- Email: [email protected]
Increased liver chemiluminescence in tumor-bearing mice
Journal of Free Radicals #1 Biology & Medicine, Vol. 1, pp. 131-138, 1985 Printed in the USA. All rights reserved.
INCREASED
LIVER
0748-5514/85 $3.00+ .00 © 1985 Pergamon Press Ltd.
CHEMILUMINESCENCE
IN TUMOR-BEARING
MICE
ALBERTO BOVERIS, SUSANA F. LLESUY, a n d C~SAR G . FRAGA lnstituto de Qufmica y Fisicoqu/mica Biol6gicas (UBA-CONICET), Facultad de Farmacia y Bioqufmiea, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina (Received 22 October 1984, Revised 19 February 1985; Accepted 22 February 1985)
Abstract--Spontaneous mouse liver chemiluminescence (109 -+ 6 cps/cm 2) was increased in the early phase after tumor implantation in a distant position with respect to the liver. A 39% increased liver chemiluminescence was observed after 5 days of the injection of Ehrlich ascites tumor cells into the peritoneal cavity, and a 64% and a 46% increased liver chemiluminescence were measured after 8 and 14 days of the implantation of a fibrosareoma and of an adenocarcinoma, respectively, in the leg. At the time of maximal stimulation of in vivo liver chemiluminescence by the distant tumors, cytosolic superoxide dismutase, catalase, and glutathione peroxidase activities were decreased by 18%, 38%, and 26% in the liver of mice bearing Ehrlich ascites tumors. The same three enzymatic activities were decreased by 21%, 19%, and 54% respectively, in the liver of fibrosarcoma-bearing mice. Total liver glutathione was decreased by 18% to 22% in the tumor-bearing animals. Hydroperoxide-initiated chemiluminescence was increased in the homogenates (105% and 45%) and mitochondria (64% and 34%) from the liver of mice bearing Ehrlich ascites tumors and fibrosarcomas, respectively, at the time of maximal in situ liver chemiluminescence. The hydroperoxide-initiated chemiluminescence of liver microsomes was decreased by 46% to 36% in the tumor-bearing animals at the same time. It is concluded that the liver of tumor-bearing animals is subjected, during the early phase after tumor implantation, to an oxidative stress with increased steady-state levels of peroxyi radicals, which are essentially responsible for the increased photoemission observed in vivo. Keywords--Free radicals, Tumor implantation, Antioxidant enzymes, Mitochondria, Cancer, Chemiluminescence, Lipid peroxidation, Microsomes
port by Neifakh s described increased chemiluminescence and lipoperoxide content in liver and brain homogenates from tumor-bearing mice and a report by Inaba et al.9 described an increased chemiluminescence in blood samples from humans with carcinomas. However, there are no integrative studies on tumor-bearing animals in which in situ liver chemiluminescence as an indication of in vivo lipid peroxidation ~° is measured simultaneously with the activity of antioxidant enzymes and the content of endogenous antioxidants. In this paper we report an increased in situ liver chemiluminescence in the early phase after tumor implantation in tumor-bearing mice which is associated with a decreased activity of the protective antioxidant enzymes and an increased hydroperoxide-initiated chemiluminescence in the homogenates and the mitochondria.
INTRODUCTION
The primary production of superoxide anion and hydrogen peroxide at the mitochondrial and endoplasmic reticulum membranes seems to support the endogenous rate of hydroperoxide production and lipid peroxidation. The process appears mediated by hydroxyl radical generation, which initiates a free radical reaction chain leading to lipid peroxidation. The enzymes superoxide dismutase, catalase, and glutathione peroxidase seem to constitute the main protective system to minimize the rate of hydroxyl radical generation and the free radical reaction of lipid peroxidation. Decades ago, Greenstein et al. 2-4 reported a decreased catalase activity in the non-tumor containing livers and kidneys of tumor-bearing mice and rats. The decrease in catalase activity was observed in animals bearing a series of tumors and was reversible upon surgical removal of the tumor. ~5 More recently, Tarakhovsky et al. 6 and Oberley 7 have reported a decrease in superoxide dismutase activity in the livers of tumorbearing mice. It is then apparent that the livers of tumorbearing animals are subjected to a decreased antioxidant protection. Concerning lipid peroxidation, a brief re-
MATERIALS AND METHODS
Animals Female inbred BALB/c mice, 2 to 3 months old and weighing 15 to 20 g from the Experimental Leukemia 131
132
A. BovEms, S. F. LLESUY, and C. G. FRAGA
Section, National Academy of Medicine (Buenos Aires, Argentina) were used. Animals were supplied with water and laboratory animal food ad libitum.
Tumors Ehrlich ascites tumor cells were implanted by intraperitoneal injection of 0.5 x 10 7 cells/mouse. Fibrosarcoma cells, originated by foreign body tumorigenesis, z~ and adenocarcinoma cells, derived from a mammary tumor, were implanted as small solid tumors of 1 to 2 mm 3 in an incision in the leg.
Liver chemiluminescence Control and tumor-bearing mice were anesthetized intraperitoneally with urethane 12% w/v at a dose of 0.1 ml/10 g of weight; the liver surface was exposed by laparotomy and chemiluminescence was measured with a Johnson Foundation photon-counter t°-~z (Johnson Research Foundation, University of Pennsylvania, Philadelphia, PA). The emission is expressed as cps/cm 2 (of liver surface). In the case of mice bearing Ehrlich ascites tumors, the ascites and the tumor cells were removed from the peritoneal cavity before the measurement of liver chemiluminescence.
Subcellular fractions Subcellular fractions from control and tumor-bearing mice were prepared from animals that were not subjected to laparotomy to avoid any effect of anesthesia and surgery on the properties of the isolated organelles. Mouse liver homogenates were prepared in a medium consisting of 140 mM KCI, 1 mM EDTA, 5 mM TrisHC1, pH 7.3, and centrifuged at 600 g for 10 min to discard nuclei and cell debris. The supernatant, a suspension of mixed and preserved organelles, was used as "homogenate."t3-t4 Mitochondrial and microsomal fractions were isolated by conventional procedures. I*-t5 The 105,000 g supernatant was used as " c y tosol" to determine cytosolic superoxide dismutase activity. The operations were carried out at 0 to 2°C. Protein was assayed by the Lowry et al. ~6method using bovine serum albumin as standard. The protein content of subcellular fractions was calculated from the ratio of cytochrome oxidase activity in the homogenate and in the isolated organelles (mitochondria) and from the isolated microsomal protein, as described previously. 14
Hydroperoxide-initiated chemiluminescence of subcellular fractions The hydroperoxide-initiated chemiluminescence of homogenates, mitochondria, and microsomes ~3.t4"uT'us
were measured in a Packard Tri-Carb model 3320 liquid scintillation counter at room temperature in the out-ofcoincidence mode. ~ Samples (4 ml of final volume) were placed in 13-mm diameter and 40-mm height flasks which were placed inside 25-mm diameter and 58-mm height glass vials. The emission is expressed in cps/ mg of protein. The background level of emission of the empty flasks and vials was 80 to 100 cps. The reaction medium consisted of 140 mM KC1, 1 mM EDTA. 5 mM Tris-HC1, pH 7.3. Chemiluminescence measurements were started by addition of 3 mM tert-butyl hydroperoxide. The protein content was adjusted at 1 nag' ml of reaction medium for homogenates and mitochondria and at 0.25 mg/ml for microsomes.
Superoxide dismutase Cytosolic superoxide dismutase activity was determined from the inhibition of the rate of autocatatytic adrenochrome formation ~9 in a reaction medium containing 1 mM epinephrine and 50 mM glycine-NaOH, pH 9.6.14
Catalase The catalase activity of the liver homogenates was determined by measuring the decrease in 240 nm absorption in a reaction medium consisting of 50 mM phosphate buffer, pH 7.2, 2 mM H202, 1% Triton X100, and 0.1 to 0.3 mg of protein/ml, determining the pseudo-first order reaction constant (k') of the decrease in H2Oz absorption (ez4o = 40 M - I cm-~) 2° and expressing the results as catalase content in mol/g tissue. The value o f k = 4.6 × 10 7 M - I S-t was used as the second-order reaction constant for pure catalase. '-~
Glutathione perox~dase This activity was determined following NADPH oxidation at 340 nm in the presence of 0.17 mM GSH, 0.2 U/ml yeast glutathione reductase, and 0.5 mM tertbutyl hydroperoxide.
Glutathione content Total glutathione (GSH plus 2 GSSG) was determined in liver homogenates after precipitation with 2% perchloric acid and using yeast glutathione reductase and 5-5'-dithio.bis-nitrobenzoic acid. 2:
Chemicals Yeast glutathione reductase, NADPH, reduced glutathione, epinephrine, 5-5'-dithio-bis-nitrobenzoic acid, and Triton X- 100 were purchased from Sigma Chemical
Chemiluminescence and cancer Co. (Saint Louis, MO); tert-butyl hydroperoxide was from Aldrich Chemical Co. (Milwaukee, WI); hydrogen peroxide was from Baker Chemical Co. (Phillipsburg, NJ); and urethane was from B.D.H. (London, England). Other reagents were of analytical grade.
133
Table i. Spontaneous Liver Chemiluminescenceof Tumor-Bearing Mice Tumor
Statistics
None (control) (n = 30) Ehrlich ascites (day 5; n = 12) Fibrosarcoma(day 8; n = 12) Adenocarcinoma (day 14; n = 12)
The figures in the text and tables indicate mean values --- S.E.M. Differences were analyzed by the variance test (ANOVA Test; Texas Instruments).
n = numberof animals. *p < .05 **p < .01
RESULTS
Liver chemiluminescence The spontaneous liver chemiluminescence of tumorbearing mice was significantly increased during the first days of growth of a distant tumor. Mice bearing either Ehrlich ascites tumors, or fibrosarcomas, or adenocarcinomas showed 39%, 64%, and 46% increased liver photoemission, respectively, at 5, 8, and 14 days after tumor implantation (Fig. 1 and Table 1). It is worth noting that the increased liver chemiluminescence was produced in an organ free of metastasis and by a distant tumor; the fibrosarcoma and the adenocarcinoma were implanted in the leg, whereas the ascites tumor cells were implanted in the abdominal cavity. The increase in liver chemiluminescence was related to tumor development; maximal liver emission was observed when the Ehrlich ascites tumors reached 18% to 22% of the body weight and fibrosarcomas and adenocarcinomas reached 3% to 5% of the body weight. The state of maximal emission was a transient state, after reaching a maximal level of the liver chemiluminescence of the 200
W o Z I,LI
_z,,~ loo X W
00
(*) 10 (o) 20 (e) TIME(doys)
I
10 20 30
Fig. !. Spontaneouschemiluminescenceof the in situ liver of normal and tumor-bearingmice. (o), normal mice; (A), mice bearing Ehrlich ascites tumor; (<3),mice bearing fibrosarcoma;and (m), mice bearing adenocarcinoma. The abscissa indicates the time after tumor implantation. Exposed liver areas were about I cm-'. Symbols indicate mean values and bars indicate S.E.M.
Chemiluminescence (cps/cm2) 109± 152 ± 179 ± 159 ±
6 19" 17"* 21'
tumor-bearing animals returned a few days later to the level showed by normal mice (Fig. 1). Mice bearing Ehrlich ascites tumor, fibrosarcoma, and adenocarcinoma died 7 to 12, 12 to 24, and 26 to 44 days, respectively, after tumor implantation.
Hydroperoxide-initiated chemiluminescence of liver homogenates and mitochondrial and microsomal suspensions The protein content of the liver subcellular fractions and the liver weight of tumor-bearing mice are shown in Table 2. The homogenate protein content was 23% decreased in the livers of mice bearing Ehrlich ascites tumors. Liver weights and the content of subcellular fractions were not different from control values in all of the other cases (Table 2). When tert-butyl hydroperoxide is added to either mouse liver homogenates or mitochondrial or microsomal suspensions, there is a burst of photoemission with different kinetics depending on the type of subcellular preparation, j3.~4.jT,18 Mouse liver homogenates show a maximum of chemiluminescence after 15 min of the addition of 3 mM tert-butyl hydroperoxide for tumor-beating mice and after 20 min for normal mice (Fig. 2). This emission declined at 20 and 25 min, respectively, and then slowly increased to reach a steady state of maximal chemiluminescence which occurs after 60 to 90 min. The liver homogenates from mice bearing either Ehrlich ascites tumors or fibrosarcomas gave 105 % and 45%, respectively, more maximal chemiluminescence than the liver homogenates from normal mice (Fig. 2 and Table 3). The mitochondria isolated from the liver of mice bearing either Ehrlich ascites tumors or fibrosarcomas gave 64% and 32%, respectively, more emission than the mitochondria isolated from the liver of normal mice, again measured at the point of maximal photoemission, after 10 to 20 min of the hydroperoxide addition (Fig. 3 and Table 3). At variance, the microsomes isolated from the liver of tumor-bearing mice were less effective in yielding hydroperoxide-initiated light emission as compared with the same organelles from normal mice.
134
A. BOVERIS, S. F. LLESUY, and C. G. FRAGA
Table 2. Liver Weight and Protein Content of the Liver Subcellular Fractions of Tumor-Bearing Mice Homogenates
Mitochondria
Liver weight (g)
Tumor
1.14 1.77 0.99 1.09
None (control) (n = 20) Ehrlich ascites (day 5; n = 12) Fibrosarcoma (day 8; n = 12) Adenocarcinoma (day 14; n = 12)
+ m -
Microsomes
(mg of protein/g liver)
0.03 0.19" 0.07 0.08
78.7 60.4 78.2 75.6
+ -+ -+
3.2 4.2 3.6 4.1
24.6 22.0 21.9 21.3
+ +-+ :'=
2+3 1.5 2.4 1.8
13.3 15.2 16.4 15.'9
: 09 + 1.7 -+ I+0 -..+- 1.2
n = number of animals. *p < .05
Maximal chemiluminescence, measured at 30 to 90 s after hydroperoxid¢ addition, was diminished by 46% and 38% in the liver microsomes from mice bearing either Ehrlich ascites tumors of fibrosarcomas (Fig. 4 and Table 3). There were not significant differences in the kinetics of the hydroperoxide-initiated chemiluminescence of homogenates, mitochondria, and microsomes in comparing the subcellular fractions obtained from normal and tumor-hearing mice. Chemiluminescence was measured at day 5 and at day 8 for Ehrlich ascites tumor- and fibrosarcoma-bearing mice, respectively.
Superoxide dismutase, catalase, and glutathione peroxidase In the days of maximal emission, these three antioxidants enzymes showed diminished activities in the livers of tumor-bearing mice (Table 4). The activity of cytosolic superoxide dismutase was decreased by 18% in the case of mice bearing Ehrlich ascites tumors and
Table 3. Hydmpcroxide-lnitiatedChemiluminescence of Liver SubcellularFractionsfrom Tumor-Bearing Mice Chemiluminescence of Subcellular Fraction Homogenate
Tumor
Mitochondria
(cps/mg protein)
None (control) Ehrlich ascites (day 5) Fibrosarcoma (day 8)
228 --- 25 467 -- 23* 330 --. 22*
806 ± 52 1320 -- 23* 1061 ± 27*
by 21% in mice bearing fibrosarcomas. Catalase activity was decreased by 38% and 19%, whereas glutathione peroxidase activity was decreased by 26% and 54% in the livers of mice bearing Ehrlich ascites tumors and fibrosarcomas, respectively. Total glutathione content
Tumor-bearing mice showed a diminished content of total liver glutathione which was reduced by 18%
1500r
~
~
2000
W¢~
¢j ~
-
,+o
I
890 -- 48 483 -+ 32* 554 - 28*
*p < .001. Nine animals in each group.
500
i
Microsomes
.
.
.
.
.
i 000
,000 / l r
-
\\
°o
i & ,..o..,..,
II
soo
t - BOOH ( mM I
('l |
l
+0
5
I
I
l
10 15 20 TIME (min)
I
I
25
30
0
10
20 30 40 T I M E (rain)
50
60
Fig. 2. tert-Butyl hydroperoxide-initia~'d chemiluminescence of liver homogenates from normal and tumor-hearing mice. (o), normal mice; (&), mice bearing Ehrlich ascitcs tumor; and (0), mice bearing
Fig. 3. tert-Butyl hydroperoxide-initiated chemiluminescence of liver mitochondria from normal and tumor-bearing mice. Symbols as in
fibrosarcoma.
Fig. 2,
Chemiluminescence and cancer
z.
tO00
"7'
O,.
-r
I
0
2
I
I
I
4 6 8 T IME (mini
I
10
,/j
I
60
Fig. 4. tert-Butyl hydroperoxide-initiated chemiluminescence of liver microsomes from normal and tumor-bearing mice. (o), normal mice; (A), mice bearing Ehrlich ascites tumor; and (e), mice bearing fibrosarcoma.
and by 22% in mice bearing Ehrlich ascites tumors and fibrosarcomas, in days 5 and 8 after tumor implantation, respectively (Table 4).
Correlation of liver chemiluminescence with the activity of protective enzymes The liver surface chemiluminescence showed a negative, statistically significative correlation with the activity of superoxide dismutase, glutathione peroxidase, and the content of total glutathione for the cases of control mice and mice bearing Ehrlich ascites tumors and fibrosarcomas. A qualitative correlation, statistically not significant, was observed between liver chemiluminescence and catalase activity (Fig. 5).
135
substance did not decrease catalase activity of liver homogenates z5 or the purified enzyme, z6 These studies were not further pursued. Other examples of altered structure and functions in distance organs of tumor-bearing mice are the decreased "y-glutamyi transferase activity with increased vacuolization in the liver, z7 and the decreased osmotic fragility and progressive decrease in the red cells content zs reported in mice bearing Ehrlich ascites tumors. Liver chemiluminescence affords an indirect assay for the in vivo steady-state level of peroxyl radicals; the assay is non-invasive, non-destructive, and specific for the organ. The results presented in this paper show that the livers of tumor-bearing mice exhibit an increased chemiluminescence in the early phase after tumor implantation. The increased photoemission reflects an increased level of oxyl and peroxyl radicals that can be understood on the basis of free-radical reactivity as an oxidative stress to the biological structure. Although it is possible to have chemiluminescence without lipid peroxidation in chemically defined and cell-free systems, 29 it is established that an increase in lipid peroxidation rate in organs and isolated cells produces a parallel increase in photoemission ~°.12.3°. It is important to consider the tissue absorption of the emitted light. Preliminary work estimates that the measured light comes from the outer 50 to 100 am of the liver; consequently, the assay would have limited validity if considerable differences exist between the activity in the outermost cells and the rest of the tissue. The reactions that lead to the spontaneous chemilurfainescence of in situ liver are understood as follows: H202 + 02 ~
DISCUSSION
The idea that tumors release a toxic substance that affects distant organs was put forward by Lucke et al. ,z3 who showed reduced catalase activity in the liver of parabiotic rats. Greenstein et al. z-4 reported that tumor extracts supressed liver catalase activity and Nakahara and Fukuoka 24 isolated a substance from neoplastic tissues, which they called toxohormone, that lowered liver catalase activity when injected to mice. However, this
) "OH + OH- + 02
(1)
~ H20 + R.
(2)
~ ROO.
(3)
2 ROO.
~ RO + ROH + ~O2
(4)
2 ROO.
~ 02 + ROH + RO*
(5)
•OH + RH R. + 02
Table 4. Superoxide Dismutase, Catalase, and Glutathione Peroxidase Activity and Glutathione Content in the Liver of Tumor-Bearing Mice
Tumor
Superoxide Dismutase (U/g liver)
Catalase (nmol/ liver)
Glutathione Peroxidase (U/g liver)
Glutathione Content (~umol/g liver)
None (control) (n = 20) Ehrlich ascites (day 5 ; n = 12) Fibrosarcoma (day 8; n = 12)
174 ± 8 143 ± 9** 137 ~ II**
0 . 9 0 ± 0.06 0.56 ± 0.03** 0.73 ± 0.03**
10.0 ± 0.7 7.4 ± 0.8* 4.6 ± 0.4**
7.6 ± 0.2 6.2 ± 0.3* 5.9 ± 0.3*
= number of animals. *p < .01 **p < .001
n
136
A. BOVERIS. S. F. LLESUY. and C. G. F ~ a 6 a
,oo
..o
l'°:
2--
"-:t-"°t
-
V°°I 0i .
•
%0
'
,~o
'
1;o
'
,;o
'
~;o
ro
o°
LIVER CHENILUHINESCENCE (cp$/cm z )
Fig, 5. Correlations between surface liver chemiluminescence and catalase, su~cro×ide dismutase and glutathione peroxidase activities and glutathione content in the liver of normal and tumor-bearing mice. Correlations values are: r = .93 and p < .05 for superoxide d i s m u t a s e activity; r = .92 and p < .05 for glutathione peroxidase activity: r = .89 and p < .05 for glutathione content; and r = .62 and p > .05 for catalase activity (dashed line).
H\
O--O H~[C_C[/H
/H
R/C'-C\R + IO2
"~ R/
\R
' (6)
+
H\
O--O H\I I/H
/H ,
,
OO"
(7)
H\
+ :=o
CO* + 02
, to2 + R / C = O
(8)
The initiation reaction is the hydroxyl radical generation, in a chelated iron-catalyzed Haber-Weiss reaction (Eq. I). Most of superoxide anion and hydrogen peroxide are generated in the membranes of mitochondria and of endoplasmic reticulum.' The hydroxyl radical will abstract hydrogen from the hydrocarbon chain of membrane polyunsaturated fatty acids (Eq. 2) yielding aikyl radicals which react with oxygen yielding peroxyl radicals (Eq. 3 ) . 31-32 These reactions will initiate a free-radical chain reaction. The termination reactions of peroxyl radicals appear essential for the photoemission process observed during lipid peroxidation, 3°-33 these reactions yield either singlet molecular oxygen (Eq. 4) or excited carbonyl groups
(Eq. 5) which are the photoemissive species. 29 Excited carbonyl groups can also be generated in the decomposition of dioxetane derivatives 34.35 formed by either addition of singlet oxygen to unsaturated hydrocarbon chains (Eq. 6 ) 34.36 o r cyclization of unsaturated tertiary peroxyl and alkyl biradicals (Eq, 7). It is interesting that singlet oxygen can generate excited carbonyl groups through dioxetane formation (Eq. 6) and excited carbonyis can produce singlet oxygen after coilisional quenching (Eq. 8)? 6 A first defense in the protective system against oxidative stress is provided by the enzymes superoxide dismutase, catalase, and glutathione peroxidase that catalyze reactions that diminish the steady-state concentrations of superoxide anion and hydrogen peroxide which are the reactants for hydroxyl radical production (Eq. 1). The activity of these antioxidant enzymes is diminished in the liver of tumor-bearing mice in the early phase of tumor development. Moreover, surface liver chemiluminescence shows a negative correlation, statistically significant, with tissue levels of superoxide dismutase, glutathione peroxidase, and total glutathione in normal and tumor-bearing mice. The diminished activity of Cu-Zn superoxide dismutase agrees with a report by Tarakhovsky 6 on decreased activity of total and Cu-Zn superoxide dismutases. Oberley 7 has reported just the: opposite, a lowered Mn-superoxide dismutase activity with a variable Cu-Zn superoxide dismutase activity. The simultaneous decrease in the activity of superoxide dismutase, catalase, and glutathione peroxidase seems to indicate that the three antioxidant enzymes are regulated synchronically; the three of them were observed increased in barbital treated mice. 37 For the in vitro hydroperoxide-initiated chemiluminescence, the hemoprotein-promoted radical for-
Chemiluminescence and cancer m a t i o n from a d d e d h y d r o p e r o x i d e ( t - B O O H ) affords the initiation reactions (Eqs. 9 - 1 l).ts'38
137
cies specific, it is p o s s i b l e that s u p p l e m e n t a t i o n with b i o l o g i c a l l y active antioxidants may afford an effective c o a d j u v a n t therapy for t u m o r - b e a r i n g m a m m a l s .
t-BOOH + Fe 3+ t-BOO- + H + + Fe 2÷ t-BOOH + Fe '+
(9) Acknowledgements--The authors wish to express their gratitude to
-~
t-BOO.(t-BO.) + RH
t-BO. + H O - + Fe ~+
(10)
> t-BOOH(t-BOH) + R.
(11)
The p e r o x y l and a l c o x y l radicals react with polyunsaturated fatty acids through Equation 11, eventually l e a d i n g to the generation o f excited species similar to the in v i v o p h o t o e m i s s i o n (Eqs. 3 - 8 ) . A s e c o n d defense system against increased levels o f o x y l radicals is p r o v i d e d by free-radical scavengers, such as r e d u c e d glutathione (in the h y d r o p h i l i c domains) and vitamins A and E (in the h y d r o p h o b i c domains). T h e s e free-radical scavengers ( A H ) act acc o r d i n g to the f o l l o w i n g reactions: ROO" + G S H 2 GS" R" + AH
2 A.
~ ROOH + GS"
(12)
' GSSG
(13)
~ RH + A. , A - A
(14) (15)
The i n c r e a s e d rates o f h y d r o p e r o x i d e - i n i t i a t e d chem i l u m i n e s c e n c e o b s e r v e d in h o m o g e n a t e s and mitoc h o n d r i a from the liver o f t u m o r - b e a r i n g mice could be u n d e r s t o o d as a l o w e r level of e n d o g e n o u s antioxidants and free-radical scavengers in the tissue. ~.~.~4Thus, a d e c r e a s e in the tissue level o f antioxidant substances will also contribute to an increased rate o f lipid pero x i d a t i o n and p h o t o e m i s s i o n in vivo. The integration o f the in v i v o and in vitro o b s e r v a t i o n s indicate an e n h a n c e m e n t o f lipid p e r o x i d a t i o n and a d i m i n i s h e d level o f antioxidant substances and antioxidant enz y m e s in the livers of t u m o r - b e a r i n g mice. The t u m o r effect m a y be p r o d u c e d either by the release o f a substance or by abstraction o f some blood c o m p o n e n t ( s ) which is essential for the maintenance and p r o p e r function o f the distant t i s s u e ? A l t h o u g h our results do not e x c l u d e either, or both p o s s i b i l i t i e s , the increase in the h y d r o p e r o x i d e - i n i t i a t e d c h e m i l u m i n e s cence o b s e r v e d in w a s h e d m i t o c h o n d r i a appears more l i k e l y to indicate an abstraction than an addition o f some substance by the tumors. If the o x i d a t i v e stress o b s e r v e d in the early phase after t u m o r implantation in the mouse liver is not spe-
Dr. Christianne Dosne de Pasqualini of the National Academy of Medicine (Buenos Aires, Argentina) for her encouragement and the generous supply of tumor-bearing animals and to Dr. Martha Dubin for the glutathione determinations. This research was supported by grants from CONICET (Consejo Nacional de lnvestigaciones Cienffficas y T(~cnicas, Argentina) and Fundaci6n Cherny. Alberto Boveris is a Career Investigator and C(~sarG. Fraga is a fellow of CONICET.
REFERENCES
I. B. Chance, H. Sies, and A. Boveris. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59:527-605 (1979). 2. J. P. Greenstein, W. V. Jenrette, and J. White. The liver catalase of tumor-bearing rats and the effect of extirpation of the tumors. J. Biol. Chem. 141:327-328 (1941). 3. J. P. Greenstein and H. B. Andervoont. Note on the liver catalase activity of pregnant mice and mice bearing growing embrionic implants. J. Natl. Cancer Inst. 4:283-284 (1943). 4. J. P. Greenstein. Biochemistry of Cancer, pp. 316-366, Academic Press, New York (1947). 5. H. Bush. An Introduction to the Biochemistry of the Cancer Cell, pp. 241-256, Academic Press, New York (1962). 6. A. M. Tarakhovskii, G. V. Glinskii, and V. A. Shliakhovenko. Superoxide dismutase activity in tumor and liver tissues of tumor-bearing animals. Ukr. Biochem. J. 52:628-632 (1980). 7. L. W. Oberley. Superoxide dismuta~ and cancer. In: Superoxide Dismutase, Volume II (L. W. Oberley, ed.), pp. 127-165, CRC Press, Boca Raton, FL (1982). 8. Y. A. Neifakh. Free radical mechanism of ultraweak chemiluminescence coupled with peroxide oxidation of unsaturated fatty acids. Biofizika 16:584-587 (1971). 9. H. Inaba, A. Yamagishi, C. Takyu, B. Yoda, Y. Goto, T. Miyazawa, T. Kaneda, and A. Saeki. Development of an ultra-high sensitive photon counting system and its application to biomedical measurements. Optics and Lasers in Engineering 3: 125130 (1982). 10. A. Boveris, E. Cadenas, R. Reiter, M. Filipkowsky, Y. Nakase, and B. Chance. Organ chemiluminescence: Non-invasive assay for oxidative radical reactions. Proc. Natl. Acad. Sci. U.S.A. 77:347-351 (1980). I I. C. D. Pasqualini, L. Sen, E Sadl, L. Schwartz, and L. Z. Tkaczevski. Tumor development in mice bearing a plastic cylinder and inoculated with human neoplastic cells. J. Natl. Cancer. Inst. $1:283-286 (1973). 12. L. A. Videla, C. G. Fraga, O. R. Koch, and A. Boveris. Chemiluminescence of the in situ rat liver after acute ethanol intoxication. Effect of ( + )-cyanidanol-3. Biochem. Pharmacol. 32: 2822-2825 (1983). 13. E. Cadenas, A. I. Varsavsky, A. Boveris, and B. Chance. Oxygen or organic hydroperoxide-induced chemiluminescence of brain and liver homogenates. Biochem. J. 198:645-652 (1981). 14. A. Boveris, C. G. Fraga, A. I. Varsavsky, and O. R. Koch. Increased chemiluminescence, and superoxide production in the liver of chronically ethanol-treated rats. Arch. Biochem. Biophys. 227:534-541 (1983). 15. A. Boveris, N. Oshino, and B. Chance. The cellularproduction of hydrogen peroxide. Biochem. J. 12,8:617-630 (1972). 16. O. H. Lowry, N. 1. Rosebrough, A. L. Farr, and R. J. Randall.
138
17.
18.
19. 20.
21.
22.
23.
24. 25. 26. 27. 28.
A. BOVERIS, S. F. LLESUY. and C G. FR~c~, Protein measurement with Folin phenol reagent. J, Biol. Chem. 193:265-275 (1951). E. Cadenas, A. Boveris, and B. Chance. Low level chemiluminescence of bovine heart submitochondrial particles. Biochem. J. 186:659-667 (1980). E. Cadenas and H. Sies. Low level chemiluminescence of liver microsomal fractions initiated by tert-butyl hydroperoxide. Eur. J. Biochem. 174:349-356 (1982). H. P. Misra and I. Fridovich. The role of superoxide anion in the autooxidation of epinephrine and a simple assay for superoxide dismutase. J. Biol. Chem. 747:3170-3175 (1972). B. Chance. Special methods: Catalase. In: Methods of Biochemical Analysis (E. Glick, ed.), pp. 408-424, Interscience Publishers, New York, NY (1954). H. Sies, T. Bucher, N. Oshino, and B. Chance. Heine occupance of catalase in hemoglobin-free perfused rat liver and of isolated rat liver catalase. Arch. Biochem. Biophys. 154:106-116 (1973). F. Tietze. Enzymatic method for the quantitative determination of nanogram amounts of total and oxydized glutathion.e.. Anal. Biochem. 27:502-522 (1969). B. Lucke, M. Bet'wick, and I. Zeckwer. Liver catalase activity in parabiotic rats with one partner tumor-bearing. J. Natl. Cancer Inst. 13:681-686 (1952). W. Nakaharaand F. Fukuoka. The newerconcept ofcancer toxin. Adv. Cancer Res. 5 : 1 5 7 - 1 7 7 (1958). W. Nakahara and F. Fukuoka. Toxohormone: A characteristic toxic substance produced by cancer tissne. Gann 40:45-69 (1949). H. Endo, T. Sugimura, and K. Konno. Catalase-depressing factors: Toxohormone and kochsaft factor. Gann 46:51-56 (1955). M. C. Cnesta Ca~do, P. Salas, and E. H. Ramos. ~t-Glutamil transpeptidase activity in mice bearing Ehrlich ascites tumor cells. Cell. Mol. Biol. 2g: 293-298 (1982). A. E. Dumont, M. S. Nachbar, and A. B. Martelli. Altered
29.
30. 3 I. 32.
33.
34. 35,
36.
37.
38.
erythrocyte osmotic fragility in mice with Ehrtich ascites tumor J. Natl. Cancer. Inst. 159:989-991 (1977). E. Cadenas and H. Sies. Low level chemiluminescence as an indicator of singlet molecular oxygen in biological systems. Meth En:ymol, 105:221-231 (1984). E. Cadenas. H. Weffers. and H, Sies. Lob level chemiluminescence of isolated hepatocytes. Eur. J. Biochem. 119: 531536 (1981). K, U. Ingold. Peroxy radicals. Accounts Chem, Res. 211 i: 1-9 (1976) S. D. Aust and B. A. Svingen. The role of iron in enzymatic lipid peroxidation. In: Free Radicals in Biology. Volume V (W. A. Pryor. ed.), pp. 1-28, Academic Press, New York (1982). A. Boveris. E. Cadenas, and B. Chance. Ultraweak chemiluminescence: A sensitive assay for oxidative radical reactions. Federation Prec. 40: 195-198 ( 1981 ). J. W. Hastings and T. Wilson, Bioluminescence and chemiluminescence. Photochem. Photobiol. 23:461-473 (1976). O. M. M, Faria Oliveira. M. Haun, N. Duran, P. J. O'Brien, C. O'Brien, E. J. H. Bechara, and G. Cilento. Enzyme-generated electronically excited carbonyl compounds. J. Biol. Chem. 253:4707-4712 (1978). C, S. Foote. Photosensitized oxidation and singlet oxygen: Consequences in biological systems. In: Free Radicals in Biology. Volume II (W. A. Pryor, ed.), pp. 85-133, Academic Press. New York (1976). C. G. Fraga, S. F. Llesuy. and A. Boveris. Increased carbon tetrachloride-stimulated chemiluminescence in the in situ liver of barbital treated mice. Acta Physiol. Pharmacol. Latinoam. 34:41-48 (1984). E. Cadenas. A. Boveris. and B. Chance. Low level chemiluminescence of hydroperoxide-supplemented cytochrome c. Biochem. J. 187:131-140 (1980).
INCREASED
LIVER
0748-5514/85 $3.00+ .00 © 1985 Pergamon Press Ltd.
CHEMILUMINESCENCE
IN TUMOR-BEARING
MICE
ALBERTO BOVERIS, SUSANA F. LLESUY, a n d C~SAR G . FRAGA lnstituto de Qufmica y Fisicoqu/mica Biol6gicas (UBA-CONICET), Facultad de Farmacia y Bioqufmiea, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina (Received 22 October 1984, Revised 19 February 1985; Accepted 22 February 1985)
Abstract--Spontaneous mouse liver chemiluminescence (109 -+ 6 cps/cm 2) was increased in the early phase after tumor implantation in a distant position with respect to the liver. A 39% increased liver chemiluminescence was observed after 5 days of the injection of Ehrlich ascites tumor cells into the peritoneal cavity, and a 64% and a 46% increased liver chemiluminescence were measured after 8 and 14 days of the implantation of a fibrosareoma and of an adenocarcinoma, respectively, in the leg. At the time of maximal stimulation of in vivo liver chemiluminescence by the distant tumors, cytosolic superoxide dismutase, catalase, and glutathione peroxidase activities were decreased by 18%, 38%, and 26% in the liver of mice bearing Ehrlich ascites tumors. The same three enzymatic activities were decreased by 21%, 19%, and 54% respectively, in the liver of fibrosarcoma-bearing mice. Total liver glutathione was decreased by 18% to 22% in the tumor-bearing animals. Hydroperoxide-initiated chemiluminescence was increased in the homogenates (105% and 45%) and mitochondria (64% and 34%) from the liver of mice bearing Ehrlich ascites tumors and fibrosarcomas, respectively, at the time of maximal in situ liver chemiluminescence. The hydroperoxide-initiated chemiluminescence of liver microsomes was decreased by 46% to 36% in the tumor-bearing animals at the same time. It is concluded that the liver of tumor-bearing animals is subjected, during the early phase after tumor implantation, to an oxidative stress with increased steady-state levels of peroxyi radicals, which are essentially responsible for the increased photoemission observed in vivo. Keywords--Free radicals, Tumor implantation, Antioxidant enzymes, Mitochondria, Cancer, Chemiluminescence, Lipid peroxidation, Microsomes
port by Neifakh s described increased chemiluminescence and lipoperoxide content in liver and brain homogenates from tumor-bearing mice and a report by Inaba et al.9 described an increased chemiluminescence in blood samples from humans with carcinomas. However, there are no integrative studies on tumor-bearing animals in which in situ liver chemiluminescence as an indication of in vivo lipid peroxidation ~° is measured simultaneously with the activity of antioxidant enzymes and the content of endogenous antioxidants. In this paper we report an increased in situ liver chemiluminescence in the early phase after tumor implantation in tumor-bearing mice which is associated with a decreased activity of the protective antioxidant enzymes and an increased hydroperoxide-initiated chemiluminescence in the homogenates and the mitochondria.
INTRODUCTION
The primary production of superoxide anion and hydrogen peroxide at the mitochondrial and endoplasmic reticulum membranes seems to support the endogenous rate of hydroperoxide production and lipid peroxidation. The process appears mediated by hydroxyl radical generation, which initiates a free radical reaction chain leading to lipid peroxidation. The enzymes superoxide dismutase, catalase, and glutathione peroxidase seem to constitute the main protective system to minimize the rate of hydroxyl radical generation and the free radical reaction of lipid peroxidation. Decades ago, Greenstein et al. 2-4 reported a decreased catalase activity in the non-tumor containing livers and kidneys of tumor-bearing mice and rats. The decrease in catalase activity was observed in animals bearing a series of tumors and was reversible upon surgical removal of the tumor. ~5 More recently, Tarakhovsky et al. 6 and Oberley 7 have reported a decrease in superoxide dismutase activity in the livers of tumorbearing mice. It is then apparent that the livers of tumorbearing animals are subjected to a decreased antioxidant protection. Concerning lipid peroxidation, a brief re-
MATERIALS AND METHODS
Animals Female inbred BALB/c mice, 2 to 3 months old and weighing 15 to 20 g from the Experimental Leukemia 131
132
A. BovEms, S. F. LLESUY, and C. G. FRAGA
Section, National Academy of Medicine (Buenos Aires, Argentina) were used. Animals were supplied with water and laboratory animal food ad libitum.
Tumors Ehrlich ascites tumor cells were implanted by intraperitoneal injection of 0.5 x 10 7 cells/mouse. Fibrosarcoma cells, originated by foreign body tumorigenesis, z~ and adenocarcinoma cells, derived from a mammary tumor, were implanted as small solid tumors of 1 to 2 mm 3 in an incision in the leg.
Liver chemiluminescence Control and tumor-bearing mice were anesthetized intraperitoneally with urethane 12% w/v at a dose of 0.1 ml/10 g of weight; the liver surface was exposed by laparotomy and chemiluminescence was measured with a Johnson Foundation photon-counter t°-~z (Johnson Research Foundation, University of Pennsylvania, Philadelphia, PA). The emission is expressed as cps/cm 2 (of liver surface). In the case of mice bearing Ehrlich ascites tumors, the ascites and the tumor cells were removed from the peritoneal cavity before the measurement of liver chemiluminescence.
Subcellular fractions Subcellular fractions from control and tumor-bearing mice were prepared from animals that were not subjected to laparotomy to avoid any effect of anesthesia and surgery on the properties of the isolated organelles. Mouse liver homogenates were prepared in a medium consisting of 140 mM KCI, 1 mM EDTA, 5 mM TrisHC1, pH 7.3, and centrifuged at 600 g for 10 min to discard nuclei and cell debris. The supernatant, a suspension of mixed and preserved organelles, was used as "homogenate."t3-t4 Mitochondrial and microsomal fractions were isolated by conventional procedures. I*-t5 The 105,000 g supernatant was used as " c y tosol" to determine cytosolic superoxide dismutase activity. The operations were carried out at 0 to 2°C. Protein was assayed by the Lowry et al. ~6method using bovine serum albumin as standard. The protein content of subcellular fractions was calculated from the ratio of cytochrome oxidase activity in the homogenate and in the isolated organelles (mitochondria) and from the isolated microsomal protein, as described previously. 14
Hydroperoxide-initiated chemiluminescence of subcellular fractions The hydroperoxide-initiated chemiluminescence of homogenates, mitochondria, and microsomes ~3.t4"uT'us
were measured in a Packard Tri-Carb model 3320 liquid scintillation counter at room temperature in the out-ofcoincidence mode. ~ Samples (4 ml of final volume) were placed in 13-mm diameter and 40-mm height flasks which were placed inside 25-mm diameter and 58-mm height glass vials. The emission is expressed in cps/ mg of protein. The background level of emission of the empty flasks and vials was 80 to 100 cps. The reaction medium consisted of 140 mM KC1, 1 mM EDTA. 5 mM Tris-HC1, pH 7.3. Chemiluminescence measurements were started by addition of 3 mM tert-butyl hydroperoxide. The protein content was adjusted at 1 nag' ml of reaction medium for homogenates and mitochondria and at 0.25 mg/ml for microsomes.
Superoxide dismutase Cytosolic superoxide dismutase activity was determined from the inhibition of the rate of autocatatytic adrenochrome formation ~9 in a reaction medium containing 1 mM epinephrine and 50 mM glycine-NaOH, pH 9.6.14
Catalase The catalase activity of the liver homogenates was determined by measuring the decrease in 240 nm absorption in a reaction medium consisting of 50 mM phosphate buffer, pH 7.2, 2 mM H202, 1% Triton X100, and 0.1 to 0.3 mg of protein/ml, determining the pseudo-first order reaction constant (k') of the decrease in H2Oz absorption (ez4o = 40 M - I cm-~) 2° and expressing the results as catalase content in mol/g tissue. The value o f k = 4.6 × 10 7 M - I S-t was used as the second-order reaction constant for pure catalase. '-~
Glutathione perox~dase This activity was determined following NADPH oxidation at 340 nm in the presence of 0.17 mM GSH, 0.2 U/ml yeast glutathione reductase, and 0.5 mM tertbutyl hydroperoxide.
Glutathione content Total glutathione (GSH plus 2 GSSG) was determined in liver homogenates after precipitation with 2% perchloric acid and using yeast glutathione reductase and 5-5'-dithio.bis-nitrobenzoic acid. 2:
Chemicals Yeast glutathione reductase, NADPH, reduced glutathione, epinephrine, 5-5'-dithio-bis-nitrobenzoic acid, and Triton X- 100 were purchased from Sigma Chemical
Chemiluminescence and cancer Co. (Saint Louis, MO); tert-butyl hydroperoxide was from Aldrich Chemical Co. (Milwaukee, WI); hydrogen peroxide was from Baker Chemical Co. (Phillipsburg, NJ); and urethane was from B.D.H. (London, England). Other reagents were of analytical grade.
133
Table i. Spontaneous Liver Chemiluminescenceof Tumor-Bearing Mice Tumor
Statistics
None (control) (n = 30) Ehrlich ascites (day 5; n = 12) Fibrosarcoma(day 8; n = 12) Adenocarcinoma (day 14; n = 12)
The figures in the text and tables indicate mean values --- S.E.M. Differences were analyzed by the variance test (ANOVA Test; Texas Instruments).
n = numberof animals. *p < .05 **p < .01
RESULTS
Liver chemiluminescence The spontaneous liver chemiluminescence of tumorbearing mice was significantly increased during the first days of growth of a distant tumor. Mice bearing either Ehrlich ascites tumors, or fibrosarcomas, or adenocarcinomas showed 39%, 64%, and 46% increased liver photoemission, respectively, at 5, 8, and 14 days after tumor implantation (Fig. 1 and Table 1). It is worth noting that the increased liver chemiluminescence was produced in an organ free of metastasis and by a distant tumor; the fibrosarcoma and the adenocarcinoma were implanted in the leg, whereas the ascites tumor cells were implanted in the abdominal cavity. The increase in liver chemiluminescence was related to tumor development; maximal liver emission was observed when the Ehrlich ascites tumors reached 18% to 22% of the body weight and fibrosarcomas and adenocarcinomas reached 3% to 5% of the body weight. The state of maximal emission was a transient state, after reaching a maximal level of the liver chemiluminescence of the 200
W o Z I,LI
_z,,~ loo X W
00
(*) 10 (o) 20 (e) TIME(doys)
I
10 20 30
Fig. !. Spontaneouschemiluminescenceof the in situ liver of normal and tumor-bearingmice. (o), normal mice; (A), mice bearing Ehrlich ascites tumor; (<3),mice bearing fibrosarcoma;and (m), mice bearing adenocarcinoma. The abscissa indicates the time after tumor implantation. Exposed liver areas were about I cm-'. Symbols indicate mean values and bars indicate S.E.M.
Chemiluminescence (cps/cm2) 109± 152 ± 179 ± 159 ±
6 19" 17"* 21'
tumor-bearing animals returned a few days later to the level showed by normal mice (Fig. 1). Mice bearing Ehrlich ascites tumor, fibrosarcoma, and adenocarcinoma died 7 to 12, 12 to 24, and 26 to 44 days, respectively, after tumor implantation.
Hydroperoxide-initiated chemiluminescence of liver homogenates and mitochondrial and microsomal suspensions The protein content of the liver subcellular fractions and the liver weight of tumor-bearing mice are shown in Table 2. The homogenate protein content was 23% decreased in the livers of mice bearing Ehrlich ascites tumors. Liver weights and the content of subcellular fractions were not different from control values in all of the other cases (Table 2). When tert-butyl hydroperoxide is added to either mouse liver homogenates or mitochondrial or microsomal suspensions, there is a burst of photoemission with different kinetics depending on the type of subcellular preparation, j3.~4.jT,18 Mouse liver homogenates show a maximum of chemiluminescence after 15 min of the addition of 3 mM tert-butyl hydroperoxide for tumor-beating mice and after 20 min for normal mice (Fig. 2). This emission declined at 20 and 25 min, respectively, and then slowly increased to reach a steady state of maximal chemiluminescence which occurs after 60 to 90 min. The liver homogenates from mice bearing either Ehrlich ascites tumors or fibrosarcomas gave 105 % and 45%, respectively, more maximal chemiluminescence than the liver homogenates from normal mice (Fig. 2 and Table 3). The mitochondria isolated from the liver of mice bearing either Ehrlich ascites tumors or fibrosarcomas gave 64% and 32%, respectively, more emission than the mitochondria isolated from the liver of normal mice, again measured at the point of maximal photoemission, after 10 to 20 min of the hydroperoxide addition (Fig. 3 and Table 3). At variance, the microsomes isolated from the liver of tumor-bearing mice were less effective in yielding hydroperoxide-initiated light emission as compared with the same organelles from normal mice.
134
A. BOVERIS, S. F. LLESUY, and C. G. FRAGA
Table 2. Liver Weight and Protein Content of the Liver Subcellular Fractions of Tumor-Bearing Mice Homogenates
Mitochondria
Liver weight (g)
Tumor
1.14 1.77 0.99 1.09
None (control) (n = 20) Ehrlich ascites (day 5; n = 12) Fibrosarcoma (day 8; n = 12) Adenocarcinoma (day 14; n = 12)
+ m -
Microsomes
(mg of protein/g liver)
0.03 0.19" 0.07 0.08
78.7 60.4 78.2 75.6
+ -+ -+
3.2 4.2 3.6 4.1
24.6 22.0 21.9 21.3
+ +-+ :'=
2+3 1.5 2.4 1.8
13.3 15.2 16.4 15.'9
: 09 + 1.7 -+ I+0 -..+- 1.2
n = number of animals. *p < .05
Maximal chemiluminescence, measured at 30 to 90 s after hydroperoxid¢ addition, was diminished by 46% and 38% in the liver microsomes from mice bearing either Ehrlich ascites tumors of fibrosarcomas (Fig. 4 and Table 3). There were not significant differences in the kinetics of the hydroperoxide-initiated chemiluminescence of homogenates, mitochondria, and microsomes in comparing the subcellular fractions obtained from normal and tumor-hearing mice. Chemiluminescence was measured at day 5 and at day 8 for Ehrlich ascites tumor- and fibrosarcoma-bearing mice, respectively.
Superoxide dismutase, catalase, and glutathione peroxidase In the days of maximal emission, these three antioxidants enzymes showed diminished activities in the livers of tumor-bearing mice (Table 4). The activity of cytosolic superoxide dismutase was decreased by 18% in the case of mice bearing Ehrlich ascites tumors and
Table 3. Hydmpcroxide-lnitiatedChemiluminescence of Liver SubcellularFractionsfrom Tumor-Bearing Mice Chemiluminescence of Subcellular Fraction Homogenate
Tumor
Mitochondria
(cps/mg protein)
None (control) Ehrlich ascites (day 5) Fibrosarcoma (day 8)
228 --- 25 467 -- 23* 330 --. 22*
806 ± 52 1320 -- 23* 1061 ± 27*
by 21% in mice bearing fibrosarcomas. Catalase activity was decreased by 38% and 19%, whereas glutathione peroxidase activity was decreased by 26% and 54% in the livers of mice bearing Ehrlich ascites tumors and fibrosarcomas, respectively. Total glutathione content
Tumor-bearing mice showed a diminished content of total liver glutathione which was reduced by 18%
1500r
~
~
2000
W¢~
¢j ~
-
,+o
I
890 -- 48 483 -+ 32* 554 - 28*
*p < .001. Nine animals in each group.
500
i
Microsomes
.
.
.
.
.
i 000
,000 / l r
-
\\
°o
i & ,..o..,..,
II
soo
t - BOOH ( mM I
('l |
l
+0
5
I
I
l
10 15 20 TIME (min)
I
I
25
30
0
10
20 30 40 T I M E (rain)
50
60
Fig. 2. tert-Butyl hydroperoxide-initia~'d chemiluminescence of liver homogenates from normal and tumor-hearing mice. (o), normal mice; (&), mice bearing Ehrlich ascitcs tumor; and (0), mice bearing
Fig. 3. tert-Butyl hydroperoxide-initiated chemiluminescence of liver mitochondria from normal and tumor-bearing mice. Symbols as in
fibrosarcoma.
Fig. 2,
Chemiluminescence and cancer
z.
tO00
"7'
O,.
-r
I
0
2
I
I
I
4 6 8 T IME (mini
I
10
,/j
I
60
Fig. 4. tert-Butyl hydroperoxide-initiated chemiluminescence of liver microsomes from normal and tumor-bearing mice. (o), normal mice; (A), mice bearing Ehrlich ascites tumor; and (e), mice bearing fibrosarcoma.
and by 22% in mice bearing Ehrlich ascites tumors and fibrosarcomas, in days 5 and 8 after tumor implantation, respectively (Table 4).
Correlation of liver chemiluminescence with the activity of protective enzymes The liver surface chemiluminescence showed a negative, statistically significative correlation with the activity of superoxide dismutase, glutathione peroxidase, and the content of total glutathione for the cases of control mice and mice bearing Ehrlich ascites tumors and fibrosarcomas. A qualitative correlation, statistically not significant, was observed between liver chemiluminescence and catalase activity (Fig. 5).
135
substance did not decrease catalase activity of liver homogenates z5 or the purified enzyme, z6 These studies were not further pursued. Other examples of altered structure and functions in distance organs of tumor-bearing mice are the decreased "y-glutamyi transferase activity with increased vacuolization in the liver, z7 and the decreased osmotic fragility and progressive decrease in the red cells content zs reported in mice bearing Ehrlich ascites tumors. Liver chemiluminescence affords an indirect assay for the in vivo steady-state level of peroxyl radicals; the assay is non-invasive, non-destructive, and specific for the organ. The results presented in this paper show that the livers of tumor-bearing mice exhibit an increased chemiluminescence in the early phase after tumor implantation. The increased photoemission reflects an increased level of oxyl and peroxyl radicals that can be understood on the basis of free-radical reactivity as an oxidative stress to the biological structure. Although it is possible to have chemiluminescence without lipid peroxidation in chemically defined and cell-free systems, 29 it is established that an increase in lipid peroxidation rate in organs and isolated cells produces a parallel increase in photoemission ~°.12.3°. It is important to consider the tissue absorption of the emitted light. Preliminary work estimates that the measured light comes from the outer 50 to 100 am of the liver; consequently, the assay would have limited validity if considerable differences exist between the activity in the outermost cells and the rest of the tissue. The reactions that lead to the spontaneous chemilurfainescence of in situ liver are understood as follows: H202 + 02 ~
DISCUSSION
The idea that tumors release a toxic substance that affects distant organs was put forward by Lucke et al. ,z3 who showed reduced catalase activity in the liver of parabiotic rats. Greenstein et al. z-4 reported that tumor extracts supressed liver catalase activity and Nakahara and Fukuoka 24 isolated a substance from neoplastic tissues, which they called toxohormone, that lowered liver catalase activity when injected to mice. However, this
) "OH + OH- + 02
(1)
~ H20 + R.
(2)
~ ROO.
(3)
2 ROO.
~ RO + ROH + ~O2
(4)
2 ROO.
~ 02 + ROH + RO*
(5)
•OH + RH R. + 02
Table 4. Superoxide Dismutase, Catalase, and Glutathione Peroxidase Activity and Glutathione Content in the Liver of Tumor-Bearing Mice
Tumor
Superoxide Dismutase (U/g liver)
Catalase (nmol/ liver)
Glutathione Peroxidase (U/g liver)
Glutathione Content (~umol/g liver)
None (control) (n = 20) Ehrlich ascites (day 5 ; n = 12) Fibrosarcoma (day 8; n = 12)
174 ± 8 143 ± 9** 137 ~ II**
0 . 9 0 ± 0.06 0.56 ± 0.03** 0.73 ± 0.03**
10.0 ± 0.7 7.4 ± 0.8* 4.6 ± 0.4**
7.6 ± 0.2 6.2 ± 0.3* 5.9 ± 0.3*
= number of animals. *p < .01 **p < .001
n
136
A. BOVERIS. S. F. LLESUY. and C. G. F ~ a 6 a
,oo
..o
l'°:
2--
"-:t-"°t
-
V°°I 0i .
•
%0
'
,~o
'
1;o
'
,;o
'
~;o
ro
o°
LIVER CHENILUHINESCENCE (cp$/cm z )
Fig, 5. Correlations between surface liver chemiluminescence and catalase, su~cro×ide dismutase and glutathione peroxidase activities and glutathione content in the liver of normal and tumor-bearing mice. Correlations values are: r = .93 and p < .05 for superoxide d i s m u t a s e activity; r = .92 and p < .05 for glutathione peroxidase activity: r = .89 and p < .05 for glutathione content; and r = .62 and p > .05 for catalase activity (dashed line).
H\
O--O H~[C_C[/H
/H
R/C'-C\R + IO2
"~ R/
\R
' (6)
+
H\
O--O H\I I/H
/H ,
,
OO"
(7)
H\
+ :=o
CO* + 02
, to2 + R / C = O
(8)
The initiation reaction is the hydroxyl radical generation, in a chelated iron-catalyzed Haber-Weiss reaction (Eq. I). Most of superoxide anion and hydrogen peroxide are generated in the membranes of mitochondria and of endoplasmic reticulum.' The hydroxyl radical will abstract hydrogen from the hydrocarbon chain of membrane polyunsaturated fatty acids (Eq. 2) yielding aikyl radicals which react with oxygen yielding peroxyl radicals (Eq. 3 ) . 31-32 These reactions will initiate a free-radical chain reaction. The termination reactions of peroxyl radicals appear essential for the photoemission process observed during lipid peroxidation, 3°-33 these reactions yield either singlet molecular oxygen (Eq. 4) or excited carbonyl groups
(Eq. 5) which are the photoemissive species. 29 Excited carbonyl groups can also be generated in the decomposition of dioxetane derivatives 34.35 formed by either addition of singlet oxygen to unsaturated hydrocarbon chains (Eq. 6 ) 34.36 o r cyclization of unsaturated tertiary peroxyl and alkyl biradicals (Eq, 7). It is interesting that singlet oxygen can generate excited carbonyl groups through dioxetane formation (Eq. 6) and excited carbonyis can produce singlet oxygen after coilisional quenching (Eq. 8)? 6 A first defense in the protective system against oxidative stress is provided by the enzymes superoxide dismutase, catalase, and glutathione peroxidase that catalyze reactions that diminish the steady-state concentrations of superoxide anion and hydrogen peroxide which are the reactants for hydroxyl radical production (Eq. 1). The activity of these antioxidant enzymes is diminished in the liver of tumor-bearing mice in the early phase of tumor development. Moreover, surface liver chemiluminescence shows a negative correlation, statistically significant, with tissue levels of superoxide dismutase, glutathione peroxidase, and total glutathione in normal and tumor-bearing mice. The diminished activity of Cu-Zn superoxide dismutase agrees with a report by Tarakhovsky 6 on decreased activity of total and Cu-Zn superoxide dismutases. Oberley 7 has reported just the: opposite, a lowered Mn-superoxide dismutase activity with a variable Cu-Zn superoxide dismutase activity. The simultaneous decrease in the activity of superoxide dismutase, catalase, and glutathione peroxidase seems to indicate that the three antioxidant enzymes are regulated synchronically; the three of them were observed increased in barbital treated mice. 37 For the in vitro hydroperoxide-initiated chemiluminescence, the hemoprotein-promoted radical for-
Chemiluminescence and cancer m a t i o n from a d d e d h y d r o p e r o x i d e ( t - B O O H ) affords the initiation reactions (Eqs. 9 - 1 l).ts'38
137
cies specific, it is p o s s i b l e that s u p p l e m e n t a t i o n with b i o l o g i c a l l y active antioxidants may afford an effective c o a d j u v a n t therapy for t u m o r - b e a r i n g m a m m a l s .
t-BOOH + Fe 3+ t-BOO- + H + + Fe 2÷ t-BOOH + Fe '+
(9) Acknowledgements--The authors wish to express their gratitude to
-~
t-BOO.(t-BO.) + RH
t-BO. + H O - + Fe ~+
(10)
> t-BOOH(t-BOH) + R.
(11)
The p e r o x y l and a l c o x y l radicals react with polyunsaturated fatty acids through Equation 11, eventually l e a d i n g to the generation o f excited species similar to the in v i v o p h o t o e m i s s i o n (Eqs. 3 - 8 ) . A s e c o n d defense system against increased levels o f o x y l radicals is p r o v i d e d by free-radical scavengers, such as r e d u c e d glutathione (in the h y d r o p h i l i c domains) and vitamins A and E (in the h y d r o p h o b i c domains). T h e s e free-radical scavengers ( A H ) act acc o r d i n g to the f o l l o w i n g reactions: ROO" + G S H 2 GS" R" + AH
2 A.
~ ROOH + GS"
(12)
' GSSG
(13)
~ RH + A. , A - A
(14) (15)
The i n c r e a s e d rates o f h y d r o p e r o x i d e - i n i t i a t e d chem i l u m i n e s c e n c e o b s e r v e d in h o m o g e n a t e s and mitoc h o n d r i a from the liver o f t u m o r - b e a r i n g mice could be u n d e r s t o o d as a l o w e r level of e n d o g e n o u s antioxidants and free-radical scavengers in the tissue. ~.~.~4Thus, a d e c r e a s e in the tissue level o f antioxidant substances will also contribute to an increased rate o f lipid pero x i d a t i o n and p h o t o e m i s s i o n in vivo. The integration o f the in v i v o and in vitro o b s e r v a t i o n s indicate an e n h a n c e m e n t o f lipid p e r o x i d a t i o n and a d i m i n i s h e d level o f antioxidant substances and antioxidant enz y m e s in the livers of t u m o r - b e a r i n g mice. The t u m o r effect m a y be p r o d u c e d either by the release o f a substance or by abstraction o f some blood c o m p o n e n t ( s ) which is essential for the maintenance and p r o p e r function o f the distant t i s s u e ? A l t h o u g h our results do not e x c l u d e either, or both p o s s i b i l i t i e s , the increase in the h y d r o p e r o x i d e - i n i t i a t e d c h e m i l u m i n e s cence o b s e r v e d in w a s h e d m i t o c h o n d r i a appears more l i k e l y to indicate an abstraction than an addition o f some substance by the tumors. If the o x i d a t i v e stress o b s e r v e d in the early phase after t u m o r implantation in the mouse liver is not spe-
Dr. Christianne Dosne de Pasqualini of the National Academy of Medicine (Buenos Aires, Argentina) for her encouragement and the generous supply of tumor-bearing animals and to Dr. Martha Dubin for the glutathione determinations. This research was supported by grants from CONICET (Consejo Nacional de lnvestigaciones Cienffficas y T(~cnicas, Argentina) and Fundaci6n Cherny. Alberto Boveris is a Career Investigator and C(~sarG. Fraga is a fellow of CONICET.
REFERENCES
I. B. Chance, H. Sies, and A. Boveris. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59:527-605 (1979). 2. J. P. Greenstein, W. V. Jenrette, and J. White. The liver catalase of tumor-bearing rats and the effect of extirpation of the tumors. J. Biol. Chem. 141:327-328 (1941). 3. J. P. Greenstein and H. B. Andervoont. Note on the liver catalase activity of pregnant mice and mice bearing growing embrionic implants. J. Natl. Cancer Inst. 4:283-284 (1943). 4. J. P. Greenstein. Biochemistry of Cancer, pp. 316-366, Academic Press, New York (1947). 5. H. Bush. An Introduction to the Biochemistry of the Cancer Cell, pp. 241-256, Academic Press, New York (1962). 6. A. M. Tarakhovskii, G. V. Glinskii, and V. A. Shliakhovenko. Superoxide dismutase activity in tumor and liver tissues of tumor-bearing animals. Ukr. Biochem. J. 52:628-632 (1980). 7. L. W. Oberley. Superoxide dismuta~ and cancer. In: Superoxide Dismutase, Volume II (L. W. Oberley, ed.), pp. 127-165, CRC Press, Boca Raton, FL (1982). 8. Y. A. Neifakh. Free radical mechanism of ultraweak chemiluminescence coupled with peroxide oxidation of unsaturated fatty acids. Biofizika 16:584-587 (1971). 9. H. Inaba, A. Yamagishi, C. Takyu, B. Yoda, Y. Goto, T. Miyazawa, T. Kaneda, and A. Saeki. Development of an ultra-high sensitive photon counting system and its application to biomedical measurements. Optics and Lasers in Engineering 3: 125130 (1982). 10. A. Boveris, E. Cadenas, R. Reiter, M. Filipkowsky, Y. Nakase, and B. Chance. Organ chemiluminescence: Non-invasive assay for oxidative radical reactions. Proc. Natl. Acad. Sci. U.S.A. 77:347-351 (1980). I I. C. D. Pasqualini, L. Sen, E Sadl, L. Schwartz, and L. Z. Tkaczevski. Tumor development in mice bearing a plastic cylinder and inoculated with human neoplastic cells. J. Natl. Cancer. Inst. $1:283-286 (1973). 12. L. A. Videla, C. G. Fraga, O. R. Koch, and A. Boveris. Chemiluminescence of the in situ rat liver after acute ethanol intoxication. Effect of ( + )-cyanidanol-3. Biochem. Pharmacol. 32: 2822-2825 (1983). 13. E. Cadenas, A. I. Varsavsky, A. Boveris, and B. Chance. Oxygen or organic hydroperoxide-induced chemiluminescence of brain and liver homogenates. Biochem. J. 198:645-652 (1981). 14. A. Boveris, C. G. Fraga, A. I. Varsavsky, and O. R. Koch. Increased chemiluminescence, and superoxide production in the liver of chronically ethanol-treated rats. Arch. Biochem. Biophys. 227:534-541 (1983). 15. A. Boveris, N. Oshino, and B. Chance. The cellularproduction of hydrogen peroxide. Biochem. J. 12,8:617-630 (1972). 16. O. H. Lowry, N. 1. Rosebrough, A. L. Farr, and R. J. Randall.
138
17.
18.
19. 20.
21.
22.
23.
24. 25. 26. 27. 28.
A. BOVERIS, S. F. LLESUY. and C G. FR~c~, Protein measurement with Folin phenol reagent. J, Biol. Chem. 193:265-275 (1951). E. Cadenas, A. Boveris, and B. Chance. Low level chemiluminescence of bovine heart submitochondrial particles. Biochem. J. 186:659-667 (1980). E. Cadenas and H. Sies. Low level chemiluminescence of liver microsomal fractions initiated by tert-butyl hydroperoxide. Eur. J. Biochem. 174:349-356 (1982). H. P. Misra and I. Fridovich. The role of superoxide anion in the autooxidation of epinephrine and a simple assay for superoxide dismutase. J. Biol. Chem. 747:3170-3175 (1972). B. Chance. Special methods: Catalase. In: Methods of Biochemical Analysis (E. Glick, ed.), pp. 408-424, Interscience Publishers, New York, NY (1954). H. Sies, T. Bucher, N. Oshino, and B. Chance. Heine occupance of catalase in hemoglobin-free perfused rat liver and of isolated rat liver catalase. Arch. Biochem. Biophys. 154:106-116 (1973). F. Tietze. Enzymatic method for the quantitative determination of nanogram amounts of total and oxydized glutathion.e.. Anal. Biochem. 27:502-522 (1969). B. Lucke, M. Bet'wick, and I. Zeckwer. Liver catalase activity in parabiotic rats with one partner tumor-bearing. J. Natl. Cancer Inst. 13:681-686 (1952). W. Nakaharaand F. Fukuoka. The newerconcept ofcancer toxin. Adv. Cancer Res. 5 : 1 5 7 - 1 7 7 (1958). W. Nakahara and F. Fukuoka. Toxohormone: A characteristic toxic substance produced by cancer tissne. Gann 40:45-69 (1949). H. Endo, T. Sugimura, and K. Konno. Catalase-depressing factors: Toxohormone and kochsaft factor. Gann 46:51-56 (1955). M. C. Cnesta Ca~do, P. Salas, and E. H. Ramos. ~t-Glutamil transpeptidase activity in mice bearing Ehrlich ascites tumor cells. Cell. Mol. Biol. 2g: 293-298 (1982). A. E. Dumont, M. S. Nachbar, and A. B. Martelli. Altered
29.
30. 3 I. 32.
33.
34. 35,
36.
37.
38.
erythrocyte osmotic fragility in mice with Ehrtich ascites tumor J. Natl. Cancer. Inst. 159:989-991 (1977). E. Cadenas and H. Sies. Low level chemiluminescence as an indicator of singlet molecular oxygen in biological systems. Meth En:ymol, 105:221-231 (1984). E. Cadenas. H. Weffers. and H, Sies. Lob level chemiluminescence of isolated hepatocytes. Eur. J. Biochem. 119: 531536 (1981). K, U. Ingold. Peroxy radicals. Accounts Chem, Res. 211 i: 1-9 (1976) S. D. Aust and B. A. Svingen. The role of iron in enzymatic lipid peroxidation. In: Free Radicals in Biology. Volume V (W. A. Pryor. ed.), pp. 1-28, Academic Press, New York (1982). A. Boveris. E. Cadenas, and B. Chance. Ultraweak chemiluminescence: A sensitive assay for oxidative radical reactions. Federation Prec. 40: 195-198 ( 1981 ). J. W. Hastings and T. Wilson, Bioluminescence and chemiluminescence. Photochem. Photobiol. 23:461-473 (1976). O. M. M, Faria Oliveira. M. Haun, N. Duran, P. J. O'Brien, C. O'Brien, E. J. H. Bechara, and G. Cilento. Enzyme-generated electronically excited carbonyl compounds. J. Biol. Chem. 253:4707-4712 (1978). C, S. Foote. Photosensitized oxidation and singlet oxygen: Consequences in biological systems. In: Free Radicals in Biology. Volume II (W. A. Pryor, ed.), pp. 85-133, Academic Press. New York (1976). C. G. Fraga, S. F. Llesuy. and A. Boveris. Increased carbon tetrachloride-stimulated chemiluminescence in the in situ liver of barbital treated mice. Acta Physiol. Pharmacol. Latinoam. 34:41-48 (1984). E. Cadenas. A. Boveris. and B. Chance. Low level chemiluminescence of hydroperoxide-supplemented cytochrome c. Biochem. J. 187:131-140 (1980).