Mutation Research, 237 (1990) 173-181
173
Elsevier MUTAGI 09055
Hyperoxia-induced clonogenic killing of HeLa cells associated with respiratory failure and selective inactivation of Krebs cycle enzymes W.G.E.J. Schoonen 1, A.H. Wanamarta 1, J.M. van der Klei-van Moorsel 2, C. Jakobs 2 and H. Joenje 1 I Institute of Human Genetics, Free Unioersity and 2 Department of Pediatrics, Free Unioersity Hospital, Amsterdam (The Netherlands)
(Received 3 May 1990) (Revision received28 June 1990) (Accepted 10 July 1990)
Keywords: Oxygentoxicity; Krebs cycle enzymes; Respiratory failure; HeLa cells; Hyperoxia; Mitochondria; Energy metabolism
Summary Cellular intoxication by elevated concentrations of 0 2 may be considered as a model for accelerated cellular aging processes resulting from excessive free radical production by normal metabolic pathways. We describe here that exposure of HeLa cell cultures to 80% 0 2 for 2 days causes progressive growth inhibition and loss of reproductive capacity. This intoxication was correlated with inhibition of cellular 0 2 consumption and inactivation of 3 mitochondrial flavoproteins, i.e., partial inactivation of N A D H and succinate dehydrogenases and total inactivation of a-ketoglutarate dehydrogenase. As a-ketoglutarate dehydrogenase controls the influx of glutamine/glutamate into the Krebs cycle, which is the major pathway for oxidative ATP generation in HeLa cells, the inactivation of a-ketoglutarate dehydrogenase was expectedly correlated with a net fall in glutamine/glutamate utilization. Furthermore, a simultaneous increase in glucose consumption and lactate production was observed, indicating that the cellular response to respiratory failure is to generate more ATP from glycolysis. In spite of this response, extensive depletion of ATP was observed. Thus, hyperoxia-induced growth inhibition and loss of clonogenicity seem to be due primarily to an impairment of mitochondrial energy metabolism resulting from inactivation of SH-groupcontaining flavoprotein enzymes localized at or near the inner mitochondrial membrane. These observations may be relevant for theories implicating loss of mitochondrial function as a prime factor in the aging process.
According to the free radical theory of aging, loss of cellular functions, a characteristic of the aging process, is a consequence of accumulating subcellular damage inflicted by reactive oxygen
Correspondence: Dr. H. Joenje, Institute of Human Genetics, Free University, Van der Boechorststraat 7, 1081 BT Amsterdam (The Netherlands).
species ('oxygen radicals') produced by normal aerobic metabolism (Harman, 1981; Mehlhorn and Cole, 1985). Oxygen radicals may be produced by enzymes localized in several compartments of the cell, i.e., cytosol, endoplasmic reticulum, and peroxisomes. Mitochondria are supposed to be a prime source of endogenous oxygen radicals, as isolated mitochondria are known to be capable of generating superoxide ( O f ) by monovalent reduc-
0921-8734/90/$03.50 © 1990 ElsevierScience Publishers B.V. (BiomedicalDivision)
174 tion of oxygen at the level of N A D H dehydrogenase, succinate dehydrogenase and ubiquinone cytochrome c reductase (Forman and Kennedy, 1974; Forman and Boveris, 1982; Turrens et al., 1985). Dismutation of superoxide yields hydrogen peroxide, which in the presence of suitable transition metals can generate the highly destructive hydroxyl radical ('OH). Exposure of cells to elevated levels of oxygen leads to enhanced fluxes of oxygen radicals at those sites where they are also produced under normal circumstances (Fridovich, 1989). By studying cells under conditions of hyperoxic stress one could thus hope to identify the consequences of increased physiological fluxes of free radicals, in terms of cellular homeostatic responses, subcellular degenerative phenomena and critical target sites. Previous studies by Balin et al. (1976, 1984a) and others on human diploid fibroblasts have documented a profound effect of oxygen concentration (in the range of 20-100% 02 , i.e., normobaric hyperoxia) on various growth characteristics. Increased oxygen concentration slows down the proliferation rate, by causing the cells to accumulate in the G 2 phase of the cell cycle (Balin et al., 1978; Poot et al., 1988), and shortens the in vitro life span (Balin et al., 1984b). Thaw et al., (1984) showed that density-arrested human glial cells, when cultured under elevated oxygen concentrations (20-40% O2), accumulate lipofuscinlike material, which could be prevented by exogenous antioxidants. In proliferating cell populations high oxygen levels are known to be genotoxic as well (Joenje, 1989). Although such effects of normobaric hyperoxia on cell cultures are now well recognized, little is known about the intracellular target sites that play a primary role in oxygen poisoning. To study this subject, permanent (transformed) cell lines have some advantage over primary cell cultures because of their superior growth characteristics and cloning efficiency. In addition, such cells are committed to continued proliferation, without the tendency to 'escape' toxicity by shifting to a nonproliferating viable state. This simplifies the assessment of cellular intoxication in cycling cells, which can be conveniently monitored at the level of clonogenic capacity. Moreover, a constitutive
cycling state demands a type of metabolism in which catabolic and anabolic pathways are tightly interconnected, so that cycling cells are expected to be highly sensitive to conditions that interfere with major pathways of energy metabolism. In the present study we document the adverse effect of hyperoxic conditions on the growth rate and reproductive capacity of HeLa cells and correlate these effects to a severe impairment of mitochondrial energy metabolism and subsequent ATP depletion. Respiratory impairment appears to be associated with a selective inactivation of 3 SH-group-containing flavoprotein dehydrogenases, localized at or near the inner mitochondrial membrane, which causes the electron transport chain to become deprived of reducing equivalents. Our data thus suggest that the inner mitochondrial membrane, a n d / o r components associated with it, constitutes a vulnerable target for reactive oxygen species produced by normal metabolic pathways. Materials and methods
Chemicals
All chemicals were of analytical grade and obtained from Sigma (St. Louis, MO, U.S.A.) or Boehringer (Mannheim, F.R.G.), except K z H P O 4, KH2PO 4, MgC12, MgSO4, EDTA and KCN, which were obtained from Merck (Darmstadt, F.R.G.), sucrose (Baker, Deventer, The Netherlands), potassium ferricyanide (Fluka, Buchs, Switzerland) and glucose and succinic acid (BDH Chemicals, Dorset, U.K.). Cell culture
HeLa cells were cultured under an atmosphere of air/2% CO 2 (normoxia) at 3 7 ° C in 75-cm 2 sealed polystyrene flasks containing 30 ml Ham's F-10 medium (Flow Labs., Irvine, Scotland), supplemented with 1 mM L-glutamate and 10% heatinactivated (30 min, 56 ° C) fetal calf serum (Flow) (complete culture medium). The method of gassing the cultures was as described before (Joenje et al., 1985). HeLa cells were routinely subcultured twice a week by trypsinization and seeding at a starting density of 7 × 105 cells/flask. Cultures were free from mycoplasma contamination, as checked at monthly intervals.
175
Experimental protocol
Oxygen consumption
HeLa cells were seeded in multiple 75-cm2 flasks (7 × 105 cells/flask) and cultured under normoxia over a period of 3 days. After 1 and 2 days subseries of flasks were gassed with an atmosphere of 80% O2/18% N2/2% CO2 (hyperoxia). All cells from 1 flask were harvested by trypsinization, suspended in 10 ml of complete culture medium and counted in a Coulter counter. Simultaneously, plating efficiencies of all cell cultures were determined by seeding 400-800 cells into Petri dishes, which were incubated at 37 ° C with 10 ml of complete culture medium. After 7 days, colonies were fixed with acetic acid and methanol (1:3, v/v), stained with Giemsa, and counted manually. In the experiments presented below the effects of 1- and 2-day hyperoxic exposure time were examined at a 3-day culturing time (see Fig. 1).
At a 3-day culturing time, cells were trypsinized, suspended in 11 ml of culture medium and counted in a Coulter counter. Subsequently, cells were centrifuged (800 × g, 5 min) and resuspended in buffer A (measurements with intact cells): 100/xl Tris-HC1 (25 mM, pH 7.4), CaC12 (1 mM), MgC12 (1 mM), NaC1 (110 mM) and KC1 (6.7 mM), or buffer B (measurements with digitonin-permeabilized cells): 100 /~1 Tris-HC1 (25 mM, pH 7.4), 0.25 M sucrose, 2 mM EDTA, 5 mM MgC12 and 10 mM K 2 H P O 4. Oxygen consumption measurements were carried out with a Clark electrode (Yellow Springs Instruments, Yellow Springs, OH) with intact or digitonin-permeabilized cells, according to Schoonen et al., (1990). With digitonin-permeabilized cells the activities of the NADH, succinate and a-glycerophosphate oxidase complexes were determined, as well as the oxygen consumption rate after uncoupling with 2,4-dinitrophenol (DNP) in the presence of the substrate a-glycerophosphate.
10"
A.
Spectrophotometric enzyme assays 0
1
~ 0.4' 1
2
;
~
Culturing time (days)
3
80"
._o .-
6040"
~-
20"
°o
Culturing time (days)
Fig. 1. Proliferation (A) and relative plating efficiency (B) of HeLa cells as a function of culturing time during normoxia (O) and the effect of shifting some of the cultures to hyperoxia for 1 ( o ) and 2 (O) days. Results are means of 4 independent measurements. Absolute plating efficiencies at normoxia were 80.5 + 3.4, 79.2 + 10.5 and 84.1 + 7.9 (means + SEM) at 1, 2 and 3 days of culturing time, respectively. Standard deviations for the total number of cells per flask were typically between 10 and 25%.
At a 3-day culturing time, cells were trypsinized and suspended in 10 ml of complete culture medium, followed by centrifugation (800 x g, 5 rain). The pellet was resuspended in 250 #l ice-cold Tris-HC1 (25 mM, p H 7.4), supplemented with 0.25 M sucrose and 2 mM EDTA, after which the suspension was sonicated with a sonifier (40,000 V, 3 × 20 s, with 40-s intervals). During the experiments, the lysates were kept on ice. All enzyme assays were carried out as described by Schoonen et al. (1990), with the exception of the N A D H dehydrogenase assay, in which 0.04% cytochrome c was used instead of 0.02%. Enzyme activities were calculated on the basis of cellular protein, determined according to Bradford (1976), with bovine serum albumin as a standard.
Assays for glucose, lactate, amino acids and ammonia Net rates of uptake or excretion of glucose, lactate, several amino acids and ammonia were calculated from changes in concentration as observed between fresh culture medium added to the cultures on day 2 and the same medium after 1 day of culturing, i.e., on day 3. On days 2 and 3 parallel cultures were trypsinized and counted to
176 correct for cell growth, as d e s c r i b e d b y R u e c k e r t a n d M u e l l e r (1960). A s t a n d a r d a u t o m a t e d assay p r o c e d u r e was used for glucose a n d a m m o n i a , an assay kit f r o m Boehringer ( M a n n h e i m , F . R . G . ) for that o f L-lactate, a n d an a m i n o acid a n a l y z e r for the p r e s e n t e d a m i n o acids.
A TP determination Cellular A T P levels were a s s a y e d b y biol u m i n o m e t r y , as d e s c r i b e d b y G i l l e et al. (1989).
2.5
_~
2.0
r,.
o
1.s
~o"
1.0
0.5'
Results
Effect of hyperoxia on cell growth and plating efficiency A s illustrated in Fig. 1, e x p o s u r e to h y p e r o x i a resulted in i n h i b i t i o n of cell p r o l i f e r a t i o n a n d a progressive r e d u c t i o n of p l a t i n g efficiency.
Oxygen consumption." intact cells H y p e r o x i c e x p o s u r e for 1 a n d 2 d a y s resulted in a progressive decrease in the rate of o x y g e n c o n s u m p t i o n (Fig. 2). U n d e r n o r m o x i a , c o u p l i n g of electron t r a n s p o r t to A T P synthesis was d e m o n s t r a t e d b y a 1.25-fold increase in r e s p i r a t i o n after a d d i t i o n of the u n c o u p l i n g agent D N P . Rem a r k a b l y , cells e x p o s e d to h y p e r o x i a failed to r e s p o n d to D N P . Theoretically, the lack of responsiveness to D N P could result from either true
~2 o" "~ E
1
0
-DNP
+DNP
Fig. 2. Hyperoxia-induced respiratory failure in HeLa cells. The effect of hyperoxia for 1 or 2 days on cellular oxygen consumption rates was assessed at 3 days of culturing time (cf. Fig. 1). Bars indicate from left to right: normoxic cells and cells kept at hyperoxia for 1 and 2 days, respectively. Results are means + SEM; n = 8. e,o significantly different from 0 and 1 day of hyperoxic exposure, respectively. Single, double and triple symbols (see also Figs. 3-5) indicate probabilities of P < 0.05, P < 0.01 and P < 0.001, respectively (Student's twotailed t-test). DNP, 2,4-dinitrophenol.
0.0
INTACT CELLS
+DIG
NADH SUCCINATE O(-GP OXIOASE OXIOASE OXlDASE
+DNP
Fig. 3. Key respiratory components in HeLa cells exposed to normoxia or hyperoxia, for 1 or 2 days. Digitonin treatment was used to measure the activities of the NADH oxidase, succinate oxidase and a-glycerophosphate oxidase complexes, in the absence and presence of DNP (25 ~tM), as described by Schoonen et al. (1990). Results are in means+ SEM; n = 8. For bars and symbols, see legend to Fig. 2. DIG, digitonin.
u n c o u p l i n g o r a lack of r e d u c i n g equivalents. As i n d i c a t e d below, the latter p o s s i b i l i t y seems to a p p l y here.
Oxygen consumption: digitonin-permeabilized cells P e r m e a b i l i z a t i o n of cells with d i g i t o n i n a c c o r d ing to a t e c h n i q u e p r e v i o u s l y d e s c r i b e d b y G r a n g e r a n d L e h n i n g e r (1982) e n a b l e d us to q u a n t i f y the capacities of the 3 c o m p l e x e s that deliver electrons to the r e s p i r a t o r y chain, i.e., the N A D H , succinate a n d et-glycerophosphate o x i d a s e complexes, b y m e a s u r i n g 02 c o n s u m p t i o n after successive add i t i o n of s u b s t r a t e s a n d inhibitors. Fig. 3 shows that cellular e x p o s u r e to h y p e r o x i a resulted in a significant decrease in o x y g e n c o n s u m p t i o n of intact a n d d i g i t o n i n - t r e a t e d cells, the effect b e i n g m o s t severe after 2 d a y s of h y p e r o x i c exposure. These c h a n g e s c o i n c i d e d with s i m u l t a n e o u s p r o gressive r e d u c t i o n s in the o x y g e n c o n s u m p t i o n rates due to the N A D H a n d s u c c i n a t e o x i d a s e complexes, which were again m o s t severe after 2 d a y s of h y p e r o x i c exposure. O n the o t h e r h a n d , the activity of the c t - g l y c e r o p h o s p h a t e o x i d a s e c o m p l e x was u n a f f e c t e d b y h y p e r o x i a , or even slightly a c t i v a t e d after 1 d a y of exposure. Strikingly, D N P clearly s t i m u l a t e d a - g l y c e r o p h o s -
177
p h a t e - d r i v e n oxygen c o n s u m p t i o n in h y p e r o x i a e x p o s e d cells, in c o n t r a s t to the oxygen c o n s u m p tion e x h i b i t e d b y intact h y p e r o x i a - e x p o s e d cells (cf. Fig. 1). This result indicates (i) that the a g l y c e r o p h o s p h a t e o x i d a s e c o m p l e x along with a m a j o r p a r t of the r e s p i r a t o r y chain, i.e., u b i q u i n o n e c y t o c h r o m e c reductase, c y t o c h r o m e c oxidase a n d A T P synthetase, is c o m p l e t e l y resistant to the c o n s e q u e n c e s of h y p e r o x i c exposure, a n d (ii) that the d i h y d r o x y a c e t o n e p h o s p h a t e / t x - g l y c e r o p h o s p h a t e shuttle p a t h w a y as a source of reducing equivalents seems to b e h a r d l y or not at all utilized in H e L a cells. T h e l a t t e r s t a t e m e n t is b a s e d n o t o n l y o n the o b s e r v a t i o n that H e L a cell r e s p i r a t i o n is i m p a i r e d u n d e r high oxygen even though exogen o u s ~ t - g l y c e r o p h o s p h a t e c a n a p p a r e n t l y be o x i d i z e d after p e r m e a b i l i z a t i o n with digitonin; in a d d i t i o n , we f o u n d that after b l o c k i n g the respiration of H e L a cells with r o t e n o n e plus m a l o n a t e r e s p i r a t i o n c o u l d n o t be r e s t o r e d b y a d d i n g excess glucose (W.G.E.J. Schoonen, u n p u b l i s h e d results).
15-
e-
D..
• •
.
NADH DH
SUCCINATE DH
(~-GP DH
•O •O •O
(~-KG DH
Fig. 4. Activities of NADH, succinate, a-glycerophosphate and a-ketoglutarate dehydrogenases in homogenates from HeLa cells exposed to normoxia or hyperoxia, for 1 or 2 days. Results are means+SEM; n =8. For bars and symbols, see legend to Fig. 2. DH, dehydrogenase; a-GP, t~-glycerophosphate; a-KG, a-ketoglutarate.
Spectrophotometric analysis I n p a r a l l e l e x p e r i m e n t s , the effect of h y p e r o x i a o n key enzymes of the r e s p i r a t o r y chain a n d tric a r b o x y l i c acid cycle was e v a l u a t e d b y assaying cell h o m o g e n a t e s using s t a n d a r d s p e c t r o p h o t o m e t ric p r o c e d u r e s . A significant, d o s e - i n d e p e n d e n t , i n a c t i v a t i o n of N A D H d e h y d r o g e n a s e a n d succ i n a t e d e h y d r o g e n a s e was o b s e r v e d in h o m o g -
enates f r o m cells e x p o s e d to h y p e r o x i a (Fig. 4), whereas the activity of m i t o c h o n d r i a l a - g l y c e r o p h o s p h a t e d e h y d r o g e n a s e was t o t a l l y unaffected. O n the o t h e r h a n d , h y p e r o x i a d r a m a t i c a l l y red u c e d the activity of a - k e t o g l u t a r a t e d e h y d r o genase in a t i m e - d e p e n d e n t m a n n e r , to a level that was b e l o w the limit of d e t e c t i o n after 2 d a y s o f h y p e r o x i c exposure. This i n a c t i v a t i o n led us to also e x a m i n e s u b u n i t E3 o f this e n z y m e c o m p l e x ,
TABLE 1 KEY ENZYMES OF GLYCOLYSIS AND SEVERAL OTHER METABOLIC PATHWAYS a Enzymes Lipoamide dehydrogenase NADH diaphorase Hexokinase Phosphofructokinase Glyceraldehyde-3-phosphate DH a-GlycerophosphateDH(cyt) Pyruvatekinase Lactate dehydrogenase Glucose-6-phosphate DH Glutathione reductase
Time of hyperoxia (days) 0
1
2
43.3 + 1.1 1407 + 41 22.8 4- 1.2 157.0 4- 4.0 819 4- 26 2038 4- 81 2669 4-120 2175 4- 85 539 ± 25 71.8 4- 1.0
38.9 4- 1.3 * 1425 4- 33 22.9 4- 0.8 159.6 4- 7.5 876 4- 33 2098 4- 77 2664 4- 71 2310 +107 556 4- 27 69.3 4- 2.7
31.2 + 1.2" * *'°°° 1439 + 71 23.6 4- 1.1 170.2 4- 8.5 831 + 42 2102 4- 98 2900 4- 84 2186 +113 541 + 21 62.3 4- 2.7" *
a Cells were cultured at normoxia or hyperoxia (1 or 2 days) and analyzed at 3 days of culturing time (cf. Fig. 1). Enzyme activities are expressed as n m o l e s u b s t r a t e / m i n / m g protein (mean + S E M ; n = 8); D H , d e h y d r o g e n a s e ; cyt, cytoplasmic. *'°Significantly different from normoxic cells and 1-day hyperoxia-exposed cells, respectively. Single, double and triple symbols indicate probabifities of P < 0.05, P < 0.01 and P < 0.001, respectively (Student's two-tailed t-test).
178
which exhibits 2 separate activities, a lipoamide dehydrogenase and a N A D H diaphorase activity. As shown in Table 1, hyperoxia did not affect N A D H diaphorase activity, but reduced the activity of the lipoamide dehydrogenase moiety; this effect may at least partly account for the observed inactivation of the entire a-ketoglutarate dehydrogenase complex. The observed suppression of oxidative metabolism apparently led to an enhanced ATP generation from glycolysis, as indicated by substantially increased glucose consumption and lactate excretion (see below). Analysis of the activities of hexokinase, phosphofructokinase, glyceraldehyde3-phosphate dehydrogenase, pyruvate kinase and lactate dehydrogenase did not reveal any significant changes in enzyme levels (Table 1), indicating that the enhanced glycolytic activity was entirely due to activation of allosteric enzymes. In view of the suspected importance of cellular NADPH and glutathione in oxidative stress we also measured the NADPH- and glutathione-recycling enzymes, glucose-6-phosphate dehydrogenase, a key metabolic enzyme of the pentose phosphate cycle, and glutathione reductase. Hyperoxia did not affect glucose-6-phosphate dehydrogenase activity and had only a marginal effect on glutathione reductase after 2 days of hyperoxia. Glucose, lactate, amino acids and ammonia
The relative contributions of glycolysis and oxidative metabolism to cellular energy supply are grossly reflected in the cellular utilization of sub-
100-
80' m
60
40 E c
20 0
Fig. 5. ATP levels of HeLa cells exposed to normoxia or hyperoxia, for ] or 2 days. Results are means_+SEM: n = 8 . For bars and symbols, see legend to Fig. 2.
strates and the excretion of products. In HeLa cells exogenous glucose is the main carbon source for glycolysis, with lactate and alanine as the main end products, while glutamine and glutamate are the most important carbon fuels for oxidative metabolism, with aspartate as one of the end products (Donnelly and Scheffler, 1976; Reitzer et al., 1979; Stanisz et al., 1983; Brand et al., 1989). Exposure to hyperoxia for 2 days resulted in a substantial increase in glucose consumption and lactate production, while exposure for 1 day had only a moderate effect (Table 2). Glutamine consumption and alanine production were unaffected by hyperoxic exposure, while a clear rise in
TABLE 2 EFFECTS OF H Y P E R O X I A ON C E L L U L A R C O N S U M P T I O N / E X C R E T I O N OF G LU C O S E, LACTATE, SEVERAL A M I N O ACI DS A N D A M M O N I A a Compound
Glucose Lactate Glutamine Glutamate Alanine Aspartate Ammonia
Time of hyperoxia (days) 0
1
2
- 6 680 ± 410 11480 _+355 -1132_+ 28 469_+ 13 212_+ 8 -54_+ 4 423-+ 6
- 7 350 + 500 13 410 _+ 65 * * - 1 1 9 3 + 30 515_+ 8** 250_+ 20 -55+ 3 449-+ 23
- 12111 _+ 1 760 *'° 17 577 + 1 045 * * ,o -1109+ 19 776_+ 16 ***'°~x~ 259+ 16 -77-+ 8 *`o 336-+ 37"°
Cells were cultured under normoxic conditions or exposed to hyperoxia for 1 or 2 days and analyzed at 3 days of culturing time (cf. Fig. 1). Concentration changes are in n m o l e / h / 1 0 7 cells ( m e a n s + S E M ; n = 4); positive sign is production; negative sign is consumption. For symbols, see Table 1.
179 glutamate production and aspartate consumption was observed after 2 days of hyperoxia. Simultaneously with these changes, the production of ammonia dropped, indicating that the influx of glutamine/glutamate into the citric acid cycle was reduced. A T P levels
To assess the net effect of impaired oxidative energy metabolism and stimulated glycolysis on the energy status of the cells, cellular ATP pools were measured as a function of hyperoxic exposure time. As shown in Fig. 5, exposure for 1 and 2 days resulted in 33% and 82% depletion of ATP, respectively. Discussion
Our results show that HeLa cells exposed to normobaric hyperoxia exhibit a decreased proliferation rate and a reduced plating efficiency, which appeared to be correlated with a progressive decline in respiratory activity. Further analysis revealed that respiratory failure was associated with selective inactivation of at least 3 mitochondrial SH-group-containing flavoprotein complexes, i.e., NADH dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase. Inactivation of the latter enzyme clearly dominated the fall in respiratory activity of intact cells, by causing a lack of NADH-reducing equivalents and consequently an apparent loss of responsiveness to the uncoupler DNP. However, in cells permeabilized with digitonin, a normal respiration rate could be induced in oxygen-poisoned cells by adding a-glycerophosphate, in which case also a clear responsiveness to DNP was restored (Fig. 3). This showed that hyperoxia did not adversely affect the electron transport chain between aglycerophosphate dehydrogenase, ubiquinone cytochrome c reductase (complex III), cytochrome c oxidase (complex IV) and ATP synthetase (complex V). The observation that inactivation of NADH dehydrogenase, succinate dehydrogenase and particularly a-ketoglutarate dehydrogenase is correlated with the loss of cell viability under hyperoxic conditions suggests that these enzymes are critical targets in oxygen poisoning.
The strong dependence of oxidative metabolism in cell cultures on glutamine a n d / o r glutamate consumption (Donnelly and Scheffler, 1976; Zielke et al., 1978; Stanisz et al., 1983; Reitzer et al., 1979), which is controlled by the influx of glutamate into the tricarboxylic acid cycle via the apparently oxygen-sensitive a-ketoglutarate dehydrogenase complex, is in line with the observed net reduction of glutamine/glutamate uptake during hyperoxic exposure. To maintain a normal energy supply an activation of glycolysis was to be expected. Such a response was indeed observed, as evidenced by an enhanced glucose consumption and lactate production. Although increased glycolysis during hyperoxia has been reported previously by Rueckert and Mueller (1960), Allen and Rasmussen (1974) and Balin et al., (1976), a clear link with an inhibition of cellular respiration was not demonstrated. Moreover, the present study shows that the hyperoxia-induced metabolic shift from oxidative metabolism towards glycolysis was not due to increased synthesis of key regulatory glycolytic enzymes, but rather to the activation of an allosteric enzyme, such as phosphofructokinase. This enzyme is known to be inhibited by ATP, the level of which clearly dropped after 1 day of hyperoxic exposure (Fig. 5). Remarkably, similar effects of normobaric hyperoxia on oxidative and glycolytic energy metabolism were observed in Chinese hamster ovary cells (Schoonen et al., 1990), suggesting that the present observations may be universally valid. The 3 enzyme complexes shown here to be inactivated by hyperoxia, i.e., NADH dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase, are localized at or near the mitochondrial inner membrane. They are equipped with essential sulfhydryl groups and, at least in case of the former 2, with iron-sulfur centers, elements that would predispose these enzymes to being vulnerable for endogenous reactive species, i.e., oxygen radicals or lipid peroxidation products (Balentine, 1982; Haugaard, 1968). Since reactive oxygen species are excessively generated by N A D H dehydrogenase, succinate dehydrogenase and ubiquinone cytochrome c reductase (Forman and Kennedy, 1974; Forman and Boveris, 1982; Turrens et al., 1985), especially under hyperoxic conditions (Jamieson, 1989; Zweier et al.,
180
1989), such a mechanism of inactivation appears to be quite plausible. In the case of lipid peroxidation, it is relevant to note that mitochondrial D N A replication occurs in association with the inner mitochondrial membrane (Nass et al., 1965; Shearman and Kalf, 1975, 1977; Albring et al., 1977). This close association of mitochondrial DNA with a membrane that is supposed to be an important cellular site of free radical production (Chance et al., 1979) raises the possibility that mitochondrial DNA, just like N A D H dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase, is subject to the attack by oxygen free radicals a n d / o r lipid peroxidation products, such as the cross-linking agent malondialdehyde (Fleming et al., 1982). Since mitochondrial DNA is not protected by histone proteins, while mitochondria are less efficient in repairing D N A damage and replication errors, the mitochondrial genome may be more susceptible to the damaging effects of free radicals and alkylating agents than the nuclear genome (Richter, 1988; Richter et al., 1988). Thus, hyperoxia may be a valuable model to evaluate the genotoxic impact of metabolically produced free radicals at the level of mitochondrial DNA.
Acknowledgements This research was supported by the Foundation for Fundamental Biological Research (BION), which is subsidized by the Netherlands Organization for Scientific Research (NWO).
References Albring, M., J. Griffith and G. Attardi (1977) Association of a protein structure of probable membrane derivation with HeLa cell mitochondrial DNA near its origin of replication, Proc. Natl. Acad. Sci. (U.S.A.), 74, 1348-1352. Allen, J.E., and H. Rasmussen (1971) The effects of oxygen on cellular metabolism, Int. J. Clin. Pharmacol., 5, 26-33. Balentine, J.D. (1982) Pathology of Oxygen Toxicity, Academic Press, New York, NY. Balin, A.K., D.B.P. Goodman, H. Rasmussen and V.J. Cristofalo (1976) The effect of oxygen tension on the growth and metabolism of WI-38 cells, J. Cell. Physiol., 89, 235250. Balin, A.K., D.B.P. Goodman, H. Rasmussen and V.J. Cristofalo (1978) Oxygen-sensitive stages of the cell cycle of human diploid cells, J. Cell Biol., 78, 390-400.
Balin, A.K., A.J. Fisher and D.M. Carter (1984a) Oxygen modulates growth of human cells at physiologic partial pressures, J. Exp. Med., 160, 152 166. Balin, A.K., A.F. Fischer, I. Leong and D.M. Carter (1984b) Physiologic oxygen tensions modulate the lifespan of human cells, Age, 7, 142. Bradford, M.M. (1976) A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72. 248-254. Brand, K., W. Fekl, J. Von Hintzenstern, K. Langer, P. Luppa and C. Shoerner (1989) Metabolism of glutamine in lymphocytes, Metabolism, 38, 29 33. Chance, B., H. Sies and A. Boveris (1979) Hydroperoxide metabolism in mammalian organs, Physiol. Rev., 59, 527605. Donnelly, M., and I.E. Scheffler (1976) Energy metabolism in respiration-deficient and wild type Chinese hamster fibroblasts in culture, J. Cell. Physiol., 89, 39-52. Fleming, J.E.. J. Miquel, S.F. Cottrell, L.S. Yengoyan and A.C. Economos (1982) Is cell aging caused by respiration-dependent injury to the mitochondrial genome?, Gerontology, 28, 44-53. Forman, H.J., and A. Boveris (1982) Superoxide radical and hydrogen peroxide in mitochondria, in: W.A. Pryor (Ed.), Free Radicals in Biology, Vol. 5, Academic Press, New York, NY, pp. 65-90. Forman, H.J., and J.A. Kennedy (1974) Role of superoxide radical in mitochondrial dehydrogenase reactions, Biochem. Biophys. Res. Commun., 60, 1044-1050. Fridovich, I. (1989) Superoxide dismutases. An adaptation to a paramagnetic gas, J. Biol. Chem., 264, 7761-7764. Gille, J.J.P., C.G.M. van Berkel, E. Mullaart, J. Vijg and H. Joenje (1989) Effects of lethal exposure to hyperoxia and to hydrogen peroxide on NAD(H) and ATP pools in Chinese hamster ovary cells, Mutation Res., 214, 89-96. Granger, D.L., and A.L. Lehninger (1982) Sites of inhibition of mitochondrial electron transport in macrophage-injured neoplastic cells, J. Cell Biol., 95, 527-535. Harman, D. (1981) The aging process, Proc. Natl. Acad. Sci. (U.S.A.), 78, 7124 7128. Haugaard, N. (1968) Cellular mechanisms of oxygen toxicity, Physiol. Rev., 48, 311-373. Jamieson, D. (1989) Oxygen toxicity and reactive oxygen metabolites in mammals, Free Radical Biol. Med., 7, 87108. Joenje, H. (1989) Genetic toxicology of oxygen, Mutation Res., 219, 193-208. Joenje, H., J.J.P. Gille, A.B. Oostra and P. van der Valk (1985) Some characteristics of hyperoxia-adapted HeLa cells. A tissue culture model for cellular oxygen tolerance, Lab. Invest., 52, 420-428. Mehlhorn, R.J., and G. Cole (1985) The free radical theory of aging: a critical review, Adv. Free Radical Biol. Med., 1, 165-223. Nass, M.M.K., S. Nass and B. Afzelius (1965) The general occurrence of mitochondrial DNA, Exp. Cell Res., 37, 516-539.
181 Poot, M., D. Schindler, M. Kubbies, H. Hoehn and P.S. Rabinovitch (1988) Bromodeoxyuridine amphfies the inhibitory effect of oxygen on cell proliferation, Cytometry, 9, 332-338. Reitzer, L.J., B.M. Wice and D. Kennell (1979) Evidence that glutamine, not sugar, is the major energy source for cultured HeLa cells, J. Biol. Chem., 254, 2669-2676. Richter, C. (1988) Do mitochondrial DNA fragments promote cancer and aging?, FEBS Lett., 241, 1-5. Richter, C., J.W. Park and B.N. Ames (1988) Normal oxidative damage to mitochondrial and nuclear DNA is extensive, Proc. Natl. Acad. Sci. (U.S.A.), 85, 6465-6467. Rueckert, R.R., and G.C. Mueller (1960) Effect of oxygen tension on HeLa cell growth, Cancer Res., 20, 944-949. Schoonen, W.G.E.J., A.H. Wanamarta, J.M. van der Klei-van Moorsel, C. Jakobs and H. Joenje (1990) Respiratory failure and stimulation of glycolysis in Chinese hamster ovary cells exposed to normobaric hyperoxia, J. Biol. Chem., 265, 11118-11124. Shearman, C.W., and G.F. Kalf (1975) DNA synthesis by a membrane-DNA complex from rat liver mitochondria, Biochem. Biophys. Res. Commun., 63, 712-721. Shearman, C.W., and G.F. Kalf (1977) DNA replication by a
membrane DNA complex from rat liver mitochondria, Arch. Biochem. Biophys., 182, 573-586. Stanisz, J., B.M. Wice and D.E. Kennell (1983) Comparative energy metabolism in cultured heart muscle and HeLa cells, J. Cell. Physiol., 115, 320-330. Thaw, H.H., V.P. Collins and U.T. Brunk (1984) Influence of oxygen tension, pro-oxidants and antioxidants on the formation of lipid peroxidation products (lipofuscin) in individual cultivated human glial cells, Mech. Ageing Dev., 24, 211-223. Turrens, J.F., A. Alexander and A.L. Lehninger (1985) Ubisemiquinone is the electron donor for superoxide formation by complex III of heart mitochondria, Arch. Biochem. Biophys., 237, 408-414. Zielke, H.R., P.T. Ozand, J.T. Tildon, D.A. Sevdalian and M. Cornblath (1978) Reciprocal regulation of glucose and glutamine utilization by cultured human diploid fibroblasts, J. Cell. Physiol., 95, 41-48. Zweier, J.L., S.S. Duke, P. Kuppusamy, J.T. Sylvester and E.W. Gabrielson (1989) Electron paramagnetic resonance evidence that cellular oxygen toxicity is caused by the generation of superoxide and hydroxyl free radicals, FEBS Lett., 252, 12-16.
173
Elsevier MUTAGI 09055
Hyperoxia-induced clonogenic killing of HeLa cells associated with respiratory failure and selective inactivation of Krebs cycle enzymes W.G.E.J. Schoonen 1, A.H. Wanamarta 1, J.M. van der Klei-van Moorsel 2, C. Jakobs 2 and H. Joenje 1 I Institute of Human Genetics, Free Unioersity and 2 Department of Pediatrics, Free Unioersity Hospital, Amsterdam (The Netherlands)
(Received 3 May 1990) (Revision received28 June 1990) (Accepted 10 July 1990)
Keywords: Oxygentoxicity; Krebs cycle enzymes; Respiratory failure; HeLa cells; Hyperoxia; Mitochondria; Energy metabolism
Summary Cellular intoxication by elevated concentrations of 0 2 may be considered as a model for accelerated cellular aging processes resulting from excessive free radical production by normal metabolic pathways. We describe here that exposure of HeLa cell cultures to 80% 0 2 for 2 days causes progressive growth inhibition and loss of reproductive capacity. This intoxication was correlated with inhibition of cellular 0 2 consumption and inactivation of 3 mitochondrial flavoproteins, i.e., partial inactivation of N A D H and succinate dehydrogenases and total inactivation of a-ketoglutarate dehydrogenase. As a-ketoglutarate dehydrogenase controls the influx of glutamine/glutamate into the Krebs cycle, which is the major pathway for oxidative ATP generation in HeLa cells, the inactivation of a-ketoglutarate dehydrogenase was expectedly correlated with a net fall in glutamine/glutamate utilization. Furthermore, a simultaneous increase in glucose consumption and lactate production was observed, indicating that the cellular response to respiratory failure is to generate more ATP from glycolysis. In spite of this response, extensive depletion of ATP was observed. Thus, hyperoxia-induced growth inhibition and loss of clonogenicity seem to be due primarily to an impairment of mitochondrial energy metabolism resulting from inactivation of SH-groupcontaining flavoprotein enzymes localized at or near the inner mitochondrial membrane. These observations may be relevant for theories implicating loss of mitochondrial function as a prime factor in the aging process.
According to the free radical theory of aging, loss of cellular functions, a characteristic of the aging process, is a consequence of accumulating subcellular damage inflicted by reactive oxygen
Correspondence: Dr. H. Joenje, Institute of Human Genetics, Free University, Van der Boechorststraat 7, 1081 BT Amsterdam (The Netherlands).
species ('oxygen radicals') produced by normal aerobic metabolism (Harman, 1981; Mehlhorn and Cole, 1985). Oxygen radicals may be produced by enzymes localized in several compartments of the cell, i.e., cytosol, endoplasmic reticulum, and peroxisomes. Mitochondria are supposed to be a prime source of endogenous oxygen radicals, as isolated mitochondria are known to be capable of generating superoxide ( O f ) by monovalent reduc-
0921-8734/90/$03.50 © 1990 ElsevierScience Publishers B.V. (BiomedicalDivision)
174 tion of oxygen at the level of N A D H dehydrogenase, succinate dehydrogenase and ubiquinone cytochrome c reductase (Forman and Kennedy, 1974; Forman and Boveris, 1982; Turrens et al., 1985). Dismutation of superoxide yields hydrogen peroxide, which in the presence of suitable transition metals can generate the highly destructive hydroxyl radical ('OH). Exposure of cells to elevated levels of oxygen leads to enhanced fluxes of oxygen radicals at those sites where they are also produced under normal circumstances (Fridovich, 1989). By studying cells under conditions of hyperoxic stress one could thus hope to identify the consequences of increased physiological fluxes of free radicals, in terms of cellular homeostatic responses, subcellular degenerative phenomena and critical target sites. Previous studies by Balin et al. (1976, 1984a) and others on human diploid fibroblasts have documented a profound effect of oxygen concentration (in the range of 20-100% 02 , i.e., normobaric hyperoxia) on various growth characteristics. Increased oxygen concentration slows down the proliferation rate, by causing the cells to accumulate in the G 2 phase of the cell cycle (Balin et al., 1978; Poot et al., 1988), and shortens the in vitro life span (Balin et al., 1984b). Thaw et al., (1984) showed that density-arrested human glial cells, when cultured under elevated oxygen concentrations (20-40% O2), accumulate lipofuscinlike material, which could be prevented by exogenous antioxidants. In proliferating cell populations high oxygen levels are known to be genotoxic as well (Joenje, 1989). Although such effects of normobaric hyperoxia on cell cultures are now well recognized, little is known about the intracellular target sites that play a primary role in oxygen poisoning. To study this subject, permanent (transformed) cell lines have some advantage over primary cell cultures because of their superior growth characteristics and cloning efficiency. In addition, such cells are committed to continued proliferation, without the tendency to 'escape' toxicity by shifting to a nonproliferating viable state. This simplifies the assessment of cellular intoxication in cycling cells, which can be conveniently monitored at the level of clonogenic capacity. Moreover, a constitutive
cycling state demands a type of metabolism in which catabolic and anabolic pathways are tightly interconnected, so that cycling cells are expected to be highly sensitive to conditions that interfere with major pathways of energy metabolism. In the present study we document the adverse effect of hyperoxic conditions on the growth rate and reproductive capacity of HeLa cells and correlate these effects to a severe impairment of mitochondrial energy metabolism and subsequent ATP depletion. Respiratory impairment appears to be associated with a selective inactivation of 3 SH-group-containing flavoprotein dehydrogenases, localized at or near the inner mitochondrial membrane, which causes the electron transport chain to become deprived of reducing equivalents. Our data thus suggest that the inner mitochondrial membrane, a n d / o r components associated with it, constitutes a vulnerable target for reactive oxygen species produced by normal metabolic pathways. Materials and methods
Chemicals
All chemicals were of analytical grade and obtained from Sigma (St. Louis, MO, U.S.A.) or Boehringer (Mannheim, F.R.G.), except K z H P O 4, KH2PO 4, MgC12, MgSO4, EDTA and KCN, which were obtained from Merck (Darmstadt, F.R.G.), sucrose (Baker, Deventer, The Netherlands), potassium ferricyanide (Fluka, Buchs, Switzerland) and glucose and succinic acid (BDH Chemicals, Dorset, U.K.). Cell culture
HeLa cells were cultured under an atmosphere of air/2% CO 2 (normoxia) at 3 7 ° C in 75-cm 2 sealed polystyrene flasks containing 30 ml Ham's F-10 medium (Flow Labs., Irvine, Scotland), supplemented with 1 mM L-glutamate and 10% heatinactivated (30 min, 56 ° C) fetal calf serum (Flow) (complete culture medium). The method of gassing the cultures was as described before (Joenje et al., 1985). HeLa cells were routinely subcultured twice a week by trypsinization and seeding at a starting density of 7 × 105 cells/flask. Cultures were free from mycoplasma contamination, as checked at monthly intervals.
175
Experimental protocol
Oxygen consumption
HeLa cells were seeded in multiple 75-cm2 flasks (7 × 105 cells/flask) and cultured under normoxia over a period of 3 days. After 1 and 2 days subseries of flasks were gassed with an atmosphere of 80% O2/18% N2/2% CO2 (hyperoxia). All cells from 1 flask were harvested by trypsinization, suspended in 10 ml of complete culture medium and counted in a Coulter counter. Simultaneously, plating efficiencies of all cell cultures were determined by seeding 400-800 cells into Petri dishes, which were incubated at 37 ° C with 10 ml of complete culture medium. After 7 days, colonies were fixed with acetic acid and methanol (1:3, v/v), stained with Giemsa, and counted manually. In the experiments presented below the effects of 1- and 2-day hyperoxic exposure time were examined at a 3-day culturing time (see Fig. 1).
At a 3-day culturing time, cells were trypsinized, suspended in 11 ml of culture medium and counted in a Coulter counter. Subsequently, cells were centrifuged (800 × g, 5 min) and resuspended in buffer A (measurements with intact cells): 100/xl Tris-HC1 (25 mM, pH 7.4), CaC12 (1 mM), MgC12 (1 mM), NaC1 (110 mM) and KC1 (6.7 mM), or buffer B (measurements with digitonin-permeabilized cells): 100 /~1 Tris-HC1 (25 mM, pH 7.4), 0.25 M sucrose, 2 mM EDTA, 5 mM MgC12 and 10 mM K 2 H P O 4. Oxygen consumption measurements were carried out with a Clark electrode (Yellow Springs Instruments, Yellow Springs, OH) with intact or digitonin-permeabilized cells, according to Schoonen et al., (1990). With digitonin-permeabilized cells the activities of the NADH, succinate and a-glycerophosphate oxidase complexes were determined, as well as the oxygen consumption rate after uncoupling with 2,4-dinitrophenol (DNP) in the presence of the substrate a-glycerophosphate.
10"
A.
Spectrophotometric enzyme assays 0
1
~ 0.4' 1
2
;
~
Culturing time (days)
3
80"
._o .-
6040"
~-
20"
°o
Culturing time (days)
Fig. 1. Proliferation (A) and relative plating efficiency (B) of HeLa cells as a function of culturing time during normoxia (O) and the effect of shifting some of the cultures to hyperoxia for 1 ( o ) and 2 (O) days. Results are means of 4 independent measurements. Absolute plating efficiencies at normoxia were 80.5 + 3.4, 79.2 + 10.5 and 84.1 + 7.9 (means + SEM) at 1, 2 and 3 days of culturing time, respectively. Standard deviations for the total number of cells per flask were typically between 10 and 25%.
At a 3-day culturing time, cells were trypsinized and suspended in 10 ml of complete culture medium, followed by centrifugation (800 x g, 5 rain). The pellet was resuspended in 250 #l ice-cold Tris-HC1 (25 mM, p H 7.4), supplemented with 0.25 M sucrose and 2 mM EDTA, after which the suspension was sonicated with a sonifier (40,000 V, 3 × 20 s, with 40-s intervals). During the experiments, the lysates were kept on ice. All enzyme assays were carried out as described by Schoonen et al. (1990), with the exception of the N A D H dehydrogenase assay, in which 0.04% cytochrome c was used instead of 0.02%. Enzyme activities were calculated on the basis of cellular protein, determined according to Bradford (1976), with bovine serum albumin as a standard.
Assays for glucose, lactate, amino acids and ammonia Net rates of uptake or excretion of glucose, lactate, several amino acids and ammonia were calculated from changes in concentration as observed between fresh culture medium added to the cultures on day 2 and the same medium after 1 day of culturing, i.e., on day 3. On days 2 and 3 parallel cultures were trypsinized and counted to
176 correct for cell growth, as d e s c r i b e d b y R u e c k e r t a n d M u e l l e r (1960). A s t a n d a r d a u t o m a t e d assay p r o c e d u r e was used for glucose a n d a m m o n i a , an assay kit f r o m Boehringer ( M a n n h e i m , F . R . G . ) for that o f L-lactate, a n d an a m i n o acid a n a l y z e r for the p r e s e n t e d a m i n o acids.
A TP determination Cellular A T P levels were a s s a y e d b y biol u m i n o m e t r y , as d e s c r i b e d b y G i l l e et al. (1989).
2.5
_~
2.0
r,.
o
1.s
~o"
1.0
0.5'
Results
Effect of hyperoxia on cell growth and plating efficiency A s illustrated in Fig. 1, e x p o s u r e to h y p e r o x i a resulted in i n h i b i t i o n of cell p r o l i f e r a t i o n a n d a progressive r e d u c t i o n of p l a t i n g efficiency.
Oxygen consumption." intact cells H y p e r o x i c e x p o s u r e for 1 a n d 2 d a y s resulted in a progressive decrease in the rate of o x y g e n c o n s u m p t i o n (Fig. 2). U n d e r n o r m o x i a , c o u p l i n g of electron t r a n s p o r t to A T P synthesis was d e m o n s t r a t e d b y a 1.25-fold increase in r e s p i r a t i o n after a d d i t i o n of the u n c o u p l i n g agent D N P . Rem a r k a b l y , cells e x p o s e d to h y p e r o x i a failed to r e s p o n d to D N P . Theoretically, the lack of responsiveness to D N P could result from either true
~2 o" "~ E
1
0
-DNP
+DNP
Fig. 2. Hyperoxia-induced respiratory failure in HeLa cells. The effect of hyperoxia for 1 or 2 days on cellular oxygen consumption rates was assessed at 3 days of culturing time (cf. Fig. 1). Bars indicate from left to right: normoxic cells and cells kept at hyperoxia for 1 and 2 days, respectively. Results are means + SEM; n = 8. e,o significantly different from 0 and 1 day of hyperoxic exposure, respectively. Single, double and triple symbols (see also Figs. 3-5) indicate probabilities of P < 0.05, P < 0.01 and P < 0.001, respectively (Student's twotailed t-test). DNP, 2,4-dinitrophenol.
0.0
INTACT CELLS
+DIG
NADH SUCCINATE O(-GP OXIOASE OXIOASE OXlDASE
+DNP
Fig. 3. Key respiratory components in HeLa cells exposed to normoxia or hyperoxia, for 1 or 2 days. Digitonin treatment was used to measure the activities of the NADH oxidase, succinate oxidase and a-glycerophosphate oxidase complexes, in the absence and presence of DNP (25 ~tM), as described by Schoonen et al. (1990). Results are in means+ SEM; n = 8. For bars and symbols, see legend to Fig. 2. DIG, digitonin.
u n c o u p l i n g o r a lack of r e d u c i n g equivalents. As i n d i c a t e d below, the latter p o s s i b i l i t y seems to a p p l y here.
Oxygen consumption: digitonin-permeabilized cells P e r m e a b i l i z a t i o n of cells with d i g i t o n i n a c c o r d ing to a t e c h n i q u e p r e v i o u s l y d e s c r i b e d b y G r a n g e r a n d L e h n i n g e r (1982) e n a b l e d us to q u a n t i f y the capacities of the 3 c o m p l e x e s that deliver electrons to the r e s p i r a t o r y chain, i.e., the N A D H , succinate a n d et-glycerophosphate o x i d a s e complexes, b y m e a s u r i n g 02 c o n s u m p t i o n after successive add i t i o n of s u b s t r a t e s a n d inhibitors. Fig. 3 shows that cellular e x p o s u r e to h y p e r o x i a resulted in a significant decrease in o x y g e n c o n s u m p t i o n of intact a n d d i g i t o n i n - t r e a t e d cells, the effect b e i n g m o s t severe after 2 d a y s of h y p e r o x i c exposure. These c h a n g e s c o i n c i d e d with s i m u l t a n e o u s p r o gressive r e d u c t i o n s in the o x y g e n c o n s u m p t i o n rates due to the N A D H a n d s u c c i n a t e o x i d a s e complexes, which were again m o s t severe after 2 d a y s of h y p e r o x i c exposure. O n the o t h e r h a n d , the activity of the c t - g l y c e r o p h o s p h a t e o x i d a s e c o m p l e x was u n a f f e c t e d b y h y p e r o x i a , or even slightly a c t i v a t e d after 1 d a y of exposure. Strikingly, D N P clearly s t i m u l a t e d a - g l y c e r o p h o s -
177
p h a t e - d r i v e n oxygen c o n s u m p t i o n in h y p e r o x i a e x p o s e d cells, in c o n t r a s t to the oxygen c o n s u m p tion e x h i b i t e d b y intact h y p e r o x i a - e x p o s e d cells (cf. Fig. 1). This result indicates (i) that the a g l y c e r o p h o s p h a t e o x i d a s e c o m p l e x along with a m a j o r p a r t of the r e s p i r a t o r y chain, i.e., u b i q u i n o n e c y t o c h r o m e c reductase, c y t o c h r o m e c oxidase a n d A T P synthetase, is c o m p l e t e l y resistant to the c o n s e q u e n c e s of h y p e r o x i c exposure, a n d (ii) that the d i h y d r o x y a c e t o n e p h o s p h a t e / t x - g l y c e r o p h o s p h a t e shuttle p a t h w a y as a source of reducing equivalents seems to b e h a r d l y or not at all utilized in H e L a cells. T h e l a t t e r s t a t e m e n t is b a s e d n o t o n l y o n the o b s e r v a t i o n that H e L a cell r e s p i r a t i o n is i m p a i r e d u n d e r high oxygen even though exogen o u s ~ t - g l y c e r o p h o s p h a t e c a n a p p a r e n t l y be o x i d i z e d after p e r m e a b i l i z a t i o n with digitonin; in a d d i t i o n , we f o u n d that after b l o c k i n g the respiration of H e L a cells with r o t e n o n e plus m a l o n a t e r e s p i r a t i o n c o u l d n o t be r e s t o r e d b y a d d i n g excess glucose (W.G.E.J. Schoonen, u n p u b l i s h e d results).
15-
e-
D..
• •
.
NADH DH
SUCCINATE DH
(~-GP DH
•O •O •O
(~-KG DH
Fig. 4. Activities of NADH, succinate, a-glycerophosphate and a-ketoglutarate dehydrogenases in homogenates from HeLa cells exposed to normoxia or hyperoxia, for 1 or 2 days. Results are means+SEM; n =8. For bars and symbols, see legend to Fig. 2. DH, dehydrogenase; a-GP, t~-glycerophosphate; a-KG, a-ketoglutarate.
Spectrophotometric analysis I n p a r a l l e l e x p e r i m e n t s , the effect of h y p e r o x i a o n key enzymes of the r e s p i r a t o r y chain a n d tric a r b o x y l i c acid cycle was e v a l u a t e d b y assaying cell h o m o g e n a t e s using s t a n d a r d s p e c t r o p h o t o m e t ric p r o c e d u r e s . A significant, d o s e - i n d e p e n d e n t , i n a c t i v a t i o n of N A D H d e h y d r o g e n a s e a n d succ i n a t e d e h y d r o g e n a s e was o b s e r v e d in h o m o g -
enates f r o m cells e x p o s e d to h y p e r o x i a (Fig. 4), whereas the activity of m i t o c h o n d r i a l a - g l y c e r o p h o s p h a t e d e h y d r o g e n a s e was t o t a l l y unaffected. O n the o t h e r h a n d , h y p e r o x i a d r a m a t i c a l l y red u c e d the activity of a - k e t o g l u t a r a t e d e h y d r o genase in a t i m e - d e p e n d e n t m a n n e r , to a level that was b e l o w the limit of d e t e c t i o n after 2 d a y s o f h y p e r o x i c exposure. This i n a c t i v a t i o n led us to also e x a m i n e s u b u n i t E3 o f this e n z y m e c o m p l e x ,
TABLE 1 KEY ENZYMES OF GLYCOLYSIS AND SEVERAL OTHER METABOLIC PATHWAYS a Enzymes Lipoamide dehydrogenase NADH diaphorase Hexokinase Phosphofructokinase Glyceraldehyde-3-phosphate DH a-GlycerophosphateDH(cyt) Pyruvatekinase Lactate dehydrogenase Glucose-6-phosphate DH Glutathione reductase
Time of hyperoxia (days) 0
1
2
43.3 + 1.1 1407 + 41 22.8 4- 1.2 157.0 4- 4.0 819 4- 26 2038 4- 81 2669 4-120 2175 4- 85 539 ± 25 71.8 4- 1.0
38.9 4- 1.3 * 1425 4- 33 22.9 4- 0.8 159.6 4- 7.5 876 4- 33 2098 4- 77 2664 4- 71 2310 +107 556 4- 27 69.3 4- 2.7
31.2 + 1.2" * *'°°° 1439 + 71 23.6 4- 1.1 170.2 4- 8.5 831 + 42 2102 4- 98 2900 4- 84 2186 +113 541 + 21 62.3 4- 2.7" *
a Cells were cultured at normoxia or hyperoxia (1 or 2 days) and analyzed at 3 days of culturing time (cf. Fig. 1). Enzyme activities are expressed as n m o l e s u b s t r a t e / m i n / m g protein (mean + S E M ; n = 8); D H , d e h y d r o g e n a s e ; cyt, cytoplasmic. *'°Significantly different from normoxic cells and 1-day hyperoxia-exposed cells, respectively. Single, double and triple symbols indicate probabifities of P < 0.05, P < 0.01 and P < 0.001, respectively (Student's two-tailed t-test).
178
which exhibits 2 separate activities, a lipoamide dehydrogenase and a N A D H diaphorase activity. As shown in Table 1, hyperoxia did not affect N A D H diaphorase activity, but reduced the activity of the lipoamide dehydrogenase moiety; this effect may at least partly account for the observed inactivation of the entire a-ketoglutarate dehydrogenase complex. The observed suppression of oxidative metabolism apparently led to an enhanced ATP generation from glycolysis, as indicated by substantially increased glucose consumption and lactate excretion (see below). Analysis of the activities of hexokinase, phosphofructokinase, glyceraldehyde3-phosphate dehydrogenase, pyruvate kinase and lactate dehydrogenase did not reveal any significant changes in enzyme levels (Table 1), indicating that the enhanced glycolytic activity was entirely due to activation of allosteric enzymes. In view of the suspected importance of cellular NADPH and glutathione in oxidative stress we also measured the NADPH- and glutathione-recycling enzymes, glucose-6-phosphate dehydrogenase, a key metabolic enzyme of the pentose phosphate cycle, and glutathione reductase. Hyperoxia did not affect glucose-6-phosphate dehydrogenase activity and had only a marginal effect on glutathione reductase after 2 days of hyperoxia. Glucose, lactate, amino acids and ammonia
The relative contributions of glycolysis and oxidative metabolism to cellular energy supply are grossly reflected in the cellular utilization of sub-
100-
80' m
60
40 E c
20 0
Fig. 5. ATP levels of HeLa cells exposed to normoxia or hyperoxia, for ] or 2 days. Results are means_+SEM: n = 8 . For bars and symbols, see legend to Fig. 2.
strates and the excretion of products. In HeLa cells exogenous glucose is the main carbon source for glycolysis, with lactate and alanine as the main end products, while glutamine and glutamate are the most important carbon fuels for oxidative metabolism, with aspartate as one of the end products (Donnelly and Scheffler, 1976; Reitzer et al., 1979; Stanisz et al., 1983; Brand et al., 1989). Exposure to hyperoxia for 2 days resulted in a substantial increase in glucose consumption and lactate production, while exposure for 1 day had only a moderate effect (Table 2). Glutamine consumption and alanine production were unaffected by hyperoxic exposure, while a clear rise in
TABLE 2 EFFECTS OF H Y P E R O X I A ON C E L L U L A R C O N S U M P T I O N / E X C R E T I O N OF G LU C O S E, LACTATE, SEVERAL A M I N O ACI DS A N D A M M O N I A a Compound
Glucose Lactate Glutamine Glutamate Alanine Aspartate Ammonia
Time of hyperoxia (days) 0
1
2
- 6 680 ± 410 11480 _+355 -1132_+ 28 469_+ 13 212_+ 8 -54_+ 4 423-+ 6
- 7 350 + 500 13 410 _+ 65 * * - 1 1 9 3 + 30 515_+ 8** 250_+ 20 -55+ 3 449-+ 23
- 12111 _+ 1 760 *'° 17 577 + 1 045 * * ,o -1109+ 19 776_+ 16 ***'°~x~ 259+ 16 -77-+ 8 *`o 336-+ 37"°
Cells were cultured under normoxic conditions or exposed to hyperoxia for 1 or 2 days and analyzed at 3 days of culturing time (cf. Fig. 1). Concentration changes are in n m o l e / h / 1 0 7 cells ( m e a n s + S E M ; n = 4); positive sign is production; negative sign is consumption. For symbols, see Table 1.
179 glutamate production and aspartate consumption was observed after 2 days of hyperoxia. Simultaneously with these changes, the production of ammonia dropped, indicating that the influx of glutamine/glutamate into the citric acid cycle was reduced. A T P levels
To assess the net effect of impaired oxidative energy metabolism and stimulated glycolysis on the energy status of the cells, cellular ATP pools were measured as a function of hyperoxic exposure time. As shown in Fig. 5, exposure for 1 and 2 days resulted in 33% and 82% depletion of ATP, respectively. Discussion
Our results show that HeLa cells exposed to normobaric hyperoxia exhibit a decreased proliferation rate and a reduced plating efficiency, which appeared to be correlated with a progressive decline in respiratory activity. Further analysis revealed that respiratory failure was associated with selective inactivation of at least 3 mitochondrial SH-group-containing flavoprotein complexes, i.e., NADH dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase. Inactivation of the latter enzyme clearly dominated the fall in respiratory activity of intact cells, by causing a lack of NADH-reducing equivalents and consequently an apparent loss of responsiveness to the uncoupler DNP. However, in cells permeabilized with digitonin, a normal respiration rate could be induced in oxygen-poisoned cells by adding a-glycerophosphate, in which case also a clear responsiveness to DNP was restored (Fig. 3). This showed that hyperoxia did not adversely affect the electron transport chain between aglycerophosphate dehydrogenase, ubiquinone cytochrome c reductase (complex III), cytochrome c oxidase (complex IV) and ATP synthetase (complex V). The observation that inactivation of NADH dehydrogenase, succinate dehydrogenase and particularly a-ketoglutarate dehydrogenase is correlated with the loss of cell viability under hyperoxic conditions suggests that these enzymes are critical targets in oxygen poisoning.
The strong dependence of oxidative metabolism in cell cultures on glutamine a n d / o r glutamate consumption (Donnelly and Scheffler, 1976; Zielke et al., 1978; Stanisz et al., 1983; Reitzer et al., 1979), which is controlled by the influx of glutamate into the tricarboxylic acid cycle via the apparently oxygen-sensitive a-ketoglutarate dehydrogenase complex, is in line with the observed net reduction of glutamine/glutamate uptake during hyperoxic exposure. To maintain a normal energy supply an activation of glycolysis was to be expected. Such a response was indeed observed, as evidenced by an enhanced glucose consumption and lactate production. Although increased glycolysis during hyperoxia has been reported previously by Rueckert and Mueller (1960), Allen and Rasmussen (1974) and Balin et al., (1976), a clear link with an inhibition of cellular respiration was not demonstrated. Moreover, the present study shows that the hyperoxia-induced metabolic shift from oxidative metabolism towards glycolysis was not due to increased synthesis of key regulatory glycolytic enzymes, but rather to the activation of an allosteric enzyme, such as phosphofructokinase. This enzyme is known to be inhibited by ATP, the level of which clearly dropped after 1 day of hyperoxic exposure (Fig. 5). Remarkably, similar effects of normobaric hyperoxia on oxidative and glycolytic energy metabolism were observed in Chinese hamster ovary cells (Schoonen et al., 1990), suggesting that the present observations may be universally valid. The 3 enzyme complexes shown here to be inactivated by hyperoxia, i.e., NADH dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase, are localized at or near the mitochondrial inner membrane. They are equipped with essential sulfhydryl groups and, at least in case of the former 2, with iron-sulfur centers, elements that would predispose these enzymes to being vulnerable for endogenous reactive species, i.e., oxygen radicals or lipid peroxidation products (Balentine, 1982; Haugaard, 1968). Since reactive oxygen species are excessively generated by N A D H dehydrogenase, succinate dehydrogenase and ubiquinone cytochrome c reductase (Forman and Kennedy, 1974; Forman and Boveris, 1982; Turrens et al., 1985), especially under hyperoxic conditions (Jamieson, 1989; Zweier et al.,
180
1989), such a mechanism of inactivation appears to be quite plausible. In the case of lipid peroxidation, it is relevant to note that mitochondrial D N A replication occurs in association with the inner mitochondrial membrane (Nass et al., 1965; Shearman and Kalf, 1975, 1977; Albring et al., 1977). This close association of mitochondrial DNA with a membrane that is supposed to be an important cellular site of free radical production (Chance et al., 1979) raises the possibility that mitochondrial DNA, just like N A D H dehydrogenase, succinate dehydrogenase and a-ketoglutarate dehydrogenase, is subject to the attack by oxygen free radicals a n d / o r lipid peroxidation products, such as the cross-linking agent malondialdehyde (Fleming et al., 1982). Since mitochondrial DNA is not protected by histone proteins, while mitochondria are less efficient in repairing D N A damage and replication errors, the mitochondrial genome may be more susceptible to the damaging effects of free radicals and alkylating agents than the nuclear genome (Richter, 1988; Richter et al., 1988). Thus, hyperoxia may be a valuable model to evaluate the genotoxic impact of metabolically produced free radicals at the level of mitochondrial DNA.
Acknowledgements This research was supported by the Foundation for Fundamental Biological Research (BION), which is subsidized by the Netherlands Organization for Scientific Research (NWO).
References Albring, M., J. Griffith and G. Attardi (1977) Association of a protein structure of probable membrane derivation with HeLa cell mitochondrial DNA near its origin of replication, Proc. Natl. Acad. Sci. (U.S.A.), 74, 1348-1352. Allen, J.E., and H. Rasmussen (1971) The effects of oxygen on cellular metabolism, Int. J. Clin. Pharmacol., 5, 26-33. Balentine, J.D. (1982) Pathology of Oxygen Toxicity, Academic Press, New York, NY. Balin, A.K., D.B.P. Goodman, H. Rasmussen and V.J. Cristofalo (1976) The effect of oxygen tension on the growth and metabolism of WI-38 cells, J. Cell. Physiol., 89, 235250. Balin, A.K., D.B.P. Goodman, H. Rasmussen and V.J. Cristofalo (1978) Oxygen-sensitive stages of the cell cycle of human diploid cells, J. Cell Biol., 78, 390-400.
Balin, A.K., A.J. Fisher and D.M. Carter (1984a) Oxygen modulates growth of human cells at physiologic partial pressures, J. Exp. Med., 160, 152 166. Balin, A.K., A.F. Fischer, I. Leong and D.M. Carter (1984b) Physiologic oxygen tensions modulate the lifespan of human cells, Age, 7, 142. Bradford, M.M. (1976) A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72. 248-254. Brand, K., W. Fekl, J. Von Hintzenstern, K. Langer, P. Luppa and C. Shoerner (1989) Metabolism of glutamine in lymphocytes, Metabolism, 38, 29 33. Chance, B., H. Sies and A. Boveris (1979) Hydroperoxide metabolism in mammalian organs, Physiol. Rev., 59, 527605. Donnelly, M., and I.E. Scheffler (1976) Energy metabolism in respiration-deficient and wild type Chinese hamster fibroblasts in culture, J. Cell. Physiol., 89, 39-52. Fleming, J.E.. J. Miquel, S.F. Cottrell, L.S. Yengoyan and A.C. Economos (1982) Is cell aging caused by respiration-dependent injury to the mitochondrial genome?, Gerontology, 28, 44-53. Forman, H.J., and A. Boveris (1982) Superoxide radical and hydrogen peroxide in mitochondria, in: W.A. Pryor (Ed.), Free Radicals in Biology, Vol. 5, Academic Press, New York, NY, pp. 65-90. Forman, H.J., and J.A. Kennedy (1974) Role of superoxide radical in mitochondrial dehydrogenase reactions, Biochem. Biophys. Res. Commun., 60, 1044-1050. Fridovich, I. (1989) Superoxide dismutases. An adaptation to a paramagnetic gas, J. Biol. Chem., 264, 7761-7764. Gille, J.J.P., C.G.M. van Berkel, E. Mullaart, J. Vijg and H. Joenje (1989) Effects of lethal exposure to hyperoxia and to hydrogen peroxide on NAD(H) and ATP pools in Chinese hamster ovary cells, Mutation Res., 214, 89-96. Granger, D.L., and A.L. Lehninger (1982) Sites of inhibition of mitochondrial electron transport in macrophage-injured neoplastic cells, J. Cell Biol., 95, 527-535. Harman, D. (1981) The aging process, Proc. Natl. Acad. Sci. (U.S.A.), 78, 7124 7128. Haugaard, N. (1968) Cellular mechanisms of oxygen toxicity, Physiol. Rev., 48, 311-373. Jamieson, D. (1989) Oxygen toxicity and reactive oxygen metabolites in mammals, Free Radical Biol. Med., 7, 87108. Joenje, H. (1989) Genetic toxicology of oxygen, Mutation Res., 219, 193-208. Joenje, H., J.J.P. Gille, A.B. Oostra and P. van der Valk (1985) Some characteristics of hyperoxia-adapted HeLa cells. A tissue culture model for cellular oxygen tolerance, Lab. Invest., 52, 420-428. Mehlhorn, R.J., and G. Cole (1985) The free radical theory of aging: a critical review, Adv. Free Radical Biol. Med., 1, 165-223. Nass, M.M.K., S. Nass and B. Afzelius (1965) The general occurrence of mitochondrial DNA, Exp. Cell Res., 37, 516-539.
181 Poot, M., D. Schindler, M. Kubbies, H. Hoehn and P.S. Rabinovitch (1988) Bromodeoxyuridine amphfies the inhibitory effect of oxygen on cell proliferation, Cytometry, 9, 332-338. Reitzer, L.J., B.M. Wice and D. Kennell (1979) Evidence that glutamine, not sugar, is the major energy source for cultured HeLa cells, J. Biol. Chem., 254, 2669-2676. Richter, C. (1988) Do mitochondrial DNA fragments promote cancer and aging?, FEBS Lett., 241, 1-5. Richter, C., J.W. Park and B.N. Ames (1988) Normal oxidative damage to mitochondrial and nuclear DNA is extensive, Proc. Natl. Acad. Sci. (U.S.A.), 85, 6465-6467. Rueckert, R.R., and G.C. Mueller (1960) Effect of oxygen tension on HeLa cell growth, Cancer Res., 20, 944-949. Schoonen, W.G.E.J., A.H. Wanamarta, J.M. van der Klei-van Moorsel, C. Jakobs and H. Joenje (1990) Respiratory failure and stimulation of glycolysis in Chinese hamster ovary cells exposed to normobaric hyperoxia, J. Biol. Chem., 265, 11118-11124. Shearman, C.W., and G.F. Kalf (1975) DNA synthesis by a membrane-DNA complex from rat liver mitochondria, Biochem. Biophys. Res. Commun., 63, 712-721. Shearman, C.W., and G.F. Kalf (1977) DNA replication by a
membrane DNA complex from rat liver mitochondria, Arch. Biochem. Biophys., 182, 573-586. Stanisz, J., B.M. Wice and D.E. Kennell (1983) Comparative energy metabolism in cultured heart muscle and HeLa cells, J. Cell. Physiol., 115, 320-330. Thaw, H.H., V.P. Collins and U.T. Brunk (1984) Influence of oxygen tension, pro-oxidants and antioxidants on the formation of lipid peroxidation products (lipofuscin) in individual cultivated human glial cells, Mech. Ageing Dev., 24, 211-223. Turrens, J.F., A. Alexander and A.L. Lehninger (1985) Ubisemiquinone is the electron donor for superoxide formation by complex III of heart mitochondria, Arch. Biochem. Biophys., 237, 408-414. Zielke, H.R., P.T. Ozand, J.T. Tildon, D.A. Sevdalian and M. Cornblath (1978) Reciprocal regulation of glucose and glutamine utilization by cultured human diploid fibroblasts, J. Cell. Physiol., 95, 41-48. Zweier, J.L., S.S. Duke, P. Kuppusamy, J.T. Sylvester and E.W. Gabrielson (1989) Electron paramagnetic resonance evidence that cellular oxygen toxicity is caused by the generation of superoxide and hydroxyl free radicals, FEBS Lett., 252, 12-16.
