- Email: [email protected]
Age-associated changes in mitochondrial mRNA expression and translation in the Wistar rat heart
Mechanisms of Ageing and Development 103 (1998) 179 – 193
Age-associated changes in mitochondrial mRNA expression and translation in the Wistar rat heart Edgar K. Hudson *, Naotaka Tsuchiya, Richard G. Hansford The Laboratory of Molecular Genetics, Gerontology Research Center, National Institute on Aging, 4940 Eastern A6enue, Baltimore, MD 21224, USA Received 23 July 1997; received in revised form 9 January 1998; accepted 25 February 1998
Abstract The purpose of this research is to determine possible causes and mechanisms involved in the age-associated decline in mitochondrial activity. We have focused on cytochrome c oxidase because it is comprised of both nuclear and mitochondrial-encoded subunits and may provide some insight into the coordination of the two genomes. In agreement with previous reports, we show an approximate 30% decrease in cardiac cytochrome c oxidase activity at 24 months compared to 6 months with no change in the activity of the nuclear encoded citrate synthase of the mitochondrial matrix. The rate of mitochondrial protein synthesis as shown by [35S]methionine incorporation decreased approximately 35% in the 24-month-old rat compared to the 6-month-old rat. The decrease in protein synthesis was associated with a 30–50% reduction in the levels of most individually radiolabeled translation products including the COX subunits and specifically, a 23% decrease in COX1 protein steady-state levels according to Western analysis. Similarly, there was a decrease in the mRNA steady-state levels of both nuclear and mitochondrial-encoded subunits of cytochrome c oxidase. These results suggest that a number of different mechanisms are involved in the age-associated decrease in heart mitochondrial activity and these are discussed. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mitochondrial protein synthesis; Cytochrome c oxidase; Mitochondrial gene expression; Aging; Heart
* Corresponding author. Tel.: +1 410 5588635; fax: + 1 410 5588157. 0047-6374/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0047-6374(98)00043-8
180
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
1. Introduction The production of ATP for cellular homeostasis is the primary function of the mitochondria. The electron transport chain or respiratory chain, located within the mitochondrial inner membrane, generates a proton gradient across the inner membrane that supplies the energy to drive the synthesis of ATP by the ATP synthase in the process known as oxidative phosphorylation. The essential elements of oxidative phosphorylation are the four protein complexes of the respiratory chain and the ATP synthase complex, all located within the inner mitochondrial membrane. In the respiratory chain, complex I is synonymous with the activity of NADH dehydrogenase, complex II with succinate dehydrogenase activity, complex III with cytochrome c reductase activity, and complex IV with cytochrome c oxidase activity. A unique feature of the mitochondrion is the presence of its own genome with the capacity to synthesize 13 polypeptides, all of which are subunits of four of the respiratory chain proteins or the ATP synthase (complex V). These polypeptides account for about 18% of the total number of subunits in the respiratory chain and ATP synthase including seven subunits of the NADH-dehydrogenase (ND1, ND2, ND3, ND4, ND4L, ND5, ND6), cytochrome b which forms part of cytochrome c reductase (CYT b), two subunits of the ATP synthase (ATP6, ATP8) and three subunits of cytochrome c oxidase (COX I, COX II, COX III). The circular DNA, approximately 16 kb in size, also encodes the 22 tRNAs and two rRNAs required for proper translation (Anderson et al., 1981). The mitochondrial genome is not self-sufficient however and is reliant on a number of nuclear-encoded enzymes for proper mitochondrial DNA replication, transcription, translation and function (Attardi, 1988; Clayton, 1991). In addition, a large number of nuclear-encoded proteins are required for mitochondrial activity while others work in conjunction with the mitochondrial-encoded subunits for the proper function of the complexes mentioned above (Poyton and McEwen, 1996). The control of mitochondrial biogenesis is complicated involving coordination of the mitochondrial and nuclear genomes (Attardi, 1988; Clayton, 1991; Poyton and McEwen, 1996). The complex interaction between the genomes and the intricacy of the system may predispose the system to errors and malfunctions that may play a role in a number of pathological states or aging. Alterations in mitochondrial DNA and changes in regulation by cytosolic factors have become the focus of several investigations looking at the role of mitochondria in the aging process and disease. The work from the laboratories of Wallace, Morgan-Hughes and DiMauro has shown that mutations in mitochondrial DNA are associated with a number of diseases (Holt et al., 1988; Moraes et al., 1992; Wallace et al., 1992). The accumulation of mitochondrial DNA mutations in post-mitotic cells such as myocytes and neurons may contribute to age-associated diseases such as Alzheimer’s (Davis et al., 1997). Another area of great interest is the role of free radical damage in the aging process and disease, specifically with
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
181
regard to the mitochondrial generation of these free radicals and their damage either to the mitochondrion itself or other cell constituents (Harman, 1993). Several reports have shown age-associated decreases in mitochondrial function and some have shown specific reductions in the activities of proteins containing mitochondrial-encoded subunits (McMillin et al., 1993; Castelluccio et al., 1994). However, most of these reports have failed to demonstrate a possible mechanism responsible for the decline in mitochondrial activity. Several reports have shown decreases in cytochrome c oxidase (complex IV) activity, including one from this laboratory, using a variety of methods and techniques (Hansford and Castro, 1982; Yamada et al., 1995). The functional importance of cytochrome oxidase and its role as a marker for mitochondrial activity and content has resulted in the development of procedures for quantitation of both its message and activity. In this report, we have confirmed the age-associated decrease in cytochrome c oxidase in the rat heart and have investigated mechanisms for this decline. We find that mitochondrial protein synthesis is diminished to a similar degree and that there are reductions in steady-state mitochondrial mRNA levels. While concentrating on the mitochondrial genome, the possibility exists that differences in nuclear-encoded protein and/or subunit expression and activity could be responsible for these changes. To this end, we also examined the activity of nuclear-encoded citrate synthase, a marker for the mitochondrial matrix, and the mRNA expression of the nuclear-encoded COX subunit IV.
2. Materials and methods
2.1. Mitochondrial preparation Mitochondria were prepared from hearts of 6- and 24-month-old male Wistar rats from the Gerontology Research Center colony by high-speed differential centrifugation and a short Nagarse digestion as reported by Hansford (1978).
2.2. Measurement of O2 consumption Oxygen partial pressure was measured at 25°C with a Clark-type oxygen electrode in a water-jacketed chamber as described by Hansford (1978).
2.3. Cytochrome c oxidase acti6ity Cytochrome c oxidase activity was measured at 25°C using a Clark-type oxygen electrode and a 2-ml system comprising 0.05 M potassium phosphate (pH 7.4), 40 mM cytochrome c, 12.5 mM ascorbate, 0.63 mM N,N,N%,N%-tetramethyl-pphenylenediamine, 0.5 mg freshly isolated mitochondria and 0.05% (v/v) Triton X-100 (Hansford and Castro, 1982).
182
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
2.4. Citrate synthase assay Citrate synthase activity was measured spectrophotometrically by thiol release at 412 nm as described by Hansford and Castro (1982).
2.5. Mitochondrial protein synthesis Mitochondrial translation was measured by the incorporation of 35S-labeled methionine as described by McKee et al. (1990) with the following modifications. The mitochondria were incubated at 37°C in a shaking water bath at 75 rpm. It was necessary to oxygenate the tube containing the reaction solution with 100% oxygen for 30 s. The tube was capped with a rubber stopper, followed by oxygenation of the tube through the rubber stopper for an additional minute. After addition of isotope or removal of an aliquot, the oxygenation step was repeated. Aliquots were removed at 30 and 60 min and added to 1 ml cold 10% trichloroacetic acid (TCA), vortexed for 30 s, then centrifuged for 4 min. The supernatant was removed and the pellet washed twice with 1 ml 10% TCA. The final pellet was taken up in 200 ml of 0.4 M NaOH, then left at room temperature for 1–1.5 h. Then 100 ml was used to quantitate radioactivity by liquid scintillation counting in 10 ml Aquasol scintillation cocktail using a Packard 2500 TR counter.
2.6. Mitochondrial translation products Upon completion of the protein synthesis assay, the remaining sample was diluted with an equal volume of 0.1 M cold methionine and incubated at room temperature for 15 min. The mix was transferred to a microcentrifuge tube and the labeled mitochondria were pelleted by centrifugation (12000× g, 4 min). The pellet was resuspended and washed with 0.6 M mannitol, 1 mM EDTA, and 5 mM methionine and then centrifuged for 4 min. The pellet was solubilized in 200 ml electrophoresis buffer (15% glycerol, 2% SDS, 1% b-mercaptoethanol, 10 mM Na phosphate, pH 7.0, 1 mM Na2EDTA, and 0.1% bromophenol blue) and stored frozen until needed, then subjected to electrophoresis over a 12% SDS-PAGE, dried onto a filter paper support and stored for 1 week at − 80°C with a phosphoimage screen (McKee et al., 1990). The bands were visualized using the Molecular Dynamics Phosphoimager after storage and the band volumes were quantitated using FragmeNT Analysis™ software (Molecular Dynamics) and identified according to the work of Polosa and Attardi (1991).
2.7. mRNA isolation and Northern blot analysis mRNA was isolated from frozen rat heart ventricle using the STAT-60 RNA isolation system (TEL-TEST, Friendswood, TX) based on the guanidinium isothiocyanate/phenol extraction technique of Chomczynski and Sacchi (1987). Total RNA and purity was quantitated spectrophotometrically using the absorbance ratio at 260/280. For the quantitation of mitochondrial-encoded mRNA levels, 10 mg
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
183
total RNA was subjected to electrophoresis in a 1.2% agarose gel containing formaldehyde while 30 mg was used for quantitation of nuclear-encoded mRNA. Fractionated RNA was transferred to a Sure-blot™ nylon hybridization membrane (Oncor) by capillary action for 24 h in 20× SSC (3 M NaCl, 0.3 M sodium citrate, pH 7.0). Once transferred, the RNA was cross-linked to the filter by ultraviolet irradiation and prepared for hybridization. The membrane was prehybridized with Church buffer (1% BSA, 7% SDS, 500 mM sodium phosphate, pH 7.0 and 1mM sodium EDTA) at 55°C for 30 min and then hybridized in Church buffer at 55°C overnight. After hybridization, the membrane was washed twice for 15 min at 55°C. The blots were then exposed to Kodak X-OMAT autoradiography film for 30 min to 6 h at −80°C. Once developed, the autoradiographs were scanned using a Molecular Dynamics Personal Densitometer, and the bands were quantitated using the FragmeNT Analysis™ software (Molecular Dynamics).
2.8. Preparation of cDNA probes The COX I and COX II cDNA were derived from rat striated muscle clones (gift of Dr Christopher Moyes) and COX IV (gift of Dr Christopher Moyes) was amplified by PCR from mouse liver cDNA. The cDNA probes were radiolabeled by random priming (Multiprime DNA labeling system, RPN1601Z; Amersham) to a specific radioactivity of 109 cpm/mg with [a-32P]dCTP (Amersham International).
2.9. Western blot analysis Standard methods of electrophoresis and Western blotting were used (Towbin et al., 1979). Thirty micrograms of isolated rat heart mitochondria were diluted 1:4 with sample buffer (62.5 mM Tris–HCl, pH 6.8, 10% glycerol, 2% SDS, 5% b-mercaptoethanol, 0.001% bromophenol blue) and then subjected to electrophoresis over a 15% SDS-polyacrylamide gel. Afterwards, the proteins were transferred overnight (4°C) to polyvinylidene difluoride (PVDF). The blots were blocked with 5% dry milk, washed, then incubated with the appropriate dilution of antibody. COX I and IV antibodies were purchased from Molecular Probes. The secondary antibodies were conjugated with alkaline phosphatase allowing visualization of the signal by the conversion of a substrate to a fluorescent product (Vistra ECF). The products were visualized using a Molecular Dynamics Fluorimager and quantitated using the Molecular Dynamics FragmeNT Analysis™ software package.
2.10. Statistical analysis of the data Data are presented as the mean9standard error (S.E.). Unless stated otherwise, significance was determined using Student’s two tailed t-test with differences of PB 0.05 considered significant.
184
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
3. Results
3.1. Mitochondrial enzymatic acti6ity A number of reports have examined the effects of age on mitochondrial oxygen consumption and in agreement with previous reports from this laboratory and others, there were no age-linked differences in either the State 3 or State 4 respiration rates using glutamate and pyruvate/malate as substrates or respiratory control ratios in the 6- and 24-month-old rat heart mitochondria (Table 1). There was, however, an approximate 30% decrease in the cytochrome c oxidase activity of the aged rat compared to the young (PB 0.03, Table 1) when measured as a function of protein content of isolated mitochondria. This is in accordance with the published results from this laboratory and others (Hansford and Castro, 1982; Yamada et al., 1995). By contrast, citrate synthase failed to show an age-associated change in activity in isolated rat heart mitochondria (Table 1). While three of 13 subunits of cytochrome c oxidase are encoded by the mitochondrial genome, citrate synthase is a product solely of the nuclear genome. Therefore, further studies were directed at determining the role of the mitochondrial genome in the decline of activity with age and/or age-associated changes in mitochondrial genetic regulation.
Table 1 Rates of O2 consumption, cytochrome c oxidase and citrate synthase of mitochondria isolated from the hearts of 6- and 24-month-old rats 6-month
24-month
O2 consumption (ng atoms/min/mg protein) Glutamate State 3 State 4 Pyruvate/malate State 3 State 4
165.28 9 10.88 25.64 9 1.61
146.20 98.78 25.28 92.09
256.71 9 12.26 30.09 9 2.11
244.01 98.67 29.13 92.19
Enzyme activity Cytochrome c oxidase (mg atoms/min/mg) Citrate synthase (mmol/min/mg)
4.816 9 0.236 4.853 9 0.543
3.461 9 0.527‡ 5.654 9 0.441
O2-uptake was measured at 30°C with either glutamate (5mM) or pyruvate (2.5 mM) plus malate (1 mM) as substrate. State 3 or active respiration was measured in the presence of 0.5 mM ADP and state 4 was the ‘resting’ respiration rate that occurred when ADP was depleted. There was no significant difference in the function of the young and old mitochondria under these assay conditions (n =7 for 6 month, n= 9 for 24 month). Cytochrome c oxidase activity decreased significantly with age (‡PB0.03, n = 12 for 6 month, n=8 for 24 month). Citrate synthase showed no age-associated decrease in activity (n =6 for both age groups). Statistical significance was evaluated using Student’s t-test for unpaired data.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
185
Fig. 1. Age-associated changes in mitochondrial protein synthesis. Heart mitochondria were isolated from 6- and 24-month-old rats, and the rate of protein synthesis was measured as described in Section 2. The rate was calculated as the slope of [35S]methionine incorporation from 0 to 60 min. Mitochondrial protein synthesis decreased by approximately 30% with age (*PB0.01, n = 10 for 6-month, n = 7 for 24-month). Statistical significance was evaluated using Student’s t-test for unpaired data.
3.2. Mitochondrial protein synthesis The next step was to quantitate mitochondrial translation rates and to examine translation products. Mitochondrial protein synthesis was assayed in the presence of cycloheximide to inhibit cytosolic protein synthesis, and it was sensitive to the mitochondrial protein synthesis inhibitor, chloramphenicol (data not shown) using isolated rat heart mitochondria. In the 24-month-old rat (aged), the rate of mitochondrial protein synthesis decreased approximately 35% compared to the 6-month-old animal (Fig. 1, P B0.01). The reduced rate of translation in the aging rat heart was reflected by the decrease in mitochondrial translation products (Fig. 2A). When old and young translation products were compared, the young rats showed greater incorporation of label. When quantitated and standardized to a common band of approximately 40 kDa in matched lanes on the coomassie stained gel, the incorporation of [35S]methionine into the mitochondrial translated proteins of the old rat heart was significantly reduced in most proteins (Fig. 2B). Some proteins failed to show a significant decrease in signal with age due to a lower degree of resolution of those bands but all showed a reduction in incorporated label. Since the mitochondrial-encoded subunits of cytochrome oxidase comprise the catalytic core of the enzyme, the decrease in mitochondrial protein synthesis is likely to contribute to the decline in cytochrome c oxidase activity. A possible
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
Fig. 2.
186
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
187
Fig. 3. Western blot analysis of cytochrome c oxidase subunits. Western blot analysis was performed on isolated mitochondria following electrophoresis on a 15% polyacrylamide-SDS gel. (A) COX I protein content decreased with age while COX IV content was unchanged. (B) To eliminate blot-to-blot variation, the 24-month- and 6-month-old were grouped in each blot (i.e., in quadruplicate), and statistical significance was evaluated by using Student’s t-test for paired samples. Mitochondrial-encoded COX I protein content decreased approximately 25% (PB 0.02, n =6 blots) while nuclear-encoded COX IV content did not change (n = 4 blots).
reason for the decrease in mitochondrial translation could be a decrease in mitochondrial mRNA, tRNA or rRNA levels or altered activity of translational factors.
3.3. Le6els of nuclear- and mitochondrial-encoded COX subunit protein Western blot analysis showed a significant decrease in COX I protein (PB 0.02) and unchanged levels of COX IV with age (Fig. 3A). For comparison, the numerical values were grouped according to age in each blot, then the means compared by a paired Student’s t-test (Fig. 3B). This was done to eliminate blot-to-blot variation. Six blots were compared for COX I and four blots compared
Fig. 2. (See left) Mitochondrial translation products. Following the protein synthesis assay, 35S-labeled proteins were collected by centrifugation, then suspended in electrophoresis buffer (see Section 2). (A) The samples were subjected to electrophoresis and then used for autoradiography. This autoradiograph represents one of four gels. Bands were identified according to the work of Polosa and Attardi (1991). In all cases, the signal was reduced in the 24-month-old rat. (B) The signals were quantitated using the Imagequant software package and then standardized to a band of approximately 40 kDa from the coomassie stained gels (determined by densitometry). Six samples were compared for each age group using four different gels. Statistical significance was evaluated using Student’s t-test for unpaired data. †P B 0.03, ‡P B0.01, *PB 0.04, PB 0.02, PB 0.007.
188
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
for COX IV. These results support the mitochondrial protein synthesis data which shows a reduction in overall protein synthesis rate as well as reduced production of the COX I subunit.
3.4. Nuclear and mitochondrial mRNA le6els Steady-state levels of mitochondrial mRNA were quantitated to determine if decreased mRNA levels were associated with the decrease in mitochondrial translation. Total mRNA was isolated from heart tissue of 6- and 24-month-old rats and subjected to Northern blot analysis using cDNA probes of cytochrome c oxidase subunits from both genomes. Message levels of the mitochondrial-encoded cytochrome c oxidase subunit I (COX I) were decreased in the aged rat heart (Fig. 4A). Similarly, nuclear-encoded cytochrome c oxidase subunit IV (COX IV) message levels were decreased in the aged rat heart (Fig. 4A). However, message levels of mitochondrial-encoded cytochrome c oxidase subunit 2 (COX II) and nuclear-encoded citrate synthase failed to change with age. These results were confirmed when the message levels were standardized to 18s ribosomal RNA (Fig. 4B). With decreases in the expression of COX I and COX IV and the unchanged expression of COX II, the likelihood that the mitochondrial genome is down-regulated with age in its entirety is reduced.
4. Discussion This report is in agreement with a number of reports which have shown age-associated decreases in cytochrome c oxidase activity (Hansford and Castro, 1982; Yamada et al., 1995). The regulation of cytochrome c oxidase activity is
Fig. 4. Northern blot analysis of cytochrome c oxidase subunits I, II and IV. Total RNA was isolated from 6- and 24-month-old rat hearts according to Section 2. For COX I and COX II, 10 mg RNA was loaded in each lane while 30 mg was used for COX IV. (A) Northern blot analysis showed decreased expression of COX I and IV and elevated COX II expression. (B) Signals were quantitated using FragmeNT Analysis™ software and standardized to the 18s rRNA signal (see Section 2). As seen in the blot, both COX I and IV levels decreased with age, whereas COX II increased with age. Due to biological variation, these changes did not reach significant levels. The case was the same for COX II which increased with age. A large number of samples (16 – 21) and a number of different gels (3 – 4) were used for these experiments.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
189
currently unknown and confounded by the fact that the function of most of the 10 nuclear-encoded subunits of cytochrome c oxidase has yet to be defined. The complexity of the mitochondrial genetic expression system makes it unlikely that a single event is responsible for the age-associated decline in mitochondrial activity shown here and elsewhere. Indeed, several different factors have been identified in the age-linked decrease in cytochrome c oxidase activity. Hayakawa et al. have shown that the decrease in the aging rat heart correlates with a decrease in mitochondrial DNA content and an increase in 8-hydroxydeoxyguanosine, an oxidative product of deoxyguanosine (Hayakawa et al., 1993). They suggest that decreased cytochrome c oxidase activity is the manifestation of an overall deterioration of the mitochondrial genome. Abu-Erreish and Sanadi reported a decrease in cytochrome c oxidase activity that paralleled a decrease in cytochrome content and suggested that the age-associated decrease in cytochrome content may be due to either changes in the rate of cytochrome synthesis or degradation (Abu-Erreish and Sanadi, 1978). Paradies et al. have shown in the hypothyroid rat heart a similar decrease in cytochrome c oxidase activity and cytochrome content but contrary to Abu-Erreish and Sanadi suggested that the decline in cytochrome content could not be entirely responsible for the decrease in cytochrome c oxidase activity. They demonstrated that the decrease in cytochrome c oxidase activity corresponded to a decrease in membrane cardiolipin content in the hypothyroid rat compared to normal and thyroid hormone treated rats. They suggested that alterations in the phospholipid environment of the inner membrane may also play a role in the regulation of cytochrome c oxidase activity (Paradies et al., 1993). When this finding is taken in conjunction with the reports of McMillin et al. which showed a decrease in cardiolipin content with age and, more recently, that of Paradies et al. which restored cytochrome c oxidase activity with the addition of exogenous cardiolipin (Paradies et al., 1997), the possibility arises that alterations in membrane composition may contribute to the regulation of cytochrome c oxidase with age. The current report demonstrates that reduced mitochondrial protein synthesis and gene expression contribute to the decline in cytochrome c oxidase activity in the aging rat heart. Since mitochondrial protein synthesis is responsible for the production of the three catalytic subunits of cytochrome c oxidase (Kim et al., 1995), changes in translational activity with age may play a major role in the decrease in cytochrome c oxidase activity. In this report, the decrease in the rate of translation and in the level of products synthesized corresponds well with a decrease in steady-state levels of COX I protein. Although the mitochondrion generates the mRNA, rRNA and tRNA required for protein synthesis, cytosolic proteins, such as initiation and elongation factors, control mitochondrial protein synthesis (Liao and Spremulli, 1991; Barker et al., 1993). Mitochondrial protein synthesis has been shown to increase in both the insulin perfused rat heart (McKee and Grier, 1990) and in the heart of thyroxine treated rats (Leung and McKee, 1990). These hormones may stimulate cytosolic synthesis of nuclear-encoded proteins necessary for mitochondrial translation. Since it has been reported that the action of these hormones is reduced with age (Mooradian and Wong, 1994; Muller et al., 1996), the possibility
190
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
exists that reduced hormone action with age may result in lowered expression of mitochondrial translation factors, hence decreased protein synthesis. Recent work from a number of laboratories has identified, purified and cloned a few of these translation factors but no work has been done to elucidate their role in aging (Liao and Spremulli, 1991; Barker et al., 1993). In addition to a majority of the subunits of Complexes I, III, IV and V, the remaining mitochondrial enzymes are all nuclear-encoded. Changes in their expression and synthesis will have dramatic effects on mitochondrial function. Ostronoff et al. have shown that inhibition of cytosolic protein synthesis with cycloheximide reduced cytochrome c oxidase activity and mitochondrial protein synthesis in liver (Ostronoff et al., 1996). Therefore, in addition to the requirement of nuclear-encoded proteins for proper mitochondrial function, they suggest that mitochondrial protein synthesis occurs at a point when synthesis of nuclear-encoded subunits increases profoundly, suggesting that some interaction between the two systems exists. However, the possibility remains that inhibition of cytosolic protein synthesis would also reduce synthesis of factors needed for mitochondrial translation. In addition to the situations discussed above, the current report shows a decrease in mitochondrial translation associated with a decrease in the levels of some mitochondrial mRNA. It does not rule out the possibility that there is a disturbance in the interaction of the two genomes or a decrease in translational factor content or action, but it presents an additional mechanism whereby reduced transcript levels may eventually alter protein synthesis. The reason for the decrease in COX I expression with age, but lack of change in COX II expression, is currently unknown. Although it has been shown that the mitochondrial genome is polycistronic with a single transcript synthesized by each strand, there are situations when the genes may be differentially expressed. Iglesias et al. have reported a decrease in mitochondrial ND3 expression in hypothyroid rats and a thyroid hormone receptor binding site in the mitochondrial ND3 gene suggesting direct transcriptional regulation of ND3 by thyroid hormone (Iglesias et al., 1995). Mouse cells treated with interferon showed a decrease in cyt b and ND5 mRNA expression while ND6 expression remained unchanged (Shan et al., 1990). Cycloheximide blocked the capacity of interferon to reduce mitochondrial gene expression suggesting the existence of an interferon-responsive nuclear gene capable of regulating mitochondrial gene expression. A recent report from Davis et al. showed an increased incidence of point mutations in the COX I and COX II genes in Alzheimer’s patients associated with a decrease in cytochrome c oxidase activity. In their report, 50% of the Alzheimer’s patients screened showed a mutation in the COX I gene while 30% had a mutation in COX II. This raises the possibility that COX I may be specifically altered with age resulting in its decreased expression and eventual impairment of the function of cytochrome c oxidase. Possible deletions of segments of the mitochondrial genome are perhaps the most striking situation whereby COX I and COX II may be differentially expressed. It has been shown that a number of diseases and aging are associated with deletions in mitochondrial DNA (Szabolcs et al., 1994; Melov et al., 1997). The accumulation of mitochondrial DNA mutations in non-mitotic cells such as neurons may contribute to a number
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
191
of age-associated diseases such as Alzheimer’s (Davis et al., 1997). All of these situations demonstrate different mechanisms whereby mitochondrial RNA may be differentially expressed. Another potential mechanism for the regulation of mitochondrial gene expression may be through the altered expression and/or action of the mitochondrial transcription factor-1 (MTF-1) and the nuclear respiratory factors-1 and -2 (NRF-1 and NRF-2). MTF-1 is a nuclear-encoded DNA-binding protein involved in both mitochondrial DNA replication and transcription through its interaction with the mt-DNA D-loop (Fisher and Clayton, 1985). In turn, MTF-1 is regulated by NRF-1 and NRF-2 (Virbasius and Scarpulla, 1994). A decrease in their expression would result in decreased mitochondrial gene expression. Interestingly, NRF-1 binds the promoter and activates the COX VIc (Evans and Scarpulla, 1990) subunit while NRF-2 activates expression of the COX IV gene (Virbasius and Scarpulla, 1991). Therefore, the corresponding decrease in COX I and COX IV expression may be the result of altered NRF-2 expression or action. In any event, NRF-1 and NRF-2 are capable of affecting both mitochondrial and nuclear encoded subunits of cytochrome c oxidase and may act as signals between the two genomes. Several different phenomena appear to be involved in the age-associated decline in cytochrome c oxidase activity in the Wistar rat heart. It is highly unlikely that a single event or disturbance leads to this decline. This report shows that mitochondrial protein synthesis and the expression of some, but not all, of the mitochondrial genome declines with age. Whether this is a consequence of free radical damage to the genome or altered transcriptional regulation is not known and the subject of current research.
References Abu-Erreish, G.M., Sanadi, D.R., 1978. Age-related changes in cytochrome concentration of myocardial mitochondria. Mech. Ageing Dev. 7, 425 – 432. Anderson, S., Bankier, A.T., Barrell, B.G., de Bruijn, M.H.L., Coulson, A.R., Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Sanger, F., Schreier, P.H., Smith, A.J.H., Staden, R., Young, I.G., 1981. Sequence and organization of the human mitochondrial genome. Nature 290, 457 – 465. Attardi, G., 1988. Biogenesis of mitochondria. Annu. Rev. Cell Biol. 4, 289 – 333. Barker, C., Markris, A., Patiotis, C., Bear, S.E., Tsichlis, P.N., 1993. Identification of the gene encoding the mitochondrial elongation factor G in mammals. Nucleic Acids Res. 21, 2641 – 2647. Clayton, D.A., 1991. Replication and transcription of vertebrate mitochondrial DNA. Annu. Rev. Cell Biol. 7, 453–478. Castelluccio, C., Baracca, A., Fato, R., Pallotti, F., Maranesi, M., Barzanti, V., Gorini, A., Villa, R.F., Castelli, G.P., Marchetti, M., Lanaz, G., 1994. Mitochondrial activities of rat heart during ageing. Mech. Ageing Dev. 76, 73–88. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156 – 159. Davis, R.E., Miller, S., Herrnstadt, C., Ghosh, S.S., Fahy, E., Shinobu, L.A., Galasko, D., Thal, L.J., Beal, M.F., Howell, N., Davis Parker, W. Jr., 1997. Mutations in mitochondrial cytochrome c oxidase genes segregate with late-onset Alzheimer disease. Proc. Natl. Acad. Sci. USA 92, 4526 – 4531.
192
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
Evans, M.J., Scarpulla, R.C., 1990. NRF-1: a trans-activator of nuclear-encoded respiratory genes in animal cells. Genes Dev. 4, 1023–1034. Fisher, R.P., Clayton, D.A., 1985. A transcription factor required for promoter recognition by human mitochondrial RNA polymerase. J. Biol. Chem. 260, 11330 – 11338. Hansford, R.G., 1978. Lipid oxidation by heart mitochondria from young adult and senescent rats. Biochem. J. 170, 285–295. Hansford, R.G., Castro, F., 1982. Age-linked changes in the activity of enzymes of the tricarboxylate cycle and lipid oxidation, and of carnitine content, in muscles of the rat. Mech. Ageing Dev. 19, 191–201. Harman, D., 1993. Free radical theory of aging: a hypothesis of senile dementia of the Alzheimer’s type. Age 16, 23–30. Hayakawa, M., Sugiyama, S., Hattori, K., Takasawa, M., Ozawa, T., 1993. Age-associated damage in mitochondrial DNA in human hearts. Mol. Cell. Biochem. 119, 95 – 103. Holt, I.J., Harding, A.E., Morgan-Hughes, J.A., 1988. Deletions of mitochondrial DNA in patients with mitochondrial myopathies. Nature 331, 717 – 719. Iglesias, T., Caubı´n, J., Zaballos, A., Bernal, J., Mun˜oz, A., 1995. Identification of the mitochondrial NADH dehydrogenase subunit 3 (ND3) as a thyroid hormone regulated gene by whole genome PCR analysis. Biochem. Biophys. Res. Commun. 210, 995 – 1000. Kim, K., Lecordier, A., Bowman, L.H., 1995. Both nuclear and mitochondrial cytochrome c oxidase mRNA levels increase dramatically during mouse postnatal development. Biochem. J. 306, 352 – 358. Leung, A.C., McKee, E.E., 1990. Mitochondrial protein synthesis during thyroxine-induced cardiac hypertrophy. Am. J. Physiol. 258, E511 – E518. Liao, H.X., Spremulli, L.L., 1991. Initiation of protein synthesis in animal mitochondria. J. Biol. Chem. 266, 20714–20719. Melov, S., Hinerfeld, D., Esposito, L., Wallace, D.C., 1997. Multi-organ characterization of mitochondrial genomic rearrangements in ad libitum and caloric restricted mice shows striking somatic mitochondrial DNA rearrangements with age. Nucleic Acids Res. 25, 974 – 982. McKee, E.E., Grier, B.L., 1990. Insulin stimulates mitochondrial protein synthesis and respiration in isolated perfused rat heart. Am. J. Physiol. 259, E413 – E421. McKee, E.E., Grier, B.L., Thompson, G.S., McCourt, J.D., 1990. Isolation and incubation conditions to study mitochondrial protein synthesis. Am. J. Physiol. 258, E492 – E510. McMillin, J.B., Taffet, G.E., Taegtmeyer, H., Hudson, E.K., Tate, C.A., 1993. Mitochondrial metabolism and substrate competition in the aging Fisher rat heart. Cardiovascular Res. 27, 2222–2228. Mooradian, A.D., Wong, N.C.W., 1994. Age-related changes in thyroid hormone action. Eur. J. Endocrinol. 131, 451–461. Moraes, C.T., Ricci, E., Petruzzella, V., Shanske, S., DiMauro, S., Schon, E.A., Bonilla, E., 1992. Molecular analysis of the muscle pathology associated with mitochondrial DNA deletions. Nat. Genet. 1, 359–367. Muller, D.C., Elahi, D., Tobin, J.D., Andres, R., 1996. The effect of age on insulin resistance and secretion: a review. Semin. Nephrol. 16, 289 – 298. Ostronoff, L.K., Izquierdo, J.M., Montoya, J.A., Cuezva, J.M., 1996. Transient activation of mitochondrial translation regulates the expression of the mitochondrial genome during mammalian mitochondrial differentiation. Biochem. J. 316, 183 – 191. Paradies, G., Ruggiero, F.M., Dinoi, P., Petrosillo, G., Quagliariello, E., 1993. Decreased cytochrome oxidase activity and changes in phospholipids in heart mitochondria from hypothyroid rats. Arch. Biochem. Biophys. 307, 91–95. Paradies, G., Ruggiero, F.M., Petrosillo, G., Quagliariello, E., 1997. Age-dependent decline in the cytochrome c oxidase activity in rat heart mitochondria: role of cardiolipin. FEBS Lett. 406, 136–138. Polosa, P.L., Attardi, G., 1991. Distinctive pattern and translational control of mitochondrial protein synthesis in rat brain synaptic endings. J. Biol. Chem. 266, 10011 – 10017. Poyton, R.O., McEwen, J.E., 1996. Crosstalk between nuclear and mitochondrial genomes. Annu. Rev. Biochem. 65, 563–607.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
193
Shan, B., Vasquez, E., Lewis, J.A., 1990. Interferon selectively inhibits the expression of mitochondrial genes: a novel pathway for interferon-mediated responses. EMBO J. 9, 4307 – 4314. Szabolcs, M.J., Seigle, R., Shanske, S., Bonilla, E., DiMauro, S., D’Agati, V., 1994. Mitochondrial DNA deletion: a cause of chronic tubulointerstitial nephropathy. Kidney Int. 45, 1388 – 1396. Towbin, H., Staehlin, T., Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350–4354. Wallace, D.C., Lott, M.T., Shoffner, J.M., Brown, M.D., 1992. Diseases resulting from mitochondrial DNA point mutations. J. Inher. Metab. Dis. 15, 472 – 479. Yamada, K., Sugiyama, S., Kosaka, K., Hayakawa, M., Ozawa, T., 1995. Early appearance of age-associated deterioration in mitochondrial function of diaphragm and heart in rats treated with doxorubicin. Exp. Gerontol. 30, 581– 593. Virbasius, J.V., Scarpulla, R.C., 1991. Transcriptional activation through ETS domain binding sites in the cytochrome c oxidase subunit IV gene. Mol. Cell. Biol. 11, 5631 – 5638. Virbasius, J.V., Scarpulla, R.C., 1994. Activation of the human mitochondrial transcription factor A gene by nuclear respiratory factors: a potential regulatory link between nuclear and mitochondrial gene expression in organelle biogenesis. Proc. Natl. Acad. Sci. USA 91, 1309 – 1313.
.
Age-associated changes in mitochondrial mRNA expression and translation in the Wistar rat heart Edgar K. Hudson *, Naotaka Tsuchiya, Richard G. Hansford The Laboratory of Molecular Genetics, Gerontology Research Center, National Institute on Aging, 4940 Eastern A6enue, Baltimore, MD 21224, USA Received 23 July 1997; received in revised form 9 January 1998; accepted 25 February 1998
Abstract The purpose of this research is to determine possible causes and mechanisms involved in the age-associated decline in mitochondrial activity. We have focused on cytochrome c oxidase because it is comprised of both nuclear and mitochondrial-encoded subunits and may provide some insight into the coordination of the two genomes. In agreement with previous reports, we show an approximate 30% decrease in cardiac cytochrome c oxidase activity at 24 months compared to 6 months with no change in the activity of the nuclear encoded citrate synthase of the mitochondrial matrix. The rate of mitochondrial protein synthesis as shown by [35S]methionine incorporation decreased approximately 35% in the 24-month-old rat compared to the 6-month-old rat. The decrease in protein synthesis was associated with a 30–50% reduction in the levels of most individually radiolabeled translation products including the COX subunits and specifically, a 23% decrease in COX1 protein steady-state levels according to Western analysis. Similarly, there was a decrease in the mRNA steady-state levels of both nuclear and mitochondrial-encoded subunits of cytochrome c oxidase. These results suggest that a number of different mechanisms are involved in the age-associated decrease in heart mitochondrial activity and these are discussed. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Mitochondrial protein synthesis; Cytochrome c oxidase; Mitochondrial gene expression; Aging; Heart
* Corresponding author. Tel.: +1 410 5588635; fax: + 1 410 5588157. 0047-6374/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0047-6374(98)00043-8
180
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
1. Introduction The production of ATP for cellular homeostasis is the primary function of the mitochondria. The electron transport chain or respiratory chain, located within the mitochondrial inner membrane, generates a proton gradient across the inner membrane that supplies the energy to drive the synthesis of ATP by the ATP synthase in the process known as oxidative phosphorylation. The essential elements of oxidative phosphorylation are the four protein complexes of the respiratory chain and the ATP synthase complex, all located within the inner mitochondrial membrane. In the respiratory chain, complex I is synonymous with the activity of NADH dehydrogenase, complex II with succinate dehydrogenase activity, complex III with cytochrome c reductase activity, and complex IV with cytochrome c oxidase activity. A unique feature of the mitochondrion is the presence of its own genome with the capacity to synthesize 13 polypeptides, all of which are subunits of four of the respiratory chain proteins or the ATP synthase (complex V). These polypeptides account for about 18% of the total number of subunits in the respiratory chain and ATP synthase including seven subunits of the NADH-dehydrogenase (ND1, ND2, ND3, ND4, ND4L, ND5, ND6), cytochrome b which forms part of cytochrome c reductase (CYT b), two subunits of the ATP synthase (ATP6, ATP8) and three subunits of cytochrome c oxidase (COX I, COX II, COX III). The circular DNA, approximately 16 kb in size, also encodes the 22 tRNAs and two rRNAs required for proper translation (Anderson et al., 1981). The mitochondrial genome is not self-sufficient however and is reliant on a number of nuclear-encoded enzymes for proper mitochondrial DNA replication, transcription, translation and function (Attardi, 1988; Clayton, 1991). In addition, a large number of nuclear-encoded proteins are required for mitochondrial activity while others work in conjunction with the mitochondrial-encoded subunits for the proper function of the complexes mentioned above (Poyton and McEwen, 1996). The control of mitochondrial biogenesis is complicated involving coordination of the mitochondrial and nuclear genomes (Attardi, 1988; Clayton, 1991; Poyton and McEwen, 1996). The complex interaction between the genomes and the intricacy of the system may predispose the system to errors and malfunctions that may play a role in a number of pathological states or aging. Alterations in mitochondrial DNA and changes in regulation by cytosolic factors have become the focus of several investigations looking at the role of mitochondria in the aging process and disease. The work from the laboratories of Wallace, Morgan-Hughes and DiMauro has shown that mutations in mitochondrial DNA are associated with a number of diseases (Holt et al., 1988; Moraes et al., 1992; Wallace et al., 1992). The accumulation of mitochondrial DNA mutations in post-mitotic cells such as myocytes and neurons may contribute to age-associated diseases such as Alzheimer’s (Davis et al., 1997). Another area of great interest is the role of free radical damage in the aging process and disease, specifically with
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
181
regard to the mitochondrial generation of these free radicals and their damage either to the mitochondrion itself or other cell constituents (Harman, 1993). Several reports have shown age-associated decreases in mitochondrial function and some have shown specific reductions in the activities of proteins containing mitochondrial-encoded subunits (McMillin et al., 1993; Castelluccio et al., 1994). However, most of these reports have failed to demonstrate a possible mechanism responsible for the decline in mitochondrial activity. Several reports have shown decreases in cytochrome c oxidase (complex IV) activity, including one from this laboratory, using a variety of methods and techniques (Hansford and Castro, 1982; Yamada et al., 1995). The functional importance of cytochrome oxidase and its role as a marker for mitochondrial activity and content has resulted in the development of procedures for quantitation of both its message and activity. In this report, we have confirmed the age-associated decrease in cytochrome c oxidase in the rat heart and have investigated mechanisms for this decline. We find that mitochondrial protein synthesis is diminished to a similar degree and that there are reductions in steady-state mitochondrial mRNA levels. While concentrating on the mitochondrial genome, the possibility exists that differences in nuclear-encoded protein and/or subunit expression and activity could be responsible for these changes. To this end, we also examined the activity of nuclear-encoded citrate synthase, a marker for the mitochondrial matrix, and the mRNA expression of the nuclear-encoded COX subunit IV.
2. Materials and methods
2.1. Mitochondrial preparation Mitochondria were prepared from hearts of 6- and 24-month-old male Wistar rats from the Gerontology Research Center colony by high-speed differential centrifugation and a short Nagarse digestion as reported by Hansford (1978).
2.2. Measurement of O2 consumption Oxygen partial pressure was measured at 25°C with a Clark-type oxygen electrode in a water-jacketed chamber as described by Hansford (1978).
2.3. Cytochrome c oxidase acti6ity Cytochrome c oxidase activity was measured at 25°C using a Clark-type oxygen electrode and a 2-ml system comprising 0.05 M potassium phosphate (pH 7.4), 40 mM cytochrome c, 12.5 mM ascorbate, 0.63 mM N,N,N%,N%-tetramethyl-pphenylenediamine, 0.5 mg freshly isolated mitochondria and 0.05% (v/v) Triton X-100 (Hansford and Castro, 1982).
182
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
2.4. Citrate synthase assay Citrate synthase activity was measured spectrophotometrically by thiol release at 412 nm as described by Hansford and Castro (1982).
2.5. Mitochondrial protein synthesis Mitochondrial translation was measured by the incorporation of 35S-labeled methionine as described by McKee et al. (1990) with the following modifications. The mitochondria were incubated at 37°C in a shaking water bath at 75 rpm. It was necessary to oxygenate the tube containing the reaction solution with 100% oxygen for 30 s. The tube was capped with a rubber stopper, followed by oxygenation of the tube through the rubber stopper for an additional minute. After addition of isotope or removal of an aliquot, the oxygenation step was repeated. Aliquots were removed at 30 and 60 min and added to 1 ml cold 10% trichloroacetic acid (TCA), vortexed for 30 s, then centrifuged for 4 min. The supernatant was removed and the pellet washed twice with 1 ml 10% TCA. The final pellet was taken up in 200 ml of 0.4 M NaOH, then left at room temperature for 1–1.5 h. Then 100 ml was used to quantitate radioactivity by liquid scintillation counting in 10 ml Aquasol scintillation cocktail using a Packard 2500 TR counter.
2.6. Mitochondrial translation products Upon completion of the protein synthesis assay, the remaining sample was diluted with an equal volume of 0.1 M cold methionine and incubated at room temperature for 15 min. The mix was transferred to a microcentrifuge tube and the labeled mitochondria were pelleted by centrifugation (12000× g, 4 min). The pellet was resuspended and washed with 0.6 M mannitol, 1 mM EDTA, and 5 mM methionine and then centrifuged for 4 min. The pellet was solubilized in 200 ml electrophoresis buffer (15% glycerol, 2% SDS, 1% b-mercaptoethanol, 10 mM Na phosphate, pH 7.0, 1 mM Na2EDTA, and 0.1% bromophenol blue) and stored frozen until needed, then subjected to electrophoresis over a 12% SDS-PAGE, dried onto a filter paper support and stored for 1 week at − 80°C with a phosphoimage screen (McKee et al., 1990). The bands were visualized using the Molecular Dynamics Phosphoimager after storage and the band volumes were quantitated using FragmeNT Analysis™ software (Molecular Dynamics) and identified according to the work of Polosa and Attardi (1991).
2.7. mRNA isolation and Northern blot analysis mRNA was isolated from frozen rat heart ventricle using the STAT-60 RNA isolation system (TEL-TEST, Friendswood, TX) based on the guanidinium isothiocyanate/phenol extraction technique of Chomczynski and Sacchi (1987). Total RNA and purity was quantitated spectrophotometrically using the absorbance ratio at 260/280. For the quantitation of mitochondrial-encoded mRNA levels, 10 mg
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
183
total RNA was subjected to electrophoresis in a 1.2% agarose gel containing formaldehyde while 30 mg was used for quantitation of nuclear-encoded mRNA. Fractionated RNA was transferred to a Sure-blot™ nylon hybridization membrane (Oncor) by capillary action for 24 h in 20× SSC (3 M NaCl, 0.3 M sodium citrate, pH 7.0). Once transferred, the RNA was cross-linked to the filter by ultraviolet irradiation and prepared for hybridization. The membrane was prehybridized with Church buffer (1% BSA, 7% SDS, 500 mM sodium phosphate, pH 7.0 and 1mM sodium EDTA) at 55°C for 30 min and then hybridized in Church buffer at 55°C overnight. After hybridization, the membrane was washed twice for 15 min at 55°C. The blots were then exposed to Kodak X-OMAT autoradiography film for 30 min to 6 h at −80°C. Once developed, the autoradiographs were scanned using a Molecular Dynamics Personal Densitometer, and the bands were quantitated using the FragmeNT Analysis™ software (Molecular Dynamics).
2.8. Preparation of cDNA probes The COX I and COX II cDNA were derived from rat striated muscle clones (gift of Dr Christopher Moyes) and COX IV (gift of Dr Christopher Moyes) was amplified by PCR from mouse liver cDNA. The cDNA probes were radiolabeled by random priming (Multiprime DNA labeling system, RPN1601Z; Amersham) to a specific radioactivity of 109 cpm/mg with [a-32P]dCTP (Amersham International).
2.9. Western blot analysis Standard methods of electrophoresis and Western blotting were used (Towbin et al., 1979). Thirty micrograms of isolated rat heart mitochondria were diluted 1:4 with sample buffer (62.5 mM Tris–HCl, pH 6.8, 10% glycerol, 2% SDS, 5% b-mercaptoethanol, 0.001% bromophenol blue) and then subjected to electrophoresis over a 15% SDS-polyacrylamide gel. Afterwards, the proteins were transferred overnight (4°C) to polyvinylidene difluoride (PVDF). The blots were blocked with 5% dry milk, washed, then incubated with the appropriate dilution of antibody. COX I and IV antibodies were purchased from Molecular Probes. The secondary antibodies were conjugated with alkaline phosphatase allowing visualization of the signal by the conversion of a substrate to a fluorescent product (Vistra ECF). The products were visualized using a Molecular Dynamics Fluorimager and quantitated using the Molecular Dynamics FragmeNT Analysis™ software package.
2.10. Statistical analysis of the data Data are presented as the mean9standard error (S.E.). Unless stated otherwise, significance was determined using Student’s two tailed t-test with differences of PB 0.05 considered significant.
184
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
3. Results
3.1. Mitochondrial enzymatic acti6ity A number of reports have examined the effects of age on mitochondrial oxygen consumption and in agreement with previous reports from this laboratory and others, there were no age-linked differences in either the State 3 or State 4 respiration rates using glutamate and pyruvate/malate as substrates or respiratory control ratios in the 6- and 24-month-old rat heart mitochondria (Table 1). There was, however, an approximate 30% decrease in the cytochrome c oxidase activity of the aged rat compared to the young (PB 0.03, Table 1) when measured as a function of protein content of isolated mitochondria. This is in accordance with the published results from this laboratory and others (Hansford and Castro, 1982; Yamada et al., 1995). By contrast, citrate synthase failed to show an age-associated change in activity in isolated rat heart mitochondria (Table 1). While three of 13 subunits of cytochrome c oxidase are encoded by the mitochondrial genome, citrate synthase is a product solely of the nuclear genome. Therefore, further studies were directed at determining the role of the mitochondrial genome in the decline of activity with age and/or age-associated changes in mitochondrial genetic regulation.
Table 1 Rates of O2 consumption, cytochrome c oxidase and citrate synthase of mitochondria isolated from the hearts of 6- and 24-month-old rats 6-month
24-month
O2 consumption (ng atoms/min/mg protein) Glutamate State 3 State 4 Pyruvate/malate State 3 State 4
165.28 9 10.88 25.64 9 1.61
146.20 98.78 25.28 92.09
256.71 9 12.26 30.09 9 2.11
244.01 98.67 29.13 92.19
Enzyme activity Cytochrome c oxidase (mg atoms/min/mg) Citrate synthase (mmol/min/mg)
4.816 9 0.236 4.853 9 0.543
3.461 9 0.527‡ 5.654 9 0.441
O2-uptake was measured at 30°C with either glutamate (5mM) or pyruvate (2.5 mM) plus malate (1 mM) as substrate. State 3 or active respiration was measured in the presence of 0.5 mM ADP and state 4 was the ‘resting’ respiration rate that occurred when ADP was depleted. There was no significant difference in the function of the young and old mitochondria under these assay conditions (n =7 for 6 month, n= 9 for 24 month). Cytochrome c oxidase activity decreased significantly with age (‡PB0.03, n = 12 for 6 month, n=8 for 24 month). Citrate synthase showed no age-associated decrease in activity (n =6 for both age groups). Statistical significance was evaluated using Student’s t-test for unpaired data.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
185
Fig. 1. Age-associated changes in mitochondrial protein synthesis. Heart mitochondria were isolated from 6- and 24-month-old rats, and the rate of protein synthesis was measured as described in Section 2. The rate was calculated as the slope of [35S]methionine incorporation from 0 to 60 min. Mitochondrial protein synthesis decreased by approximately 30% with age (*PB0.01, n = 10 for 6-month, n = 7 for 24-month). Statistical significance was evaluated using Student’s t-test for unpaired data.
3.2. Mitochondrial protein synthesis The next step was to quantitate mitochondrial translation rates and to examine translation products. Mitochondrial protein synthesis was assayed in the presence of cycloheximide to inhibit cytosolic protein synthesis, and it was sensitive to the mitochondrial protein synthesis inhibitor, chloramphenicol (data not shown) using isolated rat heart mitochondria. In the 24-month-old rat (aged), the rate of mitochondrial protein synthesis decreased approximately 35% compared to the 6-month-old animal (Fig. 1, P B0.01). The reduced rate of translation in the aging rat heart was reflected by the decrease in mitochondrial translation products (Fig. 2A). When old and young translation products were compared, the young rats showed greater incorporation of label. When quantitated and standardized to a common band of approximately 40 kDa in matched lanes on the coomassie stained gel, the incorporation of [35S]methionine into the mitochondrial translated proteins of the old rat heart was significantly reduced in most proteins (Fig. 2B). Some proteins failed to show a significant decrease in signal with age due to a lower degree of resolution of those bands but all showed a reduction in incorporated label. Since the mitochondrial-encoded subunits of cytochrome oxidase comprise the catalytic core of the enzyme, the decrease in mitochondrial protein synthesis is likely to contribute to the decline in cytochrome c oxidase activity. A possible
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
Fig. 2.
186
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
187
Fig. 3. Western blot analysis of cytochrome c oxidase subunits. Western blot analysis was performed on isolated mitochondria following electrophoresis on a 15% polyacrylamide-SDS gel. (A) COX I protein content decreased with age while COX IV content was unchanged. (B) To eliminate blot-to-blot variation, the 24-month- and 6-month-old were grouped in each blot (i.e., in quadruplicate), and statistical significance was evaluated by using Student’s t-test for paired samples. Mitochondrial-encoded COX I protein content decreased approximately 25% (PB 0.02, n =6 blots) while nuclear-encoded COX IV content did not change (n = 4 blots).
reason for the decrease in mitochondrial translation could be a decrease in mitochondrial mRNA, tRNA or rRNA levels or altered activity of translational factors.
3.3. Le6els of nuclear- and mitochondrial-encoded COX subunit protein Western blot analysis showed a significant decrease in COX I protein (PB 0.02) and unchanged levels of COX IV with age (Fig. 3A). For comparison, the numerical values were grouped according to age in each blot, then the means compared by a paired Student’s t-test (Fig. 3B). This was done to eliminate blot-to-blot variation. Six blots were compared for COX I and four blots compared
Fig. 2. (See left) Mitochondrial translation products. Following the protein synthesis assay, 35S-labeled proteins were collected by centrifugation, then suspended in electrophoresis buffer (see Section 2). (A) The samples were subjected to electrophoresis and then used for autoradiography. This autoradiograph represents one of four gels. Bands were identified according to the work of Polosa and Attardi (1991). In all cases, the signal was reduced in the 24-month-old rat. (B) The signals were quantitated using the Imagequant software package and then standardized to a band of approximately 40 kDa from the coomassie stained gels (determined by densitometry). Six samples were compared for each age group using four different gels. Statistical significance was evaluated using Student’s t-test for unpaired data. †P B 0.03, ‡P B0.01, *PB 0.04, PB 0.02, PB 0.007.
188
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
for COX IV. These results support the mitochondrial protein synthesis data which shows a reduction in overall protein synthesis rate as well as reduced production of the COX I subunit.
3.4. Nuclear and mitochondrial mRNA le6els Steady-state levels of mitochondrial mRNA were quantitated to determine if decreased mRNA levels were associated with the decrease in mitochondrial translation. Total mRNA was isolated from heart tissue of 6- and 24-month-old rats and subjected to Northern blot analysis using cDNA probes of cytochrome c oxidase subunits from both genomes. Message levels of the mitochondrial-encoded cytochrome c oxidase subunit I (COX I) were decreased in the aged rat heart (Fig. 4A). Similarly, nuclear-encoded cytochrome c oxidase subunit IV (COX IV) message levels were decreased in the aged rat heart (Fig. 4A). However, message levels of mitochondrial-encoded cytochrome c oxidase subunit 2 (COX II) and nuclear-encoded citrate synthase failed to change with age. These results were confirmed when the message levels were standardized to 18s ribosomal RNA (Fig. 4B). With decreases in the expression of COX I and COX IV and the unchanged expression of COX II, the likelihood that the mitochondrial genome is down-regulated with age in its entirety is reduced.
4. Discussion This report is in agreement with a number of reports which have shown age-associated decreases in cytochrome c oxidase activity (Hansford and Castro, 1982; Yamada et al., 1995). The regulation of cytochrome c oxidase activity is
Fig. 4. Northern blot analysis of cytochrome c oxidase subunits I, II and IV. Total RNA was isolated from 6- and 24-month-old rat hearts according to Section 2. For COX I and COX II, 10 mg RNA was loaded in each lane while 30 mg was used for COX IV. (A) Northern blot analysis showed decreased expression of COX I and IV and elevated COX II expression. (B) Signals were quantitated using FragmeNT Analysis™ software and standardized to the 18s rRNA signal (see Section 2). As seen in the blot, both COX I and IV levels decreased with age, whereas COX II increased with age. Due to biological variation, these changes did not reach significant levels. The case was the same for COX II which increased with age. A large number of samples (16 – 21) and a number of different gels (3 – 4) were used for these experiments.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
189
currently unknown and confounded by the fact that the function of most of the 10 nuclear-encoded subunits of cytochrome c oxidase has yet to be defined. The complexity of the mitochondrial genetic expression system makes it unlikely that a single event is responsible for the age-associated decline in mitochondrial activity shown here and elsewhere. Indeed, several different factors have been identified in the age-linked decrease in cytochrome c oxidase activity. Hayakawa et al. have shown that the decrease in the aging rat heart correlates with a decrease in mitochondrial DNA content and an increase in 8-hydroxydeoxyguanosine, an oxidative product of deoxyguanosine (Hayakawa et al., 1993). They suggest that decreased cytochrome c oxidase activity is the manifestation of an overall deterioration of the mitochondrial genome. Abu-Erreish and Sanadi reported a decrease in cytochrome c oxidase activity that paralleled a decrease in cytochrome content and suggested that the age-associated decrease in cytochrome content may be due to either changes in the rate of cytochrome synthesis or degradation (Abu-Erreish and Sanadi, 1978). Paradies et al. have shown in the hypothyroid rat heart a similar decrease in cytochrome c oxidase activity and cytochrome content but contrary to Abu-Erreish and Sanadi suggested that the decline in cytochrome content could not be entirely responsible for the decrease in cytochrome c oxidase activity. They demonstrated that the decrease in cytochrome c oxidase activity corresponded to a decrease in membrane cardiolipin content in the hypothyroid rat compared to normal and thyroid hormone treated rats. They suggested that alterations in the phospholipid environment of the inner membrane may also play a role in the regulation of cytochrome c oxidase activity (Paradies et al., 1993). When this finding is taken in conjunction with the reports of McMillin et al. which showed a decrease in cardiolipin content with age and, more recently, that of Paradies et al. which restored cytochrome c oxidase activity with the addition of exogenous cardiolipin (Paradies et al., 1997), the possibility arises that alterations in membrane composition may contribute to the regulation of cytochrome c oxidase with age. The current report demonstrates that reduced mitochondrial protein synthesis and gene expression contribute to the decline in cytochrome c oxidase activity in the aging rat heart. Since mitochondrial protein synthesis is responsible for the production of the three catalytic subunits of cytochrome c oxidase (Kim et al., 1995), changes in translational activity with age may play a major role in the decrease in cytochrome c oxidase activity. In this report, the decrease in the rate of translation and in the level of products synthesized corresponds well with a decrease in steady-state levels of COX I protein. Although the mitochondrion generates the mRNA, rRNA and tRNA required for protein synthesis, cytosolic proteins, such as initiation and elongation factors, control mitochondrial protein synthesis (Liao and Spremulli, 1991; Barker et al., 1993). Mitochondrial protein synthesis has been shown to increase in both the insulin perfused rat heart (McKee and Grier, 1990) and in the heart of thyroxine treated rats (Leung and McKee, 1990). These hormones may stimulate cytosolic synthesis of nuclear-encoded proteins necessary for mitochondrial translation. Since it has been reported that the action of these hormones is reduced with age (Mooradian and Wong, 1994; Muller et al., 1996), the possibility
190
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
exists that reduced hormone action with age may result in lowered expression of mitochondrial translation factors, hence decreased protein synthesis. Recent work from a number of laboratories has identified, purified and cloned a few of these translation factors but no work has been done to elucidate their role in aging (Liao and Spremulli, 1991; Barker et al., 1993). In addition to a majority of the subunits of Complexes I, III, IV and V, the remaining mitochondrial enzymes are all nuclear-encoded. Changes in their expression and synthesis will have dramatic effects on mitochondrial function. Ostronoff et al. have shown that inhibition of cytosolic protein synthesis with cycloheximide reduced cytochrome c oxidase activity and mitochondrial protein synthesis in liver (Ostronoff et al., 1996). Therefore, in addition to the requirement of nuclear-encoded proteins for proper mitochondrial function, they suggest that mitochondrial protein synthesis occurs at a point when synthesis of nuclear-encoded subunits increases profoundly, suggesting that some interaction between the two systems exists. However, the possibility remains that inhibition of cytosolic protein synthesis would also reduce synthesis of factors needed for mitochondrial translation. In addition to the situations discussed above, the current report shows a decrease in mitochondrial translation associated with a decrease in the levels of some mitochondrial mRNA. It does not rule out the possibility that there is a disturbance in the interaction of the two genomes or a decrease in translational factor content or action, but it presents an additional mechanism whereby reduced transcript levels may eventually alter protein synthesis. The reason for the decrease in COX I expression with age, but lack of change in COX II expression, is currently unknown. Although it has been shown that the mitochondrial genome is polycistronic with a single transcript synthesized by each strand, there are situations when the genes may be differentially expressed. Iglesias et al. have reported a decrease in mitochondrial ND3 expression in hypothyroid rats and a thyroid hormone receptor binding site in the mitochondrial ND3 gene suggesting direct transcriptional regulation of ND3 by thyroid hormone (Iglesias et al., 1995). Mouse cells treated with interferon showed a decrease in cyt b and ND5 mRNA expression while ND6 expression remained unchanged (Shan et al., 1990). Cycloheximide blocked the capacity of interferon to reduce mitochondrial gene expression suggesting the existence of an interferon-responsive nuclear gene capable of regulating mitochondrial gene expression. A recent report from Davis et al. showed an increased incidence of point mutations in the COX I and COX II genes in Alzheimer’s patients associated with a decrease in cytochrome c oxidase activity. In their report, 50% of the Alzheimer’s patients screened showed a mutation in the COX I gene while 30% had a mutation in COX II. This raises the possibility that COX I may be specifically altered with age resulting in its decreased expression and eventual impairment of the function of cytochrome c oxidase. Possible deletions of segments of the mitochondrial genome are perhaps the most striking situation whereby COX I and COX II may be differentially expressed. It has been shown that a number of diseases and aging are associated with deletions in mitochondrial DNA (Szabolcs et al., 1994; Melov et al., 1997). The accumulation of mitochondrial DNA mutations in non-mitotic cells such as neurons may contribute to a number
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
191
of age-associated diseases such as Alzheimer’s (Davis et al., 1997). All of these situations demonstrate different mechanisms whereby mitochondrial RNA may be differentially expressed. Another potential mechanism for the regulation of mitochondrial gene expression may be through the altered expression and/or action of the mitochondrial transcription factor-1 (MTF-1) and the nuclear respiratory factors-1 and -2 (NRF-1 and NRF-2). MTF-1 is a nuclear-encoded DNA-binding protein involved in both mitochondrial DNA replication and transcription through its interaction with the mt-DNA D-loop (Fisher and Clayton, 1985). In turn, MTF-1 is regulated by NRF-1 and NRF-2 (Virbasius and Scarpulla, 1994). A decrease in their expression would result in decreased mitochondrial gene expression. Interestingly, NRF-1 binds the promoter and activates the COX VIc (Evans and Scarpulla, 1990) subunit while NRF-2 activates expression of the COX IV gene (Virbasius and Scarpulla, 1991). Therefore, the corresponding decrease in COX I and COX IV expression may be the result of altered NRF-2 expression or action. In any event, NRF-1 and NRF-2 are capable of affecting both mitochondrial and nuclear encoded subunits of cytochrome c oxidase and may act as signals between the two genomes. Several different phenomena appear to be involved in the age-associated decline in cytochrome c oxidase activity in the Wistar rat heart. It is highly unlikely that a single event or disturbance leads to this decline. This report shows that mitochondrial protein synthesis and the expression of some, but not all, of the mitochondrial genome declines with age. Whether this is a consequence of free radical damage to the genome or altered transcriptional regulation is not known and the subject of current research.
References Abu-Erreish, G.M., Sanadi, D.R., 1978. Age-related changes in cytochrome concentration of myocardial mitochondria. Mech. Ageing Dev. 7, 425 – 432. Anderson, S., Bankier, A.T., Barrell, B.G., de Bruijn, M.H.L., Coulson, A.R., Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Sanger, F., Schreier, P.H., Smith, A.J.H., Staden, R., Young, I.G., 1981. Sequence and organization of the human mitochondrial genome. Nature 290, 457 – 465. Attardi, G., 1988. Biogenesis of mitochondria. Annu. Rev. Cell Biol. 4, 289 – 333. Barker, C., Markris, A., Patiotis, C., Bear, S.E., Tsichlis, P.N., 1993. Identification of the gene encoding the mitochondrial elongation factor G in mammals. Nucleic Acids Res. 21, 2641 – 2647. Clayton, D.A., 1991. Replication and transcription of vertebrate mitochondrial DNA. Annu. Rev. Cell Biol. 7, 453–478. Castelluccio, C., Baracca, A., Fato, R., Pallotti, F., Maranesi, M., Barzanti, V., Gorini, A., Villa, R.F., Castelli, G.P., Marchetti, M., Lanaz, G., 1994. Mitochondrial activities of rat heart during ageing. Mech. Ageing Dev. 76, 73–88. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156 – 159. Davis, R.E., Miller, S., Herrnstadt, C., Ghosh, S.S., Fahy, E., Shinobu, L.A., Galasko, D., Thal, L.J., Beal, M.F., Howell, N., Davis Parker, W. Jr., 1997. Mutations in mitochondrial cytochrome c oxidase genes segregate with late-onset Alzheimer disease. Proc. Natl. Acad. Sci. USA 92, 4526 – 4531.
192
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
Evans, M.J., Scarpulla, R.C., 1990. NRF-1: a trans-activator of nuclear-encoded respiratory genes in animal cells. Genes Dev. 4, 1023–1034. Fisher, R.P., Clayton, D.A., 1985. A transcription factor required for promoter recognition by human mitochondrial RNA polymerase. J. Biol. Chem. 260, 11330 – 11338. Hansford, R.G., 1978. Lipid oxidation by heart mitochondria from young adult and senescent rats. Biochem. J. 170, 285–295. Hansford, R.G., Castro, F., 1982. Age-linked changes in the activity of enzymes of the tricarboxylate cycle and lipid oxidation, and of carnitine content, in muscles of the rat. Mech. Ageing Dev. 19, 191–201. Harman, D., 1993. Free radical theory of aging: a hypothesis of senile dementia of the Alzheimer’s type. Age 16, 23–30. Hayakawa, M., Sugiyama, S., Hattori, K., Takasawa, M., Ozawa, T., 1993. Age-associated damage in mitochondrial DNA in human hearts. Mol. Cell. Biochem. 119, 95 – 103. Holt, I.J., Harding, A.E., Morgan-Hughes, J.A., 1988. Deletions of mitochondrial DNA in patients with mitochondrial myopathies. Nature 331, 717 – 719. Iglesias, T., Caubı´n, J., Zaballos, A., Bernal, J., Mun˜oz, A., 1995. Identification of the mitochondrial NADH dehydrogenase subunit 3 (ND3) as a thyroid hormone regulated gene by whole genome PCR analysis. Biochem. Biophys. Res. Commun. 210, 995 – 1000. Kim, K., Lecordier, A., Bowman, L.H., 1995. Both nuclear and mitochondrial cytochrome c oxidase mRNA levels increase dramatically during mouse postnatal development. Biochem. J. 306, 352 – 358. Leung, A.C., McKee, E.E., 1990. Mitochondrial protein synthesis during thyroxine-induced cardiac hypertrophy. Am. J. Physiol. 258, E511 – E518. Liao, H.X., Spremulli, L.L., 1991. Initiation of protein synthesis in animal mitochondria. J. Biol. Chem. 266, 20714–20719. Melov, S., Hinerfeld, D., Esposito, L., Wallace, D.C., 1997. Multi-organ characterization of mitochondrial genomic rearrangements in ad libitum and caloric restricted mice shows striking somatic mitochondrial DNA rearrangements with age. Nucleic Acids Res. 25, 974 – 982. McKee, E.E., Grier, B.L., 1990. Insulin stimulates mitochondrial protein synthesis and respiration in isolated perfused rat heart. Am. J. Physiol. 259, E413 – E421. McKee, E.E., Grier, B.L., Thompson, G.S., McCourt, J.D., 1990. Isolation and incubation conditions to study mitochondrial protein synthesis. Am. J. Physiol. 258, E492 – E510. McMillin, J.B., Taffet, G.E., Taegtmeyer, H., Hudson, E.K., Tate, C.A., 1993. Mitochondrial metabolism and substrate competition in the aging Fisher rat heart. Cardiovascular Res. 27, 2222–2228. Mooradian, A.D., Wong, N.C.W., 1994. Age-related changes in thyroid hormone action. Eur. J. Endocrinol. 131, 451–461. Moraes, C.T., Ricci, E., Petruzzella, V., Shanske, S., DiMauro, S., Schon, E.A., Bonilla, E., 1992. Molecular analysis of the muscle pathology associated with mitochondrial DNA deletions. Nat. Genet. 1, 359–367. Muller, D.C., Elahi, D., Tobin, J.D., Andres, R., 1996. The effect of age on insulin resistance and secretion: a review. Semin. Nephrol. 16, 289 – 298. Ostronoff, L.K., Izquierdo, J.M., Montoya, J.A., Cuezva, J.M., 1996. Transient activation of mitochondrial translation regulates the expression of the mitochondrial genome during mammalian mitochondrial differentiation. Biochem. J. 316, 183 – 191. Paradies, G., Ruggiero, F.M., Dinoi, P., Petrosillo, G., Quagliariello, E., 1993. Decreased cytochrome oxidase activity and changes in phospholipids in heart mitochondria from hypothyroid rats. Arch. Biochem. Biophys. 307, 91–95. Paradies, G., Ruggiero, F.M., Petrosillo, G., Quagliariello, E., 1997. Age-dependent decline in the cytochrome c oxidase activity in rat heart mitochondria: role of cardiolipin. FEBS Lett. 406, 136–138. Polosa, P.L., Attardi, G., 1991. Distinctive pattern and translational control of mitochondrial protein synthesis in rat brain synaptic endings. J. Biol. Chem. 266, 10011 – 10017. Poyton, R.O., McEwen, J.E., 1996. Crosstalk between nuclear and mitochondrial genomes. Annu. Rev. Biochem. 65, 563–607.
E.K. Hudson et al. / Mechanisms of Ageing and De6elopment 103 (1998) 179–193
193
Shan, B., Vasquez, E., Lewis, J.A., 1990. Interferon selectively inhibits the expression of mitochondrial genes: a novel pathway for interferon-mediated responses. EMBO J. 9, 4307 – 4314. Szabolcs, M.J., Seigle, R., Shanske, S., Bonilla, E., DiMauro, S., D’Agati, V., 1994. Mitochondrial DNA deletion: a cause of chronic tubulointerstitial nephropathy. Kidney Int. 45, 1388 – 1396. Towbin, H., Staehlin, T., Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350–4354. Wallace, D.C., Lott, M.T., Shoffner, J.M., Brown, M.D., 1992. Diseases resulting from mitochondrial DNA point mutations. J. Inher. Metab. Dis. 15, 472 – 479. Yamada, K., Sugiyama, S., Kosaka, K., Hayakawa, M., Ozawa, T., 1995. Early appearance of age-associated deterioration in mitochondrial function of diaphragm and heart in rats treated with doxorubicin. Exp. Gerontol. 30, 581– 593. Virbasius, J.V., Scarpulla, R.C., 1991. Transcriptional activation through ETS domain binding sites in the cytochrome c oxidase subunit IV gene. Mol. Cell. Biol. 11, 5631 – 5638. Virbasius, J.V., Scarpulla, R.C., 1994. Activation of the human mitochondrial transcription factor A gene by nuclear respiratory factors: a potential regulatory link between nuclear and mitochondrial gene expression in organelle biogenesis. Proc. Natl. Acad. Sci. USA 91, 1309 – 1313.
.