High throughput microplate respiratory measurements using minimal quantities of isolated mitochondria

Abstract

Recently developed technologies have enabled multi-well measurement of O(2) consumption, facilitating the rate of mitochondrial research, particularly regarding the mechanism of action of drugs and proteins that modulate metabolism. Among these technologies, the Seahorse XF24 Analyzer was designed for use with intact cells attached in a monolayer to a multi-well tissue culture plate. In order to have a high throughput assay system in which both energy demand and substrate availability can be tightly controlled, we have developed a protocol to expand the application of the XF24 Analyzer to include isolated mitochondria. Acquisition of optimal rates requires assay conditions that are unexpectedly distinct from those of conventional polarography. The optimized conditions, derived from experiments with isolated mouse liver mitochondria, allow multi-well assessment of rates of respiration and proton production by mitochondria attached to the bottom of the XF assay plate, and require extremely small quantities of material (1-10 µg of mitochondrial protein per well). Sequential measurement of basal, State 3, State 4, and uncoupler-stimulated respiration can be made in each well through additions of reagents from the injection ports. We describe optimization and validation of this technique using isolated mouse liver and rat heart mitochondria, and apply the approach to discover that inclusion of phosphatase inhibitors in the preparation of the heart mitochondria results in a specific decrease in rates of Complex I-dependent respiration. We believe this new technique will be particularly useful for drug screening and for generating previously unobtainable respiratory data on small mitochondrial samples.

Figure 1. Schematic flowchart for the isolated mitochondria assay using the Seahorse XF24 Analyzer.

Mitochondria are diluted into 1X MAS containing the substrate of choice. Initial conditions refer to any additives or compounds present at 1X at the start of the assay in addition to the substrate (e.g. drug candidate, etc.).

Figure 2. Optimization of isolated mitochondria XF assays.

2A–B, Determination of optimal µg amount of mitochondria/well. 1.25–40 µg/well of mouse liver mitochondria were attached to a V7 polystyrene XF24 plate and the coupling experiment was performed as described in Methods in the presence of succinate/rotenone. Blue vertical lines denote injections of indicated compounds. 2A shows OCR for 1.25–40 µg samples. 2B shows the absolute O₂ tension (in mm Hg) in the microchamber for 1.25–40 µg samples. Note that samples at 10 µg and above show unstable State 3 rates for OCR and depletion of O₂ in the microchamber in panels A and B, respectively. Lettering within data points indicates the group identification number.

Figure 3. Characterization of mitochondrial activity.

3A, Titration of ADP using 5 µg mouse liver mitochondria/well. ADP (0–4 mM) was injected via port A to initiate State 3 respiration and the measurement time was extended to 6 minutes. Note that 2–4 mM ADP is sufficient to maintain a relatively stable State 3 respiration rate for the duration of the measurement period, while lower concentrations show exhaustion of ADP and transition to State 4 respiration. 3B, Alkalinization of the media during phosphorylating respiration (note that unlike the OCR tracings, this data reports absolute pH rather than a rate of change in pH).

Figure 4. Isolated mitochondria remain attached to the plate for the duration of the experiment.

A, The coupling experiment was performed using 5 µg mouse liver mitochondria per well as described in Methods, however, State 3 respiration was allowed to proceed for multiple measurement periods (average OCR per measurement period shown). Note that State 3 respiration does not diminish over multiple mixing and measuring periods, indicating that the mitochondria remain attached to the well for the duration of the assay. B, State 4_o rates using different plate coatings in the presence and absence of 0.2% BSA. No significant differences in State 4_o rates were observed among different plate coatings [none, polyethyleneimine (PEI), and Cell-Tak®]. Note that the absence of BSA resulted in elevated rates of State 4_o respiration, indicative of respiratory uncoupling. C–D, Isolated mitochondria adhered to the XF24 plate as imaged by phase contrast microscopy at 20X magnification before (4C) and after the XF assay (4D).

Figure 5. Comparison of Clark electrode and XF technology shows comparable respiration data between the methods.

Mitochondria isolated from rat heart and mouse liver were used in parallel coupling experiments using either a Hansatech or Rank Clark type electrode or the XF24. Assays were performed as described in Methods for each platform, respectively. Comparison of Basal, State 3, State 4o and State 3u rates between the Hansatech and XF with rat heart using or glutamate/malate as substrate (5A) or succinate/rotenone (5B), respectively. Comparison of Basal, State 3, State 4_o and State 3_u rates between the Rank and XF with mouse liver mitochondria using succinate/rotenone (5C). Data are expressed as mean ± SD from 3 separate experiments in Fig 5A and B, and mean ± SD from 4 experiments in 5C. The high SD in 5C owes to higher rates obtained with one of the four mouse liver preps, rather than variation between methodologies on a given day. Data were analyzed using a two-factor ANOVA with repeated measures on one factor. An interaction was detected only in the data of Fig. 5A, and post-hoc paired comparisons detected lower rates in the XF24 of State 3 and 3_u, and a higher rate of State 4_o respiration with rat heart mitochondria oxidizing glutamate and malate (p<0.05).

Figure 6. Using the Coupling and Electron Flow assays in tandem to elucidate mechanistic activity of test agents.

Coupling (A, C) and electron flow experiments (B, D) were performed as described in Methods. Initial conditions are as follows (with final concentrations listed): A–B, Controls (no additives) or 20 mM sodium azide or 4 µM antimycin-A; C–D, Controls (no additives) or 10 mM malonate or 2.5 µg/ml oligomycin or 2 µM rotenone. See text for further explanation of results.

Figure 7. Assay Reproducibility.

In 7A, intra-assay reproducibility is demonstrated. Assays used 3–5 replicate wells and well-to-well variation within electron flow and coupling assays shows coefficients of variation (CV) <17% in all measurements (except where OCR has been reduced to minimal levels with rotenone or antimycin-A). In 7B, inter-assay reproducibility is demonstrated. Four separate uncoupling experiments from four different preparations of mouse liver mitochondria were averaged to illustrate reproducibility of the assay over multiple days/preparations and shows CVs <20%. The corresponding table indicates the means, standard deviations and% CV, average RCR values are given in the text.

Figure 8. Effects of Phosphatase Inhibitor (PPI) treatment on rat heart mitochondrial respiration.

Mitochondria were treated with a cocktail of phosphatase inhibitors during the isolation procedure as described in the Methods. Respiratory States 3, 4_o and 3_u were measured in the presence (+PPI) or absence (-PPI) of phosphatase inhibitor treatment in the presence of either succinate/rotenone (3 µg/well) or glutamate/malate (6 µg/well) as oxidizable substrates, and rates are expressed per µg mitochondrial protein.