Cellular phenotypes of age-associated skeletal muscle mitochondrial abnormalities in rhesus monkeys

Mutation Research 452 Ž2000. 123–138 www.elsevier.comrlocatermolmut Community address: www.elsevier.comrlocatermutres

Cellular phenotypes of age-associated skeletal muscle mitochondrial abnormalities in rhesus monkeys Marisol E. Lopez a , Nathan L. Van Zeeland a , David B. Dahl b, Richard Weindruch c , Judd M. Aiken a,) a

b c

Department of Animal Health and Biomedical Sciences, VA GRECC and Wisconsin Regional Primate Research Center, UniÕersity of Wisconsin, 1656 Linden DriÕe, Madison, WI 53706, USA Department of Statistics, VA GRECC and Wisconsin Regional Primate Research Center, UniÕersity of Wisconsin, Madison, WI, USA Department of Medicine, VA GRECC and Wisconsin Regional Primate Research Center, UniÕersity of Wisconsin, Madison, WI, USA Received 27 January 2000; received in revised form 29 March 2000; accepted 3 April 2000

Abstract Rhesus monkey vastus lateralis muscle was examined histologically for age-associated electron transport system ŽETS. abnormalities: fibers lacking cytochrome c oxidase activity ŽCOXy. andror exhibiting succinate dehydrogenase hyperreactivity ŽSDHqq.. Two hundred serial cross-sections Žspanning 1600 mm. were obtained and analyzed for ETS abnormalities at regular intervals. The abundance and length of ETS abnormal regions increased with age. Extrapolating the data to the entire length of the fiber, up to 60% of the fibers were estimated to display ETS abnormalities in the oldest animal studied Ž34 years. compared to 4% in a young adult animal Ž11 years.. ETS abnormal phenotypes varied with age and fiber type. Middle-aged animals primarily exhibited the COXy phenotype, while COXyrSDHqq abnormalities were more common in old animals. Transition region phenotype was affected by fiber type with type 2 fibers first displaying COXy and then COXyrSDHqq while type 1 fibers progressed from normal to SDHqq and then to COXyrSDHqq. In situ hybridizations studies revealed an association of ETS abnormalities with deletions of the mitochondrial genome. By measuring cross-sectional area along the length of ETS abnormal fibers, we demonstrated that some of these fibers exhibit atrophy. Our data suggest mitochondrial ŽmtDNA. deletions and associated ETS abnormalities are contributors to age-associated fiber atrophy. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Mitochondria; Aging; Skeletal muscle; Electron transport system; Atrophy

1. Introduction During oxidative respiration, mitochondria consume about 90% of cellular oxygen of which ; 2% )

Corresponding author. Tel.: q1-608-262-7362; fax: q1-608262-7420. E-mail address: [email protected] ŽJ.M. Aiken..

is converted to reactive oxygen species ŽROS. by the electron transport system wETS; w1xx. ROS involvement in the aging process was first proposed by Harman w2x who suggested that free radicals may damage cell constituents. Over two decades later, Miquel et al. w3x focused this theory by proposing that the ROS damage primarily affected mitochondrial DNA ŽmtDNA. due to its close proximity to the

0027-5107r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 0 0 . 0 0 0 5 9 - 2

124

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

inner mitochondrial membrane where ROS are produced. In addition, these investigators proposed that post-mitotic tissues that have high-energy demands Žsuch as heart, brain and skeletal muscle. are most severely affected. Deletions of the mitochondrial genome increase in abundance with age in various post-mitotic tissues of humans w4–6x, rodents w7–9x and monkeys w10,11x. Although low abundances for these deletions have been reported, these analyses were performed on tissue homogenates; such analyses assume deletions Žand associated ETS abnormalities. to be distributed uniformly among cells Žreviewed in Ref. w12x.. Two alternative approaches, examination of defined number of cells and in situ hybridization, demonstrate, however, that deletions are mosaicly distributed. Fiber bundle analysis in which defined numbers of muscle fibers were analyzed indicated that, as the numbers of fibers assayed decreased, Ži. the number of individual deletion products decreased and, Žii. the calculated abundance of individual deletion products increased w13x. In situ hybridization studies identified high accumulations of mitochondrial deleted genomes localized to individual cells and, in adjacent sections, these cells exhibited abnormalities of the ETS w14x. Taken together, these analyses indicated that mtDNA deletions are not uniformly distributed among cells but focally accumulate to high levels in a subset of cells. Activities of two complexes of the ETS are commonly analyzed: complex II Žsuccinate dehydrogenase; SDH. and complex IV Žcytochrome c oxidase; COX.. Abnormal fibers are characterized by increased succinate dehydrogenase activity wSDHqq; w15xx and, in some cases, a lack of cytochrome c oxidase activity wCOXy; w16xx. These fibers have been termed ‘‘ragged red’’ fibers ŽRRF. due to their staining pattern with modified Gomorri trichrome stain. Ultrastructural studies have revealed RRF to accumulate enlarged mitochondria w17,18x. In human quadriceps, COXy fibers appear sporadically in the third decade Ž; 0.02% of total fibers examined. but at much higher numbers Ž0.1–5% of total fibers examined. in the fifth and later decades w15,19x. Ragged red fibers were initially characterized in human mitochondrial myopathy diseases that are neuromuscular disorders associated with defects of the mitochondrial genome w20x. Some mitochondrial

myopathies ŽKearns–Sayre syndrome, progressive external ophthalmoplegia, and Pearson’s syndrome. are caused by mtDNA deletions. Others ŽLeber’s hereditary optic neuropathy; myoclonus epilepsy with ragged red fibers; and mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes. are caused by point mutations and are usually maternally inherited. In contrast to aging individuals, high levels of deleted mitochondrial molecules characterize tissues from these patients Žreviewed in Ref. w21x.. Abnormalities of the ETS are also associated with deletions of the mitochondrial genome in myopathy patients w22,23x. Mitochondrial abnormalities in skeletal muscle fibers are segmental in nature w17x; however, the length of these regions has not been clearly determined in myopathy patients or aging individuals. We have previously documented an age-associated increase in the abundance of ETS abnormalities in rhesus monkey quadriceps w24x. This study involved the analysis of single cross-sections of muscle and indicated the necessity of a longitudinal approach in which a series of adjacent cross-sections are studied through a specific length of muscle. In the present study, a longitudinal analysis of vastus lateralis Žone of the quadriceps muscles. from 11 rhesus monkeys of diverse ages was performed such that 1600 mm of muscle from each animal was characterized for abundance, length and changes in fiber cross-sectional area of ETS abnormal regions.

2. Methods 2.1. Tissue Rhesus monkey Ž Macaca mulatta. tissue biopsies were obtained from 11 animals, ages 11 to 34 years old, housed at the Wisconsin Regional Primate Research Center as described previously w10x. The maximum lifespan for rhesus monkeys at this Center is about 40 years. All procedures concerning the rhesus monkeys were approved by the Institutional Animal Care and Use Committee of the University of Wisconsin. A 5-mm3 section of vastus lateralis was placed in OCT mounting media ŽMiles, Elkhart, IN. and frozen in liquid nitrogen. The samples were stored at y808C until use.

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

2.2. Identification of ETS abnormal fibers Tissue was brought to the temperature of the cryostat Žy208C. and 8-mm transverse frozen sections Ž200 serial sections or 1600 mm per animal. were cut and placed on Probe-On Plus slides ŽFisher Scientific, Pittsburgh, PA.. Sections were stored at y808C until use. For the identification of fibers harboring ETS abnormal regions, COX and SDH enzyme activities were characterized. Enzymatic staining of COX w25x and SDH w26x activities was performed in cross-sections 56 mm Žseven sections. apart. For each animal, sections a1, a7, a13, . . . , and a199 were stained for COX activity Žtotal of 30 COX stained slides. while sections a2, a8, a14, . . . , and a200 were stained for SDH activity Žtotal of 30 SDH stained slides.. Each stained section was examined on an Olympus BH2 microscope under bright field illumination to identify fibers exhibiting abnormal COX andror SDH activities. For example, the first pair of COX and SDH stained slides Žslides a1 and 2. were examined for the presence of COXy and SDHqq fibers. All fibers exhibiting ETS abnormalities were followed in subsequent pairs of stained slides Žslides a7 and 8, 13 and 14, . . . , 199 and 200.. As these abnormalities are segmental, some of the ETS abnormal fibers found in the second pair of slides had been already recorded when examining the first pair of slides. To avoid recording an abnormal fiber more than once, fibers displaying an ETS abnormal region were marked on a digital picture of the cross-section. 2.3. Length of ETS abnormal regions To determine the region at which a fiber changed from being ETS normal to ETS abnormal, additional slides, in close proximity to those exhibiting ETS abnormalities, were stained for COX and SDH activities. The number of sections exhibiting ETS abnormalities were then counted and the length of the ETS abnormal region of the fiber determined Žeach section being 8 mm thick..

125

dilution. This antibody reacts only with type 2 fibers. Anti Mouse IgG alkaline phosphatase conjugated antibody ŽSigma, St. Louis, MO. was used as a secondary antibody at a 1:200 dilution to determine fiber type. Type 2 fibers were further classified into type 2a and 2b using the ATPase stain following the procedure of Ogilvie et al. w27x. 2.5. Fiber number Fiber number was determined from microscopic images Ž40 = magnification. of the sectioned muscle obtained with a digital camera ŽHitachi HVC20rC20M.. This allowed for the determination of the percentage of fibers exhibiting ETS abnormalities and the percentages of type 1 and type 2 fibers in the cross-section. 2.6. In situ hybridizations In situ hybridizations were performed using seven different mitochondrial probes. The preparation of 12s rRNA Žnts 781–954., ND2 Žnts 4623–5082., COX I Žnts 7170–7505., ND4 Žnts 11,347–11,565. and Cytb Žnts 14,891–15,232. was according to Lee et al. w24x. Two additional regions of the rhesus monkey mitochondrial genome ŽATPase6 and COX III. were amplified by polymerase chain reaction ŽPCR. and ligated into pGEM 3Z ŽPromega; Madison, WI.. PCR primer locations wbased on the human mtDNA numbering system; w28xx and sequences Žspecific for rhesus mtDNA. were as follows: ATPase6 Žnts 8966–8990 w5X -TCAGTCTACTTATTCAACCAGTGGC-3X x and nts 9224–9248 w5XACTGGGTTTAACTATGTGATAAGCG-3X x. and COX III Žnts 9225–9249 w5X-GCTTATCACATAGTTAAACCCAGTC-3X x and nts 9857–9879 w5XGTAGTAGTTGGCGGATGAGGCAG-3X x.. Antisense and sense RNA probes were synthesized using the MAXIscript In Vitro Transcription Kit ŽAmbion; Austin, TX. with the incorporation of 35 S a-UTP as previously described w24x. 2.7. Measurements of cross-sectional area

2.4. Fiber type Monoclonal anti-skeletal myosin Žfast. clone MY32 ŽSigma; St. Louis, MO. was used at a 1:400

Muscle fiber cross-sectional area ŽCSA. was determined in four animals from this study Ž23, 30, 32 and 33 years old.. For each animal, fiber CSA was

126

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

measured in all fibers containing ETS abnormalities and in twice that number of randomly selected ETS normal fibers. Cross-sectional area was measured using Image-Pro Plus software version 4.0 ŽMedia Cybernetics; Silver Spring, MD.. For each ETS abnormal fiber, CSA was determined in each and every slide where the fiber was abnormal. On the same slides, the CSA of control fibers Žthose not displaying an abnormality. was also determined. The percent change in CSA within an ETS abnormal region was obtained by the following method. The smallest value of CSA within an abnormal region ŽminCSA. was subtracted from the largest value of CSA within this same region ŽmaxCSA.. The percentage of this difference from the largest value of CSA was then obtained w100 = ŽmaxCSAy minCSA.rmaxCSAx. The same method was used to determine percent change in CSA for control fibers. 2.8. Statistical analyses The relationship between percentage ETS abnormal fibers and age was tested using standard regression techniques. The association between abnormality length and age was evaluated using simple linear regression and the Cox proportional hazards model w29x. Chi-square analysis was used to test if the proportion of abnormal fibers differed by fiber type. The percentage change in CSA in abnormal fibers was compared to that of control fibers using the Cox proportional hazards model. The distributions of percentage change in CSA of abnormal and control fibers were estimated using kernel densities Žarea under the curve gives the probabilities associated with the corresponding intervals along the horizontal axis.. For all statistical analyses, p values - 0.05 were considered statistically significant. All statistical analyses were performed using R Žversion 0.64.1. software ŽFree Software Foundation; Boston, MA..

3. Results 3.1. ETS abnormal fiber abundance and age Vastus lateralis biopsy samples from eleven rhesus monkeys Ž11 to 34 years old. were examined for

the presence of ETS abnormalities. Two hundred serial sections were obtained from each muscle biopsy. COX and SDH stains were performed every 56 mm. The longitudinal approach employed is illustrated in Fig. 1. By using this technique, we could follow all the fibers in a cross-section for a total length of 1600 mm. The characterization of COX and SDH enzymatic activities determined that ETS abnormalities do not occur along the whole fiber but are limited to regionŽs. within a fiber. Thus, to estimate accurately the number of ETS abnormalities present in a muscle, an analysis along the length of the muscle is required. An average of 2403 fibers Žrange: 927–6084. was examined per animal. This large range of fiber numbers is reflective of the size of the muscle biopsy obtained. The number of fibers which, somewhere along the 1600 mm region examined, displayed an ETS abnormality was determined ŽTable 1.. ETS abnormal regions were found in biopsies from all animals studied. Typically, these abnormal regions occurred only once within a fiber; however, three fibers in two old animals Žone fiber in the 33- and two fibers in the 34-year-old. harbored two ETS abnormal regions. The percentage of fibers exhibiting ETS abnormal regions increased with age Ž p 0.001; Fig. 2A.. Further, the rate of increase of ETS abnormal regions was higher in older than in younger animals. 3.2. Length of ETS abnormal regions and age The longitudinal approach used in this study allowed the determination of the length of the ETS abnormal regions. For most of fibers exhibiting ETS abnormalities, the abnormality started and ended within the 1600-mm region examined, thereby allowing a precise determination of abnormality length. A number of abnormalities Ž68 of the 283 identified., however, extended beyond the 1600-mm region examined; thus, in these fibers the data was censored, and only a minimum length of the abnormality could be determined. In spite of these constraints, muscle fibers from older animals exhibited longer ETS abnormal regions than those of younger animals ŽTable 1.. For example, the range of abnormality length was 48 to 208 mm in the 14-year-old animal while it was 32 to 1108 mm in the 34-year-old. Regression meth-

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138 Fig. 1. Segmental nature of ETS abnormalities. The diagram depicts a skeletal muscle fiber bundle with a fiber exhibiting an ETS abnormal region. The region abnormal for both COX and SDH activities is shown in blue. The regions in which the fiber changes from normal to abnormal COX and SDH activities Žtransition regions. are shown in yellow. To represent the longitudinal analysis performed, photomicrographs of cross-sections from a 32-year-old rhesus monkey stained for COX ŽA, C, E, G and I. and SDH ŽB, D, F, H and J. enzymatic activities are shown. The fiber indicated by an ‘‘x’’ is COXyrSDHqq in panels E and F but is phenotypically normal for both enzyme activities in panels A and B Ž176 mm away. and I and J Ž440 mm away.. This fiber exhibits increased SDH activity but normal COX activity in the transition regions ŽC, D and G, H.. When additional sections were analyzed Ždata not shown. it was found that the total length of the abnormality was 568 mm. The white bar in panel J represents 50 mm.

127

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

128

Table 1 Total fiber number, percentage ETS abnormal fibers and ETS abnormality length for each animal studied Age Žyears. Sex Number of fibers a ETS abnormal fibers % ETS abnormalitiesb ETS abnormal region length ŽRange; mm. c

11 M 1848 3 0.16 104–224

14 M 1318 4 0.30 48–208

20 F 6084 19 0.31 60–488

20 M 927 7 0.76 116–452

23 F 3249 18 0.55 68–544

26 M 1268 4 0.32 144–592

26 M 1073 9 0.84 64–448

30 M 1787 85 4.76 32–792

32 F 4177 46 1.10 56–1080

33 M 2586 37 1.43 76–1580

34 M 2115 51 2.41 32–1108

a

Total number of fibers present in the muscle biopsy. The number of fibers exhibiting an ETS abnormal region in the 1600-mm region examined was divided by the total number of fibers examined in the cross-section to obtain the percentage of ETS abnormal fibers. c ETS abnormality length was measured for all fibers and the range determined. b

ods determined average observed abnormality length to increase with age Ž p - 0.001; Fig. 2B, solid line.. As the regression analysis did not account for censoring, the ETS abnormality length is biased low, especially in older animals where censoring was more prevalent. To account for censoring, the Cox proportional hazards model w29x was used to relate age with ETS abnormality length. This model also found an age-associated increase in abnormality length Ž p - 0.01.; however, the increase was more pronounced in middle-aged and old animals ŽFig. 2B, dashed line.. 3.3. Age and ETS abnormal region phenotype An age-associated change in the ETS abnormal phenotype was also observed ŽFig. 3.. ETS abnormal regions from all 7 of the abnormal fibers of young adult Ž11 to 14 years old. and all 57 abnormal fibers of middle-aged Ž20 to 26 years old. animals were negative for COX activity ŽCOXy. while exhibiting SDH activity. A representative example of the ETS abnormal phenotype for young adult ŽFig. 3A and B. and middle-aged ŽFig. 3E and F. animals is shown. In old animals Ž) 30 years old., an additional ETS abnormal phenotype was observed with ; 21% Ž28r134. of the abnormal regions showing a COXyrSDHqq phenotype ŽFig. 3I and J. while the remaining abnormal fibers Ž n s 106. displayed similar phenotypes to those in younger animals. This SDH hyperreactive phenotype was only found in muscle from old animals. To determine if mtDNA deletions are associated with the COXy phenotype, sections containing ETS

abnormal fibers from all three age groups were analyzed by in situ hybridizations using seven different mitochondrial genome probes Ž12S rRNA, ND2, COX I, ATPase6, COX III, ND4 and Cytb.. A total of 25 abnormal fibers from young adult and middleaged monkeys and 50 fibers from old animals were examined within their ETS abnormal region. Deletions, indicated by a lack of signal, were associated with ETS abnormal regions of fibers ŽFig. 3C, G and K.. Deletion patterns differed among ETS abnormal regions, suggesting that different deletion events occur in different muscle fibers. To determine whether the mtDNA deletion event is clonal within the abnormal region of a fiber, in situ hybridizations were performed in two different areas Žat least 100 mm apart. of an ETS abnormal region. Five fibers containing ETS abnormalities were examined in this manner. Although differences in deletion patterns were observed between fibers, within a given ETS abnormal region, identical patterns were seen Ždata not shown., suggesting that within a given ETS abnormal region the mtDNA deletion is clonal. 3.4. Deletion genotype and ETS abnormal region phenotype Analyses along the length of fibers allow for the detection of regions Žtransition regions. in which a fiber changes from displaying normal to abnormal ETS activities. Previously, we established the presence of two different phenotypes in the transition regions of vastus lateralis ETS abnormal fibers ŽLee et al. w24x.. One phenotype occurred more frequently

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

129

COXyrSDHqq Žillustrated in Fig. 1.. As the difference between these two phenotypes is whether the fiber is COXy prior to becoming COXyrSDHqq, we initially hypothesized that if the deletion involved one of the COX subunits, a COX enzymatic deficiency would be the initial ETS abnormality observed. However, if the deletion event involved subunits other than COX, COX deficiency in the transition region would not be observed. In situ hybridizations were performed using three different mitochondrial probes ŽCOX I, COX III and Cytb. to characterize the deletion pattern of 16 ETS abnormal fibers from the 32-year-old animal. Fibers which first became COXy displayed deletions involving any of the three COX subunits. Fibers exhibiting the other transition region phenotype also showed deletions involving the three COX subunits Ždata not shown.. Therefore, no association was found between the phenotype of the transition region and the genotype of the deletion. 3.5. Fiber type and ETS abnormal region phenotype Fig. 2. ETS abnormal fiber abundance and length increase with age. ŽA. Scatter plot of percent fibers exhibiting an ETS abnormality in the 1600 mm region examined versus age. For the purposes of model fitting, the natural logarithm of percent abnormal fibers was regressed on age. The percentage of skeletal muscle fibers exhibiting ETS abnormalities increased with the animal age Ž p- 0.001.. Further, the rate of increase of fibers displaying abnormalities was higher in older animals than in younger animals. One 30-year-old animal displayed a very large percentage of abnormal fibers. Excluding this animal did not substantively change the fitted curve Žthe fitted curve shown includes the 30 year-old animal.. ŽB. Scatter plot of average observed ETS abnormality length versus age. A simple linear regression model Žsolid line. was fit to the data and indicated that as the age of the animal increased, the average abnormality length increased Ž p- 0.001.. Some abnormal regions extended beyond the 1600 mm region studied, thus for those fibers the exact length could not be determined. The Cox proportional hazards model Ždashed line., which accounts for these censored observations, also shows an age-associated increase of abnormality length Ž p0.01..

Ž70% of the fibers examined displayed this phenotype. and was characterized by the change from normal COXy COXyrSDHqq. The other phenotype was observed in 30% of the fibers and involved a transition from normal SDHqq

™

™

™

™

We next studied whether the respiratory capacity of the fiber affected the ETS abnormal phenotype. If this were true, type 1 fibers ŽFig. 4G and H., which have higher mitochondrial enzymatic activities than type 2 fibers ŽFig. 4A and B., would be expected to progress more rapidly to an SDHqq state while type 2 fibers would initially become COXy. To test this hypothesis, fiber type was determined using an antibody specific for type 2 fibers. The majority Ž) 75%. of the fibers in the vastus lateralis were type 2. The ETS abnormalities occurred preferentially in type 2 fibers Ž p - 0.001; Table 2.. Through the use of an ATPase stain, all type 2 abnormal fibers Žfrom the 32-year-old animal; n s 46. were further classified as type 2b. Fiber type analysis revealed that the phenotype of the transition region was affected by fiber type. Only type 2 fibers displayed the phenotype normal COXy COXyrSDHqq. In the ETS abnormal region, these fibers were either COXy ŽFig. 4C. or COXyrSDHqq ŽFig. 1E and F.. Type 2 fibers that were COXyrSDHqq had a tendency to have longer abnormal regions than those type 2 COXy fibers. The other phenotype Žnormal SDH qq y qq . COX rSDH was characteristic of all type 1

™

™

™

™

130 M.E. Lopez et al.r Mutation Research 452 (2000) 123–138 Fig. 3. ETS abnormal region phenotype is associated with age. The photomicrographs are cross-sections from animals 14 ŽA–D., 23 ŽE–H. and 33 ŽI–L. years old. Cross-sections were stained for COX ŽA, E and I. and SDH ŽB, F and J. activities. In situ hybridizations using probes for COXI ŽC., ND4 ŽD and H., 12SrRNA ŽG., ATPase6 ŽK. and Cytb ŽL. are shown. The fibers indicated by an ‘‘x’’ exhibit an ETS abnormal region along their length. Panels ŽA. and ŽB. represent an ETS abnormal regions phenotype from a 14-year-old rhesus monkey characterized by being negative for COX ŽCOXy . activity. Panels ŽE. and ŽF. denote an ETS abnormal phenotype found in a 23-year-old animal characterized by being COXy. Panels ŽI. and ŽJ. show an ETS abnormal region phenotype from a 33 year-old animal which is COXyrSDHqq. Low abundance of transcripts from WT mitochondrial genomes is seen in panels ŽC., ŽG. and ŽK.. Panels ŽD., ŽH. and ŽL. show reaction of the probes with transcripts from deleted mitochondrial genomes. Abnormal fibers shown are type 2 fibers. The white bar in panel L represents 50 mm.

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

131

Fig. 4. ETS abnormal region phenotype is affected by fiber type. The photomicrographs are cross-sections from skeletal muscle fibers representing a type 2 Žpanels A to F. and a type 1 Žpanels G to L. fiber exhibiting ETS abnormal regions Žfibers indicated by an ‘‘x’’.. In the ETS normal region, type 2 fibers displayed low COX ŽA. and low SDH ŽB. enzymatic activities. In the abnormal region, type 2 fibers were typically COXy ŽC.. Type 1 fibers exhibited high COX ŽG. and SDH ŽH. enzymatic activities in the normal region of the fiber. In the abnormal region, type 1 fibers showed a great decrease in COX activity, sometimes becoming COXy ŽI. and they were SDHqq ŽJ.. Results from in situ hybridization studies for the type 2 ŽE and F. and the type 1 ŽK and L. fibers shown in this example using COX III ŽE., Cyt b ŽF and K. and COX I ŽL. mitochondrial probes are presented. The white bar in panel L represents 50 mm.

fibers ŽFig. 4I and J.. These results suggest that the oxidative capacity of a fiber dictates the phenotype of the enzymatic deficiency. In situ hybridizations along ETS abnormal regions revealed differences in mitochondrial transcript

abundance between fiber types ŽFig. 4.. Within the ETS abnormal region, both fiber types exhibited an increase in transcript abundance from deleted genomes ŽFig. 4F and L.. Whereas type 2 fibers showed almost complete loss of transcripts from WT

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

132

Table 2 ETS abnormalities occur preferentially in type 2 fibers The expected numbers of type 1 and 2 abnormal fibers were obtained under the null hypothesis that the proportion of abnormal fibers in both populations Žtypes 1 and 2. are equal. The chi-square analysis shows that ETS abnormalities were found to occur preferentially in type 2 fibers Ž p- 0.001..

Type 1 Type 2

Observed ŽExpected. Observed ŽExpected.

Abnormal

Normal

13 Ž32. 270 Ž251.

2958 Ž2939. 23,096 Ž23,115.

genomes ŽFig. 4E., type 1 fibers displayed a moderate amount of transcripts from wild type genomes in

levels between that of normal type 1 and type 2 fibers ŽFig. 4K.. These results suggest that mitochondrial abnormalities have a greater impact on type 2 than type 1 fibers. 3.6. ETS abnormal fibers and intrafiber atrophy We studied fiber CSA changes for all fibers displaying ETS abnormal regions Ž n s 126. as well as in twice that number of control fibers Ži.e., not exhibiting ETS abnormalities within region examined. of four animals Ž23, 30, 32 and 33 years old.. For each ETS abnormal fiber, CSA was determined

Fig. 5. Fiber atrophy is associated with ETS abnormal regions. For each ETS abnormal fiber, CSA was determined in each slide where the fiber was abnormal. On these same slides, the CSA of control fibers Žthose not displaying an abnormality. was also determined. The distributions Žkernel density estimates. for percentage change in CSA area w100 = ŽmaxCSAy minCSA.rmax CSAx were obtained for abnormal Žsolid line. and control Ždashed line. fibers from animals 23 ŽA., 30 ŽB., 32 ŽC. and 33 ŽD. years old. The % change in CSA was higher in fibers exhibiting abnormal regions than in control fibers Ž p - 0.05.. In the old animals, some of the ETS abnormal fibers exhibited great decreases in CSA Žlong ‘‘tail’’ to the right., not found in the ETS normal fibers.

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

in each slide where the fiber was abnormal. On these same slides, the CSA of control fibers Žthose not displaying an abnormality. was also determined. The percentage change in CSA was obtained Žsee Methods. for all abnormal and control fibers. Percentage change in CSA of ETS abnormal fibers from all animals was compared to percentage change in CSA of control fibers from all animals using a Cox proportional hazards model. A statistically significant difference Ž p - 0.05. was found, with abnormal fibers showing greater changes in CSA than normal fibers.

133

To depict percentage change in CSA for each individual animal, the values of percentage change in CSA for each abnormal and control fiber were plotted and the distributions ŽKernel density: area under the curve gives the probabilities associated with the corresponding intervals along the horizontal axis. for normal ŽFig. 5, dashed curve. and ETS abnormal ŽFig. 5, solid curve. fibers obtained. For a given interval wa,bx, the area under a kernel density estimate was used to approximate the proportion of fibers exhibiting percentage change in CSA between a and b. For example, in Fig. 5, the area under 20–30% change in CSA in all four panels exceeded the area under 30–40% change in CSA; therefore, a larger fraction of the population displayed a change in CSA of 20–30% than 30–40%. Analysis of the distribution of percentage change in CSA indicated that robust changes in CSA were found only in old animals. In the 23-year-old animal ŽFig. 5A., the distribution of abnormal fibers was similar to that of the normal fibers. In the older animals ŽFig. 5B–D., the distribution of the abnormal fibers again looked similar to that of the normal fibers, with the notable exception of a long ‘‘tail’’ to the right. Each of the fibers located in the ‘‘tail’’ Žwith percentage change in CSA higher than 50%. of the kernel distributions from the old animals were further analyzed. This analysis showed that, in all cases, this population of abnormal fibers from the older animals exhibited a decrease in CSA indicating considerable intrafiber atrophy. Interestingly, some of these atrophying fibers regained their original CSA after becoming ETS normal again. Others ended at their smallest CSA suggesting that these fibers had broken. A representative example of one of these fibers from the 33-year-old animal is shown in Fig. 6.

4. Discussion Fig. 6. Fiber displaying intrafiber atrophy within an ETS abnormal region. The photomicrographs are cross-sections of vastus lateralis muscle from a 33-year-old monkey. Cross-sections were stained for COX ŽD. and SDH ŽA–C, E–H. activities. This fiber progresses from having normal SDH activity ŽA., becomes SDHqq ŽC. and COXy ŽD. and shows a decrease in cross-sectional area ŽG and H.. The fiber could not be identified in subsequent sections. Total distance between panels A and H is 504 mm. The white bar in panel H represents 50 mm.

We sought to estimate accurately the frequency and length of mitochondrial abnormalities that accumulate with age in skeletal muscle and to characterize further the intrafiber atrophy that we recently identified w24x. The segmental nature of these abnormalities necessitated the analysis of many cross-sections. In addition, differences in the ETS abnormal phenotype associated with fiber type required the

134

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

monitoring of both ETS abnormal phenotypes ŽCOXy, SDHqq .. Furthermore, the analysis of numerous sections along a fiber length provided insight on the occurrence of fiber atrophy in ETS abnormal fibers. 4.1. Abundance of fibers harboring ETS abnormalities increases with age An age-associated increase in the abundance of mitochondrial abnormalities was found in vastus lateralis muscle from rhesus monkeys, the rate of increase being higher in older animals. We followed muscle fibers throughout a 1600-mm region and found ETS abnormalities to be segmental and to extend through a portion of the fiber. Some fibers displayed more than one ETS abnormal region along their length Žparticularly in the older animals.. Extrapolating the percentage of ETS abnormalities Ž2.41%. found in the 1600-mm region examined from a 34-year old, to the entire length of the muscle fiber from the vastus lateralis w; 4 cm w30,31xx, we predict that ; 60% of the muscle fibers from this animal would contain mitochondrial abnormalities somewhere along their length. In our previous study analyzing single cross-sections Ž8 mm. of vastus lateralis from rhesus monkeys aged 2–39 years old, we did not detect ETS abnormalities in animals younger than 20 years of age w24x. However, in the present study, by analyzing a greater region of muscle Ž1600 mm., we detected abnormalities even in the youngest animals Ž11 and 14 years old. analyzed. Studies performed in human limb w15,16,19,32–34x, extraocular w35x and diaphragm w16,36x muscles also describe higher levels of ETS abnormalities in older individuals. Rifai et al. w15x report that the levels of SDHqq fibers are higher in older Ž0.33%; 61–77 years old. than younger Ž0.02%; 21–31 years old. individuals. Brierley et al. w19x found 0.1–5% COXy fibers in human quadriceps of individuals 62–85 years old. These percentages are considerably lower than our estimates and likely reflect the single cross-sections analyzed by these investigators. 4.2. Length of ETS abnormalities increases with age Studies of mitochondrial abnormalities in myopathy patients and aging individuals have found them

usually limited to regions within muscle fibers, a pattern which has been termed ‘‘segmental’’ w37x. This conclusion was reached after examining longitudinally sectioned muscle fibers and finding regions of COX deficiency located within muscle fibers w14,17,35,37–41x. Other investigators have followed individual COXy fibers a few serial cross-sections away and found that these were no longer abnormal w19,22,34,37,42x. The most extensive study analyzing serial cross-sections, to our knowledge, is that by Yamamoto and Nonaka w37x in which one muscle fiber from a myopathy patient changed from having normal COXrSDH activities to becoming abnormal followed by a return to normal activities within 610 mm. Even though the concept of ‘‘segmental’’ has been described, the length of these abnormal segments of fibers still remains unclear. In the present study, we identified all ETS abnormal regions Žthat could be detected by examining the tissue every 56 mm. present in a 1600 mm region of muscle from each of 11 rhesus monkeys. A total of 283 abnormalities was observed for all animals and these were followed along their length. The observed age-associated increase in ETS abnormality length, suggests that, with time, mitochondrial abnormalities expand along the muscle fiber encompassing longer portions of the muscle fiber in the oldest animals. The incidence of mitochondrial abnormality abundance and length increases with age. 4.3. Transition region phenotype is not associated with deletion genotype Different cellular phenotypes of mitochondrial abnormalities from myopathy patients and aging individuals have been described. Some investigators report enzymatic abnormalities focusing on a single Žeither SDH or COX. enzyme activity either in aging w16,19,32,36x or myopathies w17,34,39,43,44x. Others have characterized both enzymatic activities in the same fiber and report SDHqqrCOXy w14,15, 22,33,37,39,40,45–47x, SDHqqrCOX normal w15,48x and COXyrSDH normal w15,17,22,23,35,37,39,40, 43,44,47,49x in myopathy patients. These phenotypes, SDHqqrCOXy w14,15,33x, SDHqqr COX normal w15x and COXyrSDH normal w15,35x, have also been observed in older individuals. It has been proposed that these phenotypes may be due to

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

the specific region of the genome being lost w39,40,42,48x. For example, a COXy phenotype would be caused by the loss of one of the three mitochondrial-encoded COX subunits in the deletion event, while fibers displaying a SDHqqrCOX normal phenotype could be the result of a deletion of the genome that does not include the COX subunits. Although mtDNA deletions were localized to ETS abnormal regions of fibers, the portion of the genome deleted was not associated with the phenotype of the transition region. 4.4. Different ETS abnormal phenotypes with age and fiber type We found COXyrSDHqq fibers Ži.e., the ‘‘ragged red fiber’’ phenotype, RRF. only in old animals. Even though ETS abnormal regions were not SDHqq in young adult and middle-aged animals, we observed different phenotypes between these two age groups. Analysis along the length of the muscle allowed for the identification of increases in SDH activity in COXy regions of fibers from middle-aged but not of young adult animals. The age-associated increase in SDH activity in abnormal regions of fibers is in accord with our hypothesis of a nuclear response to the ETS abnormalities secondary to a deficit in cellular energy. These observations suggest that there is a continuous nuclear response and, thus, we see a higher accumulation of mitochondria Ži.e., as demonstrated by the higher SDH activity. in older animals, also suggesting that this process may take many years. We found that although both skeletal muscle fiber types Žtype 1 and type 2. can be affected by mitochondrial abnormalities, these abnormalities occurred preferentially in type 2 fibers. In contrast, some investigators find RRF to be preferentially type 1 w17,36,37,39,45,48–50x or type 2a w37x fibers in muscle from myopathy patients or type 1 fibers in muscle from aging individuals w36x. In addition, we found that type 1 fibers generally became SDHqq, while type 2 fibers showed different increases in SDH activity Žusually associated to abnormality length., some of them showing similar levels as type 1 abnormal fibers. Our findings emphasize the importance of screening for both COX and SDH activities. If only SDHqq ŽRRF. are screened, it is likely

135

that these will be type 1 fibers, thus, underestimating the number of fibers affected by mitochondrial abnormalities. Abnormal regions of fibers are characterized by the presence of both deleted mtDNA and WT mtDNA, a phenomenon termed heteroplasmy. In the present study, we used RNA probes to detect levels of mitochondrial transcripts. Similar results are obtained when using probes to detect the abundance of WT and deleted mtDNA compared to the abundance of their transcripts w22,40,42x. We found an increase in transcripts from deleted genomes in abnormal regions of fibers. This increase was higher in type 1 fibers as well as in older animals. Although several reports indicate that deleted genomes or their transcripts are the most abundant species in muscle fibers from myopathy patients w40,47,51x, contradictory data exist on the abundance of WT mtDNA and mtRNA within abnormal regions of fibers. Some investigators have found decreased levels of WT mtDNA w40,47,51x or mtRNA w39,42x. Others have reported normal WT mtDNA w43x or mtRNA w23,44x levels in COXy and COXyrSDHqq. Type 2b fibers are characterized by having fewer mitochondria than type 1 fibers w52x, if a mtDNA deletion occurs in a type 2 fiber, fewer deleted genomes are required to overcome wild type ŽWT. genomes. Furthermore, due to the greater abundance of transcripts in normal regions of type 1 fibers, these should have a greater accumulation of transcripts in abnormal regions. In fact, we found that abnormal regions of type 2 fibers have almost no detectable transcripts from WT genomes compared to abnormal regions of type 1 fibers. Our results agree with those of Carrier et al. w44x and Prelle et al. w47x who found greater reactivity with a probe to WT and deleted mtDNA in RRFs Žpresumably type 1. than in COXy fibers without abnormal mitochondrial accumulation Žpresumably type 2.. We observed that ETS abnormal regions ŽCOXyrSDHqq . are flanked on both sides by a transition region that is either COXy or SDHqq depending on fiber type. Transition regions have been previously identified in serial skeletal muscle sections from myopathy patients in which a gradual decrease in COX activity was found w22x. Furthermore, Moraes et al. w40x characterized longitudinally oriented sections of muscle and found COXy seg-

136

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

ments contained within longer RRF segments indicating that COX positive regions are found in the transition between RRF and normal fiber segments. In studies of single cross-sections, one can identify a fiber in the transition region or in the region where both enzymatic activities are deficient. Therefore, depending on the fiber type as well as the specific region of the fiber being analyzed, the following abnormal cellular phenotypes could be observed: C O X yr SD H n o rm a l , C O X n o rm a l r SD H qq or COXyrSDHqq. 4.5. Intrafiber atrophy is associated with ETS abnormal regions We observed a decrease in CSA in some ETS abnormal regions of fibers in old animals suggesting that fiber atrophy may be a result of mitochondrial abnormalities. This atrophy occurred within the ETS abnormal region of the affected fiber. A number of the ETS abnormal fibers declined in cross-sectional area to a level at which they were no longer detectable in subsequent sections. Lexell et al. w53x performed a study of whole human vastus lateralis cross-sections and found that both the muscle area and the total fiber number decreased by ; 40% in people from 20 to 80 years of age. Interestingly, the percentage of type 1 fibers was maintained at 50% of all fibers. This situation could arise from either a conversion of type 2 to type 1 fibers andror the preferential loss of type 2 fibers. In addition, the reduction, with age, in fiber CSA in human vastus lateralis appears to be greater for type 2 than for type 1 fibers w54,55x. We found that type 2 fibers are affected preferentially by mitochondrial abnormalities suggesting that fiber atrophy associated with mitochondrial abnormalities may be contributing to the age-related decrease in muscle mass.

5. Conclusion Taken together, these data support the following hypothesized etiology of age-associated mitochondrial abnormalities in skeletal muscle. A large deletion event in a single mitochondrial genome initiates this process. The deleted genome may have a replicative advantage, due to its smaller size, that causes it

to predominate in a region of the fiber leading to an enzymatic deficiency ŽCOXy. . The affected region accumulates mitochondria to compensate for the respiratory deficiency ŽSDHqq .. With time, mitochondria containing deleted genomes progress along the length of the fiber generating longer ETS abnormal regions. Respiratory deficiency would eventually lead to fiber atrophy. Our in situ hybridization studies suggest that the mtDNA deletion is clonal within an ETS abnormal region. In addition, the ETS abnormality formation appears to be a continuous process. First, there is an increase with age in the abundance of fibers exhibiting ETS abnormalities. Second, ETS abnormal regions of different lengths were found within individual animals and long ETS abnormal regions were only found in older animals. In addition, the ETS abnormal phenotype varied with age, with the older animals showing the greatest increases in SDH activity, indicative of a nuclear response. Finally, decreases in cross-sectional area were associated with ETS abnormal regions of older animals. These data are suggestive of a continuous process taking place over many years in rhesus monkey muscle in which mtDNA deletions cause ETS abnormalities and, eventually, fiber atrophy and loss.

Acknowledgements This work was supported by grants RO1 AG11604 ŽJ.M.A.., PO1 AG11915 ŽR.W.. from the National Institutes of Health, and AFAR Scholarship for Research in the Biology of Aging ŽM.E.L... This is publication number 39-029 from the Wisconsin Regional Primate Research Center and 00-06 from Madison VA GRECC.

References w1x D. Harman, Free radical theory of aging: consequences of mitochondrial aging, Age 6 Ž1983. 86–94. w2x D. Harman, Aging: a theory based on free radical and radiation chemistry, J. Gerontol. 11 Ž1956. 298–300. w3x J. Miquel, A.C. Economos, J. Fleming, J.E. Johnson Jr., Mitochondrial role in cell aging, Exp. Gerontol. 15 Ž1980. 575–591. w4x G.A. Cortopassi, N. Arnheim, Detection of a specific mito-

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138

w5x

w6x

w7x

w8x

w9x

w10x

w11x

w12x

w13x

w14x

w15x

w16x

w17x

w18x

chondrial DNA deletion in tissues of older humans, Nucleic Acids Res. 18 Ž1990. 6927–6933. A.W. Linnane, A. Baumer, R.J. Maxwell, H. Preston, C. Zhang, S. Marzuki, Mitochondrial gene mutation: the ageing process and degenerative diseases, Biochem. Int. 22 Ž1990. 1067–1076. G.A. Cortopassi, D. Shibata, N.-W. Soong, N. Arnheim, A pattern of accumulation of a somatic deletion of mitochondrial DNA in aging human tissues, Proc. Natl. Acad. Sci. U. S. A. 89 Ž1992. 7370–7374. S.S. Chung, R. Weindruch, S.R. Schwarze, D.I. McKenzie, J.M. Aiken, Multiple age-associated mitochondrial DNA deletions in skeletal muscle of mice, Aging ŽMilano. 6 Ž1994. 193–200. S.M. Tanhauser, P.J. Laipis, Multiple deletions are detectable in mitochondrial DNA of aging mice, J. Biol. Chem. 270 Ž1995. 24769–24775. G.C. Van Tuyle, J.P. Gudikote, V.R. Hurt, B.B. Miller, C.A. Moore, Multiple, large deletions in rat mitochondrial DNA: evidence for a major hot spot, Mutat. Res. 349 Ž1996. 95–107. C.M. Lee, S.S. Chung, K.J.M.R. Weindruch, J.M. Aiken, Multiple mitochondrial DNA deletions associated with age in skeletal muscle of rhesus monkeys, J. Gerontol. Biol. Sci. 48 Ž1993. B201–B205. C.M. Lee, P. Eimon, R. Weindruch, J.M. Aiken, Direct repeat sequences are not required at the breakpoints of age-associated mitochondrial DNA deletions in rhesus monkeys, Mech. Ageing Dev. 75 Ž1994. 69–79. C.M. Lee, R. Weindruch, J.M. Aiken, Age-associated alterations of the mitochondrial genome, Free Radical Biol. Med. 22 Ž1997. 1259–1269. S.R. Schwarze, C.M. Lee, S.S. Chung, E.B. Roecker, R. Weindruch, J.M. Aiken, High levels of mitochondrial DNA deletions in skeletal muscle of old rhesus monkeys, Mech. Ageing Dev. 83 Ž1995. 91–101. J. Muller-Hocker, P. Seibel, K. Schneiderbanger, B. Kadenbach, Different in situ hybridization patterns of mitochondrial DNA in cytochrome c oxidase-deficient extraocular muscle fibers in the elderly, Virchows Archiv. A: Pathol Anat. 422 Ž1993. 7–15. Z. Rifai, S. Welle, C. Kamp, C.A. Thornton, Ragged red fibers in normal aging and inflammatory myopathy, Ann. Neurol. 37 Ž1995. 24–29. J. Muller-Hocker, Cytochrome c oxidase deficient fibres in the limb and diaphragm of man without muscular disease: an age-related alteration, J. Neurol. Sci. 100 Ž1990. 14–21. J. Muller-Hocker, D. Pongratz, G. Hubner, Focal deficiency of cytochrome-c-oxidase in skeletal muscle of patients with progressive external ophthalmoplegia, Virchows Arch. A: Pathol. Anat. 402 Ž1983. 61–71. N. Bresolin, M. Moggio, L. Bet, A. Gallanti, A. Prelle, E. Nobile-Orazio, L. Adobbati, C. Ferrante, G. Pellegrini, G. Scarlato, Progressive cytochrome c oxidase deficiency in a case of Kearns–Sayre syndrome: morphological, immunological, and biochemical studies in muscle biopsies and autopsy tissue, Ann. Neurol. 21 Ž1987. 564–572.

137

w19x E.J. Brierley, M.A. Johnson, R.N. Lightowlers, O.F.W. James, D.M. Turnbull, Role of mitochondrial DNA mutations in human aging: implications for the central nervous system and muscle, Ann. Neurol. 43 Ž1998. 217–223. w20x I.J. Holt, A.E. Harding, J.A. Morgan-Hughes, Deletions of muscle mitochondrial DNA in patients with mitochondrial myopathies, Nature 331 Ž1988. 717–719. w21x D.C. Wallace, Mitochondrial diseases in man and mouse, Science 283 Ž1999. 1482–1488. w22x S. Mita, B. Schmidt, E.A. Schon, S. DiMauro, E. Bonilla, Detection of ‘‘deleted’’ mitochondrial genomes in cytochrome-c oxidase-deficient muscle fibers of a patient with Kearns–Sayre syndrome, Proc. Natl. Acad. Sci. U. S. A. 86 Ž1989. 9509–9513. w23x E.A. Shoubridge, G. Karpati, K.E.M. Hastings, Deletion mutants are functionally dominant over wild-type mitochondrial genomes in skeletal muscle fiber segments in mitochondrial disease, Cell 62 Ž1990. 43–49. w24x C.M. Lee, M.E. Lopez, R. Weindruch, J.M. Aiken, Association of age-related mitochondrial abnormalities with skeletal muscle fiber atrophy, Free Radical Biol. Med. 25 Ž1998. 964–972. w25x A.M. Seligman, M.J. Karnovsky, H.L. Wasserkrug, J.S. Hanker, Nondroplet ultrastructural demonstration of cytochrome oxidase activity with a polymerizing osmiophilic reagent, diaminobenzidine ŽDAB., J. Cell Biol. 38 Ž1968. 1–14. w26x V. Dubowitz, Histological and Histochemical Stains and Reactions, Bailliere Tindall, London, 1985, 720 pp. w27x R.W. Ogilvie, D.L. Feeback, A metachromatic dye-ATPase method for the simultaneous identification of skeletal muscle fiber types I, IIA, IIB and IIC, Stain Technol. 65 Ž1990. 231–241. w28x S. Anderson, A.T. Bankier, B.G. Barrell, M.H.L. de Bruijn, A.R. Coulson, J. Drouin, I.C. Eperon, D.P. Nierlich, B.A. Roe, F. Sanger, P.H. Schreier, A.J.H. Smith, R. Staden, I.G. Young, Sequence and organization of the human mitochondrial genome, Nature 290 Ž1981. 457–465. w29x D.R. Cox, Regression models and life-tables Žwith discussion., J. R. Stat. Soc. Ser. B Methodol. 34 Ž1972. 187–220. w30x T.L. Wickiewicz, R.R. Roy, P.L. Powell, V.R. Edgerton, Muscle architecture of the human lower limb, Clin. Orthop. Rel. Res. 179 Ž1983. 275–283. w31x R.R. Roy, S.C. Bodine-Fowler, J. Kim, N. Haque, D. de Leon, W. Rudolph, V.R. Edgerton, Architectural and fiber type distribution properties of selected rhesus leg muscles: feasibility of multiple independent biopsies, Acta Anat. 140 Ž1991. 350–356. w32x S.J. Oh, T.D. Thomas, H.R. Kuruoglu, Ragged red fibers and aging, Ann. Neurol. 32 Ž1992. 253, ŽAbstract.. w33x E.J. Brierley, M.A. Johnson, O.F.W. James, D.M. Turnbull, Effect of physical activity and age on mitochondrial function, Q. J. Med. 89 Ž1996. 251–258. w34x P. Chariot, E. Ruet, F.-J. Authier, D. Labes, F. Poron, R. Gherardi, Cytochrome c oxidase deficiencies in the muscle of patients with inflammatory myopathies, Acta Neuropathol. 91 Ž1996. 530–536. w35x J. Muller-Hocker, K. Schneiderbanger, F.H. Stefani, B.

138

w36x

w37x

w38x

w39x

w40x

w41x

w42x

w43x

w44x

w45x

M.E. Lopez et al.r Mutation Research 452 (2000) 123–138 Kadenbach, Progressive loss of cytochrome c oxidase in the human extraocular muscles in ageing — a cytochemical–immunohistochemical study, Mutat. Res. 275 Ž1992. 115–124. E. Byrne, X. Dennett, Respiratory chain failure in adult muscle fibres: relationship with ageing and possible implications for the neuronal pool, Mutat. Res. 275 Ž1992. 125–131. M. Yamamoto, I. Nonaka, Skeletal muscle pathology in chronic progressive external ophthalmoplegia with ragged-red fibers, Acta Neuropathol. 76 Ž1988. 558–563. K. Haginoya, S. Miyabayashi, K. Iinuma, K. Tada, Mosaicism of mitochondria in mitochondrial myopathy: an electronmicroscopic analysis of cytochrome c oxidase, Acta Neuropathol. 80 Ž1990. 642–648. S.R. Hammans, M.G. Sweeney, D.A.G. Wicks, J.A. Morgan-Hughes, A.E. Harding, A molecular genetic study of focal histochemical defects in mitochondrial encephalomyopathies, Brain 115 Ž1992. 343–365. C.T. Moraes, E. Ricci, V. Petruzzella, S. Shanske, S. DiMauro, E.A. Schon, E. Bonilla, Molecular analysis of the muscle pathology associated with mitochondrial DNA deletions, Nat. Genet. 1 Ž1992. 359–367. T. Matsuoka, Y.-I. Goto, I. Nonaka, ‘‘All-or-none’’ cytochrome c oxidase positivity in mitochondria in chronic progressive external ophthalmoplegia: an ultrastructural–cytochemical study, Muscle Nerve 16 Ž1993. 206–209. A. Oldfors, N.-G. Larsson, E. Holme, M. Tulinius, B. Kadenbach, M. Droste, Mitochondrial DNA deletions and cytochrome c oxidase deficiency in muscle fibres, J. Neurol. Sci. 110 Ž1992. 169–177. S. Collins, C. Rudduck, S. Marzuki, X. Dennett, E. Byrne, Mitochondrial genome distribution in histochemically cytochrome c oxidase-negative muscle fibers in patients with a mixture of deleted and wild type mitochondrial DNA, Biochim. Biophys. Acta. 1097 Ž1991. 309–317. H. Carrier, B. Burt-Pichat, F. Flocard, N. Guffon, B. Mousson, R. Dumoulin, C. Godinot, Molecular histology of mitochondrial and nuclear transcripts in the muscle of patients harbouring a single mitochondrial DNA deletion, Acta Neuropathol. 91 Ž1996. 104–111. M.A. Johnson, B. Kadenbach, M. Droste, S.L. Old, D.M. Turnbull, Immunocytochemical studies of cytochrome oxidase subunits in skeletal muscle of patients with partial

w46x

w47x

w48x

w49x

w50x

w51x

w52x

w53x

w54x

w55x

cytochrome oxidase deficiencies, J. Neurol. Sci. 87 Ž1988. 75–90. S. Collins, D.X.E. Byrne, S. Marzuki, Chronic progressive external ophthalmoplegia in patients with large heteroplasmic mitochondrial DNA deletions: an immunocytochemical study, Acta Neuropathol. 82 Ž1991. 185–192. A. Prelle, G. Fagiolari, N. Checcarelli, M. Moggio, A. Battistel, G.P. Comi, P. Bazzi, A. Bordoni, M. Zeviani, G. Scarlato, Mitochondrial myopathy: correlation between oxidative defect and mitochondrial DNA deletions at single fiber level, Acta Neuropathol. 87 Ž1994. 371–376. S. Collins, E. Byrne, X. Dennett, Contrasting histochemical features of various mitochondrial syndromes, Acta Neurol. Scand. 91 Ž1995. 287–293. H. Reichmann, Enzyme activity measured in single muscle fibers in partial cytochrome c oxidase deficiency, Neurology 38 Ž1988. 244–249. U.A. Walker, E.A. Schon, Neurotrophin-4 is up-regulated in ragged-red fibers associated with pathogenic mitochondrial DNA mutations, Ann. Neurol. 43 Ž1998. 536–540. M. Sciacco, E. Bonilla, E.A. Schon, S. DiMauro, C.T. Moraes, Distribution of wild-type and common deletion forms of mtDNA in normal and respiration-deficient muscle fibers from patients with mitochondrial myopathy, Hum. Mol. Gen. 3 Ž1994. 13–19. T. Ogata, Y. Yamasaki, Scanning electron-microscopic studies on the three-dimensional structure of mitochondria in the mammalian red, white and intermediate muscle fibers, Cell Tissue Res. 241 Ž1985. 251–256. J. Lexell, C.C. Taylor, M. Sjostrom, What is the cause of the ageing atrophy? Total number, size and proportion of different fiber types studied in whole vastus lateralis muscle from 15- to 83-year old men, J. Neurol. Sci. 84 Ž1988. 275–294. D.N. Proctor, W.E. Sinning, J.M. Walro, G.C. Sieck, P.W.R. Lemon, Oxidative capacity of human muscle fiber types: effects of age and training status, J. Appl. Physiol. 78 Ž1995. 2033–2038. T. Hortobagyi, D. Zheng, M. Weidner, N.J. Lambert, S. Westbrook, J.A. Houmard, The influence of aging on muscle strength and muscle fiber characteristics with special reference to eccentric strength, J. Gerontol. Biol. Sci. 50A Ž1995. B399–B406.