Identification of a cDNA clone in λgt11 for the transacylase component of branched chain ketoacid dehydrogenase

Vol. 131, No. 2, 1985 September

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

16. 1985

Pages

OF A cDNA CLONE IN lgtll OF BRANCHED CHAIN KETOACID

IDENTIFICATION COMPONENT

Stuart Division Received

Litwer

and

Dean

July

FOR THE TRANSACYLASE DEHYDROGENASE

J.

of Medical Genetics,Department Emory University, Atlanta,

961-967

Danner of Pediatrics, GA 30322

1, 1985

Two cDNA clones for the transacylase protein of the branched chain ketoacid dehydrogenase complex CE.C. 1.2.4.41 have been isolated from a human fetal liver cDNA expression library in Xgtll using antibody selection. By selective antibody elutfon from nftrocellulose filters containing the fusfon proteins* it was determined that these inserts represent the transacylase protein. These data support the hypothesis that thfs protein is synthesized in the cytosol on transcrfpts independent of the other proteins of the branched chafn ketoacid dehydrogenase complex. 0 1985 Academic Press, Inc.

Branched

chain

a multienzyme membrane

(1).

disease,

result

this

coding

associated

with

Inherited

human

disorders,

these

weight

55,000)

ketoaci

d,

whfle

El

formed

from

decarboxylase only

in

proteins

form

urine forming show

implying the

nuclear basfc

complex.

CEC 1.6.4.31, of

the

(molecular

branched

chain

dehydrogenase

forms

the

B subunits

the

syrup

families (2)

(3).

complexes.

E2 component of

46,000 Amino

phosphorylation

is

inner

proteins

patterns

activfty for

the

affected

pyruvate

a and

BCKDI

mitochondrial maple

component

protein

available

subunit

of

and acid site

BCKD

37,000

Da

sequence in

Ela

(4). Defects

result

a common

Da transacylase is

the

dehydrogenase

and

the is

Four

a-ketoglutarate

providing data

is

of

inheritance

lipoamide

CEC 1.2.4.4,

defects

analysis

proteins.

A flavoprotefn,

A 52,000

specific

Pedigree recessive

for

dehydrogenase

complex

from

complex.

autosomal

ketoacid

in

in decreased

any

of

the

three

activity

of

unique the

961

subunits

complex

All

and

of the

Copyright 0 1985 rights of reproduction

BCKD can ensuing 0006-291X/85 $1.50 by Academic Press, Inc. in any form reserved.

Vol. 131,No.

2, 1985

clinical

picture

antisera

has

subunit,

as

use

these

been

the

of provides

independently

m described in hgtll School of

to genes

a cDNA the

gene

on

cDNA

BCKD the

direct

suggests

52

evidence this

component,

clones

directing for

BCKD specific

disorder

oligonucleotide

a separate

Ilhrnr~ by Young was kindly Medicine,

inherited

develop

clone

RESEARCH COMMUNICATIONS

disease. a missing

the

select

first

and

urine

describe

for to

AND BIOPHYSICAL

syrup

to

a mechanism

of

a nuclear

maple

used

library

isolation This

of

antibodies

expression analysis

BIOCHEMKAL

protein

from

kDa,

(5). the

probes

synthesis. EZr that may

the

Here

We now Xgtll for we

component BCKD-EP be

E2

use report of

is

in

BCKD.

coded

on

made

transcript.

A cDNA library & Davis (6,7) provided by Houstonr TX.

produced essentially as from human fetal liver inserted Dr. Savio C.L. Woo, Baylor

Polyclonal antisera was produced in rabbits AJ.whlIlasi as previously described (8). Antibodies which recognize &9harIchla cc;nll proteins were absorbed from the IgG fraction by affinity gel chromatography. E. coli strain BNN97 containing Xgtll without the cDNA library was grown in LB broth to stationery log phase and the cells pelleted by centrlfugation. Cells were washed in 100 ml of phosphate buffered saline CPBSI, pelleted by centrifugation# and suspended in 40 ml of O.lM HOPS pH 7.5 for lysis by sonication. After removal of cell debrls by low speed centrifugation, the soluble proteins were mixed with BioRad Affigel 10 8 15 according to the manufacturer's instructions. Coupling was allowed to continue overnight at 4'. Unreacted coupling sites were blocked by incubation of the gel for one hour in IM glycfne ethyl ester. After extensive washing with PBS, the mixed bed Afflgel-lo,15 was incubated overnlght at 4' with anti-BCKD IgG. The gel was then poured into a 0.7 X 10 cm column and unbound antibody collected. This procedure was repeated and the solutions of unbound IgG combined for use in screening the llbrary. E. co11 strain Y1090 was infected Su!%ning ltha Ilhrary: Selection of plaques with Xgtll containing the cDNA library. producing antigenic proteins was mad8 essentially as described by deWet ti nL (9). Briefly, 2.8 X 10 recombinants were plated in top agarose on 7 150cm2 plates of LB agar. After growth for 4 hours at 42' C the plates were overlayed with a nitrocellulose filter impregnated with 10 m# isopropyl-B-D-thiogalactopyranoside CIPTGI. After 2 hours at 37' C, filters and plates were keyed and the filters removed. Each filter was soaked in 50mM TrisHClr pH 7.4 containing 150mH NaCl, 1% gelatin and 1% Tween 20 (IM-1) for 1 hr and then in 50mM Trfs-HCl, pH 7.4 containing 150 0.5H EDTA, 1% Tween 20 (IM-2) for 45 min. mM NaCl, 0.25% gelatin, Filters were then placed in IM-2 with 5% bovine serum and 25~1 of anti-BCKD IgG per 10 ml for 1 hr. After washing 3 X 10 mfn with IM-2, the filters were placed in IM-2 plus 5% bovine serum and 3 962

Vol.

131,

No. 2, 1985

BlOCHEMlCALANDBlOPHYSlCALRESEARCHCOMMUNlCATlONS

ul/lO ml goat antlrabbit IgG-peroxidase CBio-Radl for 45 min. Filters were washed 3 X 10 min with IM-2 prfor to stalning in filtered solution of 5OmM Trls-HClr pH 7.5 containing 0.3mg When the color diaminobenzidine/ml and 0.2ul/ml of 50% H202. fully developed (S-60 min), the filters were washed in deionized water and dried. This staining procedure gave lower background than using radiolabel detection. Plaques producing antlgenic fusion proteins were selected for subsequent rounds using the large bore of a sterile Pasteur pipette. Plaque number/plate reduced 10 fold for each successive round of screening beginning at 400,00O/plate down to 4O/plate.

a was

was

Selar;j&2n Qf E2 dQnas: To Select the clone for the E2 protein, plaques were grown as before and the protein production induced with IPTG impregnated nitrocellulose filters. Filters were exposed to the polyclonal BCKD-IgG for 1 hr. Unbound antibodies were washed from the filters and specifically bound antibodies eluted with 5mH glycine-HCl, pH 2.3, contalnlng 15OmM NaCl, 0.5% Triton X-100, and 100 ug BSA/ml. Washes were immediately neutralized with 1M Tris-HCl to a pH of 7.4. Eluted antibodies were used for Western blot analysis (5) of either purlfied BCKD or mitochondrial proteins. A similar procedure has been reported for the identification of the glucocordicoid receptor gene (10).

Initial positive

plaques

through all

six

were

were after

third

tested

and

of

rounds positive

expanslon. final

2.8

found

successive

plaques

positive the

screening

X

which of

for

round

recombinant

five

phage

each

Figure (6th)

lo6

of 1

of

continued selection.

five shows

selection

phage. positive At

clones stained for

Seven

and

this

remained

filters clone

point,

from 58.

Figure 1: Nltrocellulose fllters stalned with dlamlnobenzldtne as described for Western blot analysts In Methods. Left side shows the third selectlon round at a density of 4000 plaques/plate. Right side shows the final selectlon round with all plaques staining posltlve for antlgenic fusion protetn. Data is for clone 58. 963

Vol.

131,

No.

2, 1985

BIOCHEMICALAND

BIOPHYSICALRESEARCH

COMMUNICATIONS

Figure 2: Western blot of fusion protein from the five positive clones. The two outside lanes are molecular weight markers. The flve middle lanes represent fusion protein from clones 6, 58, 5A, 4 and 3 readtng left to rtght. Arrow Indtcates the posltion of 8galactosldase.

To

estimate

production these

the

of

an

clones

IPTG plates

2

hr

into

of

In

Xgtll

a

control

procedure

which were

BCKD

library

was and

from coli

with

to

antigenic

In

a

7.5% blot

identified

BNN97

kDa

by

(figure

which

the

product

single

Western

200

through

2mM

from

proteins 180

by a

washed

sulfate.

strain

carried

no

plaques

resolved

single

ranging E.

production

lauryl

revealed

antisera of

used

the

complex

selectively exposed

and

the

protein

were

buffer

substantiate

contains

induction was

detected

shown).

subunits the

the

sodium

sizes

analysis

Polyclonal

of

sample

products

and

fusion

proteins

containing

the

insert

overlaying

experiment

and not

four

(11)

with

without

(Data

Induced

these

BCKD-antibodies 2).

37'.

Laemmli

cDNA

product, by

gel

analysis

the

induced at

polyacrylamide

of

antigenic

was

for

size

BCKD

to

complex.

these bound resolved

for

To

fusion the mouse

clonal

proteins

fusion

proteins mltochondrial 964

selection

recognized

establish

which

represent, from proteins

all

proteins antibodies

clones

3 in

and a

58

Vol. 131, No. 2, 1985

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

w-=

Western blot of mouse liver mftochondrfa reacted with antf odfes selectively eluted from nftrocellulose filters containing fusion protein. Left lane represents antfbodfes from lane represents the analysis with polyclonal clone 3, middle antisera and the right lane represents antibodies eluted from clone 58.

Western

blot

staining

analysis.

patterns

Figure

3 shows

complex

or

assay

with the

human

with

the

patterns

polyclonal

selected

Similar

E2 protein.

Staining

antisera

antibodies

results

were

were

mitochondrfal

on

obtained

eluted

the

same

recognize

proteins

specifically

compared

the

the

purified

tested

antibodies

filter.

exclusively

when were

to

BCKD

by this

(data

not

shown).

Rl.ssassinn Reduced severe

activity

pathologic

Causative

genetic

questions

also

multienzyme has

been

directed

of condition,

remain

to

the

BCKD complex maple

alterations

complexes proposed

the

remain

regarding entirely

whereby inner

syrup to

from

be

of 965

of

nuclear made

humans

urine

assembly

proteins

membrane

in

results

disease.

described.

Basic

mitochondrial genes.

in

mitochondria

in

the

A mechanism

cytosol by

are

N-terminal

a

Vol. 131, No. 2, 1985

leader for

BIOCHEMICAL

sequences

(12,13).

lipoamide

unique transcripts

of for

we demonstrate

This

cDNA

from

the

c-terminal

a molecular

large

enough

insert

would

to

analysis.

two

of

pub1 ication from

disease,

Taken

together,

we

subunits

coded,

of

no

El.

dehydrogenase on

the

(figure

data

In

in El

of

precursor

transcripts

El

Western

were

present

the

B subunits

independent

appears

from

each

syrup

were that

E2

present. is

independent for

describing

it

in

E2 protein maple

transcript

reports

(14,15).

the

with

translated

the

If

by the

a hypothesis

the

kDa. component

of

a patient

a and

on

be

a previous

absence

support

must

one

In

basepair

cDNA 90

than

3).

cytosolically

view

to

53

2).

CEC 3.2.1.231

E2 protein

the

information

Da(Figure gene

80

more

for

E3,

200,000

antigenically

proteins

these

have

detected

the

membrane.

inserted of

we demonstrated

while

a nuclear

the

from

clones

ca

that

individual

B-galactosidase

116,000,

epitopes

mltochondrial

urine

made

a polypeptide

(51,

the

As yet

encode

isolated

from

of

described

suggests

mitochondrial

Since

peptides

data

been

B-galactosidase

of

Only

the

the

(6).

be

the

protein

weight

would

Our

in

in

end

contain

these

blot

as

resides

has

synthesized

assembly a fusion

library

has

be

RESEARCH COMMUNICATIONS

precursor

(14,151.

BCKD may

later

Here

protein,

A larger

dehydrogenase

proteins

AND BIOPHYSICAL

made

protein. the

~1 or

8

lipoamide

that

E3 and

other

and

E2 are

distinct

from

El. Availability containing study

of mutations

both

diesase

and

multienzyme

the the

and

for

biochemical

Sells

in

S.L.C. for

the

and BCKD

resulting

reactions

human

fibroblasts

proteins

allows

us

in

syrup

urine

involved

maple in

the

assembly

to

of

mitochondria.

We thank Dr.

clone,

mechanism

complexes

help, Susan

E2-cDNA

specific

genetic

&Kng.whdw: their

this

Drs. Woo

manuscript

E.I. for

sharing

preparation.

Ginns the

6 J.A. Xgtll

Barranger cDNA

for library

Vol. 131,No.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

2, 1985

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Dannet-, D. J. and Elsas, L.J. (1975) Biochem. Med. 1l 7-22. Rosenberg, L.E. and Striver, C.R. (1980) In wMetabolic control and disease” Ed. by P.K. Bondy and L.E. Rosenberg, Philadelphia, pp. 691-699. W.B. Saunders, Heffelfinger, S.C., Sewell, E.T. and Danner, D.J. (1983) Biochemistry 22, 5519-5522. Cook, K.G., Bradford, A.P., and Yeaman, S.J. (1984) Eur. J. Biochem. J& 587-591. Danner, D.J., Armstrong, N., Heffelfinger, S.C., Sewell, Priest, J.H., and Elsas, L.J. (1985) J. Clin. Invest. &:*;158-860. Young, R.A. and Davis, R.W. (1983) Proc. Natl. Acad. Sci. USA BQ, 1194-1198. R.A. and Davis, R.W. (1983) Science 222, 778-782. Young, Heffelfinger, S.C., Sewell, E.T., and Danner, D.J. (1983) Biochem. J. Uz 339-344. deWet, J.R., Fukushima, H., Dewji, N.N., Wilcox, E., O’Brien, J.S. and Helinski, D.R. (1984) DNA 1, 437-447. Weinberger, C.# Hollenberg, S.M.D Ong, E.S., Harmon, J.M., Brower, S.T., Cldlowski, J.# Thompson, E.B., Rosenfeld, M.G., Evans, R.M. (1985) Science 228, 740-742. Laemmli, U. (1970) Nature 221, 680-685. Felipo, V. and Grisolla, S. (1984) Curr. Top. Cell. Reg. 21, 217-249. Hay, R., Bohni, P. and Gasser, S. (1984) Biochim. Blophys. Acta lzPI 65-07. Koike, K., Tsuji, A., Urata, Y., Morlyasu, M.* and Koike, M. pp B03-806rEd. by R.C. (1984) In Elnuins an9 fln~nprn~a~~ Bray, P.C., W. de Gruyter 8 Co., New York. Matuda, S., Shirahama, T., Saheki, T., Miura, S. and Mori, M. (1983) Biochim. Biophys. Acta 141, 86-93.

967