Author Produced

Simultaneous Measurement of Superoxide/Hydrogen Peroxide and NADH Production by Flavin-containing Mitochondrial Dehydrogenases


Your institution must subscribe to JoVE's Biochemistry section to access this content.

Fill out the form below to receive a free trial or learn more about access:



Mitochondria contain several flavin-dependent enzymes that can produce reactive oxygen species (ROS). Monitoring ROS release from individual sites in mitochondria is challenging due to unwanted side reactions. We present an easy, inexpensive method for direct assessment of native rates for ROS release using purified flavoenzymes and microplate fluorometry.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Mailloux, R. J., Young, A., O'Brien, M., Gill, R. M. Simultaneous Measurement of Superoxide/Hydrogen Peroxide and NADH Production by Flavin-containing Mitochondrial Dehydrogenases. J. Vis. Exp. (132), e56975, doi:10.3791/56975 (2018).


It has been reported that mitochondria can contain up to 12 enzymatic sources of reactive oxygen species (ROS). A majority of these sites include flavin-dependent respiratory complexes and dehydrogenases that produce a mixture of superoxide (O2●-) and hydrogen peroxide (H2O2). Accurate quantification of the ROS-producing potential of individual sites in isolated mitochondria can be challenging due to the presence of antioxidant defense systems and side reactions that also form O2●-/H2O2. Use of nonspecific inhibitors that can disrupt mitochondrial bioenergetics can also compromise measurements by altering ROS release from other sites of production. Here, we present an easy method for the simultaneous measurement of H2O2 release and nicotinamide adenine dinucleotide (NADH) production by purified flavin-linked dehydrogenases. For our purposes here, we have used purified pyruvate dehydrogenase complex (PDHC) and α-ketoglutarate dehydrogenase complex (KGDHC) of porcine heart origin as examples. This method allows for an accurate measure of native H2O2 release rates by individual sites of production by eliminating other potential sources of ROS and antioxidant systems. In addition, this method allows for a direct comparison of the relationship between H2O2 release and enzyme activity and the screening of the effectiveness and selectivity of inhibitors for ROS production. Overall, this approach can allow for the in-depth assessment of native rates of ROS release for individual enzymes prior to conducting more sophisticated experiments with isolated mitochondria or permeabilized muscle fiber.


The ultimate goal of nutrient metabolism is to make adenosine triphosphate (ATP). In mammalian cells, this occurs in mitochondria, double-membraned organelles that convert the energy stored in carbon into ATP. The production of ATP begins when carbon is combusted by mitochondria forming two electron carriers, NADH and flavin adenine dinucleotide (FADH2)1. NADH and FADH2 are then oxidized by multi-subunit respiratory complexes I and II, respectively, and the liberated electrons are ferried to the terminal electron acceptor molecular oxygen (O2) at complex IV1. The thermodynamically favorable "downhill" transfer of electrons to O2 at the end of the chain is coupled to the export of protons into the intermembrane space by complexes I, III, and IV. This creates a transmembrane electrochemical gradient of protons (proton-motive force), a temporary form of Gibbs free energy that is tapped by complex V to make ATP2. Electron transfer reactions in mitochondria are not perfectly coupled to ATP production. At various points in the Krebs cycle and the respiratory chain, electrons can prematurely interact with O2 to form ROS3. The most biologically relevant ROS generated by mitochondria are O2●- and H2O2. Although O2●- is often considered the proximal ROS formed by mitochondria, it is now evident that sites of production form a mixture of O2●- and H2O2, which is associated with the free radical chemistry of flavin prosthetic groups4,5. At high levels, ROS can be dangerous, damaging biological constituents required for cell function resulting in oxidative distress6. However, at low amounts, mitochondrial ROS fulfill vital signaling functions. For instance, H2O2 release from mitochondria has been implicated in controlling T-cell activation, stress signaling (e.g., induction of Nrf2 signaling pathways), the induction of cell proliferation and differentiation, insulin signaling and release, and feeding behavior7. Considerable progress has been made in understanding the signaling function of ROS. However, important questions still remain in regard to which enzymes in mitochondria serve as the most important sources and how production is controlled.

ROS release by a site of production depends on several factors: 1) the concentration of the electron donating site, 2) the redox state of the electron donating site, 3) access to oxygen, and 4) the concentration and type of oxidation substrate3,8,9. In mitochondria, other factors like the concentration of NADH and membrane potential strength also influence ROS production8,10. For example, the rate of O2●-/H2O2 production by purified PDHC or KGDHC increases with increasing NADH availability5,11. In this scenario, electrons are flowing backwards from NADH to the FAD center in the E3 subunit of PDHC or KGDHC, the site for O2●-/H2O2 production12. Similarly, the provision of NAD+ has the opposite effect, decreasing ROS release from KGDHC12. Thus, controlling entry or exit of electrons from sites of ROS production can alter how much O2●-/H2O2 is formed. For instance, blocking the E2 subunit of PDHC or KGDHC with CPI-613, a lipoic acid analog, results in an almost 90% decrease in O2●-/H2O2 production13. Similar results can be obtained with the chemical S-glutathionylation catalysts, diamide and disulfiram, which almost abolish O2●-/H2O2 production by PDHC or KGDHC via the conjugation of glutathione to the E2 subunit14. The trade-off for blocking electron flow through the E2 subunit of PDHC and KGDHC is a decrease in NADH production which diminishes ROS formation by the electron transport chain (e.g., complex III). This can also decrease ATP output by the electron transport chain. Overall, blocking electron entry to sites of ROS production can be a highly effective means of controlling O2●-/H2O2 production.

Mitochondria can contain up to 12 potential O2●-/H2O2 sources6. Most of these sites are flavin-containing enzymes which generate a mixture of O2●- and H2O2. Use of different substrate and inhibitor combinations have allowed for the identification of which respiratory complexes and mitochondrial dehydrogenases serve as high capacity O2●-/H2O2 forming sites in different tissues3. PDHC and KGDHC have been shown to serve as high capacity O2●-/H2O2 emitting sites in muscle and liver mitochondria13,15. However, some difficulties remain in regard to the examination of the O2●-/H2O2 forming potential of individual sites in mitochondria and the impact that different substrate and inhibitor combinations have on the activities of the enzymes. This is due to the presence of unwanted side reactions (e.g., formation of O2●-/H2O2 by sites other than the enzyme of interest), contaminating endogenous nutrients (e.g.,fatty acids) or organelles (e.g., peroxisomes which also form O2●-/H2O2), and use of inhibitors that lack selectivity, and/or use of compounds that do not fully inhibit ROS production. Certain inhibitors can also alter the mitochondrial bioenergetics and direction of electron flow, and this alters ROS release from other sites of production and confounds results. Absolute rates for O2●-/H2O2 release from individual sites in mitochondria are also difficult to quantify due to the high concentration of O2●- and H2O2 eliminating enzymes in the matrix and intermembrane space. Therefore, elimination of any competing reactions that can interfere with O2●-/H2O2 release measurements can be useful when identifying high capacity O2●-/H2O2 forming sites.

Here, we present a simple method that allows for the simultaneous examination of O2●-/H2O2 production and NADH formation by purified flavin-dependent dehydrogenases. By using purified enzymes, unwanted ROS forming side reactions and ROS degrading enzymes can be eliminated allowing for a more accurate measure of native O2●-/H2O2 production rates for individual flavoenzymes. This method can be used to directly compare the O2●-/H2O2 forming capacity of different purified dehydrogenases or to screen potential site-specific inhibitors for O2●-/H2O2 release. Finally, measuring O2●-/H2O2 and NADH production simultaneously can allow for a real-time assessment of the relationship between enzyme activity and ROS release capacity.

Subscription Required. Please recommend JoVE to your librarian.


1. Chemicals and Purified Enzymes

  1. Procure the following materials: PDHC and KGDHC of porcine heart origin (or another purified mitochondrial flavoenzyme); H2O2 (30% solution), pyruvate, α-ketoglutarate, NAD+, NADH, CoASH, thiamine pyrophosphate (TPP), mannitol, HEPES, sucrose, EGTA, 3-methyl-2-oxo valeric acid (KMV), superoxide dismutase (SOD), horseradish peroxidase (HRP), 10-Acetyl-3,7-dihydroxyphenoxazine reagent (AUR), and CPI-613.

2. Planning the Assay and Reagent Preparation

  1. Set up the assay according to Table 1 in a 96-well plate.
    NOTE: The total volume for each reaction is 200 µL. Stock concentrations are adjusted such that 20 µL of reagent is added to each well. This ensures rapid addition of cofactors and substrates to each reaction well prior to starting the assay.
  2. Prepare reaction buffer containing 220 mM mannitol, 70 mM sucrose, 1 mM EGTA, and 20 mM HEPES (pH 7.4 with HCl).
    NOTE: This could change depending on the physical properties of different purified enzymes.
  3. Examine the purified KGDHC or PDHC for contamination using enzyme activity assays14,16,17 or by immunoblot or Coomassie stain5.
  4. Prepare all reagents in the reaction buffer as per the working solution concentrations in Table 1.
  5. Store all reagents as 1 mL aliquots at -20 °C. Store pyruvate and α-ketoglutarate at -80 °C in 100-µL aliquots to prevent spontaneous degradation due to auto-oxidation and repeated freeze-thaw.
  6. Prepare the AUR reagent master stock in 1 mL of DMSO and then dilute to 100 µM in reaction buffer. Store protected from light.
    NOTE: Here, the 100 µM working solution is made fresh every 2 to 3 weeks and protected from light.

3. Setting Up the Assay Template

  1. Set up the assay procedure on a monochromatic microplate reader with dual measurement capabilities.
  2. Define the reading procedure by selecting "PROTOCOL". Then click "PROCEDURE".
  3. Set "TEMPERATURE" to 25 °C (Figure 1A).
  4. Click "START KINETIC" to set read conditions (Figure 1A). Set the time and number of read intervals/experiments.
  5. Click "READ" (Figure 1B).
    NOTE: Samples are read from TOP position with a gain of 50. Time: 5 min, read at 30-s intervals. Channel 1: Resorufin fluorescence - excitation/emission = 565 nm:600 nm, wavelength width of 13.5 nm. Channel 2: NADH autofluorescence - excitation/emission = 376 nm:450 nm, wavelength width of 20 nm.
  6. Under "READ", select wells to monitor (e.g., A1-A4 and B1-B4).
  7. Select "END KINETIC".

4. Standard Curves

  1. Thaw the reagents and store on ice until needed.
  2. NADH standard curve (Figure 2A)
    1. Prepare the working solutions for NADH in the concentration range of 0.5-10 mM in reaction buffer.
    2. Add 20 µL aliquots from each NADH working solution concentration to wells containing 180 µL reaction buffer. The final concentration of NADH in each well is 0.05-1 mM.
    3. Click "READ" and set excitation/emission = 376 nm:450 nm, wavelength width of 20 nm.
      NOTE: This is an endpoint read only - kinetic parameters are not required.
  3. AUR standard curve (Figure 2B)
    1. Prepare the working stocks of hydrogen peroxide in reaction buffer; stock concentrations range from 200-4,000 nM.
    2. Add 120 µL of the reaction buffer to each well.
    3. Add 20 µL of HRP (working stock concentration = 30 U/mL, final concentration in the well = 3 U/mL), 20 µL of SOD (working stock concentration = 250 U/mL, final concentration in the well = 25 U/mL), and 20 µL of AUR (working stock concentration = 100 µM, final concentration in the well = 10 µM) to each well containing reaction buffer.
    4. Add 20 µL of each H2O2 working stock solution to each well containing buffer and assay reagents. The final reaction volume is 200 µL and the final concentration of H2O2 in each well is 20-400 nM.
    5. Incubate the plates for 1 min in the plate reader at 25 °C.
    6. Click "READ" and set excitation/emission = 565 nm/600 nm, wavelength width of 13.5 nm. Note that this is an endpoint read only - kinetic parameters are not required.

5. Measuring O2 ●- /H2 O2 Release and NADH Production by KGDHC and PDHC

  1. See Table 1 for the order of addition of the different reagents, the concentration of each working solution, and the volume added to each well.
  2. Add 40 µL of reaction buffer to each well. For assays using KMV or CPI-613, add 20 µL of buffer to each well.
  3. Add 20 µL of PDHC or KGDHC (working stock of 1 U/mL, final concentration per well = 0.1 U/mL) to each well.
  4. Incubate PDHC and KGDHC at 25 °C for 2 min.
  5. Add 20 µL of KMV (working stock = 100 mM, final concentration = 10 mM) or 20 µL of CPI-613 (working stock concentration = 1.5 mM, final concentration = 150 µM) (Table 2).
    NOTE: This step can be excluded if inhibitors are not being used for the assays.
  6. Incubate for 1 min at 25 °C.
    NOTE: This step can be excluded if inhibitors are not being used for the assays.
  7. Add 20 µL of HRP (working stock concentration = 30 units/mL, final concentration in the well = 3 units/mL) and 20 µL of SOD (working stock concentration = 250 units/mL, final concentration in the well = 25 units/mL) to each well.
  8. Add 20 µL of coenzyme A (working stock concentration = 1 mM, final concentration = 0.1 mM), 20 µL of TPP (working stock concentration = 3 mM, final concentration = 0.3 mM), and 20 µL of NAD+ (working stock concentration = 10 mM, final concentration = 1 mM) to each well.
  9. Remove the AUR reagent from protective tinfoil covering and add 20 µL to each well.
  10. Add 20 µL of pyruvate (working stock concentration ranging from 1-100 mM, final concentration for assays = 0.1-10 mM) to wells containing PDHC and α-ketoglutarate (working stock concentration ranging from 1-100 mM, final concentration for assays = 0.1-10 mM) to wells containing KGDHC.
  11. Click "READ" to start kinetic measurement.

6. Data Analysis

  1. Hold down the 'CTRL' button on the keyboard and click wells to generate one graph containing all kinetic measurements corresponding to channel 1 (Figure 3A).
  2. Click "data" in the view icon in the upper right corner for raw values for relative fluorescence units (RFU) measured at the different time points (Figure 3B).
  3. Export the values for analysis (Table 3).
  4. Click the "Graph" drop down menu on top right (Figure 3B) to access RFU values associated with the fluorescence channel 2.
  5. Repeat steps 6.1 to 6.3 for channel 2.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

Figure 3A provides a representative trace for the RFU collected during the simultaneous measurement of H2O2 and NADH production by purified KGDHC. The raw RFU data for each time interval is depicted in Figure 3B. The raw RFU data are then exported for analysis. By extrapolating from standard curves presented in Figure 2, the absolute amount of NADH and H2O2 formed at each measurement interval over the 5-min period can be determined. An example of this calculation is provided in Table 3. Once the amount of NADH and H2O2 formed by KGDHC or PDHC at the different intervals has been determined, values are copied and inputted into a graphing/statistical software for further analysis. Using the XY graph-type analysis, the amount of NADH (µmol) and H2O2 (pmol) formed by KGDHC or PDHC can be plotted as a function of time (Figure 4A). Using these values, the rate of NADH and H2O2 formation during substrate oxidation can also be calculated. As shown in Figure 4B, KGDHC and PDHC display different characteristics in terms of NADH and H2O2 forming capability.

Once it has been established that KGDHC and PDHC are enzymatically active and NADH and H2O2 formation can be tracked simultaneously, ROS release properties and enzyme activities can be measured. We chose to compare KGDHC and PDHC since they are highly homologous in basic structure and have been documented to be high capacity sites for ROS production5. Figure 5 and Figure 6 demonstrate the dependency of NADH and H2O2 production on substrate concentration. PDHC is easily saturated by low amounts of substrate with the rate of NADH and H2O2 production reaching its maximum at 0.1-0.5 mM pyruvate. KGDHC, on the other hand, displays incremental increases in NADH and H2O2 production with increasing α-ketoglutarate (Figure 5). This demonstrates that KGDHC and PDHC, although highly homologous in basic structure, have different kinetic properties in terms of how much substrate is required to induce maximal ROS release. Figure 5 also demonstrates that H2O2 release from KGDHC correlates strongly with NADH production and substrate availability. PDHC on the other does not have the same kinetic characteristics. KGDHC and PDHC also differ in terms of NADH and O2●-/H2O2 forming capacity when NAD+ concentration is varied (Figure 6).

By taking advantage of the simultaneous measurement of NADH and H2O2, potential site-specific inhibitors for ROS production can be screened. This can help determine the amount of inhibitor that is required to induce maximal inhibition of H2O2 release. In addition, by measuring NADH production at the same time, the site for inhibitor action as well as the source of ROS can be identified. Previous studies have shown that the α-ketoglutarate analog, KMV, is highly effective at inhibiting ROS release from KGDHC (≥ 90% inhibition)13,15. We also showed that CPI-613, a lipoic acid analog that blocks dihydrolipoamide of the E2 subunit of PDHC and KGDHC, can also inhibit ROS release from either complex by ~ 90%13,14. The site of action for KMV and CPI-613 are shown in Figure 7.CPI-613 is a sulfur containing molecule and it is therefore advised that thiol-rich reducing agents (dithiothreitol; DTT) or proteins (albumin) are omitted from the procedure13. Figure 7 demonstrates that both KMV and CPI-613 were both highly effective at limiting both NADH and H2O2 by KGDHC. Likewise, CPI-613 completely abolished H2O2 release from PDHC (Figure 7). Collectively, these results demonstrate that KMV and CPI-613 are highly effective inhibitors for H2O2 release and can thus be applied to experiments with isolated mitochondria. Such a screen would be highly beneficial for other potential ROS-forming enzymes since it allows for a direct assessment of the effectiveness and selectivity of different inhibitors on purified flavin-containing dehydrogenases. The simultaneous measurement of NADH also has the added benefit of allowing the preliminary identification of the potential source of ROS within an enzyme complex. Figure 7 provides a depiction of the catalytic mechanism for KGDHC, which is highly homologous to PDHC. The observation that KMV and CPI-613, which block the E1 and E2 subunit of KGDHC, almost completely abolished both H2O2 and NADH production, confirms that the E3 subunit, likely the FAD, is the ROS source18. CPI-613 exerted a similar effect on PDHC confirming that the principle site for H2O2 release is the E3 subunit18. A similar approach can be used for other dehydrogenases. This would require the identification of inhibitors that block electron flow through different parts of an enzyme complex.

Figure 1
Figure 1: Representative screen shots for assay setup. (A) Depiction of procedure set up for the kinetic measurement of changes in NADH and resorufin fluorescence. (B) Setting up the fluorescence parameters for the assay. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Representative standard curves for NADH and resorufin fluorescence. Curves are utilized to estimate the rate of NADH and O2●-/H2O2 production. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Representative traces for the raw RFU data collected during the assay. (A) Curves for each well that was read during the assay. Plate view curves are generated by holding down the CTL button and clicking on each well that is being read. (B) For data export, DATA is clicked under VIEW to generate a table containing the representative RFU values. The SPREADSHEET export button is then clicked to export the raw results. Please click here to view a larger version of this figure.

Figure 4
Figure 4: Direct comparison of the H2O2 and NADH forming rates of KGDHC and PDHC. (A) Simultaneous measurement of H2O2 and NADH forming potential of KGDHC and PDHC. (B) Calculation of the rate of H2O2 and NADH formation by KGDHC and PDHC. Direct comparison of NADH and ROS production by both flavin-containing enzymes demonstrates that KGDHC produces more H2O2 which is related to its higher enzymatic capacity. n = 4, mean ± SEM. Please click here to view a larger version of this figure.

Figure 5
Figure 5: Simultaneous measurement of H2O2 and NADH forming rates by KGDHC and PDHC oxidizing different substrate concentrations. Changes in fluorescence were measured as described previously except the final substrate concentration was varied from 0 mM to 10 mM. This approach was used to ascertain the dependency of H2O2 release on the kinetic properties of KGDHC and PDHC. n = 4, mean ± SEM. Please click here to view a larger version of this figure.

Figure 6
Figure 6: KGDHC and PDHC also display different H2O2 and NADH forming properties when NAD+ levels are changed. Changes in fluorescence were measured as described previously except the final concentration of NAD+ was varied from 0 mM to 1 mM. This approach was used to ascertain the dependency of H2O2 release on the kinetic properties of KGDHC and PDHC. n = 4, mean ± SEM. Please click here to view a larger version of this figure.

Figure 7
Figure 7: Screening of different site-specific inhibitors for NADH and H2O2 production. (A) Diagram depicting the catalytic cycle for KGDHC and the sites of action for KMV and CPI-613. Note that PDHC has a homologous catalytic cycle, except that KMV does not compete for the pyruvate binding site. E1: pyruvate or α-ketoglutarate dehydrogenase, E2: dihydrolipoamide acyltransferase, E3: dihydrolipoamide dehydrogenase. (B) Effect of KMV and CPI-613 on NADH and H2O2 production by KGDHC and PDHC. Samples were preincubated in KMV (10 mM) or CPI-613 (150 µM) and then assayed for NADH and H2O2 production. n = 4, mean ± SEM. Please click here to view a larger version of this figure.

Reagent  Working stock concentration Volume added to well Final concentration in well
KGDHC or PDHC 1 unit/mL 20 µL 0.1 units/mL
Superoxide dismutase 250 units/mL 20 µL 25 units/mL
Horseradish peroxidase 30 units/mL 20 µL 3 units/mL
Thiamine pyrophosphate 3 mM 20 µL 0.3 mM
Coenzyme A 1 mM 20 µL 0.1 mM
NAD+ 10 mM 20 µL 1 mM
10-Acetyl-3,7-dihydroxyphenoxazine  100 µM 20 µL 10 µM
Substrate 1-100 mM 20 µL 0.1-10 mM

Table 1: Example of a basic protocol set up for the measurement of O2-/H2O2 and NADH production by purified PDH or KGDH. Since 20 µL of each reagent is added to each well and the final volume/reaction is 200 µL, the final concentration of each reagent is 10-fold less than the stock concentration (e.g., the final concentration of NAD+ in the reaction mixture is 1 mM). Note that reagents should be added to the well in the order presented in the table.

Inhibitor Target
3-methyl-2-oxo valeric acid E1 subunit of KGDHC
CPI-613 E2 subunit of oxo acid dehydrogenases

Table 2: List of inhibitors and their targets. Sites of action are also depicted in Figure 8.

NADH autofluorescence
Time T° 376,450 A1 A2 A3 A4 A5 A6 A7 A8
0:00:04 24.4 99 99 91 88 85 82 58 88
0:00:34 24.4 128 107 99 108 81 80 77 78
0:01:04 24.4 124 128 117 139 97 88 97 76
0:01:34 24.4 146 142 129 132 91 97 84 78
0:02:04 24.4 161 153 148 146 90 107 95 88
0:02:34 24.4 171 160 156 145 97 103 111 122
0:03:04 24.4 176 176 161 169 109 120 99 116
0:03:34 24.4 186 180 179 167 118 123 121 119
0:04:04 24.4 201 182 187 183 127 124 119 131
0:04:34 24.4 225 190 195 186 136 143 124 132
0:05:04 24.4 227 227 203 197 115 143 121 143
Resorufin fluorescence
Time T° 565,600 A1 A2 A3 A4 A5 A6 A7 A8
0:00:00 24.4 204 178 169 152 148 113 107 90
0:00:30 24.4 224 188 167 152 144 116 124 112
0:01:00 24.4 212 186 178 161 153 114 122 120
0:01:30 24.4 212 188 165 164 160 132 127 116
0:02:00 24.4 222 203 190 169 154 128 133 127
0:02:30 24.4 217 198 193 179 165 129 147 130
0:03:00 24.4 223 218 202 179 183 142 145 137
0:03:30 24.4 239 223 208 188 179 150 155 148
0:04:00 24.4 236 219 202 183 180 163 156 153
0:04:30 24.4 249 220 214 177 199 149 150 167
0:05:00 24.4 251 232 225 196 211 170 169 161

Table 3: Spreadsheet view of the results collected from the set of experiments. Raw RFU values generated from the experiment are exported to a spreadsheet for further analysis. Using standard curves generated prior to experimentation, the rate of NADH and H2O2 production can be calculated in the spreadsheet.

Subscription Required. Please recommend JoVE to your librarian.


This protocol is advantageous since, 1) it eliminates any competing reactions that may otherwise interfere with H2O2 detection (e.g., antioxidant systems or other sources of ROS), 2) provides a direct assessment of the native rate of ROS release by a flavin-containing mitochondrial dehydrogenase, 3) allows the comparison of the native ROS release rates of two or more purified flavin-based dehydrogenases, 4) can allow for a direct comparison of the rate of ROS release and enzyme activity, and 5) allows screening of selective competitive inhibitors for ROS release. Our group and others have used this method in previous publications to examine the regulation of ROS by allosteric and redox signaling mechanisms prior to conducting experiments with mitochondria or permeabilized muscle14,16,17,19,20.

Modifications and Troubleshooting:

This procedure uses basic equipment available at almost every institution. The approach is also low cost and uses common chemicals available through various suppliers. The buffer composition may vary according to enzyme properties - although, we have found that a basic reaction buffer consisting of sucrose, HEPES, mannitol, and EGTA works well in most applications. Raw values for resorufin fluorescence are typically around 50-500 RFU for a ROS releasing enzyme diluted to 0.1 U/mL. Higher RFU values may indicate the AUR reagent has been auto-oxidized due to exposure to light or repeated freeze-thaw. Therefore, AUR working solutions should be routinely replaced every 2 to 3 weeks (depending on the frequency of use). Impurities in the enzyme preparation or denaturation can lead to lower than normal ROS and NADH readings. Enzyme purity should be routinely monitored by gel electrophoresis. Kinetic properties (Km and Vmax for substrate and NAD+) should also be assessed immediately after purchase or purification of an enzyme and then routinely monitored thereafter. Most of the reagents used in the assay are stable and can be subjected to multiple freeze-thaw cycles. Pyruvate and α-ketoglutarate should be stored in aliquots of 100 µL at -80 °C to avoid auto-oxidation. Avoid using reducing agents like DTT since it forms O2●-/H2O2 directly through radical formation in PDHC and KGDHC (and potentially other flavoenzymes)17. Assays can be conducted at 37 °C; however, our group has found that conducting these experiments at 25 °C is ideal for providing more consistent data.

Limitations of the Technique:

One major hurdle associated with this technique is access to purified flavin-containing mitochondrial enzymes. Here, we used commercially available KGDHC and PDHC to demonstrate that this technique can be used to directly compare the ROS release capacity and enzyme activity of flavin-containing mitochondrial dehydrogenases. However, most ROS producing flavin-containing enzymes are not commercially available and must be purified in the laboratory. This can be challenging considering some of these enzymes are membrane bound or absorb to the mitochondrial inner membrane. New protocols for the successful purification of membrane-bound flavin-containing enzymes that produce ROS are now available. This includes purification methods for complex I, complex II, and sn-glycerol-3-phosphate dehydrogenase21,22,23. Therefore, even though most of the listed ROS generators may not be commercially available, protocols exist for their isolation and purification. The physiological relevance of the results collected using purified flavin-based enzymes is also a limitation. Therefore, findings collected with purified enzymes should be followed up with assays using isolated mitochondria or permeabilized tissue. Five separate studies used this approach to investigate mechanisms for regulation of ROS release by KGDHC and PDHC14,16,17,19,20. Preliminary results for the role of redox switches in controlling ROS release were initially tested using purified KGDHC and PDHC, and then mechanisms for regulation were further analyzed in liver and cardiac mitochondria and muscle fiber14,16,17,20.

Significance with Respect to Existing Methods:

Measuring ROS release from individual sites of production in mitochondria can be challenging due to the presence of competing side reactions that also produce O2●-/H2O2 and antioxidant systems that may otherwise lead to the underestimation of native rates. The membrane potential can also alter ROS release rates from various sites and the use of inhibitors can alter electron flow, confounding O2●-/H2O2 formation measurements. The method presented here allows the direct evaluation of the ROS release capacity from individual enzymes while simultaneously assessing the relationship between native rates of O2●-/H2O2 and enzyme activity. This inexpensive method can serve as a tool for comparing the native ROS release rates for individual flavin-containing mitochondrial enzymes prior to conducting more sophisticated experiments with mitochondria, cells, or tissue. In addition, the method presented herein can be utilized to pre-screen the effectiveness and selectivity of inhibitors for ROS release prior to conducting assays with isolated mitochondria.

Future Applications:

The method presented herein has been used in five separate studies aimed at investigating control over ROS release from PDHC and KGDHC14,16,17,19,20. In these studies, inhibitors were screened and ROS release and NADH producing rates were evaluated using purified enzymes. Results collected with the purified enzymes were then verified using isolated mitochondria and permeabilized muscle fibers14,16,17,20. Therefore, this method can be applied to a number of future studies aimed at studying native ROS release rates from mitochondria, cells, and muscle fibers.

Critical Steps in the Protocol:

Enzyme purity and its kinetic properties should be routinely monitored. The AUR working stock reagent should also be replaced frequently. This will ensure results collected are consistent. The order of addition of reagents is vital for running a successful experiment (Table 1). Inhibitors should be titrated to determine the concentration required to elicit maximum inhibition of ROS release. In this study, we chose 10 mM KMV and 150 µM CPI-613 since we had previously determined that these concentrations induce maximal inhibition of ROS release by KGDHC and PDHC13.

Subscription Required. Please recommend JoVE to your librarian.


There is nothing to disclose.


This work was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC). Video production was carried out in collaboration with the Center for Innovation in Teaching and Learning (CITL) at Memorial University of Newfoundland.


Name Company Catalog Number Comments
Pyruvate dehydrogenase complex SIGMA P7032-10UN purified flavoenzyme
alpha-ketoglutarate dehydrogenase complex SIGMA K1502-20UN purified flavoenzyme
30% hydrogen peroxide solution SIGMA HX0640-5 reagent, standard curves
NAD+ SIGMA N0632-1G reagent, activity/ROS release assay
NADH SIGMA N4505-100MG reagent, standard curves
pyruvate SIGMA P2256-5G reagent, activity/ROS release assay
alpha-ketoglutarate SIGMA 75892-25G reagent, activity/ROS release assay
CoASH SIGMA C3019-25MG reagent, activity/ROS release assay
thiamine pyrophosphate SIGMA C8754-1G reagent, activity/ROS release assay
mannitol SIGMA M4125-100G buffer component
Hepes SIGMA H3375-25G buffer component
sucrose SIGMA S7903-250G buffer component
EGTA SIGMA E3889-10G buffer component
KMV SIGMA 198978-5G reagent, ROS release inhibitor
CPI-613 Santa Cruz sc-482709 reagent, ROS release inhibitor
SOD SIGMA S9697-15KU reagent, ROS release detection
horseradish peroxidase SIGMA P8375-1KU reagent, ROS release detection
Amplex Ultra Red Thermofisher A36006 reagent, ROS release detection
Biotech Synergy 2 microplate reader BioTek Instruments microplate reader for assays
Gen5 software BioTek Instruments software, used for collection of raw RFU
Graphpad Prism Graphpad software software, data analysis
Microsoft EXCEL Microsoft software, data analysis



  1. Mailloux, R. J. Teaching the fundamentals of electron transfer reactions in mitochondria and the production and detection of reactive oxygen species. Redox Biol. 4, 381-398 (2015).
  2. Nicholls, D. G. Forty years of Mitchell's proton circuit: From little grey books to little grey cells. Biochim Biophys Acta. 1777, (7-8), 550-556 (2008).
  3. Brand, M. D. Mitochondrial generation of superoxide and hydrogen peroxide as the source of mitochondrial redox signaling. Free Radic Biol Med. 100, 14-31 (2016).
  4. Massey, V. Activation of molecular oxygen by flavins and flavoproteins. J Biol Chem. 269, (36), 22459-22462 (1994).
  5. Mailloux, R. J., Gardiner, D., O'Brien, M. 2-Oxoglutarate dehydrogenase is a more significant source of O2(.-)/H2O2 than pyruvate dehydrogenase in cardiac and liver tissue. Free Radic Biol Med. 97, 501-512 (2016).
  6. Sies, H., Berndt, C., Jones, D. P. Oxidative Stress. Annu Rev Biochem. 86, 715-748 (2017).
  7. Kuksal, N., Chalker, J., Mailloux, R. J. Review. Progress in understanding the molecular oxygen paradox - function of mitochondrial reactive oxygen species in cell signaling. Biol Chem. (2017).
  8. Murphy, M. P. How mitochondria produce reactive oxygen species. Biochem J. 417, (1), 1-13 (2009).
  9. Wong, H. S., Dighe, P. A., Mezera, V., Monternier, P. A., Brand, M. D. Production of superoxide and hydrogen peroxide from specific mitochondrial sites under different bioenergetic conditions. J Biol Chem. (2017).
  10. Harper, M. E., Green, K., Brand, M. D. The efficiency of cellular energy transduction and its implications for obesity. Annu Rev Nutr. 28, 13-33 (2008).
  11. Ambrus, A., et al. Formation of reactive oxygen species by human and bacterial pyruvate and 2-oxoglutarate dehydrogenase multienzyme complexes reconstituted from recombinant components. Free Radic Biol Med. 89, 642-650 (2015).
  12. Tretter, L., Adam-Vizi, V. Generation of reactive oxygen species in the reaction catalyzed by alpha-ketoglutarate dehydrogenase. J Neurosci. 24, (36), 7771-7778 (2004).
  13. Slade, L., et al. Examination of the superoxide/hydrogen peroxide forming and quenching potential of mouse liver mitochondria. Biochim Biophys Acta. 1861, (8), 1960-1969 (2017).
  14. O'Brien, M., Chalker, J., Slade, L., Gardiner, D., Mailloux, R. J. Protein S-glutathionylation alters superoxide/hydrogen peroxide emission from pyruvate dehydrogenase complex. Free Radic Biol Med. 106, 302-314 (2017).
  15. Quinlan, C. L., et al. The 2-oxoacid dehydrogenase complexes in mitochondria can produce superoxide/hydrogen peroxide at much higher rates than complex I. J Biol Chem. 289, (12), 8312-8325 (2014).
  16. Mailloux, R. J., Craig Ayre, D., Christian, S. L. Induction of mitochondrial reactive oxygen species production by GSH mediated S-glutathionylation of 2-oxoglutarate dehydrogenase. Redox Biol. 8, 285-297 (2016).
  17. Mailloux, R. J., Gardiner, D., O'Brien, M. 2-Oxoglutarate dehydrogenase is a more significant source of O2(.-)/H2O2 than pyruvate dehydrogenase in cardiac and liver tissue. Free Radic Biol Med. 97, 501-512 (2016).
  18. Ambrus, A., Adam-Vizi, V. Human dihydrolipoamide dehydrogenase (E3) deficiency: Novel insights into the structural basis and molecular pathomechanism. Neurochem Int. (2017).
  19. Young, A., Gardiner, D., Brosnan, M. E., Brosnan, J. T., Mailloux, R. J. Physiological levels of formate activate mitochondrial superoxide/hydrogen peroxide release from mouse liver mitochondria. FEBS Lett. 591, (16), 2426-2438 (2017).
  20. Fisher-Wellman, K. H., et al. Mitochondrial glutathione depletion reveals a novel role for the pyruvate dehydrogenase complex as a key H2O2-emitting source under conditions of nutrient overload. Free Radic Biol Med. 65, 1201-1208 (2013).
  21. Kussmaul, L., Hirst, J. The mechanism of superoxide production by NADH:ubiquinone oxidoreductase (complex I) from bovine heart mitochondria. Proc Natl Acad Sci U S A. 103, (20), 7607-7612 (2006).
  22. Huang, L. S., Borders, T. M., Shen, J. T., Wang, C. J., Berry, E. A. Crystallization of mitochondrial respiratory complex II from chicken heart: a membrane-protein complex diffracting to 2.0. A. Acta Crystallogr D Biol Crystallogr. 61, (Pt 4), 380-387 (2005).
  23. Yeh, J. I., Chinte, U., Du, S. Structure of glycerol-3-phosphate dehydrogenase, an essential monotopic membrane enzyme involved in respiration and metabolism. Proc Natl Acad Sci U S A. 105, (9), 3280-3285 (2008).



    Post a Question / Comment / Request

    You must be signed in to post a comment. Please or create an account.

    Usage Statistics