, c = 30.27 A, beta = 102.28 degrees, with two molecules of 6000 M(r) per asymmetric unit.

Single crystals of a winter flounder antifreeze polypeptide.

Component 6 of the winter flounder's antifreeze polypeptides has been crystallized. The space group is P21, with cell parameters of a = 38.14 A, b = 37.19 A, c = 21.82 A, beta = 101.5 degrees. There are two molecules of 3300 Mr per asymmetric unit.

The chinook salmon gonadotropin II beta subunit gene contains a strong minimal promoter with a proximal negative element.

The salmon pituitary expresses two distinct gonadotropins, gonadotropin I (GTHI) and gonadotropin II (GTHII). These two hormones are synthesized in distinct pituitary cells and secreted at different stages during the reproductive cycle. To study the transcriptional regulation of the hormone-specific beta-subunit of GTHII (sGTHII beta) gene, approximately 3.5 kilobases of the 5'-flanking region was characterized and sequenced. The pituitary specificity of sGTHII beta was examined by analyzing sGTHII beta promoter activity in homologous primary pituitary cells derived from spawning rainbow trout and in a collection of heterologous cell lines. Various lengths of the 5'-flanking region of the sGTHII beta gene were ligated into a vector encoding the bacterial chloramphenicol acetyltransferase (CAT) gene, and the resulting sGTHII beta/CAT chimeric constructs were analyzed using transient expression systems. Several constructs (-3500CAT, -1260CAT, -563CAT, and -39CAT) displayed readily detectable CAT activity in the pituitary cells derived from spawning male and female rainbow trout. In contrast, three of the constructs (-3500CAT, -1260CAT, and -563CAT) were expressed only at background levels in a variety of heterologous cell lines, suggesting that the 5'-flanking sequence of sGTHII beta contains information dictating its pituitary specificity. A silencer sequence (-95 to -35, pSil) was identified, which might function to repress sGTHII beta gene expression in the nongonadotropes or in the gonadotropes at developmental stages that precede final maturation and spawning.

Differential recruitment of steroid hormone response elements may dictate the expression of the pituitary gonadotropin II beta subunit gene during salmon maturation.

The role of testosterone (T) and 17 beta-estradiol (E2) in the control of chinook salmon gonadotropin II beta subunit (sGTHII beta) gene was examined. Both E2 and T specifically stimulated GTHII beta gene expression in cultured juvenile rainbow trout pituitary cells. 5'-Flanking regions of the sGTHII beta gene linked to the chloramphenicol acetyltransferase (CAT) expression vector were transfected into these pituitary cells, and cultures were treated with steroid hormones. Estrogen-stimulated CAT activity occurred with constructs containing a 13-base pair estrogen responsive element (ERE) sequence [proximal ERE (pERE)] located at -273 to -260 upstream of its transcriptional start site. Binding specificity of pERE was confirmed by mobility shift and DNA methylation interference assays using the DNA binding domain of the human estrogen receptor. Interestingly, the pERE functioned to derepress the activity of the proximal silencer (pSil) only in the pituitary cells of juvenile trout but not in cells derived from maturing and sexually matured fish. Another potential ERE sequence comprised of three tandemly linked half-ERE palindromes was located from -2736 to -2659 [distal ERE (dERE)] of the sGTHII beta gene. Distal ERE might be responsible for the steroid responsiveness of the longest sGTHII beta/CAT construct (-3500CAT) observed in the pituitary cells of maturing fish. The function of pERE and dERE were further examined in the heterologous HeLa cells by mutagenesis and cotransfection with a rainbow trout estrogen receptor expression vector. Disruption of the palindromic structure of pERE severely impaired its function. When the sequences between pERE and dERE were deleted, a 200-fold increase in CAT activity was observed in response to E2. A model is proposed to describe the regulation of GTHII beta gene expression at different reproductive stages.

Teleost FTZ-F1 homolog and its splicing variant determine the expression of the salmon gonadotropin IIbeta subunit gene.

Steroidogenic factor 1, a member of the fushi tarazu factor 1 (FTZ-F1) subfamily of nuclear receptors, is a key regulator in mammalian reproduction. From an embryonic complementary DNA library, the zebrafish homolog of FTZ-F1 (zFF1A) and an alternatively spliced variant (zFF1B) were isolated. zFF1B represented a C-terminally truncated version of zFF1A. Whole mount in situ hybridization and reverse transcriptase-PCR analysis revealed that both zFF1A and B transcripts were present in the developing pituitaries, adult fish brain, gonads, and liver, albeit zFF1B messenger RNA was absent in testis. Comparison of the primary sequences of zFF1 with those of other FTZ-F1 subfamily members showed a close structural relationship between the mouse liver receptor homolog, which activated the alpha1-fetoprotein gene in rodent liver. However, similar to mouse steroidogenic factor 1, zFF1A regulated chinook salmon gonadotropin IIbeta subunit gene expression. On the contrary, zFF1B, which could bind a consensus gonadotrope-specific element with an affinity similar to that of zFF1A, lacked both the trans-activation function and synergistic interaction with the estrogen receptor. Furthermore, cotransfection studies in HeLa cells showed that zFF1B was a strong competitor for the action of zFF1A on the chinook salmon gonadotropin IIbeta subunit gene promoter. Our investigation suggests that 1) zFF1 represents an ancestor protein of the vertebrate FTZ-F1 homologs; 2) the antagonistic relationship between zFF1A and -B may dictate the expression of the FTZ-F1 target genes in a variety of tissues, including the pituitary; and 3) the naturally occurring zFF1B provides evidence that the C-terminal portion of zFF1A (80 amino acid residues) contains a major trans-activation function and a protein-protein interface.

Steroidogenic factor 1 and estradiol receptor act in synergism to regulate the expression of the salmon gonadotropin II beta subunit gene.

The orphan nuclear receptor steroidogenic factor-1 (SF-1) regulates the expression of several genes involved in the reproductive function and development of the adrenal, the gonads, and the pituitary gonadotropes. It also confers the gonadotrope-specific expression of the glycoprotein hormone a subunit gene by the binding to a gonadotrope-specific element (GSE). In this study, we have shown that SF-1 transactivates the salmon gonadotropin II beta subunit (sGTHII beta) gene expression. SF-1 alone offered a slight but significant enhancement on sGTHII beta promoter activity (7.2 +/- 0.6 fold). However, it stimulated sGTHII beta gene expression dramatically (127 +/- 37 fold) when combined with the estrogen receptor (ER). This synergistic interaction was specific for sGTHII beta promoter as well as for both SF-1 and ER and was estradiol-dose dependent. 5'-Deletion studies of the sGTHII beta promoter identified two putative SF-1 binding sites (GSE) and one previously identified proximal estrogen-responsive element (pERE) at -274 bp involved in this activation. The two GSE sequences located at -354 bp (sGSE(3) and -162 bp (sGSE(2) upstream of the transcription site, although imperfect as compared with the consensus GSE, bound specifically to the in vitro-translated mouse SF-1 protein. 5'-Deletion studies, competition experiments, and site-directed mutagenesis showed that binding to pERE and GSE(2) were necessary for the SF-1/ER synergistic effect. These studies suggest that the synergistic interaction of SF-1 and ER, possibly through cooperative binding or protein-protein interaction, is essential in conferring a cell type-specific expression of the GTHII beta subunit gene.

Phylogenetic specificity of prolactin gene expression with conservation of Pit-1 function.

In mammals, the pituitary POU homeodomain protein, Pit-1, binds to proximal and distal 5'-flanking sequences of the PRL gene that dictate tissue-specific expression. These DNA sequences are highly conserved among mammals but are dramatically different from PRL 5' sequences in the teleost species, Oncorhynchus tschawytscha (chinook salmon). To analyze the molecular basis for pituitary-specific gene expression in a distantly related vertebrate, we transfected CAT reporter gene constructs containing 2.4 kilobases (kb) 5'-flanking sequence from the salmon PRL (sPRL) gene into various cell types. Expression of the sPRL gene was restricted to pituitary cells, but in rat pituitary GH4 cells levels of expression were at least 90-fold lower than those obtained with a -3 kb rat PRL (rPRL) construct. Conversely, in primary teleost pituitary cells, -2.4 kb sPRL/CAT was expressed at levels about 10-fold higher than -3 kb rPRL/CAT. To determine whether species-specific transactivation by Pit-1 was sufficient to explain these species differences in PRL gene expression, we isolated a cDNA clone encoding the salmon Pit-1 POU domain and constructed a rat Pit-1 expression vector that contained salmon Pit-1 POU domain sequences substituted in frame. The chimeric Pit-1 encoded 14 amino acids unique to salmon. Coexpression of rat Pit-1 with salmon or rat PRL/CAT in transfected HeLa cells resulted in specific and strikingly comparable levels of promoter activation. Moreover, the specificity and efficacy of the chimeric salmon/rat Pit-1 was similar to wild type rat Pit-1 in activating salmon and rat PRL/CAT.(ABSTRACT TRUNCATED AT 250 WORDS)

Functional analysis of estrogen-responsive elements in chinook salmon (Oncorhynchus tschawytscha) gonadotropin II beta subunit gene.

The salmon gonadotropin II gene regulates ovulation and spawning. Analysis of the 5' flanking sequence of the hormone-specific beta-subunit of salmon gonadotropin II (sGTHII beta) gene reveals the presence of several presumptive estrogen-responsive elements (ERE). The participation of ERE in the control of sGTHII beta gene transcription was examined by the transient expression of sGTHII beta gene promoter-chloramphenicol acetyltransferase chimeric DNA constructs in HeLa cells, with the cotransfection of a rainbow trout estrogen receptor expression vector. Three ERE have been identified: the proximal ERE [pERE, at -267 base pair (bp) from the transcription start site], the distal ERE (dERE, at -2698 bp, three GGTCA motifs each separated by exactly 31 bp), and the half-ERE (1/2ERE, at -157 bp as a GGTCA motif), respectively. The pERE (TGTCAATCTGACC) represents a novel but less effective variation of the consensus ERE (cERE). The dERE is a unique estrogen-induced enhancer. It requires the participation of the pERE to be functional and the enhancer activity of pERE and dERE is promoter specific. The contribution of 1/2ERE is minor and is not cell-type specific. The activation of the ERE in the sGTHII beta gene and the synergistic cooperation between the dERE and pERE by estradiol-17 beta is dose dependent. DNA sequences in the vicinity of the ERE decreases their hormone responsiveness and the synergism between dERE and pERE. These negative regions may contribute to the quiescent endocrine state of the sGTHII beta gene during the regressive phase of the reproductive cycle in teleost.

Induction of chinook salmon growth hormone promoter activity by the adenosine 3',5'-monophosphate (cAMP)-dependent pathway involves two cAMP-response elements with the CGTCA motif and the pituitary-specific transcription factor Pit-1.

In this study, the functional role of two cAMP-response elements (CRE) in the promoter of the chinook salmon GH gene and their interactions with the transcription factor Pit-1 in regulating GH gene expression were examined. A chimeric construct of the chloramphenicol acetyltransferase (CAT) reporter gene with the CRE-containing GH promoter (pGH.CAT) was transiently transfected into primary cultures of rainbow trout pituitary cells. The expression of CAT activity was stimulated by an adenylate cyclase activator forskolin as well as a membrane-permeant cAMP analog 8-bromo-cAMP. Furthermore, these stimulatory responses were inhibited by a protein kinase A inhibitor H89, suggesting that these CREs are functionally coupled to the adenylate cyclase-cAMP-protein kinase A cascade. This hypothesis is supported by parallel studies using GH4ZR7 cells, a rat pituitary cell line stably transfected with dopamine D2 receptors. In this cell line, D2 receptor activation is known to inhibit adenylate cyclase activity and cAMP synthesis. Stimulation with a nonselective dopamine agonist, apomorphine, or a D2-specific agonist, Ly171555, suppressed the expression of pGH.CAT in GH4ZR7 cells, and this inhibition was blocked by simultaneous treatment with forskolin. These results indicate that inhibition of the cAMP-dependent pathway reduces the basal promoter activity of the CRE-containing pGH.CAT. The functionality of these CREs was further confirmed by deletion analysis and site-specific mutagenesis. In trout pituitary cells, the cAMP inducibility of pGH.CAT was inhibited after deleting the CRE-containing sequence from the GH promoter. When the CRE-containing sequence was cloned into a CAT construct with a viral thymidine kinase promoter, a significant elevation of cAMP inducibility was observed. This stimulatory response, however, was abolished by mutating the core sequence, CGTCA, in these CREs, suggesting that these cis-acting elements confer cAMP inducibility to the salmon GH gene. The interactions between CREs and the transcription factor Pit-1 in mediating GH gene expression were also examined. In HeLa cells, a human cervical cancer cell line deficient in Pit-1, both basal and cAMP-induced expression of pGH.CAT were apparent only with the cotransfection of a Pit-1 expression vector. These results taken together indicate that the two CREs in the chinook salmon GH gene are functionally associated with the cAMP-dependent pathway and that their promoter activity is dependent on the presence of Pit-1

Rapid identification and isolation of zebrafish cDNA clones.

A fast and economical approach, referred to as cDNA clone tagging, was adapted to identify and isolate zebrafish cDNA clones. The basic approach was to partially sequence the coding region of size selected cDNA clones and the partial sequences were then used as tags for identifying the clones through homology search. To benefit maximally from the tagging approach, two cDNA libraries, derived from embryonic and adult fish poly(A)+ RNAs, respectively, were constructed by unidirectional cloning; conceptually, they have the potential to represent all expressed zebrafish genes. A total of 1084 clones were sequenced from the two libraries, and 511 clones were identified, based on sequence homology. These identified clones were derived from at least 261 genes, encoding 48 translational machinery proteins, 47 cytosolic proteins, 43 cytoskeletal proteins, 41 nuclear proteins, 32 membrane proteins, 22 secreted proteins, 20 mitochondrial proteins and 8 proteins with an unknown location. Of the 261 distinct cDNA clones identified, 254 were isolated for the first time in the zebrafish. These tagged cDNA clones, identified and unidentified, provide rich resources for developmental analysis as well as mapping of zebrafish genome. The long-term objective of this study is to establish a tagged zebrafish gene library that can be accessed both by hybridization screening against the plasmid DNAs and by electronic screening using the sequence information.

The skin-type antifreeze protein gene intron of the winter flounder is a ubiquitous enhancer lacking a functional C/EBPalpha binding motif.

The winter flounder antifreeze protein (AFP) intron contains a liver-specific enhancer (Element B) which was shown earlier to bind CCAAT/enhancer binding protein (C/EBP)alpha. In contrast, as demonstrated in the present studies, the intron of the skin-type AFP gene acted as a ubiquitous enhancer and contained a TA insertion at similar region to Element B (Element S) which destroyed its interaction with C/EBPalpha. Furthermore, a TA insertion of Element B by site-directed mutagenesis decreased its liver enhancer activity. The presence or absence of C/EBPalpha binding motifs in Element B and Element S, respectively, may provide a mechanism for their differential expression.

Studies of a putative ice-binding motif in winter flounder skin-type anti-freeze polypeptide.

Winter flounder contains two distinct anti-freeze protein isoforms, which are the liver-type extracellular anti-freeze proteins and the skin-type intracellular anti-freeze protein. The skin-type anti-freeze proteins exhibit lower anti-freeze activities than the liver-type isoforms and this might be due to their lacking complete ice-binding motifs. One of the skin-type anti-freeze proteins, skin-type anti-freeze protein-3, does contain putative overlapping ice-binding motifs with the sequences '-K-DT-' and '-DT-K-'. Synthetic anti-freezes containing 0-3 repeats of the '-DT-K-' motif were tested for stability and activity. Loss of the single '-DT-K-' of skin-type anti-freeze protein-3 increases the anti-freeze activity and increasing the number of motifs to two or three lowers the activity. The decrease in activity with an increasing frequency of the motif correlates with a decrease in the helical content of these peptides at 0 degrees C.

The structure and function of a gene encoding a basic peptide from prawn white spot syndrome virus.

Prawn white spot syndrome virus (WSSV) is the major pathogen responsible for the high mortality of cultured prawns. A gene (termed as p6.8) encoding a basic peptide was found by screening the cDNA and DNA libraries of WSSV. The peptide was highly homologous with proteins rich in arginine and lysine. A fusion protein containing the p6.8 and green fluorescent protein (GFP) genes was cloned into pBV220 and expressed in E. coli. Gel mobility shift assay indicated that the peptide encoded by p6.8 had the capability of binding DNA and might be involved in DNA packaging.

Presence of distinct cis-acting elements on gonadotropin gene promoters in diverse species dictates the selective recruitment of different transcription factors by steroidogenic factor-1.

The nuclear receptor steroidogenic factor-1 (SF-1) regulates the cell-specific expression of the pituitary gonadotropin subunit genes. Several potential DNA-binding sites for SF-1, estrogen receptor (ER) and the immediate-early transcription factor NGFI-A are found in LHbeta genes from many species. In this study, we have examined the action and interaction of these transcription factors on LHbeta gene promoters from two representative vertebrate species, i.e. rat and salmon. Cotransfection studies in COS-1 cells have shown that the action of SF-1 on salmon gonadotropin IIbeta (sGTHIIbeta) gene promoter was dramatically enhanced when combined with ER. The rat LHbeta promoter was activated by SF-I or ER individually, but these two factors, however, were unable to act in synergism on this promoter. In contrast, NGFI-A, specifically in cooperation with SF-1, transactivated the rat LHbeta gene expression but was ineffective on the sGTHIIbeta gene. Gel shift experiments showed that this lack of activation was due to the low affinity of the salmon NGFI-A-responsive element for its binding protein. In conclusion, our studies demonstrate that differential recruitment of distinct transcription factors by SF-1 might be a common mechanism to activate the cell-specific gonadotropin gene expression in different species.

Salmon gonadotropin IIbeta subunit promoter contains multiple DNA elements responsible for stimulation by gonadotropin-releasing hormone through protein kinase C-dependent and -independent pathways.

Gonadotropin-releasing hormone (GnRH) stimulates gonadotropin (GTH) production by activating GTH subunit gene transcription. In salmonid fish, the expression of the beta subunit gene of GTH II (sGTH IIbeta) is stimulated by GnRH at the final stages of reproduction. DNA elements required for the GnRH stimulation were examined by analyzing sGTH IIbeta promoter activity by transfection studies in a gonadotrope-derived cell line, alphaT3-1. A GnRH analog (GnRHa) specifically stimulated the sGTH IIbeta promoter (3358 bp) expression 3.6-fold, while phorbol myristate acid (PMA) stimulated it 6.2-9-fold. Analysis of a series of 5'-deletion mutants has revealed that a proximal region (-258 to -199) was important in GnRHa stimulation through protein kinase C (PKC)-independent signal transduction pathways, because an internal deletion mutant (delta(246 - 217)/3358) showed a significant decrease in the level of GnRHa stimulation, but showed no change in stimulation by PMA. A large upstream region (-3358 to -1260) showed an enhancing activity of the GnRHa stimulation, and a far upstream 530 bp segment in this region (-3358 to -2829) may be responsible for this activity. The present results suggest that sGTH IIbeta gene may be controlled by GnRH through multiple DNA elements including those responsive to PKC-dependent and -independent signal transduction pathways.

Cloning of zebrafish activin type IIB receptor (ActRIIB) cDNA and mRNA expression of ActRIIB in embryos and adult tissues.

A full-length cDNA encoding for activin type IIB receptor (ActRIIB) was cloned from zebrafish embryos. It encodes a protein with 509 amino acids consisting of a signal peptide, an extracellular ligand binding domain, a single transmembrane region, and an intracellular kinase domain with predicted serine/threonine specificity. The extracellular domain shows 74-91% sequence identity to human, bovine, mouse, rat, chicken, Xenopus and goldfish activin type IIB receptors, while the transmembrane region and the kinase domain show 67-78% and 82-88% identity to these known activin IIB receptors, respectively. In adult zebrafish, ActRIIB mRNA was detected by RT-PCR in the gonads, as well as in non-reproductive tissues, including the brain, heart and muscle. In situ hybridization on ovarian sections further localized ActRIIB mRNA to cytoplasm of oocytes at different stages of development. Using whole-mount in situ hybridization, ActRIIB mRNA was found to be expressed at all stages of embryogenesis examined, including the sphere, shield, tail bud, and 6-7 somite. These results provide the first evidence that ActRIIB mRNA is widely distributed in fish embryonic and adult tissues. Cloning of zebrafish ActRIIB demonstrates that this receptor is highly conserved during vertebrate evolution and provides a basis for further studies on the role of activin in reproduction and development in lower vertebrates.

Two rainbow trout (Oncorhynchus mykiss) albumin genes are differentially regulated.

Two distinct albumin cDNAs (rtALB1 and rtALB2) were isolated from the rainbow trout (Oncorhynchus mykiss) liver cDNA library. The rtALB1 cDNA (2761 bp) contains a 69 bp 5' untranslated region (UTR), a 1821 bp reading region, and a long 3' UTR of 872 bp. The rtALB2 cDNA (2250 bp) contains a 78 bp 5' UTR, a 1824 bp coding region, and a 348 bp 3' UTR. The two albumins are 81.5% and 77.5% identical in their nucleotides and protein sequences, respectively. Both rtALB1 and rtALB2 genes are expressed only in the liver. The albumin mRNA was first detected in 5-week-old embryos and was tissue-specific. The two albumin genes were differentially expressed, with the rtALB1 transcripts being 3 to 10 times more abundant than the rtALB2 transcripts. This differential expression was partially regulated at the transcriptional level. Promoter analysis showed that the rtALB1 gene had a typical albumin promoter structure. However, the rtALB2 promoter was abnormal in the TATA box region and was less effective in activating the reporter gene in the mammalian cell lines. These variations in rainbow trout albumin promoter sequences might account for their differences in transcriptional efficiency.

Genomic structure of growth hormone genes in chinook salmon (Oncorhynchus tshawytscha): presence of two functional genes, GH-I and GH-II, and a male-specific pseudogene, GH-psi.

Two chinook salmon (Oncorhynchus tshawytscha) growth hormone genes (a functional GH-I gene and a pseudogene, GH-psi) were isolated and characterized. The GH-I gene sequence consists of 1.9 kb of 5'-flanking sequence, 4.1 kb of transcribed region, and 64 bp of 3'-flanking sequence, and contains 6 exons and 5 introns. The pseudogene, GH-psi, spanning 4.1 kb, has a similar structure as the GH-I gene. However, it has one wrong splicing sequence at the intron 1/exon 2 junction, one premature termination codon in exon 5, and a deletion in the last half of exon 5 and the first part of intron 5. In addition to GH-I gene and GH-psi, a third GH gene, GH-II, was identified by the polymerase chain reaction (PCR) and subsequently shown to be the second functional GH-II gene. To study the linkage arrangement of these three GH genes, 50 unrelated chinook salmon (25 males and 25 females) and one chinook salmon family were analyzed by PCR. The results showed that GH-psi exists only in males and that it segregates from father to sons. These results suggest that GH-psi is sex specific and probably resides on the Y chromosome. Together these results indicate that there are three GH genes in the genome of male chinook salmon, and only two GH genes in the females. The extra GH gene in the male is, however, a pseudogene.