Coenzyme Q2 is a universal substrate for the measurement of respiratory chain enzyme activities in trypanosomatids

The measurement of respiratory chain enzyme activities is an integral part of basic research as well as for specialized examinations in clinical biochemistry. Most of the enzymes use ubiquinone as one of their substrates. For current in vitro measurements, several hydrophilic analogues of native ubiquinone are used depending on the enzyme and the workplace. We tested five readily available commercial analogues and we showed that Coenzyme Q2 is the most suitable for the measurement of all tested enzyme activities. Use of a single substrate in all laboratories for several respiratory chain enzymes will improve our ability to compare data, in addition to simplifying the stock of chemicals required for this type of research.


Introduction
Trypanosomatids are obligatory parasites belonging to the Family Trypanosomatidae and Phylum Euglenozoa. Many representatives of this group that change between two hostsinsect vectors and higher animals or plantsare the cause of serious diseases in humans, animals and plants (e.g. sleeping sickness in Africa, Chagas disease in Latin America, and various types leishmaniases worldwide). In addition to these dixenic species, many new monoxenic species have been isolated in recent years and have only one hostinsects. Major differences in their mitochondrial metabolism [20,24] indicate that further study could be positive for both basic and applied research. The respiratory chain (RC) is the central and essential part of the mitochondrial bioenergetics of the cell and its disorders are associated with multiple metabolic diseases [13]. Therefore, measuring the activity of the RC enzymes will bring new knowledge to basic and medical research. It could potentially also help in the development of treatments for diseases caused by trypanosomatids. The RC consists of four so-called "core" enzyme complexes I-IV. Three of them (Complex I -NADH dehydrogenase, Complex IIIcytochrome c reductase, and Complex IVcytochrome c oxidase) use the energy of transferred electrons to create a proton gradient on the inner mitochondrial membrane, which is further utilised in ATP biosynthesis. Complex IIsuccinate dehydrogenase is an integral component of both the Krebs cycle and the RC, but it does not contribute to membrane potential. In addition to the core enzymes, the RC of many cells also includes several so-called alternative enzymes that transmit electrons within the RC, but without pumping protons across the inner mitochondrial membrane: for example, alternative NADH dehydrogenase (NDH2) acting in parallel with Complex I, or Trypanosome alternative oxidase (TAO) bypassing complexes III and IV. Furthermore, two low-molecular mass compounds participate in the transfer of electrons in the RC: cytochrome c and ubiquinone. The second compound belongs to a group of chemical compounds containing a quinoid ring system that can exist in several states depending on the presence or absence of electrons. Ubiquinone can be found in three different oxidation-reduction states: (i) fully oxidized formubiquinone, (ii) partially reduced and unstable semiquinone, created by the receipt of one electron, and (iii) fully reduced ubiquinol (hydroubiquinone) with two accepted electrons. Ubiquinone, which is an integral part of the RC is also called Coenzyme Q 10 (Q 10 ). The digit in its name is originally derived from number of isoprenyl subunits in its side chain; however, despite the number 10, their number in different organisms varies from 6 to 10 [27]. Q 10 is a substrate for most of the enzymes involved in electron transfer within the inner mitochondrial membrane. Some enzymes use the oxidised form of ubiquinone as an electron acceptor (e.g. both NADH dehydrogenases and succinate dehydrogenase), whereas others use its reduced form (ubiquinol) as an electron source (cytochrome c reductase and alternative oxidase). However, the hydrophobic tail of Q 10 makes this compound unsuitable for in vitro experiments due to very low solubility in aqueous solutions. That is the reason why various Q 10 analogues are used for the in vitro measurement of RC enzyme activities. The literature describes the use of different Q coenzymes for measuring RC enzyme activities; they even vary for individual enzymes. For example, NADH dehydrogenase is measured with Coenzyme Q 1 (Q 1 ) [2], Coenzyme Q 2 (Q 2 ) [7,24] and Decylubiquinone (DB) [26]; succinate dehydrogenase with Q 1 [2] and Q 2 [2,24]; cytochrome c reductase with either reduced Coenzyme Q 2 (Q 2 H) [2] or with reduced Decylubiquinone (DBH) [10,22]; and the alternative oxidase with reduced Coenzyme Q 1 (Q 1 H) and Q 2 H [12,17]. To allow comparability of the results obtained from different laboratories and to simplify stocks of chemicals needed for the assays, we tested the suitability of five commercially available ubiquinone analogues for the measurement of RC enzyme activities.
We used three different species of trypanosomatids as models. They differ in both the composition of their respiratory chain, and the strength of the activities of individual components. This makes them a suitable model for the standardisation of enzyme activity measurement. Phytomonas serpens has lost two "core" respiratory chain enzymescytochrome c reductase and cytochrome c oxidase [15]. Trypanosoma brucei (TB) does not have a fully functional complex I in the procyclic life stage (TB (PF)) [23], which dramatically lowers the activity of the respiratory chain. Reduced ubiquinone is regenerated only by TAO in the blood stream form (TB(BF)) because activity of the rest of the respiratory chain is reduced in this cell cycle stage (for review see [21]). Leishmania tarentolae has no TAO and no measurable NADH dehydrogenase activity [16,18,24]. Coenzyme Q 2 proved to be the optimal substrate for all tested enzymes in trypanosomatids. The universality of the presented results was demonstrated by the use of Coenzyme Q 2 to measure respiratory chain enzyme activities in mitochondrial lysates of a yeast (Saccharomyces cerevisiae) and a vertebrate (chicken liver).

Preparation of mitochondrial lysate
Mitochondria-enriched fractions were obtained as described previously [11]. Next, mitochondria were suspended in 0.5 M aminocaproic acid and 10% (w/v) dodecyl maltoside was added to a final concentration of 2% (w/v). Lysis was performed for 60 min on ice and the lysate was centrifuged for 10 min at maximum speed at 4°C. The supernatant was recovered and protein concentration was determined by the Bradford assay [3].

Coenzyme reduction
The reduced form of coenzyme (ubiquinol) was prepared by reduction of appropriate ubiquinone (Coenzymes Q 1 , Q 2 , Q 4 , Q 10 and DB) using a procedure adapted from [22]. Coenzyme was diluted in acidic ethanol (96% (v/v) ethanol, 1 mM acetic acid) to the final concentration 25 mM. One mL of this solution was mixed with 1 mL of 500 mM NaPi, pH 7.4 and 3 lL of 1 M HCl and the solution was sparged with nitrogen to remove oxygen. Then, 13 mg of sodium dithionite and 1 lL of 1 M HCl were added. After a short vortex, 13 mg of sodium borohydride were added. The resulting colourless solution was extracted three times with 3 mL of cyclohexane and incubated 1-2 h with 200 mg of sodium sulfate to absorb the remaining water in the organic phase. The solution was transferred to a new tube, and the cyclohexane was evaporated by nitrogen sparging. The reduced coenzyme was dissolved in acidic ethanol to the final concentration 10 mM, split into small aliquots, and stored under nitrogen at À80°C until required.
Coenzymes Q 1 and Q 2 (but not Q 4 , Q 10 and DB) could also be reduced by a simpler alternative method adapted from [2]. A 2 mL solution of coenzyme in acidic ethanol, NaPi and HCl was directly mixed with 2 mL of cyclohexane, and 20-30 mg of sodium dithionite was added. The tube was thoroughly mixed by vortex until the solution became colourless. The organic phase was removed, and the extraction was repeated twice. Collected cyclohexane containing reduced coenzyme was evaporated, as described above.

Enzymatic assays
NADH dehydrogenase activities were measured as previously described [7] with some modifications. Briefly, 5 lL of mitochondrial lysate and 5 lL of 20 mM NADH were mixed with 1 mL of NDH buffer (50 mM KPi pH 7.5; 1 mM EDTA, pH 8.5; 0.2 mM KCN). The reaction was started by the addition of 2 lL of 10 mM tested coenzyme. The reaction was followed at 340 nm for 3 min.
Succinate dehydrogenase was measured as previously described [2] with some modifications. Briefly, 5 lL of the mitochondrial lysate was mixed with 1 mL of SDH buffer (25 mM KPi, pH 7.2; 5 mM MgCl 2 ; 20 mM sodium succinate) and incubated in 30°C for 10 min. Next, antimycin A, rotenone, KCN and 2,6-dichlorophenolindophenol were separately added to a final concentration of 2 lg/mL, 2 lg/mL, 2 mM and 50 lM, respectively, and then mixed together. The reaction background was monitored at 600 nm for 5 min and its value was subtracted from measured activity. The reaction itself was started by adding of tested coenzyme to a final concentration of 65 lM, and was followed at 600 nm for 5 min.
The activity of cytochrome c reductase was measured as previously described [10] with minor modifications. Simultaneously, 2 lL of the mitochondrial lysate and 2 lL of 10 mM reduced tested coenzyme were added to 1 mL of QCR buffer (40 mM NaPi, pH 7.4; 0.5 mM EDTA; 20 mM sodium malonate; 50 lM cytochrome c; 0.005% (w/v) dodecyl maltoside) and briefly mixed. The reaction was monitored at 550 nm for 1 min.
Alternative oxidase was measured as previously described [12]. Briefly, 5 lL of mitochondrial lysate was added to 1 mL of 50 mM Tris-HCL (pH 7.4) and incubated in 25°C for 2 min. The reaction was initiated by the addition of reduced tested coenzyme to a final concentration of 150 lM, and was followed at 278 nm for 5 min.
Protein concentrations of mitochondrial lysates were about 8 mg/mL (±2.5 mg/mL) and the measured activity was converted to 1 mg of proteins. All inhibitor solutions were freshly prepared. Rotenone was dissolved in dimethylsulfoxide, DPI in methanol, sodium malonate in water, and antimycin A and salicylhydroxamic acid (SHAM) in ethanol. Inhibitors were added to the assay mixture immediately before the start of the reaction in the concentrations listed in Table 1.

Results and discussion
The respiratory chain in all three trypanosomatids has already been investigated using different Q coenzymes to Average values of enzyme activities are displayed in Figure 2 where measure individual enzyme activities. While NADH dehydrogenase, succinate dehydrogenase and cytochrome c reductase were experimentally tested in all three species, TAO activity was only measured in the case of heterologous expression of the respective T. brucei gene in E. coli [12,17], and its presence was indirectly proven in P. serpens by measurement of oxygen consumption sensitive to TAO-specific inhibitor [24]. In this study, we tested five Q coenzymes that differ in their side chain. However, the length or character of its hydrophobic tail can influence its interaction with the particular examined enzyme [9]. Therefore, we tested with all substrates not only the absolute activity of the examined enzymes, but also the sensitivity of the measured activity to inhibitors specific to individual enzymes: rotenone -Complex I, diphenyl iodonium (DPI) -NDH2, sodium malonate -Complex II, antimycin A -Complex III and SHAM -TAO. We used assays that have already been published to measure each activity. Therefore, we assumed that each individual method was already optimised.
Our goal was to test whether we could obtain comparable or better measurable values with substrates other than those that have been used so far. We confirmed that Q 10 is not suitable for the in vitro measurement of any tested enzyme activity. Similarly, Q 4 was not usable either. Obtained activities with these two substrates were not measureable, or were substantially lower (between 5-and 30-fold) than with the other three coenzymes. The probable reason for both compounds is the high hydrophobicity of their aliphatic chain (see Fig. 1). Therefore, we performed a full set of measurements only with Q 1 , Q 2 and DB (see Fig. 2 and Table 1) and the indicative activity values with Q 4 and Q 10 are given only in the text. We used two different methods to reduce all three coenzymes. While a longer method adapted from [22] totally reduced all three tested coenzymes, the simpler method described by [2] sufficiently reduced only Q 1 and Q 2 . DBH reduced by this method was not a suitable substrate for these measurements.

NADH dehydrogenase
NADH dehydrogenase activity in trypanosomatid mitochondria could correspond to two different enzymes: Complex I and NDH2. While Complex I is sensitive to rotenone [8], NDH2 is rotenone-resistant and is sensitive to DPI at concentrations that do not influence Complex I [1,4,6]. There was almost no activity with Q 4 for P. serpens (2 U) and T. brucei (1.5 U) and zero with Q 10 for both examined trypanosomatids. The highest NADH dehydrogenase activity was obtained with Q 2 as substrate (NADH dehydrogenase Fig. 2, Table 1) and absolute values, as well as sensitivity to inhibitors, are comparable with results reported in the literature [4,24]. A high activity value in T. brucei was also recorded with Q 1 , however essentially no signal in P. serpens disqualifies this coenzyme as a universal substrate for NADH dehydrogenase. Reasonably high activity values were obtained with DB in both P. serpens and T. brucei. However, very low or no efficiency of inhibition suggest that DB is not a suitable substrate for in vitro NADH dehydrogenase activity measurements in trypanosomatids. Minimal sensitivity to rotenone with DB was also demonstrated for the same activity in bovine heart mitochondria [5]. NADH dehydrogenase activity has the most striking difference in relative activity with one substrate between trypanosomatids: from no activity in P. serpens with Q 1 , through 1.5 times lower activity in P. serpens than in T. brucei with Q 2 , and almost equal activities in both species with DB. For all other enzyme substrate combinations, the relative activities between the tested species did not differ significantly. The same is true for the sensitivity to inhibitors. This may reflect the different contribution of two enzymes (Complex I and NDH2) to total activity measured in different trypanosomatid species, or might indicate that the active sites corresponding to this activity differ more than with other enzymes between trypanosomatids.

Succinate dehydrogenase
All three substrates Q 1 , Q 2 and DB were suitable for measurement of succinate dehydrogenase in all tested cell lines. The activities and their relative ratio in individual strains were comparable. Nevertheless, activity with Q 2 was the highest in all three species (Succinate dehydrogenase Fig. 2; Table 1). Inhibition with competitive inhibitor sodium malonate was more than 90% in all combination substrates and trypanosomatids. Our data suggest that succinate dehydrogenase is the enzyme with the lowest requirements to coenzyme Q specificity. Nevertheless, the activities were significantly lower with two of the most hydrophobic substrates (Q 4 : 1 U P. serpens, 7 U T. brucei and 9 U L. tarentolae; Q 10 : 0 U P. serpens, 1 U T. brucei and 4 U L. tarentolae).

Cytochrome c reductase
The relevance of Q 1 H, Q 2 H and DBH for cytochrome c reductase resembles the situation with succinate dehydrogenase.
Activities of all three substrates were comparable in both tested species. However, very low inhibition by antimycin A with Q 1 H shows that this coenzyme is not the best substrate for the enzyme (Cytochrome c reductase Fig. 2; Table 1). Measured activities of cytochrome c reductase with both Q 4 and Q 10 were zero. Although activity with DBH was slightly higher than Q 2 , both compounds are good substrates for the third respiratory chain enzyme. However, the reduction procedure for Q 2 H is simpler than DBH (see section Materials and Methods). This is why we again favour Q 2 H as the most suitable coenzyme for the measurement of cytochrome c reductase activity.

Trypanosome alternative oxidase
TAO activity has the highest differences of absolute activities between comparable cell lines. The values of P. serpens are approximately two times higher than T. brucei (BF) and even 50 times higher than T. brucei (PF) (Alternative oxidase Fig. 2, Table 1). Significantly, the lowest were activities with Q 4 (21 U with P. serpens and 5 U T. brucei (PF)) and Q 10 (23 U P. serpens and 15 U T. brucei (PF); we did not use these two substrates to measure TAO activity in T. brucei (BF). Activities with DBH are remarkably lower than with the other two hydrophylic substrates (in TB (PF) cells activity was not measurable). The highest signals were again with Q 2 H, and only slightly lower with Q 1 H. Despite the fact that in the other laboratory TAO activity was measured with Q 1 H [12,17], in TB (PF) the activity signal was not inhibited by SHAM. For this reason, we do not consider Q 1 H to be a universal substrate for this enzyme. On the basis of the obtained data, we conclude that Q 2 H is the best substrate for TAO.

Respiratory chain enzymes of S. cerevisiae and chicken liver mitochondria
To verify the general validity of our results, we applied Coenzyme Q 2 to measure activities of NADH dehydrogenase, succinate dehydrogenase and cytochrome c reductase in both mitochondrial lysates of S. cerevisiae and chicken liver ( Table 2). The values of measured activities in both organisms were roughly comparable with those obtained for the trypanosomatids investigated in this study. This also confirms the suitability of Q 2 as a substrate for the respiratory chain enzymes in these evolutionarily divergent organisms.

Conclusions
Today's practice in measuring respiratory chain enzyme activities is that different coenzyme Q 10 analogues are used for each enzyme. We have shown that a single variant can be used for all enzymes that use Q coenzymes as substrates. Our data clearly show that out of all the readily commercially available ubiquinones, coenzyme Q 2 is an optimal substrate for all tested enzymes. Q 2 is an appropriate substrate not only for trypanosomatids, but is also suitable for evolutionarily distant organisms, such as yeasts and vertebrates, thus corroborating the general validity of our conclusions.