Structure of a Fish (Rainbow Trout) Growth Hormone Gene and Its Evolutionary Implications

L. B. Agellon, S. L. Davies, T. T. Chen, and D. A. Powers

PNAS 1988;85;5136-5140
doi:10.1073/pnas.85.14.5136
This information is current as of June 2007.

This article has been cited by other articles:

www.pnas.org#otherarticles
E-mail Alerts Receive free email alerts when new articles cite this article - sign up in the box at the top
right corner of the article or click here.
Rights & Permissions To reproduce this article in part (figures, tables) or in entirety, see:
www.pnas.org/misc/rightperm.shtml
Reprints To order reprints, see:
www.pnas.org/misc/reprints.shtml

Notes:
Proc. Nati. Acad. Sci. USA
Vol. 85, pp. 5136-5140, July 1988
Evolution

Structure of a fish (rainbow trout) growth hormone gene and its

evolutionary implications
(internal repeats/nucleotide sequence/evolution)
L. B. AGELLON*t, S. L. DAVIES*t, T. T. CHENt§, AND D. A. POWERS*t§
*Department of Biology and tThe Chesapeake Bay Institute, The Johns Hopkins University, Baltimore, MD 21218; and §Center for Marine Biotechnology,
University of Maryland, Baltimore, MD 21202
Communicated by C. Ladd Prosser, March 11, 1988

ABSTRACT We have isolated and sequenced a clone from A B C D E F

a rainbow trout (Salmo gairdneri) genomic library that carries
a gene encoding a fish growth hormone (GH). This gene spans
a region of =4 kilobases, nearly twice that of mammalian GH
genes. The trout GH gene is comprised of six exons, in contrast
with five exons in mammals. The additional intron in the fish
gene interrupts translated regions that are analogous to the last
exon of its mammalian counterpart. In addition, the alleged
internally repeating sequence in mammalian GH, prolactin
(Prl), or placental lactogen (PL) is not observed in the predicted
polypeptide sequence of fish GH. Direct repeats that flank
exons I, HI, and V of the mammalian GH, Prl, and PL genes
are absent in the fish GH gene. These findings indicate that the
rainbow trout GH gene structure does not support the current
hypothesis that internally repeated regions in GH, Prl, and PL
arose from a small primordial gene.
Growth hormone (GH), prolactin (Prl), and placental lactogen
(PL) form a family of polypeptides with related structure and
function (1-3). GH and Prl are synthesized mainly by soma-
totrophs and lactotrophs, respectively, in the anterior pituitary
gland, whereas PL is synthesized by syncytiotrophoblasts in
the mammalian placenta. In vertebrates, GH is responsible for
stimulating growth, whereas Prl primarily functions as a
reproductive hormone (4). The biological activity of mamma-
lian PL is the least understood of these three hormones. One
possible function is to augment the carbohydrate-regulating
properties of GH during pregnancy (5).
The notion that these three hormones share a common
ancestry was initially suggested because of their related
biological and immunological properties (6-8). Based on the
structure of these hormones and the genes that encode them,
it has been proposed that they evolved from a common
ancestor, which, in turn, arose by repeated duplication of a
smaller gene or coding region (3, 9, 10). The majority of the
data used to generate this hypothesis has been derived from
mammalian species (6-8). To test the applicability of this
evolutionary model to vertebrates in general, we have cloned
and sequenced a gene encoding GH in the teleost fish, rain-
bow trout (Salmo gairdneri).¶
MATERIALS AND METHODS
Molecular Cloning and Isolation of Genomic Clones. DNA
used in the construction of the trout genomic library was
purified from testes dissected from a single trout. The library
was prepared by partial digestion of testes DNA with EcoRI,
cloning into Charon 4, and amplification in Escherichia coli
strain LE 392. The rainbow trout GH cDNA [the 900-base-
pair (bp) Pst I/EcoRI fragment of pAF51] (11) was labeled by
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
5136
23.1*
9.45
AM
6.50 ^,
4.3.
23
2.00
FIG. 1. Blot analysis of RTGH14B2 and trout testes DNA.
Rainbow trout testes DNA (10 Atg) and RTGH14B2 DNA (5 ng) were
treated with various restriction endonucleases, resolved on a 0.8%
agarose gel, and then transferred onto a nitrocellulose membrane.
The probe (EcoRI/Pst I fragment of pAF51) (11) was labeled to a
specific activity of 1 x 10i dpm/,ug by random priming (12) and
hybridized to the blotted DNA overnight. After washing, hybridizing
fragments were visualized by autoradiography. Lanes A-C, trout
testes DNA; D-F, RTGH14B2 DNA; A and D, EcoRI; B and E,
EcoRI and HindIII; C and F, EcoRI and Pst I. Mobility of size
markers (AHindIII fragments) is shown (in kbp) on the left.
the random priming method (12) and used as a probe for
plaque hybridization screening of the library (13). Hybrid-
ization was under stringent conditions: 50% formamide/1%
NaDodSO4/1 M NaCI/10% dextran sulfate, 42TC. DNA from
positive clones was purified and further characterized by
restriction endonuclease mapping and DNA blot analysis (14).
One clone, RTGH14B2, which carries an insert of 14.8 kbp,
was selected for sequence analysis.
Sequence Analysis. The 6.8-kbp Pst I fragment of
RTGH14B2 that hybridized with the GH cDNA probe was
subcloned into the Pst I site of plasmid pUC8 (15). Ordered
Abbreviations: GH, growth hormone; Prl, prolactin; PL, placental
lactogen.
tPresent address: Columbia University College of Physicians and
Surgeons, 630 West 168th Street, New York, NY 10032.
IThe sequence reported in this paper is being deposited in the
EMBL/GenBank data base (IntelliGenetics, Mountain View, CA,
and Eur. Mol. Biol. Lab., Heidelberg) (accession no. J03797).
 Evolution: Agellon et al. Proc. Natl. Acad. Sci. USA 85 (1988) 5137
gt caagttacag ggttgtgtct gtctgtgtga -301
-300 ctgagtgtaa ctttgttcat tcattatgtc ctagacaaca gaggtttgtg tcgtctgtgt tttgaccctc atttgtcaag tcatcgagta cgttttttgt -201
-200 ttttaggagt cacctcttcc cgaactcatg gaaaaattca tgattgattt gacgcattat actgattgtt ccatagtcac atacaaaaac aggtcccatc -101
-100 ggcgagaggt ggtacatgga gaaaatctca tgtttcctcc tgttgataca ttaaaacatg tgttctccat ctataaaaac agtggcccca aacaagcggc -1
1 AACATACTGA ACCGACCACC ACACTTTCAA GTGAAGTAAT CATCCTTGGC MTTAAGAGA AAAAMTGGG ACAAGgtaaa ccagctttta ttttattttt 100
101 ttaagtggga agtcagtgta ccatttaata ccatttaact ttaacattta atcactgagg caggggccaa gaaggcagag aaagagtgaa caagtaatgt 200
201 actgccatga gggtataatc tacttacaca gaaccacttc ctttaacaac ctaaccatgt gatctattag atttacattt gagttattta gcagagactc 300
301 ttatccagag cgacttacag gagcaattag ggttaagtgc cttgctcaag ggcacatcaa cagatttctc acctagtcag ctcagggatt caaaccapta 400
401 acctttcaet tactReccca actctcttaa tcgctaggct aatgagaaag atagcaaatt gagaatatct tactattgag aatatcttac taacatgtcg 500
501 caacatcatt tgacttactc gtttttatac atttcttatt ttctgtcatc tctcttttag TGTTTCTGCT GATGCCAGTC TTACTGGTCA GTTGTTTCCT 600
601 GGGTCAAGGG GCGGCGATGG AAAACCAACG GCTCTTCMC ATCGCGGTCA ACCGGGTGCA ACACCTCCAC CTATTGGCTC AGAAAATGTT CMCGACTTT 700
701 gtaagacagc ttttgaatct tcttttgaca cagcagataa tgtttcagag gtgtttcctc ttctttgtag acaagtgtcc tcttcacgca aaccgagcgg 800
801 caaaacattc tctctcccgt ctttgtgatt ttgtgcagGA AGGCACCCTG TTGCCTGATG AACGCAGACA GCTGAACAAG ATATTCCTGC TGGACTTCTG 900
901 TAACTCTGAC TCCATCGTGA GCCCAATCGA CMGCAGGAG ACTCAGAAGA GTTCAgtaag ttacctggct gagacaatcc tccatgatgc acgattccaa 1000
1001 catgaataat agggcatctc aatttgaaca atcgatacaa cttagtcatt agttattggg caagcagatc cccgattgtc taaactccat gggtaatata 1100
1101 ctgtagataa gcagaaccag catcatgtat ggtggaaatt aaatctagcc atgacaggga gttttaaatt gtacacttaa aatcggcagt aaaataatgc 1200
1201 tatacctcag tgccttcaaa aacaaccaca tatcatagtc cttgtaagta aaacccatca ctctctaaat cggcggtttc tctacgtcta cattctccag 1300
1301 ccatgtatca tgtaaatgat atggcatctc aagctgtaca attacatctc aacttcattt tctaataatc tgtggtttct ctacatctac acacacagGT 1400
1401 CCTGAAGCTG CTCCATATCT CTTTCCGCCT GATTGAATCC TGGGAGTACC CTAGCCAGAC CCTGACCATC TCCAACAGCC TAATGGTCAG AMCTCCAAC 1500
1501 CAGATCTCTG AGAAGCTCAG CGACCTCAAA GTGGGCATCA ACCTGCTCAT CMAGgtaatg gtcaatgacc atttgtggtg acgcactttg tcgattttta 1600
1601 actcaaatac ttctagtaag ttgaagtcag tcaatgaaaa gtcattatta cttcaaatgt ctatgtggta ctggctcaaa tctaaatgag tcacatcaat 1700
1701 gcaatttttt aaagttataa gaaatgaact ttttacccag catgctctac tacaggtaga ttttttggaa ttgtttttaa tatctgtgtt tttgcatgta 1800
1801 cagtacattg agtgattgat tgattgattc atcttatgct acacacagat atatgacgta catttttcta cattttcaca aaggtaaata acatacaagg 1900
1901 taccgaaatt ttgcaaacct acttgcaggc ctgatgtggc ctgtaaacca tgagtttcag gccactgtat ttgggaaaag ctacacctca aaataaggcc 2000
2001 ttataagata tgtaatatat tgttataaag agtttaacta taatgataat atttgcctag aaaatcactt gaaggccaca ggactgaaaa ttaatgacaa 2100
2101 caaacatgat aattctacaa ttcactcaaa aggcaaggca cacttggaaa gtatattgga gacatgtctt agtgggggca ttactaaaaa atgtcaagct 2200
2201 gataccactc aaatctcaac cctctacagg gcgactctat aggtttgagt aatgactata aaatcacttt aagcgactgt agtcagattc tgtatattaa 2300
2301 gtgcaacggt ttcctcaaaa gttttgagta atgacagcac attsgggttt acaatattgt tgttatcttc cactgacatg aaagtgaaat acaactatac 2400
2401 tttcctagtt agaaaacata at taggacc acgtatgtct cttctcagca gatctttcag cgctttccat tgtgacgggg taactcacct catcgatcat 2500
2501 cactaatagt gactatatca gtaacacccc attcaatgac'tgaatattgg cccattcaag gacatttatg catgtgtctt ttgccatatg tgcttgtaga 2600
2601 atggccaaat aaacaagtat tgttatgcac acctccaccc caccatgcat ctctctctgt ctcccacagG GGAGCCAGGA TGGCGTACTG AGCCTGGATG 2700
2701 ACAATGACTC TCAGCATCTG CCCCCCTACG GGAACTACTA CCAGAACCTG GGGGGCGACG GCAACGTCAG GAGGMCTAC GAGCTGTTGG CCTGCTTCAA 2800
2801 GAAGGACATG CATAAGgtgg aagaccatgt tgccttcaat tgcatgtgcc ttcctatatt tcctacagtg cgattgtttt tttgtgatct ctattgtgaa 2900
2901 gtatctttat aggtctttaa cccatttgtt attactatta ttgttcattg atcaagactg gtctcgagaa agtcctggtg acttagaaca ttcacattaa 3000
3001 aatgtgtcaa ctataaccta ttcttcttgt cccccccccc cccccccccc ccacccccaa gGTTGAGACC TACCTGACCG TCGCCAAGTG CAGGAAGTAT 3100
3101 CTGGAGGCCA ACTGCACTCT GTAAACGTGG GCCGGAGCGG AGCAGCAAGA GCCTGTCTCC AGGGTTCGGT TTCCCAGATA CAGATGAGAC CITTCCCTGC 3200
3201 ACTGAAGAGC ATGTTCAATT GAGATTCTCC ATTAGGCATG T1TmTlTAG TCTAGAGTAG ATTTCATTTG GATCTGGTAG AGCCTGGCTC CAGGGGTr 3300
3301 CAAGCATTTT GCATTTTGTT CTCTGAAATC AACTTTCTAT GATTTTCACT CCATTACTCG GAGCTACCAC TGATCCATGG ACATTITAGA TTAGTACATT 3400
3401 TATTGAAAAG GTTTAAAT ATGTCTTATT TAGATATATG ATTCAAGGTG GTGGTGCCAT TTATGCATAA ATTAATArTT AGAGGTGAAA TGGGAACATG 3500
3501 TAGAGCTCCA ATCTTTAGGT ACGTCCACAG ATGGATAATA TAITTTTAGAG TCATTTCCTT GAAGTATTTT CATTCCTTTA TCTTACTGTT TGAAACGAAT 3600
3601 AGTGATTCGT TTTCAAA CIrcTcTGCg gtacatgatc tcttggctac tatttgctat ctttcaaatc aacatttttg ttacaagttt ctagccccaa 3700
3701 cattcctatt gtgtcccttg gacaacttaa ggctggattc aattcgtatc gcagacgctc cattgaaatg taaaggcaat gttccctgcg ttcgtggaga 3800
3801 cggcattcac cgcaaacgct gcatatgtca ggctcgttc

FIG. 2. The nucleotide sequence of the trout GH gene was determined from a series of clones carrying deletions originating from either end
of the Pst I fragment containing the trout GH gene. The gene spans a region of -4000 bp in the trout genome. The "TATA box" and the poly(A)
signal are shown overlined and in bold print. The putative cap site (+ 1), located 22 nucleotides downstream from the TATA box, was determined
by aligning the GH genomic and cDNA sequences. Exons are shown in uppercase; introns and flanking regions are in lowercase. Sequences
capable of forming hairpin structures and a tandemly repeated sequence are underlined.
deletions (150-200 bases) originating from either end of the M13mpl9 (17). The nucleotide sequence of the GH gene was
6.8-kbp Pst I fragment were created by using BAL-31 (16) and determined from these deletion derivatives by using the
resulting fragments were recloned into either M13mpl8 or Sanger sequencing procedure (18).
5138 Evolution: Agellon et al.
RESULTS
Isolation and Characterization of the Rainbow Trout GH
Gene. Thirty clones (out of 500,000 clones screened) were
isolated from a rainbow trout genomic library by virtue of
their hybridization to the 900-bp EcoRI/Pst I fragment of the
rainbow trout GH cDNA clone (pAF51) (11). One clone
(RTGH14B2), which contains a 14.8-kbp EcoRI fragment,
was selected for detailed analysis. Fig. 1 shows the hybrid-
ization pattern of trout testes DNA and RTGH14B2 DNA.
Trout testes DNA digested with EcoRI produced several
fragments, ranging in size from 6 to 15 kbp, which hybridize
with pAF51. The largest of these fragments comigrates with
the EcoRI fragment carried by clone RTGH14B2. The hy-
bridization pattern of RTGH14B2 DNA treated with EcoRI
and HindIII or with EcoRI and Pst I is a subset of the pattern
observed in the corresponding genomic lanes (Fig. 1). The
additional hybridizing fragments in the trout testes DNA (Fig.
1, lanes A-) represent the other copy of the GH gene
(unpublished data) and not the PrI gene. Salmonid GH and Prl
share no detectable immunological relatedness (19).
The smallest single fragment of clone RTGH14B2 shown in
Fig. 1 that carries the GH transcription unit is a 6.8-kbp Pst
I fragment. After subcloning this fragment, deletion deriva-
tives originating from either end of the fragment were
constructed and sequenced (Fig. 2).
The organization of the trout GH gene was determined by
aligning the trout GH cDNA sequence (11) with the corre-
sponding regions of the GH genomic gene. The 5' region of
the GH gene contains the sequence 5' TATAAAAA 3' (20) 22
nucleotides from the putative transcription start site. Anal-
ysis of the 5' sequence, up to 332 nucleotides upstream from
the cap site, did not reveal the existence of any sequence
similar to other consensus promoter elements. The DNA
binding sites of the glucocorticoid- and thyroid hormone-
receptor complexes have been localized in the mammalian
GH genes (21, 22). Sequences with structural similarity to
these sites were not obvious in the trout GH gene. However,
it is currently not clear whether thyroid and glucocorticoid
hormones control fish GH gene expression in a manner
similar to that in mammals (23-25). The poly(A) signal 5'
AATAAA 3' (26) is located at the 3' region of the gene
(position + 3615). Thus, the rainbow trout GH gene spans a
region of =4000 bases, nearly twice the size of its mammalian
counterparts (Fig. 3).
Six exons (exon 1, 75 bp; exon II, 140 bp; exon III, 117 bp;
exon IV, 156 bp; exon V, 147 bp; exon VI, 568 bp) and five
p p
I ME I m I rtGH
E E
1 lhE..I hGH
E E
I bGH
E H
I rGH
FIG. 3. Organization of fish and mammalian GH genes. Restric-
tion fragments containing GH genes are shown. Boxes represent
exons; lines between boxes are introns. The trout GH gene contains
an additional intron (intron 5) that interrupts a region equivalent to
exon 5 of mammalian GH genes. The region of the trout gene that
encodes the untranslated portion ofGH mRNA is approximately four
times larger than its mammalian counterpart. rtGH, trout; hGH (27),
human; bGH (28), cow; rGH (9, 29), rat; E, EcoRI; H, HindlI; P,
Pst I. Scale bar represents 500 bp.
Proc. Natl. Acad. Sci. USA 85 (1988)
introns (intron 1, 485 bp; intron 2, 138 bp; intron 3, 443 bp;
intron 4, 1115 bp; intron 5, 245 bp) make up the trout GH
gene. The intron boundaries are located at positions 76/560,
701/838, 956/1398, 1555/2669, and 2817/3061. All donor and
acceptor sites resemble consensus sequences (30). The ac-
ceptor site of intron 5 is interesting in that it is preceded by
an exceptionally cytidine-rich region.
The sequence was searched for perfect and imperfect,
direct and inverted repeats >6 nucleotides. A 17-bp tandem
repeat, beginning at position +458, and a region of dyad
symmetry (stem of 8 bp, loop of 6 nucleotides) beginning at
position + 395 were found in intron 1. Two other regions of
dyad symmetry were found in intron 4, the largest of the six
introns. The first region, capable of forming a stem and loop
with 7 bp and 7 nucleotides, respectively, begins at position
+ 2394. The second region, capable of forming a stem of 11
bp and a loop of 8 nucleotides, begins at position + 2420. The
significance of these structures is unknown.
Alignment of trout and human GH gene sequences, so that
the polypeptides share maximum sequence similarity, indi-
cates that the first four exons ofboth genes encode analogous
portions of the GH polypeptide. Trout exons V and VI, which
are interrupted by the additional intron (i.e., intron 5), encode
a region ofthe GH polypeptide that is equivalent to the region
encoded by the last exon of the human GH gene. The
introduction of intron 5 into the trout gene likely took place
after the divergence of fish and tetrapods because the
mammalian GH, Prl, and PL genes all contain five exons (9,
27-29, 31). Testing of this hypothesis, however, must await
the characterization of the trout Prl gene.
DISCUSSION
The current model for GH gene evolution is schematically
represented in Fig. 4. Niall et al. (3) were first to suggest that
a small primordial gene was repeatedly duplicated to form a
larger gene that became the "precursor gene." We have
Phase / -IZI- primordial gene
-in-it Ii-
-t-I + [}C}
V- intron
+ -{I}
intron
I III IV V
~ Lii Z~uZ
rc
precursor gene
Phase //
duplication, divergence
I GH Prl PL
FIG. 4. Current model for evolution of the GH gene family,
adapted from Barta et al. (9), Miller and Eberhardt (7), Selby et al.
(10), and Slater et al. (8). During phase I, a primordial gene (or coding
region) was repeatedly duplicated. Subsequently, an intervening
sequence separating two of the duplicated domains was lost. Later,
two other domains were incorporated to form the precursor gene
containing five exons. In phase II, the precursor gene was dupli-
cated. Divergence gave rise to the genes that presently encode GH
and Prl and eventually PL. Hatched boxes represent exons that are
thought to have separate origins. Roman numerals represent exons
in the present gene structure.
 Evolution: Agellon et A Proc. Natl. Acad. Sci. USA 85 (1988) 5139

Table 1. Alignment of the internally repeating regions in human GH and corresponding regions in trout GH

Exon Encoded polypeptide sequence Identity Similarity

Human GH

E1/E2 15 L R - - : : A Hf1L H LA®I7: (i E 32 30.0 51.3

E2 93 LXJ-- -:: V LV)
(N
E5A 128 LIEDG : :WP F K(
S D 147
E5B 163 Y - - ::C F M D K V : L RRII 181

Trout GH

El 12 N R -- --V©H L H L)®QM F -i 29 10.9 39.1

E2 89 - _ _ I - - Q I 103
E5A 121 S D G V L S L D H L ft3Y i) Y 143
E5B/E6 159 i)KM3-- --CFKK(bM H KrnE( Y -L V® 176

Human GH compared to Trout GH

E1/E2 15 Lj1J- - - - A H L A F D T -
E 32 30.4 47.8
El 12 N RI- V Q
-
L H LL
- -
Q K M (i- E29

E2 93[LR - - - - S V F Nf L ®G[-@
)- S®G 110 21.7 52.2
E2 89 NJ-T-- ---TI~ S L )R IJ-9-()I 103

E5A 128 L --[P R® I F®Q Y - S L®D 147 13.0 43.5

E5A 121 S4DGVLVWL DC S Q(P9L(9P Y G N9Y 143

E5B 163 LY - - - - TFV-[ R®V 181 43.5 60.9

E5B/E6 159 W A - - - - CLFIKD M H[V - L T (y)A 176

Alignment of the internally repeated regions in human GH, equivalent regions in trout GH, and comparison between the human and trout
GH segments are shown. The internally repeated regions in human GH were taken from Niall et al. (3). Gaps (-) were introduced in the sequences
to obtain the best alignment. Each gap was counted as one position in the calculation of % similarity and % identity. Spaces (:) in the human
GH sequence were added for clarity of alignments and were not included in the calculations. Identical amino acid residues are shown in boxes.
Amino acid residues scored as similar (circled) are based on Dayhoffs highly acceptable substitutions (32).
designated these events as phase I. Subsequent duplication marks the end of phase I. At the beginning of phase II, the
and divergence of this precursor gene, which we designate as entire precursor structure was duplicated and subsequent
phase II, led to the evolution of GH, Prl, and PL. divergence gave rise to the genes that encode GH and Prl (see
The GH evolutionary model has been reviewed in detail bottom of Fig. 4).
elsewhere (6-8). We shall briefly summarize the key features The existence of an additional intron (intron 5) in the trout
of this model. Much of the rationale for phase I (Fig. 4) of the GH gene is a notable difference between fish and mammalian
GH evolutionary model is based on the existence of apparent GH genes. At first inspection, the presence of intron 5 in the
internally homologous regions within GH, Prl, and PL (3). In trout GH gene might appear to support phase I of the GH
human GH, for example, the first two of the four regions of evolutionary model. Closer examination of the sequence
internal homology are each encoded by separate exons data, however, reveals that the location of intron 5 interrupts
(exons II and IV) but the last two regions are encoded by one the last presumptive repeated segment in human GH (ESB,
larger exon (exon V). The latter exon may have been between Lys-172 and Val-173; Table 1). Intron 5 is likely a
generated after the deletion of the intervening sequence that recent introduction to the trout GH gene. Two observations
separated the two reduplicated coding domains either prior to support this hypothesis. (i) The event that gave rise to GH
or after its incorporation into the gene. The two other exons and Prl genes (phase II) occurred around 350-400 million
(exons I and III) of the human gene are thought to have years ago, before the divergence of fish and tetrapods. The
separate origins. Exon I, which eventually took over the existence of GH and Prl in these two groups of animals attests
controlling functions of the gene, was incorporated at the 5' to this notion. (it) Mammalian genes for GH and Prl are
upstream region, and exon III, which conferred the hormone organized in a similar manner-i.e., both genes have five
with carbohydrate-regulating properties, was incorporated exons and four introns. Thus, intron 5 of the trout GH gene
between the originally duplicated regions. Imperfect direct could not have existed prior to the divergence of fish and
repeats have also been found flanking exons I, III, and V of tetrapods.
rat GH, rat Prl, human GH, and human PL genes. This Comparison of regions in the trout GH polypeptide se-
observation was taken as further evidence that these regions quence, equivalent to those described by Niall et al. (3), do
arose by separate insertion events by a mechanism analogous not show the existence of internally homologous regions
to DNA transposition. The formation of the precursor gene (Table 1). Moreover, direct repeats similar to those flanking
5140 Evolution: Agellon et al.
exons I, III, and V of the rat GH, rat Prl, human GH, and
human PL genes were not found in the trout GH gene.
Therefore, although the similarity between the polypeptide
sequences of fish GH and fish PrI (33, 34) is consistent with
phase II of the model (Fig. 4), the trout GH gene structure is
inconsistent with phase I.
In human GH, the apparent internal repeats share 30%o
identical amino acid residues and 51.3% similarity when
conservative substitutions are taken into account (Table 1).
By comparison, the equivalent regions in the trout GH have
only 10.9%o identical residues and 39.1% similarity. Further-
more, the groups of similar amino acid residues among the
selected regions in trout GH are not in the same category of
acceptable substitutions (32) as those found in the apparent
internally repetitive regions of human GH. Significant se-
quence similarity, however, is observed between equivalent
regions of trout and human GH (Table 1). These comparisons
suggest that the presumed internal repeats in the human GH
are not present in trout GH.
Nicoll et al. (35) have shown that similarity between
randomly selected regions in human GH, human PL, or ovine
Prl can be demonstrated by the introduction of appropriate
gaps in the sequence. In their example, the selected regions
have at least the same extent of similarity as those described
by Niall et al. (3). Our analysis of the trout GH structure and
that of Nicoll et al. (35) do not support the hypothesis that
internally repeated regions in GH, Prl, or PL arose from a
small primordial gene (phase I, Fig. 4).
It would be premature to generate an alternate model to
account for the trout GH gene structure until the GH gene
structure from at least one organism, common to both
branches leading to teleostean and mammalian evolution, is
determined.
We are grateful to Dinsdale Gooden for providing excellent
technical assistance. This project was supported by a grant (DCB-
851-8069) from the National Science Foundation to D.A.P. and
T.T.C. This article is contribution 1410 from the Department of
Biology, The Johns Hopkins University, and contribution 109 from
the Center for Marine Biotechnology, University of Maryland.
1. Catt, K. J., Moffat, B. & Niall, H. D. (1967) Science 157, 321.
2. Sherwood, L. M. (1967) Proc. Nati. Acad. Sci. USA 58, 2307-
2314.
3. Niall, H. D., Hogan, M. L., Sauer, R., Rosenblum, I. Y. &
Greenwood, F. C. (1971) Proc. Natl. Acad. Sci. USA 68, 866-
869.
4. Bentley, P. J. (1982) Comparative Vertebrate Endocrinology
(Cambridge Univ. Press, Cambridge, England), 2nd Ed.
5. Simpson, E. R. & MacDonald, P. C. (1981) Annu. Rev. Phys-
iol. 43, 163-188.
6. Moore, D. D., Walker, M. D., Diamond, D. J., Conkling,
M. A. & Goodman, H. M. (1982) Rec. Prog. Horm. Res. 38,
197-225.
7. Miller, W. L. & Eberhardt, N. L. (1983) Endocr. Rev. 4, 97-
Proc. Natl. Acad. Sci. USA 85 (1988)
130.
8. Slater, E. P., Baxter, J. D. & Eberhardt, N. L. (1986) Am.
Zoo!. 26, 939-949.
9. Barta, A., Richards, R. I., Baxter, J. D. & Shine, J. (1981)
Proc. Natl. Acad. Sci. USA 78, 4867-4871.
10. Selby, M. J., Barta, A., Baxter, J. D., Bell, G. I. & Eberhardt,
N. L. (1984) J. Biol. Chem. 259, 13131-13138.
11. Agellon, L. B. & Chen, T. T. (1986) DNA 5, 463-471.
12. Feinberg, A. P. & Vogelstein, B. (1983) Anal. Biochem. 132,6-
13.
13. Benton, W. D. & Davis, R. W. (1977) Science 196, 180-182.
14. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular
Cloning:A Laboratory Manual (Cold Spring Harbor Lab., Cold
Spring Harbor, NY).
15. Vieira, J. & Messing, J. (1982) Gene 19, 259-268.
16. Poncz, M., Solowiejczyk, D., Ballantine, M., Schwartz, E. &
Surrey, S. (1982) Proc. Natl. Acad. Sci. USA 79, 4298-4302.
17. Norrander, J., Kempe, T. & Messing, J. (1983) Gene 26, 101-
106.
18. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.
Acad. Sci. USA 74, 5463-5467.
19. Agellon, L. B., Chen, T. T., van Beneden, R. J., Sonstegard,
R. A., Wagner, G. F. & McKeown, B. A. (1986) Can. J. Fish.
Aquat. Sci. 43, 1327-1331.
20. Breathnach, R. & Chambon, P. (1981) Annu. Rev. Biochem. 50,
349-383.
21. Moore, D. D., Marks, A. R., Buckley, D. I., Kapler, G.,
Payvar, F. & Goodman, H. M. (1985) Proc. Natl. Acad. Sci.
USA 82, 699-702.
22. Barlow, J. W., Voz, M. L. J., Eliard, P. H., Mathy-Hartert,
M., De Nayer, P., Economidis, I. V., Belayew, A., Martial,
J. A. & Rousseau, G. G. (1986) Proc. Natl. Acad. Sci. USA 83,
9021-9025.
23. Martial, J. A., Baxter, J. D., Goodman, H. M. & Seeburg,
P. H. (1977) Proc. Nat!. Acad. Sci. USA 74, 1816-1820.
24. Samuels, H. H., Stanley, F. & Shapiro, L. E. (1979) Biochem-
istry 18, 715-721.
25. Dobner, P. R., Kawasaki, E. S., Yu, L.-Y. & Bancroft, F. C.
(1981) Proc. Nat!. Acad. Sci. USA 78, 2230-2234.
26. Proudfoot, N. J. & Brownlee, G. G. (1976) Nature (London)
263, 211-214.
27. DeNoto, F. M., Moore, D. D. & Goodman, H. M. (1981)
Nucleic Acids Res. 9, 3719-3730.
28. Woychik, R. P., Camper, S. A., Lyons, R. H., Goodwin,
E. C. & Rottman, F. M. (1982) Nucleic Acids Res. 10, 7197-
7210.
29. Page, G. S., Smith, S. & Goodman, H. M. (1981) Nucleic Acids
Res. 9, 2087-2104.
30. Mount, S. M. (1982) Nucleic Acids Res. 10, 459-472.
31. Cooke, N. E. & Baxter, J. D. (1982) Nature (London) 297, 603-
606.
32. Dayhoff, M. O., ed. (1972) Atlas of Protein Sequence and
Structure (Natl. Biomed. Res. Found., Silver Spring, MD).
33. Yasuda, A., Itoh, H., & Kawauchi, H. (1986) Arch. Biochem.
Biophys. 244, 528-541.
34. Kawauchi, H., Moriyama, S., Yasuda, A., Yamaguchi, K.,
Shirahata, K., Kubota, J. & Hirano, T. (1986) Arch. Biochem.
Biophys. 244, 542-552.
35. Nicoll, C. S., Mayer, G. L. & Russell, S. M. (1986) Endocr.
Rev. 7, 169-203.