Lipid hydroperoxides potentiate mesenteric artery vasoconstrictor responses

Free Radical Biology & Medicine. Vol. 14, pp. 397-407, 1993

0891-5849/93 $6.00 + .00 Copyright~. 1993PergamonPressLtd.

Printed in the USA.All rights reserved.

Original Contribution LIPID

HYDROPEROXIDES

POTENTIATE

VASOCONSTRICTOR

MESENTERIC

ARTERY

RESPONSES

CARL A. HUBEL,* SANDRA T. DAVIDGE,t and MARGARET K. MCLAUGHLIN*t Departments of*Physiology and Biophysics and tpediatrics, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA (Received 29 July 1992: Revised l 0 November 1992"Accepted 13 November 1992)

Abstract--The aim of this study was to investigate the effects of lipid and organic hydroperoxides on vasomotor activity of isolated rat superior mesenteric arteries. Hydroperoxides did not elicit measurable responses in unstimulated (quiescent) mesenteric arteries. Contractile responses to potassium, however, were significantly potentiated by 13-(s)-hydroperoxylinoleicacid (range 3-54 uM). Potentiation of potassium responses by linoleic acid (18:2) and linolenic acid (18:3) w a s increased by pretreatment of the fatty acids with lipoxygenase (p < .01). Lipoxygenase alone had no contractile effects. Lipoxygenase-treated 18:2, left-butyl hydroperoxide, and hydrogen peroxide augmented contractile responses to phenylephrine, but to a lesser degree than corresponding augmentation of potassium responses. Contractile responses to lipoxygenase-treated 18:3 were blunted by vitamin E (p < .02) and by nitroblue tetrazolium (p < .02), whereas catalase and mannitol had no effects, implicating lipid free radicals in the contractile response. Responses to lipid hydroperoxides were not significantly altered by prostaglandin inhibitors. Endothelial cell denudation significantly enhanced the contractile responses elicited by 13-(s)-hydroperoxylinoleicacid (p < .05), indicating that lipid hydroperoxides enhance agonist-induced contractions by a direct effect on the smooth muscle. These results support a hypothesized link between lipid peroxidation and development of altered vascular function. They further suggest that the vascular endothelium may play an important role in regulation of vasomotor responses to lipid hydroperoxides. Keywords--Lipid hydroperoxides, Mesenteric arteries, Vascular endothelium, tert-Butyl hydroperoxide, Vascular smooth muscle. Vasoconstriction, Hydrogen peroxide, Free radicals

INTRODUCTION

hydroperoxides produces contractions in a variety of isolated and intact arterial, gut, uterine, and lung s m o o t h muscle preparationsJ °-14 Little i n f o r m a t i o n exists, however, c o n c e r n i n g the effects o f lipid hydroperoxides on v a s o m o t o r activity o f arteries from a circulation i m p o r t a n t to regulation o f peripheral vascular resistance. T h e present study, therefore, e x a m i n e d isometric tension responses o f the isolated rat mesenteric artery to a n u m b e r o f lipid and organic hydroperoxides.

In the presence o f free radical initiators and oxygen, unsaturated fatty acids undergo oxidative fragmentation by a chain reaction to form lipid hydroperoxides and a variety o f secondary metabolitesJ '2 Lipid hydroperoxides are formed in all organisms, and they are likely to be involved in a n u m b e r o f physiological processes. 2 A b n o r m a l l y high levels o f peroxidation products have been associated with cardiovascular diseases such as atherosclerosis 3'4 and preeclampsia, s'6 Experimental elevation o f lipid hydroperoxides in intact animals and in isolated blood vessel preparations can induce vascular endothelial cell exfoliation, reduced prostacyclin (PGI:) production, and other signs o f endothelial dysfunction. 7-9 Acute exposure to lipid

MATERIALS AND METHODS T h e experiments were performed primarily o n isolated superior mesenteric arteries taken from female 2 5 0 - 3 5 0 - g Sprague-Dawley rats (Charles River Breeders, St. Constant, Quebec). Resistance caliber mesenteric arteries ( 2 0 0 - 2 6 0 izm internal diameter) were used in a few instances to confirm that the responses o f the superior mesenteric arteries were characteristic o f the mesenteric circulation. Rats were sac-

This work was supported by National Heart, Lung, and Blood Institute Grant HL-40130. Address correspondence to: Carl A. Hubel, University of Cincinnati Medical Center, Departments of Physiology and Biophysics and Pediatrics, 231 Bethesda Avenue, Cincinnati, OH 45267-0576. 397

398

C.A. HUBEL et al.

rificed under ether anaesthesia by decapitation, and arteries were excised and placed in physiological salts solution (PSS) chilled to 4°C. Superior mesenteric arteries were cleaned of surrounding fat and connective tissue and cut transversely into two ring segments (1.8-2.2 mm length) under a dissecting microscope (Zeiss, Dedham, MA). Resistance mesenteric arteries were excised from a location immediately above the first branch of the mesenteric arcade. These vessels were treated in the same manner as the superior mesenteric arteries and cut into ring segments of 1.0-1.5 mm length. Segment lengths were measured using a calibrated filar eyepiece. Special care was taken in all cases to minimize damage to the endothelium.

Resistance vessel myograph Certain experiments used segments of resistance arteries mounted on two 32 um diameter tungsten wires in a myograph system designed by Mulvany and Halpern. ~s In brief, each wire was attached respectively to two stainless steel support blocks, one of which was attached to a force transducer and the other to a displacement device. After I h equilibration in PSS, a normalized internal circumference was determined using an exponential curve fit of passive tension versus diameter and the Law of LaPlace as previously described. Normalized vessels were equilibrated in laSS for 1 h prior to concentration-response curves.

Large vessel myograph

Hydroperoxides

Paired superior mesenteric rings were suspended from force-displacement strain gauges (Grass FT.03, Quincy, MA) for measurement of isometric tension (raN force/twice mm vessel wall length; m N / m m ) ) 5 Organ chambers housing the tings were filled with 20 mL PSS maintained at 37°C and gassed with a standard 95% 02-5% CO2 mixture through a fritted stone. Buffer pH (7.4-7.5) and gas partial pressures (pO2 = 480-580 mm Hg; pCO2 = 40-45 mm Hg) were monitored by a blood gas analyzer (Radiometer BM53 Micro Blood System, Westlake, OH). PSS had the following millimolar composition: NaCI (119); KCI (4.7); CaCI2 ( 1.6); KH2PO4 ( 1.18); MgSO4 ( 1.17); NaHCO3 (24.0); glucose (5.5). After 1 h equilibration in PSS, each ring was set to a normalized internal circumference for comparison of concentration-response characteristics between different vessels. Mesenteric arteries develop no intrinsic tone in PSS,~6 and thus the normalized internal circumference was adjusted in PSS with no exogenous relaxants. Normalization was determined using an exponential curve fit of passive tension versus diameter and the Law of LaPlace. ~5 The normalized internal circumference was defined as 0.8Lt0o, where L~oois an estimate of the internal circumference that a relaxed vessel would have when subjected to a transmural pressure of 100 mm Hg. Normalized vessels were equilibrated in PSS for 1 h before experimental protocols. Immediately before and after each experimental procedure, peak stable responses to maximally contracting potassium solution ( 125 mM KCI) were measured and used as an index of vessel stability. If responses to 125 mM KCI solution differed over time by > 20%, the vessel pair was eliminated from the data pool. (One hundred twenty-five millimolar KCI solution was PSS containing 5 mM Ca 2+ and adjusted to 125 mM K ÷ by equimolar substitution of NaCI by KCI.)

Experiments comparing the effect of hydroperoxides on rings with and without functional endothelium used 13-(s)-hydroperoxylinoleic acid (Oxford Biochemicals Research, Inc., Oxford, MI). The hydroperoxide was stored at -70°C and under N2; and working stocks were made into distilled deionized water less than 5 min prior to use. Other experiments used lipid hydroperoxides enzymatically generated by an adaptation of the method of Hamberg and Samuelsson.~7 In brief, 50 uL of unsaturated fatty acid substrate [linoleic acid (9,12-octadecadienoic acid; 18:2) or linolenic acid (9,12,15-octadecatrienoic acid; 18:3) (Nu Chek Prep, Inc., Elysian, MN)] was dissolved in an equal volume of ethanol and suspended by slow addition of 25 mL distilled deionized water with stirring. A 2-mL aliquot was then mixed with 0.95 mL of 0.2 M borate buffer solution (pH 9). Fifty uL of 1.4 × 105 IU/mL soybean lipoxygenase (linoleate:oxygen oxidoreductase; EC I. 13.11.12; Sigma Chemical Co., St. Louis, MO) in borate buffer was added to the resulting lipid suspension (4500 IU/mL final concentration), and solutions were incubated 10 min prior to use. Production of the lipid hydroperoxides was monitored spectrophotometrically using absorbance of diene conjugates at 234 nm with an average extinction coefficient of 27,500 M -~ cm -~ (Ref. 18). An iodometric technique for direct measurement of lipid hydroperoxides in solution was used to verify the diene conjugation assay. ~9 Other hydroperoxides tested were tert-butyl hydroperoxide (Sigma Chemical Co., St. Louis, MO) and hydrogen peroxide (Mallinckrodt, Paris, KY). Tertbutyl alcohol (Kodak, Rochester, NY) was used as a control for tert-butyl hydroperoxide. Hydroperoxides were tested under two different conditions, namely ( ! ) in the absence of other external stimuli (quiescent vessels), and (2) in vessels sub-

Response to hydroperoxides maximally precontracted to provide a baseline tone from which subsequent responses were measured. The second condition was achieved by flushing the bath of PSS solution and replacing with a potassium chloride-depolarizing solution (PSS made by equimolar replacement of NaCl by KC1). In some instances, vessels were activated using phenylephrine (L-phenylephrine hydrochloride, Sigma Chemical Co., St. Louis, MO). Appropriate concentrations were previously determined from complete concentration-response curves to the agonists. Hydroperoxides were cumulatively added to the baths once a stable baseline was achieved with either KCI or phenylephrine.

Antioxidants One IU/ml final concentration vitamin E (water solubilized dl-a-tocopherol; Carlson Laboratories, Arlington Heights, IL) and 0.3 mM final concentration nitroblue tetrazolium (Sigma Chemical Co., St. Louis, MO) were tested for ability to inhibit lipid hydroperoxide-mediated responses. Catalase (H2Oa:H202 oxidoreductase; EC 1.11.1.6; Sigma Chemical Co., St. Louis, MO) was added directly to vessel baths (500 IU/mL final concentration). Mannitol (Sigma Chemical Co., St. Louis, MO) was also tested for inhibition (I-10 raM). Vessels were given tone in all cases by continuous exposure to 30 mM KCI solution, resulting in an EC30(30% maximal) contraction.

Cyclooxygenase inhibition Effects of prostaglandin synthesis inhibitors on lipid hydroperoxide-induced responses were assessed using ibuprofen (from 50 mg/mL sodium ibuprofenate in 0.9% benzyl alcohol; Upjohn, Kalamazoo, MI) and indomethacin (from 10-3 M stock in 2% Na2CO3; Sigma Chemical Co., St. Louis, MO). Vessels were exposed to ibuprofen (1 x 10-5 M) or indomethacin (5 x 10-6 M and 1 x 10-s M) for45 min prior to and during treatment with hydroperoxide. Vessels were given tone by continuous exposure to 30 mM KCI solution (ibuprofen group) or 7.5 x 10-7 M phenylephrine (indomethacin group). Phenylephrine was used as the precontraction agonist in indomethacintreated vessels because the contractile response to KCI, but not phenylephrine, was significantly blunted by indomethacin alone.

Vascular endothelium Vessel rings with and without functional endothelium were used to determine the role of the vascular endothelium in the observed responses to hydroper-

399

oxides. After mounting, one ring of each pair was deendothelialized by gently rubbing the luminal surface with a tungsten rod. After l h equilibration in PSS, rings were given tone with 50 mM KCI solution to provide an ECT0 contraction baseline from which subsequent responses were measured. Functional integrity of the vascular endothelium was assessed by relaxation responses to the endothelium-dependent vasodilator methacholine ( I X l0 -6 M; methacholine chloride, Sigma Chemical Co., St. Louis, MO). Rubbed rings exhibiting less than 4% of the normal relaxation response to methacholine were considered suitable for use in the study. Following methacholine responses, the vessels were reequilibrated in PSS for l h and exposed a second time to 50 mM KCl solution. Following establishment of the contraction baseline, vessels were exposed to cumulative concentrations of 13-(s)-hydroperoxylinoleic acid. Following another reequilibration in PSS and precontraction to 50 mM KCI, endothelium-independent relaxation responses were assayed using ! x l0 -6 M sodium nitroprusside (Sigma Chemical Co., St. Louis, MO). The endothelial layer was subjected to silver staining for visualization of the endothelium by light microscopy (Zeiss, Dedham, MA) at the end of each experiment.

Data analysis Results from individual experiments were pooled to generate concentration-response curves (n refers to the number of different animals from which arteries were obtained for study). Data were plotted as the mean (+ SEM). Concentration-dependent treatment group differences were analyzed by repeated measures analysis of variance (ANOVA) of the paired data and, where validated, by Student-Newman-Keuls multiple range test on individual data points. A p value less than .05 was considered to be statistically significant. RESULTS

Lipid hydroperoxide responses Linole~c acid, linolenic acid, and their hydroperoxide derivatives augmented superior mesenteric artery contractile responses to exogenous agonists (potassium or phenylephrine). In the absence of exogenous agonists (in quiescent arteries), however, none of the lipids or their hydroperoxides elicited contractile responses. In superior mesenteric artery rings precontracted by potassium, the addition of 13-(s)-hydroperoxylinoleic acid resulted in markedly augmented contractile responses (Fig. 1). Contractions reached a plateau after approximately 5 min. The concentration of 13-(s)-hydroperoxylinoleic acid responsible for 10% of its maximal contractile effect (ECho) was 3.1

400

( . A. HUBEL e / a / 0.25 •

E E

0.20.

Z

E Z

0 01 Z I.iJ P Z

0.15.

0.10,

W Z

,< -1(,~

0.05 -

0.00

.

-6.3

.

.

.

.

i

-6.0

-5.7

13(S)-HYDROPEROXYUNOLEIC ACID (LOG M)

Fig. 1. Potentiation of KCl-induced contractions by 13-(s)-hydroperoxylinoleic acid. Superior mesenteric artery rings were exposed to 30 mM KCI solution throughout the protocol to provide a submaximal active tension baseline. Subsequent responses produced by exposure to cumulative concentrations of the lipid hydroperoxide are expressed as the change in tension from the baseline (mean +_SEM) vs. concentration ofhydroperoxide. Seven vessel pairs were

used.

uM (+0.7 uM, -0.5 uM) (geometric mean _+ SEM). The increase in isometric tension elicited by 20 uM of the hydroperoxide was 0.25 +__0.04 mN/mm, representing a 40% increase from baseline tension. In potassium precontracted superior mesenteric arteries, contractile responses to linoleic (18:2) and to linolenic acid (18:3) were significantly enhanced by prior treatment of the lipids with lipoxygenase (p < .01, repeated measures ANOVA). Lipoxygenase alone had no measurable contractile effects (Fig. 2B). Stearic acid (18:0) had no contractile effects in the presence or absence of lipoxygenase. Suspensions of lipoxygenated linoleic acid significantly potentiated contractile responses to phenylephrine in superior mesenteric arteries ( 12, 30, and 54 uM lipid; p < .05), whereas untreated linoleic acid had no detectable effect on phenylephrine-induced tone (data not shown).

Response to tert-butyl hydroperoxide In the presence of potassium (30 raM), tert-butyl hydroperoxide markedly potentiated contractile responses in superior mesenteric rings. In potassiumprecontracted arteries (Fig. 3), the ECho for tert-butyl hydroperoxide was 30 uM (+7.1 tzM, - 1.3 ~tM) (geometric mean _ SEM). The mean maximal response to 30 mM KCI + 2 mM tert-butyl hydroperoxide was within 10% of the average response to maximally con-

tracting potassium ( 125 mM KCI). The structural analog to tert-butyl hydroperoxide, tert-butyl alcohol, which possesses an alcohol group in place of the hydroperoxy group, had no contractile effects. Concentrations of ten-butyl hydroperoxide eliciting responses in resistance-size arteries (range 10-400 uM hydroperoxide; n = 6; r 2 = .90) were comparable to concentrations in superior mesenteric arteries. Tert-butyl hydroperoxide also significantly potentiated phenylephrine contractile responses. For example, 20 uM tert-butyl hydroperoxide potentiated the response to 7.5 × 10 -7 M phenylephrine by 22% (mN/ mm _+ SEM tension increase: 0.14 + 0.03: n = 5;p < .05). The maximal potentiation by tert-butyl hydroperoxide under condition of phenylephrine precontraction was, however, significantly less than under potassium precontraction (mN/mm _+ SEM: 0.3 _+ 0.04 in phenylephrine precontracted rings vs. 1.5 _+ 0. i in potassium precontracted rings; n = 5; p < .05).

Response to hydrogen peroxide Hydrogen peroxide had no contractile effects in quiescent mesenteric arteries. Figure 4, however, shows the response to graded concentrations of hydrogen peroxide in vessels under potassium-induced tone and the elimination of the response by treatment of hydrogen peroxide solutions with catalase. Catalase alone had no significant effects on baseline tone. Hydrogen peroxide also potentiated contractile responses to phenylephrine (data not shown). The maximal tension increase attributable to H202 under phenylephrine-induced tone was significantly less than maximal increases under potassium-induced tone (mN/mm __+SEM:0.2 _+ 0.05 in phenylephrine precontracted rings; 0.65 _+ 0.1 in potassium precontracted rings; n = 4; p < .05).

Dependence on preexisting tone Responses to hydroperoxides reached a maximum in 3 to 5 min and decayed slowly over time (hi2 >-- 30 min). Following removal of the hydroperoxide, normal responses to KC1 or phenylephrine were restored. The magnitude of contractions to hydroperoxides depended on the level of preexisting tone. Hydroperoxides elicited responses only when the contraction threshold to KCI or phenylephrine was exceeded. Responses to a given concentration of hydroperoxide were significantly attenuated in maximally activating solutions (125 mM KC! or 1 × 10-5 M phenylephfine) compared to submaximally activating solutions. High concentrations of hydroperoxide (i.e., > 100 uM H202 or > 1 mM tert-butyl hydroperoxide) com-

Response to hydroperoxides

E ,.0 E

T 0--0

z v

E

0.8

u,mua¢ Am ('L)

0 - - • ~

~.

T/~

u~m~st

(4)

T/l

z

0 (/3 z

r/1

oe

la.J I.0.4 z

T/

z I

,ZL/

0 "

O--0

0.8

urn.me ~

0~0 -5.0

(",j /

1.a

.

0.2

1

~ .L

.

(=)

O.S

1 r/l

/

.

.

.

.

.

!

-4.0

LINOLEIC ACID (LOG M)

o~O

~._e

/ 0.0

(m)

• u • umumE Am • ~

_r T

/•

0.2 q

1.0

0.4

/1

i

lad

"1"

•

401

o.o ~ ~ - - - ~

..o--6

. .~.~_~.zx

-5.0

-4.0

1.b LINOLENIC ACID (LOG M)

Fig. 2. Effect of lipoxygenase on linoleic acid (la) and linolenic acid (lb) potentiation of KCl-induced responses. Superior mesenteric arterT rings were exposed to 30 mM KCI solution throughout the protocol to provide a submaximal active tension baseline. Subsequent responses produced by exposure to cumulative concentrations of lipoxygenase-pretreatedlipid, untreated lipid, and lipoxygenasealone are expressed as the change in tension from the baseline (mean + SEM) vs. lipid concentration. Numbers in the figure indicate the number of vessel pairs used. Contractile responses to lipoxygenatedlipid were significantly greater than responses to untreated lipid (p < .01).

monly produced biphasic responses visualized as a contraction followed by sustained relaxation (data not shown).

mannitol, a hydroxyl radical ( ' O H ) scavenger, also did not inhibit responses (data not shown).

Effects of prostaglandin inhibitors Antioxidant effects Responses to lipoxygenated 18:3 (30 and 54 uM) in the potassium-precontracted superior mesenteric artery were reduced over 60% (p < .02; paired t test) when tested in the presence of 0.3 m M nitroblue tetrazolium (Fig. 5). Nitroblue tetrazolium alone did not alter baseline responses to potassium. The effect of vitamin E on contractile responses to lipoxygenated 18:3 in potassium-precontractied superior mesenteric arteries is shown in Fig. 6. Vitamin E was inhibitory (p < .02; repeated measures ANOVA) when tested using an N2/O2/CO2 gas mixing system (Matheson, Gloucester, MA) to lower the organ bath oxygen partial pressure (pO2 = 120-150 m m Hg; 19(202 = 38-50 m m Hg; pH = 7.35-7.45). Vitamin E did not inhibit responses of vessels gassed with 95% O d 5 % CO2. Catalase (500 I U / m L ) did not inhibit contractions by lipoxygenated 18:2 in potassium-precontracted arteries (data not shown). Under identical conditions, catalase eliminated responses to exogenously applied H202 (Fig. 5). Incubation of the vessels with 1-10 m M

Enhancement of contractile tone by iipoxygenated 18:2 was not significantly altered by exposure of vessels to ibuprofen (Fig. 7) or indomethacin (data not shown).

Vascular endothelium As shown in Fig. 8, the e n h a n c e m e n t of KCI contraction by 13-(s)-hydroperoxylinoleic acid was significantly greater in endothelium-denuded than in endothelium-intact segments (p < .05, repeated measures ANOVA). Initial baseline responses to 50 m M KCI were not significantly different between the two groups ( m N / m m __ SEM:I.6 _+0.35 without endothelium; 1.2 _+ 0.15 with endothelium; p < .2). Endothelium-intact rings displayed relaxation responses to the endothelium-dependent vasodilator methacholine (% of possible relaxation _+ SEM:58% +_ 4.9). This degree of relaxation is similar to previously reported values for vessels under potassium-induced tone. 2° Relaxation responses to methacholine were inhibited in endothelium-denuded rings (% total possible relaxation

402

('.

A.

HUBEL

C/ a t ' .

T

1.5-

E E

•-

•

0--0

,-u~wn.~

z o

(s)

/

1.0-

(/)

z

hi I--

T/i r/I

z Lid

c~ z -to

T/

(0)

T--nw~m ~ m ~ a , ~

z

E v

DISCUSSION

61 •

0.5

./7 ./O 0.0 -5.0

-4.0

-3.0

T-BUTYL COMPOUND

(M)

Fig. 3. Tert-butyl hydroperoxide potentiation of KCI-induced contractions. Superior mesenteric artery tings were exposed to KCI (30 mM) to provide a submaximal active tension baseline. Responses to tert-butyl hydroperoxide or tert-butyl alcohol are expressed as the change in tension from the baseline (mean +_ SEM) vs. tert-butyl concentration. Numbers in the figure indicate the number of vessel pairs used.

+_SEM:0.7% +_ 2.2). Silver staining confirmed the absence of endothelium in rubbed segments and the presence of endothelium (> 90% of intimal surface) in unrubbed segments. Endothelium-independent relaxation responses to sodium nitroprusside were not significantly different between endothelium-denuded and endothelium-intact segments (% of possible relaxation + SEM:96 + t.5 without endothelium; 84 + 4.2 with endothelium; p < .2).

Exposure of isolated mesenteric arteries to lipid hydroperoxides induced an increase in active tension in arteries precontracted by potassium and, to a lesser extent, by phenylephrine. Responses were also potentiated by unsaturated fatty acids, but significantly less than the same fatty acids containing hydroperoxide derivatives. In quiescent arteries, none ofthe lipids or their hydroperoxides elicited potentiation responses. It is apparent that the hydroperoxy group is the primary determinant for vasoactivity because responses were evoked by hydroperoxide compounds of varying hydrocarbon chain length, whereas they were nonexistent or markedly decreased in the absence of the hydroperoxy group. Responses to 13-(s)-hydroperoxylinoleic acid were invoked at low micromolar concentrations (< 5 uM) that are close to tissue hydroperoxide levels in certain disease states. ~ Estimation of lipid hydroperoxide concentrations in lipoxygenase-treated lipid suspensions suggest a concentration-response profile similar to 13-(s)-hydroperoxylinoleic acid. It is possible that some of the augmentation of contractile responses ascribed to lipoxygenase-generated lipid peroxides may be due to contaminants in the fatty acid preparation such as arachidonic acid or its lipoxygenase derivatives. Responses to lipoxygenase-treated lipid suspensions were, however, qualitatively similar to the other hydroperoxides tested and were significantly inhibited by antioxidants. The contractile effect of micromolar hydroperoxide concentrations (i.e., < 100 uM H202 or < 400 uM tert-butyl hydroperoxide) is a reversible phenomenon; the normal activation to KC! or phenylephrine

0.8'

EE z E

---, z

0.7.

0.6.

0.4.

z

0.3.

z

0.2

LIJ

o--o

*

1to -°

0.1

0.0

o

c,)

0.5,

_o

(g)

- IT

o-o

-0.1

-

. . . .

,

-5

.

.

.

.

.

.

,

-4 HYDROGEN PEROXIDE (LOG M)

Fig. 4. Potentiation of KCl-induced contractions by hydrogen peroxide in superior mesenteric arteries. Responses are expressed as the change in tension from the original tension baseline provided by 30 mM KC1 solution. Hydrogen peroxide was administered following pretreatment of stock solutions with (n = 4) or without (n = 5) 500 IU/mL catalase.

Response to hydroperoxides

403

0.30

E E

I

0.25

~] m

Z

E ~,

.,~,.,,~d (S) N~-W,w,k~

(s)

0.20

Z 0 (/') Z

0.15

i'-Z

0.10

I.iJ (3 Z ,<

0.05

"10 0.00

50/zM

54 p,M

CONCENTRATION LIPOXYGENATED LINOLENIC ACID Fig. 5. Effect of nitroblue tetrazolium (NBT) on the response to lipoxygenated linolenic acid. Superior mesenteric artery ring segments were activated by exposure to 30 m M KCI solution in the presence or absence of 0.3 m M NBT. Subsequent responses are expressed as the change in tension from the baseline (mean _+SEM) vs. lipoxygenated linolenic acid concentration (30 and 54 gM). Vessel pairs from five rats were used. * = p < .02 vs. NBT-untreated control.

was restored following removal of the hydroperoxide. Higher concentrations of hydroperoxides occasionally produced biphasic responses visualized as atran-

A

E E

0--0

cwmm.s(e)

l--O

,reAl,t n ( - ~

sient contraction followed by a sustained relaxation (data not shown). This was observed more frequently in phenylephrine precontracted arteries. Following

(0)

oil

0.6 "'"

~se

Z v

E

Z 0 (/I

0.4

i.i IZ i.i (.~

l/'

*

z

0

/I

l

0

°

tJl//l

8

0.2

Z

,4( 3: (J

0.0

10

,-e -5.0

.

.

t/

I/?

.

.

.

.

.

.

-4.0

LIPOXYGENATED LINOLENIC ACID (LOG M) Fig. 6. Effect o f vitamin E on the response to lipoxygenated linolenic acid. Superior mesenteric artery ring segments were activated by exposure to 30 m M KCI solution in the presence or absence of 1 I U / m L d-alpha-tocopherol. Responses are expressed as the change in tension from the baseline (mean _+ SEM) vs. lipoxygenated linolenic acid concentration. Vessel pairs from six rats were used. * = p < .05; ** = p < .03; *** = p < .01.

C.A. HUBELcta/.

404

0.8

E E z V

E

z 0 (/1 z IM I-

0 - - 0 ~lmnm~ (4) / / I

0.6

•--

•

~

(4)

f/~

o.,,

Z w

O Z

0.2

-r O

o o

-5.0

-4.0

LIPOXYGENATED LINOLEIC ACID (LOG M) Fig. 7. Lack of effect of ibuprofen on response to lipoxygenated linoleic acid. Ibuprofen (1 × 10-s M) was added to the bath 30 rain prior to and during test. Artery ring segments were exposed to 30 mM KCI solution throughout the protocol to provide a submaximal active tension baseline. Responses to lipoxygenated linoleic acid are expressed as change in tension (mean _+SEM) from the baseline vs. lipoxygenated linoleic acid concentration. Vessel pairs from four rats were used.

exposure to hydroperoxide concentrations producing relaxation, subsequent contractile responses to KCl or phenylephrine were significantly inhibited (data not shown). Further studies will be necessary to clarify the mechanism of biphasic responses. The reason for the greater enhancement of potassium compared to phenylephrine responses by hydroperoxides has not been determined. It has been shown, however, that lipid peroxides can stimulate calcium influx through voltage-operated channels in endothelial cells. 2~The possibility that lipid peroxides enhance L-type Ca 2÷ channel activity in vascular smooth muscle may warrant analysis. A number of studies have demonstrated that lipid hydroperoxides produce direct contractions in a variety of isolated and intact arteries. For example, Asano and Hidaka, ~° Sasaki et al., 12 and Koide et al. 22 have reported that lipid peroxides are potent vasoconstrictors of rabbit aortic strips, intact canine basilar arteries, and isolated canine basilar arteries, respectively. Considerable heterogeneity apparently exists, however, in the response of arteries to hydroperoxides and other oxygen-derived free radicals. 23-26 Recent studies, for example, have presented evidence for vasodilator effects of hydrogen peroxide25 and cumene hydroperoxide. 26 The response observed may differ depending on species, the type of blood vessel studied, and the type of hydroperoxide analyzed. In the present study using isolated rat mesenteric arteries, con-

tractions to submaximal concentrations of agonists (KCI, phenylephrine) were potentiated by lipid hydroperoxides, hydrogen peroxide, and tert-butyl hydroperoxide. These hydroperoxides had no effects in quiescent mesenteric arteries. Rhoades et al. 23 similarly observed that rat isolated pulmonary arteries exhibited contractile effects to low doses of H202 (< l0 -4 M) only when the smooth muscle was "primed" by precontraction with submaximal concentrations of agonist (KCI or 5-hydroxytryptamine). The oxidation of low density lipoproteins (LDL) by metal ions or cells in vitro has multiple effects, including formation of lipid peroxidation products. 27'2a Recently, Simon et al. 29have provided evidence that contractions to oxidatively modified LDL (ox-LDL) in isolated pig coronary arteries were mediated by a hydroperoxide and were dependent on preexisting tone provided by potassium or the thromboxane mimetic U46619.

Unsaturated fatty acids It is noteworthy that the unsaturated fatty acids elicited contractile responses (Fig. 2), aibiet significantly less in the absence of lipoxygenase. Contractions to the fatty acids may have been due to their oxidation in the organ chamber, although hydroperoxides were undetectable by iodometric and diene conjugation colorimetry. Formation of a common

Response to hydroperoxides 0.4

T|

E

E z

E

0 u1 z LIJ I-Z i.i 0 Z "r (.3

0--0 0--0

~ ~

(4) t4)

T

o.3

0.2

T/-o--o

a 0.1

0.0

. --6°3

.

.

.

.

f --6°0

--5°7

13(S)-HYDROPEROXYLINOLEIC ACID (LOG M)

Fig. 8. Effect of 13-(sFhydroperoxylinoleic acid in superior mesenteric arteries with and without vascular endothelium. One ring of each pair was subjected to endothelial cell denudation by rubbing the luminal surface with a tungsten rod. After I h equilibration in physiologic salts solution, rings were contracted with 50 mM KCI solution (ECTo contraction). Responses to 13-(s)-hydroperoxylinoleic acid are expressed as change in tension (mean _+SEM) from the baseline vs. concentration of the lipid hydroperoxide. Contractile responses to 13-(s}-hydroperoxylinoleic acid were significantly greater in endothelium denuded rings (closed circles) than in unrubbed rings (open circles) (p < .01 ). Four ring pairs were used per group. Endothelial integrity was assessed using endotheliumdependent and -independent vasodilators and by silver staining (see text for details).

prostaglandin metabolite as mediator of the vascular response to 18:2 and 18:3 is unlikely because 18:3 (alpha-linolenic acid) belongs to the omega-3 class of unsaturated fatty acids. This family of fatty acids cannot be converted to arachidonic acid metabolites and are competitive inhibitors ofarachidonate prostaglandin synthesis pathways. 3°

Antioxidant effects Vitamin E is thought to act by trapping lipid peroxy-radical (LOO") and lipid radical (L") species, thus breaking the peroxidation chain. 2 The blunting effect of vitamin E on the potentiation response to lipoxygenated 18:3 (Fig. 7) supports the notion that hydroperoxide species are the mediators of vasoactivity. The inhibitory effect was evident at a bath POE of 120-150 mm Hg but not at 480-580 mm Hg. The reason for the discrepancy is unknown but may involve an increased propensity for oxidative reactions, possibly including oxidation of vitamin E, at the higher pO2. The possibility was tested that hydrogen peroxide was an active contaminant or metabolite during hy-

405

droperoxide exposure. Catalase is a reductor of hydrogen peroxide and will not degrade lipid hydroperoxides; 3~ nor do lipid hydroperoxides inactivate catalase? 2 Catalase had no effect on the response to lipoxygenase-treated 18:2 in potassium-precontracted mesenteric arteries. Under identical conditions, catalase eliminated the response to exogenously applied hydrogen peroxide. The response to lipoxygenated lipid, therefore, cannot be accounted for by hydrogen peroxide. Hydroxyl radical ("OH) is also not likely to be a mediator of contractile responses because the addition of up to 10 mM mannitol had no inhibitory, effects. Nitroblue tetrazolium can scavenge a variety of radical species, including lipid radicals. 33 This agent significantly inhibited contractile responses to lipoxygenated 18:3. In addition to its ability to scavenge unsaturated fatty acid radicals, nitroblue tetrazolium is a superoxide anion (O~-) scavenger. Although our evidence suggests that lipid hydroperoxides are the mediators of vasoconstriction, it is possible that superoxide anion generated as a byproduct of peroxidation reactions may be involved in the response.

Effects of prostaglandin inhibitors Regulation of prostaglandin synthesis has been cited as a physiological role for lipid peroxidation in normal cells. 8'3° Lipid hydroperoxides can either enhance or inhibit prostaglandin synthesis depending on their concentrations and other conditions. For example, lipid hydroperoxides could promote vasoconstriction by their ability to inhibit prostacyclin (PGI2) synthetase?° In our study, incubation ofsuperior ruesenteric arteries with cyciooxygenase inhibitors (ibuprofen, cyclooxygenase) failed to blunt the contractile effects of lipid hydroperoxide. These observations suggest that the vasomotor activity of lipid hydroperoxides in isolated mesenteric arteries is not related to modulation of prostaglandin synthesis. This conclusion is consistent with data showing that hydroperoxide-induced contractions in aortic strips and in basilar artery tings are not affected by indomethacin or aspirin. 1o

Vascular endothelium Contractile responses elicited by 13-(s)-hydroperoxylinoleic acid in artery rings provided with a potassium-induced baseline tone were significantly enhanced by mechanical removal of the vascular endothelium. No significant differences in smooth muscle vasomotor function were observed between endothelium-intact and -denuded artery segments, as evi-

406

('. A. i-tUBELet al.

denced by equivalent contractile responses to KCI and relaxation responses to the endothelium-independent vasodilator sodium nitroprusside. These results indicate that contractile responses to lipid hydroperoxides in mesenteric arteries occur by direct stimulation of vascular smooth muscle and that the responses are inhibited by the vascular endothelium. The mechanism by which the endothelium decreases hydroperoxide-mediated responses remains to be elucidated, but it may involve basal release of endothelium-derived relaxing factor (EDRF). CONCLUSIONS

The mesenteric bed is a circulation important to regulation of peripheral vascular resistance. ~6 The present work demonstrates that lipid hydroperoxides potentiate contractile responses to potassium and phenylephrine in superior mesenteric artery rings. Response potentiation by t-butyl hydroperoxide observed in small (resistance size) mesenteric arteries suggests that the responses of the superior mesenteric arteries are characteristic of the mesenteric circulation in general. These results, therefore, indicate that lipid hydroperoxides are potential modifiers of peripheral vascular resistance. Contractile potentiation effects of lipid hydroperoxides were significantly enhanced by the absence of vascular endothelium. This suggests that the vascular endothelium may play an important role in regulation of vasomotor responses to lipid hydroperoxides. We have previously shown that increased lipid peroxidation, mediated by dietary vitamin E deprivation, is associated with enhanced vascular contractile responsiveness in the intact female rat and decreased relaxation responsiveness in its isolated mesenteric arteries. 7 These changes were observed in conjunction with abnormal endothelial cell morphology, supporting a hypothesized link between peroxidative stress and development of altered vascular function in v i v o . 7 Recent studies in both animals and humans indicate a possible role for lipid peroxidation in the development of atherosclerosis. 3435 ' ' Additionally, the elevated blood pressure and enhanced responsiveness to vasopressor agonists characterizing human preeclampsia is associated with elevated plasma lipid peroxidation products, endothelial damage, and decreased sera antioxidant activity. 5,6,34,36 Our study supports the hypothesis that peroxidative reactions can contribute to altered vascular function in such disorders. Acknowledgement - - We wish to thank Rollande Morrison and Nada Huron Gorman for their excellent assistance in preparing the manuscript.

REFERENCES 1. Gutteridge, J. M.; Halliwell, B. The measurement and mechanism of lipid peroxidation in biological systems. Trends Biochem Sci. 15940:129-135; 1990. 2. Kagan, V. E., ed. Lipid peroxidation in biomembranes. Boca Raton, FL: CRC Press: 1988. 3. Ledwozoyw, A.: Michalak, J.: Stephen, A.; Kadziolka, A. The relationship between plasma triglycerides, cholesterol, total lipids and lipid peroxidation products during human atherosclerosis. Clin. Chim. Acta 155:275-284; 1986. 4. Steinberg, D.: Parthasarathy, S.: Carew, T. E.; Khoo, J. C.: Witzturn, J. L. Beyond cholesterol. Modifications of low density lipoprotein that increase its atherogenicity. N. Engl. J. Meal 320:915-924:1989. 5. Hubel. C. A.: Roberts, J. M.; Taylor, R. N.; Musci, T. J.: Rogers, G. M.: McLaughlin, M. K. Lipid peroxidation in pregnancy: New perspectives on preeclampsia. Am. J. Obstet. G.vnecol. 161:1025-1034: 1989. 6. Wang, Y.; Walsh, S. W.: Guo, J.: Zhang, J. The imbalance between thromboxane and prostacyclin in preeclampsia is associated with an imbalance between lipid peroxides and vitamin E in maternal blood. Am. J. Obstet. G.vnecol. 165:16951700; 1991. 7. Hubel, C. A.: Gffggs, K. C.; McLaughlin, M. K. Lipid peroxidation and altered vascular function in vitamin E-deficient rats. Am. J. Physiol. 256:HI539-HI545; 1989. 8. Bunting, S.: Moncada, S.; Vane, J. R. The prostacyclin-thromboxane A2 balance: Pathophysiological and therapeutic implications. Br. Meal Bull. 29:27 t-276; 1983. 9. Hirai, S.: Okamoto, K.: Morimatsu, M. Lipid peroxide in the aging process. In: Yagi, K., ed. Lipid peroxides in biology and medicine. Orlando, FL: Academic Press; 1982:305-316. 10. Asano. M.: Hidaka, H. Contractile response of isolated rabbit aortic strips to unsaturated fatty acid peroxides. J. Pharm. Exp. "['her. 208:347-353: 1979. 1 I. Gurtner, G. H.: Knoblauch, A.: Smith, P. L.: Sies, H.; Adkinson, N. F. Oxidant- and lipid-induced pulmonary vasoconstriction mediated by arachidonic acid metabolites. J. Appl. Physiol. 55:949-954: 1983. 12. Sasaki, T.: Wakai. S.: Asano. T.: Watanabi. T.: Kirino, T.: Sano, K. The effect of a lipid hydroperoxide of arachidonic acid on the canine basilar artery. J. Neurosurg 54:357-365" 1981. 13. Stewart, R. M.; Weir, E. K.: Montgomery, M. R.; Niewoehner, D. E. Hydrogen peroxide contracts airway smooth muscle: A possible endogenous mechanism. Respir. Physiol. 45:333-343: 1981. 14. Cherouny, P. H.; Ghodganokar, R. B.: Gunner, G. H.; Dubin, N. H. The effect of the antioxidant, butylated hydroxyanisol, on peroxide-induced and spontaneous activity of the uterus from the pregnant rat. Biol. Reprod 40:98-103; 1989. 15. Mulvany, M. J.- Halpern. W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ. Re.s. 41:19-26; 1977. 16. McLaughlin, M. K.; Keve, T. M. Pregnancy-induced changes in resistance blood vessels. Am. J. Obstet. Gynecol. 155:12961299; 1986. 17. Hamberg, M.; Samuelsson, B. On the specificity of the oxygenation of unsaturated fatty acids catalyzed by soybean lipoxydase. J. Biol. Chem. 242:5329-5335; 1967. 18. Pryor, W. A.: Castle, L. Chemical methods for the detection of lipid hydroperoxides. In: Parker, L., ed. Methods in Enzymology. New York: Academic Press; 1984:293-299. 19. Hicks, M.; Gebicki, J. M. A spectrophotometric method for the determination of lipid hydroperoxides. Anal. Biochem 99:249-253; 1979. 20. Furchgott, R. F. Role ofendothelium in responses of vascular smooth muscle. (Trc. Res. 53:557-573; 1983. 21. Hirosumi, J.; Ouchi, Y.: Watanabe, M.; Kusunoki, J.; Nakamuff, T.; Offmo, H. Effect ofsuperoxide and lipid peroxide on cytosolic free calcium concentrations in cultured pig aortic en-

Response to hydroperoxides

22.

23.

24.

25. 26.

27. 28.

dothelial cells. Biochem. Biophys. Res. Cornmun. 152:301307; 1988. Koide, H.: Neichi, T.; Takato, M.; Matsuchita, H.: Sugioka, K.; Nakano, M.; Hata, S. Possible mechanisms of 15-hydroperoxy arachidonic acid-induced contraction of the canine basilar artery in vitro. J. Pharm. Exp. Ther. 221:481-488: 1982. Rhoades, R. A.: Packer, C. S.; Roepke, D. A.; Jin, N.: Meiss, R. A. Reactive oxygen species alter contractile properties of pulmonary, arterial smooth muscle. Can. J. Ph):~iol. Pharmacol. 68:1581 - 1589:1990. Auch-Schwelk, W.; Katusic, Z. S.; Vanhoutte, P. M. Contractions to oxygen-derived free radicals are augmented in aorta of the spontaneously hypertensive rat. Hypertension 13:859-864: 1989. Burke, T. M.; Wolin, M. S. Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation..'Ira. J. Physiol. 252:H721-H732:1987. van der Kraaij, A. M. M.: de Jonge, H. R.; Esterbauer, H.; de Vente, J.; Steinbusch, H. W. M.: Koster, J. F. Cumene hydroperoxide, an agent inducing lipid peroxidation, and 4-hydroxy2,3-nonenal, a peroxidation product, cause coronary vasodilitation in perfused rat hearts by a cyclic nucleotide independent mechanism. Cardiovasc. Res. 24:144-150; 1990. Rosenfeld, M. E. Oxidized LDL affects multiple atherogenic cellular responses. (~'irculation 83:2137-2140; 1991. l'homas, C. E.: Jackson. R. L. Lipid hydroperoxide involvement in cooper-dependent and independent oxidation of low density lipoproteins. J. Pharm Exp. Ther. 256:1182-1188: 1991.

407

29. Simon, B. C.: Cunningham, L. D.; Cohen, R. A. Oxidized low density lipoproteins cause contraction and inhibit endothelium-dependent relaxation in the pig coronary, artery. J. Clin Invest. 86:75-79; 1990. 30. Lands, W. E. M.: Kulmacz, R. J.; Marshall, P. J. Lipid peroxide actions in the regulation of prostaglandin biosynthesis. In: Pryor, W. A., ed. Free radicals in biology. New York: Academic Press; 1984:39-61. 31. Sies, H. Hydroperoxides and thiol oxidants in the study of oxidative stress in intact cells and organs. In: Sies, H., ed. Oxidative stress. London: Academic Press; 1985:73-90. 32. Arrick, B. A.: Nathan, C.; Gritfith, O. W.; Cohn, Z. A. Glutathione depletion sensitizes tumor cells to oxidative cytolysis. J. Biol. Chem. 257:1231-1237; 1982. 33. Peterson, D. A.; Gerrard, J. M.; Rao. G. H. R.; Krick, T. P.; White, J. G. Ferrous iron mediated oxidation of arachidonic acid: Studies employing nitroblue tetrazolium (NBT). Prostaglandins Meal !:304-317; 1978. 34. Roberts, J. M.; Taylor, R. N.; Musci, T. J.; Rodgers, G. M.; Hubel, C. A.; McLaughlin, M. K. Preeclampsia: An endothelial cell disorder. Am. J. Ohstet. Gynecol. 161:1200-1204; 1989. 35. Regnstrom, J.: Nilsson, J.; Tornvall, P.; Landou, C.; Hamsten, A. Susceptibility to low-density lipoprotein oxidation and coronary atherosclerosis in man. Lancet 339:1183-1186; 1992. 36. Davidge, S. T.: Hubel, C. A.; Brayden, R. N.; Capeless, E. C.; McLaughlin, M. K. Sera antioxidant activity in uncomplicated and preeclamptic pregnancies. Ohstet. G.vnecol. 79:897-901: 1992.