Vol.

167,

No.

February

BIOCHEMICAL

1, 1990

28,

BIOPHYSICAL

RESEARCH

COMMUNICATIONS Pages 280-286

1990

NATIVRIQRMOF

BlWTRRLINRRCEPT0Rl.NIlUIfANPpLAcwlTALwEHBRANEs

Shigeo

Nakajo,

Masanori

Sugiura

Department University School

Vanderbilt Received

AND

January

22,

Inagmi

and Tadashi

of Biochemistry of Medicine,

Nashville.

TN

37232

1990

sunuARY. Little is known about the native form of endothelin receptor. To clarify its functional and structural properties, we solubilized the receptor from human placenta in an active form using mild detergents CHAPS and digitonin in a specific manner, with and showed that it is able to bind 1251-endothelin-l The a pH optimum between 6 and 8 in contrast to a reported pH optimum of 4. molecular weight of the receptor was estimated as 340,000 by gel filtration of the solubilized membrane in the presence of 0.2% (w/v) digitonin. When the solubilized membranes were labeled with 1251-endothelin-l prior to gel filtration, the radioactive ligand also migrated in the position corresponding These results indicated that the native form of to a 340 kDa protein. endothelin receptor in human placenta is a 340 kDa protein. 01990 Academic Press,

1°C.

The action aunique

shows

dependence

action.

This

Specific

binding

The

of the newly

cardiac

with

of

specific

membranes

the

(2),

were

strong

proteins

renal

glomeruli

with

detergent and

extensive

characterization

clarification

of its

use of the native detergent

of biochemical

the

mesangial

researchers. and tissues.

(SDS). site.

and the form

cells

then with

(41,

and

proteins

subjected respect

to to the

rigorous

required

controversies

of the

chick

solubilized

More

receptor

in

binding

methods, exists

binding

endothelin active

of

the

(1)

physiological

of cells

studies

sulfate

properties

and biochemically

rat

Controversy the

its

ET was demonstrated

(3).

dodecyl

endothelins

attention

by cross-linking

of SDS. of

for

types

In these

1251-ET-1

properties

the for

(5).

sodium

ion

in various

binding

in the presence weight

placental

attracted

fractions

labeled

molecular

Here

rat

membrane

electrophoresis

a mild

property

peptide

calcium

of ET was demonstrated

human placental in membranes

vasoconstrictor

on extracellular

interesting

presence

discovered

and

for

the

demands

receptor

the

solubilized

by

from

human

treatment.

we report membrane

the fraction

solubilization and

its

of binding

ET

binding

properties

activity to

1251-ET-1.

The abbreviations used are: CHAPS: 3-[(3-cholamidopropyl) dimethylammoniol-lpropanesulfonic acid, ET: endothelin, PMSF: phenylmethylsulfonyl fluoride, HEPES: 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid,EGTA: ethyleneglycolbis-(B-aminoethyl ether) N,N,N',N'-tetraacetic acid. 0006-291X/90 Copyright AN rights

$1.50

0 I990 by Academic Press. Inc. of reproduction in any form reserved.

280

The

Vol.

167,

No.

1, 1990

solubilized subunits

BIOCHEMICAL

binding detected

protein

AND

showed

by cross-linking

BIOPHYSICAL

a greater followed

RESEARCH

molecular by SDS-gel

size

COMMUNICATIONS

than

the

binding

electrophoresis.

FIATRRIALS AND naTHoDS Materials. ET-l was radioiodinated with [ i251]NaI (New England Nuclear) and the monoiodinated ET-l was purified by HPLC with a Vydac Cl8 reverse as described previously (5). The specific activity of 12%g~TT-;01~ approximately 2,000 Ci/mmol. CHAPS and bis(sulfosuccinimidyl)suberate (BS3) were purchased from Pierce Chemical Co. Leupeptin, aprotinin, PMSF, bacitracin, pepstatin, heparin (4-6 kDa), poly-L-lysine (22 kDa) and digitonin were obtained from Sigma Chemical Co, Bio-Gel A-1.5m (200-400 mesh) and marker proteins for gel filtration were from Bio-Rad. ET-l was from Peptides International, Inc. All other reagents were of analytical grade. Binding Assay. Binding assay was carried out according to the previous report (5) with slight modifications. Since ET-l receptor forms aggregates readily and does not require divalent metal catious such as MgC12, MnC12 and CaC12 for stabilization or ligand binding, as shown below, we added 0.05% (w/v) Triton X100 to the binding assay buffer to prevent aggregation and omitted MgC12. Briefly, 0,.2 ml of binding assay buffer containing 30 mM HEPES (pH 7.5), 0.15 M NaCl, 0.5 mg/ml bacitracin, 0.4 mM PMSF, 1 mg/ml bovine serum albumin, 0.05% (w/v) Triton X-100 and CHAPS-solubilized membrane fraction (30 ug of protein) was incubated with 75 pM 1251-ET-1 for 2 h at 4OC. Nonsp~~i,ce~;~-~2;a.s determined in the presence of 125 nM nonradioactive ET-l. ET-l complex was separated from free 1251-ET-1 by filtration using a glass fiber filter (6). Gel Filtration of Solubilized Membrane Fraction. Membrane fraction from human placentas was prepared and solubilized with 0.75% (w/v) CHAPS as described previously (5). To the solubilized membrane fraction, digitonin was added to a final concentration of 1% (w/v), then 0.25 ml of the solubilized fraction (containing 1.25 mg of protein) was applied to a Bio-Gel A-1.5m column (1 x 48 cm) equilibrated previously with 25 mM HEPES (pH 7.5) containing 0.15 M NaCl, 3 mM EDTA, 1 mM EGTA, 0.25 mg/ml bacitracin, 3 ug/ml leupeptin, 0.4 mM PMSF, 2 rig/ml aprotinin, 3 vg/ml pepstatin and 0.2% (w/v) digitonin. The flow rate of and 0.68 ml fractions were collected. the column was approximately 4 ml/h, 1251-ET-1 binding activity in the Aliquots of 0.05 ml were used for measuring presence 0.r absence of 125 nM unlabeled ET-l. An aliquot of the membrane Binding and Cross-liuking of lz51-ET-1 Receptor. fraction solubilized with CHAPS (containing 0.75 mg of protein) was incubated with 400 pM 1251-ET-1 in 0.18 ml of binding assay buffer in the presence or absence of 600 nMunlabeled ET-l for 3 h at 4OC. After incubation, 20 ~1 of 10% (w/v) digitonin was added, then the sample was immediately applied to a Bio-Gel For cross-linking experiment, the reaction A-1.5m column as described above. mixture was further incubated with 0.8 mM BS3 for 25 min at 4OC after binding of 1251-ET-1 t o solubilized receptor, and subsequently quenched by adding The sample was' treated with 1% (w/v) digitonin, then glycine (10 mM final). analyzed by gel filtration as described above.

RESULTS Specific studied such These

under as Mg2+, divalent

binding the Mn2+

of

various

1251-ET-1 conditions

and Ca2+,

cations

did

not

or

to

the

as shown EDTA did

affect

not

1251-ET-1 281

solubilized

membrane

in Table alter

the

binding

I. ET-l

fraction

Divalent

metal

binding

to intact

was ions

activity.

membranesnot

Vol.

167,

No.

Table

1, 1990

I.

BIOCHEMICAL

Effects

of

%Control

Control

Poly-Lys

solubilized assay

Llglml)

(10

vg/ml)

(100

ug/ml)

98 100 96

(0.5

P-f)

101

(2.0 uM)

101

(10

102

PM)

detergent.

buffer

or

binding

chloride

polycationic

of 1251-ET-1. 1).

an optimum

Addition

resulted

Sodium

activities.

(Fig.

(1

by the

pH range

of

in

105 (Mg2+),

also

showed

activity between

specific

binding

(1 (10 (10 (10

mM) mM) ml-l) IIN)

102 101 99 104

NaCl

(0.2 (0.5

M) M)

100 97

(1.0 M)

92

10 mM each

of divalent

97 (Mn2+)

a symmetrical membranes

1.0 M.

showed binding bell-shaped

(without

01

on the

investigated

was

pH profile

detergent

with

treatment)

CHAPS Diaitonim Triton X-100 Lubrol-PX

CAPS

5

7

9

II Concentration

PH Fig. 1. Effect of pH on specific membrane fraction (30 vg of protein) h at 4OC at various pH. Fig. 2. solubilized ET-l for

% binding

03

binding of 1251-ET-1. was incubated with

Effect of detergent on specific membrane fraction (30 ug of protein) 2 h at 4'C in the presence of various

282

(%)

CHAPS-solubilized 75 pM 1251-ET-1 for

binding of lz51-ET-l. was incubated with detergents.

to

Polyanionic

no effect

. A 0 .

q

cation

and 96 (Ca2+)

up to

substance

Intact

to

EDTA M&l2 MnC12 CaC12

of pH on the specific showed

6-8.

COMMUNICATIONS

%Control

no effect

(poly-L-lysine) The effect

The binding

RESEARCH

Condition

100

Heparin

(heparin)

BIOPHYSICAL

various conditions on the 1251-ET-l solubilized particulate fraction

Condition

binding

AND

CHAPS75 pM 12'1-

2

Vol.

167,

were for

No.

also the

ET-1

1, 1990

subjected

binding

(Fig.

2).

to

In order digiton eluted

below

prior

gel

binding solubilized

activity

peak

membrane

in

inhibition

X-100

a similar

causing

pH-profile

inhibited

the

dose-dependent

than even

0.8%

while

1251-

manners

any other at

half-maximal

respectively,

the native with

filtration,

under

showed partially

remained

and 0.2% (w/v),

"MATERIALS

as a single

also

COMMUNICATIONS

detergents (w/v).

The

inhibition those

were

of Lubrol-PX

0.1%.

solubilized to

under

RESEARCH

fractions

less

activity

to determine

fraction

It detergents

showed

binding

0.5% (w/v)

9!J were

BIOPHYSICAL

membrane

of CHAPS and Triton

approximately

described

Various

solubilized

of

AND

analysis.

lz51-ET-l.

the

56%

concentrations

membrane

similar

Among them digiton and

and Brij

to

of

binding

tested,

BIOCHEMICAL

molecular CRAPS was

then

the fractions

labeled

40

20 Fraction

stable

Number

of

ET-l

treated

a Bio-Gel

specific

weight

remained were

to

ET-l

a molecular peak

further

applied

AND METHODS". with

weight

receptor, with

A-1.5m

binding 340,000

(Fig.

for

several

days

1251-ET-1

1%

(W/V)

column

activity

of

with

the

by

3). at

The O'C.

cross-linking

60 (0,68ml/tube)

Fig. 3. Gel filtration of solubilized membrane fraction. CHAPS-solubilized membrane fraction (1.25 mg of protein) was treated with 1% (w/v) digitonin and then applied on a Bio-Gel A-1.5m column (1 x 48 cm) in the presence of 0.2% (w/v) digitonin. Fractions were assayed for specific binding of lz51-ET-l as described under MATERIAL AND METHODS". The inset shows the molecular weight determination of ET-l receptor. Standard proteins (0) included were 1, myoglobin (17 kDa), 2, ovalbumin (44 kDa), 3, y-globulin (158 kDa), and 4, thyroglobulin (670 kDa). The molecular weight of ET-l receptor (x) was Vo and Vt indicate the elution positions of blue estimated to be 340,000 . dextran and cyanocobalamin. respectively. 283

as was If

Vol.

167.

No.

BIOCHEMICAL

1. 1990

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

6 A.

B. Noncovalent

Covalent

* -a-

i 3 4

+ cold - cold

+ -

+ cold - cold

4 !

6

Fig.

20 Fraction

4.

40

Gel

membrane fraction the (A)

BS3

Fig.

4A.

prior

of

125-I-ET-i-receptor

(0.75 mg of protein) or absence (0) of 600

and then subsequently AND METHODS".

using

20 Fraction

(0.6BmlItube)

filtration

presence (01 or without (B)

0

60

Number

filtration,

The molecular

of

respectively. cold ET-l

prior

When lz51-ET-1 binding to the chromatography,

contrary,

the

28 kDa peak was not

130

kDa

peak

increased

the

nonspecifically

labeled

from

gel

of

receptor

(Fig.

4B).

ET-l

receptor

filtration These

on human of

(cross-linked)

the

with

indicate

that

.placental

membranes membrane as shown

that complex

presence ET-l,

in

28,000,

of excess On the furthermore,

130

kDa and

28

The

same result

kDa was

between

lz51-ET-1 and molecular form of

the native

is

and

disappeared.

cold

lz51--ET-1.

solubilized

in with

as shown

340,000

out in the

by excess

noncovalent

results

gave a 32 kDa band

were

indicating

were

identified

340 kDa peak

abolished

proteins

400 pM 1251-ET-1 and was reacted

were

peaks

was carried

obtained

electrophoresis

two

.significantly,

CHAPS-solubilized

with ET-1

peaks

the

60 ml/tube)

treated with 1% (w/v) digitonin as described under "MATERIALS

two major

weights

40 (0.6B

Number

complex.

was incubated nM unlabeled

BS3. The sample was further subjected to gel filtration

to gel

n

a 340

kDa protein.

fractions

labeled

previously

SDS gel

with

1251-ET-1

(5).

DISCDSSION The present in a stable been the

demonstrated isolation Treatment

ET-l

receptor

results

indicate

and functionally with

that

active

intact

membranes.

and characterization of the

placental

completely

which

the receptor

form.

To date

could

with

be observed 284

ET-l

can be solubilized

many ET binding

The present

of solubilized membranes

for

results

ET-l 1% (w/v)

studies

open the

have

way for

receptor. digitonin

as a sharp

solubilized

340 kDa peak upon

Vol.

167,

gel

No,

1, 1990

filtration.

Confirmation

protein

was also

1 with

or

the

obtained

without

On the other of

BIOCHEMICAL

hand

binding

amount

of

by the

340

BIOPHYSICAL

kDa

0.75% (w/v)

CHAPS produced

activity

at

eluting

4A and

incomplete

position

as a broad

ET-l

prelabeled

BS3 (Fig.

the

COMMUNICATIONS

as specific

membranes

reagent,

eluting

RESEARCH

protein

solubilized

cross-linking activity

of binding

AND

binding

with

4B,

1251-ET-

respectively).

solubilization

of

void

peak with

with

volume

65%

and a small

a molecular

weight

range

of 420,000-350,000. The binding or

divalent

contrast

to the

to rat

renal

Mn2+

for

the

pH 4.0

papillary the

showed

in

Present

optimal

that

for

weight

Taken placental binding

detergent

a neutral

(8) have

weight of

immunoglobulin.

This

E'reviously as ET-l subunit

binding

of

EDTA

membranes

(7).

are

The lack

EDTA was confirmed

with

The reason

1251-ET-1

to

rat

to

be optimum

human placental Again,

the

in

1251-ET-1

was reported

pH range.

shown that

ET-1

we

membranes

chick

papillary at

membranes

reason

in chick

protein

for

the

32

kDa

was

cross-linking membrane

of

that subunits

340

differed

Ca2+ channel greater on

based The of

that

in the present human

procedure in the present

kDa ET-l

including

than

obtained

protein

solubilized

indicate

heart

ultracentrifugation.

by chemical

in the

composed

by heart

of the result

identified

results are

estimated in

is in support

have

binding these

was

receptor

result

receptor

voltage-dependent

which

ET-l

was obtained

together

of

membranes.

or intact

dihydropyridine-sensitive,

molecular

similar

The

require

binding

as solubilized

of a detergent

using

the

Ca2+ or

not

observations

clear.

et al.

the

membranes

Mg2+,

did

These

or human placental

clear.

at

Ca2+.

Mn2+ stimulates

as well

absence

membranes

or

by Mn2+,

studies

not

molecular study.

not

the

binding is

Miyazaki from

is

receptor

(3)

binding membranes

discrepancy

discrepanc,y

that

membranes

of

(3).

as Mg2, reports

human placental

membrane

on

to th e solubilized

such

previous

of stimulation intact

of 1251-ET-1

cations

receptor

32 kDa protein

placental (5).

A

studies. on human as ET-l

subunit.

We are grateful to Dr. F. Boehme of the Department of Gynecology and Obstetrics at Vanderbilt University for placentas and to Ms. Doris Bruey for typing the manuscript. This research was supported by research grants HL-14192 and HL-35323 from the National Institutes of Health.

1.

Yanagisawa, Mitsui, Y., 415.

M., Kurihara, Yazaki, Y.,

H., Kimura. S., Goto, K. and Masaki,

2.

Watanabe, H., Miyazaki, H., Kondoh, Yanagisawa, M., Masaki, T. and Murakami, Commun. 161, 1252-1259.

285

Tomobe, Y., Kobayashi, T. (1988) Nature 332,

M., Masuda, Y., K. (1989) Biochem.

Kimura. Biophys.

M., 411S., Res.

Vol.

167,

No.

1, 1990

BIOCHEMICAL

E.R., Marsden, Biophys. Res.

AND

BIOPHYSICAL

3.

Martin, Biochem.

4.

Sugiura, M., Snajdar, R.M., Schwartzberg, M., Badr, K.F. (1989) Biochem. Biophys. Res. Commun. 162, 1396-1401.

5.

Nakajo, S., Sugiura, Biochem. Biopphys.

6.

Takayanagi, R., Inagami, Misono, K. (1987) J. Biol.

7.

Fischli,

8.

Miyazaki, H., Takahashi, M., J. Cardiovascular

W., Clozel,

P.A., Brenner, B.M. Commun. 162, 130-137.

RESEARCH

M., Snajdar, Res. Commun.

R.M., Boehm, 164, 205-211.

T., Snajdar, Chem. 262,

M. and Guilly,

and Ballermann,

F.H.

R.M., Imada, 12104-12113. C. (1989)

COMMUNICATIONS

Life

B.J.

and Inagami,

and Inagami. T., Sci.

(1989)

T. (1989)

Tamura, 44,

T.

M. and

1429-1436.

Kondoh, M.. Watanabe, H., Hayashi, T., Murakami, K., Yanagisawa, M., Kimura, S., Goto, K. and Masaki, M. (1989) Pharmacol. (Suppl. 51, S155-S156.

286