Free Radical Biology & Medicine, Vol. 10, pp. 93-100, 1991 Printed in the USA. All rights reserved.

0891-5849/91 $3,00 + .00 Copyright © 1991 Pergamon Press plc

Original Contribution HYDROPEROXIDE-INITIATED CHEMILUMINESCENCE: AN ASSAY FOR OXIDATIVE STRESS IN BIOPSIES OF HEART, LIVER, AND MUSCLE BEATRIZ GONZALEZ FLECHA,* SUSANA LLESUY, a n d ALBERTO BOVF-~S Institute of Biochemistry and Biophysics, School of Pharmacy and Biochemistry, University of Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina (Received 5 June 1990; Revised and Accepted 13 September 1990) Abstract--Hydroperoxide-initiated chemiluminescence was standardized as a microassay to evaluate the occurrence of oxidative stress in human biopsies. Samples of 10 to 50 mg of rat liver or heart were homogenized, diluted in reaction medium, added with tert-butyl hydroperoxide, and assayed for chemiluminescence in a liquid scintilation counter in the out-of-coincidence mode. Optimal conditions for the assay were: 0.3 to 1.2 mg/mL of homogenate protein in 120 mM KC1, 30 mM phosphate buffer (pH 7.4), and 3 mM tert-butyl hydroperoxide at 30°C. In these conditions, maximal chemiluminescence values were 550 ± 30 and 1100 ± 40 cps/mg protein, for liver and heart homogenates, respectively. Liver and heart homogenates were subjected to in vitro oxidative stresses such as supplementation with organic hydroperoxide or with enzymatic systems generating superoxide anion or hydrogen peroxide. Chemiluminescence was higher in the poststress samples than in the control ones. The ratio: poststress chemiluminescence/ control chemiluminescence (B/A) was about 1.4 or higher for both tissues. Human heart biopsies were utilized to investigate the occurrence of oxidative stress after clinical situations associated to ischemia-reperfusion. BIA ratios were 2.1 ± 0.4, 1.4 -+ 0.1, and 2.8 ± 0.4 for human heart, liver, and skeletal muscle, respectively. Keywords--Hydroperoxide-initiated chemiluminescence, Oxidative stress, Biopsies, Oxygen free radicals

INTRODUCTION

product of the reduction of organic hydroperoxides and hydrogen peroxide, seems restricted to red blood cells and liver, 5 in the latter case, with the possibility of measuring it in the bile. Organ chemiluminescence, has the determination of the fotons produced in the termination reactions of peroxyl radicals, is organ specific and nondestructive with respect to the organ. 6'7 However, glutathione release into the bile and organ chemiluminescence imply surgery and exposition of the organs. The determination of tert-butyl hydroperoxide-initiated chemiluminescence appears as a sensitive assay that has been applied to detect the existence of oxidative stress associated to experimental pathological situations such as chronic ethanol treatment in rats, s barbital treatment in mice, 9 tumor-bearing in mice, 1° and adriamycin cardiotoxicity in rats, mice, and rabbits 11A2 In this paper, we report the standardization of these techniques as micro-assay to facilitate its proper utilization in human biopsies. The assay, adapted as will be described here, has been used to detect oxidative stress in human heart during revascularization surgery, 13 in human liver during tumor resection surgery and in human muscle during ischemia-reperfusion cycles, as well as the prevention of heart oxidative stress by the addition of mannito114 and desferoxamine in the cardiople-

There is ample evidence indicating the participation of oxygen free radicals in several human pathologies as reviewed by Halliwell 1 and Chance et al. 2 Increased steady-state levels of oxyradicals and of alcoxyl and peroxyl radicals in the tissues are understood as an oxidative stress due to the damaging potential of these species) However, the determination of occurrence of oxidative stress in human tissues has several limitations due to the lack of assays being simultaneously sensitive and specific. 1 There are some techniques such as: a) hydrocarbon gas exhalation, 4 b)release of oxidized glutathione, 5 and c) organ chemiluminescence6'7 which seem sensitive and specific enough to be applicable to physiological conditions. These assays are based upon the determination of stable products or byproducts of the reaction chain of lipid peroxidation, a process that follows to the production of oxygen free radicals in intact biological systems. 2 Hydrocarbon (ethane, propane, etc.) exhalation implies a whole body determination that lacks organ specificity and it is subjected to interference. 4 Release of oxidized glutathione, a metabolic *Author to whom correspondence should be addressed. 93

B. GOSrZAt~ZFI~CV.Aet al.

94

gic solution and by the previous treatment of the patients with vitamins A and E. MATERIALS AND METHODS

Tissue biopsies Full thickness needle biopsies of human heart, liver, or muscle (10-50 mg) were obtained, immediately immersed in saline solution, and frozen.

action medium consisting of 120 mM KC1, 30 mM phosphate (pH 7.4) added with either tert-butyl hydroperoxide (0-500 ixM), or 5 mM glucose and glucose oxidase (0-30 mU/mL), or 500 txM xanthine and xanthine oxidase (0-10 mU/mL). After incubation of 30 rain, for the case of tert-butyl hydroperoxide, or of 10 rain for the case of the two enzymatic systems, samples were taken, protein concentration was adjusted to 0.5 mg/mL with reaction medium at 30°C, and 3 mM tertbutyl hydroperoxide was added to determine chemiluminescence.

Animals Male wistar rats of 150-180 g fed with a conventional laboratory diet and water ad libitum were used.

Protein determination

Tissue homogenates

Protein was measured by the Lowry et al. method 15 using bovine serum albumin as standard.

Samples of 10 to 50 mg of wet weight were homogenized in 2 mL of 120 mM KC1, 30 mM phosphate buffer (pH 7.4) at 0-4°C and centrifuged at 600g for 10 min at 0--4°C. The pellet was discarded and the supernatant was used as homogenate, s

Tert-butyl hydroperoxide-initiated chemiluminescence It was measured in a Packard Tri-Carb model 3320 or in a Beckman LS 100 liquid scintilation counter set in the out-of-coinicidence mode and in the tritium channel, s--14 These counters have photomultipliers responsive in the range 380--620 nm. Samples were placed in 10 ram-diameter and 35 mm-height flasks which were placed inside 25 mm-diarneter and 50 mm-height low potassium glass vials. The vials were kept in the dark up to the moment of assay and determinations were carded out in a dark room in order to avoid vial phosphorescence activated by fluorescent light. A dim light from tungsten lamps was used. The dark current value in the absence of vials was 40 __+ 1 cps and the emission from the empty flasks and vials was 50 ___ 1 cps in a Packard Tri-Carb counter. In a Beckman LS 100 counter the corresponding values were 200 ___ 10 and 225 ___ 10 cps, respectively. Assay conditions were: 0.3 to 1.2 mg/ mL of homogenate protein in a reaction medium consisting of 120 mM KC1, 30 mM phosphate buffer (pH 7.4) with 3 mM tert-butyl hydroperoxide, in a final volume of 2 mL, samples were kept at constant temperature (30°C). The results are expressed in counts per second (cps) or cps/mg protein.

Electron microscopy Biopsies of human heart for transmition electron microscopy were immersed in cold 3% glutaraldehyde in 0.1 M phosphate buffer pH 7.4, fixed in glutaraldehyde, posffixed in 1% osmium tetroxide, dehydrated, and embedded in Araldite. Ultrathin sections were stained with lead citrate and orange acetate and examined with a Siemens Elmiskop 101 electron microscope. Electron micrographs were examined morphometrically under a grid. According to the criteria of Kloner, 14 the severity of mitochondrial swelling was graded 0 through 4 as follows: 0 = normal mitochondria, 1 = early swelling as manifested by separation of cristae and clearing of matrix density, 2 = more marked swelling than in grade 1, 3 = massive swelling with membrane disruption, and 4 = the findings in grade 3 plus rupture of inner and outer mitochondrial membranes.

Chemicals Tert-butyl hydroperoxide was from Aldrich Chemicals Co. (Milwaukee, WI), glucose oxidase, xanthine, xanthine oxidase, glutathione, superoxide dismutase, catalase, and tocopherol were purchased from Sigma Chem. Co. (St. Louis, MO). Other reagents were of analytical grade.

Statistics In vitro oxidative stress of liver and heart homogenates Rat liver and heart homogenates (4 mg of protein/ mL) were incubated at 370C with gentle stirring in a re-

The values in the text and tables indicate mean values---SEM. The significance of differences between means were analyzed by the ANOVA-Dunnett's test of variance analysis. 17

Oxidative stress determination

95 RESULTS

tS ,$ I

Hydroperoxide-initiated chemiluminescence

o

~" 3O

?

/

-I

Liver

u w w w

N15 IAI X

X 141

°o

,o

2'o TIME

(rain)

Fig. l. Time course of ten-butyl hydroperoxide-initiated chemiluminescence in tissue homogenates of human biopsies. 0.3 mg homogehate protein/mL

Light emission shows a kinetic which is characteristic of each tissue and of each experimental condition. 8-11 When homogenate s obtained from biopsies of human heart, liver, and muscle are added with tert-butyl hydroperoxide the emission increased with time to reach a maximal level after about 20 min for the heart, 30 rain for the liver, and 50 min for the skeletal muscle (Fig. 1). Maximal emission was linearly related to the protein concentration in the range of 0.2 to 1.2 mg/mL. It was found difficult to reproduce emissions at protein concentrations lower than 0.2 mg/mL and a progressive decrease of photoemission was observed when protein concentrations were higher than 1.2 mg/mL, this latter probably due to turbidity quenching. It is worth noting that freezing of the biopsies at - 20°C for 72 h did not modify the hydroperoxide-initiated chemiluminescence of the homogenates (data not shown).

1200

HEART

"3 1000 L

E e~

800 LIVER

w

Z

tu

u

u1 uJ

600 1200

Zm

t000

X _.1

i

S00'

~oo

bJ Z

600 1,00

200

200 20 |

O0

I

1

2

i.

tert - BUTYL

40 60 80 TIME (mini

HYOROPF'ROXIOE

100 I

6 (raM)

Fig. 2. Effect of ten-butyl hydroperoxide concentration on maximal photoemission. Inset: Effect of ten-butyl hydroperoxide concentration on the time course of chemiluminescence in heart homogenates. (D 1 raM, • 2 mM, [] 3 mM and • 6 mM ten-butyl hydroperoxide. Protein concentration was 0.5 mg/mL.

96

B. GONZALEZFLECrtA et al.

1600

chemiluminescence at a hydroperoxide concentration of of 0.9 mM and a plateau at hydroperoxide concentrations higher than 2 mM in the case of rat liver, and 3 mM in the case of rat heart (Fig. 2). The initiation reactions as measured by initial slopes of the chemiluminescence versus time plots (units: c.s -2) also depended on hydroperoxide concentration (Fig. 2, inset). The optimal pH value was about 7.47.6, at lower and higher pH values photoemission decreased (Fig. 3). Different reaction media were tested for the assay. Maximal chemiluminescence was observed in the presence of electrolyte solutions and buffers. The presence of organic substances such as Tris, sucrose, mannitol, and so forth, which can act as free radical scavengers, markedly diminished chemiluminescence (Table 1). The rate of the chain reaction depended highly on the assay temperature. At temperatures lower than 25°C, the reaction was very slow and it took at least 90 min to reach maximal emission. The values of maximal emission also depended on temperature. Optimal temperature seemed to be 30°C (Fig. 4).

HEART

Q

1200 E

1000

u

m Z W W Z

800

600

z &OO

z

W Z u

200

O:

0

I

I

I

I

I

6.0

6.5

7.0

7.5

8.0

pH

Fig. 3. Effect ofpH. Sampleswere processedand assayedin 100 mM phosphate buffers of different pH. Protein concentration was 0.4 mg/ mL. Duplicate samples were measured.

Effect of inhibitors Hydroperoxide-initiated chemiluminescence was inhibited significantly when free radical scavengers such as tocopherol, ascorbic acid, and reduced glutathione were present in the reaction medium. In contrast, superoxide dismutase and catalase in concentrations approximately 10 times higher than the ones corresponding to the intact tissues, did not affect light emission (Fig. 5).

Optimal experimental conditions Small samples (10-50 mg) of rat heart and liver were used to set up the assay conditions. Maximal chemiluminescence showed an hyperbolic dependence on hyd r o p e r o x i d e c o n c e n t r a t i o n with h a l f m a x i m a l

6000 c o o u~ 6

HEAR

E

6000 u W Z W w

Z

"312

'

3'.,

'

3.6

2000

~

L

I

V

E

R

I

W

~

•.//

0 0

i

10

20 TEMPERATURE

30

temperature on maximalchemiluminescence in rat liver and heart homogenates. Inset: ICI/molfor both tissue homogenates. Protein concentration: 0.5 mg/mL. Duplicate sampleswere measured. Fig. 4. Effect of the

40

(°C) Arrhenius plot. Apparent E* is 140

Oxidative stress determination

Table 1. Effect of the Composition of the Reaction Medium on Hydroperoxide-initiated Chemiluminescence of Rat Tissue Homogenates. (In all eases, pH was 7.4.)

BufferComposition 120 mM KC1, 30 mM phosphate. 100 mM phosphate. 120 mM KC1, 30 mM phosphate, 1 mM EDTA. 140 mM KCI, 5 mM Tris-HC1. 140 mM KC1, 5 mM Tris, 1 mM EDTA. 100 mM KC1, 30 mM phosphate, 70 mM Sucrose. 40 mM KC1, 30 mM phosphate, 200 mM mannitol. 230 mM mannitol, 70 mM sucrose, 20 mM Tris-HCl. 70 mM sucrose, 230 mM rnannitol, 5 mM Tris, 1 mM EDTA

97

0.0

CATALASE 0.1

I'

(jcM)(a) 0.2

I

0.3

I

1

SOD ( ~ M I ( A )

CHEMILUMINESCENCE (cps/rng protein) Liver Heart

0

2

4.

6

8

10

I

I

I

|

I

I

690 +__ 60 555 --- 30 590 +_ 55

1400 ± 20 1350 ± 70 1130 ... 60

0

1

2

3

t

,

i

i

590 ± 35 230 --- 20* 590 ___ 70

980 ± 60 870 ± 50* 1050 ± 70

620 +-- 35

1200 ... 70

390 +__ 35*

590 ± 80*

200 ___ 40*

690 ± 60*

*p < 0.01

ASCORBATE ( m M ) ( z ~ )

100

An

A

&

C~

--"

•

80 UJ U X UJ U~ UJ Z

60

40

=)

This may indicate that superoxide anion and hydrogen peroxide are not implicated in the main pathway of the hydroperoxide-initiated chain reaction.

Hydroperoxide-initiated chemiluminescence in homogenates subjected to in vitro oxidative stress Rat liver and heart homogenates were subjected to oxidative stress, by supplementation with either hydroperoxide or with enzymatic systems that generate superoxide anion or hydrogen peroxide. These model systems simulate oxidative stress situations corresponding to excess production of either peroxyl radicals, superoxide anion or hydrogen peroxide (for tert-butyl hydroperoxide, xanthine-xanthine oxidase and glucose-glucose oxidase, respectively). The hydroperoxide-initiated chemiluminescence of homogenates subjected to oxidative stress was higher than the photoemission of the control homogenate. The chemiluminescence increase was proportional to hydroperoxide concentration or to the enzymatic activity of the generating system, that is, to the rate of oxyradical production. It was found that due to slight changes in the assay conditions, it was convenient to run samples subjected to oxidative stress in parallel with control samples and to express the resuits as the ratio poststress/control (or B/A), meaning the ratio of the chemiluminescence of the homogenate of tissue exposed to oxidative stress (poststress sample) to the chemiluminescence of the control homogenates (control sample). A lineal dependence was found between the ratio B/A and both the hydroperoxide concentration added to the homogenate and the enzymatic activities of the used generating systems (Fig. 6).

LU

20

I

I

1

1

I

0

10

20

30

hO

VITAMIN E ( n m o l / m g p r o t e i n ) ( e ) I

I

0

1

I

I

2 3 GSH ( m M ) ( o )

I

I

6

5

Fig. 5. Effect of glutathione, ascorbate, tocopherol, superoxide dismutase, and catalase on hydroperoxide-initiated chemiluminescence.

Correlation between chemiluminescence and morphological damage after oxidative stress in human biopsies The assay was applied to estimate the oxidative stress associated to the ischemia-reperfusion process of human heart during revascularization surgery, of the human liver during tumor resection surgery, and human muscle during ischemia and reperfusion. Reperfusion biopsies showed a higher chemiluminescence as compared with control (preischemic) biopsies (Table 2). In the case of human heart, simultaneous determinations of mitocondrial damage by electron microscopy showed that in postreperfusion samples (sample B), the percentage of severely damaged mitochondria (grades 3 and 4) was higher than in control samples (sample A). The ratio between the percentage of mitochondria with grade 3 and 4 of damage in (B) and (A) samples showed a marked and significant positive correlation with the similar B/A ratio for chemiluminescence (Fig. 7).

B. GONZAL~ FLORA et al.

98

0 |

GLUCOSE O X I D A S E ( m U / m l ) ( A ) 1S 30 I

I

I

1.6

I

I

zx

,,

0

GLUCOSE O X I D A S E ( m U / m l ) ( z x | 15 30

I'

I

I

|

I

I

1./t

LIVER 1.5

HEART

/

1.3

1.¢

O

o i-,-

1.3

•

m 1.2

A

~

o

•

o

/

n, 1.2 < m

.Y

•

1.1 1.1

"~

°

1.0

I

I

I

0

0

I

I

I

I

I

I

I

I

0 100 300 300 ~00 500 [tert-BUTYL HYDROPEROXIDE] (,~I,M } ( e ]

I

0 100 200 300 400 500 [tert-BUTYL HYDROPEROXIDE ](j~M)(•)

0

•

1.0

I

0

I

100 200 XANTHINE 0XIDASE ( m U / m l } ( o )

I

I

I

I

I

2 4 6 8 10 XANTHINE OXIDASE ( m U / m l ) ( o )

Fig. 6. Poststress chemiluminescence/control chemiluminescence ratio (B/A ratio) in rat liver and heart homogenates subjected to in vitro oxidative stress. Homogenates were incubated at 37°C with stirring in the presence of either tert-butyl hydroperoxide or glucose-glucose oxidase or xanthine-xanthine oxidase.

DISCUSSION

In this paper we present an assay applicable to human biopsies to characterize situations of oxidative stress. Some of the experiments needed to properly set up the experimental conditions have been performed using animal tissues. The conditions defined here have al-

ready been used to evaluate oxidative stress in human biopsies of heart, liver, and muscle in cases of ischemia-reperfusion. The assay appears as useful to evaluate the integral level of the nonenzymatic antioxidant defenses of a tissue. 8-]4 The substances that constitute the nonenzymatic defenses (tocopherol, retinol, etc.) would decrease

Table 2. tert-Butyl Hydroperoxide Initiated-Chemiluminescence in Human Tissues Subjected to In Vivo Oxidative Stress

TISSUES

n

EXPER/MENTAL SITUATION OR TREATMENT

Heart

6 6 6 8

Ischemia-reperfusion Ischemia-rep. + rnannitol Ischemia-rep. + deferoxamine Ischemia-rep. + vit (A + E)

Liver

4

Ischemia-reperfusion

1.4 ± 0.1

Muscle

6

Ischemia-reperfusion

2.8 -+ 0.4

~from ref. 13 2from ref. 14

RATIO POSTSTRESS/CONTROL 2.1 1.2 1.1 0.9

--- 0.4 ] -.+ 0.22 --- 0.3 ± 0.2

Oxidative stress determination •

s4D / / /

0

/ 1 I f