Identification of insulin binding activity and isolation of endogenous insulin from rat liver

Vol. 44, No. 1, 1971

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

IDENTIFICATION OF INSULIN BINDING ACTIVITY AND ISOLATION OF ENDOGENOUS INSULIN FROM RAT LIVER* John G. Hemington** and Arnold Dunn Department of Biological Sciences University of Southern California Los Angeles, California 90007

Received Mayl2,

1971

SUMMARY: A d i s c r e t e peak o f immunoinsulin a c t i v i t y ( m o l e c u l a r w e i g h t 300,000) has been s e p a r a t e d by g e l f i l t r a t i o n from l i v e r homogenates. This f r a c t i o n chromatographs w i t h i n s u l i n a s e a c t i v i t y and c y c l i c n u c l e o t i d e p h o s p h o d i e s t e r ase. Endogenous i n s u l i n has been s e p a r a t e d from t h i s i n s u l i n b i n d i n g a c t i v i t y by v a r i o u s methods. Our r e s u l t s s u g g e s t t h a t when bound, i n s u l i n i s n o t d egraded by t h e i n s u l i n a s e a c t i v i t y . Much o f t h e i n s u l i n

r e l e a s e d by t h e p a n c r e a s i s t ak en up by t h e l i v e r .

A l t h o u g h i n s u l i n has m u l t i p l e e f f e c t s t i o n and u l t i m a t e f a t e a r e unknown. by r e d u c t i o n o f i n t r a c e l l u l a r levels. 1'2

on h e p a t i c m e t a b o l i s m , i t s These i n s u l i n e f f e c t s

cyclic 3',5'

of ac-

a d e n o s i n e monophosphate ( c y c l i c AMP)

One mechanism s u g g e s t e d f o r t h i s e f f e c t

tion of cyclic nucleotide phosphodiesterase

site

might be m e d i a t e d

is insulin

(PDE).

induced a c t i v a -

Insulin activation

has been r e p o r t e d in l i v e r 3 and f a t c e l l 4 , 5 p r e p a r a t i o n s . sues c o n t a i n two forms o f c y c l i c AMP p h o s p h o d i e s t e r a s e :

o f PDE

Extrahepatic tis-

a high molecular weight

form (400,000) and a low m o l e c u l a r w e i g h t form ( 2 0 0 , 0 0 0 ) .

P r e v i o u s l y only the

h i g h m o l e c u l a r w e i g h t form, which h y d r o l y z e s b o t h c y c l i c AMP and c y c l i c 3 ' , 5 ' g u a n o s i n e monophosphate ( c y c l i c GMP), has been r e p o r t e d i n l i v e r 6 ; c e n t l y r e p o r t e d the p r e s e n c e , cific

in l i v e r ,

we have r e -

o f th e low m o l e c u l a r w e i g h t form s p e -

f o r c y c l i c AMP h y d r o l y s i s . 7 The e x p e r i m e n t s d e s c r i b e d in t h i s paper i n d i c a t e

that insulin

can be s e p -

a r a t e d from l i v e r homogenates in a s s o c i a t i o n w i t h a h i g h m o l e c u l a r w e i g h t fraction,

presumably p r o t e i n .

(by g e l f i l t r a t i o n ) t h a t when i n s u l i n

The i n s u l i n

containing fraction

w i t h PDE and i n s u l i n a s e a c t i v i t y . is associated with this protein,

chromatographs

Our r e s u l t s

suggest

i t c a n n o t be degraded by

the i n s u l i n a s e o

* Taken from a dissertation submitted by J.G.H. to the Graduate School of the University of Southern California in partial fulfillment of the requirement for the degree of Doctor of Philosophy, January 1972. ** Present address: Johnson Research Foundation, University of Pennsylvania, Philadelphia, Pennsylvania 19104. This investigation was supported by Research Grant AM-07215 from the Division of Arthritis and Metabolic Diseases of the U.S.P.H.So, by Biomedical Sciences Support Grant RR-07012-04, National Institutes of Health and a Grant from the Diabetes Society of Southern California.

71

Vol. 44, No. 1, 1971

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Materials and Methods Following sodium pentobarbital anesthesia, livers were removed from male Sprague-Dawley rats (220-300 g) and homogenized in 8 volumes (w/v) of homogenizing medium in a Sorvall Omni-Mixer.

The preparation and assay of PDE were

performed according to the methods of Thompson and Appleman 8, except that the homogenizing medium contained 10.9% sucrose, 36.5 mM Tris-acetate buffer, pH 6.0, and 2.82 mM 2-mercaptoethanol.

In this method, approximately 7% of the

total liver protein and 100% of the total liver PDE activity is applied to agarose gel columns (Biogel A-5m).

Protein was measured by the Lowry method. 9

The PDE assay is a two-step enzymatic conversion of tritiated cyclic nucleotide to labeled adenine.

Assay concentrations of cyclic AMP and cyclic GMP

were 1.25 and 2.5 x 10 -7 M, respectively.

A unit of PDE activity equals one

picomole of cyclic nucleotide hydrolyzed per minute in an initial assay volume of 0.4 ml.

Results are expressed as units/ml.

Fractions were concentrated

using an Amieon Ultrafiltration Cell (UM20E membrane- 20,000 molecular weight "cut-off") and the pH was reduced to 3.0 with acetic acid.

Insulin was ex-

tracted from these concentrated fractions by various means: ultrafiltration, acid-ethanol extraction as modified by Arquillal09 and by chromatography on Sephadex G-75 (fine grade) with 0.167 M acetic acid.

Acetic acid solutions

were neutralized with 2 M Tris prior to insulin and enzyme assays. Insulin was assayed using a radioimmunoassay "kit" (Schwarz/Mann BioResearch, Inc., Orangeburg, New York).

Results are expressed as ~units/ml.

sulinase activity was measured by two different procedures.

In-

In the first

method, i00 ~units of standard human insulin (Schwarz/Mann) was added to fraction aliquots and incubated overnight at 4°C.

The insulin remaining at the

end of the incubation was assayed immunologically as above.

Results are ex-

pressed as the difference between ~units added and ~units remaining after the incubation.

In the second method, 1251-insulin (Schwarz/Mann) was added to

aliquots and incubated for i0 min at 30°C. of 5% trichloroacetic acid (TCA).

The reaction was stopped with i ml

After centrifugation, 0.5 ml aliquots of

the TCA supernatant were counted by liquid scintillation spectroscopy.

Re-

suits are expressed as the percent of the initial counts per min 1251 (corrected for blank) not precipitated by the TCA. Rats were made diabetic by intraperitoneal injections (50 or i00 mg/kg body weight) of streptozotocin. II Purina Lab Chow ad libitum.

They were maintained on 0.45% saline and

All diabetic animals tested positively for uri-

nary glucose, had plasma glucose concentrations greater than 400 mg %, exhibited weight loss (50-60 g) and reduced plasma insulin levels. were used 3 weeks after onset of diabetes.

?2

The animals

Vol. 44, No. 1, 1971

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Results Figure i compares column profiles IA) and diabetic

(Fig. IB) livers.

and 4 diabetic rats.

(Biogel A-5m) obtained

from normal

(Fig.

Similar results were obtained in 8 normal

All preparative

same time; equal amounts of protein

steps and assays were performed at the

(150 mg) were applied to the columns.

The

only apparent difference between normal and diabetic livers is the marked reduction in immunoinsulin

content of the diabetic profile.

as measured by the disappearance by streptozotocin

of added standard insulin, was not affected

induced diabetes.

In both normal and diabetic profiles,

Normal

90'-

Insulinase activity,

A

Diabetic

in-

B

E

"" 8 0 -

E

iI

~

=E

r.

~ ~o- ~ 6o-

g ~o~l'i~

•-- 50--0 50~, 2C '~40= "- ~a40o ~, i o -

E 30

~ 3o-

2c

~ 20-

IO

1 X Cycl;c AMP substrate ~ / ' A lmmunoinsulin iX / • Insulinase

i

II

I0 -

,

m .o EVo

so

,o

RACTION

,o

8o

NOMBER

mg

50 60

EVo FRACTION

TO BO NUMBER

Figure i. Chromatographic profiles of cyclic AMP phosphodiesterase, immunoinsulin and insulinase activity measured immunologicaliy on agarose gel (Biogel A-Sm) from normal (A) and diabetic (B) rat livers. Equal amounts of protein (150 mg) applied to columns. Buffer is 50 mM Tris-acetate, pH 6.0, 3.75 mM 2-mercaptoethanol.

sulinase activity does not appear as a single peak as do PDE and immunoinsulin activities.

In the preparation

from normal rat liver,

insulinase activity

peaks in fraction 50, decreases until fraction 54 and then increases again to a second peak in fraction 64.

This "notch" pattern is characteristic

sulinase activity when measured by this immunological method, both normal and diabetic preparations.

of in-

and occurs in

As can be seen in Figures i and 2 (B

and C), the "notch" in insulinase activity is coincident with the rise in immunoinsulin.

The immunoinsulin

at 37 ~U/ml, apparently Figure lB.

fraction in normal rat liver in Figure 2 peaks

similar to the amount shown for the diabetic liver in

This is due to different amounts of protein placed on the column

?3

Vol. 44, No. 1, 1971

BIOCHEMICAL AND BIOPHYSICALRESEARCHCOMMUNICATIONS

0

= o

Cyclic

GMP substrote

A

X Cyclic.AMP substrote

0.9

-20 o.~_= "= = -I0

- 0.6 o oJ
Q.

"°~ETM~ ~-10"20 ~

i x

~

"

-60

o~

llnsulinose-pU p -'11~

= ~ ° o~ ~' ' 4 0 0 [ n s u l i n a s e - % =

_%

jl

~

eo i

6 !

T

I

I

I

~

20-

Io 1

T

o

-

i

30 40 50 60 70 80 90 100 II0 FRACTION NUMBER Figure 2. Chromatographic profiles of protein (A280) , cyclic AMP phosphodiesterase and cyclic GMP phosphodiesterase (A), immunoinsulin (B) and insulinase measured both immunologically and by % non-precipitable CPM (C) from the same normal rat liver preparation. 70 mg of protein applied to column (Biogel A-5m). Buffer is 50 mM Tris-acetate, pH 6.0, 3.25 mM 2-mereaptoethanol.

(70 mg from normal liver vs. 150 mg from diabetic liver).

In most cases, the

immunoinsulin peak occurs in the same fraction as the lowest point of the "notch".

When insulinase is determined by measuring the 1251 remaining in the

TCA supernatant, the "notch" is absent.

With the exception of this "notch",

insulinase activity profiles, measured by both methods can be superimposed (Fig. 2C).

Also in Figure 2A, the low molecular weight form of PDE activity

(cyclic AMP hydrolysis only) appears as a shoulder of the total PDE activity between fractions 76 and 95.

Both insulinase and immunoinsulin appear as dis-

crete peaks of activity between the two forms of PDE and are not coincident with either form.

The estimated molecular weight of these peaks is 300,000;

presumably, the insulin is bound to a protein.

?4

Vol. 44, No. 1, 1971

A

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

1.5 ml conc.agarose gel fraction, pH 3.0 0.8

-30

A

A I mmunoinsulin

0.6 0

~(~

GO OJ

0.4<

~ 3

'~ =

no -

02

-

Lilly, insulin, glucagon low No detectable absorbance at 280 nrn

B -30 -20 -I0 I 6

A 8

J IO

~ 12

i 14

A ~ '~ A A~AZ~ 16 18 20 22 24 26 FRACTION NUMBER

~ 28

J 30

J ;52

~ 34

Figure 3. Camparison of chromatographic properties on Sephadex G-75 of immunoinsulin activity between concentrated agarose gel fractions (A) and standard insulin (Lilly, glucagon low) (B). Medium is 0.167 M acetic acid.

Figure 3 compares the results of chromatography on Sephadex G-75 of a standard insulin (Lilly, glucagon low) (Fig. 3B) with the concentrated agarose gel fractions (Fig. 3A)°

These samples were chromatographed on the same col-

umn with the same flow rate.

Both insulin peaks were detected immunologically

and have similar chromatographic properties.

The high molecular weight pro-

tein fractions (200,000 - 400,000) are excluded from the Sephadex gel.

Enzy-

matic activity (PDE and insulinase) is not detectable after adjusting the pH to 3.0, followed by neutralization. not have insulinase activity.

The immunoinsulin peak in Figure 3A does

In addition, after acidification immunoinsulin

and not insulinase activity or other larger proteins pass through the UM20E membrane. able.

Similarly, after acid-ethanol extraction only insulin is detect-

The possibility of contamination of liver preparations by circulating

insulin seems remote since plasma or red blood cells prepared in a similar manner do not have comparable activity. move blood still gave comparable results.

Also, livers perfused in situ to reFurthermore, the insulin appears

first with the 300,000 molecular weight fraction. Discussion Since both insulin and insulinase occur in the same agarose gel column fractions, the insulinase activity could give a false positive inmnunoassay for insulin by degrading added iodinated insulin.

?5

The comparison of insulin and

Vol. 44, No. 1, 1971

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

insulinase content of normal and diabetic liver fractions indicates, however, that such interference may not occur.

Although the insulinase activity of nor-

mal and diabetic liver fractions is the same, the amount of immunoinsulin in the diabetic fraction is reduced, as expected.

The estimated molecular weight

of the immunoinsulin fraction is 300,000 indicating that the endogenous insulin is probably bound to a high molecular weight fraction, probably a protein. In addition, insulin has been dissociated from the binding protein and from insulinase activity by acidic conditions and then separated on Sephadex G-75 (Fig. 3), acid-ethanol extraction and ultrafiltration.

We propose that the

endogenous insulin, when bound in this form, is protected from the insulinase activity present in the same fractions, but can still be recognized by the insulin antibody.

The different results obtained by the two insulinase methods

support this hypothesis.

Insulinase, when measured by degradation of exogenous

iodinated insulin, has a single peak of activity.

Insulinase measured by the

disappearance of exogenous human insulin, detected immunologically, should have the same result; the "notch" is interpreted as bound insulin which the insulinase could not degrade. Since no attempt was made to separate sub-cellular components, we cannot assign a cellular locus of the insulin binding protein.

It is possible that

this might be a membrane insulin receptor, as reported for fat cells. 12,13 The nucleus might be a second possible site since insulin has been extracted from isolated rat liver nuclei. I0

The insulin binding protein and insulinase

are discrete peaks and represent only a small portion of the total protein present in the homogenate.

Furthermore, we find that these proteins are either

absent or present in only small amounts in tissues known to be insensitive to insulin (brain, kidney, plasma, and red blood cells).

These findings suggest

that the insulin binding protein and insulinase might be involved with the physiological action of insulin in the liver.

References i.

Exton, J.H., S.B. Lewis, and C.R. Park. In press.

1970.

Ann. N.Y. Aead. Sci.

2.

Jefferson, L.S., J.H. Exton, R.W. Butcher, E.W. Sutherland, and C.R. Park. 1968. J. Biol. Chem. 243: 1031.

3.

Schultz, G., G. Senft, and K. Munske.

1966.

4.

Loten, E.G., and J.G.T. Sneyd.

Biochem. J. 120: 187.

5.

Senft, G., G. Schultz, K. Munske, and M. Hoffman. 322.

1970.

6.

Thompson, WoJ., and M.M. Appleman.

7.

Hemington, J.G., W.J. Thompson, and A. Dunn.

1971.

8.

Thompson, W.J., and M.M. Applemano

1971.

76

Naturwissenschaften 53: 529.

1968.

Diabetologia ~:

J. Biol. Chem. 246: 3145. Fed. Proc.

In Press.

Biochemistry IO: 311.

Vol. 44, No. 1, 1971

9. i0.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Lowry, O.H., N.J. Rosebrough, A.L. Farr, and R.J. Randall. Chem. 193: 265. Arquilla, E.R.

II.

Recup, C.C.

12.

Kono, T.

13.

Cuatrecasas, P.

1971.

1970.

1969.

Personal communication.

Pharmacol.

Rev. 2_22:485.

J. Biol. Chem 244: 5777. 1969.

Proc. Nat. Aead. Sci. USA 6_~3: 450.

77

1951.

J. Biol