Insulin-receptor interactions in liver cell membranes

Vol. 44, No. 2, 1971

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

INSULIN-RECEPTOR INTERACTIONSIN LIVER CELL MEMBRANES

P. Cuatrecasas, B. Desbuquois and F. Krug Departments of Medicine, and Pharmacology and Experimental Therapeutics, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 Received May 21, 1971 SUMMARY. The specific binding of 1251-insulin to liver cell membranesis a Binding is displaced by low concensaturable process with respect to insulin. trations of native insulin but not by biologically inactive insulin derivatives or by other peptide hormones. The rate constants of association (3.5 x 106 mole-1 set 'l) and of dissociation (2.7 x 1O-4 set-1) of the insulin-membrane complex can be determined independently. The dissociation constant of the complex, determined from the rate constants and from equilibrium data, is about 7 x lo-l1 M. Complex formation does not result in degradation of the insulin molecule. The binding interaction is a dissociable process involving a homogeneous membranestructure which is almost certainly the biologically significant receptor. The kinetic properties, and the effects of enzymic perturbations of the membrane, suggest that the insulin receptors of liver and of adipose tissue cells may be very similar structures. Large agarose beads containing like biological

activity

covalently

bound insulin

when tested on isolated

possess insulin-

adipose tissue cells,

sug-

gesting that the receptors for this hormone are located in superficial of the cell

(1).

Sensitive

techniques have been developed which permit direct

and accurate measurementsof specific receptor

interactions

interaction

in intact,

metabolically

by standard kinetic

The'insulin

receptor of fat cells

identical

properties,

fat cell

(2).

The

process which can be

appears to be located exclusively activity

in the membranefraction

and the effect

in the intact

insulin-

responsive fat cells

(2). The insulin-binding

is recovered quantitatively

kinetic

significant

expressions.

the plasma membraneof these cells cells

and biologically

is a simple, homogeneousand dissociable

characterized

regions

(2).

of intact

A number of

of chemical and enzymic perturbations, and in the membranefraction

is reasonable assurance that study of the binding of insulin branes by these procedures reflects

biologically

333

specific

in

(3).

are

Thus there

to fat cell mem-

receptor processes.

Vol. 44, No. 2, 1971

BIOCHEMICAL

At present the interaction

AND BIOPHYSICAL

of insulin

with receptors in liver

be studied only by measuring the specific cell preparations.

The properties

with the properties

of an intact

accurately

insulin

tem from liver in liver

cell

liver

binding of labeled hormone to broken-

system since it is not yet possible to measure in a simple, biologically

It is hence difficult

fragments measure specific

describes the basic properties membranepreparations.

remarkable similarities

to ascertain

receptor

of the interaction

between the properties

responsive sys-

that binding studies

interactions.

This report

between insulin

The nature of the interaction

observed in fat cell membranes, indicate ceptor of liver

cells can

of this binding cannot be compared directly

binding properties

cells.

RESEARCH COMMUNICATIONS

and isolated

itself,

of this interaction

that the biologically

and those

important re-

cells is being studied.

METHODS. The liver

membranes(from Sprague-Dawley rats) were prepared (4)

from homogenates by differential

centrifugation

in 0.25 M sucrose solutions;

they were separated from the major nuclear and mitochondrial

elements.

tein was determined by the method of Lowry -et al. (5). 125 I-Insulin was prepared and purified as described in detail (2). (2).

and the

This 125I-insulin

is biologically

indistinguishable

elsewhere

from native insulin

125 binding of I-insulin

The assay used to measure specific

Pro-

to membranes

was similar

to that described for measuring binding to intact fat cells (2) and 125 -10 fat cell membranes(3). Membraneswere incubated with I-insulin (10 M to lo-l2

M) to equilibrium,

Every experimental point binding of insulin

then filtered

and washed on EGWPMillipore

(performed in triplicate)

by performing identical

(20 to 60 pg per ml) of native

insulin

is corrected

filters.

for nonspecific

incubations in which large amounts

are added before the iodoinsulin

(2).

RESULTS. Specific

binding of insulin to liver cell membranesis detected at -11 very low concentrations (10 M) of the hormone (Figure 1). The insulin-10 M 125 binding sites in the membraneare saturated at 3 x 10 I-insulin. The maximal binding capacity of these membranepreparations insulin

per mg of membraneprotein.

334

is about 0.1 pmole of

Vol. 44, No. 2, 1971

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

to Liver Membranesby Peptide Displacement of 125I-Insulin-Binding Hormones Liver membranes(71 pg of protein) were incubated for 30 minutes at 24" in 0.2 ml KRB buffer, 1% albumin, a peptide hormone, and 1.7 x 10-11 M 1*51-insulin. TABLE I.

binding of 125I-insulin

Specific

Addition

nmoles x 10e6/mg protein No additions

4.2

+ O.l*

Native insulin,

2 rnlg per ml 8 rnvg per ml 0.4 pg per ml Proinsulin, 0.2 pg per ml 10 yg per ml Desoctapeptide insulin, 5 pg per ml

3.0 2.1 0.1

+ 0.1 i 0.1 i 0.1

3.4 0.9

-I- 0.2 i 0.1

4.0

+ 0.2

Glucagon, 50 pg per ml

4.3

+ 0.1

Growth Hormone, 50 pg per ml

4.2

+ 0.1

"2 standard error of the mean (three observations).

I I

I

2 [‘WI-INSULIN].

I

t

3

4

M x 1010

to liver membranes Figure 1 - Dependenceof specific binding of 1*5I-insulin Membranes(50 Kg of protein) were incubated on the concentration of insulin. for 40 minutes at 24" in 0.2 ml of Krebs-Rin er-bicarbonate (KRR) buffer, 1% Three ml of ice-cold (w/v) albumin, and the indicated amount of 195I-insulin. KRRbuffer, 0.1% albumin, were added to each tube, and the suspension was rapidly filtered on EGWPMillipore filters and washed with an additional 10 ml of ice-cold buffer. Values are corrected for nonspecific adsorption of insulin, as described in the text. 335

Vol. 44, No. 2, 1971

BIOCHEMICAL

AND BIOPHYSICAL

binding of 125 I-insulin

The specific

placed by low concentrations

of native

to liver insulin

of glucagon or growth hormone (Table I), high as 5% (w/v). sulin,

The biologically

(devoid of biological Proinsulin

inactive,

activity)

insulin.

Insulin

reduced or oxidized

that its affinity

the binding of 125 I-insulin

follows first-order Table II.

kinetics

The dissociation

stants (7.7 x 10-"

binding (Table I).

is 16 times less than that of native dissociates

the insulin-membrane complex.

to liver

is confirmed by kinetic (2).

(4 x

Desoctapeptide insulin

cell membranesis a time-

dependent process which obeys second-order kinetics.

the studies of fat cells

chains of in-

from the membrane(Table I) but the displace-

antiserum readily

of this interaction

of albumin as

does not compete for insulin

binding of 125I-insulin

The specific

dis-

but not by high concentrations

as high as 0.2 mg per ml.

125 displaces I-insulin

ment curve indicates

membranesis readily

or by concentrations

alone or in combination, do not affect

-11 M) at concentrations 10

RESEARCH COMMUNICATIONS

Dissociation (Figure 2).

The bimolecular nature

analyses similar

to those used in

of the insulin-membrane complex The rate constants are summarized in

constant calculated

from the ratio

of the rate con-

M) is nearly the sameas that (6.7 x 10-11 M) calculated

from the equilibrium

data presented in Figure 1.

These values are similar

to

those of the insulin-fat cell complex, which have been reported (2) as 5 x -11 M (from rate constants) and 8 x 10-11 10 M (at equilibrium).

TABLE II.

Kinetic Parameters of Insulin-Receptor Membranes (at 24")

Rate of dissociation constant,

k,l/kl

At equilibrium

No appreciable destruction tions of incubation.

in Liver

3.5 (+ 0.4) x lo6 mole-1 set -1 -4 -1 2.7 (+ 0.2) x 10 set

Rate of association

Dissociation

Interaction

7.7 x lo-l1 M 6.7 (+_0.8) x lo-l1

of medium125I-insulin

A suspension containing 336

M

occurs during the condi-

0.5 mg of membraneprotein per ml

Vol. 44, No. 2, 1971

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Figure 2 - Semi-log plot of the dissociation of 1251-insulin bound to liver membranesas a function of time at 24'. Membranes (0.1 mg of protein) were incubated for 40 minutes at 24" in 0.2 ml of KRB buffer, 1% albumin, and 3.5 x lo-lo M 1251-insulin. Three ml of KRRbuffer, 1% albumin, (at 24') and native insulin (10 pg per ml) were added. At the indicated times the suspensions were rapidly filtered and washed on Millipore filters. -10 M 12511% albumin, and 2.2 x 10 125 was incubated for 40 minutes at 24". The I-insulin recovered after

of Krebs-Ringer-bicarbonate insulin filtration

on Millipore

(KRH) buffer,

filters

was 95%as effective

as native

125 I-insulin

when tested for binding to fresh membranes. These results are in contrast

TABLE III.

Properties Complex

of 125I-Insulin

Eluted from Insulin-Liver

to

Membrane

Twelve ml of KRB buffer, 1% albumin, containing 6 mg of liver membraneprotein, were incubated for 20 minutes at 24" with 3.0 x lo-10 M 1251-insulin. The suspension was centrifuged at 22,000 x g for 20 minutes. The membraneswere resuspended in 12 ml of the samebuffer (4") and centrifuged again. 1251-insulin was eluted from the pellet by suspending in 0.8 ml of 0.2 N HCl containing 10% albumin and incubating for 30 minutes at 24“. The suspension was centrifuged, and the supernatant was neutralized and tested by physical methods (2) and binding. 125 I-Insulin Properties

Not used

% Precipitable by 8% TCA % Adsorption to talc % Adsorption to silica Specific binding to membranes* i-2' . = $11 -11 ; 125;-;;?;:: 125 ales

of 125I-insulin

97

98

96 97

96 98

3.6 1.1 x lO+j 10-5

per mg of membraneprotein. 337

Eluted from membranes

1.2 x 1O-6 3.8 1o-5

Vol. 44, No. 2, 1971

BIOCHEMICAL

the marked ability

of liver

insulin

tightly

insulin

is similar

its

ability

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

membranesto inactivate

bound to the liver to native

glucagon (4,6).

The 12'1-

membranescan be eluted with acid.

125 I-insulin

by several physical criteria

to bind to fresh membranes(Table III).

The eluted and by

The binding process is not

associated with degradation of insulin. The specific by digesting

binding of 125I-insulin

to liver

cell membranesis enhanced

the membraneswith phospholipase C or phospholipase A, and by in-

creasing the ionic strength of the medium (Table IV). erties

are very similar

to those of the insulin

fat cell membranes(2,3). similar

to the effect

TABLE

IV.

The effect

on fat cells

These rather unique prop-

receptor of intact

of trypsin

digestion

fat cells

and

(Table IV) is also

(3).

Effects of NaCl and Enzymic Digestions of Liver Membraneson the Specific Binding of 1251-Insulin

Liver membranes(0.5 mg per ml) were incubated for 40 minutes at 37' in EBB buffer, 1% albumin, containing phospholipase C (50 pg per ml), phospholipase A (70 pg per ml), trypsin (0.1 mg per ml), and in somecases CaC12 (10 mM). The enzymes were removed by centrifugation, suspension in buffer, and re-centriMembrane suspensions (60 ug per ml) were tested for specific 1251fugation. insulin (1.1 x lo-10 M) binding as described in Figure 1. Phospholipase A effects are only observed in the presence of CaC12; Ca* has no effect on phosThese effects will be described in more detail elsepholipase C digestion. where (3). MembraneTreatment

Specific

Binding of 1251-Insulin

nmoles x 105/mg protein No enzymic digestion Phospholipase C, 5 pg per ml 50 pg per ml Phospholipase A, 20 ug per ml (no Ca*) 20 pg per ml (Ca*) Trypsin, 0.1 mg per ml NaCl, 0.5 M 1.0 M 2.0 M 3.0 M

DISCUSSION. Binding of iodoinsulin recently

(7,8),

but the properties

The insulin-membrane interactions

5.1 9.4 9.8 5.2 7.1 0.3 9.1 12.4 19.6 10.7

to liver

f + + T T ? T T 7 I

0.2 0.3 0.2 0.3 0.2 0.1 0.3 0.1 0.2 0.3

membraneshas been reported

of the binding have not been characterized. presented in this report almost certainly 338

Vol. 44, No. 2, 1971

BIOCHEMICAL

represent

interaction

for

the specific

this

hormone.

binding

Specific

is observed

interaction

is

tKd> 7 x lo-l1 There

are

between

binding

only

with

a saturable

biologically

active

reversible

M) near

the physiologic

process

in

of hormone

analogs,

enzymes

insulin

with

liver

and with

intact

little

doubt

that

iments

on intact

ceptors

of liver

cells

and adipose

cells

The present

tissue

process

The

of insulin

in serum

(kinetics, of cell

are being suggest if

not

for

of insulin. constant

and fat

may be similar,

receptor

a dissociation

strength)

data

cell

and competition

derivatives

having

interactions

(2).

and the liver

the properties

and ionic fat

insulin-receptor

RESEARCH COMhtUNlCATlONS

concentration

similarities

effects

fat

insulin

is

a simple,

striking

AND BIOPHYSICAL

(9).

affinity

the interaction membranes. studied that

and of

There

is

in the experthe insulin

identical,

re-

chemical

structures. Acknowledgements: We thank Miss Elizabeth O'Connell for technical assistance. This work was supported by NIH grant AM14956 and the Kroc Foundation, Santa Ynez, California. P. Cuatrecasas is a U.S.P.H.S. Research Career Development Awardee (AM31464). B. Desbuquois is on leave from Institut National de la Sante et de la Recherche Medicale, France.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9.

Cuatrecasas, P., Proc. Nat. Acad. Sci. U.S.A., 63. 450 (1969). Cuatrecasas, P., Proc. Nat. Acad. Sci. U.S.A., in press (June, 1971). Cuatrecasas, P., manuscript in preparation. Desbuquois, B., Krug, F., and Cuatrecasas, P., manuscript in preparation. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., J. Biol. Chem., 193. 265 (1951). Rodbell, M., Krans, H.M.J., Pohl, S.L., and Birnbaumer, L., J. Biol. Chem., 246, 1861 (1971). and Weidemann, M.J., Biochem. Biophys. Res. Commun., 41, House, P.D.R., 541 (1970). Freychet, P., Roth, J., and Neville, D.M., Biochem. Biophys. Res. Commun., 43- 400 (1971). Cahill, G.F., Jr., Lavis, V., Soeldner, J.S., Sloan, D., and Steimke, J., J. Metab. Clin. Exptl., 3 769 (1964).

339