Production of Biologically Active Recombinant Tilapia Insulin-Like Growth Factor-II Polypeptides in Escherichia coli Cells and Characterization of the Genomic Structure of the Coding Region

(1)

Volume 16,Number7,1997 MaryAnnLiebert,Inc.

Pp.

883-892

Production of

_Biologically

Active Recombinant

_Tilapia

Insulin-Like Growth

Factor-II

_Polypeptides

in

Escherichia coli

Cells

and

Characterization of the Genomic Structure

of the

_{Coding Region}

JYH-YIH

_CHEN,1

CHI-YAO

_CHANG,2

JIAN-CHYI

_CHEN,2

SHIH-CHIEH

_SHEN,1

and JEN-LEIH

WU2-3

ABSTRACT

Insulin-like

_growth

factor-II

(IGF-II)

is

a

fetal

growth

factor

in

humans,

but has not been

clearly

identified

in

fish

uptonow. Fora

detailed

understanding

of

the

physiological

response of

fish

IGF-II,

the first

step

was

to clone

_tilapia

IGF-II cDNA from

the

brain

cDNA

_{library, coding}

the

_region

of

_genomic

DNA,

and also

ex-pressing tilapia

IGF-II

_polypeptides

from Escherichia coli.

_Tilapia

cDNA sequences total

1,977

bp,

and pre-dicted

nucleotide

sequences and

amino acid sequences of

_tilapia

share

77.9%

and

90.7%

homology

_identity

with

rainbow

trout

_IGF-II,

_{respectively.}

The

_genomic

structure

of

the

_{tilapia prepro-IGF-II coding region}

is

very

difficult

tosequence in mammals and

birds.

The cloned

_tilapia

IGF-II

gene

_{coding region}

appears much

more

complex

than

in other vertebrates. In

_tilapia

IGF-II,

the

first

_coding

exonI

encoding

part

of

the

signal

peptide

sequence is 25

amino

acids shorter than

the first

_coding

exon

of

mammals and

birds.

The other

23

amino acids

of

the

_{signal peptide,}

and the first

amino acids of

the B

domain

and

C domain

are encoded

by

tilapia coding

exon2.

The

C, A,

and

D

domains,

and

the

first 20 amino acids of

the F

peptide

areencoded

by

tilapia coding

exon

3.

The other E

peptides

and the

3'

untranslated

region

(UTR)

region

areencoded

by tilapia

coding

exon4. These

data

show that the

IGF-II genes

have

significantly differing

structuresin vertebrate

evo-lution,

and therearedifferences of

_interrupting

introns

in the

IGF-I

_genomic

structure

_compared

with

mam-mals. To obtain recombinant

_biologically

active

_{polypeptides,}

_tilapia

IGF-II B-C-A-D domains were

ampli-fied

_using

the

_polymerase

chain reaction

(PCR),

then

_ligated

with

_glutathione

S-transferase

_(GST,

pGEX-2T

vector).

Tilapia

recombinant IGF-II

protein

was

purified

and characterized in E.

coli.

The fusion

protein

was

also

_digested

with thrombin

and

_appeared

as a

recombinant IGF-II

polypeptide single

band

with

a

molecu-larmassof 7

kD. The recombinant

tilapia

IGF-II

protein biological

function

wasmeasured

by

stimulation of

[3H]thymidine

incorporation.

The assay

concentration

wassetup

from 0

to

120

nMto

stimulate

_tilapia

ovary cell line

_(TO-2)

_{significantly}

to

_{uptake thymidine.}

The

results

_suggest

that the

recombinant IGF-II

_protein

was dose

dependent.

Brickell,

1996),

Sparus

aurata

_(Duguay

et

_{al., 1996),}

_sheep

(Brown

et

_{al, 1990;}

Demmer et

al,

1993),

rainbow trout

(Shamblott

and

Chen,

1992),

mice

_(Stempien

et

al,

1986),

hu-mans

(Bell

et

al, 1984;

Jansen et

al,

1985),

andrats

(Dull

et

al,

1984).

IGF-II is

_thought

to

_play

an

_important

role in

mam-malian fetal

_{development (Gray}

et

al,

1987;

Cohick and

Clem-mons,

1993);

the

highest expression

of IGF-II mRNA occurs

in the fetus andneonate

_(Soares

et

_{al, 1985),}

_{and gene}

target-'Instituteof

_Zoology,

National Taiwan

_{University, Taipei,}

Taiwan,R.O.C. instituteof

_Zoology,

AcademiaSinica,

Taipei,

Taiwan,R.O.C.

instituteof FisheriesScience,National Taiwan

_{University, Taipei,}

Taiwan,R.O.C.

INTRODUCTION

Insulin-like

growth factor II

(IGF-II)

is a

_single-chain

polypeptide

that contains the

NH2-B-C-A-D-COOH

domain. The

_{signal peptide}

and E domainareremoved untilmature

pep-tide

_production.

IGF-II structure consists of three disulfide bonds

_(Blundell

et

_al,

_1978),

_{and cDNA sequences}are

highly

conservedindifferent

animals,

including

chickens

_(Darling

and

(2)

ing experiments

have shown that IGF-II is a

key

_component

regulating

fetal

growth

(DeChiara

et

_al,

_1990),

soIGF-II is also

calledafetal

_growth

factor. Administration of IGF-II in the rat's central nervous _system can increase food intake and

change

feeding

behavior

_(Lauerio

et

al,

1987).

As mentioned

above,

IGF-II has

_potent

_mitogenic

and

meta-bolic effects ininvivo andinvitrosystems

(Humbel

1990;

Luthi

et

al, 1992;

Jones and

_{Clemmons, 1995).}

In mouse

IGF-II,

mRNAis

_{strongly expressed throughout}

embryogenesis;

abun-dant IGF-II mRNA wasfound in all

mesodermally

and

endo-dermally

derived organs, butwasnotdetected in the

develop-ing

nervous

_{system (Lee}

et

al,

1990).

Cultured

_sheep

choroid

plexus epithelial

cells can

synthesize

and secrete IGF-II and IGF

_{binding protein-2,}

which

_suggests

that the choroid

_plexus

epithelium

is the

_important

_organ

_secreting

these

_polypeptides

(Holm

et

_al,

_1994).

In

mammals,

IGF-II is an

imprinting

_gene

(DeChiara

et

_{al, 1991;}

_Rappolee

_et

_{al, 1992;}

_Willison,

_1991),

with IGF-II transcribed from the

_paternal

_{copy and the} IGF-II/mannose

_6-phosphate

_receptor

transcribed from thematernal copy. The

biological

purpose of

imprinting

is

_{supported by}

the

survival

hypothesis

for

_placental

mammals

(Haig

and

Graham,

1991),

butthe sameisnotknown about fishes.

IGF-II has

positive

effectsonfetal

_growth

and

regulation

in

mammals. These

_multiple

functions are reflected in the

com-plex

gene structure; the human IGF-II gene consists of 10 ex-onsof about 30 kb in

_length

ofDNA

(de

Pagter-Holthuizen

et

al, 1987, 1988;

Holthuizenet

al,

1990).

The sequence

encod-ing

the mature,

circulating

70-amino-acid

_polypeptides

are

con-tained withinexons

_{8, 9,}

and 10. Rat andmouseIGF-II genes

consist of 6exons_{and span about}12kb ofDNA

(Rotwein

and

Hall, 1990;

Ikejri

et

al, 1990,

1991).

The

_{prepropeptide}

is also contained within exons

_{4, 5,}

_{and 6. The ovine IGF-II gene is}

comprised

of 9exonsthat _span

_{approximately}

25

kb,

and the

coding region

is contained within exons

8, 9,

and 10

(Ohlsen

et

al,

1994).

In

fish,

the_{IGF-II cDNA sequence from liver is}

only

presentin

Sparus

aurata

_(Duguay

et

_al,

₁₉₉₆₎

and rain-bow trout

(Shammblott

and

Chen, 1992);

cloning

and

se-quencing

ofthe

_piscine

IGF-II gene hasnotbeen

_reported,

and

no_papers were found

reporting

onthe

_production

of the

bio-logically

active fish IGF-II recombinant

polypeptides

from

Es-cherichia coli. In

fish,

IGF-II-like

_peptides

are

reported

in

in-sulin cells of _{the elasmobranchian endocrine pancreas}

(Reinecke

et

al,

1994). Furthermore,

it isnotclear if the func-tion of IGF-II in fish brain cell is autocrineor

paracrine.

So

re-combinant

_tilapia

IGF-II

_protein

should be

_produced

as soon as

possible

to

_investigate

the

_{physiological}

functions of

_piscines,

especially

fish brain cell

_{physiological}

_{functions. It is very}

im-portant

toause a

_protein

_expression

_system

that

produces

high-quantity

and

_{quality tilapia}

recombinant IGF-II

_proteins

for

an-tibody preparation,

immunohistochemical

_{study, biological}

activity

analysis,

radioimmunoassay

(RIA),

and

enzyme-linked

immunosorbent assay

(ELISA).

In this _paper, we first _report

cloning

and

_sequencing

of the IGF-II cDNA _{gene from} the

tilapia (hybridized species)

brain cDNA

_library

and the

_cloning

and

_sequencing

ofthe

_tilapia

(Oreochromis

mossambicus)

IGF-II _{gene of the}

_{coding region.}

We found that the

_coding

exon

arrangementsare_{very different from those of mammalian}

IGF-II genes, and the

expression

of the

_tilapia

IGF-II mature

pep-tide

_by

theE. coli gene

expression

system.

_{Biological activity}

analysis

was conductedwith the

_tilapia

ovary cell line

(TO-2

cell

line).

The

stimulatory

effectof recombinant

_tilapia

IGF-II

polypeptides

wasconcentration

dependent

with

_{high activity}

in

[3H]thymidine

incorporation,

however,

suggesting

recombinant

tilapia

IGF-II

_polypeptides

canstimulatecellular

_{proliferation.}

MATERIAL AND

METHODS

Isolation

of tilapia

IGF-II

cDNA clones

Acanthopagrus

schlegeli

IGF-I cDNA

_(unpublished

data)

from the S domain to E domain was

amplified by

the

poly-merase chain reaction

(PCR).

The PCR

product,

about a

551-bp

DNA

_fragment,

was

_{purified by}

electroelution and used as a

probe

for

_isolating

clones from the

_tilapia

(hybrid)

brain

cDNA

_library

by

the

_{plaque hybridization}

method

(Maniatis

et

al,

1982).

About 1 millionrecombinant

_{bacteriophages}

were

seeded on 12LB

_plates

and transferred to

_nylon

membranes. After

_{denaturing, renaturing,}

and

_{cross-linking,}

the membranes

were

hybridized

to the

_probe

of

_{Acanthopagrus}

_schlegeli

IGF-I S domain to E domain PCR

_{products. Hybridization}

buffer used was 40% formamide

_containing

NaDodS04 (7

grams/100

ml),

0.5 M EDTA

pH

8.0

(100

/al/100

ml),

50% PEG8000

₍₂₀

ml/100

_ml),

40%formamide

(40

ml/100

ml)

at

37°C for 16 hr. After

_{hybridization}

filterswere washedin 2X

SSC,

0.1%

NaDodSCv,

0.5X

SSC,

0.1%

_NaDodS04;

0.1X

SSC,

0.1%

NaDodS04

at room

_{temperature, 37°C,and}

_40°C.

Positive

_plaques

allowed in vivo excision of the

_pBluescript

phagemid

from theUni-ZAPvector.

Isolation

_{of tilapia}

IGF-II

_genomic

clones

Approximately

1 million recombinant

_{bacteriophages}

from

the

_tilapia

(O. mossambicus)

_{genomic library}

were screened with

32P-labeled

_tilapia

IGF-II cDNA

_fragments.

The

_tilapia

ge-nomicDNA

_library

wasconstructedin

_phage

charon 40

_cloning

vector.

_{Hybridization}

buffer used was 50% formamide

(hy-bridization buffercontentswerethesamewith the isolationof

tilapia

IGF-IIcDNAclone

conditions),

at42°C for 16

hr;

after

hybridization,

filters were washed four times in 0. IX

SSC,

0.1%

NaDodS04

at 65°C.The

_positive

plaques

were

purified

andrestriction

_mapping

ofDNA from

_{plaque-purified positive}

isolateswasdone

_by

Southern

_blotting

with

32P-labeled

_tilapia

IGF-II cDNA

_probes.

Nucleotide

_sequencing

_and

_analysis

cDNA

clones,

containing pBluescript

double-stranded

phagemids

with the cloned DNA

insert,

appeared

onthe

LB-ampicillin

plate.

Atotal of four cloneswereobtained and used for

_{large-scale plasmid preparation.}

The entire

cDNA,

digested

with Pst Irestriction enzyme, was subcloned into

_pUC18

and transformed into

JM109,

and then

sequenced by

the

_Sänger

dideoxy

chain-terminationmethod

_(Sanger

et

_al,

₁₉₇₇₎

_and

se-quenase kit

(USB,

version

2.0).

The

_tilapia

IGF-II

_phage

DNAs

were

digested

with Sac

I,

then subcloned into the

pBluescript

vectorand transformed intoXL1 BlueE. coli host cells. Next the

QIAGEN

plasmid

extraction mini-kit was used to extract

DNA,

andonestrandwas

sequenced by

anABIautosequencer. The_{nucleic acid sequences} were

compared

with all

published

(3)

Construction

_of

recombinant

_tilapia

IGF-II

expression

vector

Tilapia

IGF-II,

fromB domaintoD

domain,

was

amplified

by

PCR and two

_{oligonucleotide}

_primers:

5'-CG-GAATTCATATGGAAATGGCCTCGGGCGGAGACGC;

and 5'

-CGGAATTCCTCATTCGGACTTGGCAGGTTTG-GCAC. Thesetwo

_primers

containedtheEcoRIsite. TheATG

initiation codonwas

designed

in the front of the

tilapia

IGF-II

B

_domain,

andthestopcodonwasconnected after the final

se-quence ofthe

tilapia

IGF-IIDdomain. The PCR

_products

were

constructed with the

_{glutathione-S-transferase}

(GST)

gene fu-sion system

(Pharmacia

Biotech)

of

_expression

vector

(pGEX-2T).

We chose the GST gene fusion

system

(pGEX-2T)

to

express

tilapia

IGF-II as afusion

protein

with

glutathione-S-transferase because the GST_gene fusion

_protein

would be

ex-pected

tobe

efficient,

the fusion

_proteins

tendtobe

soluble,

it isnot_necessarytoisolatethemfrom inclusion

bodies,

and

_they

are

produced

in

_{high yields.}

The thrombin

_{site-specific}

cleav-age of GST fusion

protein

thencanbe

rapidly purified

by

glu-tathione-agarose affinity chromatography

(Koland

et

al, 1990;

Hooveret

al, 1991;

Kwang

et

al,

1991).

The

_expression

vec-torwastransformedinto

BL21(DE3)

E.coli cells andselected

by

ampicillin. Colony hybridization

and

_sequencing

wereused

to

_identify

thecorrectdirection for insertion.

Expression

and

_{purification of}

_tilapia

IGF-II recombinant

_protein

A

_{single colony}

of

BL21(DE3)

E.coli cells

_containing

a

re-combinant

pGEX-IGF-II

plasmid

was inoculated in 100 ml of

2X YTA medium

₍₁₆

_grams/liter

_tryptone;

10

_grams/liter

yeast

extract; 5

_grams/liter

NaCl;

100

pglm\ ampicillin)

andincubated

at37°C for 10hrwith

_shaking.

Theculturewasthentransferred

into 500 ml of2X YTA medium and incubated at 37°C with

shaking

until the absorbanceat600nm was 1.1. Then 0.1 mA/

isopropyl-thio-D-galactoside

was added to the culturemedium

andthe culturewasincubatedat25°C for3 hrwith

shaking.

The culturewasthen

_centrifuged

at

_7,700

X_gfor 10 minat4°Cand

resuspended

by

50

p\

of ice-cold 1X

_{phosphate-buffered}

saline

(PBS)

per milliliter of culture. The cellswere

disrupted by

son-ication and 1% Trition X-100

_(final

concentration)

was added.

The

lysed

cellswere

centrifuged

at

12,000

Xgfor10 minat4°C

and the

_supernatant

was

passed

_through

a

_0.45-/nm

filter andthen

aspirated

intoa column

(Pharmacia Biotech).

Thecolumn was

washed

_by

1X PBS and thenathrombin solutionwasloadedinto the columnat25°C for 16 hr. The reaction

mixture,

which

con-tainedthe IGF-II

protein,

was collected. The recombinant

pro-teinwas run on

NaDodS04-PAGE gel,

and then transferredtoa

PVDF membrane for amino acid

sequencing.

Recombinant

tilapia

IGF-II

_polypeptides

and GST

_protein

were

_separated

on

15%

_{polyacrylamide}

gels,

and the

_protein

wasblottedontoa

Hy-bond ECL nitrocellulose membrane

(Amersham

Life

Science).

The

protein

was detected

by

anti-IGF monoclonal antibodies

(kindly

provided by

Dr. Chi-Yao

_Chang

of the Institute of

Zo-ology,

Academia

Sínica,

R.O.C).

The ECLWestern

_blotting

was

performed

according

toAmersham Life Science

_protocols.

Bioassay

The

_bioactivity

of

_tilapia

IGF-II

_protein

wasmeasuredwith

anin vitro_assay.

_{[3H]Thymidine}

_{incorporation}

into DNAina

TO-2cell line

(Chen

et

al,

1983)

wasstudied. TO-2 cells

(3

X

104 cells/well)

were seeded in 24-well

_plates

in MEM/F12 medium

supplemented

with 10%

(BSA)

for24hr. After 24 hr of serum-free

incubation,

the cellswereincubated withor

with-outvariousamountsof IGF-II

(0-120

x\M)

for18 hr.The cells

werethen

pulse-labeled

with

[3H]thymidine

(2

pCilmX)

for 2

hrat25°C. Then 0.5 mlof 0.3NNaOHwas

added,

and after 5 min the mixture was transferredto scintillation vials. After addition of 5 ml of

_aquasol,

the solutionwascounted ina

scin-gaattcgcggccgcctaactcacctgcaatcacaccaaccaaataattcccaacattttg 61 actactgccatctgacatggaaacccagcaaagatacggacatcactcactttgccacoc 1 METQQRYGHHSLCHT 121 ctgccggagaacgcagaacagcagaatgaaggtccagaggatgtcttcgacgagtcgggc 16 _{CRRTQNSRMKVQRMSSTSRA} 181 gctgctctttgcactggccctgacgctctacgcagtggaaatggcctcggcggagacgct 36 LLFALALTLYVVEMASAETL 241 _{gtgtgggggagaactggtggatgcgctgcagtttgtctgtgaagacagaggc'ttttattt} 56 CGGELVDALQFVCEDRGFYF 301 cagtaggccaaccagcaggggtaacaaccgacgcccccagacccgtgggatcgtagagga 76 SRPTSRGNNRRPQTRGIVEE 361 gtgttgtttccgtagctgtgacctcaacctactggagcagtactgtgccaaacctgccaa 96 CCFRSCDLNLLEQYCAKPAK 421 gtccgaaagggacgtgtcagccacctccctacaggtcataccggtgatgcccgcactaaa xl6 _{SERDVSATSLQVIPVMPALK} 481 acaggaagttccgaagaagcaacatgtgaccgtgaagtactccaaatacgaggtgtggca 136 _{QEVPKKQHVTVKYSKYEVWQ} 541 gaggaaggcggcccagcggctccggaggggtgtccccgccattctgagggccagaaagta !56 _{RKAAQRLRRGVPAILRARKY} 601 taagaggcacgcggagaagattaaagccaaggagcaggctatcttccacaggcccctgat 176 _{KRHAEKIKAKEQAIFHRPLI} 661 cagccttcctagcaagctgcctcccgtgttactcaccacggacaactttgtcagtcacaa 196 SLPSKLPPVLLTTDNFVSHK 721 _{atgagcccgctgccagccctttgcacagacaagagttttgagggtgaaaaaaagactagg} 781 ggattatagctttgtctctgacgtcatttcagtggcagtcctctttgacctcccctgccc 841 tgtccgagctcaccaatccctccccctgcacatatccactacgtcttgaacccctggccc 901 ttttctaatgacccnnttaaacccgaactcccccctccccaccaacccaccctcctctgg 961 _{cacacagacatgccttcacattcttcctgtctgaactctttctctcccaccctctttcag} 1021 tcactgatacaaaaggcacaaacacaaaacgtcgaacaaaaagttaacaatttggctgaa 1081 tgcggttcaggtggatccttaagcaaaagacaaaaagagaagggaaaaagaagatgaaag 1141 agatctgtcgtttgcaagtgtcaagaggacacctagcggaatgttttttgtccttgtgga 1201 agacaactgaaagtgaagagctgcttgcatgaaagaatccattccacctcattttcctga 1261 ggcaaaagaaaatctccgttagtcetttagtctgcacctctacctgtaatgggactteca 1321 _{cactgtaaggaattattttgtaaaattagattcctgttccagcaccttttgatcacaaac} 1381 aaaaagcagaaaagagtctgcaaaattgcacattgccacggattacgtctttgtaagaaa 1441 aaaatgggcactattttttcatgaacaatgaacgtgtagcttaaaaaaatgtcacggtgc 1501 _{tagctttgggaatggactcaaagaagaggtggaaaagcacgtttttttttctttgaatta} 1561 ataattaaagctttccgttttaaggaaagtgtgactttttaaaaaaaggaaaattttgga 1621 tatgggggagctctggcagtggcaatgtcaagggggaaagagtcactgaggaaaaatatg 1681 _{ggctgtgttggcatctaggctcatggtgagtnctagcggctgctatttactagtttgcca} 1741 gcataagncagcaagggatgacccgagacctagtccctgttcctcctgtccctctgaggc 1801 tgctggacacatggagcactatggggacacatacgggacaccatggaccacctggattgg 1861 _{gacagtactatagttcggggacagtacaacctgtttgccatggctttgcggactgttctg} 1921 _{gcaggaagtaacatggcatggactaagaacgagtggggcggccgcgaattc}

FIG. 1. Nucleotide sequence of

tilapia (hybrid)

IGF-II cDNA and the

_predicted

_{amino acid sequence of the hormone. The} nucleotideswerenumbered

beginning

with the first nucleotide

atthe 5' end.The numberonthe second line indicatesthe

or-der of the amino acid

_position.

IGF-II contains a

_signal

pep-tide of 47amino acid residues

_(1-47),

aB domain

_peptide

of 32 amino acid residues

(48-79),

a C domain

_peptide

of 11

amino acid residues

_(80-90),

anAdomain

_peptide

_(91-111),

a D domain

peptide

of6 amino acid

residues

(112-117),

andanEdomain

peptide

(118-216).

The5' _{UTR sequence contained}atotalof 76 nucleotides in

_length.

The 3'UTRsequence containeda

to-tal of

1,247

nucleotides in

length.

Asterisk

(*),

Start

_{codon; #,}

(4)

Tilapia

Sparus

aurata Rainbow trout Chicken Human Rat Mouse

Sheep

B domain C domain

EMAS..AETLCGGELVDALQFVCEDRGFYFSRPT

SRGNNRRPQTR

EVAS..AETLCGGELVDALQFVCEDRGFYFSRPT

SRGNNRRPQNR

EVAS..AETLCGGELVDALQFVCEDRGFYFSRPT SRSNSRRSQNR

AYGTAETLCGGELVDTLQFVCGDRGFYFSPRV

GRNN.RR.INR

AYRPSETLCGGELVDTLQFVCGDRGFYFSRPA

SRVS.RR..SR

AYRPSETLCGGELVDTLQFVCSDRGFYFSRPS

SRAN.RR..SR

AYGPGETLCGGELVDTLQFVCSDRGFYFSRPS

SRAN.RR..SR

AYRPSETLCGGELVDTLQFVCGDRGFYFSRPS

SRIN.RR..SR A domain Ddomain

GIVEECCFRSCDLNLLEQYCA

KPAKSE

GIVEECCFRSCDLNLLEQYCA

KPAKSE

GIVEECCFRSCDLNLLEQYCA

KPAKSE GIVEECCFRSCDLALLETYCA KSVKSE GIVEECCFRSCDLALLETYCA TPAKSE GIVEECCFRSCDLALLETYCA TPAKSE GIVEECCFRSCDLALLETYCA TPAKSE GIVEECCFRSCDLALLETYCA APAKSE Ancestral vertebrate

_{??A????ETLCGGELVD?LQFVC?DRGFYFSR??}

?R???RR???R GIVEECCFRSCDL?LLE?YCA ???KSE

Tilapia

Sparus

Sheep

signal peptide

METQQRYGHHSLCHTCRRTQNSRMKVQRMSSTSRALLFALALTLYVV

METQQRHGRHSLCHTCRRTESSRMKVKKMSSSSRALLFALALTLYVV

METQKRHEYHSVCHTCRRTENTRMKVKMMSSSNRVLVIALALTLYIV

MCAARQILLLLLAFLAYALDSAA

MGIPMGKSMLVLLTFLAFASCCIA MGIPVGKSMLVLLISLAFALCCIA MGIPVGKSMLVLLISLAFALCCIA MGITAGKSMLALLAFLAFASCCYA

Tilapia

Sparus

Sheep

Edomain

RDVSATSLQVIPVMPALKQEVPKKQHVTVKYSKYEV1FQRKAAQRLRRGVPAILRARKYKRHAEKIKAKEQA.IFHRPLI

RDVSATSTQVLPVMPPLKQEVSRKQHVTVKYSKYEVWQRKAAQRLRRGVPAILRAKKYRRQAEKIKAQEQA.

IFHRPLI

RDVSATSLQ11PMVPTIKQDVPRK.HVTVKYSKYEAIQRKAAQRLRRGVPAILRARKFRRQAVKIKAQEQA.MFHRPLI

RDLSATSLAGLPALN..KESFQKPSH..AKYSKYNVWQKKSSQRLQREVPGILRARRYRWQAEGLQAAEEARAMHRPLI

RDVS.TPPTVLP.DNFRR..

YPVGKFFQYDTW.KQSTQRLRRGLPALLRARRGHVLAKELEAFREAKR.HRPLI

RDVS.TSQAVLP.DDFPR..YPVGKFFKFDTW.RQSAGRLRRGLPALLRARRGRMLAKELEAFREAKR.HRPLI

RDVS.TSQAVLP.DDFPR..YPVGKFFQYDTW.RQSAGRLRRGLPALLRARRGRMLAKELKEFREAKR.HRPLI

RDVS.ASTTVLP.DDFTA..YPVGKFFQSDTW.KQSTQRLRRGLPAFLRARRGRTLAKELEALREAKS.HRPLI

Tilapia

Sparus

aurata Raiinbow trout Chicken Human Rat Mouse

Sheep

SLPSKLPPVLLTTDNFVSHK*.. SLGSKLPPVLLATDNYVNHK*.. TLPSKLPPVLPPTDNYVSHN*..

SLPSQRPPAPRASPEATGPQE*.

ALPTQDPA.HGGAPPEMASNRK*

VLPPKDPA.HGGASSEMSSNHQ*

ALPTQDPATHGGASSEASSD*..

FIG. 2.

_Comparison

of theamino_{acid sequence of}

_tilapia

IGF-II,

S. aurata

_IGF-II,

rainbowtrout

IGF-II,

human

IGF-II,

rat

IGF-II,

mouse

_IGF-II,

_sheep

_IGF-II,

and chicken IGF-II.

_Sequences

start atthe first methionine

_peptide

amino acid residue. The

IGF-II

_{prepropeptide}

is divided intothe

signal peptide,

and theB,

C,

A, D, andEdomain. Adot

_{(•) represents}

a

_{gap/deletion.}

Hypothetical

ancestral vertebrate IGF-II BdomaintoDdomain sequencesareshown below. dilation counter.

_Every

_parameter

in this

_experiment

was

re-peated

three times.

RESULTS

Isolation and

characterization

_{of tilapia}

IGF-II cDNA _gene

of

_Sparus

aurata

_(Duguay

_et

_al,

₁₉₉₆₎

and rainbowtrout

(Shamblott

and

Chen,

1992)

IGF-IIcDNA

geneswerefromthe livercDNA

_library.

But

here,

wescreened about 1 million recombinant

_{bacteriophages}

from the

_tilapia

(hybrid)

brain cDNA

library,

andwe

finally

obtained four

pos-itive colonies. The recombinant

_plasmids

of each of these clones

were excisedin

vivo, extracted,

and sized

_by

1% agarose

gel

electrophoresis.

One of the four

clones,

designated

as

12-1,

was

chosen for further studies. The size of the cDNA

_appeared

to

be about 2 kb and

_{by sequencing}

wasidentifiedas

_tilapia

IGF-II. The nucleotide sequences were

_originally

cloned into the

Eco RI site ofthe

phage

ZAP vector. The recombinant DNA

(5)

Ex 1

Ex

2 Ex

3 Ex 4

Tilapia

Lys

S25 ValS26 Ser B29

Arg

B30

Ex8

Ex9

Ex

10

Sheep

Ser B29 (A) SerB29

(GC

Pro Ell

_AspE12

FIG. 3.

_Comparison

of

sheep

_{coding region}

structureandthe

_organization

of the

tilapia

IGF-II

_{coding region.}

Exonsareshown

by

boxes,

and introns and

_flanking

_sequenceare shown

by

thin lines. At the bottom of each structureare the relative locations

oftheexonand intron boundaries.

in

_Fig.

1. _{IGF-II cDNA gene contains 76}

_bp

in5' untranslated

region

(UTR),

1,246

bp

in the 3'

_UTR,

and the

_{coding region}

hasa

length

of 645

bp.

The B toDdomains of the IGF-II

ma-ture

peptide

translated into a 70-amino-acid residue. The first

47 amino acid residues

possibly comprise

the

signal

peptide,

whereas the last 98 amino acid residues

comprise

theEdomain.

The IGF-II amino acid

_comparison

of different animals is

shown in

_Fig.

2.

_Comparison

of

_predicted

amino acid

_tilapia

IGF-IIBto Ddomains with rainbowtroutIGF-II B toD do-mains shows 95.7%

_similarity

and 92.9%

_identity.

In

addition,

the

_length

of the

_tilapia

IGF-II StoEdomains

_compared

tothat of rainbow trout S to E domains shows 90.7%

_similarity

and

81.8%

_identity.

_However,

_tilapia

IGF-IIBtoDdomains

com-pared

to

_chicken,

_human,

rat,mouse,and

_sheep

IGF-II BtoD

domains,

possess similarities of

83.1%,

79.1%, 80.6%, 83.6%,

and

80.6%,

and identities of

78.5%,

77.6%, 79.1%, 79.1%,

and

79.1%,

respectively.

With the

predicted

amino acid sequence

comparison

between fish

_species,

weinferred that the ancestral

fish IGF-IIBtoDdomainswere

_highly

_conserved,

andinamino acid sequence

comparisons

with mammalianIGF-IIBtoD do-mains hada 3-codon

insertion,

and in theB domain had a 2-codon deletion. These

_phenomena

also existed between

_tilapia

and rainbowtrout.

_Figure

2 shows that the

tilapia

mature

IGF-II

_peptide

has 5 amino acids different from the othertwo

pub-lished fish IGF-IImature

_{peptides. They}

are locatedonB2

(for

tilapia

it is

Met;

inrainbowtrout and S.aurata,

they

are

Val),

C3

(for

tilapia

and S.aurata,

_they

are

Gly;

inrainbow trout, it

is

Ser),

C5

(for

tilapia

andS. aurata,

they

are

_Asn;

inrainbow trout, it is

Ser),

C8

_(for

_tilapia

and S. _aurata,

_they

are

_Pro;

in

rainbow trout, it is

Ser),

and C10

(for

_tilapia,

it is

Thr;

in rain-bowtroutandS. aurata,

they

are

_Asn).

Isolation and

characterization

_of

tilapia

IGF-II_gene

coding

region

About 1 million recombinant

_{bacteriophages}

froma

tilapia

(6)

TilapiaIGF-IICodingExon1

1 TACACTGCGTAAACGTGGAAAA TGCCCA TGGAAGTCTTCCA TA TTTTGTGACTCTCACC 60 CTCTTATTTCTCCCTTCAAGCACTTTCA TAAAACGrCTCTCCGCCTTTTTTTTTTCATC 119 GGCGAAGAGGAGGAGCAAGGGGTGGGGTCGGTGTAAGGCGCGTGCTTTAGTATATAATA 178 CCTCTCCCTGAGAAGTTTTGCCTGTCGCCTAGTCTTTGGCACAGCTTCTCACTCACCA T 237 CrCTATACTTTAACCCAACTGGGAAACTfiKCTUCCTGCUTUCKCUtiCUkMkKI 296 TCCCAACATTTTGACTACTCCCATCTCACATCGAAACCCAGCAAAGATACCGACATCAC I METQQRYGHH 355 _{TCACnTGCCACACCTGCCGGAGAACGCAGAACAGCAGAATGAAGgtaaccaaagaaca} II SLCHTCRRTQNSRMK 414 _{agcaaattgttttatactctccggctctgccgtgcgcgtaatgnaagagtat.} .- 0.8 Kb -.

TilapiaIGF-IICodingExon 2

1 tgtacctcttcgtctgaaaaaaaaaaaaaaaatctggctgattttgattaaaaaaatgg 60 tatttaactgtcattaactgttattttgttaacgatttctgtatgccacaactttctgc 119 atatcatgggtacatttggtgaaccccatgcttcattccgcagGTCAAGAAGATGTCTT 26 V K K M S S 178 CCACGAATCCCGCGCTGCTCTTTGCACTCGCCCTGACGCTCTACCTAATGGAAATGGCC 32 TNPALLFALALTLYLMEMA 237 TCCGCGGAAACCCTGTnGGGGGAAAACTGGTGGATGCGCTGCAATTTGTCTGTGAAGA 51 _{SAETLFGGKLVDALQFVCED} 296 CAGAAGCTTTTATTTCAgtaagtttcaaagcattacnagtttccccaatggctgcgtga 71 R S F V F S 355 ttgctcatttgcctgttgaatctctctgttgtgcccttgcacacatctgtttggagcaa 414 aagtgggaagttacccactacnaatacttcgttactgtactccagtatagttttcagtt 473 agaatttttgccccctacatttttaaacagatatctgtactttctactcc. .- 2.8Kb -.

1 _{gatgttgtgtttgcagtccctaacctntacgtcttcattcctttttgtgtttttcctca} 60 gGTAGGCCAACCAGCAGGGGTAACAACCGACGCCCCCAGACCCGTGGGATCGTAGAGGA 77 RPTSRGNNRRPQTRG1VEV 119 GTGTCTTTTCTGTAGCTGTGACCTCAACCTACTGGAGCAGTACTGTGCCAAACCTGCCA 96 CLFCSCDLNLLEQYCAKPAK 178 AGTCCGAAAGGGACGTGTCAGCCACCTCTCTACAGGTCATACCGGTGATGCCCGCACTA 116 SERDVSATSLQV1PVMPAL 237 AAACAGgtacgtctaagcaacaacaacaacaggccagtatgggaaatagtgctaatccc 135 K Q 296 agctctatctgtcctcccatctcctgtgcccccattcacctctgaggctagcccctatg 355 tcactgactcUgagtagagtgtacccacgctaacgcagttatatctagUaattggcc 414 aatggaaagcactcaacttacaaagaaagtgctgacagtcatggaaaacattacaaaag 473 tcacaacacgttatattcaggaaaaggaatgtgttaagtgcgtatatgaaggaa. .- 1.3 Kb -.

1 attgaacaatatnttatnaccntaatgaatgatccttcttttccctttttcttctattt 60 tcgcccgcacgccacaatagGAAGTTCAGAAGAAGCAACATGTGACCGTGAAGTATTCC 137 E V Q K K Q H V T V K Y S 119 AAATACGAGGTGTGGCAGAGGAAGGCGGCCCAGCGGCTCCGGAGGGGTGTCCCCGCCAT 150 KYEVWQRKAAQRLRRGVPAI 178 TCTGAGGGCCAGAAAGTATAAGAGGCACGCGGAGAAGATTAAACCCAAGGAGCAGGCTA 170 LRARKYKRHAEKIKAKEQAI 237 TCTTCCACACGCCCCTGATCAGCCTTCCTAGCAAGCTGCCTCCCGTGTTGCTCACCACG 190 FHRPL1SLPSKLPPVLLTT 296 GACAACTTTGTCAGTCACAAATGAGCCCGCTGCCAGCCCTTTGCACAGACAAGAGTTTT 209 D N F V S H K * 355 GAGGGTGAAAAAAAGACTAGGGGATTATAGCTTTGGTCTTCTGACGTCATTTCTGTGGC 414 AGTCCTCTTTGACCTCCCCTCCCCTGTCCGAGCTC

FIG. 4. Partial nucleotide sequence of the O. mossambicus IGF-II gene

coding region.

The uppercase lettersrepresent

tran-scribed

regions

and lowercase letters

_represent

intronsor

flank-ing

sequences. The italic letters

represent

the

_regions

that

re-main uncertain

_by

cDNA

_sequencing.

Amino acids number:

1^-7,

signal peptide

sequence;

48-79,

B domain sequence;

80-90,

C _{domain sequence;}

91-111,

A domain _sequence;

112-117,

D domain sequence;

118-215,

E domain sequence. Thestopcodon is indicated withanasterisk

(*).

positive

colonieswereobtained. The four

_positive

colonieswere

extracted and restriction enzyme

mapping

of the DNA from the four

_{purified plaques}

was

performed.

The

_genomic

DNA

se-quences of the

tilapia coding region

weredivided into four ex-ons and were

mapped

as shownin

_Fig.

3. The

_{tilapia coding}

exons were foundto _span a

region

of

approximately

12.9 kb.

In mammalian IGF-II genes, the

coding region

is

_comprised

of threeexonsbutin

_tilapia

the

_{coding region}

is

_comprised

of four

exons.The

tilapia

IGF-II

coding

exon 1 wasfrom the 5' UTR

(compared

cDNA

_sequence)

tothe

_signal

_peptide

(S25);

cod-ing

exon2was from_partof the

signal peptide

(S26)

toB do-main

(B28);

coding

exon 3was from the C domain

(Cl)

toE domain

_(E20);

and

_coding

exon 4was from the Edomainto

part

of the 3'UTR

_(compared

cDNA

sequence).

In

_tilapia,

cod-ing

exon 1contains 25 amino

acids,

coding

exon2contains 51

amino

acids,

coding

exon3contains60amino

acids,

and

cod-ing

exon4 contains 80 amino acids.Betweenexon 1 andexon

2,

there isan

interruption by

anintron of about 0.8

kb;

between exon2 andexon

3,

there isan

_{interruption by}

anintron of about

2.8

kb;

and betweenexon3 andexon4there isan

_interruption

by

an intron ofabout 1.3 kb. _{The sequence of}thefour

_tilapia

coding

exons andpartof the

_flanking

_{sequence of introns}are

shown in

_Fig.

4. The sequences of the exon-intron

junctions

are often describedas

conforming

tothe GT-AG rule

(Breath-nach et

al, 1978;

Breathnach and

Chambón,

1981).

The

cod-ing region

of

tilapia

IGF-II gene

predicts

a

prepropeptide

of 215

amino

acids,

including

a 47-amino-acid

signal peptide,

a

70-amino-acidmature

_peptide,

anda98-amino-acidE

peptide.

The

predicted

amino acid sequences, 5' UTR and 3' UTRDNA

se-quences for

tilapia

IGF-II,

arecontrasted with those determined

by

cDNA sequence data. The

genomic

structure of the

_tilapia

IGF-II

coding region

showsgreatvariationwith

_previously

pub-lished mammal IGF-II gene

organization.

Construction and

_{expression of}

IGF-II

expression protein

In the first step, PCR was used to

_amplify

IGF-II mature

polypeptides

ofDNA

_fragments,

whichwerethen

digested

with

Eco RI. Inthe second

_step,

weconstructed PCR

_products

with

the

pGEX-2T

vectorof Eco RI

_digestion.

The

_ligation

products

were transformed into

BL21(DE3)

E. coli cells.

_Colony

hy-bridization andDNA

_sequencing

wereusedto

_identify

the DNA sequence and orientation.Toexpress the

tilapia

IGF-II

recom-binant

_{polypeptides,}

the

BL21(DE3)

E. coli cells

_including

re-combinant IGF-II

plasmids

wereinduced with0.1 mMIPTG and grown for 3 hrat22°C.

Figure

5 shows the total cell

_proteins

extracted from induced and noninducedE. coli cell

culture;

aclear band of 36kDwas

detected after 0.1mMIPTG induction for 30-180 min. The

36-kD

protein

was not

_digested

_by

thrombin,

as afusion

_protein

with GST. Lanes 2 to 7 show the

_protein

band

_present

in E.

coli cell

_{containing pGEX2T}

vextor

_ligated

withthe

_tilapia

IGF-II mature

_polypeptide

_{DNA sequence. Lane} 1 shows the pro-tein band

_{containing pGEX-2T}

vector

_only.

The final

_purified

IGF-II

_protein

was

compared

by

denatured

polyacrylamide gel.

This

_protein

is not found

abundantly

in inclusion bodies.

In-duced with 0.1 mM

IPTG,

afusion

protein

was

produced

with an_apparentmolecular

weight

of 36 kD. The final recombinant

(7)

12 3 4

5 6 7

12 3 4 5 6

30kDa—

GST

+

IGF-II

GST

FIG. 5.

_Expression

of

_tilapia

EGF-IImature

_peptide

inE.coli

BL21(DE3).

Cells were cultured in 2YT medium with 100

¿ig/ml ampicillin

at37°C until the

OD0oo

reached 0.4-0.6. Then the

_temperature

was shifted to 22°C and 0.1 mMIPTG was

added for induction of

_tilapia

IGF-Imature

_{peptide synthesis.}

The cells wereharvested after

30, 60, 90, 120, 150,

and 180

min

induction,

and the total

_protein

was extracted

by lysis

buffer and

analyzed by

NaDodSOa-PAGE

on a 10%

gel

with Coomassie Blue

_staining.

Lane

1,

Protein

_{expressed by}

pGEX-2Tvector

_alone;

lane

2,

cells

containing

IGF-I insert after

in-duction for30

min;

lane

3,

cells

_containing

IGF-I insert after induction for 60

min;

lane

4,

cells

_containing

IGF-I insert

af-terinduction for 90

min;

lane

5,

cells

_containing

IGF-I insert afterinductionfor 120

min;

lane

6,

cells

containing

IGF-I

in-sertafter induction for 150

min;

lane

7,

cells

containing

IGF-Iinsert after induction for 180min.

46kDa— 30kDa-~ 21.5kDa — 14.3kDa — 6.5kDa— IGF-II + GST — IGF-II

FIG. 6.

Expression

of

_tilapia

IGF-Imature

_peptide

inE.coli

BL21(DE3).

The E. coli cell culture conditionswerethesame asin

_Fig.

5. When the cell cultures wereharvested after 180 min of

induction,

the fusion

_proteins

were

_digested

with throm-bin and

_{purified by}

RedPack Module

(Pharmacia Biotech).

The

proteins

were

analyzed by

_{NaDodSOa-PAGE}

on a15%

_gel

with Coomassie Blue

_staining

(A)

and

_{immunoblotting}

(B).

Lane

1,

E. coli culture

_containing

the

cloning

vector

_(pGEX-2T)

alone;

lane

2,

E. coli culture

_containing

the

_tilapia

IGF-IImature pep-tideDNA_{sequence constructed with}

_{pGEX-2T cloning}

vector; lane

3,

IGF-II mature

_{peptide digested by}

thrombin and

puri-fied

by

RedPack

Module;

lanes

4—6,

monoclonal antibodies raised

_against

the

_tilapia

IGF for ECL immunoreaction. The

loading

order of lanes 4-6 arethe same anin lanes

1-3,

re-spectively.

single

band. The

_pGEX-2T

vectorin cells alone and GST fu-sion IGF-IImature

_polypeptide

with and without thrombin

di-gestion

were run ona

NaDodS04-PAGE gel

as shown in

Fig.

6. ECLwestern

_{blotting analysis explained}

thatthe

protein

can

be detected with the

_specifically

monoclonal

_anti-tilapia

IGF

antibody

(Fig.

6).

The recombinant IGF-II

_polypeptides

were identified with amino acid

sequencing.

The

_expression

ofthe IGF-II

polypep-tide and the

_predicted

IGF-II mature

_{polypeptide comparison}

are shown in Table 1. In Table 1, the

expression

of IGF-II

polypeptide

shows anadditionalsevenamino acids before the

mature

_{polypeptides.}

Of thesevenamino

acids,

sixare a

pGEX-2T multiclonal siteDNAsequencetotranslate amino

acids,

and

oneis the ATGstartcodon.

Characterization

_{of tilapia}

IGF-II

recombinant

polypeptides

After thrombin

_digestion,

the

_tilapia

_{IGF-II recombinant} pro-tein revealeda

single

band of7 kDon adenatured

NaDodSOa

polyacrylamide gel.

Totestthe

_biological

function of

_tilapia

re-combinant IGF-II

_{polypeptides, they}

were

_{analyzed by}

in vitro

assay of

incorporation

of

[3H]thymidine

intoDNA differentia-tion. The cell line

designated

TO-2wasestablished from ovaries

of

healthy

adult

_{tilapia hybrids}

(Tilapia

mossambicaX T.

nilot-ica)

as an

_experimental

cell line. Thetest concentrationswere

between 0 and 120n/Vfand

_{incorporation ability significantly}

increased over _{this concentration range}

(ANOVA;

F =

4.46;

df=

6.14;

_p<

0.05).

Figure

7. shows that Duncan

multiple

Table 1. Different Amino Acid

Sequences

Betweenthe Predicted and

Actual ExpressionofRecombinant Polypeptides

amino

acid

_sequence

Predicted

_expression

amino

acid

EMASAETLCGGEL.

Actual

_expression

amino acid

GSPGIHMEMASAETLCGGEL.

(8)

rangetestdetecteda

significant

difference in IGF-II

concen-tration between0and 120nM.

DISCUSSION

The novel

_findings

inthe

_present

_experiments

arethe

com-plete coding region sequencing

and

_analysis

of

_tilapia

IGF-II from braincDNA

_library

isolated

_{by plaque hybridization.}

The

Acanthopagrus schlegeli

IGF-I cDNA

_{coding region}

wasused

as a

hybridization probe

under

_{low-strength hybridization}

con-ditions.

_{Surprisingly,}

when the nucleotide sequencewas

com-pared by

the GCG GenBank program, the

greatest

homology

wasshown

_by

rainbowtrout

IGF-II,

and the secondwas

_by

hu-manIGF-II sequences. This extendsto

_eight

the numberof

an-imals

_species

forwhich

_published

_{IGF-II sequence is available.} In

_tilapia

and rainbow trout, the

_peptides

are

highly

conserved. Predicted amino acids between fish and mammalsdiffered

_by

the addition of2 amino acids in the B

domain;

in the C

do-main,

fish and mammals differed

by

an increase of 3 amino acids.A

change

inthe IGF-I aminoacid sequenceat

_position

B23-B25

_{(Phe-Tyr-Phe)}

will

_give

risetoadecreasein

ability

of

_binding

tothe IGF-I

_receptor

(Cascieri

et

al,

1988).

These amino acid sequencesarealso

_present

atB26-B28inthe

_tilapia

IGF-IImature

_peptide.

These

phenomena

existin all animals whoseIGF-II sequence

hasbeen

_published,

and

_they

are

_strictly

conserved.In mouse

andrat

IGF-II,

position

B22 is

_changed

from Serto

_Gly

com-pared

with

humans;

Gly

is found at

_positive

B22 in insulins

from all

_species

_except

_{hystricomorphs}

(LeRoith, 1991).

But in

10000r-1 . * 9000 L T

2

:

J?

Q 8000 7

/

1

.2 7000

_!"

*

yS

W

jr 9 6000 -/ O * /Y

S

5000

_~:

_7*^

M

4000

-/

. / >ï ₃₀₀₀ -Öl

*/

H : r V SC 2000 r

/

1000 '-oF. , . i i i i i i . . i , . . i . . , i , 0 20 40 60 80 100 120 IGF-II

_(nM)

FIG.7. Effects ofrecombinant

_tilapia

IGF-II

_polypeptide

on

stimulated

_tilapia

_{ovary cell}

(TO-2

cell

line)

proliferation.

The

following

effectsweremeasured with different concentrations

of recombinant

_tilapia

IGF-II

_{stimulating incorporation}

of

[3H]thymidine

intoDNA

_synthesis.

The data show that the TO-2cell membranemusthaveanIGF_receptor, andso

_represent

a

dose-dependent

effect.

fish,

this

_position

B22 is

Glu,

Why

is itnot

_Gly

orSer? The real mechanism is

_presently

unknown.In

_tilapia

IGF-I,

there is

an inference A/-linked

_{glycosylation}

site

(Asn-X-Ser/Thr),

but

it isnotfound in the

_tilapia

IGF-II

_peptide.

The role of these IGF-II

_peptides,

whether

_they

haveaA/-linked

glycosylation

site or_{not, remains for the}mostpartunknown in fish.

InVitro,IGFs have very

important

functions and actionson

neuronal and

_glial

cellfunction. The ribonuclease

_protection

as-say, insitu

_{hybridization,}

and

_{immunohistochemistry}

wereused

todemonstrate thatIGF-IIis

_{synthesized predominantly}

inthe

leptomeninges,

choroid

_plexus,

and

_parenchymal

microvascu-lature in rats, which

_presumably

_represents

the site of IGF-II

bioactivity

within the brain

(Logan

et

al,

1994).

In adult rats, IGF-II mRNA canbe detected in brain and other organs, and alsocanbe detected in rainbowtrout.

_Except

in the

liver,

lev-els of IGF-IImRNA in brain hasa

higher

_expression

than

to-talIGF-I in rainbowtrout

_(Chen

et

_al,

_1994).

_So,

there is no

doubtthata

_tilapia

IGF-II clonecanbeobtained fromabrain

cDNA

library.

Butinmost

fishes,

the

_{adenohypophysis}

is dif-ferentiated intoarostralanda

proximal

_{pars distalis and}a_pars

intermedia. Whether IGF-II has any function infish

adenohy-pophysis

is still unclear.

The exon

_organization

of the

_tilapia

IGF-II

_{coding region}

gene is very dissimilartothat of mammalian and avian IGF-II genes. In

sheep

IGF-II genes, the

promoter

directs the

tran-scription

of six

_noncoding

exonsand

alternatively

splices

tothe sharedexons

_{8, 9,}

and 10

(Ohlsen

et

_al,

_{1994). Up}

_{to now,}it has _{been determined that the IGF-II genes of mammals}

(hu-man, mouse,

_sheep,

rat) (Frunzio

et

al, 1986;

de

Pagter-Holthuizenet

al, 1987;

Soareset

al, 1986;

Rotwein and

Hall,

1990;

Holthuizen et

al, 1990;

_Ikejiri

et

al, 1990, 1991;

van

Dijk

et

al, 1991;

Ohlsen

étal,

1994)

and birds

(chicken)

(Dar-ling

and

Brickell,

1996)

have three

_coding

exons of similar

structure; but in fish

(O.

mossambicus)

the IGF-II gene has four

coding

exons. A

separation

of IGF-II_genestructure

_strategy

is

suggested

basedontherateof evolution of verterbrates.

Com-mon

_{evolutionary history}

for the insulin/IGF

family

_{genes may}

be duetothe

_phylogeny

of_{derived amino acid sequences.}

In-sulin and IGF genesarebelieved tohave evolved

by

repeated

duplication

and

divergence (Ellsworth

et

al,

1994).

The IGF-I

andIGF-II

_separation

isconcludedtohavetaken

_place

about 70_{million years}_ago,which is aboutthesametime asthe

ap-pearance of

placental

mammals

(Rinderknecht

and

Humbel,

1988).

These

_assumptions

arebasedonthe

publication

of IGF

sequences. It would seem that the

_{dissimilarity}

ofthe

struc-ture/function of the

_coding

exon arrangement of

tilapia

(Eu-teleostei)

compared

tobirdsandwarm-blooded vetebrates may have resulted from

_homoplastic

evolution.

Tounderstand the IGF-II

_protein

regulation

of fish

physiol-ogy, we have

_developed

the GST-IGF-II fusion

_protein

ex-pression

system. This isa

_{single-step purification}

of

polypep-tides

expressed

in E. coli as fusion with

glutathione-S-transferase

(Smith

and

Johnson,

1988).

The

low-temperature

induction of fusion

_{protein synthesis}

can

_improve

soluble

pro-tein

_production

_(Hartman

et

al,

1992).

Wetried many

temper-atureconditions and found that 22CCwassuitable for

purifica-tion of fusion

_proteins.

The novel

_tilapia

IGF-II

_protein

was

expressed

in E. coli andwas

highly

activein the TO-2 cell line.

Inrats, IGF-II

(50

ng/ml)

stimulates

_oocyte

maturation

_(Feng

(9)

and IGF-Iweredetected

throughout sheep preimplantation

de-velopment

from the one-cellto the

_blastocyst

_{stages (Watson}

etal,

1994).

IGF-II has

_specific

and

_{high-affinity binding}

sites for IGF-II

_receptors

onwhole ovarian membranes andonovarian

sec-tions,

suggesting

the IGF-II/M6P

_receptors

in ovarian tissuecan

remodel and mediate IGF-II actionon

folliculogenesis

(Teissier

et

al,

1994).

In most

fish,

the structureof the ovarian follicle is similar.The ovary consists ofa

_granulosa

cell

_layer

andone ortwo outer

_sublayers

of theca cells. The theca and

_granulosa

layers

aredivided

_by

abasement membrane. In

humans,

IGF-II mRNA is

_expressed

in newborn ovarian stromaand in both newborn and adult ovaries

(Zhou

and

_Bondy,

1993).

Intherat

ovary, in situ

_{hybridization}

and RNase

_protection

_assays

sug-gested

that the IGF-II

expression

in theca-interstitial cells is

specific

tocelltype

(Hernandez

et

al,

1990).

In contrast, the amino acid sequences of

tilapia,

avian,

and mammalian IGF-II

mature

_peptides

are _{very conserved}

_(general

amino acid

se-quences similaritiesare79% and

above).

Given the

above,

these

data

_suggest

thatmature

_tilapia

IGF-II

_polypeptides

_{may have} similar ovarian functions in fish as

compared

with those in mammals and birds. This

_explains

_why

the recombinant

_tilapia

IGF-II

_protein

canstimulate

_{thymidine incorporation}

_and

pre-sent

_{dose-dependent}

effects in the

_tilapia

_{ovary cell line.}

ACKNOWLEDGMENTS

We thank Dr. Thomas T.

Chen,

Dr.

Ching-Ming

Kuo,

and

Dr. Cho-Fat Hui for their

_appropriate

and concise comments

aboutthis

_experiment

and

_manuscript.

We thank Dr. Wei-Yuan

Chow for

_{kindly providing}

the

_{tilapia (hybrid)}

brain cDNA

li-brary

and Oreochromis mossambicus

_genomic

DNA

_library.

We thank Mr.

_Hung-Chih

Chen for

_providing

the

Acanthopa-grus

schlegeli

IGF-I cDNA

plasmid.

We thank Dr. I-Chiu Liao

for his

_support

andencouragement.This

_project

was

_supported

by

NSC

_grants

NSC 85-2321-B-001-007-A15

(R.O.C),

and NSC 86-2311-B-001-048-B24

(R.O.C).

to Dr. Jen-Leih Wu.

REFERENCES

BELL, G.I.,MERRYWEATHER, J.P., SÁNCHEZ-PESCADOR, R.,

STEMPIEN, M.M., PRIESTLEY, L., SCOTT, J.,andRALL, L.B. (1984).

Sequence

ofacDNA clone

_encoding

human

preproinsulin-like

_growth

factor II. Nature310,775-777.

BLUNDELL, S., BERDARKA, E., RINDERKNECHT, and

HUM-BEL,R.E. (1978). Insulin-like

_growth

factor: A model for

_tertiary

structure

_accounting

for

_{immunoreactivity}

and _receptor binding. Proc. Nati. Acad. Sei. USA75, 180-184.

BREATHNACH, R.,and_CHAMBÓN,P.(1981).

Organization

and

ex-pression

of

_{eucaryotic split}

genes

coding

for

_proteins.

Annu. Rev. Biochem.50,349-383.

BREATHNACH, R., BENOIST, C, O'HARE, K., GANNON, F.,and

CHAMBÓN,P.(1978). Ovalbumingene:evidence foraleader

se-quence in mRNA and DNA sese-quencesatthe exon-intron boundaries.

Proc. Nati. Acad. Sei. USA75,4853-4857.

BROWN, W.M., DZIEGIELEWSKA, K.M., FOREMAN, R.C., and

SAUNDERS,N.R.(1990).The nucleotide and deducedaminoacid sequences of insulin-like

growth

factor II cDNAs from adult bovine and fetal

_sheep

liver. Nucleic Acids Res. 18,4614.

CASCIERI, M.A., CHICCHI, G.G., APPLEBAUM, J., HAYES, N.S., GREEN, B.G.,andBAYNE,M.L. (1988).Insulin-like

_growth

fac-torIwith reduced

_affinity

for the_type 1 insulin-like

_growth

factor

receptor.

Biochemistry

27,3229-3223.

CHEN, S.N., CHI-UENO, S.C.,andKOU,G.H.(1983).Acell line

de-rived from

_Tilapia

_ovary.Fish.Pathol.18(1), 13-18.

CHEN, T.T., MARSH, A., SHAMBLOTT, M., CHAN, K.M.,TANG,

Y.L., CHENG, CM.,andYANG, B.Y.(1994).Structureand

evo-lutionof fish

_growth

hormone and insulinlike

_growth

hormone and

insulinlike

_growth

factor_{genes. In Fish}Physiology,vol. Xin. N.M. Sherwood and C.L. Hew, eds. (Academic Press, New York) pp.

179-209.

COHICK, W.S., and CLEMMONS, D.R. (1993). The insulin-like

growth

factors.Annu. Rev.

_Physiol.

55,131-153.

DARLING, D.C.,andBRICKELL,P.M.(1996).Nucleotide sequence

and

_genomic

structureof the chicken insulin-like

_growth

factor-II (IGF-II)

coding

region.Gen.

_Comp.

Endocrinol. 102,283-287.

DeCHIARA, T.M., EFSTRATIADIS, A., and ROBERTSON, E.J. (1990). A

_{growth deficiency phemotype}

inheterozygousmice

car-rying

an insulin-like

_growth

factor_{II gene}

disrupted by targeting.

Nature345,78-80.

DeCHIARA, T.M., ROBERTSON, E.J., and EFSTRATIADIS, A. (1991). Parental

_imprinting

ofthemouseinsuline-like

_growth

factor

Hgene. Cell64,849-859.

DEMMER, J., HILL, D.F.,andPETERSEN,G.B.(1993). Characteri-zation oftwo

_sheep

insulin-like

_growth

factor-II cDNAs with

dif-ferent5'-untranslated

regions.

Biochim.

_Biophys.

Acta1173,78-80. De PAGTER-HOLTHUIZEN, P., JANSEN, M., VAN SCHAIK,

F.M.A., VAN DER KÄMMEN, R.A., OOSTERWIJK, C, VAN DENBRANDE, J.L.,andSUSSENBACH,J.S.(1987).The human

insulin-like

_growth

factorIIcontainstwo

_{development-specific}

pro-moters.FEBSLett.214,259-264.

DE PAGTER-HOLTHUIZEN, P., JANSEN, M., VANDER

KÄM-MEN, R.A., VAN SCHAIK, R.M.A., and SUSSENBACK, J.S. (1988). Differential

_expression

of the human IGF-II _gene: charac-terizationof theIGF-IImRNAs andanmRNA

_encoding

a

_putative

IGF-II associated

_protein.

Biochim.

_Biophys.

Acta. 950,282-295.

DUGUAY, S.J., LAI-ZHANG,J.,STEINER,D.F., FUNKENSTEIN, B.,andCHAN,S.J.(1996).

Developmental

and

_{tissue-regulated}

ex-pression

ofIGF-I and IGF-II mRNAs in

_Sparus

aurata.J. Mol. En-docrinol. 16,123-132.

DULL, T.J., GRAY, A., HAYFLICK, J.S.,andULLRICH,A.(1984). Insulin-like

_growth

factorIIprecursorgene

organization

in relation

toinsulin gene

family.

Nature

_310,

777-781.

ELLSWORTH, D.L., HEWETT-EMMETT, D.,andLI,W.H.(1994). Evolution of base

_composition

in the insulin and insulin-like

_growth

factor genes. Mol. Biol. Evol. 11,875-885.

FENG, P., CATT, K.T., and KNECHT, M. (1988).

Transforming

growth

factor-beta stimulates meiotic maturation of therat_oocyte.

Endocrinology

122,181-186.

FRUNZIO, R„ CHIARIOTTI, L., BROWN, A.L., GRAHAM, D.E., RECHLER,M.M.,andBRUNI,C.B. (1986).Structure and

expres-sion of theratinsulin-like

_growth

factorII(rIGF-II)gene. rIGF-II

RNAs are transcribes from two _promoters. J. Biol. Chem. 261, 17138-17149.

GRAY, A., TAM, A.W., DULL, T.J., HAYFLICK, J., PINTAR, J., CAVENEE, W.K., KOUFOS, A.,andULLRICH,A.(1987).

Tissue-specific

and

_{developmentally}

regulated

_{transcription}

ofthe insulin-like

_growth

_{factor 2 gene. DNA}6,283-295.

HAIG, D., and GRAHAM, C. (1991). Genomic

_imprinting

andthe

strangecaseoftheIGF-II_receptor.Cell64,1045-1046.

HARTMAN, J., DARAM, P., FRIZZELL, R.A., RADO, T., BEÑOS, DJ., andSORSCHER,E.J.(1992).

Affinity purification

of

insolu-blerecombinantfusion

_{proteins containing glutathione-S-transferase.}

Biotechnol

_Bioeng.

39,828-832.

(10)

D.,andADASHI,E.Y.(1990). Ratovarian insulin-like

_growth

fac-tor_{II gene}

_expression

istheca-interstitial cell-exclusive:Hormonal

regulation

and_receptordistribution.

_{Endocrinology}

127,3249-3251.

HOLM, N.R., HANSEN, L.B.H.,NILSSON,C,andGAMMELTOFT,

S. (1994).Gene

_expression

and secretion of insulin-like

_growth

fac-tor-II and insulin-like

_{growthfactor binding protein-2}

from culture sheepchoroid

_{plexus epithelial}

cells. Brain Res. Mol. Res.21,67-74.

HOLTHUIZEN, P., VAN DER LEE, F.M., IKEJIRI, K.,

YA-MAMOTO, M„andSUSSENBACH,J.S.(1990). Identification and initial characterization ofafourth leaderexonand_promoterof the

human IGF-II gene.1087,341-343.

HOOVER, D., FRIEDMANN, M., REEVES, R., andMAGNUSON,

N.S.(1991).Recombinant human

_pim-1

exhibits serine/threonine

lá-ñase

_activity.

J.Biol. Chem.266,14018-14023.

HUMBEL, R.E. (1990). Insulin-like

_growth

factors I and II. Eur J

Biochem.

_190,

445^162.

IKEJRI, K., UENO, T., MATSUGUCHI, T., TAKAHASH1, K., ENDO, H.,andYAMAMOTO,M.(1990).The

_primary

structureof

theratinsulin-likegrowthfactor_{II gene}

_region.

Biochim.

_Biophys.

Acta1049,350-353.

IKEJRI, K., FURUICHI, M., UENO, T., MATSUGUCHI, T., TAKA-HASHI,K.,ENDO,H,and_YAMAMOTO,M. (1991).The

pres-ence andactive

_{transcription}

of three

_independent

leaderexons in

themouseinsulin-like

_growth

factor II gene. Biochim.

Biophys.

Acta 1089,77-82.

JANSEN, M., VAN SCHAIK, F.M., VAN TOL, H, VAN DEN

BRANDE, J.L., and SUSSENBACH, J.S. (1985). Nucleotide

se-quencesofcDNAs

encoding

precursorsofhuman insulin-like

growth

factorII(IGF-II)andanIGF-II variant. FEBSLett. 179,243-246.

JONES, J.I.,andCLEMMONS,D.R.(1995).Insulin-like

_growth

fac-torsand their

_{binding proteins: Biological}

actions. Endocrine Rev.

16,

3-34.

KOLAND, JG., O'BRIEN, K.M., andCERIONE, R.A. (1990).

Ex-pression

of

_{epidermal growth}

factor_receptor_sequenceasE. coli fu-sion

_proteins:

Applicationin the

_study

of

_tyrosine

kinase function. Biochem.

_Biophys.

Res. Commun. 166,90-100.

KWANG, J., BOLINE, S., LITTLEDIKE, E.T., DUBOVI, E.J., and DONIS, R.O.(1991).

Expression

of the

_p80

_region

of bovine viral diarrhea virus and identification of

_specific

antibodies to this

re-combinant

_protein

inbovinesera.Biochem.

_Biophys.

Res. Commun. 178, 1326-1334.

LAUTERIO, T.J., MARSON, L„ DAUGHADAY, W.H.,and_BAILE, C.A._(1987). Evidencefor the role ofIGF-II in the control offood intake.

_Physiol

Behav.40,755-758.

LEE, J.E., PINTAR, J.,andEFSTRATIADIS,A.(1990).Patternofthe insulin-like

_growth

factor II gene

expression during early

mouse

em-bryogenesis. Development

110, 151-159.

LeROITH,D.(1991).Insulin-like GrowthFactors:Molecularand Cel-lularAspects. (CRCPress,BocaRaton,FL).

LOGAN, A., GONZALEZ, A.M., HILL, D.J., BERRY, M.,

GREG-SON, N.A.,andBAIRD,A.(1994).Coordinated_pattern of

expres-sion and localization of insulin-like growth factor-II (IGF-II) and

IGF-binding protein-2

in the adult rat brain.

_{Endocrinology}

135, 2255-2264.

LUTHI, C,ROTH,B.V.,andHUMBEL,R.E.(1992).Mutants of

hu-manIGF-II. Eur. J.Biochem.205,483^190.

MANIATIS, T., FRITSCH, E.F.,andSAMBROOK,J. (1982).

Mole-cularCloning:A

_Laboratory

Manual. Cold

_Spring

Harbor

Labora-tory,Cold

_Spring

_Harbor,NewYork).

OHLSEN, S.M., LUGENBEEL, K.A.,andWONG,E.A.(1994).

Char-acterizationofthe linked ovine insulin and insulin-like

_growth

fac-torII genes. DNA Cell Biol.13,377-388.

RAPPOLEE, D.A., STURM, K.S., BEHRENDTSEN, O., SCHULTZ, G.A., PEDERSEN, R.A.,andWERB,Z.(1992).Insulin-like

_growth

IIactsthroughan

endogenous growth pathway regulated by

imprit-ing

in

_early

mouse

_embryos.

Genes & Dev.6,939-952.

RINDERKNECHT, E., andHUMBEL,R.E.(1978).The amino acid sequenceofhuman insulin-like

growth

factor I and its structural

ho-mology

with

_proinsulin.

J.Biol. Chem. 253,2769-2776.

REINECKE,M., WEIMAR, E., MAAKE, C, DRAKENBERG, K., FALKMER, S.,andSARA,V.R.(1994).IGF-2-like

_peptides

are

pre-sentin insulin cellsoftheelasmobranchian_{endocrine pancreas: An} immunohistochemical and

_{Chromatographie study. Histochemistry}

102,365-371.

ROTWEIN, P.,andHALL,L.J. (1990).Evolution of IGF-II: charac-terization of themouse_{IGF-II gene and}identificationoftwo

pseudo-exons.DNACell Biol.

_9,

725-735.

SANGER, F., NICKLIN, S., andCOULSEN, A.E. (1997). DNA

se-quencing

with

_{chain-terminating}

inhibitors. Proc. Nati. Acad. Sei.

USA74,5463-5467.

SHAMBLOTT, M.J.,andCHEN,T.T.(1992).Identification ofa

sec-ondinsulin-like

_growth

factor inafish

_species.

Proc. Nati. Acad. Sei.

USA89, 8913-8917.

SMITH, D.B.,andJOHNSON,K.S.(1988).

Single-step purification

of

polypeptides expressed

in Escherichia coliasfusions with

glutathione

S-transferase. Gene67,31-40.

SOARES, M.B., ISHII, D.N., and EFSTRATIADIS, A. (1985).

De-velopment

and

_{tissue-specific expression}

ofa

_family

of

_transcripts

relatedto ratinsulin-like

_growth

factor II mRNA. Nucleic Acids Res. 13, 1119-1134.

SOARES, M.B.,TÜRKEN,A., ISHII, D„ MILLS, L., EPISKOPOU, V., COTTER, S., ZEITLIN, S.,andEFSTRATIADIS,A.(1986).Rat insulin-like

_growth

factor II gene: A

single

gene withtwo

promot-ers

_expressing

a

_{multitranscript family.}

J. Mol. Biol.192,737-752.

STEMPIEN, M.M., FONG, N.M., RALL, L.B.,andBELL,G.I.(1986).

Sequence

of a

_placental

cDNA

_encoding

the mouse insulin-like

growth

factor II precursor. DNA.5,357-361.

TEISSIER, M.P., MONGET, P., MONNIAUX, D„andDURAND,P. (1994).

Changes

in insulin-like

_growth

factor-II/mannose-6-phos-phate

receptor

during growth

and atresia of ovineovarian follicles.

BiolReprod.

50, 111-119.

VAN DIJK, M.A., VAN _{SCHAIK, F.M.A., BOOTSMA, H.J.,}

HOLTHUIZEN, P.,andSUSSENBACH,J.S.(1991).Initial charac-terization of the four_promotersofthe human insulin-like

_growth

fac-torII_{gene. Mol. Cell}Endocrinol.81,81-93.

WATSON, A.J., WATSON, P.H., ARCELLANA-PANLILIO, M., WARNES, D., WALKER, S.K., SCHULTZ, G.A., ARMSTRONG, D.T.,andSEAMARK,R.F.(1994).A

_growth

factor

_phenotype

_map forovine

_{preimplantation development.}

Biol.

_Reprod.

50,725-733.

WILLISON,K. (1991).

Opposite imprinting

ofthemouseIGF-II and IGF-II_receptor_{genes. Trends Genet.}7,107-109.

ZHOU, J.,andBONDY,C._{(1993). Anatomy}of the insulin-like

_growth

factor systemin the human testis. FértilSteril.60,897-904.

Address

_reprint

_requests

to:

Dr. Jen-Leih Wu

Institute

of Zoology

Academia Sínica

Nankang, Taipei

Taiwan

11529,

Republic

of

China Received for

_publication

October