Cholesterol hydroperoxides inhibit calmodulin and suppress atherogenesis in rabbits

BIOCHEMICAL

Vol. 146, No. 3, 1987 August 14. 1987

RESEARCH COMMUNICATIONS Pages 1166-1172

AND BIOPHYSICAL

CHOLESTEROL HYDROPEROXIDES INHIBIT CALMODULIN AND SUPPRESS ATHEROGENESIS IN RABBITS Carl

L. Tipton, Pak C. Leung, Joni S. Johnson, Robert J. Brooks, and Donald C. Beitz

Department of Biochemistry and Biophysics, Iowa State University, Ames, Iowa 50011 Received

June

26,

1987

A mixture of cholesterol autoxidation products, prepared from an aged sample of cholesterol by recrystallization from methanol, inhibits calmodulin irreversibly in a Ca2+-dependent reaction. Inhibitory activity is lost after treatment with NaBH4, NaCNBH3, or NaI, from which we conclude that calmodulin inhibition is due to one or more cholesterol hydroperoxides. Partially purified cholesterol hydroperoxides, with or without cholesterol, were fed to young adult white rabbits. Cholesterol in the diet caused extensive atheroma formation in the aortas, but the addition of cholesterol hydroperoxides markedly reduced lesion formation. A cholesterol hydroperoxide preparation that was reduced by treatment with NaI was not effective in preventing atheroma formation. Cholesterol hydroperoxides did not lower cholesterol concentrations in blood plasma, liver, or heart. D 1987 AcademicPress, Inc. The

of

role

cholesterol

been a controversial cholesterol

oxidation

topic.

Peng

oxidation

responsible

for

products,

an initial

atherosclerosis. are

cell

other

markedly

less

Taylor

hand,

in

(11,

some

arterial

On the

cholesterols

products

and

of

which

injury

that

Higley

atherogenic

et

to

atherogenesis

for are al.

rabbits

has long

example,

argue

cytotoxic,

eventually (2)

may be results

find

than

that

that

highly

in

oxidized purified

cholesterol. Cholesterol of which

autoxidation

extensive

interest

in

to determine observations,

the metabolic

cholesterol

inhibit

tested

produces

are hydroperoxides,

for

Materials

reported their

effects

from cholesterol

calmodulin

the

of that

the

are

products,

cholesterol

cholesterol

suggested

number

others

oxidation

effects here,

a large

which

of products, derived

first

Despite has been done

little

hydroperoxides.

hydroperoxides that

the

(3).

these

The

in autoxidized

compounds

should

be

on atherogenesis.

and Methods

The starting material for isolation of cholesterol hydroperoxides was USP cholesterol purchased from Nutritional Biochemicals Inc. and stored at room temperature for approximately 20 years. The purification procedure was adapted from van Lier and Smith (4) and Teng et al. (5). The yields given in the following procedure are averages from 14 preparations. Lots of 50 g of the aged cholesterol were dissolved in 1.5 L of hot methanol and cooled, yielding 37 g crystalline cholesterol. The filtrate was reduced to about 0006-291X/87 Copyright AN rights

$1.50

0 1987 by Academic Press, Inc. of reproduction in any form reserved.

1166

Vol. 146, No. 3. 1987

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

500 ml, and a second crop of crystals (5.0 g) was collected. The filtrate then was taken to dryness in vacua, redissolved in ethyl ether, extracted with 0.1 N NaOH, washed with water, and dried over anhydrous Na2S04. After removal of the ether, the residue was dissolved in methylene chlorideacetone 9:l (v/v) and chromatographed on a column of Sephadex LH-20. Fractions were examined by TLC on silica gel, developed with benzenemethanol 8:l (v/v). Cholesterol and other sterols were detected with I2 vapor, and hydroperoxides were detected by spraying the plates with N,N-dimethyl-p-phenylenediamine (6). Fractions eluting after the last detectable cholesterol were combined, yielding 2.03 g of a mixture enriched in cholesterol hydroperoxides (hereinafter referred to simply as cholesterol hydroperoxides) . Measurement of tri-iodide formed by reaction with KI (7) indicated that about 40% of the material was sterol hydroperoxide, and thinlayer chromatograms showed at least three hydroperoxide bands. A “reduced hydroperoxide” preparation was obtained by treating a solution of the cholesterol hydroperoxides in methanol with a 1 .5-fold molar excess of NaI (calculated on the assumption that all the starting material was hydroperoxide). After 3 hr, an excess of Na2S203 was added. When the yellow color was gone, the methanol solution was diluted with diethyl ether, washed several times with water, dried over Na2S04, and evaporated to dryness. The initial experiments showing inhibition of calmodulin were carried out using the supernatant from recrystalization of aged cholesterol without further purification. This solution contained 26.8 mg/mL solid residue. Testosterone, 4-cholesten-j-one and cholesterol-5a6a-epoxide were obtained from Sigma Chemical Co., St. Louis, MO. Cholesterol-7a-hydroperoxide was synthesized as described by Schenck et al. (8). Calmodulin was assayed as described before (9). Young adult New Zealand white rabbits were used for the feeding experiments, four animals for each treatment, except the controls, which had three. The basal diet was Purina rabbit chow pellets. Sterols to be added to the diet were dissolved in diethyl ether and poured over the pellets. The solvent was allowed to evaporate under a stream of N2. After 61 days (1st experiment) or 56 days (2nd experiment), the animals were anesthetized with The aortas were split longisodium pentothal and killed by exsanguination. Lipids were extracted by the tudinally and stained with Sudan IV (10). method of Bligh and Dyer (ll), and the concentrations of free and esterified cholesterol were determined enzymatically (12). Results As

seen

in

Table

recrystallization

of

trations.

37

The

PM.

then

The active

their

below

resulted,

inhibition form.

inhibit

incubated Reduction

molecular

that

only, in the with

with

concentration

inhibition if the however, presence borohydrides 1167

of

supernatant at

of 400 for assay

component

the

the

calmodulin

weight

was pre-incubated

an effective

in

in the

of any single

showing occurs

are

cholesterol concentration

When calmodulin

cholesterol

remaining

concentration

diluted

inhibition

materials

an average

products,

much lower. and

aged

Assuming

autoxidation to

1,

of

concen-

cholesterol is

from

the mixture

0.37

must

autoxidized

cholesterol

of autoxidized

cholesterol,

is irreversible calmodulin and

CaC12, or with

low

the

mixture

after

with

(Table 2). autoxidized

calmodulin

NaI destroyed

be

in the

the

inhibi-

BIOCHEMICAL

Vol. 146, No. 3, 1987 Table

AND BIOPHYSICAL

Inhibition of calmodulin-dependent by cholesterol autoxidation

1.

RESEARCH COMMUNICATIONS

CAMP phosphodiesterase products nmoles

Autoxidized cholesterol ug/mL in assay mixture

min

P1/30

+CaM

0

-CaM 46.6

43.7 45.6 15.4

1.5 15 150

4.0

3.5

4.3

6.9

8.3

4.7

4.6

Autoxidized cholesterol in 0.5 UL methanol was incubated with 80 VL phosphodiesterase assay mixture (+/- 4 units of calmodulin) for 30 min at 30°C. The assay was then started Each datum by addition of CAMP and continued for 30 min. represents a separate experiment.

tory and

activity not

(Table

epoxides

as one or more ides (8)

are is

3). or

cholesterol

effective; (data

not

fed

hydroperoxides,

Sudan

areas

IV are

by cholesterol tration

is

prevented

the

addition

that

have

of been

all

with

a test

reduced

Table

the

of by

2.

NaI.

shown

in Figure

Additions mM CaC12, 0.25 PL methanol 6.7 pg autoxidized 2 mM EGTA, 6.7 pg autoxidized No calmodulin

1. with

with Areas but

The cholesterol was

repeated

of feeding

cholesterol

of calmodulin) assays shown.

darkly

of

cholesterol

hydroperoxides (Figure

2)

with

hydroperoxides

hydroperoxides

again

30.1,

1168

infil-

cholesterol

cholesterol cholesterol

assayed

by

supplemented

of calmodulin by autoxidized is Ca2+-dependent ~nmoles P1/30 min

were

without

the lipid

9.5, 31.4, 6.0,

33.8

9.5 31.5 6.4

Samples of 3.34 pig calmodulin in 100 uL of 20 mM Tris-HCl, PH 7.0, 1 mM Mg acetate, and 1 mM imidazole plus the additions shown in the table were incubated at 30°C. After 90 min, aliquots were diluted to a calmodulin concentration of 2 ng/pL. Ten-pL aliquots of the diluted samples (4 units duplicate

were

or

stained

a diet

50 mg/day

1

1 mM CaC12,

activity 5-cholesten-3g-

formation,

by including

effects

Inhibition cholesterol

hydroperox-

shown).

Animals lesion

cholesterol.

The

not

material

photochemically

inhibitory

1% cholesterol,

The experiment the

active

cholesterol

without

122 uM (data

extensive

effects.

hydroperoxides

the

synthesized

containing

completely

only

4-cholesten-3-one,

infiltration.

have

along

at are

almost

have no obvious

Also

diets

of lipid

alone

hydroperoxides

to reduce identifying

Not

6a-epoxide, all

of rabbits

cholesterol

alone

shown).

or testosterone,

Aortas

thus

hydroperoxides.

cholesterol-5a,

ol-7-one,

be expected

compounds,

cholesterol-7a-hydroperoxide,

inactive

cholesterol,

NaI would

carbonyl

(9).

Results

of

pre-

Vol. 146, No. 3, 1987

BIOCHEMICAL Identification cholesterol

Table 3. autoxidized Reducing autoxidized

agent added cholesterol

AND BIOPHYSICAL

of hydroperoxides responsible for

RESEARCH COMMUNICATIONS

as the calmodulin

vented

lipid with

responsible

100

3 15 8

infiltration NaI for

Figure

1.

Figure

2.

the

was

of

Percent inhibition of calmodulin

to

None NaBH4 NaBH3CN NaI

reduced

component inhibition

of ineffective.

reduced

lipid

the

aorta, Thus,

but

the

hydroperoxides

material

that

had

evidently

infiltration.

Aortas stained with Sudan IV to reveal lipid infiltration, first experiment. Diets of the rabbits were (left to right): control, 1% cholesterol, cholesterol hydroperoxides, cholesterol plus cholesterol hydroperoxides. same as Figure 1 except the last two groups, Second experiment, received reduced cholesterol hydroperoxides and bottom row, cholesterol plus reduced cholesterol hydroperoxides, respectively.

1169

been are

BIOCHEMICAL

Vol. 146, No. 3, 1987 Table _-___--

Blood

4.

plasma

AND BIOPHYSICAL

cholesterol first experiment

Treatment

and

RESEARCH COMMUNICATIONS

cholesterol

mg/dL 11 10 446 524

~Ester

Cholesterol 2 S.D.

Cholesterol

Control + hydroperoxides + cholesterol + both

esters,

k3 + 2 f 163 t 113

8?5 8+3 894 844

+ 214 f 328

-Table

Blood

5.

plasma

cholesterol and experiment

cholesterol

esters,

second

Treatment

Cholesterol

Cholesterol +_ S.D.

mg/dL

11 +a

Control + hydroperoxides + reduced hydroperoxides + cholesterol + cholesterol + hydroperoxides + cholesterol + reduced hydroperoxides

The effect effect

of cholesterol

on lipid

peroxides prevent

uptake

do not lipid

lower

a28 820

f f

459 58

901 1430

* 486 f 252

484

f

104

810

+ 473

hydroperoxides

plasma

wall

liver

may be quite cells

cholesterol

in the

* 20 + 2

27 f 7

5k2

by artery

deposition

30 21

7+-3

Ester

because

levels

the

(Tables

and heart

specifically

(Table

an

cholesterol

4 and 5)

hydro-

nor

do they

6).

Discussion Although do not

have

vation

that

it the

quite

protective

effect Bell

Table

6.

in

of

unexpected.

is

related

and

Schaub

to the (13)

and

cholesterol

autoxidation,

the

We have

no direct

of the

for

this

and of Kaul

g wet

liver

have a that

comes (14)

and

that

this

from chlorpro-

heart,

HEART Cholesterol Esters

tissue

199 229 234 1217

+ -t + f

36 23 22 615

17 21 14 1941

+ ? i f

12 11 a a40

75 107 41 160

f f f t

1539

f

93

3638

f

528

201

k 26

1170

obser-

hydroperoxides

__ Cholesterol

mg/lOO

Zero-time control Control diet + hydroperoxides + cholesterol + cholesterol + hydroperoxides

evidence

possibility

LIVER Cholw Esters

Cholesterol

our

hydroperoxides,

cholesterol

in

products

(21,

and Kukreja

cholesterol esters first experiment

---

autoxidation

atherosclerosis

ability

Some support

Cholesterol

that

initiating

products

was

calmodulin. of

likely

role

first

effect,

to inhibit

now

an important

protective

reports

is

19 30 11 24

11 f 5 15 * 7 3k2 99 f 44 163

+ 68

the

BIOCHEMICAL

Vol. 146, No. 3, 1987 mazine

and trifluoperazine,

enriched

in

reducing

blood

inhibitors is

cholesterol,

dependent

of

upon

essential

artery

wall

might for

this

exist

(LDL)

if

in which

the

their the

these

aortas

diets without

drugs

are

potent

recently

now

preparing

We are

of atherosclerosis

A2 plays

has

not

cholesterol

used

purified it

macrophages been

an

charac-

hydroperoxides

lipid

deposition were

in the

calmodulin-

(17).

material

that

wall

A2 involved

exits

shown

by aorta

to prevent

phospholipase

We have and

development

the

phospholipase

of

ability

some precedent

mixture

of

A2 involved

ability

hydroperoxide

(unpublished). on the

lipoprotein phospholipase

and

The cholesterol a complex mixture. from

consuming

in

Both

between

which

deposition

modification The

(16).

calmodulin

dependent,

to rabbits

concentrations.

low-density

A connection

inhibit

lipid

RESEARCH COMMUNICATIONS

(15).

oxidative

role

administered

reduce

cholesterol

uptake

terized. to

when

of calmodulin

The

AND BIOPHYSICAL

is

in

experiments

is

25-hydroperoxy-cholesterol

a potent

to

these

test

inhibitor this

of

material

for

calmodulin its

effect

in rabbits.

Acknowledgements Journal Experiment Graduate

Paper Station,

We wish paring

feed,

Iowa

to

J-12658

of

Iowa,

Project

Ames,

College,

Woltanski,

No.

thank

feeding Nicholas

State the the

Hogg,

the

Iowa

Agriculture

No. 2817.

Supported

and Home Economics in

part

by the

University. following rabbits

persons

for

and in tissue

Doug Johnson,

Jeff

their

sampling:

Seeling

assistance Bryan

in Miller,

preMike

and Sue North.

References (1)

(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

Peng, S.K., and Taylor, C.B. (1983) In Dietary Fats and Health (E.G. Perkins and W.J. Visek, eds.) pp. 919-933. American Oil Chemists’ Society, Champaign, Ill. Higley, N.A., Beery, J.T., Taylor, S.L., Porter, J.W., Dziuba, J.A., and Lolich, J.J. (1986) Atherosclerosis 62, 91-104. Smith, L.L. (1981) Cholesterol Autoxidation. Plenum Press, New York. van Lier, J.E., and Smith, L.L. (1970) J. Org. Chem. 35, 2627-2632. Teng, J. J., Kulig, M-J., Smith, L.L., Kan, G., and van Lier, J.E. (1973) J. Org. Chem. 38, 119-123. Smith, L.L., and Hill, F. L. (1972) J. Chromatogr. 66, 101-109. Pryor, W.A., and Castle, L. (1984) Methods Enzymol. 105, 293-299. and Eisfeld, W. (1958) Justus Liebigs Schenck, G.O., Neumuller, O.-A., Ann. Chem. 618, 202-210. Leung, P.C., Taylor, W.A., Wang, J.H., and Tipton, C.L. (1984) J. Biol. Chem. 259, 2742-2747. Holman, R.L., McGill, H.C. Jr., Strong, J. P., and Geer, J.C. (1958) Lab. Invest. 7, 42-47. Bligh, E.G., and Dyer, W.J. (1959) Can. J. Biochem. Physiol. 37, 911-917. (1984) Omodeo Sale, F., Marchesini, S., Fishman, P.H., and Berra, B. Anal. Biochem. 142, 347-350. 1171

Vol. 146, No. 3, 1987

(13) (14)

(15) (16) (17)

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

F.P., and Schaub, R.G. (1986) Arteriosclerosis 6, 42-48. Kaul, D., and Kukreja, R.S. (1987) Atherosclerosis 64, 211-214. Gietzen, K. (1986) in Intracellular Calcium Regulation (H. Bader, K. Gietzen, J. Rosenthal, R. Rudel, and H. U. Wolf, eds.) pp 405-423. Manchester University Press, Manchester, U.K. Parthasarathy, S., Steinbrecher, U.P., Barnett, J., Witzturn, J.L., and Steinberg, D. (1985) Proc. Natl. Acad. Sci. USA 82, 3000-3004. Moskowitz, N., Andres, A., Silva, W., Shapiro, L., Schook, W., and Puszkin, S. (1985) Arch. Biochem. Biophys. 241, 413-417. Bell,

1172