mitochondrial dna remains intact during drosophila aging ... · these changes correlate with the...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 6 1993 by The American Society for Biochemiatry and Molecular Biology, Inc.

Vol. 268, No. 25. Issue of September 5, pp. 18891-18897,1993 Pnnted m U. S. A.

Mitochondrial DNA Remains Intact during Drosophila Aging, but the Levels of Mitochondrial Transcripts Are Significantly Reduced*

(Received for publication, March 26, 1993)

Manuel CallejaS, Pilar PeiiaSO, Cristina UgaldeS, Carmen Ferreiroll, Roberto MarcoS, and Rafael GaresseSII

Autdmmu de Madrid, c/Arzobispo Morcillo 4, Madrid 28029, Spain From the $Departanento de Bwquimica and Znstituto de Investigaciones Bwmedicas (CSIC) Facultad de Medicina, Uniuersidad

It has been suggested that mutations accumulated in mitochondrial DNA during the aging process may be causally related to the decreased physiological response of the senescent organisms. We have quantified and evaluated the integrity of the mitochondrial genome during the life span of Drosophila melanogaater. Its amount remains fairly constant representing roughly 1% of the total DNA at all ages. Southern experiments have also revealed a high stability and integrity of the mitochondrial DNA (mtDNA). How- ever, we have detected an important decrease in the steady-state levels of all mitochondrial transcripts investigated: 16 S ribosomal RNA (lBSrRNA), cytochrome c oxidase, cytochrome b, and B H+-ATP synthase subunit. These changes correlate with the shape of the life span curve, preceding the decrease in survival of the male flies used in the study, and at least in the case of 16SrRNA, is tissue-specific. Although mitochondrial DNA remains unchanged in heads, thoraces, and abdomens, 16SrRNA levels decrease more severely in heads and thoraces and much less conspic- uously in abdomens. On the other hand, control non- mitochondrial transcripts investigated remain essentially unaffected. These results suggest that in Dro- sophila the main effect of aging on the mitochondrial genetic system is downstream from mtDNA itself. The decline in the levels of 0 H+-ATPase transcript, nuclear-encoded, suggests that not only the mitochondrial machinery, but also the nuclear one involved in mitochondrial biogenesis, is affected during aging.

Aging is an extremely complex process affecting all known biological organisms with the consequence of reducing their functional capacity, especially against injury and illness in a species-specific temporal pattern. An insufficiently character- ized interplay of genetic and environmental factors are thought to be involved in producing this complex response (1-3). Although some efforts have been carried out in clari-

* This work was supported by the Fondo de Investigaciones Sani- tarias de la Seguridad Social (FISss) Grants 89/0637 and 90/0262 and the Plan Nacional del Especio Grant 91/627. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a fellowship from the FISss. ll Present address: Centro de Investigaciones Biol6gicas (CSIC), c/

Velazquez 144, 28006 Madrid, Spain. 11 TO whom correspondence should be addressed: Departamento de

Bioquimica & Instituto de Investigaciones Biomkdicas (CSIC), Fa- cultad de Medicina U.A.M., c/Arzobispo Morcillo 4, Madrid 28029, Spain. Fax: 34-1-5854587.

fying the cellular and molecular alterations in aged animals, still relatively little is known about the changes at these basic levels which are produced during aging. Since originally proposed by Miquel (4, 5), the “mitochondrial mutation theory of aging” places one of the key sites of cellular damage within the mitochondria. Mitochondria consume about 90% of total oxygen and are a continuous source of oxygen radicals (6) which can damage, among other molecules, the mtDNA.’ An impairment of the mitochondrial function and the subsequent reduced supply of ATP could have far reaching pleiotropic effects at the cellular level. In accordance with these ideas, several theories have been proposed recently considering the mtDNA an important contributor to the aging phenomena. For instance, Ritcher (7) has suggested that damaged mtDNA could accumulate in senescent cells and that mitochondrial fragments escaping from these organelle could become inte- grated in the nuclear genome causing mutations in their insertion places. Linnane et al. (8) have suggested that accumulation of mtDNA mutations could also be one of the main causes of diseases associated with aging. In their view, mutated mtDNA would progressively accumulate from concep- tion to death of an organism. Its cellular segregation will produce a decline in the bioenergetic capacity of the affected tissues with the consequence of, among other syndromes, muscle weakness and neurodegenerative diseases frequently associated with senescence.

Several features make mtDNA especially susceptible to mutations: it is small, highly compacted, and unprotected with histones. DNA repair and recombination mechanisms are lacking in mitochondria (9). Consequently, mtDNA mutations in mammals accumulate at least 5-10 times faster than in single copy nuclear DNA (lo), and in man, several degenerative diseases affecting muscle, nervous system, or both have been directly associated with mutations in the mtDNA (11). Furthermore, the transfer of DNA between mitochondria and nucleus is well documented in several organisms (12), although only in lower eukaryotes a direct correlation between mtDNA instability and senescence has been demonstrated (13).

A link between aging and mitochondrial impairment has been recently proposed in humans, where an age-related decrease in mitochondrial enzyme activities in both muscle (14) and liver (15) have been found, and the number of skeletal and heart muscle fibers deficient in cytochrome c oxidase increases with age (16, 17). In addition, mtDNA deletions accumulate in old humans, mostly in energy-demanding tissues such as brain or muscle (18). However, the amount of

‘The abbreviations used are: mtDNA, mitochondrial DNA; 16SrRNA, 16 S ribosomal RNA; Cytb, cytochrome b; COI, cytochrome c oxidase subunit I; @-ATPase, mitochondrial H+-ATP synthase subunit @; kb, kilobase (8); bp, base pairs.

18891

18892 Drosophila Mitochondrial Gene Expression and Aging

deleted mtDNA is extremely low compared with wild-type &DNA (usually less than 0.1%), and therefore the physiological significance of the increase in the defective molecules is not clear. In vitro 'experiments using human po cells and quantitative analysis of mi~chondrial myopathies have shown that a considerable excess of mutated mtDNA is necessary for causing oxidative phosphorylation defects (19, 20). Therefore, additional mechanisms, nuclear or mitochondrial, have to be found altered to explain the decrease of mitochondrial function with age.

The progressive alteration of mitochondria with age has been also detected in other animals. In rodents, for example, there is also an increase of mutated mtDNA with age (21,22), and the rate of mtDNA transcription is reduced in old animals (23, 24). Among invertebrates, Drosophila melanogaster has been the organism more extensively used in gerontological research, and its validity in these studies has been widely documented (25). Many observations have shown that oxygen consumption is inversefy correlated with D. melanogaster life apans, and there are important morphological alterations in the mitochondria of aged flies (26-28). In an effort to under- stand how the mi~chondrial genetic system is affected during the aging process, we have evaluated the level and integrity of the mtDNA and the steady-state level of some mitochondrial transcripts during the life span of D. melunogaster.

~ P ~ I ~ E ~ A L PROCEDURES

Drosophila Cultures-Oregon R wild-type D. melarwgaster males were maintained at 25 "C in 25-ml tubes with Drosophila culture medium at an initial density of 20 flies/tube. The survival curve was determined by counting the number of flies still alive at 24-h intervals. Flies were transferred to new vials with fresh medium two times/ week. The composition of the medium (per liter) is as follows: 100 g of bakers' yeast, 8 g of brown sugar, 12.5 g of agar, 50 g of flour, and 5 ml of propionic acid.

Dmsophla Dissection-Flies stored at -70 "C were placed on a steel plate maintained on dry ice. Heads, thoraces, and abdomens were carefully dissected under a dissecting microscope, immediately frozen at -70 "C, and stored until used.

Enzyme Activities Determination-Flies were homogenized, 20 in a typical experiment, in 1 ml of buffer (10 mM Tris-HC1, pH 7.5, 20 mM EDTA, 0.25 M sucrose, 2.25 mM Gaelp) and the debris removed by low speed centrifugation at 500 X g during 2 min. The homogenate was centrifuged at 13,000 X g during 10 min, and the particulated fraction containing the mitochondria was resuspended in 100 pl of the homogenization buffer. Cytochrome c oxidase and glutamate dehydrogenase were assayed in this fraction and glucose phosphate isomerase in the supernatant fraction as described previously (29- 31). Protein content was measured by the procedure of Bradford (32).

Qunntification of mtDNA-Total DNA was prepared from male flies of different ages using a modification of the procedure of Binghan et al. (33). Typically, 20 males were homogeni~d in 1 ml of buffer (10 mM Tris-HC1, pH 7.5, 10 mM EDTA, 60 mM CaCla, 0.5% SDS) and the debris removed by low speed centrifugation (500 x g during 2 rnin). The homogenate was incubated with 10 pl of DNase-free RNase for 15 min at 37 "C and the DNA extracted once with pheno~chloroform (111). Different amounts of total DNA were blotted on a nylon membrane (Amemhan) using the Manifold I1 system (Schleicher & Schuell), and fixed on the filter following the manufacturer's indications. To quantify the mtDNA, the filters were probed at high stringency with [a-32P]dCTP labeled D. meknogaster-specific mitochon~ial DNA clones (see below), at 10' cpmlml, specific activity >lo7 cpmlpg. After several washes with 0.1 X SSC, 0.1% SDS at 65 "C, the filter was autoradiographed at room temperature for different intervals of time. Known amounts of purified Drosophila mtDNA (34) were also included in parallel as standard. After identi- fication of the optimal exposure time to be in the linear range of resDonse. the intensity of the hybridization signal was densitometri- caliy quantified.

Restriction Pattern Analysis of D. melanogoster mtDNA-For Southern experiments, 10 pg of total DNA were digested with the selected restriction enzyme (Boehringer) in the optimal conditions recommended by the ~ n u f a c t u r e r , electrophoresed on agarose gel,

and blotted on nylon membranes. The filters were probed, washed, and autoradiographed in the same conditions as described above for the slot blots.

Specific Mitochondrial Probes-Specific probes to detect the different ~ ~ h o n ~ ~ gene transcripts 16 S rRNA, cytochrome c oxidase subunit I (COI), and cytochrome b (Cytb) have been prepared from defined D. melarwgaster mitochondrial DNA clones. The 16SrRNA probe is a 618-bp EcoRI-RsaI fragment, position 7253-8141 (as numbered in Ref. 351, containing the 3' region of the 16SrRNA gene. The Cytb probe is an 854-bp SacI-PuuII fragment, positions 5400-6249 (35), containing most of the Cytb gene. The COI probe is a 970-bp EcoR V-Sac1 fragment, position 957-1927 (as numbered in Ref. 36) containing half of the tRNAC" gene, the tRNA- gene, and the 5'- half of the COI gene. The fragments were subcloned in Bluescript vector (Stratagene) using standard methods (37). The constructions were checked by sequencing of the clones. The @-A TP synthase pmbe (@-ATPase) is a cDNA clone that contains almost the complete coding sequence of the D. melarwgaster ATP synthase p subunit gene? The sequence has the accession number X71013.

RNA Extraction and Northern Analysis-The protocol described by Sambrook et ai. (37) was used as follows. 40 Dmsophla males were homogenized in 1.5 ml of 10 mM EDTA, pH 8, 0.5% SDS, the homogenizer washed with 2 ml of 0.1 M sodium acetate, pH 5.2, 10 mM EDTA, and the homogenate phenolized once. After the addition of 440 pl of 1 M Tris HCl, pH 8, and 180 pl of 5 M NaCl, the aqueous phase was precipitated with cold ethanol, the pellet resuspended in 200 p l of H20 (MilliQ purity), 600 p l of 4 M sodium acetate, pH 6, added and the sample kept at 4 "C overnight. Finally, the sample was centrifuged at 13,000 X g for 10 min, the pellet resuspended in 100 pl of HzO, and the final concentration of RNA was spectrophotometri- cally measured.

For Northern hybridizations, total RNA was electrophoresed on 1.2% agarose, 1.8 M formaldehyde gels, blotted on a nylon membrane, and probed with the [a-32P]dCTP-labeled mitochondrial probes, using the same conditions as described for the Southern experiments. After several washes in high stringency conditions, the fdters were autoradiographed during different intervals of time to identify the optimal conditions of the exposure. The intensity of the signals was then densitometrically quantified.

RESULTS

Mitochondrial Enzyme Activity during the Aging Response of Drosophila-The half-life of the D. melanogaster Oregon R strain in our culture conditions is 37 days, and the maximum life span is 55 days (Fig. 1). Males were selected in these studies to avoid the complications arising from the important ovarian content in mitochondria in females. As mentioned in the introduction, several reports have shown important age- related morphological and biochemical alterations in mitochondria (26-28) and a decrease in mit~hondrial proteins (38) during Drosophila aging. To confirm these observations in our hands and as a first approximation for evaluating the mitochondrial function during this process, we have measured in young and old flies, the activity of two m i ~ h o n ~ i a l enzymes: the electron transport chain enzyme cytochrome c oxidase, an inner membrane multisubunit complex, three of whose subunits are codified in the mitochondrial genome, and glutamate dehydrogenase, a soluble enzyme present in the mitochondrial matrix, encoded in the nuclear genome. Fig. 1 shows that there is a decrease in both activities at the end of the life spans of our Drosophila populations. Roughly, there is a 25% decrease in cytochrome c oxidase and glutamate dehydrogenase activities as detected in older animals, when compared with the activity in young animals. A similar result has also recently been described in humans (14,15), where a negative correlation between respiratory enzyme activities and age has been found. As a control of our observations and to rule out that they are part of a general decrease in cellular enzyme activities, we have also evaluated the activity of a cytosolic enzyme, glucose phosphate isomerase. As can be

P. Peiia and R. Garesse, submitted for publication.

Drosophila Mitochondrial Gene Expression and Aging 18893

80 -

5 4 0 -

20 - c . I . , . # . , . , . I . .

0 5 1 0 1 5 20 25 30 3 40 45 5 0 55 7 “ ’ . ‘ . ’

Age (in days)

Cytochrome c Oxidase Glutamate Dehydrogenase Glucose Phosphate Isomerase

200

150 1 0 20 30 40 50 60

500 1500 1 0 20 30 40 50 6 0 1 0 2 0 30 40 50 6 0

Age (in days) Age (in days) Age (in days)

FIG. 1. Survival curve and enzyme activity determinations. The survival curve of Oregon R wild-type D. mekanogaster males were determined as described under “Experimental Procedures.” Cytochrome c oxidase, glutamate dehydrogenase, and glucose phosphate isomerase were assayed as described (29-31). Each point represents the average of three determinations (mean f standard error). Data were statistically analyzed. Cytochrome c oxidase and glutamate dehydrogenase, p < 0.0005. Using a multiple comparison test (Fisher LSD), the differences among the values were significant (>95%, indicated as *) in all cases except in the comparison between the points corresponding to 15 and 24 days (cytochrome c oxidase) and 15,24, and 33 days (glutamate dehydrogenase). In the case of glucose phosphate isomerase, there are no significant differences ( p > 0.4).

appreciated in Fig. 1, its activity remains constant throughout the whole life period investigated, indicating that the decrease in cytochrome c oxidase and glutamate dehydrogenase activities in Drosophila flies is not just one component of a general decrease in enzymatic activities at the cellular level.

Levels and Integrity of Drosophila Mitochondrial DNA during the A g i n g Response-In accordance with the current ver- sion of the mitochondrial theory of aging, it is postulated that damaged mtDNA molecules should accumulate during the aging process. In our attempt to substantiate these ideas in Drosophila, the first observation was that the amount of total DNA recovered from flies sampled during the life span of Drosophila is the same at all ages (1 f 0.2 pglfly), indicating that generalized cellular death is not an important phenomenon occurring in flies of increasing age. Different amounts of total DNA (50, 250, and 500 ng) were slot-blotted and probed with each one of the four mitochondrial clones (H, C, B, or D), expanding most of the coding sequence of the Drosophila mtDNA as schematized in Fig. 2A. Several experiments have been carried out (covering the complete life spans of three different populations assayed two or more times each of them). In all experiments and using every one of the four mitochondrial clones, the same conclusion was obtained. The amount of mtDNA per pg of total DNA remains constant during the complete life span, representing 1% of total DNA (10 ng of mtDNA/pg of total DNA). One representative example of these experiments is shown in Fig. 2B.

Drosophila is a tiny animal, not easily subjected to experimental manipulation and isolation of individual tissues and organs in quantities sufficient for biochemical assays as per- formed in this work. Therefore, as an approximation to de-

termine if there is some tissue-specific variation in the level of mtDNA, the same kind of analysis was carried out in DNA extracted from heads, thoraces, and abdomens of young and old flies. In the head the nervous system and some sensory organs predominate, the thorax is rich in muscle fibers, especially the potent flight musculature, and in the abdomen a series of internal organs are present, namely, intestine, mal- pighian tubules, and generative organs. The results are shown in Fig. 2C. The same amount of mtDNA is essentially detected in the three samples. In the same experiment, total DNA extracted from whole flies of the same age was simultaneously measured. As expected from the previous results, no variation in mtDNA was again detected.

On the other hand and as indicated in the Introduction, even though the quantity of the mtDNA was maintained during aging, alterations in the mtDNA integrity could be involved in the aging response. To investigate this possibility, the mtDNA restriction pattern was studied in Southern blot experiments of DNA samples covering the whole life span of Drosophila. Fig. 3, A and B, show the results obtained using two restriction enzymes, EcoRI and HaeII1, that cut infre- quently the D. melamgaster mtDNA. The pattern with both enzymes is the expected one (35, 36): lo-, 5.4-, 1.8-, and 0.4- kb fragments with EcoRI and 8.8-, 6-, and 3.5-kb fragments with HaeIII. No detectable age-associated change was found. Even after very long exposures of the autoradiographs (data not shown), no additional fragments that could indicate the occurrence of insertion/deletions or reorganizations in the mtDNA during this process were detected. Fig. 3C shows a similar analysis using the enzyme AZuI that cuts very frequently in the D. melanogaster mtDNA. The complexity of


A

D n C B

4-e l k b

C H T

=.

I 1 ng rnt-DNA - 5 ng mt-DNA

0 10 ng mt-DNA

TOTAL

0 6 Days

0 41 Days

FIG. 2. Quantification of mitochondrial DNA. A shows a sche- matic representation of the D. melanogaster mitochondrial genome. The gene order is presented using standard abbreviations for the names of the genes. The vertical bars indicate the position of restriction enzymes HindIII, HaeIII, and EcoRI. The location within the molecule of the clones H, C, B, and D used in slot blot, and Southern experiments are indicated by wide horizontal bars. They cover most of the coding sequence, and a more detailed description of their origin and sequence can be found elsewhere (35, 36). B, different amounts (50, 250, and 500 ng) of total DNA obtained from males of the age indicated were blotted on a nylon membrane and hybridized with [a- 32P]dCTP-labeled B clone as described under “Experimental Proce- dures.” As standard, 1,5, and 10 ng of highly purified D. melanogaster mtDNA were also included. C, 15 Drosophila males were dissected, equivalent amounts of total DNA extracted from heads (H), thoraces (T), and abdomens ( A ) and slot-blotted as described in “Experimental procedures”. The filter was hybridized and washed using the same conditions described in B. As a control, total DNA extracted from flies of the same ages was also included.

the pattern of low molecular weight bands fits well with the result expected from the Drosophila mitochondrial sequence (35,36), and it is again the same in all samples independently of the age of the flies.

We have extended the analysis using other restriction enzymes (MspI, HpaII, and HindIII; data not shown), again with no variations in the restriction pattern at different ages of the flies. Thus, these experiments allow us to conclude that the great majority of mitochondrial molecules remains unaltered during Drosophila aging, and therefore, an accumulation of them was not likely involved in the aging process itself.

Steady-state Level of Mitochondrial Transcripts during Dro- sophila Aging-The possibility remained that the alterations were at downstream steps of the mitochondrial gene expression machinery. As a first approximation to evaluate the effect of senescence on mtDNA transcription, we have quantified

the steady-state levels of three mitochondrial transcripts encoded in the mitochondrial genome, 16SrRNA, Cytb, and COI, and one mitochondrial transcript encoded in the nuclear genome, the B-ATPase. Equivalent amounts of total RNA extracted from flies of different ages were analyzed in North- ern experiments using specific probes prepared as described under “Experimental Procedures.’’ Several experiments carried out in different populations of flies have shown that the total amount of RNA extracted per fly was the same at all ages. In accordance with this finding, the main component of the RNA, the cytoplasmic ribosomal RNA, remains unaltered during the Drosophila life span. Contrary to this, the steady- state level of mitochondrial transcripts significantly decreases in old animals. Fig. 4A shows the variation observed in the steady-state level of three transcripts, 16SrRNA, Cytb, and &ATPase. A continuous decline in the amount of these transcripts can be clearly appreciated. Densitometric analysis of several Northern experiments using these probes have shown that the more dramatic decrease is detected at the level of 16SrRNA. Only 20% of the amount present in young flies remains in older animals. For Cytb and @-ATPase transcripts the decline is consistent in all experiments but more variable, ranging from 30 to 60%. The reason of this variability is at present unknown. A similar experiment analyzed with a COI- specific probe is shown in Fig. 4B, and again a marked decrease is observed in older animals.

In addition to the main RNA component, the cytoplasmic rRNA, the cytoplasmic actin messenger RNA has been quantified as a second control during the aging response of Dro- sophila. Fig. 4, A and B, show that in this case, its steady- state level also remains constant, reinforcing the conclusion that the decrease in mitochondrial transcripts is singular and does not reflect a generalized decrease in the level of cellular RNA during aging.

A final point was to investigate if the decrease detected in mitochondrial transcripts is tissue-specific. For the same rea- sons as we have argued above, we have determined the level of 16SrRNA in total RNA extracted from heads, thoraces, and abdomens of young and old flies. We have selected this transcript because it is expressed at the highest level among mitochondrial transcripts, and as has been shown above, shows the more important decrease during Drosophila life span. The result is shown in Fig. 5. Interestingly, although the decrease is dramatic in heads and thoraces, it is more moderated in abdomens. In parallel, and as a control of loading, the same filter was hybridized with a cytoplasmic ribosomal probe. Densitometric analysis reveals that the ratio 16SrRNA/cytoplasmic 18SrRNA decreases 3.5 times in heads, more than 5 times in thoraces and only 1.5 times in abdomens. Therefore, although the decline in the steady-state level of the 16SrRNA transcripts is general in the fly, it is quantitatively more important in highly energetic and func- tionally essential tissues such as brain and muscle.

DISCUSSION

Numerous studies carried out in several organisms have shown the presence of morphological alterations of senescent mitochondria (26-28) and a continuous decline in the bioenergetic capacity with age, especially in the most energy- demanding tissues (14, 15, 20). In this paper we present evidence corroborating that in Drosophila there is also an age- dependent decrease in the mitochondrial function, as measured using two mitochondrial enzymes as markers, cytochrome c oxidase and glutamate dehydrogenase. Although the decrease is moderate (25%)) its physiological significance should be relevant. For example in humans a similar reduction

Drosophila Mitochondrial Gene Expression and Aging 18895

A

10 Kb+ 5.4 Kb+

1.8 Kb+

0.4 Kb+

Age (in days) B

Age (in days)

1 1 6 2 0 3 0 3 4 41 1 1 6 2030 34 41 -

6 Kb -+ 3 . 5 K b -+ m-m

c Age (in days)

1 1 6 2 0 3 0 3 4 4 1 “0”-

0.5 Kb+

0.1 Kb+

FIG. 3. Restriction pattern analysis of D. melanogaeter mtDNA. For Southern experiments, 10 pg of total DNA were digested with the selected restriction enzyme in the optimal conditions recommended by the manufacturer, electrophoresed on an agarose gel, blotted to a nylon membrane, and hybridized in the same conditions described for the slot blots. The figure shows the results obtained with three enzymes: EcoRI ( A ) , Hue111 ( E ) , and Alu I (C).

A 5 15 25 35 40 50 Days

“ . _ . . . . 16s rRNA

*LIY”-

* .. .. ’. Cy1 b

1% H*ATPase

c 4 w u $& d

B Days

15 24 33 40 51

w w m b . COI

Actin

FIG. 4. Steady-state levels of mitochondrial transcripts. A, total RNA (5 pg) prepared from flies of the selected age were electrophoresed overnight on 1.2% agarose-formaldehyde gel and blotted on a nylon membrane. 32P-Labeled probes were specific for 16 S rRNA, Cytb, @ATPase (see “Experimental Procedures”) and actin 5C (46). E shows a similar experiment probed with 32P-labeled COI and actin 5C clones. The hybridization and washing conditions were the same described under “Experimental Procedures.” The lower lane in both panels shows the fluorescence of the cytoplasmic rRNA band.

in mitochondrial activity has been reported to be responsible, at least in part, of the late onset of appearance of symptoms observed in several neurodegenerative diseases caused by mutations in mtDNA (11,20).

To investigate the role that the mitochondrial genetic system may play in the loss of mitochondrial function, we have initially quantified the amount of mtDNA during the life span of Drosophila. Its level remains constant at all ages, representing roughly 1% of total DNA, a percentage similar to that found in somatic cells of a variety of organisms. As deduced from experiments carried out with DNA extracted from head, thorax, and abdomen of young and old flies, the amount of mtDNA does not seem to change in specific tissues. Further- more, restriction enzyme analysis shows that the integrity of

Days 2 51 2 51

H

T

A

16s rRNA 18s rRNA FIG. 5. Steady-state levels of 16SrRNA in heads, thoraces,

and abdomens of Drosophila adults. Five Drosophila males were dissected, and the total RNA extracted from heads, thoraces, and abdomens was slot-blotted on a nylon membrane. The filter was hybridized and washed with the 32P-labeled 16SrRNA probe as described under “Experimental Procedures” and hybridized again in the same conditions with a 32P-labeled 18 S rRNA probe.

the mitochondrial genome remains unaltered during the aging process. Although with this kind of analysis is not possible to rule out the accumulation of point mutations or the presence of a very low levels of altered molecules, the data strongly suggest that the great majority of the mitochondrial genome remains unaltered during Drosophila aging. Interestingly, a recent report on the behavior of the mitochondrial DNA during cell apoptosis (39) also indicates that in this process leading to the cellular death no mitochondrial DNA fragmen- tation occurs, although it involves severe nuclear DNA deg- radation. Furthermore, since no differences were found in the restriction fragments of Drosophila mtDNA produced by two sets of enzymes, MspI and HpaII differentially affected by methylation of the bases in their restriction sites suggest that no changes in methylation occurs during the aging response, although this possibility was very unlikely, since no methylation has been detected yet in arthropod DNA (40, 41).

The amount of mtDNA during the aging process has been estimated previously by other authors in Oregon R and Swed- ish-C D. melamgmter strains using equilibrium density centrifugation (42, 43). Significant decreases in the amount of mtDNA were reported in these experiments, although a direct implication of this decrease in aging could not be established


(44). Using a more direct approach as described here, we have been unable to reproduce these findings. Interestingly, a recent study (45) has demonstrated that although the amount of rat liver mtDNA does not change during aging, there are changes in its buoyant density. In younger animals the mtDNA density is homogeneous, but in older ones a hetero- geneous pattern is observed, probably due to the presence of proteins bound to the mtDNA of old animals. The possibility remains that the same phenomenon could be occurring during Drosophila aging, thus explaining the discrepancy between our results and those reported previously (42-44).

In senescent rats evidence showing that there is a decrease in the steady-state levels of mitochondrial transcripts, tissue- specific due to a reduced rate in the mtDNA transcription, has been reported (23, 24). We have explored this possibility in Drosophila quantifying the amount of several mitochondrial transcripts during its whole life span. Interestingly, we have found that this is indeed the case. There is a continuous decline in the levels of the measured mitochondrial transcripts, three encoded in the mitochondrial genome, lGSrRNA, Cytb, COI, and one encoded in the nuclear genome, the &ATPase. The fact that the levels of cytoplasmic rRNA and actin mRNA do not change, indicates that the effect is not general, although whether this decrease affects all mitochondrial transcripts remains to be directly proved. The similar response of all mitochondrial transcripts studied, inde- pendent of the genome encoding them, probably is a consequence of the necessary coordination between the two genomes, nuclear and mitochondrial, in the biogenesis of mitochondria. This idea is reinforced by the finding that not only cytochrome c oxidase activity, which contains subunits codified in the mitochondrial genome, but glutamate dehydrogenase, exclusively encoded in the nucleus, decreases during the aging of Drosophila. It is important to point out that the decrease of the mitochondrial transcript investigated occurs in advance to the decrease in the survival curve of the Dro- sophila population used in our experiments, a fact that suggests a possible cause-effect relationship between both proc- esses. The availability of several strains of Drosophila selected for different life spans (3) could strengthen the significance of this correlation, if the decrease in their mitochondrial transcripts levels also precedes and matches their survival curves. We are currently addressing this issue in our labora- tory, as well as whether transcription itself or RNA stability is the mechanism responsible of the decrease in the steady- state levels of mitochondrial transcripts.

The main effect we have detected in this study is an important reduction in the levels of lGSrRNA, that in old animals is only 20% of that present in young ones. This change should affect the efficiency of the mitochondrial trans- lation machinery, with the consequence of reducing the synthesis of all polypeptides within the mitochondria. The rate of protein synthesis in isolated organelles from D. melanogaster has been measured previously (38) and matches closely our own measurements of mtrRNA levels. Protein synthesis decreases about 25% during the first half of the adult life, reaching 40% in old animals. The results presented here on the changes in 16SrRNA levels during the life span of Dro- sophila explains this continuous reduction in the rate of mitochondrial protein synthesis.

Even more interesting is the finding that the decrease, at least at the level of lGSrRNA, is tissue-specific, with an important reduction of the amount present in highly energy- demanding tissues such as brain and muscle, being more moderate in the abdomen organs. A similar tissue specificity has been observed in mammals, suggesting that the effect of

aging on mitochondrial function should be universal throughout the animal kingdom. In humans, the loss of oxidative phosphorylation capacity observed in senescence has been related to alterations in mtDNA. As discussed in the introduction the amount of deleted mtDNA is extremely low, although it is possible that it affects mitochondria in specific cells with important consequences for the physiology of the tissue. The data presented here show that in Drosophila the main effect of aging on mitochondria is on the steady-state level of mitochondrial transcripts, a similar situation as has been described in rats (23,24). As suggested by Cantatore and colleagues (22), this result is better explained by an age- dependent mitochondrial metabolic alteration than by mitochondrial DNA mutations, but the precise mechanism remains to be investigated. Drosophila due to its excellent manipulability, molecular and genetical, and short life span provides a suitable experimental model system to explore, in- depth, the mechanism and significance of these age-related changes.

Acknowledgments-We thank J. Miquel for helpful comments during this work and J. Miquel and M. Quintanilla for the critical reading of earlier versions of the manuscript. The invaluable help of Antonio Fernandez in the preparation of figures is fully acknowledged. We also thank Dr. Cruces for the generous gift of the Drosophila cytoplasmic rRNA clone.

REFERENCES 1. 2.

3.

5. 4.

6.

8. 7.

10. 9.

11. 12. 13.

15. 14.

16.

17. 18.

19.

20. 21.

22.

23.

24.

25.

26. 27.

28.

29.

30.

31.

32. 33. 34.

35.

Medawar, P. (1952) An Unsolved Problem of Bwlo H. K. Lewis, London Finch, C. E. (1990) Lowevih: Senescence and the%rwme, The University

of Chicago Press, Chibago Rose, M. R. (1991) Euolutwnaty Biology of Aging, Oxford University Press,

Nlrm Ynrk Miquel, J., and Fleming, J. (1984) J. ET. GerontoL 19 , 31-36 Mi el, J., and Fleming, J. (1986) Mech. Ageing Res. 8. 51-74 Lezo, S. A., Lorentzen, R. J., and Ts’o, P. 0. P. (1980) Biochemistry 19,

_.-.. ””

Ritcher C. (1988) FEBS Lett. 241 , 1-5 Linnan;, A. W., Marzuki, S., Ozawa, T., and Tanaka, M. (1989) Lancet i,

3023-3028

Brown, W. M. (1981) Ann. N. Y. Acad. Sci. 361 , 119-132 Brown, W. M., George, M. J. R., and Wilson, A. C. (1979) Proc. Natl. Acad.

Wallace, D. C. (1992) Science 266,628-632 Gellisen, G., and Michaelis, G. (1987) Ann. N. Y. Acad. Sci. 603,391-401 Turker, M. S., Domenico, J. D., and Cummings, D. J. (1987) J. BioL C k m .

Trounce, I., Byme, E., and Marzuki, S. (1989) Lancet i, 637-639 Yen,. T. C., Chen, Y. S., King, K. L., Yeh, S. H., and Wei, Y. H.0989)

Miiller-H&ker. J., Schneiderbanger, K., Stefani, F. H., and Kadenbach, B.

Byme, E., and Dennett, X. (1992) Mutotion Res. 275,125-131 Cortopassi, G. A.! Shibata, D., Soong, N. W., and Amheim, N. (1992) Pmc.

Hayashi J. I., Ohta, S., Kikuchi, A., Takemitau, M., Goto, Y. I., and

Wallace, D. C. (1992) Annu. Rev. Biochem. 61,1175-1212 Pik6,L., Hougham, A. J., and Bullpit, K. J. (1988) Mech. Ageing Dev. 4 3 ,

Gadaleta, M. N., Rainaldi, G., Lezza, A. M. S., Milella, F., Fracaso, F., and

Gadaleta, M. N., Petruzzella, V., Renis, M., Fracasso, F., and Cantatore, P. Cantatore, P. (1992) Mutation Res. 276,181-193

Fernandez-Silva, P., Petruzella, V., Fracasso, F., Gadaleta, M. N., and (1990) Eur. J. Brochem 187,501-506

Baker G. T. 111 Jacobson M and Mo skl, G. (1985) in H d h k of Cantatore, P. (1991) Biochem. Bmphys. Res. Commun. 176,645-653

CedBioloh a h Aging (Cri&falo, U y e d ; pp. 511-578, CRC Press, Boca Raton, FL

642-645

Sei. U. S. A. 76,1976-1971

262,2250-2255

Blochem. Bmphys. Res. Commun. 166,994-1003

(1992) Mutation Res. 276,115-124

NatL Acad. Sa. U. S. A. 8 9 , 7370-7374

Non&a, I. (1991) Proc. NatL Acad. Sc i U. S .A. 88,10614-10618

279-293

Vann, A. C., and Webster, G. C. (1977) Exp. Gerontol. 12 , l -5 Miquel, J., Pro,, J., Bensch, K. G., and Johnson, J. E., Jr. (1977). in Free

Radzc& m Bwlogy (Pryor, W., ed) Vol. 3, pp. 133-182, Academtc Press, New York

Takahashi, A., Philpott, D. E., and Miquel, J. (1970) J. Gerontol. 2 6 , 222- 228

Vallejo, C. G., Sillero, M. A. G., and Marco, R. (1979) Cell. MOL Biol. 26 ,

Valle’o, C. G , Marco, R., and Sebastih, J. (1970) Eur. J. Biochem. 14 ,

Kramer, J. J., Henslee, J. G., Weabayashi, Y., and Jones, M. E. (1983) in Methods of Enzymtrc Analysw (Ber eyer H. U Be eyer J., and

Bradford, M. M. (1976) Anal. Biochem. 72,248-254 GraKl, M., e&) pp. 200-210, V e r l a g - ~ m i e , ’ W e i n ~ e e i m , ~ ; l y

Binghan P. M., Levis, R., and Rubm, G. M. (1981) Cell 26.693-704 Bultman;, H., and Laird, C. D. (1973) Biochim. Biophys. Acta 299, 196-

Garesse, R. (1988) Geneties 118,649-663

113-124

47$-485 ’

209

Drosophila Mitochondrial Gene Expression and Aging 18897 36. de Bmijn, M. H. L. (1983) Nature 304,234-241 37. Sambrook, J., Fritsch, F., and Maniatis, T. (1989) Molecular Cloning; A 42. Massie,’H. R.,’Baird, P B . , &d McMahon, M. M. (1975) Gerontobgiu 21,

‘Old "ring Harbor ‘Old "ring Harbor’ 43. MFsie,-H. R., Colacicco, J. R., and Willians, T. R. (1981) Age (Omaha) 4,

41. Warner A. H. and B haw J. C. (1984) Deu. EioL 102,264-267

231-238 NY

mitochondrial dna remains intact during drosophila aging ... · these changes correlate with the...

Documents