Decreased Activities of Ubiquino1:Ferricytochrome c Oxidoreductase (Complex 111) and Ferrocytochrome c:Oxygen Oxidoreductase (Complex IV) in Liver Mitochondria from Rats with Hydroxycobalamin[c-lactam]-induced Methylmalonic Aciduria*

Rats treated with hydroxycobalamin[c-lactam] (HCCL), a cobalamin analogue that induces methylmalonic aciduria, have increased hepatic mitochon- drial content and increased oxidative metabolism of pyruvate and palmitate per hepatocyte. The present studies were undertaken to characterize oxidative metabolism in isolated liver mitochondria from rats treated with HCCL. After 5-6 weeks, state 3 oxidation rates for diverse substrates are reduced in mitochon- dria from HCCL-treated rats. Similar reductions of mitochondrial oxidation rates are obtained with dini-trophenol-uncoupled mitochondria excluding defective phosphorylation as a cause for the observed decrease in mitochondrial oxidation. The activities of mitochon- drial oxidases are reduced in HCCL-treated rats and demonstrate a defect in complex IV. Investigation of the complexes of the respiratory chain reveals a 32% decrease of ubiquino1:ferricytochrome c oxidoreduc- tase (complex 111) activity and a 72% decrease of ferrocytochrome c:oxygen oxidoreductase (complex IV) activity in mitochondria from 5-6-week HCCL- treated rats as compared with controls. Liver mitochondria from HCCL-treated rats also demonstrate de- creased cytochrome content per mg of mitochondrial protein


Decreased Activities of Ubiquino1:Ferricytochrome c Oxidoreductase (Complex 111) and Ferrocytochrome c:Oxygen Oxidoreductase (Complex IV) in Liver Mitochondria from Rats with Hydroxycobalamin[c-lactam]-induced Methylmalonic Aciduria*
(Received for publication, March 15, 1991) Stephan KrahenbuhlS, Mei Chang, Eric P. Brass Rats treated with hydroxycobalamin[c-lactam] (HCCL), a cobalamin analogue that induces methylmalonic aciduria, have increased hepatic mitochondrial content and increased oxidative metabolism of pyruvate and palmitate per hepatocyte. The present studies were undertaken to characterize oxidative metabolism in isolated liver mitochondria from rats treated with HCCL. After 5-6 weeks, state 3 oxidation rates for diverse substrates are reduced in mitochondria from HCCL-treated rats. Similar reductions of mitochondrial oxidation rates are obtained with dinitrophenol-uncoupled mitochondria excluding defective phosphorylation as a cause for the observed decrease in mitochondrial oxidation. The activities of mitochondrial oxidases are reduced in HCCL-treated rats and demonstrate a defect in complex IV. Investigation of the complexes of the respiratory chain reveals a 32% decrease of ubiquino1:ferricytochrome c oxidoreductase (complex 111) activity and a 72% decrease of ferrocytochrome c:oxygen oxidoreductase (complex IV) activity in mitochondria from 5-6-week HCCLtreated rats as compared with controls. Liver mitochondria from HCCL-treated rats also demonstrate decreased cytochrome content per mg of mitochondrial protein (25% decrease of cytochrome b and 52% decrease of cytochrome a + as as compared with control rats). The HCCL-treated rat represents an animal model for the study of the consequences of respiratory chain defects in liver mitochondria.
Vitamin B12 deficiency decreases the activity of the two cobalamin-dependent enzymes L-methylmalonyl-CoA mutase and methionine synthetase in humans (1) and in experimental animals (2, 3). L-Methylmalonyl-CoA mutase catalyzes the conversion of L-methylmalonyl-CoA to succinyl-CoA which is the rate-limiting step in cellular propionate utilization (4). In rats, vitamin B12 deficiency resulting in decreased activity of L-methylmalonyl-CoA mutase can be achieved by feeding a vitamin B12-free diet (2, 3). Administration of the vitamin B12 analogue hydroxycobalamin[c-lactam] (HCCL)' is an alternative method to produce decreased activity of L-methylmalonyl-CoA mutase (5). Surprisingly, rats treated with HCCL not only showed metabolic consequences of L-methylmalonyl-CoA mutase inhibition (hepatic accumulation of abnormal acyl-CoAs and methylmalonic aciduria) (6, 7) but also had increased oxidative metabolism of pyruvate and palmitate per hepatocyte (7, 8), which was explained by increased hepatocellular mitochondrial content (8). In contrast to the increased oxidative metabolism of palmitate and pyruvate per hepatocyte, state 3 oxidation rates in isolated liver mitochondria from HCCL-treated rats (expressed per mg of mitochondrial protein) for pyruvate or palmitoylcarnitine as substrates were decreased (8), suggesting a defect in oxidative metabolism of liver mitochondria.
The aim of the present investigation was to characterize the function and to define any enzymatic defect in the electron transport chain of isolated liver mitochondria from HCCLtreated rats. The results of the current studies show that HCCL treatment over 5-6 weeks leads to decreased activities of ubiquino1:ferricytochrome c oxidoreductase (complex 111) and ferrocytochrome c:oxygen oxidoreductase (complex IV) of the electron transport chain of liver mitochondria and is associated with decreased mitochondrial cytochrome b and cytochrome a + a3 contents.

MATERIALS AND METHODS
Animak-Male Fischer 344 rats (Charles River Laboratories, Portage, MI) were used for all experiments. Two to three rats were housed per cage with free access to drinking water and normal rat chow (Lab Chows, Purina Mills Inc., St. Louis, MO). After an acclimation period of 10 days, an osmotic mini-pump (Aha Corp., Palo Alto, CA, model 2002) containing either saline (control animals) or hydroxycobalamin[c-lactam] (HCCL-treated animals) was implanted subcutaneously under ether anesthesia. Hydroxycobalamin[c-lactam] was delivered at a rate of 2 pg/h. In animals treated for 5-6 weeks, a second mini-pump (content identical to the first mini-pump) was placed 3 weeks after the first pump implantation. At the time point of the first mini-pump implantation, the animals weighed 250 ? 20 g ( n = 24). In agreement with previous studies (5-7), both body weights and liver weights were not different between HCCL-treated and control rats after 2-3 weeks and after 5-6 weeks of treatment with HCCL (data not shown).
Mitochondrial Isolation and Ozidatiue Metabolism-Liver mitochondria were isolated as described by Hoppel et al. rats were killed by decapitation between 0800 and 09:OO a.m. The liver was quickly removed and placed in ice-cold MSM buffer (220 mM mannitol, 70 mM sucrose, 5 mM MOPS, pH 7.4). The liver was rinsed, blotted, weighed, minced, and washed with cold MSM buffer. A 10% suspension (w/v) of the minced liver containing 2 mM EDTA was prepared using a Potter-Elvehjem homogenizer with a loose fitting pestle, Nuclei and cell debris were removed by centrifugation at 700 X g for 10 min, and mitochondria were isolated by centrifugation of the supernatant at 7000 X g for 10 min. The resulting mitochondrial pellet was washed twice with MSM buffer and finally diluted to contain approximately 50 mg of mitochondrial protein per ml.
Oxygen consumption by intact mitochondria was measured in a chamber equipped with a Clark-type oxygen electrode (Yellow Springs Instrument Co.) at 30 "C. The incubations contained 1 mg/ml mitochondrial protein in 80 mM KCl, 50 mM MOPS, 1 mM EGTA, 5 mM KH2P04, pH 7.4. Defatted bovine serum albumin (1 mg/ml, w/v) was added to the incubations containing fatty acids as substrates. The final volume of the incubations was 500 p1 for incubations containing fatty acids as substrate and 1000 pl for all other substrates. After depletion of endogenous substrates by the addition of ADP, the substrate was added to the incubation, and state 3 respiration was initiated by the addition of ADP (final concentration, 100 nmol/ml). State 3 and state 4 respiration were defined and calculated according to Chance (10) as ADP-stimulated and ADP-limited respiration, respectively. Respiratory control ratios (ratio of state 3 to state 4 respiration) were calculated according to Estabrook (11).
Mitochondrial oxidase activities were determined with freezethawed mitochondria as described by Blair et al. (12) and were performed at 30 "C using the oxygen electrode. Incubations contained 20 mM potassium phosphate (pH 7.41, 0.1 mM EDTA, 0.15 mM oxidized cytochrome c, mitochondrial protein, and substrates (added last) in a final volume of 500 p1. Activities were determined in the absence and in the presence of specific inhibitors and were calculated as the difference of uninhibited minus inhibited rates. NADH oxidase activity was measured with 2.8 mM NADH as substrate, with and without 7.5 p~ rotenone as inhibitor. Succinate oxidase was measured in the presence of 0.8 mM duroquinone and 40 mM succinate, with and without 10 pg of antimycin A as inhibitor. Duroquinol oxidase was measured with 2 mM duroquinol as substrate, with and without 10 pg of antimycin A as inhibitor. Cytochrome c oxidase was measured in the presence of 0.24 mM N,N,N',N'-tetramethyl-p-phenylenediamine (TMPD) and 7.2 mM L-ascorbate, with and without 2 mM sodium azide as inhibitor.
Succinate:2,6-dichloroindophenol (DCIP) oxidoreductase (complex 11) activities were determined spectrophotometrically. Incubations contained 100 pg of mitochondrial protein, 0.1 mM EDTA, 0.1% bovine serum albumin (w/v), 3 mM sodium azide, 100 p~ DCIP, and 50 mM potassium phosphate, pH 7.4, in a final volume of 1 ml. After an equilibration period of 5 min, reactions were started by the addition of 20 mM succinate, and the decrease in absorbance at 600 nm was monitored. Activities were calculated using an extinction coefficient of 21 mM" cm" for DCIP (19). Theonyltrifluoroacetone (TTFA) was used as a specific inhibitor of succinate:DCIP oxidoreductase activity (19,20). TTFA-sensitive activities were obtained by subtraction of the rate determined in incubations containing 1 mM TTFA from the rate in parallel incubations without TTFA. Similar assays were performed containing duroquinone (added to the reaction mixture before the equilibration period at a final concentration of 0.5 mM).
With or without addition of duroquinone, 1 mM TTFA led to a 85% reduction of succinate:DCIP oxidoreductase activity; in contrast, 1 mM TTFA inhibited succinate dehydrogenase activity by less than 10%. Ubiquinokferricytochrome c oxidoreductase (complex 111) activities were determined spectrophotometrically. Incubations contained 25 pg of mitochondrial protein, 0.1 mM EDTA, 0.1% bovine serum albumin (w/v), 150 p~ cytochrome c, 3 mM sodium azide, and 50 mM potassium phosphate, pH 7.4, in a final volume of 1 ml. After a 5min equilibration period, the reactions were started by the addition of 100 p~ ubiquinol-4, and the increase in absorbance at 550 nm was monitored. Activities were calculated using an extinction coefficient of 19.1 mM" cm" for reduced cytochrome c. Antimycin A-sensitive activities were obtained by subtraction of the rate determined in incubations containing 10 pg of antimycin A from the rate in parallel antimycin A-free incubations. Ubiquinol-4 was synthesized according to Ragan et al. (21) by reduction of ubiquinone-4 (10 pmol) with NaBH4 in 2 ml of a 1:1 ethanoEH20 mixture (v/v, pH 2). The ubiquinol-4 formed was extracted twice with 1 ml of diethyl etherisooctane 2:l (v/v). The combined organic phases were washed with 2 ml of 2 M NaCl and evaporated to dryness at room temperature under a stream of nitrogen. The residue was dissolved in ethanol, and the resulting light yellow solution was acidified by the addition of 10 pl of 0.1 M HC1. This solution was stable for at least 4 months when stored at -20 'C under light protection.
Enzyme activities are expressed as milliunits (nmol/min) per mg of mitochondrial protein or as milliunits per g of liver, wet weight. For cytochrome oxidase, the first order rate constant was determined (expressed as l/min/mg of mitochondrial protein).
Mitochondrial cytochrome content was determined spectrophotometrically at room temperature as described by Williams (22).
Protein concentrations in mitochondrial preparations were determined using the biuret method with bovine serum albumin as a standard (23).
Reagents-Ubiquinone-4 was purchased from Fluka (Buchs, Switzerland). TTFA and duroquinone were obtained from Aldrich. Rotenone and antimycin A were purchased from Sigma. DCIP was obtained from Fisher. Hydroxycobalamin[c-lactam] was prepared by the reduction of the previously synthesized cyanocobalamin analogue (24,25) and was kindly provided by Dr. R. H. Allen, University of Colorado Health Sciences Center, Denver, CO. All other chemicals were of reagent grade.
Statistical Methods-Data are presented as mean & S.D. Means of two groups were analyzed using Student's t test after comparing the variance of the two groups by an F-test (26). In the case of unequal variance, a modified t test was used (26). Means of three groups were compared by analysis of variance (26). Student's t test with Bonferroni correction was used to compare individual means in the case of a significant F in the analysis of variance.

RESULTS
Rats were treated with minipumps containing either saline (control group) or the vitamin BI2 analogue HCCL for 2-3 weeks or for a total of 5-6 weeks. Treatment of rats with HCCL reduces the activity of L-methylmalonyl-CoA mutase (5) leading to metabolic abnormalities similar to human methylmalonic aciduria (5-7) and is associated with increased hepatic mitochondrial content (8). In agreement with a previous study (B), mitochondrial protein content (corrected for incomplete recovery by using activities of the mitochondrial enzymes citrate synthase and succinate dehydrogenase) was increased in livers from 5-6 week HCCL-treated rats as compared with control rats. The calculated mitochondrial protein content was 70 f 5 mg/g of liver in HCCL-treated ( n = 9) and 50 f 5 mg/g of liver in control rats (n = 9, p < 0.05).
As illustrated in Fig. 1, the electron transport chain is located in the inner mitochondrial membrane and transports electrons from NADH or FADH to molecular oxygen and produces a transmembranous proton gradient used for the generation of ATP (27). Oxidative metabolism was investigated in intact liver mitochondria isolated after 2-3 weeks or after 5-6 weeks of HCCL treatment. There were no differences in state 3 oxygen consumption for P-hydroxybutyrate, pyruvate, L-glutamate, succinate, and fatty acids as substrates with mitochondria isolated from 2-3 week HCCL-treated rats as compared with mitochondria from control rats (Table I). However, after a treatment period of 5-6 weeks, state 3 oxidation rates were reduced in mitochondria from HCCLtreated as compared with control rats (Table I). The reduction in state 3 oxidation rates was approximately 30% for pyruvate, L-glutamate, and fatty acids (requiring complexes I, 111, and IV of the electron transport chain), 38% for succinate (requiring complexes 11, 111, and IV), 47% for duroquinol (requiring complexes I11 and IV), and 51% for TMPD/L-ascorbate (requiring complex IV). State 4 oxygen consumption was not different between mitochondria isolated from control and HCCL-treated rats; thus, the respiratory control ratios showed the same decline as the state 3 oxidation rates (data not shown). When oxidative phosphorylation was uncoupled with dinitrophenol, state 3 oxidation rates were reduced by 36% for L-glutamate and by approximately 50% for duroquinol and TMPD/L-ascorbate as substrates (Table 11) Activities of mitochondrial oxidases (measured polarographically) were investigated to help localize the affected site and to exclude changes in the composition of the inner mitochondrial membrane and/or transport defects of substrates as causes for the observed decrease in state 3 oxidation rates with liver mitochondria from HCCL-treated rats. Consistent with the state 3 oxidation rates obtained with intact mitochondria, oxidase activities were reduced in mitochondrial preparations from HCCL-treated rats as compared with preparations from control rats (Table  111). The reductions in oxidase activity with mitochondrial preparations from HCCLtreated as compared with control rats were 60.2% for NADH oxidase (requiring complexes I, 111, and IV), 47.5% for succinate oxidase (requiring complexes 11, 111, and IV), 69.9% for duroquinol oxidase (requiring complexes I11 and IV), and 57.4% for cytochrome c oxidase (requiring complex IV).
The results obtained by the polarographic methods demonstrated a defect at complex IV in liver mitochondria from HCCL-treated rats; however, additional defects in the other complexes of the electron transport chain could not be excluded. Therefore, activities of discrete components of the electron transport chain were determined. Consistent with the polarographic studies, the defect at complex IV was confirmed spectrophotometrically (Table IV). The first order rate constant of cytochrome c oxidase expressed per mg of mitochondrial protein was decreased by 72% in mitochondrial preparations from HCCL-treated rats as compared with control rats. Activities of rotenone-sensitive NADH:ferricytochrome c oxidoreductase (requiring complexes I and 111) and succinate:ferricytochrome c oxidoreductase (requiring complexes I1 and 111) were reduced by 76 and 42%, respectively, in mitochondrial preparations from HCCL-treated as compared with control rats. In contrast, rotenone-insensitive NADH:ferricytochrome oxidoreductase activity, which requires NADH:cytochrome bs reductase in the outer mitochondrial membrane (16), was not affected by HCCL treatment. Similarly, activities of complex I (NADH:ferricyanide oxidoreductase and NADH:duroquinone oxidoreductase) and of complex I1 (succinate dehydrogenase and succinate:DCIP oxidoreductase) were not different between mitochondria from HCCL-treated and control rats (Table IV). The suspected defect at complex I11 was confirmed directly by the 32%  decrease of the ubiquino1:ferricytochrome c oxidoreductase activity in mitochondrial preparations from HCCL-treated rats as compared with preparations from control rats (Table  IV).
Reduced activities of complex I11 or complex IV of the electron transport chain are frequently associated with decreased contents of mitochondrial cytochrome b or cytochrome a + a3 (28-32). As compared with control rats, the content of cytochrome a + a3 (expressed per mg of mitochondrial protein) was decreased by 20% in 2-3 week HCCL-

TABLE I1
Oxygen consumption by uncoupled isolated rat liver mitochondria Uncoupling was performed by the addition of 0.1 mM dinitrophenol to the mitochondrial incubations. State 3 oxygen consumption was measured in a test chamber equipped with a Clark electrode at 30 "C. The number of incubations with individual mitochondrial preparations was 6 for both groups. Results are expressed as nanoatoms/ min/mg of mitochondrial protein and are presented as mean f S.D.
Significant differences ( p < 0.05) between control and HCCL-treated rats are indicated by *.  (Table V). After a treatment period of 5-6 weeks, the mitochondrial content of both cytochrome a + a3 and cytochrome b were decreased by 52 and by 25%, respectively, and the mitochondrial cytochrome c1 content was increased by 36% in HCCL-treated as compared with control rats.

DISCUSSION
The function of the respiratory chain was investigated in isolated liver mitochondria from rats treated with the vitamin Blz analogue HCCL. HCCL-treated rats showed decreased activities of ubiquino1:ferricytochrome c oxidoreductase (complex 111) and ferrocytochrome c:oxygen oxidoreductase (complex IV) with decreased function of the electron transport chain after 5-6 weeks of HCCL treatment. These defects were associated with decreased mitochondrial content of cytochrome a + a3 and cytochrome b.
After a treatment period of 2-3 weeks, state 3 oxygen consumption by isolated mitochondria was not different between HCCL-treated and control rats. In contrast, after 5-6 weeks of HCCL treatment, state 3 oxidation rates for mitochondria from HCCL-treated rats were reduced for most substrates studied (Table I). The reduction in state 3 oxidation rates was greater for substrates with high oxidation rates (duroquinol and TMPD/ascorbate) than for substrates showing lower oxidation rates (pyruvate, L-glutamate, succinate, and fatty acids). In contrast to mitochondrial state 3 oxygen consumption, mitochondrial cytochrome a + a3 content was already reduced after 2-3 weeks of HCCL treatment, and a further reduction was observed after 5-6 weeks of HCCL treatment, combined with a reduction in mitochondrial cytochrome b content. The finding that the mitochondrial cytochrome a + a3 content can be reduced without impairment of the function of the electron transport chain is consistent with observations made in human electron transport chain defects (28-32) and suggests that complex IV is normally not ratelimiting for the activity of the electron transport chain. Since complex I11 is proposed to be rate-limiting for the activity of the electron transport chain in liver mitochondria (33), the observed decrease in oxidative metabolism of isolated mitochondria from 5-6 week HCCL-treated rats probably reflects decreased activity of complex I11 and not of complex IV.

TABLE IV
Activities of subunits of the respiratory chain Cholate-solubilized mitochondria (1 mg of mitochondrial protein per ml of 100 mM phosphate buffer, pH 7.4, containing 1% cholate, w/v) were used for these measurements. Activities were determined spectrophotometrically as described under "Materials and Methods." The number of determinations with individual mitochondrial preparations was 6 for both groups. Results are presented as mean f S.D. and are expressed as milliunits/mg of mitochondrial protein (except for ferrocytochrome c:oxygen oxidoreductase where the first order rate constant is given as l/min/mg of mitochondrial protein). Significant differences between control and HCCL-treated rats are indicated by *. as nmol/mg of mitochondrial protein and are presented as mean f S.D. Significant differences ( p < 0.05) between HCCL-treated and control rats are indicated by *, and between the two groups of HCCL-treated rats by +. Decreased mitochondrial ubiquinone content could contribute to the observed decrease of the electron transport chain activity of isolated mitochondria from HCCL-treated rats. To approach this possibility, succinate:DCIP oxidoreductase activity (complex 11) was determined both with and without addition of duroquinone (Table IV). Since DCIP accepts electrons directly from ubiquinol (34, 35), succinate:DCIP oxidoreductase activity measured without addition of exogenous ubiquinone is decreased in endogenous ubiquinone deficiency and can be normalized by the addition of exogenous ubiquinone (36). Succinate:DCIP oxidoreductase activities obtained in the current studies (no differences between HCCL-treated and control animals both in the absence and in the presence of 0.5 mM duroquinone, Table IV) are therefore consistent with a normal functional ubiquinone pool in mitochondria from HCCL-treated rats.
Consistent with a previous report in HCCL-treated rats (8), mitochondrial protein content per g of liver was increased in 5-6 week HCCL-treated as compared with control rats. Previous studies in HCCL-treated rats showed an increase in mitochondrial protein content as early as 2-3 weeks after the start of HCCL treatment (8), a time point where the function of the respiratory chain is unaltered ( Table I). These findings suggest that increased mitochondrial content in livers from HCCL-treated rats is not a direct consequence of defects in the electron transport chain. It is possible that the metabolic abnormalities produced by the inhibition of L-methylmalonyl-CoA mutase by HCCL (hepatic accumulation of unusual acyl-CoAs and methylmalonic aciduria), which appear early after HCCL administration (7), are responsible for the mitochondrial proliferation in HCCL-treated rats. When mitochondrial biogenesis is accelerated, decreased mitochondrial cytochrome a + u3 and/or cytochrome b content could result from an insufficient availability of the respective apoproteins and/or heme groups. In support of this hypothesis, hepatic mitochondrial proliferation in rats induced by hyperthyroidism is associated with decreased mitochondrial cytochrome b content (37). It is possible that the decrease in mitochondrial cytochrome a + u3 and cytochrome b content in human mitochondrial disorders (28-32) is a consequence of mitochondrial proliferation which is frequently observed in human defects of the electron transport chain (30, 32, 38).
In humans, decreased complex 111 activity is a recognized cause for mitochondrial myopathy, mitochondrial cardiomyopathy, and/or mitochondrial encephalomyopathy (28-32). A possible impairment of hepatic mitochondrial function has only been described in one of these patients (31). Similar to 5-6 week HCCL-treated rats (6), plasma glucose and ketone body concentrations were unaltered during starvation in this patient (31), ruling out severe defects in hepatic intermediary metabolism. In rats treated with HCCL over 5-6 weeks, compensatory mechanisms including mitochondrial prolifer-ation (this study and Ref. 8) and increased coenzyme A biosynthesis (6) are proposed to protect hepatic energy metabolism. Similar mechanisms may moderate the defects in hepatic energy metabolism in humans with mitochondrial disorders (38).
In conclusion, isolated liver mitochondria from 5-6 week HCCL-treated rats demonstrate decreased activities of complex 111 and complex IV with a consequent decrease of the function of the respiratory chain. Decreased complex I11 and complex IV activities may result from decreases in mitochondrial cytochrome b and cytochrome u + u3 contents. HCCLtreated rats provide an interesting animal model for the study of the development and the metabolic consequences of hepatic mitochondrial disorders and of the molecular mechanisms of mitochondrial biogenesis.