• 沒有找到結果。

Expression, Characterization, and Genomic Structure of Carp JAK1 Kinase Gene

N/A
N/A
Protected

Academic year: 2021

Share "Expression, Characterization, and Genomic Structure of Carp JAK1 Kinase Gene"

Copied!
18
0
0

加載中.... (立即查看全文)

全文

(1)

Volume 15,Number 10,1996

Mary

AnnLiebert,Inc.

Pp.

827-844

Expression,

Characterization,

and

Genomic Structure of

Carp

JAK1 Kinase Gene

MAU-SUN

CHANG,1

GEEN-DONG

CHANG,2

JIANN-HORNG

LEU,3

FORE-LIEN

HUANG,1-3

CHEN-KUNG

CHOU,4

CHANG-JEN

HUANG,3

and

TUNG-BIN

LO2-3

ABSTRACT

A

3.7-kb cDNA encodes the carp

JAK1

kinase of

1,156

amino

acid residues. The overall amino acid

sequence

identity

between carp

JAK1

and

murine

JAK1, JAK2, JAK3,

and

human

TYK2

is

57%, 35.5%, 31.3%,

and

42.4%,

respectively.

In

addition,

carp

JAK1

shows

higher

sequence

homology

to

mammalian

JAK1

in

both

the

kinase-like

(JH2)

and kinase

(JH1)

domains

(approximately

70%

identity).

Therefore,

carp

JAK1

is

a

ho-molog

of mammalian

JAK1.

To

investigate

the

possible

function of

JH2

domain,

full-length,

and various

trun-cated

forms of

carp

JAK1

were

produced

in the baculovirus

system.

Our

results demonstrate

that

c-JHl

and

C-JH2

associate with each other and

C-JH2

can be

tyrosine-phosphorylated

by

c-JAKl

and

by

c-JH(l

+

2).

The JAK

I gene was

also isolated from

a

carp

genomic library

and characterized. This

gene

is divided into 24

exons

spanning

atleast

31

kb

of

genomic

DNA. Exon 1 contains

the

5'-untranslated

region

and

exon 2

con-tains the

putative

translation initiation site. The

2.5-kb DNA

region

upstream

of the

transcription

initiation

site contains

numerous

potential binding

sites for

transcription

factors

including

NF-IL6, HNF-5, API,

GHF-5,

and

E2A. When this

DNA

fragment

was

placed

upstream

of the

chloramphenicol acetyltransferase

(CAT)

reporter

gene

and

transfected into

a carp

CF

cell

line,

it could drive the

synthesis

of CAT enzyme 16 times

more

efficiently

than

the

promoterless

pCAT-Basic.

Deletion

analysis

defined

a

positive

regulatory

region

be-tween

—1,023

and

—528.

A smaller

region

(—181

to

+59)

without

any

typical

TATA-box sequences, G

+

C-rich sequences,

or other

binding

sequences

for

known

transcription

factors still

had

promoter

activity.

Constructs without this

region

did

not have detectable

promoter

activity.

This

suggests

that

this

region

of

DNA may

play

an

important

role

in

the

expression

of

carp

JAK

I gene.

INTRODUCTION

of the external

signals

tothe cell.

Upon binding

of the

ligand,

thekinase

activity

of the

receptor

PTK is

induced,

resulting

in

TYROSiNE

PHOSPHORYLATION

catalyzed by protein tyrosine

ki-

phosphorylation

of the

receptor

and cellular substrateson

tyro-nases

(PTKs)

isa

key

step

in

transducing signals

fromex- sine residues

(Ullrich

and

Schlessinger,

1990).

By

contrast, the ternal stimulitothe nucleus.

Therefore,

PTKs

play

very

impor-

nonreceptor PTKs,

duetothe lack of extracellular

domains,

can-tantroles in the

regulation

of cell

proliferation

and differentiation not

directly

interact with extracellular

ligands.

Therefore,

the

(Hunter

and

Cooper,

1985;

Hanks et

al,

1988).

On the basis of

physiological

functions of these

nonreceptor

PTKs remained ob-their structural

similarities,

PTKscanbe divided intotwo

major

scurefor

quite

some time until the observation that

p56lck

was groups:receptorandnonreceptorPTKs.

Receptor

PTKs contain associated with the CD4 and CD8

co-receptors

(Veillette

et

al,

extracellular, transmembrane,

and

cytoplasmic

domains. Theex-

1988).

This

suggests

that

by

binding

nonreceptor PTKs,

some tracellular domainsareabletoassociate with

ligands,

where the transmembranereceptors

lacking

TKdomainscanactina

man-cytoplasmic catalytic

domains are

responsible

for transmission ner

analogous

tothereceptorPTKs.

'Department

of

Zoology,

and2GraduateInstitute of BiochemicalSciences,National Taiwan

University,

Taipei,

Taiwan,

instituteof

Biological Chemistry,

AcademiaSinica,

Taipei,

Taiwan.

4Department

of MedicalResearch,VeteransGeneral

Hospital, Taipei,

Taiwan.

(2)

The

polymerase

chainreaction

(PCR) (Mullís

and

Faloona,

1987)

has been

employed

toclone

potential

PTKs,

using

de-generate

oligonucleotide primers

corresponding

to conserved

peptide fragments

inkinase domains. The JAK

(Janus kinase)

family

wasfirst identified

by

thisstrategy

(Wilks,

1989;

Wilks et

al, 1991).

This

family belongs

tothe

nonreceptor

PTKs and

currently

consists of JAK1

(Wilks

et

al, 1991;

Yang

et

al,

1993),

JAK2

(Harpur

et

al,

1992),

JAK3

(Witfhuhn

et

al,

1994),

and TYK2

(Firmbach-Kraft

et

al,

1990)

in mammals. These kinases lack the Src

homology

domains SH2 and SH3

(Pawson

and

Schlessinger,

1993)

butbear an unusual feature of

having

akinase-like domain. The functional role of TYK2 was first demonstrated

by

its

participation

ininterferon

(IFN)

signal

transduction

pathway

and its association with the IFN-a

receptor

(Velazquez

et

al,

1992).

Recently,

many

cytokine

re-ceptors

have been foundtoassociate with one ortwo or more members of the JAK

family, including

the

receptors

for

ery-thropoietin

(EPO) (Witthuhn

et

al,

1993),

growth

hormone

(Argetsinger

et

al,

1993),

prolactin (Campbell

et

al, 1994;

Da Silva et

al, 1994;

Rui et

al,

1994),

interleukin 2

(IL-2)

(Miyazaki

et

al,

1994;

Russell et

al, 1994;

Witthuhnet

al,

1994),

IL-3

(Silvennoinen

et

al,

1993),

IL-4

(Witthuhn

et

al,

1994),

IL-6

(Stahl

et

al,

1994),

IL-12

(Bacon

et

al,

1995),

on-costatin

M,

leukemia

inhibitory

factor

(LIF)

(Stahl

et

al,

1994),

granulocyte-macrophage colony-stimulating

factor

(GM-CSF)

(Quelle

et

al,

1994),

granulocyte

colony-stimulating

factor

(G-CSF) (Nicholson

et

al,

1994),

and

IFN-y (Muller

et

al, 1993;

Watling

et

al,

1993).

Although lacking

protein

kinase

domains,

the above members of the

cytokine

receptor

superfamily

can

couple ligand binding

with the

tyrosine phosphorylation

of downstream

effectors,

which is mediated

by

association with the JAKs. The activated JAKs then

phosphorylate cytoplasmic

transcription

factors STATsor the

signal

transducer and

acti-vators of

transcription.

After

tyrosine

phosphorylation,

these

transcription

factorsarehomo-or

heterodimerized,

translocated into the

nucleus,

and induce

transcriptional

responses

(Shuai

et

al, 1993;

Darnellet

al, 1994;

Ihleet

al, 1994;

Schindler and

Darnell,

1995).

In

fish,

growth

hormone and

prolactin

from many

species

have been isolated and cloned

(Chang

et

al, 1992a,b;

Bemardi et

al,

1993),

whereas

only

oneIFNfrom flatfish has been

pu-rified and cloned

(Tamai

et

al,

1993).

Basedon a

variety

of studies from

mammals,

the JAK

family

membersareshownto

be involved in the

signal

transduction of

growth

hormone

(Argetsinger

et

al,

1993),

prolactin (Campbell

et

al, 1994;

DaSilvaet

al,

1994),

and interferons

(Velazquez

et

al,

1992).

Therefore,

as aninitial

step

tounderstand the

signal

transduc-tion in

fish,

westartedtoisolate the carp JAK kinase

family

by

PCR. As

reported

in this paper,amember of the JAK

family

wascloned and

designated

carp JAK1 kinase. The carp JAK1 kinase has a

higher

sequence

homology

in both JH1 and JH2 domains

(70%

identity)

tohuman and murine JAK1

(Wilks

et

al, 1991;

Yang

et

al,

1993).

Because the function of the JH2 domain is

unknown,

we

explored

its

activity

inabaculovirus

expression

system,which has been usedtoexpress mammalian

JAK2 kinase

(Quelle

et

al,

1994;

Duhe and

Farrar,

1995).

We made constructs that encode

full-length

JAK1

(c-JAKl),

the JH1 and JH2 domains alone

(c-JHl

and

c-JH2),

and the JH1 and JH2 domains in combination

[c-JH(l

+

2)]

and

expressed

them in the baculovirus system. Coinfection and

immunopre-cipitation experiments

showed thatc-JHl and C-JH2 could as-sociate with each

other,

and that C-JH2 could be

tyrosine-phos-phorylated by

c-JAKlor

c-JH(l

+

2).

Our resultsmay

provide

clues

concerning

thefunction of the JH2 domain.

Studies

by

others showed that

expression

ofmostmembers of the mammalian JAK kinase

family

suchas

JAK1, JAK2,

and

TYK2,

is

ubiquitous

whereas

only

JAK3 is

mainly expressed

in

hematopoietic

cells

(Ihle

and

Kerr,

1995).

As afirst

step

to

explore

the molecular basis

underlying

the

regulation

of carp JAKl gene

expression,

we also cloned and characterized the carp JAKl gene.We have

mapped

the carp JAKl

transcription

start site and characterized its

promoter.

With nested deletion

mutantsof the

5'-flanking

region

fusedtothe

chloramphenicol

acetyltransferase

(CAT)

reporter gene, the promoter

activity

was characterized

by

transfection into afish cell line. To our

knowledge,

this is the first

report

tocharacterize the

genomic

structure and the

promoter

region

of JAK kinases.

MATERIALS AND METHODS

Materials

Insect

Spodoptera

frugiperda

(Sf9)

cells were grown in Grace's medium

containing

10% fetal calf serum

(FCS;

Life

Technologies,

Inc.,

Gaithersburg,

MD).

Monoclonal

anti-phos-photyrosine IgG

clone 4G10 was

purchased

from

Upstate

Biotechnology

Inc.

(Lake

Placid,

NY).

Construction

of

a

random-primed

cDNA

library

Total RNAwasisolated from carp brain

by

the

guanidinium

isothiocyanate

method

(Chomczynski

and

Sacchi,

1987),

fol-lowed

by

the use of a

QuickTrack

mRNA isolation kit

(Stratagene,

La

Jolla,

CA).

The isolated mRNA was used for the

synthesis

of

random-primed

cDNA.

Amplification of

DNA

fragments

by

PCR

Degenerate

primers

(Walks, 1989)

were

designed

tofit the amino acid sequences thatare conserved among PTKs

(Hanks

et

al,

1988; Hanks,

1991).

The amino acid sequences of the

two

opposing primers

are HRDLAAR

(subdomain

VI in

Fig.

2,

below)

and DVWSFG

(subdomain IX).

The first-strand carp brain cDNA was usedas

template.

The conditions for the re-action were94°C for 1

min,

45°C for 1

min,

and 72°C for 2 min in 30

cycles.

cDNA

library screening

An

oligo-(dT)-primed

carp liver cDNA

library

in the lamda ZAP II

(Short

et

al,

1988)

obtained from

Stratagene

was screened

by using

a

PCR-amplified

DNA

fragment

as a

probe

(see Results).

The

probe

was labeled

using

a DIG DNA

Labeling

Kit

(Boehringer

Mannheim, Mannheim,

Germany).

Hybridization

and

washing

werecarried out

according

tothe standard

procedures

(Sambrook

et

al,

1989).

Signals

were de-tected

using

the DIG Luminescent Detection Kit for Nucleic Acids

(Boehringer Mannheim).

Putative

positive plaques

were

purified

through

several rounds of

rescreening

atlower densi-ties and then excisedas

pBluescript plasmids according

tothe manufacturer's instructions. To isolate the

full-length

cDNA,

(3)

FISH TYROSINE KINASE

JAKl

the

300-bp

DNA

fragment

from the5' end of

pJ21 (Fig.

1)

was

amplified

by

PCR, DIG-labeled,

and then usedas a

probe

to

screenanother

random-primed

carp brain cDNA

library

as de-scribed above.

Sequence

analysis

cDNA clonesweresubclonedinto

plasmid pUC18

(Yanisch-Perron et

al,

1985)

and

sequenced by

the

dideoxy

chain-terminationmethod

(Sanger

et

al,

1977)

using

a

Sequenase

kit

(United

States Biochemical

Corp.

Cleveland,

OH).

Several pro-grams from IntelliGenetics

(Mountain

View,

CA)

wereusedto

analyze

the nucleotide sequences.

DNA

constructions

DNA

fragments encoding

the entire carp JAKl were

ampli-fied with

overlap

extension PCR

(Ho

et

al,

1989)

fromJ9 and J21 cDNA

(Fig.

1)

by using

the

following primers: primer

F1.5'-ATATCAGGATCCATGCCAGAACTAGCAGTCATG-3' and

primer

Rl,

5'-GTGGAAACTCAGCTGGCTGAC-3';

primer

F2,

5'-GTCAGCCAGCTGAGTTTCCAC-3'and

primer

R2,

5'

-CAGTAC

AAGÇTTCTC

ATCTTATTTCCAT

AGTTA-3'. The nucleotide sequences of

primers

Rl and F2 are

com-plementary

and

they

are locatedatthe end of clone J9. Other

DNA

fragments encoding

the kinase-like

domain,

kinase

do-main,

and both domainswere

amplified

with the PCR from J21 cDNA

by

using

the

following primers:

c-JH2

(residues

572-875),

5'

-ATATCAGGATCCCAGCTGAGTTTCC

ACC-GCATC3'(F3)

and

5'-CAGTACAAGCTTGAGGAACCTC-TTCTCAAACAC-3'(R3);

c-JHl

(residues 869-1,156),

5'-ATATC

AGGA1ÇCGTGTTTG

AGAAGAGGTTCCTC-3'

(F4)

and 5'-CAGTAC

AAGÇTTCTC

ATCTTATTTCCAT

AGTTA-3'(R2); c-JH(l

+

2) (residues 572-1,156),

5'-ATATCA-GGATCCCAGCTGAGTTTCCACCGCATC-3'

(F3) and

5'-CAGTACAAGCTTCTCATCTTATTTCCATAGTTA-3'(R2).

Each

primer

encodesa

unique

restrictionsite

(underline) (Bam

HI, GGATCC;

and Hind

III,

AAGCTT).

The PCR

products

bases o carp JAKl

s'-f

—n 3500 3730

K

J9 J21 HI

FIG.1. Restriction maps of carp JAKl cDNA clones. The ki-nase-like

(JH2)

and kinase

(JH1)

domains are

represented

by

the

horizontally striped

and black

boxes,

respectively.

Restriction enzyme sites are shown above carp JAKl cDNA. The

poly

(A)

tail is

represented by

AAA.The cDNA

clones,

Jl 1 and

J21,

were isolated from acarp liver

library

whereas the clone J9 wasisolated fromacarp brain

library.

werethen restricted with Bam HI and HindIIIand

ligated

into

pQE30

(QIAGEN

Inc., Chatsworth,

CA).

DNAsequence

analy-sis was

performed

to confirm the accuracy of the PCR

prod-ucts.

Expression

in

Escherichia coli

All the

resulting plasmids (pQEJAKl, pQEJH(l

+

2),

pQEJHl,

and

pQEJH2)

were transformed into E. coli strain JM109. Thetransformantsweregrown

overnight

at37°C in LB broth and diluted 1:10 into fresh medium. After incubation at

37°C for1

hr,

isopropyl-/3-D-thiogalactopyranoside

(IPTG)

was added to a final concentration of 1

mM,

and incubation was continued for 3 hr. Cellswerecollected

by centrifugation.

Due

toall the recombinant

proteins

containastretchof six

histidines,

therefore these

proteins

canbe

purified by using

Ni2+-nitrilo-triacetate-agarose (Ni-agarose) (QIAGEN

Inc.)

according

tothe

procedures

asdescribed

(Huang

et

al,

1995).

Antibodies

Purifiedrecombinant

protein

c-JHl orc-JH2wasdissolved in

phosphate

buffersaline

(PBS)

and mixed

thoroughly

withan

equal

volume of Freund's

complete adjuvant

for the first

in-jection

of Freund's

incomplete

adjuvant

for the second and third

injection. Approximately

100 pg of recombinant

protein,

for each

injection,

was

subcutaneously injected

into the back of

guinea pig.

Expression

in

the

baculovirus

system

All of the

pQE30

constructs described in the E. coli

ex-pression

systemwere

digested

with EcoRIand XbaIand then

ligated

into the transfer vector PVL1393

(PharMingen,

San

Diego,

CA).

Plasmids

(0.5

pg of

each)

were co-transfected into Sf9 cells with 0.1 pg of

"Baculogold"

virus DNA

(PharMingen),

whichcontainedalethaldeletion.The

parental

baculovirus is unabletoinfect insect cells and

only

recombi-nant baculovirus

containing

the

plasmid

constructcan

repli-catein Sf9 cells.

Sf9 cells were infected with various baculovirusat a

multi-plicity

of infection

(moi)

of 10

plaque-forming

units/cell.

After 60

hr,

cells were

harvested, washed,

and

lysed

inabuffer

con-taining

150 mM

NaCl,

1% Triton

X-100,

and50 mM Tris

pH

7.5. The Triton-insoluble cell

pellets

weresolubilized inabuffer

containing

1.5%

/V-lauroylsarcosine,

25 mM

triethanolamine,

and 1 mM EDTA

pH

8.0

(Frankel

et

al,

1991).

Both Triton-soluble and -insoluble fractions were incubated with the

Ni-agarose.

After extensive

washing

with 0.5 M

NaCl,

proteins

wereeluted

by

0.1 M EDTA

pH

8.0. An

equivalent

amountof

protein

was

subjected

to

NaDodS04-polyacrylamide

gel electrophoresis

(PAGE),

blottedtonitrocellulose paper, and

probed

with differ-entantibodies.

Western

blot

The methodofTricine

NaDodSGj-PAGE (Schagger

andvon

Jagow, 1987)

wasused. The

gel

concentrationwas7.5%. After

NaDodS04-PAGE,

the

proteins

were

analyzed

by

immunoblot

using

anti-PYmAbor

polyclonal

antibodies

specific

for JH1 or JH2 domain

according

tothe

procedures

asdescribed

(Huang

et

(4)

Immunoprecipitation

Sf9 cellswerecoinfected with vAcJH2 and vAcJHl. After 60

hr,

cellswerecollectedand solubilized

by

resuspension

and incubation in the

immunoprecipitation

buffer

(50

mM Tris

pH

7.4,

150 mM

NaCl,

1 mM

Na3V04

and 1%

NP-40).

Immunoprecipitation

was

performed

by

the addition ofanti-JHl oranti-JH2 antiserumatadilution of 1:100orwith

preimmune

serum as a control reaction. The mixture

(200

pi)

was incu-batedat4°C for 2 hrfollowed

by

anincubation with 50

pi

of 50

mg/ml slurry

of Protein

A-Sepharose

(Pharmacia,

Uppsala,

Sweden)

for 1 hr. The

Sepharose

was collected and washed three times with the

immunoprecipitation

buffer andtwotimes with 20 mM Tris

pH

7.4,

1 mM

EDTA/1

mMEDTA.

Samples

werethen boiled ina0.5%

NaDodS04

sample

bufferto

disrupt

the

complexes.

Productswere

subjected

to

NaDodS04-PAGE

and

analyzed by

immunoblot

using

anti-PY mAbor

polyclonal

antibodies

specific

forthe JH1 orJH2 domain.

Reverse

transcription

and

PCR

mRNA

(0.5 pg)

isolated from carp brain and liver tissues wasincubatedat65°C for 5 min inabuffer

containing

50 mM

Tris-HCl,

75 mM

KC1,

3 mM

MgCl2,

10 mM

dithiothreitol,

2 units of RNasin and 1.25 mM

deoxynucleotide

triphosphatates

(dGTP,

dATP, dTTP,

and

dCTP),

and then

kept

onice. Two hundred units of

Superscript Moloney

murine leukemia virus reverse

transcriptase

(RT;

GIBCO

BRL,

Gaithersburg,

MD)

and

oligo(dT)i2_i8 primer

(2 pg)

and random hexamer

primers

(2

pg)

wereadded and incubatedat 37°C for 1 hr. Thereaction was then

stopped by

incubationat95°C for 5 min.

Fivesets of

specific primers,

Fl/Rl, F1/R2, F3/R3, F4/R2,

and

F3/R2

wereused and their sequencesweredescribed in the DNA Construction section. PCRwasthen carriedout

by

addi-tion of the heat-treated RT

mixture,

PCR

buffer,

and

Taq

poly-merase

(Promega,

Madison, WI).

Only

the

amplified

DNA

frag-ments

by

F3/R3

and

F4/R2

from brain tissues and PCR

product

by

Fl/Rl

from liver tissueswereisolated fromanagarose

gel

and subcloned into the

pGEM-T

vector

(Promega).

Other

ana-lytical

methods were the same as described in the section of

Sequence Analysis.

Isolation

of

carp

JAKl

genomic

clones

A

commercially

available carp liver lambdaFIX II

genomic

library (Stratagene)

was usedto isolate 15-to 18-kb

genomic

DNAclones

containing

the gene that encodes carp JAKl.

Full-length

carp JAKl cDNA was labeled

using

a DIG DNA

Labeling

Kit

(Boehringer

Mannheim).

Approximately

1 X

106

independent

cloneswere

plated

ata

density

of 5 X

104

plaque

forming

units/150-mm

Petri dish.

Hybridization

and

washing

werecarriedoutasdescribed above.

Signals

weredetected

us-ing

the DIG Luminescent Detection Kit for Nucleic Acids

(Boehringer

Mannheim).

Putative

positive plaques

werefurther

purified through

several rounds of

rescreening

atlower densi-ties. Four distinct carp

genomic

DNAclones

(designated

Jl, J2,

J3,

and

J4)

were

isolated,

and DNAwas

prepared

from each clone. These carp

genomic

clones were

digested

with various restriction endonucleases. The DNA

fragments

were fraction-ated

by electrophoresis

in 1% agarose

gels

and transferredto

nylon

filters

(Sartorius

AG,

Goettingen, Germany).

The

result-ing

DNAblotswere

probed

under theabove conditions of

high

stringency

with the DIG-labeled whole carp JAKl cDNA.

Sequence analysis

of

genomic

clones that encode

carp JAKl

Phage

DNA was

digested

with Not

I,

Sac

I, 5a/1,

and Xba

I and subcloned into either the

pUC18

or

pBluescript

vector.

Each subclone was

sequenced by

the

dideoxy

chain-termina-tion method

(Sanger

et

al,

1977)

using

the

Sequenase

proto-cols from United States Biochemical. Thenucleotide sequences of each DNA

fragment

were determined

using

T3 and T7

primers

from both ends. Additional nucleotide sequenceswere determined

using

different sets of 20-nucleotide

primers

syn-thesized basedonthe known sequences from both ends of each DNA

fragment.

The sizes of the intronsweredetermined either

by

restriction enzyme

mapping

of

genomic

clones or

by

PCR

using

sense and antisense

specific

primers

that are from two

differentexons.The

genomic

DNAsequencewas

analyzed

with the Genetics

Computer Group

software program. The

tran-scription

factor

recognition

site data bases

(releases

7.3 and

6.5)

wereusedto

identify potential transcription

factor motifs within the

5'-flanking region

of the carp JAKl gene.

Primer extension

analysis

An antisense

oligonucleotide

(5'-GlTT

AATTTCGATCAT-TACAGACGTTTG-3')

complementary

to the 5' end of carp JAKl cDNA clone J9

(Fig.

1)

was labeledatthe 5'end with

["y-32P]ATP

by

T4

polynucleotide

kinase and

purified

on a

Stratagene

NucTrap probe purification

column. The labeled

primer

was annealedto5pg of

poly(A)+RNA

prepared

from adult carp liver totalRNAand extendes with

Moloney

murine leukemia virus

(M-MLV)

reverse

transcriptase

(GIBCO BRL)

asdescribed

previously

(Sambrook

et

al,

1989).

The extended

products

were

analyzed

on a5%

polyacrylamide,

7Murea

se-quencing

gel.

The size of the extended

products

was inferred froma

sequencing

ladder of the JAKl gene obtained from the same

primer

used for

primer

extension.

Plasmid construction

for

CAT

functional

analysis

A 3-kbSal I

fragment encompassing

for

5'-flanking

region

was subcloned into

pBluescript

and various deletion mutants

were

generated

from this clone

by

PCR

using specific primer

FIG. 2. Amino acid sequence

alignment

of the carp,mouse

(Yang

et

al,

1993)

and human

(Wilks

et

al,

1991)

JAKl

proteins.

Alignments

were

initially

made

by

computer

analysis

andwere

subsequently aligned by inspection. Gaps

areintroducedto

op-timize

alignment

and shownas dots. The identical residuesare

represented by

dashes. The

putative

kinase domains are delin-eated witharrows. Subdomains Ia-XIa of the kinase-line domain and subdomains I-XI of the kinase domain aredenoted

ac-cording

to the

previous

reports

(Hanks

et

al, 1988; Hanks,

1991).

The sequence of carp JAKl has been

deposited

in the

(5)

JAKl

cJAKl MPELAVMELGRQLCGKMKKQRKAEMTVLTVMKGLEIHFYLPDTHQLEYFKDCHTAEDL 58 mJAKl MQYLNIKEDCNAMAF-A--RSFK-T-VKQWPEP-V-VT-L-REP-RLGSGEY-E- 60 hJAKl MQYLNIKEDCNAMAF-A--RSSK-T-VNLEAPEP-V-VI-S-REP-RLGSGEY-E- 6 0 cJAKl CVEGCQEDATSHLCADNLFALSEESQDLWYAPNHAFKITEETSIKLHYRMRFYFTNWHAP 118 mJAKl -IRAA-ECSI-P--H.-YD--TK-RIITVDDK--LR-GT 119 hJAKl -IRAA-ACRI -P- -H.-YD-NTK-RTITVDDKM-LR-GT 119 cJAKl VRTESPVWRHSLFKHKGVSVSPKGPEGTPLLDAASLEYLFAQGQYDFLRGLAPVRAPQNE 178 mJAKl NDN-QS-PK-Q-NGYEKKRV-EA-S-LIKC-I-D-KT- 17 9 hJAKl NDN-QS-PK-Q-NGYEKKKI-DA-S-LVKC-I-D-KT- 179 cJAKl AEKHEIENECLGMAVLAITHHAKSNDLPLSGVGAETSYKRFIPDSLNRTIKQRNFSHVYV 23 8 mJAKl QDG-D-S-MMKKMQ-PELPKDI-Y--ET--KS-R-LLTRMR 23 9

hJAKl QDG-D-S-MMKKMQ-PELPKDI-Y--ET- -KS-R-LLTRMR 23 9

cJAKl YNNVFKNFLNEFNSKTIQDSNITLYDLKVKYLSTLETLTQGLGRETIEPKILKVSGESDG 2 98

mJAKl I-D--K-N-C--SVSTH-A-KHY-A-IF-TSM-LI-S-NEL 2 99

hJAKl I-D--K-N-C--SVSTH-A-KHY-A-IF-TSM-LI -S-NEM 299

cJAKl SPALTLPSGDDG.LGYEVQVSGTTGISWRRKPVPNILIVKDKTKSKKNKADKQSKKEMTK 357

mJAKl -RCHSND--NVL.YEVM-TGNLGIQWRQKPNV-PVEKEKNKLKRK-LEYNKHKKDD-RN- 358

hJAKl NWFHSNDG-NVLYYEVM-TGNLGIQWRHKPNV-SVEKEKNKLKRK-LEYNKDKKDE-KN- 3 5 9

cJAKl RKTVMTIFSDFFEITHIVIKESCATIYSQDNKTMELDLFYRDAALSFAALVDGYFRLTVD 417

mJAKl LREEWNN--Y-P-WS-NK--N-N-K-SSHEE-VS-A- 418

hJAKl IREEWNN--Y-P-WS-NK--K-N-K-SSHEE-VS-A- 419 cJAKl AHHYLCTEVAPSSWQNLENGCHGPICTEYAIHKLRQEGNEEGTYVLRWSCTDYNYIIMT 4 77 mJAKl -D-PLI-H-IQ-N-S-M-FDN-L-- 4 78 hJAKl -D-PLI-H-IQ-N-S-M-FDN-L-- 4 79 CJAKl WCIEMDLCESRPVPQYKNFQIETSPQGYRLYGTDTFRPTLKELLEHLQGQNLRTENLRF 53 7 mJAKl -T-F-KSE.VLGGQK-F-VQKGR-S-H-SMDHF-S-RD-MN--KK-I-D-IS- 53 7 hJAKl -T-F-KSEQVQGAQK-F-VQKGR-S-H-S-RSF-S-GD-MS--KK-I-D-IS- 53 9 »*** ja **** CJAKl QPVLVGLGQPRKISNLLVMTRDREPDSQRQPQVSQLSFHRILKEEIVQGEHLGRGTRTNI 5 97

mJAKl VLKRCCQPK--E-A-KKAQEW.-PVYSM-D-KDLI-H- 596

hJAKl MLKRCCQPK- -E-A-KKAQEW

.

-PVYPM-D-KDLI-H- 598

**IIa** *** Ilia *** ***j_va***

cJAKl YSGVLKLKSEDDDDMGGYSQEVKVILKVLGSGHRDISLAFFETASMMRQISHKHIALLYG 6 57 mJAKl -T-LDYKDEEGIAEEKK. . I-DPS-A-V-VY- 6 54 hJAKl -Y-MDYKD-EGTSEEKK. . I-DPS-A-V-VY- 6 56 ******Va****** *****yj;a*****+ cJAKl VCVRHQENIMVEEFVQYGPLDLFMRRQSIPLSTAWKFQVAKQLAGALSYLEDKKLVHGNV 717 mJAKl -DV-EG-H-K-DA-T-P-K-S-D- 714 hJAKl -DV-EG-H-K-DV-T-P-K-S-D- 716 ++**Viia**** * *VIIIa** *** cJAKl CSKNILVARDGLDGEGGPFIKLSDPGIPITVLSREECVDRLPWIAPECVQDTANLSIAAD 777

mJAKl -T--L-L--E-I-SDI-VS--T-Q--IE-I-E-SK-V- 774

hJAKl -T--L-L--E-I-S-C-IT-Q--IE-I-E-SK---V--- 776 *j_Xa**** ***** xa *** ***xia*** cJAKl KWGFGTTLWEICYNGEIPLKDKKLTEKERFYAAQCQLASPDCEELAKLMTHCMTYDPRQR 837 mJAKl --S-T-I-ESR-RPVT-S-K-D-R--N-N-- 834 hJAKl --S-T-I-ESR-RPVT-S-K-D-R--N-N-- 836 < i i to ********** j **** CJAKl LFFRAIVRDIDMVEKQNPSIQP...VPMLEVDPTVFEKRFLKKIRDLGEGHFGKVELCRY 8 94 mJAKl P-M-NKL-E-D-VS.EKQPTT-H-R- 8 93 hJAKl P-M-NKL-E-D-VSRKKNQPT-H-R- 896

** ***ij**** **III* * **IV**** ***

cJAKl DPRGDRTGELVAVKSLKPENREEQSSNLWREIHILRELYHENIVKYKGIWHEEGGRSIKL 954 mJAKl --E--N-Q-SGGNHIAD-KK--E-N-CM-D--NG- 953 hJAKl --E--N-Q-SGGNHIAD-KK--E-N-CM-D--NG- 956 ***** v ******* *** * ** VI ********* CJAKl IMEFLPAGSLKEYLPRNKAHIDLKTLLNYAVQICQGMDLLASRNYIHRDLAARNVLVENE 1014 mJAKl -S-K--NK-N--QQ-K--I-K-Y-G--Q-V-S- 1013 hJAKl -S-K--NK-N--QQ-K--V-K-Y-G--Q-V-S- 1016 ***YII**** ***YIII***** ***jx***** CJAKl NTVKIGDFGLTKSIKDNEGYYTVKDDLDSPVFWYAPECLIHCKFYRASDVWSFGVTMYEL 1074 mJAKl HQ-A-ETDKE-R-Q-I-LH-- 1073 hJAKl HQAETDKERMQSILH -1076 * *x**** **xi** CJAKl LTYCDISCSPMSVFLTMIGPTHGQMTVTRLVKVLEEGKRLPKPDGCSDRLYCLMRRCWEA 1134 mJAKl -SDF-AL--K-T-K-C-PN-P-EV-Q-K-F 113 3 hJAKl -SDS-AL--K--.NT-K-C-PN-P-EV-Q-K-F 113 6 cJAKl TPEKRIDFKGLIANFQQMIDNQ 1156 mJAKl Q-SN-TT-QN--EG-EALLK 1153 hJAKl Q-SN-TS-QN--EG-EALLK 1156

(6)

582 JH2 847 875 JH1 1156 c-JAKl

C-JH2

c-JHl

c-JH(l +2)

FIG. 3. Schematic

diagrams

of

full-length

and the various truncated forms of carp JAKl constructs. Thekinase-like and kinase domainsare

represented by horizontally striped

and dark

box,

respectively.

Truncated moleculeswereconstructedas de-scribed in thetext.

normalize the

samples

for differencesintransfection

efficiency.

CAT and

/3-Gal

activities in the extracts were measured

ac-cording

tothe

previous procedures

(Herbomel

et

al,

1984).

The

acetylated

products

of the CAT assaywere

separated by

thin-layer chromatography developed

with chloroform-methanol

(95:5, vol/vol),

visualized

by autoradiography

and

quantitated

by using

a

Phospholmager (Bio-Imaging

Analyzer

BAS

2000,

Fuji, Japan).

RESULTS

Isolation

of

carp

JAKl cDNA

We

amplified

carp brain cDNA

by

PCR

using

degenerate

primers,

an

approach

showntobeeffective in

isolating

human and murine JAKl gene

(Wilks

et

al, 1991;

Yang

et

al,

1993).

sets. PCR

products

were then cloned into

polylinker regions

of the

reporter

vector

pCAT-Basic (Promega).

Clone

pJPl

contains the

flanking region

fragment

JP1,

nucleotides

—2,541

to +59. JP1 was

synthesized using

PCR

primers

5'

-TTG

AAGÇTTTCCTCCTAGG

ATC AGG-C AG A-3'

(nucleotides

-2,541

to

-2,522)

and 5'-GGGTCTAGAT-GCTTCAGTCGT-CATGATCAA-3'

(nucleotides

+39 to

+59)

onthe 3-kb Sal I

fragment.

The

oligonucleotide primers

have additional sequences of Hind III and Xba I sites ateach

end,

respectively.

Fragment

JP1 was

purified

from the

gel,

di-gested

with Hind III and Xba

I,

and then cloned intothesame sites of

pCAT-Basic.

Clone

spJP2, pJP3, pJP4, pJP5,

and

pJP6

were

similarly

constructed, except

that the JP1

fragment

was

replaced

with JP2

(nucleotides

-2,541

to

-181),

JP3

(nu-cleotides

-2,541

to

-901),

JP4

(nucleotides -1,023

to

+59),

JP5

(nucleotides

-528to

+59),

and JP6

(nucleotides

-181 to

+59),

respectively.

The

pRSV-CAT

(Gorman

et

al,

1983)

was usedas a

positive

control.

Transfection,

CAT,

and

ß-galactosidase

assay

Carp

fin

epitheloid

cells,

CF

(Chen

and

Kou,

1986),

were maintained in Leibovitz's L-15 medium

supplemented

with 10% fetal calfserum

(FCS)

at27°C. Subconfluent cultures

(ap-proximately

60%

confluent,

24 hr after

plating)

in60-mm

cul-turedishes werewashed twice with Leibovitz's L-15

medium,

and incubated with

DNA-Lipofectamine complexes containing

20 pg of the different CATconstructs

together

with 5 pg of

pSV-/3-galactosidase (ß-GAP)

vector

(Promega)

in

duplicate.

Transfection was carried out for 5

hr,

and cellswere washed with fresh Leibovitz's L-15 medium and fed with the same medium

supplemented

with 10% FCS. After 40

hr,

cellswere

harvested,

washed in

phosphate-buffered

saline

(PBS),

and

lysed

with 25 mMTris

phosphate pH

7.8,

containing

2 mM

dithiothreitol,

2 mM

EDTA,

10%

glycerol,

and 1% Triton X-100 at room

temperature

for30 min. The total

lysates

were

scraped

fromthe dish and transferredto

microcentrifuge

tubes. Cell debriswasremoved

by centrifugation

and theextractswere frozenat

70°C untiluse.Protein concentrationwasmeasured

by

the Bio-Rad

protein

concentration

quick-assay

method

(Bio-Rad, Richmond,

CA).

The

pSV-ß-Gal

vector

(Promega)

carry-ing

the SV40

promoter

linked tothe

/3-Gal

gene wasused to

¿w»

kDa M 1 94 75 45 28 22 17

ML

— C-JH1/JH2

#vvs

M c-JHl

B

kDa 94 75 45 28 22 17 anti-PY

FIG. 4. Demonstration of

tyrosine

kinase

activity

of c-JHl in E. coli. Strain JM109 cellsweretransformed with

pQE30

(lane

1),

pQEJH2

(lane 2),

and

pQEJHl

(lane 3),

respectively.

After induction with 1 mM IPTG for 3

hr,

proteins

were

prepared

and

separated by

NaDodS04-PAGE

and stained with Coomassie blue

(A)

orimmunoblotted with anti-PY mAb

(B).

Positions of c-JHl and c-JH2 are shown on the

right.

Lane

M,

Prestained molecularmassmarkers.

(7)

FISH TYROSINE KINASE

JAKl

Amplified fragments

ofabout220

bp

were

purified

and

ligated

into

pGEM-T (Promega). Among

the 20clones

sequenced,

15 clones contained a

single

JAK-related sequence.

Using

this DNA

fragment

as a

probe,

we screened an

oligo(dT)-primed

carp liver cDNA

library.

Schematics of thetwo

representative

carp liver JAKl clonesobtainedfrom thisscreen areshown in

Fig.

1. On the basis ofDNAsequence

analysis,

thesetwo

clones,

pJl

1 and

pJ21,

areidenticalexceptthat

pJl

1 is0.7 kb whereas

pJ21

is 2.1 kb in

length.

It was found that

pJ21

encodes the carp

homolog

of mammalian

JAKl,

containing

residues from

465 to

1,156.

To obtain the

full-length

cDNA

clone,

a

300-bp

DNA

fragment

from the 5' end of

pJ21

wasusedas a

probe

to

screen acarp brain cDNA

library.

This

screening yielded pJ9

(1.9

kb),

asshown in

Fig.

1.

Therefore,

the

full-length

carp

JAKl cDNA sequencesarederived from the two

overlapping

clones,

pJ9

and

pJ21,

and containa5'-untranslated

region

of 186

nucleotides,

a

coding

region

of

3,468 nucleotides,

anda

3'-untranslated

region

of65nucleotides. The total

length

of carp JAKl cDNA is

3,719

nucleotides.

(The

sequence has been

de-posited

in GenBank withanaccession number

L24895.)

Structure and the

activity

of

carp

JAKl cDNA

The

complete

sequenceof carp JAKl cDNA encodesanopen

reading

frame of

1,156

amino acid residues witha

predicted

mol-ecularmassof 129 kD and the

protein

wastermed carp JAKl.

Allmembers of the JAK

family

haveseven

homologous

domains in the molecule that have been namedasJHsorJAK

homology

domains. JH1 is a

carboxy-terminal kinase-catalytic

domain whereas JH2 isakinase-like

domain;

the other five JHsare

pre-sentin the farammo-terminal

part.

Aminoacidsequence

com-parison

of carp JAKl with human and murine JAKl

(Wilks

et

al, 1991;

Yang

et

al,

1993)

is shown in

Fig.

2. There isa

higher

sequence

homology

in both JH1 and JH2

(70%

identity).

The overall sequence

identity

between carp and

human/murine

JAKl is about 57%. When the deduced amino acid sequences of carp JAKl were

compared

with those of murine JAK2

(Harpur

et

al,

1992),

murine JAK3

(Witthuhn

et

al,

1994),

and human TYK2

(Firmbach-Kraft

et

al,

1990),

itwasfound that the sequence

ho-mology

is lower in JH1

(50,

46,

and 56%

identity, respectively)

andJH2

(45,

43,

and51%

identity, respectively).

The overall se-quence

identity

between carp JAKl and murine

JAK2,

murine

JAK3,

and humanTYK2is

35%, 31%,

and

42%,

respectively.

Totestwhether the JH1orJH2 domains possess

kinase-cat-alytic activity,

we

generated His-tag

fusion

proteins

of the JH2 domain

(c-JH2)

and theJH1 domain

(c-JHl)

ofcarp JAKl

(Fig.

3),

and

expressed

thesefusion

proteins

in E.coli

(Fig.

4A).

The

tyrosine

kinase

activity resulting

from each fusion

protein

can be detected

by anti-phosphotyrosine

monoclonal antibodies

(anti-PY mAb).

Duetothe lack of

endogenous tyrosine

kinases in E.

coli,

there is littleornocross-reactive

background

for the anti-PY Western blot

(Fig.

4B,

lane

1).

Among

the

expressed

fusion

proteins (Fig.

4A),

only

c-JHl

displayed tyrosine

kinase

activity

and several

tyrosine-phosphorylated

bands

including

a

band

corresponding

toc-JHl weredetected

(Fig.

4B,

lane

3).

By

contrast, C-JH2

(lane

2)

didnotexpress any observable

ty-rosinekinase

activity.

These resultsareconsistentwithan

ear-lier

report

(Wilks

et

al, 1991)

andsuggestthat

only

JH1 isa

functionally

active kinase domain. c-JHl and C-JH2were fur-ther

purified by Ni-agarose affinity chromatography

and used

togenerate

polyclonal

antibodies

(see

below).

Full-length

and various

truncated

forms of

carp

JAKl

are

produced

in insect

cells

using

the

baculovirus

system

Figure

5,

A and

B,

show that

c-JAKl,

c-JH(l

+

2),

c-JHl,

and C-JH2were

present

inboth the soluble and

Triton-A kDa 94 75 45 . 28 22 . 17 B kDa

VN^?V

„(*»»

sS» vN°'

JB^WW*1

10

„KcVO^

( anti-JH2 S&

„ce«»

3ft0+ï>

..mJ**1

-« c-JAKl ., c-JHO+2) -« pp60 -. c-JHl c kDa anti-JHl vf* 10 — c-JAKl — c-JH(l+2) -. pp60 — c-JHl anti-PY

FIG.5.

Expression

of

full-length

and various truncated forms

ofcarp JAKl in thebaculovirus system.Sf9 cellswereinfected

with

wild-type

baculovirus,

AcMNPV

(lanes

1 and

2)

and a

variety

of recombinantbaculovirus

(lanes 3-10),

respectively.

After 60

hr,

cells wereharvested and

separated

into Triton-soluble

(lanes

1, 3, 5, 7,

and

9)

andTriton-insoluble fractions

(lanes

2, 4, 6, 8,

and

10).

Both fractionswerefurther boundto

Ni-agarose gels

asdescribedin thetextand

equivalent

amounts

ofbound

patients

were

separated

by NaDodS04-PAGE,

trans-ferredto

nitrocellulose,

and immunoblotted with

polyclonal

an-tibodies

specific

for JH2 domain

(anti-JH2;

A),

for JH1 domain

(anti-JHl; B),

and a mAb for

phosphotyrosine

(anti-PY; C).

Positions of

prestained

molecularmassstandardsand their

(8)

insoluble fractionsas revealed

by

antibodies

specific

for JH2

(Fig.

5A)

and for JH1

(Fig.

5B).

Again,

the anti-PYmAbswere usedtodetect the

tyrosine

kinase

activity

of eachfusion

pro-tein. As a

control,

Sf9 cells were infected

by

wild-type

bac-ulovirus and there was little orno cross-reactive

background

for the anti-PY Western blot

(Fig.

5C,

lanes1 and

2).

As shown in

Fig.

5C

(lanes

3 and

4),

c-JH2 didnot

display tyrosine

ki-nase

activity.

Other fusion

proteins,

c-JAKl,

c-JH(l

+

2),

and c-JHl all

expressed tyrosine

kinase

activity (Fig.

5C,

lanes

5-10)

and

autophosphorylation by

themselves seemedtooccur as we

compared

their

electrophoretic

mobilities in

Fig.

5,

A and

B,

and the

corresponding

mobilities in the anti-PY Western blot

(Fig.

5C).

Interestingly,

in cells

expressing

c-JAKl and

c-JH(1

+

2),

anadditional

protein

of60 kDwasdetected

by

anti-JHl antibodies and anti-PY mAb

(Fig.

5B,

lanes

7-10,

and

Fig.

5C,

lanes

7-10).

Moreover,

there were more

tyrosine-phos-phorylated proteins

bound to

Ni-agarose

in the Triton-insolu-ble fraction of insect cells

expressing

c-JAKl and

c-JH(l

+

2)

(Fig.

5C,

lanes 8 and

10)

than cells

expressing

c-JHl. The pro-teins

analyzed

werefractions boundto

Ni-agarose, presumably

viathe

His-tag

of recombinant

proteins.

Therefore,

they

were either

cleavage products

of JAKl related

proteins

orassociated

proteins.

c-JHl and C-JH2

interact

with each other

To demonstrate the association of c-JHl and

c-JH2,

anti-JHl

antibody

was used to

precipitate

c-JHl from the extract of vAcJHl- and vAcJH2-co-infected

cells,

and the associated c-JH2wasdetected

by blotting

withanti-JH2

antibody.

As shown in

Fig.

6A,

C-JH2 is

co-precipitated by

c-JHl.

Similarly,

c-JHl

is also

co-precipitated

by

c-JH2

(Fig.

6B).

Because both

pro-teins are

highly expressed

under these

conditions,

there is the

possibility

that this interaction is somehow

nonspecific.

Therefore,

another recombinant baculovirus

vAcCAT,

which carries the

chloramphenicol acetyltransferase

(CAT)

gene un-der the control of the same

polyhderin

promoter,

was usedto coinfect with vAcJHl or vAcJH2. The

immunoprecipitation

data showed that neither anti-JHl noranti-JH2

antibody

was ableto

precipitate

CAT

protein

from thecoinfected cellextract

(data

not

shown).

Moreover,

neither anti-JH2noranti-JHl

an-tibody

wasableto

precipitate

c-JHlorC-JH2 from theextracts

from cellsinfectedwith vAcJHl orvAcJH2 aloneasshown in

Fig.

6

(lane 1).

Therefore,

C-JH2seemstointeract

specifically

with c-JHl and

possibly by

tyrosine-phosphorylated by

c-JHl

(see

texts

below).

JH2 domain

is

tyrosine-phosphorylated

by

c-JAKl and

c-JH(l +2)

Because c-JHl and C-JH2were associated with each other

(Fig.

6),

we

sought

to

investigate

whether c-JH2 isasubstrate forc-JH1

by

co-infection

experiments

in which insect cellswere co-infected with vAcJH2 and vAcJAKl or vAcJH2 and

vAcJH(l

+

2),

instead of co-infection with vAcJH2 and vAcJHl. This avoided

ambiguity

indata

interpretation

due to

the similar

mobility

of c-JHl andC-JH2on

NaDodS04-PAGE.

The

experimental procedures

werethesame asthose described in the vAcJHl and vAcJH2 co-infection

experiment.

Proteins boundto

Ni-agarose gels

were

analyzed by blotting

with either anti-JH2

antibody

oranti-PY mAb. The control

experiment

was carried out

by infecting

insect cells with eithervAcJAKl or

##

B

kDa

94

-75 -45

-28

-22 -17

-IP

F

& &

1 2 3

#^

ir jo w *

*

Blot

:

anti-JH2

kDa

94

-75 -45 -28 -22 -17

-IP:

9

Cv ¡y » « S

S¡ S

Blot

:

anti-JHl

FIG. 6. Association ofc-JHl with C-JH2. Sf9cells wereinfectedwith vAcJH2 alone

(a,

lane

1),

vAcJHl alone

(B,

lane

1),

and co-infected with vAcJHl and vAcJH2

(lanes

2and

3),

respectively.

TheTriton-soluble fractionswere

immunoprecipitated

with normal

guinea pig

antiserum

(pre-imm;

Iane2),

anti-JHl antibodies

(A,

lanes 1 and

3)

and anti-JH2 antibodies

(B,

lanes 1 and

3).

The associated

proteins

were

analyzed by

NaDodS04-PAGE,

transferredto

nitrocellulose,

and immunoblotted with anti-JFK

(A)

oranti-JHl antibodies

(B).

Arrowheadindicates

positions

of c-JHl and

C-JH2,

respectively.

(9)

JAKl

vAcJH(l

+

2)

alone. As shown in

Fig.

7B,

C-JH2indeedwas

tyrosine-phosphorylated

by

c-JAKl

(lane 2)

and

by

c-JH(l

+

2)

(lane 4).

This suggests that

transphosphorylation

of C-JH2

by

c-JHl mayoccurwhen

they

are

expressed

in

high

levels

to-gether.

RT-PCR

of

brain and liver mRNA

Asdescribed

above,

two

overlapping

clones,

pJ9

and

pJ21,

wereisolated from different tissues. To

investigate

whether

tis-sue-specific

alternative

splicing

occursandto

analyze

the un-cloned

regions

inthetwo

tissues,

fivesets of

specific primers

wereusedto

perform

RT-PCRonmRNAderived from

com-4.4 kb—

2.3 kb

_

2.0kb—

0.6 kb—

*

if

<o

&

& .

# %

Q %

&

BLBLBL BL BL kDa 94 _ 75 -45 _ 28 _ 22 _ 17

-B

kDa 94 . 75 -45 . 28 -22 -17

c#

<^.

V

Ä^>*>

MF

»i* tr

<fif?

- c-JAKl <—— ^

c-JH(l+2)

anti-JH2

Hi

&

^c^>:>

4*"

4P

MF

-, c-JAKl • -«

c-JH(l+2)

-*

pp60

» -« C-JH2 anti-PY

FIG. 7.

Transphosphorylation

ofC-JH2

by

c-JAKl and

c-JH(1

+

2).

Sf9 cellswereinfectedwith vAcJAKl alone

(lane

1),

both vAcJAKl and vAcJH2

(lane

2), vAcJH(l

+

2)

alone

(lane 3),

and both

vAcJH(l

+

2)

and vAcJH2

(lane 4),

respec-tively.

The Triton-insoluble fractions were

prepared

and then incubated with

Ni-agarose

gels

and

equivalent

amounts of bound

proteins

were

separated by

NaDodS04-PAGE,

trans-ferredto

nitrocellulose,

andimmunoblottedwith JH2 anti-bodies

(A)

or anti-PY mAb

(B).

Prestained molecular mass markers

(in kilodaltons)

areshownonthe left.

FIG. 8. RT-PCR

analysis

ofthe carp JAKl kinase

transcripts.

mRNAs derived from brain

(B)

and liver

(L)

tissues were

primed

with

oligo(dT)

and random

primers

and

subjected

to re-verse

transcription.

The

resulting

cDNAwas

amplified

with five

primer

sets,

i.e., F3/R3,

R4/R2,

F3/R2,

Fl/Rl,

and

F1/R2,

and thePCR

products

were

analyzed

by electrophoresis

on a 1% agarose

gel.

The sequence of each

primer

wasdescribed in the

DNAConstructionsection.

moncarp brain and liver tissues. PCR

products

from both tis-sues were identical in

length (Fig.

8).

Only

the uncloned

re-gions

inthetwo

tissues,

i.e.,

PCR

products

from

primers

F3/R3

and

F4/R2 (from brain),

and the DNA

fragment

from

Fl/Rl

(from

liver)

were

sequenced.

All of the nucleotide sequences werethesame as thosewe

deposited

in GenBank withan ac-cession number L24895. These sequences,aswellasthose from two

overlapping

clones,

pJ9

and

pJ21,

suggestthatatleasttwo

identical

transcripts

from carp brain and liver tissues encode the same

full-length

carp JAKl kinase.

Genomic

organization

of

the

carp

JAKl

gene

Asaninitial

step

to

investigate

the

regulation

of carp JAKl gene

expression,

we also cloned and characterized the carp JAKlgene. Four

positive phage

clones,

termed Jl to

J4,

were

isolated from a

Stratagene

carp liver

genomic library

with a

DIG-labeled

full-length

carp JAKl cDNAas a

probe.

We are unable tofill the gap between

phage

cloneJl and J2

by

PCR

amplification

of

genomic

DNA with a setof

oligonucleotides

that

correspond

to one end of clone Jl andthe other end of cloneJ2.To locate allexons,these

phage

cloneswere

analyzed

by

Southern

blotting, subcloning,

and

sequencing.

As shown in

Fig.

9,

the restriction map of each

genomic

clone was

con-structed

by digesting

the

phage

DNA with a

panel

of restric-tion enzymes

separately

orin variouscombinations:

5a/1,

Bgl

n,

Hind

III,

Xho

I,

and Eco RI. On thebasis of the nucleotide sequences of subcloned

fragments,

the carp JAKl gene is

com-posed

of24exonsthat spans atleast 31 kb ofDNA

(Fig.

9).

Thesequencesaroundthe exon-intron boundarieswere deter-mined andareshown in Table 1. Allexon-intron boundaries

identifiedconformedtothe

GT/AG

splice donor/acceptor

rule

(Breathnach

et

al,

1978).

Some exons were

relatively

small

(88-108

bp),

whereas the first and the sixth exons were

large

(237

bp

and 349

bp).

The size ofintronsvaried

considerably,

ranging

from >3 kb

(intron

1)

to 100

bp

(intron 19).

The first exoncontainedthe5'untranslated

region,

andthe secondexon

(10)

exon untranslatedexon

I_L

J_L

i i

2S

o— CJ00 Wffl

_u_

> Di o -Ü CO — -, Sx m a

_LL

J_L

> oí

Exon

no l 2 3 4 5 6 7 8 9 10 11 1213 14

JAK,

^JIJUL/IIII

IJUJU1JLJUI

15 1617181920 21 22 23 24 2.5kb >3kb 1.8kb lkb 1.2kb 1.5kb 2.3kb 2.2kb 1.5kb 1.8kb 1.5kb 1.4kb

Jl

/A

J2

J3

J4

"//

FIG. 9. Genomic

organization

of the carp JAKl gene. Exons are indicated

by

boxes numbered from 1 to24.Solid boxes

in-dicatethe carp JAKl

coding region

whereas open boxes

represent

the 5' and3' untranslated

regions.

Intronsand the5'-and

3'-flanking regions

are indicated

by

the solid lines. The entire gene spans atleast 31 kb in

length

andcontains 24exons. A re-striction mapwasshown above the

genomic

structure.Three

overlapping phage

clones,

J2-J4 andone

nonoverlapping

clone, Jl,

were isolated fromaLambda FIXII carp

genomic library.

Agap between

phage

clones Jl and J2was notobtainable

by

PCR

amplification

of

genomic

DNA.

containedthe

putative

translation initiation site. The

largest

in-tron

(>3 kb)

separates

exons1 and 2. The JH2 domainwas lo-catedon exons 11-17 and the

catalytic

JH1 domainwaslocated on exons 18-24. Exon 24containedthe last 33 amino acidsas

wellas the 3' untranslated

region.

The

promoter

and

exon/in-tronsequences of carp JAKl kinase gene have been

deposited

in GenBank with 10 serial accession

numbers,

from U53685to U53694.

Table 1. Exon-Intron Organizationof theCarp JakI Gene

exon exon size

number

(bp)

3'end of

5'end of

approximate

3'end of

5'end of

amino

acid

the

exon

the intron

size

(bp)

the intron

the

exon

interrupted

1 242 TGC CTG ACG AG gtaaggacga >3000 tatcctgcag T GTC TGG ATG

2 208 CAA GAA GAT GCC gtaagcgaga 1800 ttttccccag ACA TCT CAC Ala67 (3)

3 125 TAC CGT ATG AG gtgagtgggc 1000 gtttacacag G TTT TAT TTT Argl09 (2)

4 154 CTG TTT GCA CAG gtgaggtaca 400 tatgttttag GGC CAG TAT Glnl60 (3)

5 164 GCTGAG ACC AG gtgaggtaca 148 atctttgtag C TAC AAG CGA Ser215 (2)

6 349 AAG CCT GTA CCG gtgagtttag 250 cttttgatag AAT ATT CTG Pro331 (3)

7 180 AAT AAA ACT ATG gtaatttcca 600 ttgcccacag GAG TTG GAC Met391 (3)

8 158 GGA CCT ATC TG gtatgaccat 1200 ttctctacag C ACA GAG TAT Cys444 (2)

9 115 GTC TGC AH GAG gtacacacac 1500 tctctctcag ATG GAC CTA Glu482 (3)

10 199 CAA CCC AGA A gtaccagcac 2300 tttccttcag AA ATT TCC AAC Lys549 (1)

11 110 GAG ATT GTA CAG gtgatatttt 900 ttttattcag GGT GAG CAT Gln585 (3)

12 150 GAT ATC TCT CTG gtaagatgca 187 ctcctctcag GCT TTC TTT Leu635 (3)

13 89 CAT CAG GAG AA gtaagtacct 2200 ctgtctgtag T ATC ATG GTG Asn665 (2)

14 127 CTC AGC TAT CTG gtaaagaaac 1500 aatcctgtag GAG GAC AAG Leu707 (3)

15 136 AGCAGA GAA G gttgagagat 820 gtgtctacag AG TGT GTG GAC Glu753 (1)

16 152 AAGCTC ACA GAG gtaacagcat 111 ctctgagcag AAG GAG AGG Glu803 (3)

17 152 GAG AAA CAG AA gtatgaccta 457 ttctctccag T CCT TCC ATT Asn854 (2)

18 88 GAT CTTGGA GAG gtattttctg 110 ctcttttcag GGT CAC TTT Glu883 (3)

19 193 CAC GAA GAA G gtaaagccac 100 ttgtttcaag GT GGG AGA TCC Gly948 (1)

20 125 CAG ATC TGC CAG gtaacatcat 1800 ttaactccag GGC ATG GAC Gln989 (3)

21 173 CCA GTG TTCTG gtaagaatca 170 attttcccag G TAT GCC CCA Trpl047(2)

22 118 AGC CCT ATG TCG gtaagtggcc 1500 aaccattcag GTG TTC CTT Serl086(3)

23 111 TGC TCA GAC AGG gtaatatata 1400 gttttgacag TTG TAC TGT Argll23(3)

24 164 ATG AH GAC AAT CAG TAACTATGGAAATAAGATGAGATGCAGACTCCACCTCTTTTGTAAGAGGAAGTTCCAGAGAGACCAAAAAAAAAAA

_

(11)

FISH TYROSINE KINASE

JAKl

.NF-IL6 . -2541 TCCTCCTACGATCAGGCAGAAAATGTTTTnGACATTTATTTTGACATTATGTTGTTGTTTTGGGAAATAAAAAATCATTAAAATATTCACnnCACTC -2441 ATGTTTTTCCTGTAGACATCACGACTATAAAGAAAATTAAGACATCATTTCAGTTTGCAnTCATTTAATGTTnCTACAATTTCnTTCTTTCTGTGTC -2241 TTTTGGCACAGAAATATCCTTTTCAAATAGACACnGTAATCAAACTTTGCACAGACTTTAGTnCTGTGTTTnCAAAGACTACAAAATGTATATAAGT HNF-5. ... -2141 TAAAACAACAnGTACATCCTTATTTAnATGCATATnGTGTAAATATTAGCAGAAATGGGAAnCCATCCTTnACATATAnAACATTTGnTTAGG -2041 GCGACnACTACAAAAAGATGnTACnATATGAAACAATTTGACAACCAAACMATAGAAGTTTTAAAAAAAAAATAATACCTGCTTTTCTGTTCCCAT -1941 AAAACTGCTGAnTGTGATGTCAGACATnAAAAGCCATGAAGAGAAAGATATCATGGCCAAACCCCAAAACATTTGGAATCTCTGAAAAGCCCAATATG NF-IL6 .... -1741 CAGnCTAGGAGGnAATATAnGAATATCATTTTGTACAATATCAAAGCATATATGATGTATAAATTAGTTACTAAACAnnAnCAATTCATAGCAT -1641 nGGACACACACCATGAAGATnACAGTGTAGTATGATAAGTCnnCTGCCCATnCGTnATATAAnATAGACCATnTCCCATGATGTTAGTAAAT .API -1541 TAGACCCAGGTATTATCCTCTGTCACTTGCCCCnGnACATCAGCCCTnAACTCTTGAGTCATCTGAACATTTTAGAATCATATAGTAAGGTGAGTAA -1441 GTGGTCTGTAATGnGTGATATATGGAAATGCAGnATACAATnATTTTTTCnTTITnCAGAAAATAAAAATGTTTTCTCATCTGAAAAGCTGGGTC NF-IL6 . HNF-5 .... -1341 ATTnCAAnTGATGTTTACATGTATTTCGGAAATCCTCATCTAAATATTTGCTGAATnCTGTCTTAGCTTTTAGATGAGTGTTTATGGCAGTAnGCA . API -1241 CGTAGACTCATCACGGACCAAnATAGCCCATGACTCATCCAACGATAGGAGTTTCCAACTGGACTTTTTCCCCCCAGTGTnGTTTAAATAnAnGTC -1141 AACATATCTCAACACCATATTTGACATAAATATGTAnGTTGCAAATCATTnGACATGTCTGACAACATTTATATGGAAATGTTGATAAGGATAGACAA -1041 AAATGTATTTAATAGTAGAATnGAACTGAACGTTnACTCTAAAAACCTATTAAACTCCAATACACACAGTCTAATAnGTTTATATATCTACCCGAAC -941 nCTGnGAACTTAnGATCAGTCACAnaCCCAATGCTCTGCAGCCCACAGTTTCTGGCAACTCATATGATGCTGAGGACGGAAGTCCCTCCCTTTCA -841 CGGCAGGCnAAACCTCATGTCnAAAGACCATCTGCGTAATCAGAGAGGGGTTTGTGTTnGCTGATACCTATTTTAAAAACAGACACGTACCTGTAAA E2A ... -741 GCGGAAACAGCTGATTGATTMCTTAATGTnAGGAnAGGACTTrATGGACAAAAATAAACAACAGTAATAAAATGTAACAGnCAnAATAATATTTC NF-IL6 . . GHF-5 ... -641 TATTTACATTTTAnTCTGAAATCTAAAnATnATGAnATATCCAnATnATGAATCTGCAAMCATCTTTTACAGTGnGTGCTGTTACTGTTTAT -541 ATAGACTAGATAACTGCAGTATCTAAGCGnACnTTAAATGAGAAGCATAATAATGCnAGnAGCAGTAAnATTTAATATATGAATCAAnATTAAT -441 GGTTGTTATATnAAAGTGGTCATGTGGCATACCAnCAATGGACAAAATTnATCCGAGATATCTTTnAATACTACATATTTAACGAAGTGAAATCAG . TATA-box. -341 CGCnAAGGATCnCATGTAnATGAnCTAATCAnTTCAAACCATTTGAATCAnGGAGAGAGAGAAAAAAAAATGATATGTATAAATATGATGTTiT NF-IL6 ... -241 GGGAAATTTATGCATTTATGTTTACATTACAGCAATGCACTCTAAGCGAGAGTTTTGATCGCATGCATAAATAGTAAnCGTCTTTTAACTATATACAGT -141 ACAGAGGCAATGCGTCCAGTAGCCACAACAAAAAATCGCTGGCTTACTCTCTGCnAAGGGnAAAGCCAnGGTACAGAATTTGAnGACAGCCGCATG . CCAAT ... .+1 .... . -41 AGCCAATAAGCGGCATGnACAGACCTCGTCGnACCCTATTnCGGGAATGTGTTCGACGCTTGACAGGAAAGGTTAGTTGATCATGACGACTGAAGCA

.*

( transcription

start site of carp JAKl gene

)

.

+60 CGGnTGTAGCACTCTGAAGTGAAGTnGTnTGAAAATAAACCTCGTGGAATAAACCAnATACTATCGGACAnAnnGGAGTGAAATAACGAGCAC

+250 +160 TTTTACATCAACAAAAAACCAAGAGGAAGTGTTCAAACGTCTGTAATGATCGAAAnAAACGGAAGTAGCTTTGCCTGACGAGTGTCTGGATGCCAGAAC

(

translation start site

)

FIG. 10. Nucleotidesequence of the

5'-flanking region

of the carp JAKl gene. A 28-mer antisense

oligonucleotide

usedfor

primer

extension

analysis

is underlined. The candidate

transcription

startsite

by primer

extension

(see

Fig.

11)

is indicated with anucleotidenumber

(+1)

andan

asterisk,

which is locatedat249

bp

upstream

tothetranslationstartsite.Potential

binding

sites fora

variety

of

transcription

factors arealso marked and underlined. The

promoter

andexon 1 sequences of carp JAKl kinase gene has been

deposited

in

GenBank,

accessionnumber U53685.

Determination

of

the

transcription

initiation site

The

transcription

startsitewasdetermined

by

primer

exten-sion

analysis using

poly(A)+RNA

from carp liver. We useda 28-mer

oligonucleotide

labeled with

32P

atthe5' end. The

ex-act

position

of the extended

product

wasdetermined

by

align-ing

the

sequencing

ladderobtainedwith thesame

primer.

One

major

extended

product

wasrevealed and

corresponded

tothe siteat249

bp

upstream

tothe initiatormethionine codon. For

describing

the

5'-flanking region

of the carpJAKl gene, we

數據

FIG. 1. Restriction maps of carp JAKl cDNA clones. The ki- ki-nase-like (JH2) and kinase (JH1) domains are represented by
FIG. 4. Demonstration of tyrosine kinase activity of c-JHl in E. coli. Strain JM109 cells were transformed with pQE30 (lane 1), pQEJH2 (lane 2), and pQEJHl (lane 3), respectively
Fig. 1. On the basis of DNA sequence analysis, these two clones, pJl 1 and pJ21, are identical except that pJl 1 is 0.7 kb whereas
Fig. 6 (lane 1). Therefore, C-JH2 seems to interact specifically
+7

參考文獻

相關文件

You are given the wavelength and total energy of a light pulse and asked to find the number of photons it

Reading Task 6: Genre Structure and Language Features. • Now let’s look at how language features (e.g. sentence patterns) are connected to the structure

The Secondary Education Curriculum Guide (SECG) is prepared by the Curriculum Development Council (CDC) to advise secondary schools on how to sustain the Learning to

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =&gt;

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement.. 

• Formation of massive primordial stars as origin of objects in the early universe. • Supernova explosions might be visible to the most