Development of a new assay for complex I of the respiratory chain.
Disorders of complex I are a frequent cause of mitochondrial diseases and can manifest with a wide variety of clinical symptoms, ranging from simple muscle lethargy to whole organ malfunctions (3). Recently, a deficiency of complex I has been implicated in neurodegenerative disorders, such as Parkinson disease (4).
For the diagnosis of complex I deficiency, detection of functional protein activity is preferable to a mutational screen of all 41 subunit transcripts or coding sequences. The technique used currently is a spectrophotometric assay, which measures the oxidation of NADH to [NAD.sup.+] at 340 nm by skeletal muscle homogenate prepared from patient biopsies (5-7). The disadvantage with this method has been its insensitivity. One reason for the insensitivity is that it requires large quantities of tissue, which make the assay solution turbid and difficult to analyze by optical methods. Another of the reasons for the low sensitivity of the spectrophotometric assay is that NADH is oxidized not only by complex I but also by the many other dehydrogenases located in skeletal muscle or other tissues.
It is well established that dehydrogenases such as complex I exhibit stereospecificity regarding the oxidation of NADH. Studies with submitochondrial particles from beef heart and rat liver have revealed a specificity of the respiratory chain complex I for the 4B hydrogen atom of NADH (8). At the same time, it was found that most other NADH dehydrogenases (microsomal, mitochondrial outer membrane, malic acid, and cytochrome c reductase) were 4A-[sup.3]H specific.
Because, as mentioned above, one of the disadvantages of the spectrophotometric assays is the high background activity of dehydrogenases other than complex I, the availability of [4B-[sup.3]H]NADH could potentially produce a more specific assay. We therefore prepared radioactive isoforms of NADH tritiated in the respective 4A-[sup.3]H or 4B-[sup.3]H positions and tested both as suitable substrates for complex I activity. Enzyme in the form of purified mitochondria or homogenate was prepared from rat or human skeletal muscle and incubated with the respective radioisotopes. The product ([sup.3][H.sub.2]O) was collected after charcoal adsorption of unreacted [sup.3]H-NADH and measured as an indicator of the amount of NADH oxidation. Sensitivity to the inhibitor rotenone was used as the indicator of complex I specific activity (9).
Materials and Methods
All reagents were of analytical grade and purchased from either Merck or Sigma unless otherwise specified.
PREPARATION OF 4A AND 4B STEREOISOMERS OF NADH
Tritiated [4-[sup.3]H]NAD-ammonium salt was obtained from Amersham at a specific activity of 1.35 Ci/mmol. The respective isoforms of NADH were prepared enzymatically from stereospecific dehydrogenases and then purified on an anion-exchange column, as described previously by Fisher and Guillory (10).
[4A-[sup.3]H]NADH was synthesized with 1 mg of the enzyme L-glutamate dehydrogenase (45 U/mg; Sigma) and 60 [micro]mol L-glutamic acid (monosodium salt). The reaction was carried out in a mixture containing 40 [micro]mol of Tris-HCl, 112 [micro]mol of sucrose, and 360 [micro]mol of hydrazine hydrate. Five micromoles of unlabeled and 15 nmol of tritiated [NAD.sup.+] were added to a total volume of 1.25 mL (final specific activity, 4 mCi/mmol) and incubated at room temperature for 1.5 h.
[4B-[sup.3]H]NADH was synthesized with 200 [micro]mol of ethanol and 0.2 mg of alcohol dehydrogenase (305 U/mg; Fluka) in a mixture containing 24 [micro]mol of Tris-HCl, 70 [micro]mol of sucrose, and 240 [micro]mol of hydrazine hydrate. Five micromoles of unlabeled and 15 nmol of tritiated [NAD.sup.+] were added to a total volume of 2.5 mL (final specific activity, 4 mCi/mmol) and incubated at room temperature for 1 h. Reactions were stopped by heating to 95[degrees]C for 3 min, the protein was pelleted by centrifugation at 15 000g for 5 min, and the supernatant was used for isolation of the respective reaction products.
The product from each of the enzyme reactions was loaded onto an A-25 DEAE Sephadex (Amersham Pharmacia) column (1 x 4.5 cm) equilibrated with starting buffer (25 mmol/L Tris-HCl, pH 8). The column was washed with 100 mL of water, and then unreacted [NAD.sup.+] was eluted with incremental concentrations (20 mL of 0.05 mol/L, 20 mL of 0.1 mol/L, and 10 mL of 0.2 mol/L) of Tris-HCl, pH 7.
The remaining bound [sup.3]H-NADH isomer was eluted with 0.5 mol/L Tris-HCl, pH 7, in 1-mL fractions. The fractions were checked spectrophotometrically at 340 nm, and the final concentration of the [sup.3]H-NADH isomer was established against a calibration curve of unlabeled NADH in elution buffer. The specific activity of the product was assessed by scintillation counting in a BETA V liquid scintillation counter (Kontron Instruments) with 5 mL of IRGA-safe plus (Packard) scintillation fluid in polyethylene vials. [4A-[sup.3]H]NADH and [4B-[sup.3]H]NADH were then diluted with commercially available NADH and water to a specific activity of 1 mCi/mmol and a concentration of 1 mmol/L. As shown in the Results, unlabeled [4A]NADH cannot be differentiated from [4B]NADH by complex I, allowing us to use commercial NADH for dilution of the respective [[sup.3]H]NADH isotopes. Aliquots were stored at either 4 or -20[degrees]C. NADH stored at 4[degrees]C degraded over periods longer than 3 weeks, but stocks stored at -20[degrees]C were still stable after storage for at least 4 months. The column was refreshed between runs with 0.25 mol/L HCl and then rinsed with 100 mL of [H.sub.2]O and titrated back to pH 8 with starting buffer.
Unlabeled [4A]NADH and [4B]NADH was prepared by the same method as described for the respective tritiated isoforms for initial experiments to demonstrate that complex I is not able to differentiate between these two isoforms.
PREPARATION OF MITOCHONDRIA
Mitochondria were prepared from rat skeletal muscle as described by Hoppel et al. (11). The final yield was 5.5 mg of mitochondrial protein per gram of skeletal muscle wet weight as determined by the Bradford protein assay (12). Mitochondria were stored in 25 mmol/L potassium phosphate buffer at -20[degrees]C until needed. Usually, 2 [micro]g of mitochondrial protein was used per assay unless otherwise specified.
PREPARATION OF HOMOGENATE
Skeletal muscle samples from human biopsies or rats were stored frozen at -70[degrees]C until needed and then were homogenized with a motorized glass/glass homogenizer (Eurostar) at a concentration of 20 g tissue/L in 25 mmol/L potassium phosphate buffer until complete tissue disruption was achieved. There was no change in activity observed after freezing and thawing. The standard assay measured 40 [micro]g of tissue per assay unless otherwise specified.
Complex I activity was measured in a 50 mmol/L potassium phosphate buffer, pH 6.5, containing 5 mmol/L Mg[Cl.sub.2], 2.5 g/L bovine serum albumin, 100 [micro]mol/L decylubiquinone, 0.02 g/L antimycin A, and 0.0125 g/L KCN. Buffer was made fresh each day and used within 4 h. KCN and antimycin A were added to prevent backflow of electrons along the respiratory chain.
[sup.3]H-NADH (specific activity 1 mCi/mmol) in a final concentration of 100 [micro]mol/L in either its 4A-[sup.3]H or 4B-[sup.3]H form was added to the buffer and preincubated at 37[degrees]C for 3 min in a final volume of 98 [micro]L. Control reactions contained 2 [micro]mol/L rotenone as a specific inhibitor of complex I activity. The reaction was started by addition of 2 [micro]L of enzyme preparation.
Reactions were allowed to proceed for 5 min before being stopped by the addition of 1 mL of 75 g/L activated charcoal in water and immediately vortex-mixed for 20 s. Samples were centrifuged at maximal speed on a bench top centrifuge for 5 min, and duplicate 300-[micro]L aliquots were taken for counting with 5 mL of scintillation fluid.
Spectrophotometric determination of the activity of complex I was performed as described previously (6) but with 100 [micro]mol/L decylubiquinone instead of ubiquinone-1 as the electron acceptor.
DATA PRESENTATION AND STATISTICAL ANALYSIS
For all points represented in the graphs, the averaged value from triplicate reactions in the presence of rotenone was subtracted from a minimum of three individual reactions in the absence of the inhibitor. Activities were adjusted to [micro]mol of NADH oxidized per minute per milligram of protein or tissue and taken as only that portion of activity inhibitable by rotenone, unless otherwise specified.
Data are given as mean [+ or -] SD. Michaelis-Menten kinetics were fitted using nonlinear regression analysis (Sigmaplot for Windows 4.0; Jandel Scientific).
Although not optically active, the two hydrogens in position 4 of the nicotinamide ring of NADH can be distinguished by dehydrogenases. Using NADH that was tritiated specifically in the 4 position, Ernster et al. (8) showed that most dehydrogenases (e.g., microsomal, mitochondrial outer membrane, malic acid, and cytochrome c reductase) preferentially use the hydrogen in the 4A position, whereas complex I and alcohol dehydrogenase use the hydrogen in the 4B position. An assay based on this specificity and using [4B-[sup.3]H]NADH as a substrate would therefore presumably be more specific for complex I activity.
The synthesis of the two radioactive isomers of NADH was achieved enzymatically from tritiated [NAD.sup.+] and either L-glutamate dehydrogenase (4A-[sup.3] H) or alcohol dehydrogenase (4B-[sup.3]H) as described in Materials and Methods. The reaction product was adsorbed onto a DEAE Sephadex anion-exchange column, and the tritiated NADH was eluted with 0.5 mol/L Tris buffer, pH 7. The concentrations of the resulting fractions were typically 0.5-1 mmol/L, and the specific activity was 4 mCi/mmol NADH.
Using isolated rat skeletal muscle mitochondria, we optimized assay conditions for temperature (data not shown) and pH (Fig. 1). The highest activity without loss of sensitivity to rotenone was observed at pH 6.5. This pH was therefore chosen for all further experiments. The optimal temperature was 37[degrees]C, with increasing activity over the range investigated (20-37[degrees]C). As shown in Fig. 2A, the reaction was linear for 10 min with isolated mitochondria as the enzyme source, and a similar result was obtained with rat skeletal muscle homogenate (Fig. 2B). With an incubation time of 5 min, the reaction was linear for up to 4 [micro]g of mitochondrial protein or 80 [micro]g of skeletal muscle. Similar results were obtained for isolated rat liver mitochondria and liver homogenate (data not shown). Replacing the phosphate buffer with Tris-HCI did not produce higher activities (data not shown).
[FIGURE 1 OMITTED]
Activity at all times was taken as a subtraction of the rotenone-insensitive portion from the total observed; i.e., total activity observed minus activity in the presence of 2 [micro]mol/L rotenone. The background at zero time or in the absence of enzyme was directly proportional to the total tritium added to the reaction and represented 5-10% of the total radioactivity used. This background typically represented 20-40% of the final activity (after charcoal adsorption) in the presence of enzyme and after incubation for 5-10 min. Rotenone-sensitive NADH oxidation was between 60% and 80% of the total activity after subtraction of background for both isolated mitochondria and muscle homogenate.
Complex I is a multisubstrate enzyme, requiring a suitable quinone as an electron acceptor. Michaelis-Menten kinetics were observed for both decylubiquinone (Fig. 3) and [4B-[sup.3]H]NADH (Fig. 4). The corresponding [K.sub.m] values were 48 [micro]mol/L for decylubiquinone and 35 [micro]mol/L for [4B-[sup.3]H]NADH, and the [V.sub.max] values were 89 and 90 [micro]mol * [min.sup.-1] * mg [protein.sup.-1] for decylubiquinone and [4B-[sup.3]H]NADH, respectively. When [4A-[sup.3]H]NADH was used as a substrate, rotenone-insensitive but no rotenone-sensitive activity could be observed with isolated mitochondria, demonstrating the specificity of the assay (Fig. 4). However, as shown directly by spectrophotometric analysis, both 4A-NADH and 4B-NADH are substrates for complex I and produce the same activity (data not shown). These experiments demonstrate that complex I specifically uses the hydrogen of NADH that is located in the 4B position.
[FIGURE 2 OMITTED]
Studies of variability in the assay were conducted with multiple homogenates of the same sample (rat skeletal muscle) on different days, and the standard deviation was taken as a gauge of the reproducibility of the assay. Five separate homogenates were prepared individually on 3 different days, with the mean activity of complex I showing an overall variation of 6% of the mean (2.66 [+ or -] 0.17 U/g tissue). On the 3 individual days, the mean activities were 2.48 [+ or -] 0.95, 2.79 [+ or -] 0.79, and 2.71 [+ or -] 0.59 U/g tissue, respectively.
[FIGURE 3 OMITTED]
Simultaneous determination of complex I activity in rat skeletal muscle (3.1 [+ or -] 1.2 U/g) and mitochondria isolated from the same muscle (76 [+ or -] 13 U/g protein) allowed us to estimate the amount of mitochondrial protein per gram of muscle. According to these activities, 1 g of skeletal muscle yielded 42 mg of mitochondrial protein.
Given that the overall aim of this project was to establish a workable method for human tissue homogenate, we assayed complex I activity in frozen tissue samples (stored for 1 month to 2 years at -70[degrees]C) of different adult patients. The reaction demonstrated Michaelis-Menten kinetics with similar kinetic constants as found in rat skeletal muscle. The reaction was linear at 5 min for up to 60 [micro]g of tissue per assay and at 10 min for up to 40 [micro]g of tissue per assay. Maximal activities were determined for seven different human tissue homogenates, using the conditions worked out for rat tissues. The activities were close to those in rat skeletal muscle homogenate, averaging 4.2 [+ or -] 2.4 U/g tissue. The same tissues were also analyzed by the spectrophotometric method, which yielded an activity of 3.2 [+ or -] 1.8 U/g tissue. Unfortunately, no skeletal muscle samples of patients with complex I deficiency were available for testing.
The fundamental importance of the respiratory chain and its increasing implication in several diseases has led to the need for development of reliable assays to test the function of each complex on a direct and individual basis. The development here of a reproducible, sensitive, and specific assay for complex I of the respiratory chain is a valuable contribution to the detection of this disorder in patients with metabolic diseases.
[FIGURE 4 OMITTED]
We developed an approach to determining complex I activity that is entirely different from the spectrophotometric methods used currently. This approach is based on the ability of the NADH dehydrogenases to differentiate between the two hydrogens in position 4 of the dihydropyridine ring of NADH. This specificity of the NADH dehydrogenases is most likely related to steric hindrance of the rotation of the dihydropyridine ring, which prevents free access of both hydrogens to the active center of the respective dehydrogenase. Regarding unlabeled NADH, the existence of the 4A- and 4B-isoforms is functionally not important because both positions contain hydrogen atoms. We have shown this directly with the spectrophotometric method, which indicated identical activities for complex I for both the 4A- and 4B-NADH isoforms. Therefore, [4B-[sup.3]H]NADH can be diluted with commercial NADH for the assay of complex I activity as described by us. As demonstrated by our experiments, the existence of the 4A- and the 4B-NADH isoforms becomes an important diagnostic tool, however, when one of the hydrogens is labeled and can therefore be differentiated from the other one.
The 4A- and 4B-isoforms must be differentiated from the [alpha]- and [beta]-isoforms of NADH, which are commercially available. The [alpha]- and [beta]-isoforms of NADH are true stereoisomers, differing in the position of the adenosine moiety in relation to the plane of the sugar ring. This position is above the plane of the sugar ring in the case of the [alpha]-isomer and below the plane in the case of the [beta]-isomer. In all cases, we used the more common [beta]-NADH isomer.
Our studies show that [4B-[sup.3]H]NADH can be used reliably as a substrate for the measurement of complex I activity. We have demonstrated that the assay is linear with time and protein or tissue content for isolated mitochondria and tissue homogenates from rats and humans, and that the results are reproducible. The reactions of the two substrates can be described by Michaelis-Menten kinetics, and the [K.sub.m] values are in the expected range. Previous reports of the activity of complex I have varied according to the enzyme preparations and the ubiquinone species used. Values range from 0.3-1 [micro]mol * [min.sup.-1] * mg [protein.sup.-1] for submitochondrial particles from bovine heart (13,14) or yeast (15) to 0.082 [micro]mol * [min.sup.-1] * mg [protein.sup.-1] for human skeletal muscle mitochondria (7). This value is close to the [V.sub.max] obtained with rat skeletal muscle mitochondria in our studies (0.090 [micro]mol * [min.sup.*-1] * mg [protein.sup.-1]). The ubiquinone species used is another source of variation of the [V.sub.max] for this enzyme reaction. The longer chain ubiquinones are problematic because of their insolubility in water (7), and other ubiquinones are problematic because they render complex I activity less sensitive to rotenone (16). Because we (17) and others (14,15,18) have used decylubiquinone successfully for the spectrophotometric determination of complex I activity, we considered it a suitable substrate for the current study.
The background was consistently 5-10% of the radioactivity added to the assay and was not dependent on the reaction time, the amount of enzyme present, or the concentration of unlabeled NADH. It therefore most likely represents an impurity of the synthesized isotope. In support of this assumption, the background rose over the period of storage, parallel with the decay of NADH. After preparation, [sup.3]H-NADH should therefore be stored in aliquots at -20[degrees]C. Under these conditions, [sup.3]H-NADH is stable for months and the background remains low.
After subtraction of the background, the rotenone-sensitive [4B-[sup.3]H]NADH activity accounted for 60-80% of the total activity observed, depending on the enzyme preparation investigated. The remaining activity can be explained by incomplete inhibition by rotenone and/or antimycin and cyanide, or by the presence of other dehydrogenases that use [4B-[sup.3]H]NADH as a substrate (e.g., [alpha]-lipoyl dehydrogenase) (8). When compared with the spectrophotometric methods, rotenone is a more efficient inhibitor with [4B-[sup.3]H]NADH as the substrate, in particular with tissue homogenates as the enzyme source. This finding represents an advantage of the current assay over the spectrophotometric methods, and can be explained by the specificity of complex I for [4B-[sup.3]H]NADH.
In contrast to [4B-[sup.3]H]NADH, [4A-[sup.3]H]NADH was not a substrate for complex I because no rotenone-sensitive activity was observed with isolated mitochondria as the enzyme source. On the other hand, a small rotenone-insensitive activity was detectable with isolated mitochondria as the enzyme source. This activity can be explained by mitochondrial dehydrogenases that use [4A-[sup.3]H]NADH as a substrate, e.g., 3-phosphoglycerol dehydrogenase and malate dehydrogenase (8).
A disadvantage of the new assay of complex I activity is the initial preparation of [4B-[sup.3]H]NADH. However, once set up, the column can be stored under ethanol and refreshed before each new run. Preparation of [4B-[sup.3]H]NADH can be achieved easily within 1 day, along with the determination of the concentration of NADH and its specific activity. As discussed above, [4B-[sup.3]H]NADH is stable for months if stored at -20[degrees]C.
Based on the results of our study, we believe that this novel assay for complex I has several advantages over the currently used spectrophotometric assays and could therefore potentially replace these methods, in particular for the determination of complex I activity in tissue homogenates.
These studies were supported by grants from the Swiss National Foundation for the Study of Muscle Diseases and from the Swiss National Science Foundation (31-46792.96). We thank Professor C. Hoppel for initial constructive comments concerning the protocol of this study.
Received October 26, 1999; accepted December 17, 1999.
(1.) Loeffen J, Triepels R, van den Heuvel LP, Schuelke M, Buskens C, Smeets R, et al. cDNA of eight nuclear encoded subunits of NADH:ubiquinone oxidoreductase: human complex I cDNA characterisation completed. Biochem Biophys Res Commun 1998;253: 415-22.
(2.) Hatefi Y. The mitochondrial electron transport and oxidative phosphorylation system. Annu Rev Biochem 1985;54:1015-69.
(3.) Bindoff LA, Birch-Machin MA, Cartlidge NE, Parker WD Jr, Turnbull DM. Respiratory chain abnormalities in skeletal muscle from patients with Parkinson's disease. J Neurol Sci 1991;104:203-8.
(4.) Schapira AH. Mitochondrial involvement in Parkinson's disease, Huntington's disease, hereditary spastic paraplegia and Friedreich's ataxia. Biochim Biophys Acta 1999;1410:159-70.
(5.) Krahenbuhl S, Schafer T, Wiesmann U. Determination of the activities of the enzyme complexes of the electron transport chain in human fibroblasts. Clin Chim Acta 1996;253:79-90.
(6.) Veitch K, Hombroeckx A, Caucheteux D, Pouleur H, Hue L. Global ischaemia induces a biphasic response of the mitochondrial respiratory chain. Anoxic pre-perfusion protects against ischaemic damage. Biochem J 1992;281:709-15.
(7.) Birch-Machin MA, Briggs HL, Saborido AA, Bindoff LA, Turnbull DM. An evaluation of the measurement of the activities of complexes I-IV in the respiratory chain of human skeletal muscle mitochondria. Biochem Med Metab Biol 1994;51:35-42.
(8.) Ernster L, Hoberman HD, Howard RL, King TE, Lee CP, Mackler B, Sottocasa G. Stereos pecificity of certain soluble and particulate preparations of mitochondrial reduced nicotinamide-adenine dinucleotide dehydrogenase from beef heart. Nature 1965;207:940-1.
(9.) Degli Esposti M, Ngo A, McMullen GL, Ghelli A, Sparla F, Benelli B, et al. The specificity of mitochondrial complex I for ubiquinones. Biochem J 1996;313:327-34.
(10.) Fisher RR, Guillory RJ. Resolution of enzymes catalyzing energy-linked trans hydrogenation. 3. Preparation and properties of Rhodospirillum rubrum transhydrogenase factor. J Biol Chem 1971; 246:4687-93.
(11.) Hoppel CL, Kerr D, Dahms B, Roessmann U. Deficiency of the reduced nicotinamide adenine dinucleotide dehydrogenase component of complex I of mitochondrial transport. J Clin Investig 1987:80:71-7.
(12.) Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248-54.
(13.) Degli EM, Ghelli A, Ratta M, Cortes D, Estornell E. Natural substances (acetogenins) from the family Annonaceae are powerful inhibitors of mitochondrial NADH dehydrogenase (complex I). Biochem J 1994;301:161-7.
(14.) Fato R, Estornell E, Di Bernardo S, Pallotti F, Parenti CG, Lenaz G. Steady-state kinetics of the reduction of coenzyme Q analogs by complex I (NADH:ubiquinone oxidoreductase) in bovine heart mitochondria and submitochondrial particles. Biochemistry 1996; 35:2705-16.
(15.) Zhu QS, Beattie DS. Direct interaction between yeast NADH-ubiquinone oxidoreductase, succinate-ubiquinone oxidoreductase, and ubiquinol-cytochrome c oxidoreductase in the reduction of exogenous quinones. J Biol Chem 1988;263:193-9.
(16.) Chen S, Guillory RJ. Identification of the NADH-[NAD.sup.+] transhydrogenase peptide of the mitochondrial NADH-CoQ reductase (complex I). A photodependent labeling study utilizing arylazido-[beta]-alanyl [NAD.sup.+]. J Biol Chem 1984;259:5124-31.
(17.) Krahenbuhl S, Schafer T, Wiesmann U. Determination of the activities of the enzyme complexes of the electron transport chain in human fibroblasts. Clin Chim Acta 1996;253:79-90.
(18.) Zheng XX, Shoffner JM, Voljavec AS, Wallace DC. Evaluation of procedures for assaying oxidative phosphorylation enzyme activities in mitochondrial myopathy muscle biopsies. Biochim Biophys Acta 1990;1019:1-10.
HILARY BROOKS and STEPHAN KRAHENBUHL *
Department of Clinical Pharmacology, University of Berne, Murtenstrasse 35, CH-3010 Berne, Switzerland.
* Author for correspondence.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Enzymes and Protein Markers|
|Author:||Brooks, Hilary; Krahenbuhl, Stephan|
|Date:||Mar 1, 2000|
|Previous Article:||Correlation of antemortem serum creatine kinase, creatine kinase-MB, troponin I, and troponin T with cardiac pathology.|
|Next Article:||Estimating the long-term effects of storage at -70[degrees]C on cholesterol, triglyceride, and HDL-cholesterol measurements in stored sera.|