Thermal sensitivity of mitochondrial functions in permeabilized muscle fibers from two populations...

transcript

doi:10.1152/ajpregu.00542.2010 301:R48-R59, 2011. First published 30 March 2011;Am J Physiol Regul Integr Comp Physiol

Nicolas Pichaud, J. William O. Ballard, Robert M. Tanguay and Pierre U. BlierDrosophila simulans with divergent mitotypespermeabilized muscle fibers from two populations of Thermal sensitivity of mitochondrial functions in

You might find this additional info useful...

54 articles, 15 of which can be accessed free at:This article cites http://ajpregu.physiology.org/content/301/1/R48.full.html#ref-list-1

including high resolution figures, can be found at:Updated information and services http://ajpregu.physiology.org/content/301/1/R48.full.html

can be found at:and Comparative PhysiologyAmerican Journal of Physiology - Regulatory, Integrativeabout Additional material and information

http://www.the-aps.org/publications/ajpregu

This infomation is current as of January 8, 2012.

ISSN: 0363-6119, ESSN: 1522-1490. Visit our website at http://www.the-aps.org/.Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2011 by the American Physiological Society. ranging from molecules to humans, including clinical investigations. It is published 12 times a year (monthly) by the Americanilluminate normal or abnormal regulation and integration of physiological mechanisms at all levels of biological organization,

publishes original investigations thatAmerican Journal of Physiology - Regulatory, Integrative and Comparative Physiology

on January 8, 2012ajpregu.physiology.org

nloaded from

CALL FOR PAPERS Mitochondrial Function/Dysfunction in Health and Disease

Thermal sensitivity of mitochondrial functions in permeabilized muscle fibersfrom two populations of Drosophila simulans with divergent mitotypes

Nicolas Pichaud,1 J. William O. Ballard,2 Robert M. Tanguay,3 and Pierre U. Blier1

1Laboratoire de Biologie Intégrative, Département de Biologie, Université du Québec a Rimouski, Rimouski, 3Laboratoire deGénétique Cellulaire et Développementale, Département de Biologie Moléculaire, Biochimie Médicale, et Pathologie, Institutde Biologie Intégrative et des Systèmes, Université Laval, Laval, Quebec, Canada; and 2School of Biotechnology andBiomolecular Sciences, University of New South Wales, Sydney, New South Wales, Australia

Submitted 18 August 2010; accepted in final form 26 March 2011

Pichaud N, Ballard JWO, Tanguay RM, Blier PU. Thermal sensi-tivity of mitochondrial functions in permeabilized muscle fibers from twopopulations of Drosophila simulans with divergent mitotypes. Am J PhysiolRegul Integr Comp Physiol 301: R48–R59, 2011. First published March 30,2011; doi:10.1152/ajpregu.00542.2010.—In ectotherms, the external tem-perature is experienced by the mitochondria, and the mitochondrialrespiration of different genotypes is likely to change as a result. Usinghigh-resolution respirometry with permeabilized fibers (an in situapproach), we tried to identify differences in mitochondrial perfor-mance and thermal sensitivity of two Drosophila simulans popula-tions with two different mitochondrial types (siII and siIII) andgeographical distributions. Maximal state 3 respiration rates obtainedwith electrons converging at the Q junction of the electron transportsystem (ETS) differed between the mitotypes at 24°C. Catalyticcapacities were higher in flies harboring siII than in those harboringsiIII mitochondrial DNA (2,129 vs. 1,390 pmol O2·s�1·mg protein�1).The cytochrome c oxidase activity was also higher in siII than siIIIflies (3,712 vs. 2,688 pmol O2·s�1·mg protein�1). The higher catalyticcapacity detected in the siII mitotype could provide an advantage interms of intensity of aerobic activity, endurance, or both, if theintensity of exercise that can be aerobically performed is partlydictated by the aerobic capacity of the tissue. Moreover, thermalsensitivity results showed that even if temperature affects the catalyticcapacity of the different enzymes of the ETS, both mitotypes revealedhigh tolerance to temperature variation. Previous in vitro study failedto detect any consistent functional mitochondrial differences betweenthe same mitotypes. We conclude that the in situ approach is moresensitive and that the ETS is a robust system in terms of functionaland regulatory properties across a wide range of temperatures.

metabolism; mitochondrial DNA; mitochondrial respiration; temper-ature

THE CONTROL OF OXIDATIVE PHOSPHORYLATION (OxPhos) and reg-ulation of mitochondrial respiration are thought to be shared bydifferent complexes: the phosphorylation system, the electrontransport system (ETS), and the reactions that provide sub-strates and electrons to the system (14). Since all these pro-cesses are expected to be affected differently by temperature,the distribution of control strength may also be altered bytemperature (18). Investigation of catalytic capacities of thedifferent enzymes of the ETS at different temperatures is

therefore important to understand management of mitochon-drial respiration control and, plausibly, adaptation to environ-mental temperature. The goal of this study is to identifydifferences in mitochondrial performance and thermal sensi-tivities associated with mitochondrial DNA (mtDNA) haplo-type divergences in the fly Drosophila simulans. This is animportant biological question, because it has been suggestedthat selection on mtDNA might lead to haplotypes adapted todifferent environments (12, 23).

It has been suggested that thermally induced changes incytochrome c oxidase (COX) activity act on regulation ofrespiration mainly through impact on the redox state of ETS(11). For example, in Salvelinus fontinalis, it has been shownthat the COX maximal activity (Vmax) measured at differenttemperatures should reach a significantly high level of inhibitionbefore impairment of mitochondrial respiration at each tempera-ture tested (11). This excess of the catalytic capacity of COX wassuggested to be required for the adequate functioning of mito-chondria at different temperatures encountered by the species tomaintain the ETS mainly in an oxidized state and, consequently,to ensure a sharp thermodynamic gradient in the ETS under mostthermal conditions (11). It is therefore relevant to measure thisapparent COX excess capacity to investigate the impact oftemperature on the regulatory properties of respiration throughthe redox state of the ETS. This can be done by calculating the“biochemical threshold effect” (for a review see Ref. 45) usingincreasing concentrations of a specific inhibitor of a specificcomplex (complex IV in our case) while measuring individualactivity of this complex and the OxPhos pathway in mitochon-dria respiring at maximum capacity at different temperatures.In the case of high electron input in the ETS using multiplesubstrates (pyruvate � malate � proline � sn-glycerol 3-phos-phate), our laboratory showed that an important threshold and,subsequently, a high apparent COX excess capacity werepresent at 12°C but were nonexistent at higher temperatures(42). These differences in biochemical threshold and apparentCOX excess capacity were proposed to be associated withalteration of the distribution of control along the ETS anddehydrogenase processes with temperature changes (42). In-deed, another approach to relate the impact of temperature oncatalytic capacity of COX to the respiration rate and theactivity of the ETS is assessment of the flux control coefficient(Ci) of COX at these temperatures. Ci for a given enzyme isdefined as the change in pathway flux upon a small change inenzyme activity, and it follows that enzymes with a high Ci are

Address for reprint requests and other correspondence: P. U. Blier, Labo-ratoire de biologie intégrative, Département de biologie, Université du Québeca Rimouski, 300 allée des Ursulines, Rimouski, QC, Canada G5L 3A1 (e-mail:pierre_blier@uqar.qc.ca).

Am J Physiol Regul Integr Comp Physiol 301: R48–R59, 2011.First published March 30, 2011; doi:10.1152/ajpregu.00542.2010.

nloaded from

significant in regulation of metabolism (21, 32, 38). In aprevious study, we showed that the level of control of mito-chondrial respiration by COX increases significantly with in-creasing temperatures, confirming the importance of this pa-rameter in mitochondrial respiration during temperature-in-duced changes (42).

We sought to identify differences in mitochondrial perfor-mance and thermal sensitivities associated with mtDNA hap-lotype divergences in the fly D. simulans. D. simulans harborsthree geographically distinct, subdivided haplogroups (siI, siII,and siIII) with �3% interhaplogroup divergence (3, 4, 6) butlacks any nuclear subdivision at any nuclear-encoded locitested to date (31, 39). The siII and siIII mitotypes live insympatry in Kenya, where the frequency of the siIII type is�40% (5). In four wild-caught siII and four siIII fly lines, lowamino acid variation was observed within mitotypes in a 4.5-kbregion from position 1450 to position 5983 of the mtDNAgenome. In contrast, large divergences were noted between themitotypes (2.04 � 10�2% nucleotide divergence and 1.21 �10�2% amino acid divergence in the same region) (7). Specif-ically, there are 1 � 10�2 average pair-wise amino aciddifferences at the level of complex I, 0.3 � 10�2 at the level ofcomplex III, and 0.39 � 10�2 at the level of complex IVbetween both mitotypes (4). Recently, it has been shown that,at the level of complex IV, which has its three larger subunits(COI, COII, and COIII) encoded by mtDNA, there are 76synonymous and 2 nonsynonymous fixed differences in COIIbetween flies harboring siII and those harboring siIII mtDNA(7, 8, 39). One nonsynonymous change occurs in a loop towardthe intermembrane space and causes an isoleucine-to-threoninereplacement (7), and the second change causes an isoleucine-to-valine replacement in a �8 strand of the cupredoxin foldtoward the intermembrane space (8). Therefore, there aresignificant probabilities that these amino acid substitutionsmight be important in terms of physicochemical properties ofthe peptides. Consequently, complex IV is an ETS locus ofinterest to delineate the functional mitochondrial differencesassociated with mtDNA between the two mitotypes.

The thermal sensitivity of D. simulans mitotypes has beenaddressed in a previous study using isolated mitochondria (42).This approach is valuable for assessing mitochondrial func-tional integrity and maximal capacity of OxPhos, as well asimport of mitochondrial proteins (1, 9–11, 14, 20, 28, 36, 42,50). However, it may be biased under some circumstances (15,19, 34, 40, 44, 46). Kuznetsov et al. (34) emphasized thatproperties of mitochondria may differ in vivo and in vitro;therefore, an in situ approach could be more relevant thanstudying isolated mitochondria in some circumstances. Here,we develop an in situ approach to study mitochondrial respi-ration in permeabilized fibers and then compare the results withour previous findings from studies of isolated mitochondria(42). This in situ approach differs from our previous experi-ment using isolated mitochondria (42), since we measured themitochondrial respiration during high electron input throughthe ETS using multiple substrate-uncoupler-inhibitor titrationprotocols (25) adapted to Drosophila. In the present study, weused a more complex combination of substrates than that usedin our previous study (42) to maximally reduce the differentcomplexes of the ETS, and we attempted to maintain thenatural cellular conditions and minimize stress on mitochon-dria by working on permeabilized fibers. We measured mito-

chondrial respiration at 12°C, 18°C, 24°C (considered anoptimal temperature), and 28°C in permeabilized fibers. Thethermal sensitivity of mitochondrial respiration was measuredwith different substrate combinations of complex I, complex II,and complex III through the Q pool and their respectiveinhibitors. We also determined functional integrity of the outermitochondrial membrane and thermal sensitivity of uncoupledrespiration, the biochemical threshold and the apparent excesscapacity of COX at high convergent pathway flux, and the Ci

of COX using inhibitor titration experiments to compare over-all mitochondrial performance at different temperatures of bothmitotypes. Pyruvate dehydrogenase (PDH) activity was alsomeasured to investigate electron entry in the ETS at the levelof complex I.

MATERIALS AND METHODS

Fly Mitotypes

D. simulans (STURTEVANT 1919) siII and siIII, both collected inKenya, were used. For each mitotype, four isofemale lines (2KY0412,2KY0415, 2KY0418, and 2KY0421 for siII and 3KY0410, 3KY0412,3KY0414, and 3KY0420 for siIII) were reared from flies collected inNairobi in November 2004 (7). The mtDNA haplotype was deter-mined using allele-specific PCR, as described previously for eachisofemale line (17), and Wolbachia infection was tested using con-served 16S rDNA primers (29). According to Ballard et al. (7), thereis only one nonsynonymous change within siII lines and one nonsyn-onymous change within siIII lines over 5,500 bp of mtDNA analyzed.In contrast, large divergences were noticed between the mitotypes(2.04 � 10�2% nucleotide divergence and 1.21 � 10�2% amino aciddivergence in the same region). PCR assays showed that all isofemalelines belonged to the corresponding mtDNA haplotype and were notinfected by Wolbachia.

Flies were reared at constant temperature (24.0 � 0.1°C), humidity(50% relative humidity), diurnal cycle (12:12-h light-dark), and den-sity (�100 flies in 25 ml of standard cornmeal medium), as describedpreviously (42). Flies were reared in standard cornmeal mediumcontaining a mixture of 10 g of agar-agar, 12 g of sugar, 54 g of driedyeast, and 106 g of cornmeal flour dissolved in 2 liters of tap water,and propionic acid (8 ml) and methyl-4-hydroxybenzoate [10% (wt/vol) in ethanol (32 ml)] were added to the mixture to avoid mite andmold contamination. Flies used for experiments were 10-day-oldmales. To avoid fitness problems associated with aging, the parents ofthe experimental flies were �14 days old. Each population was rearedin two different incubators (A and B) to allow replicates of experi-mental lines and to avoid any “incubator” effect. Flies used for fiberpermeabilization and enzymatic analyses were weighed individuallywith an electronic balance (Mettler-Toledo, Columbus, OH) with aresolution of 0.1 �g.

Preparation of Permeabilized Muscle Fibers

All procedures were conducted at 4°C. Flight muscles from four oreight thoraxes (1 for 28°C or 2 for 12, 18, and 24°C from each of the4 isofemale lines of the same mtDNA type) of 10-day-old D. simulanswere taken for fiber permeabilization. Thoraxes were separated fromabdomens and heads and then placed onto a petri dish with 2 ml ofrelaxing solution (BIOPS) (37, 52) containing 2.77 mM CaK2EGTA,7.23 mM K2EGTA, 5.77 mM Na2ATP, 6.56 mM MgCl2, 20 mMtaurine, 15 mM Na2phosphocreatine, 20 mM imidazole, 0.5 mMdithiothreitol, and 50 mM K-MES, pH 7.1. Individual fiber bundleswere separated with two pairs of sharp forceps, achieving thin musclefiber bundles. The fiber bundles were transferred to a petri dishcontaining 2 ml of BIOPS and 81.25 �g/ml saponin and gently mixedat 4°C for 30 min. Preliminary experiments optimizing the maximalrespiratory control ratio (RCR) at 24°C determined the saponin

R49MITOCHONDRIAL TEMPERATURE SENSITIVITY

AJP-Regul Integr Comp Physiol • VOL 301 • JULY 2011 • www.ajpregu.org

nloaded from

concentration (n 3 for each concentration: RCR 3.3, 6.9, 7.5, and6.2 for 23.75, 50, 81.25, and 110 �g/ml, respectively). Fiber bundleswere then rinsed at 4°C for 10 min in respiration medium containing120 mM KCl, 5 mM KH2PO4, 3 mM HEPES, 1 mM EGTA, 1 mMMgCl2, and 0.2% BSA (wt/vol), pH 7.2, and immediately transferredto a respirometer (OXYGRAPH-2K, Oroboros Instruments, Inns-bruck, Austria) filled with air-saturated respiration medium. The O2

electrodes were calibrated with air-saturated respiration medium ateach experimental temperature, and zero-O2 measurements weretaken after addition of sodium dithionite. O2 solubility for mediumrespiration was calculated for the four experimental temperaturesaccording to Rasmussen and Rasmussen (43).

High-Resolution Respirometry

For high-resolution respirometry experiments, we used saturatingconcentrations of different substrates. We rationalized that even if itdoes not reflect the physiological input of electrons into the ETS and,consequently, the respiration rates, this approach is required to max-imize resolution and increase our ability to identify the locus ofdivergence between the two mitotypes.

O2 flux measurements with substrates and uncoupler. Respirationwas measured at 12°C, 18°C, 24°C (with permeabilized fibers from 8thoraxes), and 28°C (with permeabilized fibers from 4 thoraxes), andDatLab software (Oroboros Instruments) was used for data acquisitionand analysis. According to the temperature and the mitotype tested,mitochondrial respiration of 7–20 different preparations was averagedfor each incubator (19 and 12 preparations for siII and siIII, respec-tively, at 12°C; 16 and 20 preparations for siII and siIII, respectively,at 18°C; 13 and 19 preparations for siII and siIII, respectively, at24°C; and 18 and 7 preparations for siII and siIII, respectively, at28°C). After addition of pyruvate (10 mM), malate (10 mM), andL-proline (10 mM), fiber bundles were transferred into the respirationchambers to achieve state 2. The following substrates and uncouplerwere then sequentially added to the chamber: ADP [5 mM, to achievestate 3 respiration for complex I (CI)], cytochrome c from equine heart[10 �M, as an index of functional integrity of the outer mitochondrialmembrane (CIc)], succinate [10 mM, to reach state 3 respiration forcomplex I � complex II (CIc � CII)], sn-glycerol 3-phosphate [20mM, to achieve state 3 respiration for complex I � complex II �glycerol 3-phosphate dehydrogenase (CIc � CII � G3PDH)], and2,4-dinitrophenol [uncoupler, optimum concentration 5–15 �M, toreach maximum O2 flux (CIc � CII � G3PDH � U)]. According toGnaiger (27), this protocol allowed us to calculate RCR for complexI (CI/state 2), cytochrome c effect (CIc/CI), uncoupling control ratio(UCR, CIc � CII � G3PDH � U/CIc � CII � G3PDH), O2 flux (forCI, CIc, CIc � CII, CIc � CII � G3PDH, and CIc � CII � G3PDH � U),and substrate control ratio (SCR, ratios of fluxes in the same couplingstate for CIc and CIc � CII, with CIc � CII � G3PDH as thereference state). From this point, 1) inhibitors were sequentiallyintroduced to inhibit the overall mitochondrial respiration, and COXactivity was determined before sodium azide titration, or 2) mitochon-drial respiration inhibition was measured using sodium azide titration.

Inhibitors and COX activity. After mitochondria were uncoupled,the following inhibitors and substrates were introduced in the follow-ing order: rotenone [0.5 �M, inhibitor of complex I (CII � G3PDH �U)], malonate [5 mM, inhibitor of complex II (G3PDH � U)], andantimycin A (2.5 �M, inhibitor of complex III). The inhibition ofcomplexes I, II, and III allowed us to measure the residual O2

consumption (ROx) due to residual oxidative side reactions in per-meabilized fibers (27). N,N,N=,N=-tetramethyl-p-phenylenediamine(TMPD) � ascorbate (0.5 and 2 mM, respectively) were then addedto measure COX activity. After each introduction, O2 consumptionwas measured and chemical backgrounds (oxidation rates due toautoxidation of TMPD, ascorbate, and cytochrome c), as well asinstrumental backgrounds, were subtracted from the activity (24).These subtractions allowed us to calculate O2 flux (for CII � G3PDH �

U, G3PDH � U, ROx, and COX) and SCR (for CII � G3PDH � Uand G3PDH � U, with CIc � CII � G3PDH � U as the referencestate).

Azide titration experiment. After TMPD � ascorbate were intro-duced, sodium azide was progressively added at appropriate intervalsto inhibit COX activity. The concentrations used for titration were asfollows: 1, 2, 7, 12, 32, 52, 102, 152, 252, 352, 852, 1,352, 3,852, and8,852 �M or until maximum inhibition was achieved. Alternatively,after mitochondria were uncoupled, the same sodium azide titrationwas performed to progressively inhibit the maximum pathway flux.This titration of maximum pathway flux, as well as azide titration ofCOX activity, allowed us to construct plots of relative respiration rateagainst the percentage of inhibition of COX activity at the same azideconcentration (38, 53). The inhibition constant (Ki) was calculatedfrom transformed data using the Dickson linearization (33). As azideis a noncompetitive inhibitor, Ci was calculated as Ci �(dJ/J)/(dI/Ki), where J is the respiration flux, dJ is the decrement of respirationflux caused by increment of inhibitor addition, dI (between 0 and 50�M), and Ki is the inhibition constant for sodium azide (33).

Ratios of the different O2 fluxes, CIc to COX, CIc � CII to COX,and CIc � CII � G3PDH to COX, were also calculated to emphasizethe possible different proportions of each complex at the differenttemperatures tested.

Measurements of mitochondrial respiration were averaged for eachmitotype at each temperature and for each incubator. Data are ex-pressed as mean respiration rates in picomoles of O2 consumed persecond per milligram of protein for each mitotype at each temperatureand for each incubator.

Enzymatic Analyses

PDH activity was measured using a UV/VIS spectrophotometer(Lambda 11, Perkin Elmer, Foster City, CA) equipped with a ther-mostated cell holder and a circulating water bath. PDH was measuredfrom 16 whole flies (4 from each isofemale line of each mitotype)homogenized in 1,600 �l of ice-cold buffer containing 50 mMTris·HCl, 1 mM MgCl2, 0.1% Triton X-100, and 1 mg/ml BSA, pH7.8, and centrifuged for 3 min at 750 g. Each measurement was donein duplicate, and five different analyses were averaged for eachincubator, temperature, and mitotype. Enzymatic activities are ex-pressed in units per milligram of protein, where unit is 1 �mol ofsubstrate transformed per minute at each temperature tested (12, 18,24, and 28°C).

PDH activity was measured according to Thibeault et al. (49).NADH production was followed by coupling it to the reduction ofiodotetrazolium violet (INT). Supernatant (100 �l) was incubatedwith ice-cold buffer supplemented with 2.5 mM NAD, 0.5 mMEDTA, 0.1 mM coenzyme A, 0.1 mM oxalic acid, 0.6 mM INT, 6U/ml lipoamide dehydrogenase, and 0.2 mM thiamine pyrophosphate,pH 7.8. The reaction was initiated with the addition of 5 mM pyruvate(omitted from the control), and the change in absorbance was mea-sured for 5 min at 500 nm (INT ε500 15.4 ml·cm�1·�mol�1).

Protein Content

At the end of each mitochondrial respiration measurement, fiberbundles were removed and homogenized with a Tekmar homogenizer,and the homogenates were immediately frozen at �80°C for furtheranalyses. Total protein content was determined for homogenates fromfiber bundles and from whole flies in duplicate by the bicinchoninicacid method (47). Because of addition of cytochrome c duringexperiments and the presence of BSA in the ice-cold buffer and in therespiration medium, the protein content of the buffer was subtractedfrom the fiber preparations.

Temperature Coefficient Values

Temperature coefficient (Q10) values for mitochondrial functions(CI, CIc, CIc � CII, CIc � CII � G3PDH, CIc � CII � G3PDH �

R50 MITOCHONDRIAL TEMPERATURE SENSITIVITY

nloaded from

U, CII � G3PDH � U, G3PDH � U, COX activity, and PDH) werecalculated as Q10 (rate 2/rate 1)10/(t2�t1), where rate 1 is the meanrate of the parameter measured at temperature t1 and rate 2 is themean rate of the parameter measured at temperature t2. As thedifferent parameters were not measured on the same individuals, Q10

was calculated using the mean of each parameter measured at eachtemperature.

Chemicals

All chemicals were purchased from Sigma-Aldrich (Mississauga,ON, Canada).

Statistical Analyses

All statistical analyses were performed with SAS software (version9.1.3, SAS Institute, Cary, NC). O’Brien’s test was used to verifyhomogeneity of data. No differences between incubators were de-tected for any parameters measured; consequently, ANOVA with twoindependent variables (temperature and mitotype; the incubator vari-able was discarded) was performed using general linear models(GLMs) with the least-square means method for multiple comparisontests. This allowed us to determine any interactions between thevariables and the effect of each variable on the different parametersmeasured. Significance was defined at P � 0.05.

RESULTS

Data from incubators A and B were pooled, giving 14–40different preparations according to the temperature and mito-type tested (38 and 24 preparations for siII and siIII, respec-tively, at 12°C, 32 and 40 preparations at 18°C, 26 and 38preparations at 24°C, and 36 and 14 preparations at 28°C) and10 different homogenates for enzymatic analyses of eachmitotype at each temperature. Consequently, F and P valuesare presented only for the factors “temperature” and “mito-type” and their interaction (Table 1). Statistical values cited in

the text are those obtained with the least-square means methodto simplify presentation of results.

For O2 fluxes and consequent RCR, UCR, and SCR, ROxwas subtracted from initial O2 fluxes measured.

Fly Sampling

All lines from the two groups (siII and siIII) were pooledbecause of low intrahaplogroup variation (31). In the resultingpools of isofemale lines, the mean mass was similar betweenmitotypes or between incubators (0.685 � 0.070 mg for siIIincubator A, 0.674 � 0.061 mg for siII incubator B, 0.679 �0.079 mg for siIII incubator A, and 0.686 � 0.071 mg for siIIIincubator B).

RCRs were calculated for complex I as the ratio of CI tostate 2 (26), and results are presented in Fig. 1. RCRs obtainedusing substrates for complex I indicate a well-coupled respi-ration. No significant differences were detected between mito-types. RCRs were only influenced by temperature (Table 1)and were maximum at 24°C, but differences among tempera-tures were significant only for siII between 12 and 24°C (P �0.01) and between 24 and 28°C (P � 0.0001).

Cytochrome c Effect and UCR

To determine if mitochondrial integrity was disrupted, O2

flux was compared before and after cytochrome c injection.Ratios of CIc to CI showed little effect of cytochrome caddition on the mitochondrial respiration (CIc-to-CI ratioswere not significantly different from 1; results not shown),suggesting functional integrity of the outer mitochondrialmembranes. No significant differences between mitotypeswere observed, although a slight (but not significant) decreasewas detected from 12 to 28°C.

O2 flux was compared before and after the addition of anuncoupler to see if ATP synthase and adenine nucleotide

Fig. 1. Effect of temperature on respiratory control ratio (RCR CI/state 2)when fiber preparations from the 2 mitotypes, siII and siIII, were supplied withpyruvate � malate � L-proline (state 2) � ADP (CI). Values are means � SEfor 14–40 fiber preparations. Significance was set as P � 0.05. Letters denotedifferences between temperatures: a is statistically different from b and c, andb is statistically different from c.

Table 1. Results from ANOVA showing F ratios withvariables mitotype and temperature

Error (df)Mitotype(df 1)

Temperature(df 3)

Mitotype �Temperature

(df 3)

RCR 190 0.11 2.68* 0.15Respiration rates

CI 98 5.41 5.4* 1.79CIc 98 6.53* 3.75 1.53CIc � CII 98 6.16* 2.67 1.03CIc � CII � G3PDH 98 10.07* 2.52 1.09CIc � CII � G3PDH � U 98 10.38* 3.14* 1.29CII � G3PDH � U 98 7.43* 3.36* 1.13G3PDH � U 98 8.32* 2.17 1.21COX 98 5.4* 7.33* 0.72

SCRCIc 222 0.47 2.65* 1.01CIc � CII 222 0.33 1.67 0.3CII � G3PDH � U 103 3.55 11.88* 0.7G3PDH � U 103 0.76 6.31* 1.1

Metabolic controlKi 102 1.45 93.81* 0.17Ci 125 27.58* 157.87* 1.22

Enzymatic analysisPDH 65 0.03 26.17* 0.33

RCR, respiratory control ratio; CI and CII, complex I and II; G3PDH,glycerol 3-phosphate dehydrogenase; U, uncoupler; COX, cytochrome c oxi-dase; SCR, substrate control ratio; PDH, pyruvate dehydrogenase; Ki, inhibi-tion constant; Ci, flux control coefficient. *P � 0.05.

nloaded from

translocase (ANT) exerted a limitation on mitochondrial res-piration. No significant differences were detected betweenmitotypes or among temperatures, and all UCRs were close to1 (results not shown), which suggests that ATP synthesiscapacity (ATP synthase), as well as ANT, can support themaximum respiration rates observed with the experimentalsubstrates used. These results are consistent with previousresults on mitochondrial preparations (42).

Respiration Rates

Comparisons of respiration rates for the in situ approachwith those for the in vitro approach (42) showed that respira-tion rates were almost always lower in permeabilized fibersthan in isolated mitochondria (results not shown). GLMsshowed that mitotype and temperature influenced the respira-tion rates for CIc � CII � G3PDH � U, CII � G3PDH � U,and COX, while CI was only influenced by temperature andCIc, CIc � CII, CIc � CII � G3PDH, and G3PDH � U wereonly influenced by mitotypes (Table 1). At 24°C, significantdifferences were detected between mitotypes (Fig. 2) for CIc �CII � G3PDH, CIc � CII � G3PDH � U, CII � G3PDH �U, G3PDH � U, and COX (P 0.02, P 0.015, P 0.027,P 0.031, and P 0.036, respectively). A difference wasdetected between temperatures for complex IV in siIII, withCOX activity being significantly higher at 28°C than at other

temperatures (P � 0.012). In siII, all parameters were signif-icantly higher at 28°C than at 12 and 18°C (P � 0.0415).

SCR allowed us to determine the effect of each substrateaddition on mitochondrial respiration. For CIc and CIc � CII,the reference state was CIc � CII � G3PDH; for CII �G3PDH � U and G3PDH � U, the reference state was CIc �CII � G3PDH � U. Temperature was the only factor thatinfluenced SCR at the level of CIc, CII � G3PDH � U, andG3PDH � U (Table 1). Results for CIc and CIc � CII showedno significant SCR differences between mitotypes at any tem-perature (Fig. 3). SCR was lower in siII than siIII for CII �G3PDH � U, as well as for G3PDH � U, respiration rates at24°C (P 0.037 and P 0.048, respectively). Interestingly,significant increases in SCRs were observed when succinatewas added as substrate in siII at 12 and 18°C (P � 0.048) andin siIII at all temperatures (P � 0.042).

COX Excess, Biochemical Threshold, and Ci

The state 3-to-COX ratios for CIc � CII and for CIc � CII �G3PDH (Table 2) were not significantly influenced by anyassayed parameter, as determined by GLMs and the least-square means method. The Vmax was higher for COX than forCIc, CIc � CII, and CIc � CII � G3PDH, denoting that, in our

Fig. 2. Mitochondrial functions measured at 4different temperatures in permeabilized fibersfrom siII and siIII. O2 fluxes are expressed aspmol O2 consumed·s�1·mg protein�1, with pyru-vate � malate � L-proline � ADP (CI), �cyto-chrome c (CIc), �succinate (CIc � CII), �sn-glycerol 3-phosphate (CIc � CII � G3PDH),�2,4-dinitrophenol [CIc � CII � G3PDH �uncoupler (U)], �rotenone (CII � G3PDH �U), �malonate (G3PDH � U), �antimycin A(ROx), and �N,N,N=,N=-tetramethyl-p-phenyl-enediamine (TMPD) � ascorbate [cytochrome coxidase (COX)], and were corrected for residualO2 consumption (ROx). Values are means � SEfor 7–20 fiber preparations. Significance was setas P � 0.05. *Significant difference betweenmitotypes. Letters denote differences betweentemperatures: a is statistically different from b.

nloaded from

measurements of maximal respiration rates, mitochondria uti-lize �31–57% of the COX activity available according to thesubstrate combinations used (Table 2).

Next, using a combination of substrates that maximallyreduce complexes I and II and G3PDH, we examined thebiochemical threshold effect at high flux through the ETS.Azide titration resulted in hyperbolic inhibition of COX.Threshold plots display the overall flux as a function of COXactivity; the threshold for inhibition of COX is defined as theintercept of the initial slope with the linear fit of the final slope(Fig. 4A). The value of the threshold corresponds to the

percentage of the maximal COX activity that has been inhib-ited before impairment of the maximal respiration rate ofmitochondria; therefore, the higher the threshold value, thehigher the excess of COX activity. Consistent with our previ-ous results using the in vitro approach (42), we detected athreshold (�67% for siII and 75% for siIII). Subsequently, wedetected an apparent COX excess capacity, i.e., the intercept ofthe extrapolation of the linear regression for the final slope withthe axis at zero COX inhibition, at 12°C for both mitotypes[188% (R2 1) for siII and 230% (R2 0.995) for siIII]. Noapparent COX excess capacity was detected at 18, 24, and28°C for both mitotypes, and comparisons were made betweenthe different Ki and Ci (Fig. 4, B and C). Ki was only influencedby temperature (Table 1), and no significant differences wereobserved between mitotypes. Ki of both mitotypes increasedwith increasing temperatures, with significant differences at 24and 28°C compared with the two other temperatures (P �0.036). Interestingly, Ci was influenced by mitotype and tem-perature (Table 1). Significant differences between mitotypeswere found at all temperatures tested (P � 0.041), and a signifi-cant increase was also detected for both mitotypes between 24 and28°C (P � 0.001). At all temperatures, Ci at the COX level washigher for siIII, which might be reflected by the lower apparentCOX excess capacity at 12°C, as well as the lower activity for thismitotype.

Fig. 3. Substrate control ratio (SCR) calculatedat 4 different temperatures in permeabilized fi-bers from siII and siIII, with CIc � CII �G3PDH as reference state for CIc and CIc � CIIand with CIc � CII � G3PDH � U as referencestate for CII � G3PDH � U and G3PDH � U.Values are means � SE for 7–20 fiber prepara-tions. Significance was set as P � 0.05. *Sig-nificant difference between mitotypes. Lettersdenote differences obtained after addition ofsuccinate: a is different from b.

Table 2. Mitochondrial state 3 for CIc�, CIc � CII�, andCIc � CII � G3PDH-to-COX respiration rate ratio at 12,18, 24, and 28°C in permeabilized fibers ofDrosophila simulans

Temperature, °C

CIc/COXCIc �

CII/COXCIc � CII �G3PDH/COX

siII siIII siII siIII siII siIII

12 36 34 41 39 53 5118 37 35 41 39 57 5324 34 33 38 38 54 4928 35 31 36 34 50 44

Values are percent maximum activity (Vmax).

nloaded from

Enzymatic Analyses

The enzymatic activity of PDH was measured to ensure thatdivergences between haplotypes could not be dictated bydifferences in the rate of entry of carbon substrates in thetricarboxylic acid (TCA) cycle. PDH activity was only influ-enced by temperature (Table 1). No significant differenceswere detected between mitotypes at any temperature (Fig. 5).PDH activity was, however, affected by temperature and sig-nificantly increased between 12 and 18°C, as well as between24 and 28°C (P � 0.037).

Effects of Temperature

Results for Q10 are presented in Fig. 6. Calculation of Q10

leads to the elimination of interindividual variations in the data,making statistical analysis impossible. Results can, however,be used to discuss some general patterns. For both mitotypes,Q10 values of ETS complexes appear to be lower with lowertemperatures, and the highest Q10 values are reached at 24–28°C. The most evident difference between mitotypes is thelevel of COX thermal sensitivity at the highest temperaturerange, with Q10 of 2.49 for siIII compared with 1.51 for siII.The same pattern is observed for PDH, with higher Q10 athigher temperatures, while divergences between mitotypeswere observed (with higher Q10 for siII than siIII). Not sur-prisingly, temperature influenced most parameters: the respi-

ration rates for CI, CIc � CII � G3PDH � U, CII � G3PDH �U, and COX, the SCRs for CIc, CII � G3PDH � U, andG3PDH � U, and Ki and Ci, as well as PDH activity (Table 1).

DISCUSSION

In this study, using an in situ approach (whole muscle), closeto natural physiological conditions, we sought to identifydifferences in mitochondrial performance and thermal sensi-tivity between populations of D. simulans with divergentmitochondrial haplotypes. We also compare the results ob-tained previously using an in vitro approach (on isolatedmitochondria). To our knowledge, this is the first study to usepermeabilized fibers of invertebrates to evaluate the impact oftemperature variations on mitochondrial performance. We de-veloped a new technique that allowed us to accurately comparedifferent steps of the ETS, even with small differences in O2

consumption, which may be applied to investigate impacts ofenvironmental conditions on mitochondrial function. More-over, genetic divergences in mtDNA, which may be related todivergences at the level of metabolic efficiency, can be ad-dressed as an evolutionary determinant of metabolic adapta-tions.

The RCRs obtained with this new method showed well-coupled respiration (between 4.0 and 7.8 according to themitotype and the temperature), consistent with previous RCRsmeasured in isolated mitochondria (ranging from 5.3 to 8.3)

Fig. 4. A: apparent COX excess capacitymeasured at 12°C in siII and siIII. COXexcess capacity is represented by thresholdplot, which shows relative flux through theelectron transport system (ETS) as a functionof relative inhibition of COX at similar so-dium azide concentrations. Two linear re-gressions were calculated from initial andfinal slopes and extrapolated to zero-COXinhibition (R2 0.9962 and 1, respectively,in siII; R2 0.9951 and 0.9954, respectively,in siIII). Intercepts represent COX excesscapacity (see below). Values are means �SD. B: inhibition constant (Ki) of sodiumazide calculated from transformed data usingDickson linearization. C: flux control coeffi-cient (Ci) calculated according to Eq. 1 in thesiII and siIII mitotypes at the 4 differenttemperatures. Values are means � SE for7–20 fiber preparations for mitochondrialrespiration inhibition and 7–20 fiber prepara-tions for COX activity inhibition. Signifi-cance was set as P � 0.05. *Significantdifference between mitotypes. Letters denotedifferences between temperatures: a is statis-tically different from b and c, and b is statis-tically different from c.

nloaded from

using the same substrates and temperatures (42). The cyto-chrome c effect was very low at all experimental temperatures,confirming functional integrity of the outer mitochondrialmembrane. These results confirm that the in situ method wasappropriate for evaluation of mitochondrial function and mightbe more relevant than an in vitro approach, since the cellularenvironment is preserved.

UCR results showed that when the ETS is “nearly saturated”with electrons from substrates, no effect of uncoupling can bedetected, suggesting that ATP synthesis capacity and ADPsupply by ANT can support the maximum electron fluxthrough the ETS with the combination of substrates that weused (42). Since O2 fluxes for CIc � CII � G3PDH and CIc �CII � G3PDH � U are almost identical, phosphorylation ofADP does not appear to be a limiting step under these exper-imental conditions. Consequently, the control of mitochondrialrespiration may be exerted mostly upstream of ATP synthaseand ANT.

Unfortunately, it is difficult to compare the respiration ratesof the present study with those of the previous study of isolatedmitochondria, since rates of respiration have been normalizedin two different ways. In both studies, respiration was normal-ized with total protein content. However, in the previous study(42), total protein content was measured in mitochondrialpreparations; in the present study, we measured the totalprotein content of the permeabilized fibers. Therefore, wefocused on the comparisons of patterns of thermal sensitivitybetween the two mitotypes that we obtained in the two studies,instead of comparisons of the O2 fluxes between the twomethods.

In our previous study (42), we investigated the entry ofelectrons through the ETS at the level of complex I (withpyruvate � malate � L-proline) or at the level of complex III

via G3PDH (with sn-glycerol 3-phosphate). However, it wasnot possible to measure the simultaneous entry of electrons atthe level of complexes I and III. This was mainly because theO2 concentration in the chamber was exhausted within a fewminutes because of the system used (mitochondrial prepara-tions with high content of mitochondria), restricting the exper-iment and preventing simple protocols with few titration steps.This was not the case with high-resolution respirometry usingpermeabilized fibers; therefore, extended substrate-uncoupler-inhibitor titration protocols may be applied (26). Moreover, ithas been shown that in situ mitochondrial functions are usuallymore stable in permeabilized fibers than in isolated mitochon-dria, allowing longer and more complex protocols (27, 34). Inphysiological conditions, mitochondrial respiration is sup-ported by electron input from different NAD�- and FAD-linked complexes (mainly complex I, complex II, and G3PDH)into the Q cycle. Moreover, NAD�- and FAD-linked substratesenter the ETS at the same time, since NADH and FADH2 areproduced when the TCA cycle is activated. In the presentexperiments, we fed the TCA cycle by providing pyruvate,malate, and L-proline, which allowed generation of NADHthrough PDH, malate dehydrogenase, and glutamate dehydro-genase, whereas succinate and sn-glycerol 3-phosphate wereused to generate FADH2 via succinate dehydrogenase andG3PDH. The additive effect of substrate combinations on theO2 flux using strategic design of complex I � complex IIprotocols has been the approach of different studies (2, 13, 27,35) using different fiber preparations from various mammaliantissues to delineate the implication of the different complexesin the level of state 3 respiration.

In Drosophila, the impact of substrate combination has beeninvestigated for NAD�-linked substrates (22, 31, 41), but theimportance of FAD-linked substrates is still poorly understood.

Fig. 5. Pyruvate dehydrogenase (PDH) activities of siII and siIII measured in10 different homogenates per mitotype per temperature. Values are means �SD. Significance was set as P � 0.05. Letters denote differences betweentemperatures: a is statistically different from b and c; b is statistically differentfrom c.

Fig. 6. Temperature coefficients (Q10) for mitochondrial respiratory fluxes withpyruvate � malate � L-proline � ADP (CI), �cytochrome c (CIc), �succi-nate (CIc � CII), �sn-glycerol 3-phosphate (CIc � CII � G3PDH), �2,4-dinitrophenol (CIc � CII � G3PDH � U), �rotenone (CII � G3PDH � U),�malonate (G3PDH � U), �antimycin A (ROx), �TMPD � ascorbate(COX) were corrected for ROx and for pyruvate dehydrogenase (PDH)activity. Q10 values for mitochondrial functions were calculated according toEq. 2. Values are means for 7–20 fiber preparations.

nloaded from

Since the rate of mitochondrial respiration with succinate isquite low, it has been suggested that succinate does not readilycross the inner mitochondrial membrane (22, 51), while sn-glycerol 3-phosphate is preferred as FAD-linked substrate (butsee Ref. 55). Indeed, if respiration rates for succinate were verylow, it can be greatly increased by breaking up mitochondria(51), leading to the conclusion that insect mitochondria areimpermeable to succinate. With SCR as an expression of therelative control exerted by variation of experimental substratesat a fixed coupling state (27), our results clearly demonstratethat the addition of succinate has a significant impact on therespiration rate (Fig. 3) of Drosophila, and it is thereforeessential to take into account complex II as a convergingbranch to the Q junction. Moreover, complex II is only en-coded by the nuclear genome, and since no differences be-tween mitotypes were detected in SCR or in O2 fluxes for CIc �CII (Figs. 2 and 3) compared with CIc alone, we suggest thatthe impact of the nuclear genes on divergences of mitochon-drial functions between these two populations is weak. Theseresults are in line with enzymatic assays, since PDH, alsoencoded by nuclear genes, did not show differences betweenmitotypes at any of the temperatures tested. These results arealso consistent with those from our previous study, since weshowed that citrate synthase and aconitase (two other enzymesstrictly encoded by nuclear DNA) activities were not differentbetween mitotypes (42).

Study of the functional design of the OxPhos system re-quires a conceptual transition from a strict analysis of the ETSto a perspective envisioning the convergent structure of elec-tron flow to the Q junction (27). Comparative analyses of ETScatalytic capacity, therefore, required investigation of the elec-tron flux at the Q junction, with maximum capacities of allcontributory branches converging at this Q junction. In ourstudy, the protocol was designed to evaluate the substratecombination effect in terms of O2 consumption with threedifferent electron inputs (complex I, complex II, and G3PDH)through the ETS. Respiration rates showed that when ETS wassupplied with substrates from three branches (CIc � CII �G3PDH), significant differences between mitotypes were de-tected at 24°C, and these differences were also observed whenan uncoupler was added (CIc � CII � G3PDH � U), as wellas when inhibitors of complex I or II were used (CII � G3PDH �U and G3PDH � U). These results suggest a higher catalyticcapacity of the ETS from the siII population. This divergenceis clearly significant at 24°C and is still apparent at 28°C.These differences are significant at the highest electron flux(CIc � CII � G3PDH and CIc � CII � G3PDH � U), whencomplexes I and II are inhibited (CII � G3PDH � U andG3PDH � U), and for COX activity. It is still possible that thehigher flux for siII is induced by a higher rate of electron entryat the G3PDH level, but this enzyme is only encoded bynuclear DNA, and it is unlikely that G3PDH is differentbetween the two mitotypes. This leaves us to suspect that thecatalytic advantage should be associated with the ability ofcomplexes III and IV to deplete electrons, which is in line withthe higher level of COX activity in siII.

Maximal COX activity is higher in siII at 24°C (Fig. 2),highlighting the importance of COX catalytic capacity tomodulate the overall respiration rate. The ratios of mitochon-drial respiration to COX showed no differences among tem-peratures or between mitotypes. The biochemical threshold and

the apparent COX excess capacity were measured at highpathway flux, with the three branches fully fueled with elec-trons. A relatively important biochemical threshold of �67%in siII and 75% in siIII was observed only at 12°C, with anapparent COX excess capacity of �188% in siII and 230% insiIII (Fig. 4A). From 18 to 28°C, no excess capacity of COXwas detected, suggesting that the maximal, uninhibited COXactivity could not support higher respiration rates than maximalstate 3 measured at these temperatures. However, at 12°C, theexcess revealed much higher catalytic capacity than is requiredto support the maximum state 3. Limitation should thereforecome from reactions upstream of cytochrome c oxidation,since even at this temperature, uncoupler did not increaserespiration rate, indicating no limitation by ATP synthase orANT. Since the relative activity of the different complexesdoes not vary compared with COX at the different tempera-tures (Table 2), this discounts the probability that a majorlimitation arises from other complexes of ETS, leaving us toconclude that it could be induced by impairment at the levelsof substrate transports, TCA cycle, or dehydrogenases, whereelectrons are provided. We cannot confirm this hypothesis, butPDH activity strongly suggests that at 12°C this enzyme cannotsupport a high flux of entry electron into the ETS, leading tothe observed apparent excess COX capacity.

We previously demonstrated a more marked apparent COXexcess capacity at low temperature in isolated mitochondria(excess of 604% for siII and 613% for siIII) (42). These resultssuggest that, at low temperature, isolated mitochondria mightnot support the state 3 level of respiration achieved by mito-chondria in permeabilized fibers within similar conditions. Incombination, these data strongly suggest that isolated mito-chondria do not accurately reflect the functional properties ofaerobic metabolism at low temperature, nor do they documentthe thermal sensitivity of cellular respiration in ectotherms.Previous studies demonstrated higher apparent COX excesscapacity in isolated mitochondria than in permeabilized fibers(32, 54). These differences between the in vitro and the in situstudies and the much higher apparent COX excess capacity inmitochondrial preparations likely arise from different experi-mental conditions, as well as from the appropriate microenvi-ronmental conditions of the in situ experiment.

Unfortunately, the present calculation of the Ci for COX wasmade at maximal state 3 with saturating concentration of ADP,which is not relevant to normal physiological conditions. In thepresent study, only the Ci for COX was measured, and werationalized that this analysis did not allow any inferencesabout the flux control at lower respiratory rates. However, itremains a good indicator of physiological and biochemicaldivergences between the two mitotypes, even if its relevance in“normal” physiological conditions remains to be evaluated byfurther experimentation. At 12, 18, and 24°C, COX exerts arelatively low control over the OxPhos capacity (Fig. 4C)compared with 28°C. At this higher temperature, OxPhoscapacity seems highly controlled by COX. The control ofOxPhos capacity should therefore be more evenly distributedover the different complexes of the ETS at 12, 18, and 24°C.Even at 12°C, where high apparent COX excess capacity wasmeasured, COX maintained a significant control over state 3respiration. This is quite intriguing, since Chamberlin (14), ina top-down approach, noted that although temperature affectedthe kinetics of all subsystems (substrate oxidation, phosphor-

nloaded from

ylation, and proton leak), the control of respiration did notchange with temperature. However, this study considers thesubstrate oxidation system (including TCA cycle, electrontransport chain, and metabolite transporters) as an integratedsystem. While it is possible that the proportion of the controlcarried out by this system does not vary with temperature, it isstill possible that the distribution of control among the com-plexes or reactions inside this system does vary.

In terms of thermal sensitivity, important divergences weredenoted between mitotypes that should be interpreted cau-tiously because of the method we used to calculate Q10. Bothmitotypes show low thermal sensitivity at 12–18°C, and siIIIalso showed a weak impact of temperature at 18–24°C (Q10

close to 1). At 18–24°C, Q10 is 1.5- to 2-fold higher in siII thansiIII, whereas at 24–28°C, Q10 is highest in siIII for almost allparameters except CI and CIc. In this temperature range, theQ10 values for PDH are high in both mitotypes. Moreover, at24–28°C, the Q10 values for COX in siIII were the highest,which suggests a strong influence of temperature on thismitotype, but mostly for complex IV. Since ratios of mitochon-drial respiration to COX for CIc, CIC � CII, and CIc � CII �G3PDH were not significantly different between temperatures(Table 2), we could suggest that the differences between bothmitotypes are induced by divergences in the functional prop-erties of the component of the ETS, and not by simple quan-titative adjustments. Even if this suggestion seems weak be-cause of the method used to calculate Q10, temperature influ-ence determined by ANOVA (Table 1) also supports thishypothesis. The overall lower Q10 values at 12–18°C mightallow siII to colonize different spatial niches, which is in linewith its wider worldwide distribution, while siIII is endemic tocontinental East Africa, Madagascar, and Reunion Island (5).This spatial niche differentiation hypothesis does not, however,exclude the possibility of temporal niche differentiation (diur-nal or seasonal differences in haplotype performance/abun-dance) (42).

No distinctions between the ratios of each complex to COXwere detected among experimental temperatures (Table 2).This is quite surprising, since the different enzymes of the ETSshould have different thermal sensitivities, because they arepartly governed by the catalytic properties of enzymes and,particularly, by the free energy of activation of the reaction (16,48). Temperature-induced changes should therefore have dif-ferential effects on the different enzymes of the ETS; conse-quently, control over respiration should be differently distrib-uted along the ETS at different temperatures. According to ourresults, it appears that the control exerted by COX is minimallyaffected by the wide temperature range (12–24°C) encounteredby this species. Taken together, these results highlight therobust nature of the functional and regulatory properties of theETS, which helps support a wide range of temperaturechanges.

Perspectives and Significance

We have shown that an in situ approach for the assessmentof mitochondrial performance is appropriate for invertebratespecies and may be more relevant than the in vitro approach onisolated mitochondria, since cellular interactions are muchmore preserved in permeabilized fibers. We also showed thatwhen succinate is provided, respiratory rates increase in Dro-

sophila (it may be only detected in permeabilized fibers withhigh-resolution respirometry, as other studies on mitochondrialisolations did not detect any effects). Therefore, it must betaken into account in evaluation of the capacity of all branchesof the ETS converging at the Q junction and the consequentialimpact on mitochondrial respiration. The apparent COX excesscapacity detected at 12°C may be due to impairment in pro-cesses upstream that drive the electron input into the ETS.These results are partly corroborated by assays of a specificdehydrogenase (PDH) that showed low activity at 12°C.Higher catalytic capacities of the ETS were detected at 24°C inthe siII mitotype, and this catalytic advantage could be asso-ciated with the ability of complexes III and IV to depleteelectrons.

Our study clearly demonstrates that the ETS is a robustsystem in terms of functional and regulatory properties, whichis supported across a wide range of temperatures, and thatcatalytic capacities at 24°C are higher in the siII than the siIIIhaplotype. Even if both mitochondrial mitotypes revealed hightolerance to temperature variation, it is clear that the catalyticcapacity of mitochondria from the siII population outcompetesthat of mitochondria from the siIII population at 24°C; there-fore, sensitivity of mitochondrial respiration diverges betweenboth populations at �24°C. This higher catalytic capacity inthe siII mitotype could result in an advantage in terms ofintensity of aerobic activity, endurance, or both, if the intensityof exercise that can be aerobically performed is partly dictatedby the aerobic capacity of the tissue. We could also suggestanother advantage of the siII over the siIII mitotype whenindividuals are performing an equivalent workload. For a givenmechanical work, mitochondria from an siII mitotype individ-ual may be solicited at a lower ratio of their Vmax, since theirmitochondrial catalytic capacity is higher. This might drive alower rate of reactive oxygen species production, since it isproportional of the electron flux and of the redox state of theETS. In both cases, further experiments are required to clearlydelineate the adaptive advantage of siII catalytic capacity.While the different thermal sensitivities of the two mitotypes’ETS complexes could result from selection on these mitotypesto better perform in their specific environments, we cannotconclude that the difference resides entirely at the level ofmtDNA. It is plausible that undetected differences in nucleargenes may exist. Introgressions between mitotypes may shedlight on the adaptive value of Drosophila mitochondria inco-adapted cellular environments and will allow us to delineatethe relationship between respiration rates and mtDNA diver-gences.

Few studies have been able to relate divergences amongmitochondrial haplotypes of natural populations to any meta-bolic and phenotypic traits with potential adaptive value (30).Evaluation of the potential adaptive value of mtDNA haplo-types is of paramount importance, since this will highlight thepossibility or limitation of the evolutionary plasticity of mito-chondrial metabolism. Our observations on mitochondrial cat-alytic capacity divergence associated to mitochondrial mito-types bring new experimental opportunities to test adaptivevalues of mtDNA.

ACKNOWLEDGMENTS

We thank H. Lemieux for help with experimental protocols and E. HébertChatelain for help with animal care.

nloaded from

GRANTS

This study was supported by research grants from the Natural Sciences andEngineering Research Council to P. U. Blier and Canadian Institute of HealthResearch to R. M. Tanguay.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

REFERENCES

1. Abele D, Heise K, Pörtner HO, Puntarelo S. Temperature-dependenceof mitochondrial function and production of reactive oxygen species in theintertidal mud clam Mya arenaria. J Exp Biol 205: 1831–1841, 2002.

2. Aragonés J, Schneider M, Van Geyte K, Fraisl P, Dresselaers T,Mazzone M. Deficiency or inhibition of oxygen sensor Phd1 induceshypoxia tolerance by reprogramming basal metabolism. Nat Genet 40:170–180, 2008.

3. Ballard JWO. Comparative genomics of mitochondrial DNA in membersof the Drosophila melanogaster subgroup. J Mol Evol 51: 48–63, 2000.

4. Ballard JWO. When one is not enough: introgression of mitochondrialDNA in Drosophila. Mol Biol Evol 17: 1126–1130, 2000.

5. Ballard JWO. Sequential evolution of a symbiont inferred from the host:Wolbachia and Drosophila simulans. Mol Biol Evol 21: 428–442, 2004.

6. Ballard JWO. Drosophila simulans as a novel model for studyingmitochondrial metabolism and aging. Exp Gerontol 40: 763–773, 2005.

7. Ballard JWO, Melvin RG, Katewa SD, Maas K. Mitochondrial DNAvariation is associated with measurable differences in life-history traits andmitochondrial metabolism in Drosophila simulans. Evolution 61: 1735–1747, 2007.

8. Ballard JWO, Melvin RG. Linkin the mitochondrial genotype to theorganismal phenotype. Mol Ecol 19: 1523–1539, 2010.

9. Blier PU, Guderley HR. Mitochondrial activity in rainbow trout redmuscle: the effect of temperature on the ADP-dependence of ATP syn-thesis. J Exp Biol 176: 145–157, 1993.

10. Blier PU, Guderley HE. Effects of pH and temperature on the kinetics ofpyruvate oxidation by muscle mitochondria from rainbow trout (On-corhynchus mykiss). Physiol Zool 66: 474–489, 1993.

11. Blier PU, Lemieux H. The impact of the thermal sensitivity of cyto-chrome c oxidase on the respiration rate of Arctic char red musclemitochondria. J Comp Physiol B 171: 247–253, 2001.

12. Blier PU, Dufresne F, Burton RS. Natural selection and the evolution ofmtDNA-encoded peptides: evidence for intergenomic co-adaptation.Trends Genet 17: 400–406, 2001.

13. Boushel R, Gnaiger E, Schjerling P, Skovbro M, Kraunsoe R, Flem-ming D. Patients with type 2 diabetes have normal mitochondrial functionin skeletal muscle. Diabetologia 50: 790–796, 2007.

14. Chamberlin ME. Top-down control analysis of the effect of temperatureon ectotherm oxidative phosphorylation. Am J Physiol Regul Integr CompPhysiol 287: R794–R800, 2004.

15. Csordás G, Renken C, Várnai P, Walter L, Weaver D, Buttle KF,Balla T, Mannelle CA, Hajnóczky G. Structural and functional featuresand significance of the physiological linkage between ER and mitochon-dria. J Cell Biol 174: 915–921, 2006.

16. Daniel RM, Danson MJ, Eisenthal R, Lee CK, Peterson ME. The effectof temperature on enzyme activity: new insights and their implications.Extremophiles 12: 51–59, 2008.

17. Dean MD, Ballard KJ, Glass A, Ballard JWO. Influence of twoWolbachia strains on population structure of east African Drosophilasimulans. Genetics 65: 1959–1969, 2003.

18. Dufour S, Rousse N, Diolez P. Top-down control analysis of temperatureeffect on oxidative phosphorylation. Biochem J 314: 743–751, 1996.

19. Dzeja PP, Bortolon R, Perez-Terzic C, Holmuhamedov EL, Terzic A.Energetic communication between mitochondria and nucleus directed bycatalyzed phosphotransfer. Proc Natl Acad Sci USA 99: 10156–10161,2002.

20. Fangue NA, Richards JG, Schulte PM. Do mitochondrial propertiesexplain intraspecific variation in thermal tolerance? J Exp Biol 212:514–522, 2009.

21. Fell DA. Metabolic control analysis: a survey of its theoretical andexperimental development. Biochem J 286: 313–330, 1992.

22. Ferguson M, Mockett RJ, Shen Y, Orr WC, Sohal RS. Age-associateddecline in mitochondrial respiration and electron transport in Drosophilamelanogaster. Biochem J 390: 501–511, 2005.

23. Fontanillas P, Depraz A, Giorgi MS, Perrin N. Nonshivering thermo-genesis capacity associated to mitochondrial DNA haplotypes and genderin the greater white-toothed shrew, Crocidura russula. Mol Ecol 14:661–670, 2005.

24. Gnaiger E, Kuznetsov AV. Mitochondrial respiration at low levels ofoxygen and cytochrome c. Biochem Soc Trans 30: 252–258, 2002.

25. Gnaiger E, Wright-Paradis C, Sondergaard H, Lundby C, Calbet JA,Saltin B. High-resolution respirometry in small biopsies of human muscle:correlations with body mass index and age. Mitochondr Physiol Network10: 14–15, 2005.

26. Gnaiger E. (Editor). Mitochondrial Pathways and Respiratory Control(1st ed.). Innsbruck: OROBOROS MiPnet, 2007.

27. Gnaiger E. Capacity of oxidative phosphorylation in human skeletalmuscle: new perspectives of mitochondrial physiology. Int J Biochem CellBiol 41: 1837–1845, 2009.

28. Hardewig I, Peck LS, Pörtner HO. Thermal sensitivity of mitochondrialfunction in the Antarctic Notothenioid Lepidonotothen nudifrons. J CompPhysiol B 169: 597–604, 1999.

29. James AC, Ballard JWO. Expression of cytoplasmic incompatibility inDrosophila simulans and its impact on infection frequencies and distribu-tion of Wolbachia pipientis. Evolution 54: 1661–1672, 2000.

30. Jha M, Côté J, Blier PU, Stewart DT. Sperm motility in Mytilus edulisin relation to mitochondrial DNA polymorphisms: implications for theevolution of doubly uniparental inheritance in bivalves. Evolution 62:99–106, 2008.

31. Katewa SD, Ballard JWO. Sympatric Drosophila simulans flies withdistinct mtDNA show difference in mitochondrial respiration and electrontransport. Int Biochem Mol Biol 37: 213–222, 2007.

32. Kunz WS, Kudin A, Vielhaber S, Elger CE, Attardi G, Villani G. Fluxcontrol of cytochrome c oxidase in human skeletal muscle. J Biol Chem274: 27741–27745, 2000.

33. Kuznetsov AV, Clark JF, Winkler K, Kunz WS. Increase of flux controlof cytochrome c oxidase in copper-deficient mottled brindled mice. J BiolChem 271: 283–288, 1996.

34. Kuznetsov AV, Veksler V, Gellerich FN, Saks V, Margreiter R, KunzWS. Analysis of mitochondrial function in situ in permeabilized musclefibers, tissues and cells. Nat Protoc 3: 965–976, 2008.

35. Lemieux H, Garedew A, Blier PU, Tardif JC, Gnaiger E. Temperatureeffects on the control and capacity of mitochondrial respiration in perme-abilized fibers of the mouse heart. Biochim Biophys Acta 14: 201–202,2006.

36. Lemieux H, Tardif JC, Dutil JD, Blier PU. Thermal sensitivity ofcardiac mitochondrial metabolism in an ectothermic species from a coldenvironment, Atlantic wolfish (Anarhichas lupus). J Exp Marine Biol Ecol384: 113–118, 2010.

37. Letellier T, Malgat M, Coquet M, Moretto B, Parrot-Roulaud F,Mazat JP. Mitochondrial myopathy studies on permeabilized musclefibers. Pediatr Res 32: 17–22, 1992.

38. Letellier T, Heinrich R, Malgat M, Mazat JP. The kinetic basis ofthreshold effects observed in mitochondrial diseases: a systemic approach.Biochem J 302: 171–174, 1994.

39. Melvin RG, Katewa SD, Ballard JWO. A candidate complex approachto study functional mitochondrial DNA changes: sequence variation andquaternary structure modeling of Drosophila simulans cytochrome coxidase. J Mol Evol 66: 232–242, 2008.

40. Milner DJ, Mavroidis M, Weisleder N, Capetanaki Y. Desmin cyto-skeleton linked to muscle mitochondrial distribution and respiratory func-tion. J Cell Biol 150: 1283–1298, 2000.

41. Miwa S, St-Pierre J, Partridge L, Brand MD. Superoxide and hydrogenperoxide production by Drosophila mitochondria. Free Radic Biol Med35: 938–948, 2003.

42. Pichaud N, Hébert Chatelain E, Ballard JWO, Tanguay R, MorrowG, Blier PU. Thermal sensitivity of mitochondrial metabolism in twodistinct mitotypes of Drosophila simulans: evaluation of mitochondrialplasticity. J Exp Biol 213: 1665–1675, 2010.

43. Rasmussen HN, Rasmussen UF. Oxygen solubilities of media used onelectrochemical respiration measurements. Anal Biochem 319: 105–113,2003.

44. Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM,Tuft RA, Pozzan T. Close contacts with the endoplasmic reticulum asdeterminants of mitochondrial Ca2� responses. Science 280: 1763–1765,1998.

45. Rossignol R, Faustin B, Rocher C, Malgat M, Mazat JP, Letellier T.Mitochondrial threshold effects. Biochem J 370: 751–762, 2003.

nloaded from

46. Saks VA, Veksler VI, Kuznetsov AV, Kay L, Sikk P, Tiivel T, TranquiL, Olvares J, Winkler K, Wiedemann F, Kunz WS. Permeabilized celland skinned fiber techniques in studies of mitochondrial function in vivo.Mol Cell Biochem 184: 81–100, 1998.

47. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH,Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk MD.Measurement of protein using bicinchoninic acid. Anal Biochem 150:76–85, 1985.

48. Somero GN. Adaptation of enzymes to temperature: searching for basic“strategies.” J Comp Physiol B 139: 321–333, 2004.

49. Thibeault M, Blier PU, Guderley H. Seasonal variation of musclemetabolic organization in rainbow trout (Oncorhynchus mykiss). FishPhysiol Biochem 16: 139–155, 1997.

50. Ulrich PN, Marsh AG. Thermal sensitivity of mitochondrial respirationefficiency and protein phosphorylation in the clam Mercenaria merce-naria. Mar Biotechnol 11: 608–618, 2009.

51. Van Den Bergh SG, Slater EC. The respiratory activity and permeabilityof housefly sarcosomes. Biochem J 82: 362–371, 1962.

52. Veksler VI, Kuznetsov AV, Sharov VG, Kapelko VI, Saks VA. Mito-chondrial respiratory parameters in cardiac tissue: a novel method ofassessment by using saponin-skinned fibers. Biochim Biophys Acta 892:191–196, 1987.

53. Villani G, Attardi G. In vivo control of respiration by cytochrome coxidase in wild-type and mitochondrial DNA mutation-carrying humancells. Proc Natl Acad Sci USA 94: 1166–1171, 1997.

54. Villani G, Greco M, Papa S, Attardi G. Low reserve of cytochrome coxidase capacity in vivo in the respiratory chain of a variety of human celltypes. J Biol Chem 273: 31829–31836, 1998.

55. Walker DW, Hájek P, Muffat J, Knoepfle D, Cornelison S, Attardi G,Benzer S. Hypersensitivity to oxygen and shortened lifespan in a Dro-sophila mitochondrial complex II mutant. Proc Natl Acad Sci USA 103:16382–16387, 2006.

nloaded from

Thermal sensitivity of mitochondrial functions in permeabilized muscle fibers from two populations...

Documents