Journal of Gerontology: BIOLOGICAL SCIENCES 1998, Vol. 53A, No. 4, B24O-B244

Copyright 1998 by The Gemntological Society of America

Effects of Oxygen on Protein Carbonyl and Aging in Caenorhabditis elegans Mutants With Long (age-1) and Short (mev-1) Life Spans Hiroshi Adachi,1-2 Yoshisada Fujiwara,3 and Naoaki Ishii2 'Life Science Research Center, Lion Corporation, Kanagawa, Japan. department of Molecular Life Science, Tokai University School of Medicine, Kanagawa, Japan, department of Radiation Biophysics and Genetics, Kobe University School of Medicine, Kobe, Japan.

Protein carbonyl accumulation is an indicator ofoxidative damage during aging. The relationship between oxidative stress and protein carbonylation during aging was studied by using a long (age-1) and a short (mev-1) life span mutant of Caenorhabditis elegans. Protein carbonyl concentrations were similar in young adults of both mutants and wild type; however, the subsequent age-dependent accumulation was different with the genotype. The mev-1 mutant (with 50% superoxide dismutase activity) accumulated protein carbonyl at a faster rate than did wild type, whereas the age-1 mutant exhibited no obvious increase except a significant accumulation at the end of extended life span. Exposure to 70% oxygen between the ages of 4 and 11 days caused a far greater accumulation of carbonyl in mev-1 than in wild type, but not in age-1. In addition, rates of aging were enhanced by oxygen in a concentration-dependent fashion. The age-1 mutant was more resistant to, but mev-1 was more sensitive to, such oxygen enhancements of aging than was wild type. These results provide further evidence that oxidative damage is one of the major causal factors for aging in C. elegans, and that the age-1 and mev-1 genes govern resistance to oxidative stress.

T

HERE are a variety of hypotheses that seek to explain the mechanism of aging (1-3). One such hypothesis is that oxygen free radicals are causally involved in the aging process in aerobic organisms (4). This hypothesis is supported by the following lines of evidence: (i) reactive intermediates of oxygen reduction are constantly generated in aerobic organisms (5,6), (ii) the concentration of reaction products with free radicals increases during aging under physiological conditions (7,8), and (iii) environmental factors that produce oxidative stress enhance the accumulation of oxidative molecular damage (9). Insufficiency in both or either antioxidative defense mechanisms and repair systems may cause accumulation of oxidative molecular damage, which consequently leads to homeostatic disturbance during the aging process. However, the direct causal relationship needs to be established among oxidative stress, antioxidant defenses, and accumulation of oxidative molecular damages in the aging process. A genetic approach is useful for determining free radical involvement in aging. In particular, the nematode Caenorhabditis elegans offers several distinct advantages such as (i) a short life span of approximately 20 days, (ii) a 3-day life cycle, (iii) an excellent system for genetic analysis (10-12), and (iv) an excellent system for detecting cumulative age-related cellular alterations because the somatic tissues of the nematode consist of long-lived postmiotic cells. A number of life span (LS) mutants have been isolated in which life span is either longer or shorter than that of wild type (13-16). The age-1 mutant exhibits a 65% extension of mean LS and a greater elevation of both catalase and su-

B240

peroxide dismutase (SOD) activities at old age as compared with wild type (14,15,17,18). In contrast, the mev-1 mutant shortens the mean LS by 30% and reduces the SOD activity to about 50% of wild type (16). Stadtman and associates have demonstrated that protein carbonyl derivatives are formed in vitro as a result of metalcatalyzed oxidations and accumulate during aging in disparate model systems (19-23). These results indicate that protein carbonyl modifications can be a specific indicator of oxidized protein. Sohal et al. (9) have shown that protein carbonyl content was not only inversely correlated with the life expectancy of the houseflies, but increased under sublethal hyperoxia condition (100% oxygen). Therefore, the mev-1 short LS and age-1 long LS mutants are useful for elucidating the role of oxidative stress in the accumulation of protein carbonyl during the aging process in C. elegans. Here we measured carbonyl contents during aging in age-1 and mev-1 under both normal atmospheric and hyperoxic conditions to understand the role of oxidative stress at the protein level in aging of C. elegans. MATERIALS AND METHODS

Nematode Strains The C. elegans strains used were N2 (wild type), which carries the normal alleles of age-1 and mev-1; TJ401 (long LS mutant), which carries age-1 {hx546)\ and TK22 (short LS mutant), which carries mev-1 (knl). The wild-type N2 strain was from the Caenorhabditis Genetic Center (University of Minnesota, St. Paul, MN). The age-1 mutant was kindly sup-

PROTEIN CARBONYL AND AGING OF C. ELEGANS

plied by Dr. Thomas Johnson (University of Colorado, Boulder, CO). The mev-1 mutant was isolated from ethyl methanesulfonate-treated wild-type animals by Ishii and colleagues (16), and is sensitive to methyl viologen (paraquat) that generates superoxide anions. Stock maintenance and handling were carried out as described by Brenner (24). Culture Conditions Hermaphrodites of the wild type and mutants were grown at 20°C on NG agar plates, with live bacteria (Escherichia coli OP50, a uracil auxotroph) as food (24). Synchronous cultures were obtained as previously described (25). In brief, eggs were collected from NG agar plates by using sodium hypochlorite (26) and allowed to hatch by incubation overnight at 20°C in S buffer (0.1 M NAC1, .05 M potassium phosphate buffer, pH 6.0) (24). The newly hatched LI larvae were cultured on NG agar plates (using 150-mm plastic plates for protein carbonyl measurement and 35-mm plates for determining LS). To prevent progeny production, 5-fluoro-2'-deoxyuridine (FudR; Sigma Chemical, St. Louis, MO) was added to the NG agar at a final concentration of 40 uM after animals had reached adulthood. To measure survival and mean LS, animals were counted every day by eliminating dead ones. Death was recognized as the loss of spontaneous movement and lack of response to touch with a probe.

B241

per 1 mg of protein. Protein concentration was determined by the Pierce BCA method (28). Oxygen Exposure Animals cultured on the NG agar plates were exposed to various concentrations of oxygen from 4 days after hatching to the end of LS in an air-tight plastic chamber (29). Either 100% oxygen or 100% nitrogen was driven at a rate of 2 liters/min into an air atmosphere of each chamber until the desired concentration was achieved. The closed chambers were kept at 20°C, and gas within chambers was replaced once a day. Statistical Analyses Statistic comparisons of mean LS values among strains under atmospheric condition, and among various concentrations of oxygen at each strain, were determined by Duncan's Multiple Range test. For age-dependent changes in protein carbonyl contents, the statistical significance of differences among all strain by time combinations were determined by Duncan's Multiple Range test. Data of protein carbonyl contents under hyperoxia condition were subjected to two-way analysis of variance (ANOVA) by oxygen concentrations and strains, and a Least Significant Difference test was used for means separation at/? < .05 within strain. RESULTS

Measurement of Protein Carbonyl Content Animals were collected from the NG agar plates and washed several times with S buffer. They were placed on a 40-um mesh filter in a small chamber filled with S buffer, and only living animals that swam out through the mesh within 10 min were collected. Live animals were washed several times with S buffer, resuspended with 5 mM EDTA (pH 8.0), and frozen at -80°C until use. In each experiment a 10% (vol/vol) homogenate in S buffer was made using a Teflon homogenizer. Protein carbonyl content was measured by a slightly modified method of Levine et al. (27) using 2,4-dinitrophenylhydrazine (DNPH). A 1 ml aliquot of 20% trichloroacetic acid (TCA) was added to 1 ml of the homogenate to precipitate the protein, and the mixture stood for 15 min at 4°C. After centrifugation at 2,000 X g for 15 min, the supernatant was discarded, and 0.5 ml of 10 mM DNPH dissolved in 2 M HC1 was added to the protein fraction. In parallel, a blank was prepared by treatment with 2 M HC1 instead of DNPH. After incubation at 27°C for 60 min, 0.5 ml of TCA was added to the sample, and the mixture stood for 15 min at 4°C. After centrifugation at 40,000 X g for 15 min, the supernatant was discarded. The pellet was washed three times with 1 ml of a 1:1 (vol/vol) mixture of 20% TCA with 100% ethanol and ethyl acetate to remove free reagent. The sample was allowed to stand for 15 min at 4°C and then centrifuged at 40,000 X g; the supernatant was discarded each time. The final precipitated protein was redissolved in 1.5 ml of 6 M guanidine hydrochloride solution. After incubation for 5 h at 37°C, the sample was centrifuged at 600 X g for 10 min. Absorbance at 380 nm was determined and, using a molar absorption coefficient of 21 mM"1 • cnr1, the carbonyl content was calculated as nmol of DNPH incorporated (protein carbonyls)

Life Spans of age-\ and mev-l Figure 1 shows the representative survival curves of N2, age-J, and mev-1. The mean LS values were 20.6 days for N2, 33.1 days for age-J, and 16.1 days for mev-1. Thus, the mean LS was 61% longer in age-J, but 22% shorter in mev-1 than in N2 at 20°C in air (21% oxygen). These results are similar to values reported previously for these strains (15,16). Effect of Age on Protein Carbonyl Content Carbonyl contents of N2, age-J, and mev-1 were measured at five different times in animals grown in air (21%

10 15 20 25 30 35 40 45 50 55 60 Age (days) Figure I. Survival curves in air (21% oxygen). # , Wild type N2 (n = 107); U, age-1 (n = 104); D, mev-l (n = 100). Mean LS values with different letters are significantly different {p < .01) according to Duncan's Multiple Range test. Animals were kept at 20"C through the life span.

ADACHIETAL.

B242

oxygen). As shown in Figure 2, the protein carbonyl contents in young N2, age-], and mev-1 adults at the age of 4-8 days were similar (between 2.2 and 2.5 nmol/mg protein). Afterwards, different accumulations of carbonyl were observed with the genotype and increasing age. In N2, an age-dependent accumulation in carbonyl content was observed until the end of LS around 20 days to reach 4.7 ± 0.4 nmol/mg protein, whereas in mev-1 it occurred at a faster rate to reach 5.7 ± 0.5 nmol/mg protein at the end of LS (age 15-16 days) than in N2. In contrast, age-J showed no obvious increase in carbonylated protein (2.4-2.8 nmol/mg) until age 18-20 days, followed by a significant increase to 4.0 ± 0.4 nmol/mg at the end (39 days) of LS. Furthermore, carbonyl contents at the ages giving equisurvivals of 90-100% and 40-60% were the highest in mev-1 and the lowest in age-1 (Figure 2). Thus, the shorter life span is well associated with the greater carbonyl accumulation and vice versa. Effect of 70% Oxygen on Carbonyl Content To reveal the defense ability in age-1 and mev-1 operated under higher levels of oxidative stress, protein carbonyl contents were compared after exposure to 70% oxygen between age 4 and 11 days. Figure 3 shows the results measured at the age of 11 days. Such hyperoxia caused 100% (p < .001) and 31% (p < .05) increases in carbonyl in mev-1 and N2 over the respective basal levels in the ambient atmosphere of 21% oxygen, respectively. However, there was no significant increase even by 70% oxygen in age-1 at the age of 11 days. Thus, mev-1 and age-1 are also sensitive and resistant to the hyperoxia-induced protein oxidation, respectively.

Effects of Different Oxygen Concentrations on Aging The above differences at the protein level may be reflected in differential rates of aging. To establish the relationship between intensity of oxidative stress and aging, we measured survival curves of N2, mev-1, and age-1 by continuous exposure to 1, 21, 40, 60, and 80% oxygen between the age 4 days and the end of LS. As shown in Figure 4, the 1% hypoxic condition extended mean LS 23% for mev-1, 26% for N2, and 21% for age-1. The mean LS values of N2 and age-1 (20 and 35 days, respectively) under oxygen between 21% and 40% remained unchanged. On the other hand, the mean LS of mev-1 was shorter under 40% oxygen than under 21% oxygen. Exposure to 60% and 80% oxygen shortened the mean LS values of all three strains in comparison with 21% oxygen. It is remarkable that the 80% oxygen survival curve of age-1 resembled the 1 % hypoxic survival curve of mev-1 and the air (21% oxygen) survival curve of N2. This pattern correlates with the antioxidant capacities of the respective strains, as age-1 has high levels of SOD and catalase (17), whereas mev-1 possesses 50% SOD as wild type (16). DISCUSSION

Stadtman, Oliver, and colleagues have shown that protein carbonyl derivatives are formed in vitro and in vivo as a result of metal-catalyzed oxidation and accumulate during (he aging process (19-23, 30). Orr and Sohal (31) have shown that overexpression of SOD and catalase retards protein carbonyl accumulation and extends LS (6-33%) of transgenic houseflies. The present results have further demonstrated the relevance of protein oxidation to the aging process by showing that the protein carbonyl content is affected by two genes that regulate resistance to oxidative stress. The carbonyl contents at the equi-survival levels of 40-60% and 90-100% were highest in mev-1 and lowest in age-1 (Figure

b

• • . •

S

21 % oxygen 70% oxygen

b

a

a T

a o 4-8 days

•s

11-13 15-16 days days

18-20 days

39 days

100 90-100 40-60 100 100 90-100 mev-1 100 90-100 40-60 0

30-50

-N2

100 age-1 100

U

0

I

I

I

I

I

I

15

I

2 -

0

0

I

30

45

Figure 2. Age-dependent changes in protein carbonyl contents in homogenates of wild-type N2, age-1, and mev-l. Symbols for the strains are indicated in the figure. Protein carbonyl contents were measured as described in Materials and Methods. Each value of mean ± SD at each different age group was determined by four to five separate determinations. Protein carbonyl contents among all strains by time combination with different letters are significantly different (p < .01) according to Duncan's Multiple Range test. Survival fractions (%) for each group are indicated at the bottom of the figure.

T

i

•

aee-1

•

a

N2

mev-1

Figure 3. Effect of 70% hyperoxia on carbonyl contents in age-1, wildtype N2, and mev-1. Protein carbonyls were measured after exposure to 21 or 70% oxygen from 4 days after hatching to 11 days. Each histogram for mean ± SD was derived from three separate determinations. Protein carbonyl contents are significantly different within strain (p < .001) and within oxygen concentration (p < .001) according to two-way ANOVA. Data of protein carbonyl contents within strain with different letters are significantly different (p < .05) based on a Least Significant Difference test analysis.

PROTEIN CARBONYL AND AGING OF C. ELEGANS

02 rneanLS±SD 1% 43.4 ±5.4 a 21% 35.4 ± 8.7 b 40% 373 ± 5.5 b 60% 27 J ± 6.1c 80% 13.9 ± 1.2 d

60 neanLS±SD O2 H 1% 25.9 ±3.5 a 21% 20.9 ± 4.0 b 40% 21.2 ± 3.3 b 60% 14.6 ± 3.1c 80% 10.8 ± 1.4 d

20

30

40

'""'3!tev

50

60

02 nK uLS—SP a

1% 18.0 ±4.4 a

life

16.1 ±3 J b

40% 12.9 ± 2.9 c 60% 115 ± 1.7 d 80% 10.8 ± 1.6 d

60%

V* •

•

10

!

20

30

40

50

60

Age (days) Figure 4. Survival curves of age-l, wild-type N2, and mev-I under various concentrations of oxygen continuously present later than 4 days after hatching. (A) age-1, (B) Wild-type N2, (C) mev-1. Oxygen concentrations: 1% (•), 21% ( 0 ) , 40% (•), 60% (OX and 80% (•). Mean LS values with different letters at each strain are significantly different (p < .01) according to Duncan's Multiple Range test. About 100 animals were used for each experiment.

2), which represents a positive correlation between LS and carbonyl level. The accumulation of protein carbonyl content is repressed in the long LS mutant age-1 but enhanced in a short LS mutant mev-1 (Figures 2 and 3) by the respective high and low antioxidant activities. In this regard, Larsen (17) observed an aging-associated increase in activities of SOD and catalase in age-1, despite the identical superoxide levels in age-1 and wild type for the first 2 weeks (18). The mev-1 mutant has only half of the wild-type SOD activity (16) and enhances aging (Figures 1 and 3). Therefore, the above results strengthen the link between oxidative stress-mediated protein oxidation and aging. In the survival curves in Figure 4 A-C, aging was en-

B243

hanced differently in age-1, N2, and mev-1 by oxygen in a concentration-dependent manner. These results likely reflect variable responses to different intensities of oxidative stress (16) (Figure 4). Interestingly, no difference in the protein carbonyl content was detected among N2, mev-1, and age-1 in the young adult phase (Figure 2). Honda et al. (29) have shown that exposure of mev-1 to 1% oxygen only during an early phase (from age 4-10 days after hatching) does not extend its LS. Exposure to 1% oxygen during a post-early phase (from age 11 days after hatching to the end of LS) (29) or continuous 1 % oxygen exposure during the life time later than age 4 days (Figure 4) can extend LS of mev-1, suggesting that even the mev-1 can keep the homeostasis during early phase in LS, despite a low level of SOD activity (16). Therefore, mev-1 that has a low antioxidant activity can tolerate a low level of oxidative stress produced by 1% oxygen. The hyperoxia-induced elevations of carbonyl in C. elegans (Figure 3) are similar to those reported previously in the houseflies: the carbonyl content of houseflies increased in response to hyperoxia (100% oxygen), the highest increase occurring between day 2 and day 3 of exposure (9). The increases in carbonyl levels induced at 11 days by 70% oxygen were 100% in mev-1 and 30% in N2 over the levels under 21% oxygen, indicating that mev-1 is more sensitive to excess oxidative stress than is N2. However, such a hyperoxia-induced increase was not detected in age-1, so that age-1 has a far greater defense activity against protein oxidation by hyperoxia than does N2 (Figure 3). In conclusion, oxidative damage is one of the major causes of aging, and oxygen is one determinant of aging in C. elegans. The age-1 and mev-1 gene products govern resistance to oxidative stress. Because the SOD and catalase activities at 11 days in age1 are identical with those in wild type (17), additional factors may exist for such a lower carbonyl content in age-1; for example, turnover of carbonylated proteins may be rapid or the antioxidant mechanisms may be inducible by oxidative stress. Morris et al. (32) have recently shown that the age-1 gene encodes a homolog of mammalian phosphatydilinositol-3-OH kinase catalytic subunit. The age-1 gene is identical to the daf-23 gene (which is required for dauer formation, developmental stage) in C. elegans. The normal age-1 gene regulates SOD and catalase negatively and causes normal aging (33,34) (Figures 1 and 4). The daf groups (daf23, daf-2, daf-12, and daf-18) are also required for normal aging (33,34), suggesting that signal transduction-mediated cellular processes that are genetically controlled are also important for the aging process. The relationship between signal transduction and antioxidant activity with respect to aging remains to be solved.

ACKNOWLEDGMENTS

This work was supported in part by a Grant in Aid for Aging Research from the Ministry of Health and Welfare, Japan, to Y.F. and N.I. and by a Grant in Aid for Scientific Research to N.I. (Project no. 08308032) from the Ministry of Education, Science, Sports, and Culture, Japan. We thank P. Hartman (Department of Biology, Texas Christian University) and K. Suzuki (Department of Molecular Life Science, Tokai University School of Medicine) for the critical reading of the manuscript. The ne-

B244

ADACHIETAL.

matode N2 strain was obtained from the Caenorhabditis Genetic Center under the auspices of the National Center for Research Resources. Address correspondence to Dr. Hiroshi Adachi, Life Science Research Center, Lion Corp., 100 Tajima, Odawara, Kanagawa 257-0811, Japan. REFERENCES

1. Roy AK, Chatterjee B, eds. Molecular Basis of Aging. Orlando, FL: Academic Press; 1984. 2. Macieira-Coelho A, ed. Molecular Basis of Aging. Boca Raton, FL: CRC Press; 1995. 3. Masoro EJ, ed. Handbook of Physiology. New York: Oxford University Press; 1995, Section II. 4. Harman D. The aging process. Proc Natl Acad Sci USA. 1981 ;78: 7124-7128. 5. Sohal RS, Sohal BH. Hydrogen peroxide release by mitochondria increases during aging. Mech Ageing Dev. 1991 ;57:187-202. 6. Chance B, Sies H, Boveris A. Hydroperoxide metabolism in mammalian organs. Physiol Rev. 1979;59:527-603. 7. Koppang N. Canine ceroid-lipofuscinosis-A model for human neuronal ceroid-lipofuscinosis and aging. Mech Ageing Dev. 1973;2: 421-445. 8. Strehler BL, Mark DD, Mildvan AS, Gee MV. Rate and magnitude of age pigment accumulation in the human myocardium. J Gerontol. 1959; 14:257-264. 9. Sohal RS, Agarwal S, Dubey A, Orr WC. Protein oxidative damage is associated with life expectancy of houseflies. Proc Natl Acad Sci USA. l993;90:7255-7259. 10. Klass MR. Aging in the nematode Caenorhabditis elegans: major biological and environmental factors influencing life span. Mech Ageing Dev. 1977;6:4l3-429. 11. Bolanowski MA, Russell RL, Jacobson LA. Quantitative measures of aging in the nematode Caenorhabditis elegans. I. Population and longitudinal studies of two behavioral parameters. Mech Ageing Dev. 1981; 15:279-295. 12. Blum J, Fridovich I. Superoxide, hydrogen peroxide, and oxygen toxicity in two free-living nematode species. Arch Biochem Biophys. l983;222:35-43. 13. Klass MR. A method for the isolation of longevity mutants in the nematode Caenorhabditis elegans and initial results. Mech Ageing Dev. l983;22:279-286. 14. Friedman DB, Johnson TE. A mutation in the age-l gene in Caenorhabditis elegans lengthens life and reduces hermaphrodite fertility. Genetics. 1988; 118:75-86. 15. Friedman DB, Johnson TE. Three mutants that extend both mean and maximum life span of the nematode, Caenorhabditis elegans, define the age-l gene. J Gerontol Biol Sci. 1988;43:B 102-B109. 16. Ishii N, Takahashi K, Tomita S, Keino T, Honda S, Yoshino K, Suzuki K. A methyl viologen-sensitive mutant of the nematode Caenorhabditis elegans. Mutat Res. 1990;237:165-171. 17. Larsen PL. Aging and resistance to oxidative damage in Caenorhabditis elegans. Proc Natl Acad Sci USA. 1993;90:8905-8909. 18. Vanfleteren JR. Oxidative stress and ageing in Caenorhabditis elegans. Biochem J. I993;292:6O5-6O8.

19. Stadtman ER, Oliver CN. Metal-catalyzed oxidation of proteins. J Biol Chem. 1991;266:2005-2008. 20. Stadtman ER. Protein oxidation and aging. Science. 1992;257: 1220-1224. 21. Starke-Reed PE, Oliver CN. Protein oxidation and proteolysis during aging and oxidative stress. Arch Biochem Biophys. 1989;275:559—567. 22. Fucci L, Oliver CN, Coon MJ, Stadtman ER. Inactivation of key metabolic enzymes by mixed-function oxidation reactions: possible implication in protein turnover and ageing. Proc Natl Acad Sci USA. 1983;80:1521-1525. 23. Oliver CN, Ahn B, Moerman EJ, Goldstein S, Stadtman ER. Age-related changes in oxidized proteins. J Biol Chem. 1987;262:5488-5491. 24. Brenner S. The genetics of Caenorhabditis elegans. Genetics. 1974; 77:71-94. 25. Mitchell DH, Stiles JW, Santelli J, Sanadi DR. Synchronous growth and aging of Caenorhabditis elegans in the presence of fluorodeoxyuridine. J Gerontol. l979;34:28-36. 26. Emmons SW, Klass MR, Hirsh D. Analysis of the constancy of DNA sequences during development and evolution of the nematode Caenorhabditis elegans. Proc Natl Acad Sci USA. 1979;76:1333-1337. 27. Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz A, Ahn B, Shaltiel S, Stadtman ER. Determination of carbonyl content in oxidatively modified proteins. Methods Enzymol. 1990; 186:464-478. 28. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein using bicinchoninic acid. Anal Biochem. 1985; 150:76-85. 29. Honda S, Ishii N, Suzuki K, Matsuo M. Oxygen-dependent perturbation of life span and aging rate in the nematode. J Gerontol Biol Sci. 1993;48:B57-B61. 30. Carney JM, Starke-Reed PE, Oliver CN, Landum RW, Cheng MS, Wu JF, Floyd RA. Reversal of age-related increase in brain protein oxidation, decrease in enzyme activity, and loss in temporal and spatial memory by chronic administration of the spin-trapping compound JV-terr-butyl-a-phenylnitrone. Proc Natl Acad Sci USA. 1991 ;88: 3633-3636. 31. Orr WC, Sohal RS. Extension of life-span by overexpression of superoxide dismutase and catalase in Drosophila melanogaster. Science. 1994;263:l 128-1130. 32. Morris JZ, Tissenbaum HA, Ruvkun G. A phosphatidylinositol-3-OH kinase family member regulating longevity and diapause in Caenorhabditis elegans. Nature. l996;382:536-539. 33. Kenyon C, Chang J, Gensch E, Rudner A, Tabtiang R. A C. elegans mutant that lives twice as long as wild type. Nature. I993;366: 461-464. 34. Larsen PL, Albert PS, Riddle DL. Genes that regulate both development and longevity in Caenorhabditis elegans. Genetics. 1995; 139:

1567-1583.

Received March 4, 1997 Accepted February 3, 1998