Determination of Mitochondrial Function and Site-Specific Defectsin Electron Transport in Duodenal Mitochondria in Broilers

with Low and High Feed Efficiency1

C. P. Ojano-Dirain,*,2 M. Iqbal,* D. Cawthon,* S. Swonger,* T. Wing,† M. Cooper,† and W. Bottje*

*Department of Poultry Science, University of Arkansas, Fayetteville, Arkansas 72701;and †Cobb Vantress Inc., Siloam Springs, Arkansas 72761

ABSTRACT Duodenal mitochondria were isolatedfrom broiler breeder males with high (0.79 ± 0.01, n = 9)and low (0.63 ± 0.02, n = 9) feed efficiency (FE) to assessrelationships of FE with duodenal mitochondrial functionand site-specific defects in electron transport. Sequentialadditions of adenosine diphosphate (ADP) resulted in 1)higher respiratory control ratio (RCR; an index of respira-tory chain coupling) in high FE mitochondria providedsuccinate, and 2) higher ADP to oxygen ratio (ADP:O; anindex of oxidative phosphorylation) in low FE mitochon-dria provided NADH-linked substrates (malate, pyr-uvate, or both). Basal electron leak, measured as H2O2

production, was greater in low FE mitochondria providedsuccinate (P = 0.08) or NADH-linked substrates. As H2O2

(Key words: broiler, duodenum mitochondria, electron leak, feed efficiency, respiratory control ratio)

2004 Poultry Science 83:1394–1403

INTRODUCTION

As feed represents about 50 to 70% of the cost of broilerproduction, feed efficiency (FE, gain to feed) or feed con-version ratio (FCR, feed to gain) remains one of the mostimportant traits in commercial animal breeding pro-grams. Genetic selection for increased broiler perfor-mance has resulted in a 250 to 300% improvement inbody weight and FE in 1991 and 2001 broiler strains com-pared with a 1957 random bred control population (Ha-venstein et al., 1994, 2003; Chapman et al., 2003).Availability of feed additives that support optimumgrowth and advances in broiler production such as feedformulation that allow close approximation of the nutri-ent requirements have also improved FE. However, de-spite these advances, as much as 10% variation in growthand FE still exists within broiler lines (Emmerson, 1997).

2004 Poultry Science Association, Inc.Received for publication February 5, 2004.Accepted for publication April 4, 2004.1This research was funded by USDA-NRI grant (#2001-03443) and

Cobb-Vantress Inc. to W. Bottje, and is published with the support ofthe Director of the Agriculture Research Experiment Station, Universityof Arkansas, Fayetteville, AR.

2To whom correspondence should be addressed: cojano@uark.edu.

1394

levels were elevated in low FE compared with high FEmitochondria by complex I (P ≤ 0.07) and complex IIinhibition, the higher basal electron leak in low FE mito-chondria was apparently due to site-specific defects inelectron transport at complexes I and II. Elevations inH2O2 above basal levels indicated that high FE mitochon-dria may also exhibit electron transport defects at com-plexes I and III. Despite an ability to produce adenosinetriphosphate (ATP) that was equal or superior to thatdemonstrated in high FE duodenal mitochondria, lowFE mitochondria exhibited a greater inherent degree ofelectron leak. The results provide insight into the rolethat duodenal mitochondria play in the phenotypic ex-pression of FE in broilers.

Recent evidence suggests that mitochondrial function orbiochemistry may be associated with FE in broilers (Bottjeet al., 2002; Iqbal et al., 2004) and rats (Lutz and Stahly,2003). Whereas various studies have reported differencesin mitochondrial oxygen utilization with different breedsof animals (e.g., Mukherjee et al., 1970; Dziewiecki andKolataj, 1976; Brown et al., 1986) or dietary manipulations(e.g., Renner et al., 1979; Toyomizu et al., 1992a,b), to ourknowledge, the reports of Bottje et al. (2002) and Lutzand Stahly (2003) are the only studies that have suggesteda direct relationship between mitochondrial function andFE that have excluded breed and dietary effects.

Mitochondria are organelles found in all eukaryoticcells whose major function is to generate cellular energy(adenosine triphosphate; ATP). Approximately 90% ofthe total ATP production from the complete oxidationof glucose to carbon dioxide and water is generated by

Abbreviation Key: ADP = adenosine diphosphate; ADP:O = adeno-sine diphosphate to oxygen ratio; ATP = adenosine triphosphate; DCFH-DA = 2′,7′-dichlorofluorescin diacetate; EGTA = ethylene glycol-bis (β-aminoethylether)-N,N,N′,N′ tetraacetic acid; FE = feed efficiency;HEPES = N-[2-hydroxyethylpiperizine]-N′-[2-ethanesulfonic acid];O2

�- = superoxide; RCR = respiratory control ratio; ROS = reactive oxygenspecies; TTFA = thenoyltrifluroacetone.

FEED EFFICIENCY AND MITOCHONDRIAL FUNCTION 1395

FIGURE 1. Diagrammatic representation of the respiratory chainadapted from Lehninger et al. (1993). The respiratory chain consists of5 multiprotein enzyme complexes, complexes I, II, III, IV, and V [ATPase;adenosine triphosphate (ATP) synthase], and 2 mobile electron carriers,coenzyme Q (CoQ) and cytochrome c (cyt c). Electrons from NADH-linked and FADH2-linked substrate donate electrons at complexes I andII, respectively. Electrons (e-) are transferred to O2, the terminal electronacceptor, which is subsequently reduced to water. Electron flow throughcomplexes I, III, and IV is accompanied by proton (H+, dashed arrows)transfer from the mitochondrial matrix to the inner membrane spacecreating a proton-motive force, which drives proton back into the matrixthrough ATP synthase and provides the energy for ATP synthesis.

mitochondria via oxidative phosphorylation (Lehningeret al., 1993). The inner mitochondrial membrane bears thecomponents of the electron transport chain, also known asthe respiratory chain or oxidative phosphorylation sys-tem. The respiratory chain consists of 4 multisubunit en-zyme complexes (I, II, III, and IV), 2 electron carriers,coenzyme Q (CoQ) and cytochrome c (cyt c), and ATPsynthase (ATPase or Complex V), the ATP-synthesizingenzyme complex (Figure 1). Electrons from NADH-linkedsubstrates such as malate and pyruvate enter the chain atcomplex I, whereas succinate, a FADH2-linked substrate,donates electrons at complex II. Electrons are transferredalong the electron transport chain through the proteincomplexes to O2, the terminal electron acceptor that issubsequently reduced to water. Electron flow throughcomplexes I, III, and IV is accompanied by proton transferfrom the matrix to the inner membrane space creating aproton-motive force, which provides the energy for ATPsynthesis as protons flow back into the matrix throughATP synthase (Lehninger et al., 1993).

The mitochondrial respiratory chain has also been rec-ognized as a major site of reactive oxygen species (ROS)production and, therefore, is a major source of endoge-nous oxidative stress. It is estimated that 2 to 4% of thetotal oxygen consumed by mitochondria is converted toROS due to univalent reduction of oxygen to form super-oxide (O2

�-) as a result of electron leakage from the electrontransport chain (Boveris and Chance, 1973; Chance etal., 1979; Turrens and Boveris, 1980). The O2

�- is quicklydismutated by the mitochondrial superoxide dismutaseto produce ROS such as H2O2. Increased ROS may over-

3Cobb Vantress, Inc., Three Springs Farm, OK.

whelm the cell’s antioxidant protection and can causedamage to critical structures in the cell including proteins,DNA, and lipids (Yu, 1994). In addition, elevated ROSmay also induce cell death (Miwa et al., 2000).

Mitochondrial dysfunction such as increased ROS pro-duction have been linked to the pathogenesis of manyhuman diseases, including but not limited to, diabetes,Alzheimer’s, Parkinson’s, ischemic reperfusion injury,and in aging (e.g., Fiegal and Shapiro, 1979; Shigenagaet al., 1994; Benzi and Moretti, 1995; Herrero and Barja,1998; Madesh et al., 2000; Rustin and Rotig, 2001). Inbroilers, mitochondrial dysfunction has been implicatedto pulmonary hypertension syndrome (Cawthon et al.,1999, 2001; Tang et al., 2000, 2002; Iqbal et al., 2001).Bottje et al. (2002) also reported that muscle mitochondrialdysfunction including increased ROS production maycontribute to the phenotypic expression of low FE in broil-ers. As the small intestine is the major site of nutrientabsorption requiring considerable amounts of ATP, thepurpose of this study was to extend our findings fromthose of the muscle by evaluating duodenal mitochon-drial function in broilers from the same genetic line andfed the same diet. Our hypothesis is that intestinal mito-chondrial function could also be important in the pheno-typic expression of FE in broilers. Major objectives of thisstudy were: 1) to evaluate relationships between duode-nal mitochondrial function (respiratory chain couplingand oxidative phosphorylation) and FE and 2) to assesssites and amounts of electron leak in duodenum mito-chondria by measuring H2O2 production in low and highFE broilers.

MATERIALS AND METHODS

Birds and Management

In this study, eighteen 7-wk-old birds that exhibitedthe highest or lowest FE were identified within a groupof 100 breeder male replacement stock3 tested for FE from6 to 7 wk of age. The birds were color-coded such thatwe were not aware which color belonged to the high andlow FE groups until completion of the experiments. Birdswere transported to the University of Arkansas, wherethey were housed individually in similar cages (51 × 51× 61 cm) and environmental conditions (25°C; 15L:9D)and were fed the same diet provided during the FE trial(20.5% protein, 3,280 kcal/kg). Birds were provided freeaccess to feed and water. After a 5-d acclimation, onebird per day was randomly selected from each group forisolation of duodenal mitochondria, with group selectionalternated on each day. Each bird was euthanized withan overdose of sodium pentobarbital by intravenous in-jection into the wing vein. Data were also obtained forweight, length, and diameter of the duodenal loop andmucosa weight.

Duodenal Mitochondria Isolation

Mitochondria were isolated by standard differentialcentrifugation as described by Lawrence and Davies

OJANO-DIRAIN ET AL.1396

TABLE 1. Six- to 7-wk growth performance of broilerswith low or high feed efficiency (FE) 1

High FE Low FEVariable (n = 9) (n = 9) P-value

6-wk BW, g 2,331 ± 28 2,310 ± 21 0.5447-wk BW, g 3,283 ± 39 3,065 ± 48 0.003BW gain, g 951 ± 30 756 ± 36 0.001Feed intake, g 1,205 ± 30 1,190 ± 47 0.790FE, g of gain/g of feed 0.79 ± 0.01 0.63 ± 0.01 <0.001FCR,2 g of feed/g of gain 1.27 ± 0.01 1.58 ± 0.02 <0.001

1For details see Materials and Methods. Values are mean ± SEM ofn shown in parentheses.

2Feed conversion ratio.

(1986) with modifications. Briefly, the duodenal loop wasquickly excised from the anesthetized bird and immedi-ately placed in a beaker containing ice-cold oxygenatedsolution A [160 mM sucrose, 110 mM mannitol, 2 mMHEPES, 11 mM Tris-HCl, 0.25 mM ethylene glycol-bis (β-aminoethylether)-N,N,N′,N′ tetraacetic acid (EGTA), and0.1mM phenylmethylsulfonyl fluoride (PMSF), pH 7.4].The pancreas was carefully removed, and the duodenalloop was cut in half. The luminal side was flushed withsolution A, and the remaining digesta was washed withthe same solution and then wiped gently with Whatman#2 filter paper to blot solution A and wipe off mucin. Themucosa was separated from the musculature by scrapingwith a microscope slide.

Approximately 10 g of mucosa was mixed with DEAE-cellulose suspension [8 g of DEAE-cellulose in 80 mL ofisolation medium A (70 mM sucrose, 220 mM mannitol,2 mM HEPES, 0.5 mM EGTA, 0.1 mM phenylmethylsulfo-nyl fluoride, and 0.37 g fatty acid-free BSA/100 mL, pH7.4), 175 U/mg heparin, and 1mM dithiothreitol]. After2 min, 50 mL of isolation medium A was added to thesuspension, and the mucosal cells were disrupted by gen-tle homogenization with 6 to 7 strokes in a Potter-Elvej-ehm vessel with a Teflon pestle. The homogenate wasdiluted by a further addition of 150 mL of isolation me-dium A and centrifuged at 750 × g for 10 min. The super-natant containing mitochondria was filtered throughnylon cloth4 and centrifuged at 9,800 × g for 7 min. Themucus layer from the mitochondrial pellets was removed.The mitochondria were then resuspended in 35 mL ofisolation medium A and centrifuged again at 12,100 × gfor 7 min. The mitochondrial pellet was further enrichedby subjecting it to a final spin (12,100 × g for 7 min) in35 mL of isolation medium B (70 mM sucrose, 220 mMmannitol, 2 mM HEPES, and 1.2 g of fatty acid-free BSA/100 mL, pH 7.4). The final mitochondrial pellet was sus-pended in isolation medium B and stored on ice for func-tional assays. All procedures were carried out at 4°C.

Mitochondrial protein was determined with a Lowryassay,5 with BSA as a standard, according to Lawrence

4Performix 900, Berkshire Corporation, Great Barrington, MA.5Kit 610-A, Sigma Chemical Co., St. Louis, MO.6Yellow Springs Instruments Co., Yellow Springs, OH.

TABLE 2. Duodenal loop (DL) physical characteristics of broilerswith low or high feed efficiency (FE)1

High FE Low FEVariable (n = 7) (n = 8) P-value

DL weight, g 43 ± 4 46 ± 4 0.61DL length, cm 46 ± 2 49 ± 2 0.42DL diameter, cm 0.72 ± 0.04 0.73 ± 0.03 0.89DL area, cm2 53 ± 3 56 ± 2 0.38Mucosa weight, g 11 ± 1 10 ± 1 0.61Mucosa/DL area, g/cm2 0.21 ± 0.01 0.19 ± 0.02 0.30Mitochondrial protein, mg/mL 2.1 ± 0.2 2.3 ± 0.2 0.46

1 For details see Materials and Methods. Values are means ± SEM ofn shown in parentheses.

and Davies (1986). The purity of isolated mitochondriawas evaluated for each mitochondrial preparation bymeasuring the activity of citrate synthase (mitochondrialmarker) according to Srere (1969), with modifications.Citrate synthase activities (U/mg of protein) were 257± 23 and 265 ± 20 for high and low FE mitochondria,respectively. As mitochondrial protein and values of ci-trate synthase activity were not different between the highand low FE birds, the purity of mitochondrial preparationwas considered to be similar between groups.

Assessment of Mitochondrial Function

Duodenal mitochondrial function was assessed bymonitoring oxygen consumption polarographically ac-cording to Estabrook (1967). Oxygen consumption rate(expressed as nmol of monomeric oxygen/min per milli-gram of protein) was measured with a Clark-type oxygenelectrode6 equipped with a magnetic stirrer and thermo-statically controlled respiration chamber set at 37°C.Equal amounts (based on protein concentration) of freshlyisolated mitochondria were added to the respirationchamber containing 900 µL of RCR reaction buffer (220mM D-mannitol, 70 mM sucrose, 2 mM HEPES, 2.5 mMKH2PO4, 2.5 mM MgCl2, 0.75 mM EDTA, 0.5 mM, and0.13 g BSA/100 mL, pH 7.4) and 10 mM each of succinate(an FADH2-linked energy substrate), pyruvate, or malateor 10 mM malate and 2.5 mM pyruvate (NADH-linkedenergy substrates).

In the presence of respiration buffer and an energysubstrate, isolated mitochondria exhibit an initial slowrate of oxygen uptake (state 2). Upon addition of adeno-sine diphosphate (ADP), an immediate increase in oxygenutilization is observed as ADP is converted to ATP (state3 or active respiration). As ADP levels become limitingfollowing conversion to ATP, mitochondria enter restingrespiration (state 4). The respiratory control ratio (RCR),an index of respiratory chain coupling, is calculated bydividing state 3 by state 4 respiration rate. The ADP tooxygen ratio (ADP:O ratio), an index of oxidative phos-phorylation or the efficiency of ATP synthesis coupled tocell respiration, is determined by dividing the quantityof ADP added by the amount of oxygen consumed duringactive (state 3) respiration (Estabrook, 1967).

Mitochondrial function was assessed with sequentialadditions of ADP (220 and 440 µM) to mimic a repeated

FEED EFFICIENCY AND MITOCHONDRIAL FUNCTION 1397

FIGURE 2. Diagrammatic representation of the electron transportshowing sites of chemical inhibition used in this study. Electron (e-)transfer was blocked at complex I by rotenone, at complex II by malonate(competitive inhibitor of succinate dehydrogenase) and 4,4,4-trifluoro-1-[2-thienyl]-1,3-butanedione (TTFA; complex II to ubiquinone), and atcomplex III by myxothiazol (center o) and antimycin A (center i). If asite-specific defect exists in the electron transport at any of these sitesof chemical inhibition, electrons will leak (dotted arrows) from therespiratory chain and consume oxygen by univalent reduction thatresults in the formation of superoxide (O2

�) that, in turn, can be convertedto reactive oxygen species such as H2O2. CoQ = coenzyme Q; SOD =superoxide dismutase.

demand for energy as previously described (Cawthon etal., 2001; Iqbal et al., 2001). Mitochondrial preparationswith an RCR of less than 2.0 (from 1 bird only) indicatingpoorly coupled mitochondria (Estabrook, 1967) were notconsidered viable and were excluded from the experi-ment. All functional measurements were made in tripli-cate and averaged and were completed within 3 h ofmitochondrial isolation.

Mitochondrial H2O2 Productionand Site-Specific Defectsin Electron Transport

Figure 2 represents the electron transport chain (ETC)showing sites of chemical inhibition used to determinesites and amounts of electron leak in this study. An in-crease in H2O2 generation following chemical inhibitionindicates that the site of electron leakage is between thesite of inhibition and entry of substrate into the electrontransport chain (Barja, 1999). In the presence of FADH2-linked substrate, electron transfer from complex II to ubi-quinone was inhibited with 4,4,4-trifluoro-1-[2-thienyl]-1,3-butanedione (TTFA). Malonate, a competitive inhibi-tor of succinate dehydrogenase, was also used to inhibitthe entry of electrons to complex II. With NADH-linkedsubstrates, rotenone was used to block electron transferfrom complex I to ubiquinone, and myxothiazol and anti-mycin A inhibited electron flow into center o (toward theouter mitochondrial membrane) and center i (toward the

7Molecular Probes, Eugene, OR (www.probes.com).8Bio-Tek Instruments, Inc., Winooski, VT.9SAS Institute Inc., Cary, NC.

inner mitochondrial membrane) of complex III, respec-tively. If a site-specific defect exists in the electron trans-port at any of these sites of chemical inhibition, electronswill leak from the respiratory chain and consume oxygenby univalent reduction that results in the formation ofsuperoxide (O2

�-). In the presence of superoxide dismu-tase, O2

�- is dismutated to H2O2 (Chance et al., 1979).The mitochondrial production of H2O2 was measured

using dichlorofluorescin diacetate (DCFH-DA)7 fluores-cent probe as previously described (Bass et al., 1983; Iqbalet al., 2001). DCFH-DA is a nonpolar compound thatcan readily diffuse into cells, where it is hydrolyzed byesterases to the nonfluorescent dichlorofluorescin(DCFH). The DCFH is then oxidized by H2O2 to a fluores-cent product, dichlorofluorescein, which fluoresces whenexposed to ultraviolet light in the presence of H2O2 andperoxidase (Rota et al., 1999; Bilski et al., 2002). The in-crease in dichlorofluorescein fluorescence was found tobe highly correlated with H2O2 concentrations (Iqbal etal., 2001). Briefly, 45 µL of mitochondrial sample (2 mg/mL) was added in a 96-well microtiter plate containing48 µL of H2O2 buffer (145 mM KCl, 30 mM HEPES, 5 mMKH2PO4, 3 mM MgCl2, and 0.1 mM EGTA; pH 7.4), 52µM DCFH-DA, and 10 mM each of succinate, malate, orpyruvate or 10 mM malate and 2.5 mM pyruvate as en-ergy substrates. Superoxide dismutase (10 U/well) wasadded to each well to convert all O2

�- to H2O2 with valuescorrected for blanks and residual fluorescence by catalase(Iqbal et al., 2001). Final concentrations of inhibitors usedwere: 10 µM of rotenone; 13 µM each of antimycin A,myxothiazol and TTFA; and 7 µM of malonate. The reac-tion mixture was incubated at 37°C, and the change influorescence was recorded for 20 min with a FLX800 Mi-croplate Fluorescence Reader8 set at a sensitivity of 110and excitation/emission wavelengths of 485 and 528 nm,respectively. H2O2 values were calculated from a stan-dard curve with known amounts of H2O2, and resultswere expressed as nanomoles of H2O2 per minute permilligram of mitochondrial protein.

Statistical Analyses

Figure 3 was analyzed with regression analysis and allother data were analyzed with one-way ANOVA usingJMP 5.0 statistical software.9 Means were separated byStudent’s t-test, and data are presented as the mean ±SEM. With the H2O2 data, multiple comparisons wereperformed with the same mean and errors were calcu-lated from the observed errors on the original values, anderrors were also reported as the standard error of themeans. A probability level of P ≤ 0.05 was consideredsignificant unless stated otherwise.

RESULTS

Growth Performanceand Duodenal Loop Data

Initial BW (at 6 wk) was not different, but the high FEbirds were heavier at 7 wk due to faster growth rate with

OJANO-DIRAIN ET AL.1398

FIGURE 3. Relationships between 6 to 7 wk total feed consumptionand body weight gain in male broiler breeders with high (n = 9) andlow (n = 9) feed efficiency (FE). Regression equations shown weresignificant (P < 0.05).

no differences in feed intake (P = 0.79) resulting in higher6 to 7 wk FE (g of gain:g of feed) in the high FE than inthe low FE group (Table 1). Feed conversion ratios (FCR,g feed:g gain) are also shown in Table 1. Figure 3 illus-trates the dependency of body weight gain on feed intakeand indicates that low FE birds needed to consume about100 to 150 g more feed to gain the same BW as the highFE birds. There were no differences in the weight, length,diameter, area, and mucosa weight of the duodenum orin mitochondrial protein (Table 2).

Mitochondrial Function

Values for respiration rates, expressed as natoms ofoxygen per minute per milligram of protein are providedin Table 3. With succinate, an FADH2-linked substrate,respiration rates were not different after the first additionof ADP. A lower state 4 respiration rate was observed inhigh FE duodenum mitochondria following the secondaddition of ADP. When NADH-linked substrates (malate,malate-pyruvate, and pyruvate) were used as electron

TABLE 3. Mitochondrial oxygen consumption during state 2, 3, and 4 respiration after sequential addition of adenosine diphosphate (ADP,1, 2) in duodenum mitochondria from broilers with high and low feed efficiency (FE) provided with succinate, malate,

pyruvate or pyruvate-malate as energy substrates1

Succinate Malate Pyruvate-malate Pyruvate

Variable High FE Low FE High FE Low FE High FE Low FE High FE Low FE

(natoms O/min per mg of protein)State 2 27.6 ± 2.1 22.0 ± 2.6 8.4 ± 0.2 7.5 ± 0.4 9.1 ± 1.2 8.2 ± 0.4 10.7 ± 1.1 8.7 ± 0.4State 3 (1) 74.4 ± 7.1 60.3 ± 7.9 31.9 ± 4.3 29.2 ± 3.5 28.7 ± 6.1 24.9 ± 3.8 26.3 ± 5.5 26.7 ± 3.1State 3 (2) 76.4 ± 8.9 66.4 ± 8.1 32.9 ± 4.8 30.4 ± 4.7 30.1 ± 6.8 25.2 ± 4.6 28.2 ± 7.0 31.1 ± 5.5State 4 (1) 22.4 ± 1.8 18.3 ± 1.6 7.2 ± 0.9 5.8 ± 0.5 7.3 ± 1.2 6.2 ± 0.4 6.4 ± 0.8 5.3 ± 0.6State 4 (2) 8.7 ± 1.1b 11.5 ± 0.7a 3.6 ± 0.7 3.3 ± 0.5 4.1 ± 0.8 3.3 ± 0.3 3.4 ± 0.8 3.2 ± 0.3

a,bMean respiration values within an energy substrate with different letters are significantly different (P < 0.05).1 For details, see Materials and Methods. Values represent the mean ± SEM for high (n = 6) and low (n = 7) FE broilers. Active (state 3) and

resting (state 4) respiration rates were measured after the first (1) and second (2) addition of ADP in duodenum mitochondria provided withsuccinate, malate, pyruvate-malate, or pyruvate as energy substrates.

donors, no differences were observed in respiration ratesbetween the 2 groups.

Mitochondrial function data are provided in Table 4.Values are shown for initial assessment of electron trans-port chain coupling (RCR 1) and oxidative phosphoryla-tion (ADP:O 1) and after sequential additions of ADP(RCR 2 and ADP:O 2). When duodenal mitochondriawere provided succinate, which donates electrons to therespiratory chain at complex II, there were no differencesin RCR 1 values, but coupling was improved after thesecond addition of ADP (RCR 2) and was greater in thehigh FE compared with low FE mitochondria (P < 0.001).With NADH-linked substrates, there were no differencesin RCR 1 or RCR 2 values between the low and highFE mitochondria. However, after the second addition ofADP, regardless of substrate used, RCR was improvedabout 2-fold compared with that after the first additionin the high and low FE mitochondria. The acceptor controlratio (state 3/state 2), which is similar to RCR, was alsonot different between the 2 groups with either substrate(data not shown).

No differences were observed in ADP:O 1 between the2 groups with either substrate. As with the RCR values,the ADP:O ratio was also improved (33 and 21% withFADH2-linked and NADH-linked substrates, respec-tively) following the second addition of ADP (ADP:O 2).The ADP:O 2 values were higher with malate, malate-pyruvate, and pyruvate in low FE compared with highFE mitochondria. There were no differences in ADP:O 2values between groups when succinate was provided asthe energy substrate.

Mitochondrial H2O2 Production

In this paper, the terms electron leak and H2O2 produc-tion are used synonymously. Results of duodenal mito-chondrial H2O2 production with FADH2 (succinate) andNADH (malate, pyruvate, and malate-pyruvate)-linkedsubstrates are shown in Figure 4. Basal H2O2 productionis represented in mitochondria that were provided withno inhibitor (NI). As shown in Figure 4A, basal H2O2production was marginally greater (P < 0.08) in low thanin high FE mitochondria provided succinate. Inhibition

FEED EFFICIENCY AND MITOCHONDRIAL FUNCTION 1399

TABLE 4. Mitochondrial function in duodenal mitochondria from broilers with high and low feed efficiency (FE)1

Succinate Malate Malate-pyruvate Pyruvate

Variable High FE Low FE High FE Low FE High FE Low FE High FE Low FE

RCR 1* 3.41 ± 0.32* 3.40 ± 0.32* 4.65 ± 0.47* 4.77 ± 0.31* 4.00 ± 0.52* 3.90 ± 0.48* 4.20 ± 0.40* 5.00 ± 0.50*RCR 2 8.89 ± 0.26a 6.77 ± 0.81b 9.77 ± 0.85 9.1 ± 0.46 7.24 ± 0.71 7.47 ± 0.81 9.11 ± 0.84 8.87 ± 0.45ADP:O 1† 1.19 ± 0.07† 1.25 ± 0.13‡ 2.35 ± 0.05† 2.67 ± 0.15‡ 2.21 ± 0.18† 2.53 ± 0.22‡ 2.33 ± 0.21† 2.62 ± 0.15†ADP:O 2 1.72 ± 0.19 1.57 ± 0.10 2.83 ± 0.05b 2.97 ± 0.03a 2.79 ± 0.08b 2.97 ± 0.02a 2.81 ± 0.05b 2.98 ± 0.02a

a,bMeans within rows are significantly different between FE group within an energy substrate (P < 0.05).1For details, see Materials and Methods. Values represent the mean ± SEM for high (n = 6) and low (n = 7) FE broilers. Functional measurements

include the respiratory control ratio (RCR) and the adenosine diphosphate (ADP):O ratio, after first (1) and second (2) addition of ADP in duodenalmitochondria provided with succinate, malate, pyruvate, or malate-pyruvate as energy substrates.

2The ADP:O 2 were significantly higher than ADP:O 1 within each column (†P < 0.05, ‡P < 0.08).*The RCR 1 values were significantly lower than RCR 2 values within each column (P < 0.005).

of complex II with malonate and TTFA resulted in higherH2O2 values in low FE compared with high FE mitochon-dria, but there were no differences between groups withinhibition of complex III with antimycin A and myxothia-zol. Low FE mitochondria treated with TTFA showed anincrease (P < 0.06) in H2O2 production compared with itsbasal H2O2 values.

The basal level of H2O2 production in low FE mitochon-dria was greater than high FE values with malate (Figure4B), malate-pyruvate (Figure 4C) and pyruvate (Figure4D). When mitochondria oxidized malate, inhibition ofelectron transport at complex I with rotenone and at com-plex III with antimycin A (center i) and myxothiazol (cen-ter o) significantly increased H2O2 production in high FEmitochondria. Low FE mitochondrial H2O2 values withcomplexes I and III inhibition were also numerically in-creased compared with the basal values, but the increasein magnitude was not significant. When the 2 groupswere compared within an inhibitor treatment, H2O2 val-ues were higher (P < 0.07) in low FE compared with highFE mitochondria with complex I inhibition (Figure 4B).With malate-pyruvate, increases in electron leak abovebasal levels were observed in both groups with rotenone,antimycin A, and myxothiazol inhibition, but there wereno differences between groups within inhibitor (Figure4C). With pyruvate as an energy substrate, H2O2 produc-tion was increased above basal values in both groupswith rotenone and antimycin A inhibition but not withmyxothiazol. The H2O2 values were also higher (P < 0.06)in low FE compared with high FE mitochondria aftercomplex I inhibition with rotenone (Figure 4D). We alsoobserved that the magnitude of H2O2 production was 10-fold lower in mitochondria provided pyruvate (Figure4D) compared with malate (Figure 4B), with the pyruvate-malate combination (Figure 4C) attenuating electron leak.

DISCUSSION

Cellular energy (ATP) generated in mitochondria viaoxidative phosphorylation fuels nutrient absorption andtransport in the small intestine. A considerable amountof energy is used by the gut in carrying out its functionand for the maintenance of this tissue because the intesti-nal epithelium is continuously renewed. New cells are

produced in the crypts, migrate to villus tip, and aresloughed off in 3 to 5 d. McBride and Kelly (1990) esti-mated that 11 to 18% of the whole energy expenditurein ruminants is used by the gut, and the majority of thisenergy is spent for Na+-K+-ATPase activity (6 to 12%)and protein synthesis (4.0 to 4.6%). For this reason, theabsorptive capacity of the small intestine is directly influ-enced by the availability of ATP to fuel Na+-K+-ATPaseand to renew epithelial cells. The gastrointestinal tractconsumes a large quantity of energy and uses a largeproportion of the body’s oxygen supply. The contributionof the gastrointestinal tract to the whole body oxygenconsumption has been reported to be approximately 20%in ruminants (Webster, 1980; Huntington and McBride,1988), 25% in pigs (Yen et al., 1989), and 6 to 8% in broilerbreeder hens (Spratt et al., 1990). Hence, inefficiencies inmitochondrial function such as increased mitochondrialROS production may limit the amount of nutrients ab-sorbed and may reduce the efficiency of converting feedinto demand tissues or eviscerated body mass.

Unlike reports demonstrating the impact of breed (e.g.,Mukherjee et al., 1970; Dziewiecki and Kolataj, 1976;Brown et al., 1986) or diet (e.g., Renner et al., 1979; Toyo-mizu et al., 1992a, b) on mitochondrial function, experi-ments conducted by Bottje et al. (2002), by Lutz and Stahly(2003), and in this study used animals from the samegenetic line and provided the same diet. Similar to previ-ous findings (Bottje et al., 2002), the high FE birds in theseexperiments showed higher body weight gain at 7 wk dueto greater FE. Evaluation of duodenum weight, length,diameter, area, or mucosa weight did not show differ-ences between the low and high FE broilers. Thus, physi-cal attributes of the intestines could not account fordifferences in mitochondrial function or H2O2 productionobserved in this study.

Mitochondrial function in isolated duodenal mitochon-dria was studied using established indices of mitochon-drial function: RCR, for respiratory chain coupling andADP:O ratio for oxidative phosphorylation (Estabrook,1967). The RCR and ADP:O ratio was improved in bothgroups following a sequential addition of ADP (repeatedenergy demand), and these improvements were similarto observations in liver and lung mitochondria (Cawthonet al., 2001; Iqbal et al., 2001). However, with succinate,

OJANO-DIRAIN ET AL.1400

FIGURE 4. The H2O2 production (nmols/min per mg of mitochondrial protein) in duodenal mitochondria obtained from broilers with high(shaded bars) and low (open bars) feed efficiency (FE) and provided with different energy substrates: A) succinate, B) malate, C) malate-pyruvate,and D) pyruvate. The H2O2 production was measured with a 2′,7′-dichlorofluorescin diacetate (DCFH-DA) probe as described in the Materialsand Methods. Each bar represents the mean ± SEM for high (n = 6) and low (n = 5) FE birds. a,bBars with different letters within an inhibitorgroup are significant between FE groups (P < 0.05, or as stated). *Values differ from no inhibitor (NI) values within FE group (P < 0.05).†Values differ from no inhibitor (NI) values within FE group (P ≤ 0.07). AA = antimycin A; MXO = myxothiazol; MALO = malonate; TTFA =thenoyltrifluroacetone; Rot = rotenone.

low FE mitochondria exhibited a lower RCR after thesecond addition of ADP due to a higher state 4 (resting)respiration rate, indicating that electron transport was lesstightly coupled in the low FE than in high FE duodenalmitochondria. Lower RCR values as a result of increasedstate 4 respiration rate was also observed in muscle andliver mitochondria (Davies et al., 1982) and heart mito-chondria (Ji and Mitchell, 1994) after exercise, which sug-gested possible inner mitochondrial membrane leakage(Ji, 1999). During exercise demand for oxygen consump-tion greatly increases. As the gut also consumes an enor-mous amount of oxygen and as there was greater electronleak in low FE mitochondria provided succinate (Figure4A), there may be similar mechanisms in mitochondrial

membrane leakage during exhaustive exercise and inhighly metabolically active mitochondria found in intesti-nal tissue. With NADH-linked substrates, the RCR valueswere not different between groups either after the firstor second addition of ADP. This result is in contrast tothe findings of Bottje et al. (2002) who reported that RCR1 was lower in low FE breast muscle mitochondria. Re-sults of the functional studies suggest that low FE duode-nal mitochondria exhibit inefficiency in respiratory chaincoupling associated with complex II.

Despite the higher RCR 2 value observed in high FEmitochondria oxidizing succinate, there were no differ-ences in the ADP:O values between the high and low FEduodenal mitochondria. The ADP:O ratio after the second

FEED EFFICIENCY AND MITOCHONDRIAL FUNCTION 1401

addition of ADP, however, was significantly higher inlow FE mitochondria with malate, malate-pyruvate, andpyruvate as energy sources. Thus, low FE duodenal mito-chondria exhibited an ability to carry out oxidative phos-phorylation that was equal or superior to that of high FEmitochondria depending upon the energy substrate. Weinitially hypothesized that the higher ADP:O ratio in lowFE mitochondria after the second addition of ADP withNADH-linked substrates might be a compensatory mech-anism for a respiratory chain defect. Furthermore, themechanism responsible for the improvement of RCR andADP:O ratio in low and high FE mitochondria after thesequential ADP addition is also not clear at this time. Butas differences in low and high FE mitochondrial functionwere observed only after the second addition of ADP,it appears that the 2 groups respond differently duringrepeated demand for energy. More experiments will beconducted to help us understand the basis of these obser-vations.

Generation of ROS is associated with the normal func-tioning of mitochondria. Impairment of the mitochondrialrespiratory chain leading to increased electron leak (Weiet al., 2001) and increased ROS level pose a serious threatto the cellular antioxidant defense system and increase thesusceptibility of various cellular components to oxidativedamage (Kristal et al., 1997; Ji, 1999). Similar to previousresults in muscle (Bottje et al., 2002), basal H2O2 produc-tion was greater in low than in high FE duodenal mito-chondria with either NADH- or FADH2-linked substrate(see Figure 4). This may in turn predispose the low FEmitochondria to greater oxidative stress. With succinate,inhibiting electron transfer from complex II to ubiquinonewith TTFA raised H2O2 levels above basal values in lowFE mitochondria, and these values were higher comparedwith those of high FE mitochondria. Low FE mitochon-dria also exhibited higher H2O2 levels compared withhigh FE mitochondria with malonate, another complex IIinhibitor, but inhibited values were not elevated abovebasal levels. A possible explanation for this differenceis that, as an analog of succinate, malonate is a strongcompetitive inhibitor of succinate dehydrogenase thatblocks the tricarboxylic acid cycle (Koeppen and Riley,1987; Lehninger et al., 1993) and, therefore, would not beexpected to raise H2O2 values above basal levels. Theresults of site-specific defects evaluation with succinatesuggest that the lower RCR and higher state 4 respirationafter the second ADP addition observed in low FE mito-chondria may be due to the higher electron leak withincomplex II of the electron transport chain. Earlier studies(Iqbal et al., 2001; Bottje et al., 2002; Tang et al., 2002) alsosuggested that lower coupling might be associated withhigher electron leak.

On the other hand, when complex I or NADH-linkedsubstrates were used as an energy source, both the highand low FE mitochondria exhibited increased H2O2 pro-duction after inhibition of electron transport at complexI with rotenone and within center i and center o of com-plex III with antimycin A and myxothiazol, respectively.These results indicated that site-specific defects at com-

plexes I and III in the electron transport chain appear tooccur in both the high and low FE mitochondria. Theseresults differ from findings of Bottje et al. (2002) in high FEmuscle mitochondria that did not show increased H2O2

production after inhibition of electron transport at com-plexes I and III. This difference in terms of ROS produc-tion between muscle and intestine may be due to thephysiological function of these tissues. Several authorshave suggested that ROS can induce apoptosis in manydifferent cell systems (e.g., Simon et al., 2000). A similarROS-mediated mechanism may be instrumental in medi-ating rapid turnover of intestinal epithelial cells, whichwould not be desirable in muscle tissue. The magnitudeof H2O2 production at complex I with malate or pyruvateand at complex II with succinate, however, was signifi-cantly greater in the low FE compared with high FE duo-denal mitochondria, which indicated the possibility ofgreater site-specific defects in low FE duodenal mitochon-dria. Attenuation of H2O2 production at complex I wasobserved when a combination of malate and pyruvatewas used as energy substrate. The higher ADP:O ratioafter the second addition of ADP observed in low FEmitochondria with malate or pyruvate but not with ma-late-pyruvate combination may indeed be a compensa-tory mechanism in low FE mitochondria as a result ofhigher electron leak or H2O2 production with these sub-strates.

The well-known sites of O2�- and H2O2 generation are

complexes I and III of the electron transport chain (Chanceet al., 1979; Turrens and Boveris, 1980; Nohl et al., 1996).Site-specific defects have been observed at complexes Iand III in lung, heart, and breast muscle mitochondria inbroilers with pulmonary hypertension syndrome (Iqbalet al., 2001; Tang et al., 2002) as well as in low FE breastmuscle mitochondria (Bottje et al., 2002). Conversely,Cawthon et al. (2001) reported site-specific defects at com-plex II in liver mitochondria from PHS broilers. In thecurrent study, site-specific defects in electron transportwere observed at complexes I, II and/or III (depending onsubstrate) in the high and low FE duodenal mitochondria.Combined, these results imply that the site of electronleak in the mitochondrial respiratory chain may be tissuedependent as suggested earlier by Kwong and Sohal(1998) and may reflect the metabolic function of the tissue.

The H2O2 data also suggest that electron leak in duode-nal mitochondria may depend on the energy substrate. Inaddition, H2O2 values with pyruvate were 10-fold lowercompared with succinate, malate, or malate-pyruvate inboth FE groups. Because substrates used in this studywere chosen based on acceptable RCR values during testruns, the lower magnitude of H2O2 with pyruvate wasunexpected. However, in studies on lymphoid cell lines(Miwa et al., 2000) and in inflammatory joint diseases(Herz et al., 1997), it has been reported that pyruvatecan directly scavenge H2O2 and protect cells from ROS-mediated oxidative damage. In plant mitochondria, pyr-uvate can also inhibit H2O2 formation by activating thealternative oxidase (Braidot et al., 1999). Thus, lower lev-els of H2O2 formation with pyruvate observed in this

OJANO-DIRAIN ET AL.1402

study may be due to its action as an antioxidant orthrough its ability to stimulate other antioxidant defensesin the cell.

In summary, the results of these experiments demon-strated that mitochondrial electron transport chain cou-pling was higher in high FE mitochondria followingrepeated demand for energy with FADH2- but not withNADH-linked substrates. However, with NADH-linkedsubstrates low FE duodenum mitochondria exhibited anability to synthesize ATP in vitro that was superior tothat in high FE mitochondria. Basal H2O2 production,as an indicator of electron leak, was higher in low FEmitochondria, but site-specific defects in electron trans-port were present in both low and high FE duodenalmitochondria at complexes I and III. The magnitude ofsite-specific defect at complexes I and II, however, wassignificantly higher in the low FE mitochondria, sug-gesting that increased electron leak at these sites wasassociated with low feed efficient broilers. The findingsin this study provide insight into understanding the rela-tionship between FE and mitochondrial function in duo-denal tissue. A better understanding of cellular andbiochemical mechanisms associated with FE could helpin developing tools for aiding selection programs in iden-tifying breeding replacement stock with superior FE.

ACKNOWLEDGMENTS

The authors thank H. Brandenburger for technical edit-ing, M. Davis for proofreading this manuscript, and M.Toyomizu (Tohoku University, Japan) for technical ad-vice regarding procedures associated with isolation ofduodenal mitochondria. This research was presented inpart at the Southern Poultry Science Society Meeting inAtlanta, GA (January 2003).

REFERENCES

Barja, G. 1999. Kinetic measurement of mitochondrial oxygenradical production. Pages 533–534 in Methods in Aging Re-search. P. B. Yu, ed. CRC Press LLC, Washington, DC.

Bass, D. A., J. W. Parce, L. R. Dachatelet, P. Szejda, M. C. Seeds,and M. Thomas. 1983. Flow cytometric studies of oxidativeproduct formation by neutrophils: A graded response tomembrane stimulation. J. Immunol. 130:1910–1915.

Benzi, G., and A. Moretti. 1995. Age- and peroxidative stress-related modifications of the cerebral enzymatic activitieslinked to mitochondria and glutathione system. Free Radic.Biol. Med. 19:77–101.

Bilski, P., A. G. Belanger, and C. F. Chignell. 2002. Photosensiti-tized oxidation of 2′,7′-dichlorofluorescin: Singlet oxygendoes not contribute to the formation of fluorescent oxidationproduct 2′,7′-dichlorofluorescein. Free Radic. Biol. Med.33:938–946.

Bottje, W., M. Iqbal, Z. X. Tang, D. Cawthon, R. Okimoto, T.Wing, and M. Cooper. 2002. Association of mitochondrialfunction with feed efficiency within a single genetic line ofmale broilers. Poult. Sci. 81:546–555.

Boveris, A., and B. Chance. 1973. The mitochondrial generationof hydrogen peroxide. Biochem. J. 134:707–711.

Braidot, E., E. Petrussa, A. Vianello, and F. Macri. 1999. Hydro-gen peroxide generation by higher plant mitochondria oxi-dizing complex I or complex II substrates. FEBS Lett.451:347–350.

Brown, D. R., S. K. DeNise, and R. G. McDaniel. 1986. Hepaticmitochondrial activity in two breeds of chicken. Poult. Sci.65:613–615.

Cawthon, D., K. Beers, and W. G. Bottje. 2001. Electron transportchain defect and inefficient respiration may both underliepulmonary hypertension syndrome (PHS)-associated mito-chondrial dysfunction in broilers. Poult. Sci. 80:474–484.

Cawthon, D., R. McNew, K. W. Beers, and W. G. Bottje. 1999.Evidence of mitochondrial dysfunction in broilers with pul-monary hypertension syndrome (ascites): Effect of t-butylhydroperoxide on function, glutathione and related thiols.Poult. Sci. 78:114–125.

Chance, B., H. Sies, and A. Boveris. 1979. Hydroperoxide metab-olism in mammalian organs. Physiol. Rev. 59:527–605.

Chapman, H. D., Z. B. Johnson, and J. L. McFarland. 2003.Improvements in the performance of commercial broilers inthe USA: Analysis for the years 1997 to 2001. Poult. Sci.82:50–53.

Davies, K. J. A., T. A. Quintanilha, G. A. Brooks, and L. Packer.1982. Free radical and tissue damage produced by exercise.Biochem. Biophys. Res. Commun. 107:1198–1205.

Dziewiecki, C., and A. Kolataj. 1976. Rate of oxygen uptake byliver mitochondria in purebred chickens and in their hybrids.Genet. Pol. 17:219–224.

Emmerson, D. A. 1997. Commercial approaches to genetic selec-tion for growth and feed conversion in domestic poultry.Poult. Sci. 76:1121–1125.

Estabrook, R. W. 1967. Mitochondrial respiratory control andthe polarographic measurement of ADP:O ratios. MethodsEnzymol. 10:41–47.

Fiegal, R. J., and B. L. Shapiro. 1979. Mitochondrial calciumuptake and oxygen consumption in cystic fibrosis. Nature.278:276–277.

Havenstein, G. B., P. R. Ferket, and M. A. Qureshi. 2003. Growth,livability, and feed conversion of 1957 versus 2001 broilerswhen fed representative 1957 and 2001 broiler diets. Poult.Sci. 82:1500–1508.

Havenstein, G. B., P. R. Ferket, S. E. Scheidler, and B. T. Larson.1994. Growth, livability, and feed conversion of 1957 and1991 broilers when fed ‘typical’ 1957 versus 1991 broiler diets.Poult. Sci. 73:1785–1794.

Herrero, A., and G. Barja. 1998. Hydrogen peroxide productionof heart mitochondria and aging rate are slower in canariesand parakeets than in mice: sites of free radical generationand mechanisms involved. Mech. Aging Dev. 103:133–146.

Herz, H., D. R. Blake, and M. Grootveld. 1997. Multicomponentinvestigations of the hydrogen peroxide- and hydroxyl radi-cal-scavenging antioxidant capacities of biofluids: the rolesof endogenous pyruvate and lactate. Relevance to inflamma-tory joint diseases. Free Radic. Res. 26:19–35.

Huntington, G. B., and B. W. McBride. 1988. Ruminant splanch-nic tissues. Energy cost of absorption and metabolism. Pages313–327 in Biomechanisms Regulating Growth and Develop-ment. G. L. Steffens and T. S. Rumsey, ed. Beltsville Symposiain Agricultural Research, Beltsville, MD.

Iqbal, M., D. Cawthon, R. F. Wideman, Jr., and W. G. Bottje. 2001.Lung mitochondrial dysfunction in pulmonary hypertensionsyndrome. I. Site-specific defects in electron transport chain.Poult. Sci. 80:485–495.

Iqbal, M., N. Pumford, K. Lassiter, Z. Tang, T. Wing, M. Cooper,and W. G. Bottje. 2004. Low feed efficient broilers within asingle genetic line exhibit higher oxidative stress and proteinexpression in breast muscle with lower mitochondrial com-plex activity. Poult. Sci. 83:474–484.

Ji, L. L. 1999. Antioxidants and oxidative stress in exercise. Proc.Soc. Exp. Biol. Med. 222:283–292.

Ji, L. L., and E. W. Mitchell. 1994. Effects of adriamycin on heartmitochondrial function in rested and exercised rats. Biochem.Pharmacol. 47:877–885.

Koeppen, A. H., and K. M. Riley. 1987. Effect of free malonateon the utilization of glutamate by rat brain mitochondria. J.Neurochem. 48:1509–1515.

FEED EFFICIENCY AND MITOCHONDRIAL FUNCTION 1403

Kristal, B., S. Koopmans, C. T. Jackson, Y. Ikeno, B. Par, and B.P. Yu. 1997. Oxidant-mediated repression of mitochondrialtranscription in diabetic rats. Free Radic. Biol. Med.22:813–822.

Kwong, L. K., and R. S. Sohal. 1998. Substrate and site specificityof hydrogen peroxide generation in mouse mitochondria.Arch. Biochem. Biophys. 350:118–126.

Lawrence, C. B., and N. T. Davies. 1986. A novel, simple andrapid method for the isolation of mitochondria which exhibitrespiratory control from rat small intestinal mucosa. Biochim.Biophys. Acta 848:35–40.

Lehninger, A. L., D. L. Nelson, and M. M. Cox. 1993. Principlesof Biochemistry. 2nd ed. Worth Publishers, New York.

Lutz, R. T., and T. S. Stahly. 2003. Quantitative relationshipbetween mitochondrial bioenergetics and efficiency of ani-mal growth. J. Anim. Sci. 81(Suppl. 1):141. (Abstr.)

Madesh, M., A. Ramachandran, A. Pulimood, M. Vadranam,and K. A. Balusubramanian. 2000. Attenuation of intestinalischemia/reperfusion injury with sodium nitroprusside:studies on mitochondrial function and lipid changes. Bio-chim. Biophys. Acta 1500:204–216.

McBride, B. W., and J. M. Kelly. 1990. Energy cost of absorptionand metabolism in the ruminant gastrointestinal tract andliver. A review. J. Anim. Sci. 68:2997–3010.

Miwa, H., J. Fujii, H. Kanno, N. Taniguchi, and K. Aozasa. 2000.Pyruvate secreted by human lymphoid cell lines protects cellsfrom hydrogen peroxide mediated cell death. Free Radic. Res.33:45–56.

Mukherjee, T. K., R. W. C. Stevens, and M. P. Hoogendoorn.1970. Oxygen uptake of mitochondrial isolates from twobreeds of chickens and their F1 cross. Poult. Sci. 49:1130–1131.

Nohl, H., L. Gille, K. Schonheit, and Y. Liu. 1996. Conditionsallowing redox-cycling of ubisemiquinone in mitochondriato establish a direct redox couple with molecular oxygen.Free Radic. Biol. Med. 20:207–213.

Renner, R., S. M. Innis, and M. T. Clandinin. 1979. Effects ofhigh and low erucic acid rapeseed oils on energy metabolismand mitochondrial function of the chick. J. Nutr. 109:378–387.

Rota, C. C., C. F. Chignell, and R. P. Mason. 1999. Evidence forfree radical formation during the oxidation of 2′-7′-dichloro-fluorescin to the fluorescent dye 2′-7′- dichlorofluorescein byhorseradish peroxidase: Possible implications for oxidativestress measurements. Free Radic. Biol. Med. 27:873–881.

Rustin, P., and A. Rotig. 2001. Inborn errors of complex II—Unusual human mitochondrial diseases. Biochim. Biophys.Acta 1553:117–122.

Shigenaga, M. K., T. M. Hagen, and B. N. Ames. 1994. Oxidativedamage and mitochondrial decay in aging. Proc. Natl. Acad.Sci. USA 91:10771–10778.

Simon, H. U., A. Haj-Yehia, and F. Levi-Schaffer. 2000. Role ofreactive oxygen species in apoptosis induction. Apoptosis5:415–418.

Spratt, R. S., B. W. McBride, H. S. Bayley, and S. Leeson. 1990.Energy metabolism of broiler breeder hens. 2. Contributionof tissues to total heat production in fed and fasted hens.Poult. Sci. 69:1348–1356.

Srere, P. A. 1969. Citrate synthase. Methods Enzymol. 13:3–26.Tang, Z., M. Iqbal, D. Cawthon, and W. Bottje. 2000. Defects in

heart and breast muscle mitochondrial electron transport ofbroilers with pulmonary hypertension syndrome. Free Radic.Biol. Med. 29:S25.

Tang, Z., M. Iqbal, D. Cawthon, and W. Bottje. 2002. Heartand breast muscle mitochondrial dysfunction in pulmonaryhypertension in broilers (Gallus domesticus). Comp. Biochem.Physiol. 132:527–540.

Toyomizu, M., D. Kirihara, M. Tanaka, K. Hayashi, and Y. Tom-ita. 1992a. Dietary protein level alters oxidative phosphoryla-tion in heart and liver mitochondria of chicks. Brit. J. Nutr.68:89–99.

Toyomizu, M., K. Mehara, T. Kamada, and Y. Tomita. 1992b.Effects of various fat sources on growth and hepatic mito-chondrial function in mice. Comp. Biochem. Physiol.101A:613–618.

Turrens, J. F., and A. Boveris. 1980. Generation of superoxideanion by the NADH dehydrogenase of bovine heart mito-chondria. Biochem. J. 191:421–427.

Webster, A. J. L. 1980. Energy cost of digestion and metabolismin the gut. Pages 469–84 in Digestive Physiology and Metabo-lism in Ruminants. Y. Ruckebusch and P. Thivend, ed. MTPPress, Lancaster, UK.

Wei, Y. H., C. F. Lee, H. C. Lee, Y. S. Ma, C. W. Wang, C. Y.Lu, and C. Y. Pang. 2001. Increases of mitochondrial massand mitochondrial genome in association with enhanced oxi-dative stress in human cells harboring 4,977 BP-deleted mito-chondrial DNA. Ann. NY Acad. Sci. 928:97–112.

Yen, J., J. Nienaber, D. Hill, and W. Pond. 1989. Oxygen con-sumption by portal vein-drained organs and by whole animalin conscious growing swine. Proc. Soc. Exp. Biol. Med.190:393–398.

Yu, B. P. 1994. Cellular defenses against damage from reactiveoxygen species. Physiol. Rev. 74:139–162.