Mutation Research, 275 (1992) 267-279 1992 Elsevier Science Publishers B.V. All rights reserved 0921-8734/92/$05.

1)0

267

MUTAGI 0248

Role of oxidative stress in Drosophila aging

J.E. F l e m i n g , I. R e v e i l l a u d a n d A. N i e d z w i e c k i
Linus Pauling Institute of Science and Medicine, Palo Alto, CA 94306, USA (Received 1 March 1992) (Revision received 18 May 1992) (Accepted 19 May 1992)

Keywords: Drosophila; Ageing; Oxidative stress; Heat shock; Superoxide dismutase; Oxygen radicals

Summary
We review the role that oxidative damage plays in regulating the lifespan of the fruit fly, Drosophila melanogaster. Results from our laboratory show that the lifespan of Drosophila is inversely correlated to its metabolic rate. The consumption of oxygen by adult insects is related to the rate of damage induced by oxygen radicals, which are purported to be generated as by-products of respiration. Moreover, products of activated oxygen species such as hydrogen peroxide and lipofuscin are higher in animals kept under conditions of increased metabolic rate. In order to understand the in rive relationship between oxidative damage and the production of the superoxide radical, we generated transgenic strains of Drosophila melanogaster that synthesize excess levels of enzymatically active superoxide dismutase. This was accomplished by P-element transformation of Drosophila melanogaster with the bovine cDNA for CuZn superoxide dismutase, an enzyme that catalyzes the dismutation of the superoxide radical to hydrogen peroxide and water. Adult flies that express the bovine SOD in addition to native Drosophila SOD are more resistant to oxidative stresses and have a slight but significant increase in their mean lifespan. Thus, resistance to oxidative stress and lifespan of Drosophila can be manipulated by molecular genetic intervention, in addition, we have examined the ability of adult flies to respond to various environmental stresses during senescence. Resistance to oxidative stress, such as that induced by heat shock, is drastically reduced in senescent flies. This loss of resistance is correlated with the increase in protein damage generated in old flies by thermal stress and by the insufficient protection from cellular defense systems which includes the heat shock proteins as well as the oxygen radical scavenging enzymes. Collectively, results from our laboratory demonstrate that oxidative damage plays a role in governing the lifespan of Drosophila during normal metabolism and under conditions of environmental stress.

The hypothesis that free radicals, or more specifically oxygen radicals, give rise to the biological damage that causes the physiological de-

Corrcspoadence: James E. Fleming, Department of Biology, Eastern Washington University, Cheney, WA 99004, USA.

cline characteristic of oldvr organisms is only now beginning to be adequately appreciated. Although modern molecular biological approaches to this question are now being rigorously applied, the theory has its beginnings around the turn of the century. Historically, the concept can be traced back at least as far as 1908 when Rubner published a

268 paper suggesting that th~ total amount of energy consumed by different mammalian species remained relatively constant throughout life (Rub~ her, 1908; see also Sohal, 1976). Attempts to relate similar factors to lifespan in Drosophila were initiated by Loeb and Northrup (1917). They noted that, since these animals are poikilotherms, ambient temperature affects their physiological activity and especially their !ifespan. It should be pointed out that the use of Drosophila was recognized very early in the development of gerontological research. It is now clear that this organism provides the scientific community with an excellent model for such studies since (1) it has a relatively short lifespan; (2) large colonies can be maintained under controlled laboratory conditions; (3) the species is available in highly inbred and outbred strains (with differing lifespans); (4) the adult appears to show many of the manifestations of cellular senescence observed in mammals; and (5) DNA-mediated germline transformation is possible (Miquel et al., 1981; Baker et ai., 1985; Reveillaud et al., 1991). Following the early studies of Rubner (1908) and Loeb and Northrup (1917), Raymond Pearl (1928), on the basis of his own data, published the rate of living theory of aging. In the case of Drosophila, this theory postulates that adult flies have a shorter duration of life at high temperature because of a higher rate of living; that is, it is more metabolically active. Since then, numerous observations have been reported regarding the relationship between temperature and more importantly, metabolic rate and lifespan in Drosophila (Sohal, 1976; Miquel et al., 1976, 1983; Fleming et ai., 1981; Sohal, 1987). In early work from our laboratory, we showed that 0 2 consumption, a measure of metabolic rate, is inversely correlated to Drosophila lifespan (Miquel et al., 1976; Fleming et al., 1981, 1988). In analogous studies, Sohal and coworkers provided convincing support for the rate of living theory in the housefly, Musca domestica (Sohal and Bucham, 1981). The biochemical implications for the relationship between lifespan and metabolic rate (02 consumption) were not fully appreciated until it was shown that oxygen toxicity is related to the metabolic generation of the superoxide radical (O~-) (Fridovich, 1975; McCord and Fridovich, 196~). Thus, the finding that aerobic cells produce oxygen radicals as a by-product of respiration provided a conceptual link between the rate of living theory and the generation of reactive oxygen species. In fact, as far back as 1954, Rebecca Gerschman (1962, 1981) had proposed that oxygen toxicity might be caused by the generation of free radicals, and she even suggested that a continuous small 'slipping' in the antioxidant defenses of organisms could be a factor contributing to aging and death (see Miquel et al., 1980). Although the conclusion was inherent in her argument, apparently she did not actually make the specific connection between free radical damage and aging. This connection was anticipated by Denham Harman in 1956 when he published the farsighted free radical theory of aging (Harman, 1956, 1982). Basically the theory stated that 'aging results from the deleterious effect of free radicals produced in the course of cellular metabolism'. Conceptually, the free radical t~heo~ as first spelled out by Harman has changed little in the last 35 years. Presently, many groups are using molecular biological techniques to search out the specific genes that may be involved in the generation or scavenging of free radicals; however, the relationship between lifespan and metabolic rate still provides strong evidence in support of the free radical theory of aging in Drosophila. increased metabolic rate results in oxidative stress and, if not adequately scavenged, accelerates certain age-related changes including a reduction in lifespan (Phillips et al., 1989). Moreover, the products of activated oxygen species such as hydrogen peroxide (H202), oxidized glutathione, compounds that react with thiobarbituric acid and lipofuscin are higher in animals kept under conditions of increased metabolic rates. The discovery by McCord and Fridovich of superoxide dismutase (SOD), an enzyme that catalyzes the dismutation of the superoxide radical, produced strong indirect support for the concept that oxygen radicals must play a deleterious role in biological systems (McCord and Fridovich, 1969). The awareness that aerobic cells possess a wide variety of mechanisms, both enzymatic and nonenxymatic, that protect against oxygen toxicity provided further evidence for the involvement of oxygen radicals in pathologic processes, including

269

aging (Fridovich, 1988). It is now fairly well established that the superoxide radical is produced in vivo by a number of biological reactions that include enzymatic, spontaneous and photochemical oxidations (Fridovich, 1988). The specific role of 02 free radicals in the aging process is underscored by the observation that at least 1% of the 0 2 consumed by a respiring cell is incompletely reduced in the mitochondrial electron transport chain (Boveris, 1976). Over time, this small 'leakage' of O 2 radicals can result in irreversible cellular damage (Miquel et al., 1980).

Free radical theory of aging

As a first approximation, the free radical thepry of aging predicts that the lifespan of organisms with similar metabolic rates should be correlated to their level of antioxidant defenses. Of course, one aspect of the free radical theory that is not taken into account in this perspective is that the rate of free radical production may be as important in governing the rate of damage during aging as the levels of defense. Variations in the efficiency of oxidative phosphorylation or microsomal drug metabolism, for example, may contribute to the different levels of free radical generation for a given metabolic rate. Despite the potential importance of free radical synthesis rates, we will focus our discussion on the role of oxygen radical scavenging in the aging process. A modest amount of research has been carried out on the relationship between longevity and antioxidant defenses. The first studies to report on the comparison of SOD to lifespan in Drosophila were carried out by Bartosz et al. (1979). These authors reported that a short lived mutant of Drosophila was deficient in total SOD activity. However, Massie and coworkers (1975) have noted that SOD activity is not significantly different between wild strains of Drosophila whose lifespans may differ by as much as 40%. We have also noted that the Samarkand strain of Drosophila, a long lived stock, has nearly twice as much catalase as the short lived Swedish C strain, but no difference in SOD activity (unpublished data). In addition, the Samarkand strain is more efficient at scavenging hydroxyl radicals generated in vitro than the Swedish C strain (Fleming

et al., 1988). Ruddle et al. (1988) reported that propyi gallate, a phenolic antioxidant, significantly increases both the mean and maximum lifespan of Drosophila when they are fed for their entire lifespan with a concentration of 0.3% as a dietary supplement. Phillips et al. (1989) have reported that CuZn SOD null mutants in Drosophila are hypersensitive to paraquat and have drastically reduced longevity (see also Peng et al., 1986). More recently, Arking et al. (1991) have shown that paraquat resistance is associated with extended longevity in a genetically selected long lived strain of Drosophila. Undoubtedly, SOD plays some role in regulating the lifespan of Drosophila and, in concert with catalase, controls the level of toxic 0 2 species in the cell which contributes to some aspects of cellular senescence. It should be pointed out that glutathione peroxidase may also be of considerable importance as a free radical scavenger in this scheme since it is apparently responsible for reducing the level of organic peroxides. However, the existence of this enzyme in Drosophila has not been adequately demonstrated.

SOD transgenics have increased lifespans

Until recently, a direct test of this concept, i.e., to upregulate the intraceilular levels of these enzymes and then determine the consequence on aging, was not possible. Such experiments can now be accomplished due to the development of technologies that permit the introduction of new or additional genes into cells or even whole animals. With the discovery of the P-transposable element in Drosophila melanogaster by Rubin (1986) and Spradling (1986), the method of germline gene transfer has become relatively routine. We have exploited this technology in order to examine the role of specific antioxidant enzymes on the aging process in Drosophila. Although catalase, glutathione peroxidase and vitamins C and E probably play major roles in the intracellular free radical detoxification, we initially focused on CuZn SOD (Reveillaud et al., 1991). Thus, to understand the in vivo relationship between an efficient dismutation of 0 2 and senescence, we generated transgenic strains of Drosophila melanogaster overproducing CuZn

270

100

L _. J~"~l]~_
j ~ ~

80
>o 60
40

-0- Control males -e- Transgenic males --

100 ~ 1 80. i

~,~

-e- w67c23 + ct25 -e- P[bSOD]5

*

"
4o

2O
O I I I

2O 0

10

20 30 Paraquat, [raM]

40
Fig. 2. Survival curves and recipient flies that atmosphere for 20 min in Reveillaud et al.

Age, Days
of populations of transgenic, control have been exposed to a 100% oxygen each day. These strains are described (1991): w67c23, recipient strain of Drosophila melanogaster used for microinjection; Ct25, control strain as described in Fig. 2; P[bSOD]5 and P[bSOD]I 1 are transgenic strains that express the bovine superoxide dismutase in addition to the native Drosophila SOD.

Fig. 1. Dose dependent survival curves of adult flies transgenie for the bovine superoxide dismutase gene and control flies in the presence of paraquat. Transgenic flies expressed SOD at an average increase of 32.5% over control and recipient flies. Control flies are strains that were injected with the recombinant plasmids not containing the eDNA for the bovine superoxide dismutase. Recipient strains are those that were used for the microinjection. The paraquat studies were carried out by placing groups of 30 flies in 35-ml glass vials that held a piece of filter paper wetted with a 15% sucrose solution containing the indicated concentrations of paraquat. A stock paraquat solution was made fresh every day, and the filter paper was rewetted with the new solutkm. Each point is the mean value standard error for more than 300 flies. Reprinted from Reveillaud et el. (1991)with permission.

SOD. Thi~: was achieved by microinjecting Drosophila embryos with P-elements containing the bovine CuZn SOD eDNA under the control of the Drosophila actin 5c gene promoter (Fig. 1) (Reveillaud et el., 1991). The insertion of the bovine CuZn SOD eDNA into the genome of Drosophila resulted in a mean increase of 32.5%
TABLE !

in the level of total CuZn SOD activity in transformed adults compared with that of control flies (Reveillaud et al., 1991). Twenty different strains of Drosophila expressing the bovine form of CuZn SOD were generated and several of these stocks were examined for their resistance to oxidative stress and lifespan. As a test of resistance to superoxide mediated toxicity, flies were given various concentrations of paraquat, a compound that has been shown to generate O~" when metabolized in vivo (Hassan and Fridovieh, 1978). Fig. 1 shows the dose dependent survival curves of

LIFESPAN DATA FOR SOD TRANSGENIC AND CONTROL POPULATIONS Strain Number examined 173 178 169 158 168 131 162 Quantiles" (days) 25% 50% 39 4! 53 39 51 37 48 46 47 53 53 52.5 47 52.5 Mean time to death (days) SEM 45.55 0.97 44.46 0.87 53.31 0.81 49.89 0.80 52.22 0.78 45.580.93 52.44 0.71
P vs.

75% 53 5i 60 53 58 51 58

control

Ctr4 (control) W67C23 PbSOD 3 PbSOD 5 PbSOD 11 PbSOD 12 PbSOD 18

0.52 < 0.0001 0.0057 < 0.0001 0.72 < 0.0001

a Time until 25, 50, and 75% of flies were dead,

271

100~ ~ 1

~ "=" Ctr 4

iooI
,o t
0

>e8o1

-,9 W67C23

\
,

-. C,r4

80

~o
,o
0

20

40

60

80

100

20

40

60

80

100

,']
2O

.oof_
0
0 20

h._

4" Ctr 4

"~'s'~l

100

80 ,o

"=" 4 1 -e- Ctr P[bSOD]I

-t
0

,0]
I I | I

40

60

80

100

20

40

60

80

100

100
80

Ctr 4 P[bSOD]12

-n- Ctr 4
80 60

i
100

40
2O

20 0 20 40 60 Age, days 80 100 0 0 20 40 60 Age, days

80

Fig. 3. Lifespan curves of adult transgenic, control and recipient strains raised and maintained at 25oc. Flies were collected at eclosion and the number of dead was counted every 3 days. A summary of the statistical evaluation of these data is shown in Table 1. Reprinted from Reveillaud et al. (1991)with permission.

2~ bovine SOD transgenics and control flies fed the indicated concentrations of p raqaat. Transgenic males are significantly more resistant to paraquat induced killing than either the controls or the recipient lines. In this case, control flies have been injected with unrecombined P-elements that do not contain the bovine SOD eDNA and thus have normal levels of SOD. Theoretically, this is not unexpected; however, the specific role of SOD in protecting against paraquat toxicity remains unresolved, especially in studies involving bacteria or cultured cells (Carlioz et al., 1986; Scott et al., 1987; Krail et al., 1988). The lifespan data for transgenic lines that have been measured are shown in Fig. 3 and Table 1. There was a slight but statistically significant increase in the mean lifespan of these strains. However, with the exception of one strain (data unpublished), we did not find that the maximum lifespan of our transgenics was increased. We have examined two of our strains in more detail to determine their resistance'to specific oxidative and nonoxidative stresses. When flies are exposed to 100% oxygen atmospheres, a drastic reduction in lifespan is observed; however, this reduction is significantly less for transgenic Drosophila (Fig. 2). The mean lifespan for control strains is 3.62 days whereas the lifespan of transgenic strain is 4.35 days. It is interesting that transgenic flies are more sensitive to H202 than control flies. The median lifespan of transgenic flies fed 1% H202 is 64 h and for controls it is 72 h. These findings are consistent with results on transformed cell culture lines overproducing human CuZn SOD and with data on Down's syndrome patients, in whom a 50% increase in cytoplasmic SOD has been associated with free radical induced pathologic changes (Sinet et al., 1979; Elroy-Stein et al., 1986; Krall et al., 1988). Increased levels of SOD in the absence of a compensatory upregulation of catalase is thought to result in an accumulation of H202 which may give rise to cytotoxic effects such as that seen in Down's syndrome. This is supported by our observation that catalase activity is the same for transgenics as it is for control flies. Using starvation resistance as a non-oxidative stress, we found that both transgenie and control flies maintained on water only had identical survival curves of 72 h. Also, throughout the lifespan, no significant difference in physical activity was displayed for transgenic versus control flies. Thus, lifespan and resistance to oxidative stress (induced by paraquat or hyperoxia), but not to nonoxidative stress, can be enhanced by increasing the endogenous synthesis of SOD. Seto et al. (1990) reported that the introduction of an extra gene for Drosophila Cu2,n SOD does not result in an increased lifespan. They argued that SOD transgenics have the same lifespan as control flies. It is surprising that these authors did not provide a statistical analysis of their mortality data since these data formed the basis of their conclusion. Their graphs show that the median lifespan for transgenics is ~ 43 days and for controls is ~ 40 days. The maximum lifespan for their controls is 50 days, their transgenic strain SOD + 4 is 51 days and their SOD + 5 is 56 days. Although these are clearly not dramatic increases in lifespan, they may be significant which emphasizes the need for statistical treatment of the data. Also, their results are interesting when compared to our data since we reported similar values for the lifespan of our transgenic lines; however, we found that the increases in lifespan observed in many of our transgenie lines were statistically significant (Table 1). As previously pointed out, flies that are null for the SOD gene have drastically reduced lifespans and are extremely sensitive to oxidative stresses (Phillips et al., 1989). We recently introduced the bovine gene for SOD into these null mutants by genetic crosses with our transgenic flies in an attempt to rescue the reduced lifespan and sensitivity to oxidative stress. The rescued strain displayed increased resistance to paraquat and hyperoxia as compared to the null; however, lifespan was not restored to normal levels even though there was a slight increase over the nulls (unpublished data), it is not as yet known why the lifespan of these flies is unrestored to normal levels but it may be related to the fact that several genes contribute to the shortened lifespan of the mutant or to the differential expression of the bovine SOD which is driven by the actin promoter. So far, SOD appears to play a significant role in the lifespan of Drosophila. But it is becoming

273

more evident that there is a narrow window of expression which is compatible with a normal lifespan for Drosophila. Very low levels of expression are clearly detrimental despite the observation that only 50% expression results in a normal lifespan (Phillips et al., 1989). And although increases of expression up to about 35% of normal provide added protection against oxidative stress and resuR in an increased lifespan, levels of SOD above this concentration appear to be toxic. This latter observation may be related to the finding that very high levels of SOD are toxic due to the increased generation of H20 2. This is supported by our observation that SOD overexpression resulted in the death of pupae during the process of eclosion for some transgenic strains. Morphological analysis of malpighian tubules from imagoes that died during eclosion showed that these cells contained abundant lipofuscin (age pigment) (Reveillaud et al., 1991). Typically, lipofuscin, an indicator of oxidative damage, only occurs in senescent or oxygen stressed flies (Miquel et al., 1977). Thus, a change in the optimum concentration of CuZn SOD, whether through underexpression or overexpression, apparently leads to an accumulation of 0 2 or H202, the toxic compounds of oxygen metabolism. There may be threshold levels for superoxide dismutase above or below which viability is affected. These results are consistent with the data from Down's syndrome patients, in whom the 50% increase in cytoplasmic SOD has been aP,sociated with free radical induced pathologic changes (Sinet et al., 1979; Elroy-Stein et al., 1986; Krali et al., 1988). These results point out the need to generate transformed flies in which functionally coupled activities of the free radical scavenging pathway are elevated in a balanced way. Thus, we are in the process of constructing flies with increased activities of catalase and glutathione peroxidase in conjunction with the CuZn SOD. It is prudent to consider also that introduction of the gene for Mn SOD, the mitochondrial form of the enzyme, may reduce the intraceUular damage resulting from metabolically generated free radicals. We have suggested previously that the mitochondria may be the 'Achilles heel' of the aging cell, a concc.pt also anticipated by Harman (1982), Munkres (1979) and Miquel et al. (1980). We

proposed that oxygen radical damage to the mitochondrial genome may be the initial site of senescent degeneration of the cell. This was supported by the observation that ,'m'tochondria! DNA may be particularly vulnerable to oxidative damage since it is located in close proximity to the inner mitochondrial membrane, the site of free radical generation (Fleming et al., 1982; Shearman and Kalf, 1977). Mitochondrial DNA has been shown to be quite sensitive to damage by toxic chemicals (Wunderlich, 1971/1972); the mitochondrial genome appears to be more vulnerable to mutagenesis than the nuclear genome since it is not protected by histones or by competent repair mechanisms (Fleming et al., 1982; Miquel and Fleming, 1986; Fukanaga and Yielding, 1979). In support of this view, Ames has recently shown that the mitochondrial genome is damaged at a rate that is more than 10 times that of the nuclear DNA (Ames, 1990). Heat shock proteins in aging Drosophila The relationship between heat and oxidative stress and the regulation of organismal response to these stresses may be particularly important for aging research since old animals are typically less resistant to physiological and biochemical stress than their younger counterparts. According to the concept of senescence championed by Medawar (1952), aging does not inevitably lead to death but makes an organism more susceptible to potentially lethal stresses. In addition to the well known antioxidant enzymes just discussed, it is worthwhile to consider that another class of proteins, the heat shock proteins, may play a role in protecting Drosophila against oxygen radical damage, especially during periods of increased stress. The first report on the effect of heat stress on these organisms originated with Ritossa (1962, 1963, 1964), who observed the induction of puffs on the giant salivary gland chromosomes in flies exposed to increased temperature. The heat induced puffs are sites of intense RNA transcription of active individual genes. In addition, he noticed that the puffing pattern seen after temperature elevation could also be induced by treating the isolated salivary glands by 2,3-dinitrophenol, an uncoupler of ox-

274

idative phosphorylation, sodium azide, an inhibitor of cellular respiration, or by release from anoxia, all suggesting the possibility of the existence of a direct connection between heat shock and the cellular respiratory chain. Recent intensive research in this area provided evidence that the stress of both heat and oxidation led to similar cellular effects. These include: (1) both lead to accumulation of adenylated nucleotides or 'alarmones' (Bochner et al., 1984; Guedon et al., 1986); (2) antioxidant enzyme activity is induced by both types of stress (Morgan et al., 1986; Christman et al., 1985; Loven et al., 1985; Hass et al., 1988); (3) both treatments result in the induction of at least some common heat shock proteins (Courgeon et al., 1988; Hahn et al., 1991); (4) the inhibition of antioxidant defenses stimulates the production of heat shock proteins and increases susceptibility to killing by heat shock (Levinson et al., 1978; Omar et al., 1986); (5) thermotolerance can be induced in response to oxidative stress (Lee et al., 1983). In this latter case however, the possible cross-protection between heat stress and oxidative injury has not been completely established (Mitchell et al., 1983). In bacteria for instance, heat shock induces thermotolerance but not resistance to H202 (Christman et al., 1985). in Chinese hamster fibroblasts, resistance to H202 w a s observed after exposure to heat, but no resistance to heat after exposure to H202 (Spitz et al., 1987), On the other hand, the coordinate upregulation of enzymes with antioxidant potential (heme oxygenase and SOD) and of classical heat shock proteins in human monocytes during phagocytosis support the hypothesis that heat shock proteins may be induced as part of a cellular mechanism for protection from oxidative stress, The important implication of these studies is related to the question of whether such protection occurs in rive and what mechanisms are involved. Support for the induction of heat shock proteins following oxidative stress stems from the observations that reperfusion injury, inflammation, or return from anoxia induces the h e a t shock response (Morimoto et al., 1990). The mechanism involved in protection from oxygen free radicals by heat shock proteins may include, among others, polypeptide stabilization (Pelham, 1986), indue-

tion of classical scavengers such as SOD and GSH (Andreoli et al., 1986; Freeman and Meredith, 1989) or maintenance of normal cellular ui-

A.
Heat shock (min)
0 30 60 0 30 60 0 3060

23s 16s

Age (days)

28

47

B.

4O

+I
so
0

young I! middle age old

2o

10
R

30 rain

60 rain

Time of heat shock

Fig, 4, Changes in the level of hspT0 RNA in heat shocked Drosophila with age. (A) Northern blot analysis of hspT0 RNAs from control flies and flies heat shocked for 30 rain and 60 rain. RNA was separated on formaldehyde-agarose gel, immobilized on Biotrans membrane and hybridized to the 32p-labeled Drosophila hspT0 probe. The size of hybridization band was estimated to be -- 2.3 kb. (B) Percent of hsp70:actin RNA calculated from Northern blots for heat shocked flies of different ages. The membrane was reprobed with Drosophila actin 42A probe, and hybridization bands were scanned and integrated. Data from five different hybridization experiments were calculated as percent of hsp70:actin ratios expressed in young (1-4 days old), middle aged (23-28 days old), and old (47 days old and older) flies for 30 and 60 rain of heat stress. The graph bars show mean + S.E. Reprinted from Niedzwiecki et al. (1991)with permission.

275

trastructure (Thomas et al., 1982; Falkner et al., 1981). We initially examined the heat shock response in aged Drosophila by high resolution two dimensional gel electrophoresis (Fleming et al., 1988). We found that old flies apparently respond to heat shock by synthesizing significantly more heat shock proteins than young insects. Moreover, a heat shock response similar to that observed in senescent Drosophila could be induced in young organisms by feeding them canavanine, an arginine analog. This result suggested that posttranslational modifications of protein, some of them possibly by oxidative processes, may be responsible for the unusual stress response in aging flies. More recently our laboratory determined the effect of aging on the expression of one of the heat shock genes, hsp70 (Niedzwiecki et al., 1991). In these studies we found that the level of hsp70 RNA increases in flies up to middle age (28 days), but then declines for senescent flies (47 days old and older) (Fig. 4). Taking into account that expression of hsp70 has been implicated in the protection of cellular proteins against heat induced aggregation and denaturation (Pelham,

-hs
I I I

+hs
I

kD

-106 - 80

50

32 27

-18
| g | |

100"

Fig. 6. Effect of heat stress on formation of ubiquitin complexes with proteins synthesized before heat shock in aging Drosophila. Young (2-4 days old) and old (60-61 days old) flies were exposed for 3 h to '~SS-methionine, then transferred to cold methionine and heat shocked for 30 min at 37C. Control flies were kept for 30 min at 25C, Protein extracls from these flies were immunoprecipitated with polyclonal antiubiquitin antibodies, separated on 12% SDS polyacrylamide gels and exposed to autoradiography.

7$"

SO"

i
25"

16

20

s0

4o

so

66

Age (days) Fig. 5, Survival curve for aging Drosophila. Flies of different ages were exposed to 37([: for 30 min (rl) and 60 rain (e). Each data point represents the calculated mean percent of survival for 50-150 flies. Reprinted from Niedzwiecki et al. (1991) with permission.

1986) this age related increase in hspT0 in heat shocked Drosophila supports our prediction of an accumulation of altered proteins with age. However, in old flies, the expected increase in protein damage generated by heat actually results in less protection from hsp70. We have also noticed that in old Drosophila recovering from heat stress, the hspT0 message persists for a longer time than it does in young flies. This message is continually translated into protein which apparently accounts for the observation that old flies 'make' more heat shock proteins than the young ones. Again, the prolonged expression of hsp70 during recovery from heat shock was also observed in young flies fed canavanine. The variation in persistence of hsp70 after heat stress has been observed in

276

Drosophila cell culture and related to the severity of the stress (Yost et al., 1990). Consequently, changes in the steady state hsp70 level in aging flies may be related to multiple mechanisms, both the transcription and degradation of hsp70 RNA may be involved. The longer expression of hsp70 in old flies may also be related to increased protein damage and a 'slower' repair process and/or less functional hsp70 generated in senescent organisms. This latter suggestion is supported by reports showing a decreased activity of some cellular repair enzymes and an increased proportion of inactive proteins with age (Stadtman et al., 1986). The accumulation of conformationally altered proteins plays a role i~l the regulation of hsp70 expression in Drosophila, however, it is still not clear how these abnormal proteins affect thermotolerance. We observed that old Drosophila do not survive heat shock as well as young flies (Fig. 5). This would imply that the stress induced by heat in aging organisms is more severe due to the labile state of certain cellular proteins. Accordingly, we observed an elevated proteolytic rate in old Drosophila during heat stres:, (Nicdzwiecki et

al., 1990) and an increased level of proteinubiquitin conjugates formed in these flies (Fig. 6). However, the proteins affected by this process have not yet been identified. We have also investigated how heat shock affects the activity of the intracellular scavenging system. The delicate balance between the hydrogen peroxide generating (SOD) and the hydrogen peroxide detoxifying systems (catalase, glutathione peroxidase) seems to be of great importance. Recently, we examined the age dependent expression of RNA and the enzymatic activity of CuZn SOD and catalase in aging Drosophila exposed to heat stress. Expression of the CuZn SOD gene was stimulated by heat in flies of all ages and it paralleled the age dependent changes in hsp70 RNA synthesis (Fig. 7). CuZn SOD RNA was elevated in heat shocked middle aged flies, but decreased in old Drosophila. However, the changes at the transcriptional level were not accompanied by significant differences in the total CuZn SOD activity. Catalase gene expression and enzymatic activity were not affected by age in h:at stressed Drosophila. On the other hand, we observed some physico-chemical changes in both

Heat
Shock rain 0 30 60 0

hsp70
30 60 0 30 60

Cu/ZnSOD
0 30 60 0 30 60 0 30 60 0

catalase
30 60 0 30 60 0

30 60

16S 23S

young

middle age

old

young

middle ego

old

young

middle age

old

Fig. 7. Changes in the expression of hspT0, CuZn SOD and catalase in heat stressed aging flies. Flies 2 days old (young), 23-24 days old (middle age) and 49-50 days old (old) were exposed to 37C for 0 rain, 30 rain, and 60 rain, After that time RNA was isolated, separated on formaldehyde-agarose gels, and transferred to Biotrans membrane, Membranes were hybridized with 0.48-kb Drosophila CuZn SOD, 1,4.kb human catalase and 5,4-kb Drosophila hsp70 8erie.

277

SOD and catalase with age following heat shock. In the case of catalase, the increased production of H202 in old heat stressed flies seems to be involved (unpublished data). Conclusion Collectively, many of the recent data from our laboratory demonstrate that oxidative stress plays a major role in governing the lifespan of Drosophila both during normal metabolism and under conditions of stress. For example, the metabolic rate measured as 02 consumption of these flies housed under various conditions inversely correlates with their lifespan. Evidence of oxidative stress such as lipofuscin has been shown to accumulate with age and under conditions of increased oxidative stress. In some instances, the level of antioxidant protection inherent in different strains of insects correlates with their mean iifespan. The introduction of antioxidants into the food of flies has been reported, at least in the case of propyl gallate, to extend the maximum lifespan of a population of flies. Moreover, the insertion of foreign genes that code for antioxidant enzymes using P-element mediated transformation results in protection from oxidative stress and significantly increases their lifespan. Resistance to oxidative stress, such as that induced by thermal shock, is drastically reduced in senescent flies and correlates with the increased protein damage generated in old flies by stress and/or insufficient protection from cellular defense systems. References
Ames, B. (1990) Oxidative mitochondrial DNA damage, Age, 13, 103. Andreoli, S.P., C.P. Malett and J.M. Bergstein (1986) Role of glutathione in protecting endothelial cells against hydrogen peroxide oxidant injury, J. Lab. Clin. Med., 108, 190198. Arking, R., S. Buck, A. Berries, S. Dwyer and G.T. Baker IlI (1991) Elevated paraquat resistance can be used as a bioassay for longevity in a genetically based long-lived strain of Drosophila, Dev. Genet., 12, 362-370. Baker ilk G.T., M. Jacobsen and G. Mokrynski (1985) Aging

in Drosophila, in: V. Christofalo (Ed.), Handbook of Cell Biology of Aging, CRC, Boca Raton, F L pp. 571-578. Banosz, G., W. Leyko and R. Fried (1979) Superoxide dismutase and life-span of Drosophila melanogaster, Experientia, 35, 1193-1194. Bochner, B.R., P.C. Lee, S.W. Wilson, C.W. Cutler and B.N. Ames (1984) AppppA and related adenylated nucleotides are synthesized as a consequence of oxidation stress, Ceil, 37, 225-232. Boveris, A., N. Oshino and B. Chance (1972) The cellular production of hydrogen peroxide, Biochem. J., 128, 617630. Carlioz, A., and D. Touati (1986) Isolation of superoxide dismutase mutants in Escherichia coil: is superoxide dismutase necessary for aerobic life?, EMBO J., 5, 623-630. Christman, M.F., R.W. Morgan, S.F. Jacobsen and B.N. Ames (1~'85) Positive control of a regulon for defenses against oxidative stress and some heat shock proteins in Salmonella typhimurium, Cell, 41,753-762. Courgeon, A.-M., E. Rollet, J. Becket, C. Maisonhaute and M. Best-Belpomme (1988) Hydrogen peroxide (H20 2) induces aetin and some heat-shock proteins in Drosophila cells, Eur. J. Biochem., 171, 163-170. EIroy-Stein, O., Y. Bernstein and Y. Groner (1986) Overproduction of human CnZn superoxide dismutase in transfected cells: extenuation of paraquat-mediated cytotoxicity and enhancement of lipid peroxidation, EMBO J., 5, 115122. Falkner, F.G., H. Saumweber and H. Biessman (1981) The Drosophila melanogaster proteins related to intermediate filament proteins of vertebrate cells, J. Cell. Biol., 91, 175-183. Fleming, J.E., H.A. Leon and J. Miquel (1981) Effects of ethidium bromide on development and aging of Drosophila: ]replications for the free radical theory of aging, Exp. Gerontol., 16, 287-293. Fleming, J.E., J. Mique], S.F. Cottre]l, L.S. Yensoyan and A.C. F~onomos ([982) ]s cell aging caused by respirationdependent injury to the mitochondrial genome?, Gerontology, 28, 44-53. Fleming, J,E., P.B. Melnikoff and K.G. Bensch (1984) Identification of mitochondrial proteins in two dimensional electrophoretic gels of adult Drosophila melanogaster, Biochim. Biophys. Acta, 802, 340-345. Fleming, J.E., J. Miquel and K.G. Bensch (1985) Age dependent changes in mitochondria, in: Molecular Biology of Aging, Plenum, New York, NY, pp. 143-155. Fleming, J.E., R. Shibuya and K.G. Bensch (1987) Lifespan, oxygen consumption and hydroxyl radical scavenging capacity of two strains of Drosophila melanogaster, Age, 10, 86-89. Fleming, J.E., J.K. Walton, R. Dubitsky and K.G. Bensch (1988) Aging results in an unusual expression of Drosophila heat shock proteins, Prec. Natl. Acad. Sci. (U.S.A.), 85, 4099-4103. Freeman, M.L.. and M.J. Meredith (1989) Glutathione conjugation and induction of a 32,000 dalton stress protein, Biochem. Pharmaeol., 38, 299-304.

278 Fridovich, I. (1975) Superoxide dismutase, Annu. Rev. Biochem., 44, 147-159. Fridovich, !. (1988) The biology of oxygen radicals: general concepts, in: B. Halliwell (Ed.), Oxygen Radicals and Tissue Injury, Upjohn, Bethesda, MD, pp. 1-5. Fukanaga, M., and ICL. Yielding (1979) Fate during cell growth of yeast mitochondria and nuclear DNA after photolytic attachment of the mono~zide analog of ethidium, Biochem. Biophys. Res. Commun., 90, 582-586. Gerschman, R. (1962) Man's dependence on the earthly atmosphere, in: ICS. Schaefer (Ed.), Proceedings of the 1958 [First] International Symposium on Submarine and Space Medicine, McMillan, New York, NY, pp. 475-492. Gerschman, R. (1981) Historical introduction to the 'Free radical theory of oxygen toxicity', in: D.L. Gilbert (Ed.), Oxygen and Living Processes: An Interdisciplinary Approach, Springer Verlag, New York, NY, pp. 44-46. Guedon, G.F., G.T. Gilson, J.P. Ebel, N.M. Befort and P.M. Remy (1986) Lack of correlation between extensive accumulation of bisnucleotide phosphates and the heat shock response in eukaryotic cells, J. Biol. Chem., 261, 1645916465. Gutteridge, J.M., C.T. Westermarck and B. Halliwell (1986) Oxygen radical damage in biological systems, in: J.E. Johnson Jr. (Ed.), Free Radicals, Aging and Degenerative Diseases, Alan Liss, New York, NY, pp. 99-139. Hahn, G.M., E.C. Shiu and E.A. Auger (1991) Mammalian stress proteins HSP70 and HSP28 coinduced by nicotine and either ethanol or heat, Mol. Cell. Biol., 11, 6034-6040. Harman, D. (1956) Aging: a theory based on free radical and radiation ct,emistry, J. Gerontol., 11,298-300. Harman, D. (1982) The free radical theory of aging, in: W.A. Massie, H.R., M.B. Baird and M.M. McMahon (1975) Loss of mitochondrial DNA with aging, Gerontology, 21,231-237. McCord, J.M., and !. Fridovich (1969) Superoxide dismutase: an enzymatic function for erythrocuprein (hemocuprein), J. Biol. Chem., 224, 6049-6055. Medawar, P.B. (1952) An Unsolved Problem of Biology, Lewis, London. Melhorn, R.J., and G. Cole (1985) The free radical theory of aging: a critical review, Adv. Free Radical Biol. Med., 1, 165-223. Miquel, J., and J.E. Fleming (1984) A two-step hypothesis on the mechanism of in vitro cell a~h,g', .Cell. d~ffer~nti~.*i~n followed by intrinsic mitochondrial mutagenesis, Exp. Gerontol., 19, 31-36. Miquel, J., and J.E. Fleming (1986) Theoretical and experimental support for an oxygen radical-mitochondrial injury hypothesis of cell aging, in: J.E. Johnson, R. Walford, D. Harman and J. Miquel (Eds.), Free Radicals, Aging and Degenerative Diseases, Alan R. Lisa, New York, NY, pp. 51-74. Miquel, J., P.R. Lundgren, K.G. Bensch and H. Atlan (1976) Effects of temperature on the life span, vitality and fine structure of Drosophila melanogaster, Mech. Ageing Dev., 5, 347-370. Miquel, J., J. Or6, K.G. Bensch and J.E. Johnson Jr. (1977) Lipofuscin: fine-structural and biochemical studies, in: W.A. Pryor (Ed.), Free Radicals in Biology, Academic Press, Hew York, NY, pp, 133-182. Miquel, J., P.R. Lundgren and J.E. Johnson (1978) Spec-

Pryor (Ed.), Free Radicals in Biology, Vol. V, Academic Press, New York, NY, pp. 255-275, Hass, M.A., and D. Massaro (1988) Regulation of the synthesis of superoxide dismutases in rat lungs during oxidant and hyperthermic stresses, J. Biol. Chem., 263, 776-781. Hassan, H.M,, and I, Fridovieh (1978) Superoxtde radical and the oxygen enhancement of the toxicity of paraquat in Escherichia colt, J. Biol, Chem., 253, 8143-8148, Krall, J., A.C. Bagley, G.T. Mullenbach, R.A. Hallewell and R.E. Lynch (1988) Superoxide mediates the toxicity of paraquat for cultured mammalian cells, J. Biol. Chem., 263, 1910-1914. Lee, P.C., B.R. Bochner and B.N. Ames (1983) AppppA, heat-shock stress, and cell oxidation, Proc. Natl, Acad. Sci. (U.S.A.), 80, 7496-7500. Levinson, W., P. Mikdens, H. Opperman and J. Jackson (1978) Effect of antabuse (disulfiram) on Roos sarcoma vires an~ on eukaryotic cells, Bioehim. Biophys. Acta, 519, 65-75. Loeb, J., and J.H. Northrup (1917) On the influence of food and temperature on the duration of life, J. Biol. Chem., 32, 103-121. Loven, D.F, D.B. Leeper and L.W. Oberley (1985) Superoxide dismutase levels in Chinese hamster oval, cells and ovarian carcinoma cells after hyperthermia or exposure to cycloheximide, Cancer Res., 45, 3029.-3033.

trophotofluoromctric and electron microscopic study of lipofuscin accumulation in the testis of aging mice, J. Gerontol,, 33, 5-19. Miquel, J., A.C. Economos, K,G. Bensch, H, Allan and J,E. Johnson Jr, (1979) Review of cell aging in Drosophila and mouse, Age, 2, 78-88,
Miquel, J., A.C Economos. J. Fleming and J.E. Johnson Jr. (1980) Mitochondrial role in cell aging, Exp. Gerontol., 15, 575-591, Miquel, J., A.C Economos and K.G. Bensch (1981) Insect vs mammalian aging, in: J.E. Johnson Jr. (Ed.), Aging and Cell Structure, Plenum, New York, NY, pp. 347-379. Miquel, J., J.E. Fleming and A.C. Economos (1982) Antioxidants, mitochondrial respiration and aging in Drosophila, Arch, Gerontol, Geriatr., 1, 349-363. Miquel, J., R. Binnard and J.E. Fleming (1983) Role of metabolic rate and DNA repair in Drosophila ageing: Implications for the mitochondrial mutation theory of cell aging, Exp. GerontoL, 18, 161-171. Miquel, J,, A.C Economos and J.E. Johnson Jr. (1984) A systems-thermodynamic view of cell and organismic aging, in: J.E, Johnson Jr, (Ed,), Aging and Cell Function, Plenum, New York, NY, pp. 247-280. Mitchell, J,B., A, Russo, T,J. Kinsella and E. Giatstein (1983) Glutathione elevation during thermotolerance induction and thermosensitization by glutathione depletion, Cancer Res., 43, 987-991. Morgan, R.N., M.F. Christman, F.S. Jacobson, G. Story and B.N. Ames (1986) Hydrogen peroxide-indocible proteins

279 in Salmonella typhimurium overlap with ~eat shock and other stress proteins, Prec. Natl. Acad. Sci. (U.S.A.), 83, 8059-8063. Morimoto, R.I., A. Tissi~res and G. Georgopoulos (1990) The stress response, function of proteins and perspectives, in: R.I. Morimoto, A. Tissi~res and G. Georgopoulos (Eds.), Stress Proteins in Biology and Medicine, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 1-36. Munkres, K.D. (1979) Ageing of Neurospora crassa. IX. Microviscosity properties of mitochondrial membranes during normal and abnormal growth and development of an inositel auxotroph, Mech. Ageing Dev., 10, 173. Niedzwiecki, A., and J.E. Fleming (1990) Changes in protein turnover after heat shock are related to accumulation of abnormal proteins in aging Drosophila melanogaster, Mech. Ageing Dev., 52, 295-304. Niedzwiecki, A., A.M. Kongpachith and J.E. Fleming (1991) Aging affects expression of 70-kDa heat shock proteins in Drosophila, J. Biol. Chem., 266, 9332-9338. Nohl, H., and D. Hegner (1978) Do mitochondria produce oxygen radicals in vitro?, Eur. J. Biochem., 82, 563-567. Omar, R., S. Yano and Y. Ki!dcawa (1986) Oxygen free radicals in hyperthermic injury, Fed. Prec., 45, 452. Pearl, R. (1928) The Rate of Living, Knopf, New York, NY. Pelham, H.R.B. (1986) Speculation on the functions of the major heat shock and glucose regulated proteins, Cell, 46, 959-961. Peng, T.X., A. Moya and F.J. Ayala (1986) Irradiation resistance conferred by superoxide dismutase: possible adaptive role of a natural polymorphism in Drosophila melanogaster, Prec. Natl. Aead. Set. (U.S.A.), 83, 684-687. Phillips, J.P., S.D. Campbell, D, Michard, M. Charbonneau and A.J. Hilliker (1989) Null mutation of copper/zinc superoxide dismutase in Drosophila confers hypersensitivity to paraquat and reduced longevity, Prec. Natl. Acad. Sci. (U.S.A.), 86, 2761-2765. Polla, B.S., A,M. Healy, E.P. Amento and S.M. Krune (1986) 1,25-Dihydroxyvitamin D3 maintains adherence of human monocyles and protects them from thermal injury, J. Clin. Invest., 77, 1332-1339. Polla, B.S., A.M. Healy, W.C. Wojmo and S.M. Krane (1987) Hormone la,25.dihydroxyvitamin D 3 modulates heat shock response in monocytes, Am. J. Physiol., 252, C640C649. Polla, B.S., J.V. Bonventre and S.M. Krane (1988) 1,25.Dihydroxyvitamin D3 increases the toxicity of hydrogen peroxide in the human monocytic line U937: the role of calcium and heat shock, J. Cell. Biol., 107, 373-380. ReveiUaud, I., A. Niedzwieeki, G.G. Bensch and J.E. Fleming (1991) Expression of bovine superoxide dismutase in Drosophila melanogaster augments resistance to oxidative stress, Mol. Cell. Biol., ! !, 632-640. Ritossa, F.M. (1962) A new puffing pattern induced by a temperature shock and DNP in Drosophila, Experientia, 18, 571-573. Ritossa, F.M. (1963) New puffs induced by temperature shock, DNP and salicylate in salivary chromosomes of Drosophila melanogaster, Drosophila inf. Serv., 37, 122-]23. Ritossa, F.M. (1964) Specific loci in polytene chromosomes of Drosophila, Exp. Cell Res., 35, 601-607. Rubin, G.M. (1986) Lab Methods Book, Berkeley, CA. Rubner, M. (1908) Das Problem der Lebensdauer und seine Beziehungen zu Wachstum und Ern~hrung, R. Oldenburg, Munich. Ruddle, D.L., LS. Yengoyan, J. Miquel, R. Marcuson and .I.E. Fleming (1988) Propyl gallate delays senescence in Drosophila melanogaster, Age, 11, 54-58. Scott, M.D., S.R. Meshnick and J.W. Easton (1987) Superoxide dismutase-rich bacteria: paradoxical increase in oxidant toxicity, J. Biol. Chem., 262, 3640-3645. Seto, N.W., S. Hayashi and G.M. Tener (1990) Overexpression of Cu-Zn superoxide dismutase in Drosophila does not affect life-span, Prec. Natl. Acad. Sci. (U.S.A.), 87, 4270-4274. 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. Sinet, P.M., L. Lejeune and H. Jerome (1979) Trisomy 21 (Down's syndrome) giutathione peroxidase, hexose monopbosphate shunt and IQ, Life Sci., 24, 29-34. Sohal, R.S. (1976) Metabolic rate and lifespan, in: R. Wirier (Ed.), Cellular Aging: Concepts and Mechanisms, Karger, Basel, pp. 25-40. Sohai, R.S. (1987) The free radical theory of aging: a critique, Rev. Biol. Res. Aging, 3, 431-449. Sohal, R.S., and P.B. Bucham (1981) Relationship between physical activity and life span in the adult housefly Musca domestica, Exp. Gerontol., 16, 159-167. Spitz, D.R., W.C. Dewey and G.C. Li (1987) Hydrogen peroxide or heat shock induces resistance to hydrogen peroxide in Chinese hamster fibroblasts, J. Cell. Physiol., 131,364373. Spradling, A.S, (1986) P.clement-mediated transformation, in: D.B. Roberts (Ed.), Drosophila: A Practical Approach, IRL, Oxford, pp. 175-197. Stadtman, E.R. (1986) Oxidation of proteins by mixed-function oxidation systems: implication in protein turnover, ageing and neutrophil function, Trends Biochem. Sci., I l, 11-12. Thomas, G.P., WJ. Welch, M.P. Mathews and J.R. Feramisco (1982) Molecular and cellular effects of heat shock and related treatments of mammalian tissue culture cells, Cold Spring Harbor Symp. Quant. Biol., 46, 985-996. Wunderlich, V., I. Tetzlaff and A. Graffi (1971/1972) Studies on nitrosodimethylamine: Prefereptial methylation of mitochondrial DNA in rats and hamsters, Chem.-Biol. Interact., 4, 81-89. Yost, HJ., R.B. Petersen and S. Lindquist (1990) Posttranscriptional regulation of heat shock protein synthesis in Drosophila, in: R.I. Morimoto, A. Tissi~res and G. Georgopoulos (Eds.), Stress Proteins in Biology and Medicine, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 379-409.