Methods, compositions and kits for assaying mitochondrial function

Abstract

The invention provides methods, compositions, devices, and kits relating to the use of cholesterol-dependent cytolysins (e.g., PFOs) for measuring intracellular mitochondrial activity.

Claims

1. A method for introducing an exogenous agent into a permeabilized cell comprising: (a) incubating a cell with a cholesterol-dependent cytolysin that lacks cysteine residues and that lacks an N-terminal signal sequence, thereby selectively permeabilizing the plasma membrane of the cell; and (b) introducing an exogenous agent into the permeabilized cell.

2. The method of claim 1, wherein the exogenous agent is a nucleic acid.

3. The method of claim 1, wherein the exogenous agent is protein.

4. The method of claim 1, wherein the exogenous agent is an antibody.

5. The method of claim 1, wherein the exogenous agent is a dye or fluorescent protein.

6. The method of claim 1, wherein the dye is a calcium sensitive dye or a pH-sensing dye.

7. The method of claim 1, wherein the exogenous agent is a reducing agent, a respiratory chain inhibitor, an oxidative phosphorylation inhibitor, an uncoupling agent, a transport inhibitor, an ionophore or a Krebs cycle inhibitor.

8. The method of claim 1, wherein the exogenous agent is a test agent.

9. The method of claim 1, further comprising measuring the effect of the exogenous agent on the cell.

10. The method of claim 9, wherein said measuring comprises measuring an intracellular function of the cell.

11. The method of claim 10, wherein the intracellular function is a mitrochondrial function.

12. The method of claim 1, wherein the cell is a mammalian cell.

13. The method of claim 1, wherein the exogenous agent cannot cross the plasma membrane of a non-permeabilized cell.

14. The method of claim 1, wherein the cholesterol-dependent cytolysin is a variant of a Perfringolysin O (PFO).

15. The method of claim 1, wherein the exogenous agent is a substrate for a mitochondrial enzyme.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 provides a non-limiting overview of glucose metabolism in a typical mammalian cell. Relationships of the glycolysis, TCA cycle and OxPhos system are shown. The reactions generating and oxidizing NADH and FADH.sub.2 are shown. CC: Cytochrome c; Q: ubiquinone; CI-V: OxPhos Complexes I-V; MAS: malate-aspartate shuttle: GPS: glycerol-3-phosphate shuttle. The measurements of lactate, O.sub.2 and CO.sub.2 levels in the immediate surroundings of cells are expected to give the rates of glycolysis, respiration and the TCA cycle.

(2) FIG. 2 illustrates digitonin (DIG)-based cell permeabilization. Respiration of INS1E was measured in Ca2+-free buffer containing succinate as substrate using the XF-analyzer. After measuring basal respiration rate, cells were permeabilized with DIG (1st arrow) and then ADP, with or without Cytochrome c (CC) was added (2.sup.nd arrow). Second additions were buffer, ADP, and ADP+CC for groups DIG, DIG+ADP, and D1G+ADP+CC respectively.

(3) FIGS. 3A and 3B illustrate PFO performance compared with digitonin performance. FIG. 3A shows primary pancreatic -cells. FIG. 3B shows Chinese hamster lung fibroblasts. Injections were as follows: a was 1 mM ADP+10 M Cytochrome c (Cyt C or CC) with nPFO (25 nM) or Digitonin (DIG, 0.010% in FIG. 3A, 0.005% in FIG. 3B) in the presence of succinate as respiratory substrate; and b was 1 g/ml oligomycin. OCR rates either shown as absolute value (FIG. 3A) or normalized to 3rd rate (FIG. 3B).

(4) FIG. 4 shows that PFO-based assays do not need exogenous Cytochrome c. Succinate supported, ADP stimulated INS1E cells respiration is shown. Injections were as follows: a was nPFO, b was ADP with or without Cytochrome c.

(5) FIGS. 5A-5C illustrate OxPhos capacity and respiratory (ETC/RC) capacity. ADP- and FCCP-stimulated respirations give OxPhos and respiratory capacities, respectively. FIG. 5A shows relative levels of ADP- and FCCP-stimulated respiration in Chinese hamster lung fibroblasts, the V79-G3. Injections were as follows: a was 1 mM ADP; b was 2 M FCCP. FIG. 5B shows oligomycin sensitivity (coupling test) of the ADP-stimulated respiration in permeabilized V79-G3 cells compared to the sensitivity of basal respiration in intact cells. Injections were as follows a was 1 mM ADP or ADP+nPFO; and b was 2 g/ml oligomycin. FIG. 5C shows respiratory decline following nPFO permeabilization is due to ADP limitation resulting from the dilution of the cytosol. Injections were as follows: a was 5 nM nPFO; and b was buffer (nPFO) or 1 mM ADP (nPFO+ADP) or 1 mM ADP+104M Cytochrome c (nPFO+ADP+CC).

(6) FIGS. 6A and 6B show comparable levels of ADP-stimulated and FCCP-stimulated respiration in rat pancreatic cells. FIG. 6A shows rat insulinoma INS1E cells. Injections were as follows: a was nPFO+ADP; and b was FCCP. FIG. 6B shows primary rat 0/islet-cells. Injections were as follows a was nPFO. b was succinate+ADP, and c was FCCP. The lower rates of basal respiration in primary cells are due to pre-starvation and the absence of any exogenous substrates prior to injection arrow b.

(7) FIGS. 7A-7C show functional assays for Complex I deficiency using the PFO-based method. FIG. 7A shows results of a complex I mutant showing the lack of respiration on Complex I substrates. FIG. 7B shows complemented control cells showing the restoration respiration on Complex I substrates. FIG. 7C shows the data from panel B with the basal rate before arrow a is set to 0. Injections were as follows: a was nPFO, and b was succinate (Suc), or glutamate+malate (Glu+Mal).

(8) FIG. 8 illustrates limitations in substrate supply to -cell ETC/RC following oligomycin treatment, and sensitivity to Ca2+-mediated mitochondrial permeability transition. All groups were treated with oligomycin (2 g/ml) and then after >60 min cells were succinate and/or FCCP with or without nPFO. Groups are as followings. Oligo_Buffer indicates oligomycin treated only; Oligo_SF indicates oligomycin+succinate+FCCP only; Oligo_PF indicates oligomycin+nPFO+FCCP followed by succinate addition later, Oligo_PSF indicates oligomycin+nPFO+succinate+FCCP; Oligo_PSFCa indicates oligomycin and 1.3 mM CaCl.sub.2 added after nPFO+succinate+FCCP additions.

(9) FIG. 9 provides an overview of a PFO purification scheme.

(10) FIG. 10 provides an outline of a working model showing that -cell respiration is primarily supported by the redox shuttles when cells are intact. Complex I-dependent respiration should drop significantly when -cells are permeabilized if it is primarily supported by redox shuttle.

(11) FIGS. 11A and 11B show that -cells display low level of Complex I-dependent respiration. FIG. 11A shows Complex I, II, & III-dependent respiration determined by measuring respiratory inhibition with the increasing concentrations of rotenone (Rot), thenoyltrifluoroacetone (TTFA), antimycin A (Ant A) respectively compared to control (Con). FIG. 11B shows percent maximal respiratory inhibition by rotenone, TTFA and antimycin A in INS1E cells.

(12) FIGS. 12A-12D show that permeabilized -cells do not display Complex I-dependent respiration: FIG. 12A shows results from primary -cells. FIG. 12B shows results from primary astrocytes. FIG. 12C shows results from CCL16 lung fibroblasts. FIG. 12D shows NAD(P)H levels in isolated mitochondria in the presence of glutamate+malate and rotenone. Arrows indicate the following: a indicates cell permeabilization; and b indicates substrate+ADP (for FIGS. 12A,12B) or ADP (for FIG. 12C). In FIG. 12C, all substrates were added in the respiration buffer before permeabilization. Pyruvate, glutamate, malate, isocitrate, -ketoglutarate and rotenone are indicated with one letter symbols: P, G, M, I, K and R respectively.

(13) FIGS. 13A-13C show that -cell's relative respiratory (ETC/RC) and oxidative phosphorylation (OxPhos) capacities are comparable across cell types. FIG. 13A shows INS1E cells. FIG. 13B shows V79 Chinese hamster lung fibroblasts. FIG. 13C shows primary -cells. Arrows indicate the following: a indicate permeabilization with ADP (for FIGS. 13A, 13B) or without ADP (for FIG. 13C); b indicates FCCP (for FIGS. 13A,13B) or ADP (for FIG. 13C); c indicates FCCP. Maximal respiratory stimulations with ADP and FCCP were taken as the indices of the oxidative phosphorylation and respiratory capacities, respectively.

(14) FIGS. 14A and 14B show that -cell respiration gradually declines when oxidative phosphorylation is inhibited. Test of substrate limitation vs. respiratory chain dysfunction. FIG. 14A shows that glucose (Glu) and pyruvate (Pyr) stimulate respiration in INS1E cells, which gradually declines in the presence of oligomycin. Lo & Hi indicates 2 mM and 16.7 mM glucose. FIG. 14B shows the respiratory decline in the presence of oligomycin is due to the limitations in substrate supply to the respiratory chain. Letters P, S, F indicate permeabilization, succinate and FCCP, respectively.

(15) FIG. 15 shows inter-relationships among the glycolysis, TCA cycle and OxPhos. Emphasis is given to reactions that generate electron donors such as NADH and FADH.sub.2 to support OxPhos as indicated. The NADH and FADH.sub.2 generating reactions are shown by broken lines, while those consuming them are shown by solid lines. Reversible reactions are shown by lines with arrowheads on both ends. F-1, 6-BP, fructose-1, 6-bisphosphate; G-3-P, glyceraldehyde-3-phosphate; GPS: glycerol-3-phosphate shuttle; MAS: malate aspartate shuttle; CI-V: OxPhos complexes I-V; .p: proton motive force; mitochondrial membrane potential; pH: pH difference across mitochondrial inner membrane; Pi: inorganic phosphate; Q: ubiquinone.

(16) FIGS. 16A-16F show results of an assay of mitochondrial function with digitonin-permeabilized cells. cells were grown as described in Example 8. Ca2+-free LKB buffer containing 10 mM succinate was used in the presence of 2 mM (FIGS. 16A-16C) or 15 mM glucose (FIGS. 16D-16F). FIG. 16A shows exogenous Cyt C used for maximal respiration. cells were first permeabilized with 0.01% digitonin (DIG) and then 1 mM ADP either alone (ADP) or with 10 M Cyt C (ADP+CC) was added at right arrow. Control group did not receive either ADP or Cyt C. FIG. 16B shows respiratory coupling with ATP synthesis under the assay conditions used in FIG. 16A. Cells were first permeabilized with 0.01% DIG in the presence (Oligo-FCCP) or absence (Control, Oligo, FCCP) of 2 g/ml Oligo and then following additions were made: Control, which is ADP+CC; Oligo, which is control plus Oligo; FCCP, which is control plus 2 M FCCP; Oligo-FCCP, which is Oligo plus 2 M FCCP. FIG. 16C shows data from FIG. 16B with basal rate set at 100% for the comparison of ADP- and FCCP-stimulated respiration rates. FIGS. 16D-16F show results from primary rat astrocytes (FIG. 16D), hamster B2-MWFE cells (FIG. 16E) and human H1080 cells (FIG. 16F) that were first permeabilized with indicated concentrations of digitonin (DIG) and then 1 mM ADP with 10 M Cyt C (ADP+CC) was added.

(17) FIGS. 17A-17F show results of an assay of mitochondrial function with nPFO-permeabilized cells. Cells were prepared as described in Example 8 and the assays were performed in Ca2+-free LKB buffer containing 2 mM glucose (FIGS. 17A-17D) or 15 mM glucose (FIG. 17E-17F). The ADP- and FCCP-stimulated respiration over basal respiration rates (BRR) were indicative of Spare OxPhos capacity (SOC) and spare respiratory capacity (SRC), respectively. FIG. 17A shows the respiratory response of primary pancreatic cells permeabilized with 25 nM nPFO or 0.01% digitonin (DIG). Succinate and ADP were added at the time of permeabilization in the presence of 10 M Cyt C (Succ+ADP). 1 g/ml oligomycin (Oligo) was added to determine the coupling efficiency. FIG. 17B shows results of nPFO titration using INS1E cells. Assay conditions were the same as in FIG. 17A except Cyt C was not added. ADP with varying concentrations of nPFO (@ADP) and 2 M FCCP (@FCCP) were added. FIG. 17C shows SOC compared with SRC in INS1E cells. FIG. 17D shows nPFO concentration compared with SOC in INS1E cells. FIG. 17E shows SOC compared with SRC in V79-G3 cells. FIG. 17F shows nPFO concentration compared with SOC and SRC in V79-G3 cells.

(18) FIGS. 18A-18H show an assessment of mitochondrial integrity and respiratory coupling in nPFO-permeabilized cells. After growing cells as described in Example 8 respiration was measured in Ca2+-free LKB buffer containing 10 mM succinate with 2 mM glucose (FIGS. 18A-18C) or 15 mM glucose (FIGS. 18D-18H) cells were first permeabilized with at least 1 nM nPFO, and then changes in respiration rates were measured following addition of the indicated compounds. FIG. 18A shows mitochondrial integrity in nPFO-permeabilized INS1E cells (ADP indicates 1 mM ADP alone; ADP+CC indicates 1 mM ADP+10 M Cyt C). FIG. 18B shows respiratory coupling in nPFO-permeabilized INS1E cells (ADP indicates 1 mM ADP alone; oligo indicates 1 mM ADP with 1 g/ml oligomycin; ADP+FCCP indicates 1 mM ADP+1 g/ml oligomycin+2 M FCCP). FIG. 18C shows data from FIG. 18B normalized with basal rate set at 100% for comparison of ADP- and FCCP-stimulated respiration rates. FIG. 18D shows integrity in nPFO-permeabilized V79-G3 cells (control indicates buffer only; ADP indicates 1 mM ADP; ADP+CC indicates 1 mM ADP with 10 M Cyt C). FIG. 18E shows respiratory decline in nPFO-permeabilized (nPFO) compared with oligomycin-treated intact (Oligo) V79-G3 cells. FIG. 18F shows oligomycin-insensitive respiration in intact compared with permeabilized V79-G3 cells (Oligo indicates 1 mM ADP followed by oligomycin; nPFO+ADP_Oligo indicates nPFO with 1 mM ADP followed by oligomycin). FIGS. 18G and 18H show mitochondrial integrity in nPFO-permeabilized HEK293 and SHSY-5Y cells, respectively (ADP indicates 1 mM ADP; ADP+CC indicates 1 mM ADP+10 M Cyt C).

(19) FIGS. 19A-19F show estimations of the spare (reserve) and total capacities of OxPhos and ETC/RC. cells were grown as described in Example 8 and assays were measured in Ca2+-free LKB (FIGS. 19A, and 19D-19F) or LPBT (FIGS. 19B and 19C) buffers containing 15 mM glucose. 10 mM succinate was added with 1 mM ADP to measure spare OxPhos capacity (SOC). After measuring the ADP-stimulated respiration FCCP was added to measure spare ETC/RC capacity (SRC). In FIGS. 19A-19C, the succinate and ADP were added with nPFO, while in FIGS. 19D-19F they were added after permeabilization with nPFO. FIG. 19A shows SOC and SRC in intact (SRCi) and permeabilized (SRCp) HEK293 cells. In Control, 2 M FCCP was added to non-permeabilized cells; In nPFO, 3 M FCCP was added to permeabilized cells after measuring succinate+ADP-stimulated respiration. FIG. 19B shows an estimation of total OxPhos (TOC) capacity in HEK293 cells by adding SOC to the oligomycin-sensitive portion of basal respiration rate. Total respiratory capacity is estimated by the maximal respiration rate in nPFO-permeabilized cells (PSA_F indicates nPFO+succinate+ADP followed by FCCP addition; O_PSAF indicates oligomycin followed by nPFO+succinate+ADP+FCCP addition). FIG. 19C shows an coupling efficiency of HEK293 cells determined by monitoring oligomycin-sensitive portion of the ADP-stimulated respiration (PSA_O indicates nPFO+succinate+ADP followed by oligomycin addition; O_PSAF indicates oligomycin followed by nPFO+succinate+ADP+FCCP addition). FIG. 19D-19F shows a strategy for determination of TOC by serial additions of nPFO, ADP, and FCCP in different cells such as HEK293 (FIG. 19D). SHSY-5Y (FIG. 19E) and cells (FIG. 19F).

(20) FIGS. 20A-20I show the effects of different factors on OxPhos capacity. Experimental conditions were as described in Example 8 and with reference to FIG. 19. Respiratory substrates (10 mM each) and ADP (1 mM) were either added simultaneously with nPFO (FIGS. 20A, 20B, 20C, 20G, 20H) or afterwards (FIGS. 20E, 20F). FIG. 20A depicts substrate-dependent variation in OxPhos capacity of primary cells in Ca2+-free LKB buffer containing 2 mM glucose (succ indicates succinate; G+M indicates glutamate+malate; G3-P indicates glycerol-3-phosphate). FIG. 20B shows the effect of free Ca2+ on succinate-supported OxPhos in cells under assay conditions used in the experiments associated with FIG. 20A (control indicates succinate+ADP; CaCl.sub.2 indicates 700 M CaCl.sub.2 added with succinate+ADP). FIG. 20C shows the effect of different respiration buffers on the OxPhos capacity of HEK293 cells. Cells were incubated in different buffers (see Table 6) and then ADP-stimulated respiration was measured. 2 g/ml oligomycin was added to test the coupling efficiency. FIG. 20D shows coupling efficiency (CE) in different buffers. Data from FIG. 20C are plotted. FIG. 20E shows the effect of phosphate (KH.sub.2PO.sub.4) on OxPhos and respiratory capacity. See Table 6 for LKB, HKB, and MAS buffers. KH.sub.2PO.sub.4 is the LKB with 10 mM KH.sub.2PO.sub.4 instead of 0.4 mM. FIG. 20F shows data from FIG. 20E re-plotted after setting basal respiration rate to 100%. FIG. 20G shows, for INS1E cells, spare OxPhos capacity in different buffers, comparing LKB with LPBT, and HKB with HPBT. Basal respiration rate before permeabilization was set at 100%. FIG. 20H shows differences in the basal and ADP-stimulated respiration rates of INS1E cells in high Na+ and high K+ buffers. Actual respiration rates in LPBT and HPBT buffers for FIG. 20G are shown. FIG. 20I shows the effect of different buffers on Complex I- and Complex II-dependent respirations in INS1E cells.

(21) FIGS. 21A-21F show data relating to mitochondrial dysfunction and specific features of mitochondrial metabolism. In FIGS. 21A-21C, Complex I dependence of the glutamate+malate (Glu+Mal)-supported respiration is shown. Assays were performed in free Ca2+LKB buffer containing 15 mM glucose. cells were permeabilized with 1 nM nPFO and ADP-stimulated respiration was measured using the indicated substrates. In FIG. 21A a rotenone sensitivity test of the Glu+Mal-supported respiration is shown. 2 M rotenone (Rot) was used to inhibit Complex I. In FIG. 21B, a lack of Glu+Mal-supported respiration in Complex I-deficient CCL16-B2 cells is shown. FIG. 21C shows complex I activity rescued in B2-MWFE cells (which are CCL16-B2 cells complemented with wild type Chinese hamster Ndufa1 cDNA). FIG. 21D provides data showing that malate effects efficient utilization of NADH generating substrates such as pyruvate (P), isocitrate (I), glutamate (G), and .-ketoglutarate (K (PF, IF, GF and KF indicate, respectively, P, I, G, and K being added separately with 2 M FCCP; PGMIKF indicates P, I, G, and K added together with FCCP). Malate was added last (3rd arrow from left) in all groups except in PGMIKF, which received it with other substrates. FIG. 21E shows a lack of Complex I-dependent respiration on individual substrates (P, G, M, I, K) in primary cells (PA, GA, MA, IA and KA indicate, respectively, P, G, I, M, and K added separately with 1 mM ADP). Succinate-supported respiration (SA) was used as positive control. While the SA group received 2 M FCCP at 3rd injection, other groups received 10 mM succinate. FIG. 21F shows complex I-dependent respiration in rat astrocytes compared with INS1E cells. Glu+Mal supported respiration was monitored in astrocytes and IS1E cells in the presence (Astro_GM+ROT, INS1E_GM+ROT) and absence (Astro_GM, INS1E_GM) of 2 M rotenone.

(22) FIGS. 22A-22B show the effects of oligomycin treatment on substrate supply to cell ETC/RC. INS1E cells were grown and starved as described in the Examples. FIG. 22A shows progressive respiratory decline in INS1E cells in the presence of oligomycin. LKB buffer containing 2 mM glucose in the presence of 1.3 mM CaCl.sub.2 was used for the assays; EGTA was not added. After measuring the basal respiration rates in 2 mM glucose, 14.7 mM glucose or 10 mM pyruvate were added (left arrow) to measure respiratory stimulations. Subsequently, approximately 90 min later, 2 g/ml oligomycin (Oligo) was added. Control group received buffer only. FIG. 22B shows effects of oligomycin on ETC/RC function of INS1E cells. Ca2+-free LKB buffer with 16.7 mM glucose and without added EGTA was used for the assays. cells were incubated for approximately 60 min in 16.7 mM glucose before the assay was performed. All groups were treated with 2 g/ml oligomycin (Oligo) at indicated time point, and then received 10 mM succinate and/or 2 M FCCP with nPFO. Control indicates buffer only; Succ+FCCP indicates succinate+FCCP only (no PFO); nPFO+FCCP_Succ indicates nPFO+FCCP followed by succinate addition; nPFO+Succ+FCCP indicates nPFO+succinate+FCCP added together.

(23) FIGS. 23A-23B show results of FCCP titrations using intact cells. HEK293 cells were seeded at indicated cells densities, and then respiration assay was performed as described in Example 8 using LKB buffer with 1.3 mM CaCl2 and 15 mM glucose without added EGTA.

(24) FIG. 23A shows an actual graph obtained from the XF24 analyzer. Successive additions of 1 M FCCP were made using ports A, B, C, and D resulting in cumulative concentrations of 1, 2, 3, and 4 M FCCP respectively. FIG. 23B shows a linear relationship of the basal and maximal respiration rates with cell density. Data from rates 3 (without FCCP) and 6 (with 2 M FCCP) in FIG. 23A were used.

(25) FIGS. 24A-24B shows the relative performance of certain cell permeabilizing agents compared to digitonin. INS1E cells were grown and assays were performed as described in Example 8. cells were permeabilized with indicated agents and respiration was monitored in the presence of succinate (10 mM), ADP (1 mM) and Cyt c (10 M). FIG. 23A shows relative mitochondrial function in digitonin (0.01% DIG) compared with saponin (5-25 g/ml SAP) permeabilized cells. FIG. 23B shows relative mitochondrial function in digitonin (0.01% DIG) compared with alamethicin (3-30 g/ml ALA) permeabilized cells.

(26) FIGS. 25A-25F shows comparative performances of different PFO variants and the effects of DTT on respiration. Respiratory rates in nPFO- and rPFO-permeabilized HEK293 cells were determined in Ca2+-free LKB (FIG. 25A) and HKB (FIG. 25B) buffers. The rPFO data from panels FIGS. 25A and 25B are shown after setting basal respiration rate to 100% in FIG. 25C. FIG. 25D shows a comparison of nPFO and dbPFO using INS1E cells. Assays were performed in Ca2+-free LKB with 2 mM glucose and cells were permeabilized with 1 nM nPFO or 1 nM dbPFO+1 mM DTT. FIG. 25E shows the effects of 1 mM DTT on respiration of permeabilized INS1E cells. FIG. 25F shows the effect of 1 mM DTT on the respiration of intact INS1E cell.

(27) FIG. 26A shows rPFO-permeabilized INS1E cells. Cells permeabilized with 1 nM rPFO were observed under microscope at the end of experiments in V7 culture plates. Propidium iodide (PI) staining was used to confirm complete permeabilization (bottom), which was evident from cytoplasmic swelling (top). FIG. 26B shows dbPFO-mediated permeabilization of B2-MWFE cells. 100 nM DTT was used to induce the pore formation in 1 nM dbPFO-treated cells.

(28) FIG. 27 shows cholesterol-dependence of Cysteine-free PFO derivatives for membrane binding. The fraction of bound PFO derivatives (0.1 M final concentration) to liposomes of varying cholesterol content and POPC, POPE and SM in a constant 1:1:1 ratio (0.2 mM total lipid final concentration) was determined using intrinsic Trp fluorescence as described in experimental procedures. The cholesterol-dependent binding isotherms for various derivatives: nPFO (PFO) (squares), rPFO (circles), rPFO.sup.D434S (upward triangles) and dbPFO.sup.D434S-C459A (downward triangles) are shown. Data points are the average of at least two measurementsstandard deviation.

(29) FIGS. 28A-28D show the relative performance of Cysteine-free PFO derivatives in mitochondrial function assays. HEK293 cells were permeabilized using nPFO, rPFO and rPFO.sup.D434S (see Table 5) at two different concentrations (0.1, 1.0 nM). Subsequently succinate (Succ: 10 mM) and FCCP (3 mM) were added together to measure maximal respiratory activity. FIG. 28A shows the relative performance at 0.1 nM. FIG. 28B shows the relative performance at 1.0 nM. FIG. 28C shows a comparison of activity of nPFO at both 0.1 and 1.0 nM concentrations. FIG. 28D shows a comparison of activity among PFO variants at both concentrations (0.1 & 1.0 nM). Assays were performed in LPBT buffer (see Table 6).

(30) FIGS. 29A-29F show that digitonin mediated cell permeabilization does not give stable respiration. Cells were permeabilized in the presence of different concentrations of digitonin (% DIG) and ADP stimulated respiration using succinate as substrate was monitored. Assays were performed either in LKB or LPBT buffer (see Table 6). FIGS. 29A and 29B show inverse relationships between digitonin concentration and respiratory activity in LKB buffer. In FIG. 29A, succinate was present in the medium before permeabilization and in FIG. 29B it was added with ADP. Compare arrows ADP and Succ+ADP in both panels. Assay was done with starved INS1E cells. FIGS. 29C and 29D show effects of the respiratory buffer used in non-starved INS1E cells. Respiration declines faster in LKB buffer (FIG. 29C) compared to that in LPBT (FIG. 29D) at all digitonin concentrations tested. FIGS. 29E and 29F show that Cytochrome c loss with digitonin is associated with respiratory decline. Addition of exogenous Cytochrome c (CC) with succinate+ADP inhibits the respiratory decline even at highest digitonin concentration used (0.01%) in both LKB & LPBT buffers.

(31) FIGS. 30A and 30B show effects of low K+(LPBT buffer) compared with high K+(HPBT) buffer on INS1E cell bioenergetics. FIG. 30A shows respiratory decline following the addition of oligomycin (Oligo; 1 g/ml) and response to uncoupler FCCP (2 M). FIG. 30B shows data in FIG. 30A re-plotted with base line set to 100% at rate 4. A faster decline and no response to FCCP is depicted in high K+ buffer (HPBT).

(32) FIGS. 31A-31F show dose-dependent effects of rotenone on overall cellular respiration in intact cells and on Complex I activity in PFO permeabilized cells. FIGS. 31A and 31D show base-line normalized respiration rates of rotenone-treated HEK293 human cells in LKB (FIG. 31A) and LPBT (FIG. 31D) buffers (Rot indicates 0, 10, 20, 50 & 1000 nM rotenone added at arrow; P/GM/A indicates 1 nM rPFO, glutamate+malate (10 mM each), 1 mM ADP & 3 M FCCP; succ indicates 10 mM succinate). FIGS. 31B and 31E show respiration rates and rotenone concentrations plotted from data in FIGS. 31A and 31D, respectively (Basal indicates decline in respiration of intact cells; Glu+Mal indicates Complex I dependent respiration in the presence of FCCP; and succ indicates Complex II-dependent respiration in the presence of FCCP). FIG. 31C shows percent inhibition of Complex I and HI-dependent respirations at different rotenone doses (LKB-CI, CII indicates Complex I and II functions in LKB buffer; and LPBT-CI, II indicates Complex I and II functions in LPBT buffer). FIG. 31F shows a relationship between the basal Complex I inhibition (in intact cells) and maximal Complex I inhibition (in permeabilized cells) in LKB (broken lines) and LPBT buffer (solid lines). At maximal rotenone dose (1 M) no significant difference was observed between basal and maximal as well as between the buffers (see FIGS. 31B and 31E).

(33) FIGS. 32A-32D show reduced respiratory activity in Ndufa1S55A-derived MEFs. FIGS. 32A-32B show differences in the basal and maximal respiration rates of WT and S55A MEFs. Maximal respiration was determined by using the increasing dose of FCCP (2-4 M). FIGS. 32C-32D show respiratory activity in cells that were permeabilized by recombinant perfringolysin O (rPFO). Complex I and II activities were measured by using specific substrates in the presence of ADP+FCCP (Glu+Mal indicates glutamate+malate and succ indicates succinate). 100,000 cells/well were spun down in V7 plates and measurements were made after 4 hours of culture.

DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS OF THE INVENTION

(34) In some embodiments, aspects of the invention relate to simple and reproducible assays for mitochondrial function. In some embodiments, one or more assay components can be provided as kits. Aspects of the invention relate to the surprising finding that cytolysin-based (e.g., perfringolysin O (PFO)-based) cell permeabilization eliminates the need for isolating mitochondria, and maintains the cellular microenvironment around mitochondria. This allows functional assays to be performed at close to physiological conditions.

(35) According to some aspects, the invention relates to cholesterol-dependent cytolysins. As used herein, cholesterol-dependent cytolysins are members of a family of proteins that form pores in lipid-based membranes in a cholesterol sensitive manner. In some embodiments, cholesterol-dependent cytolysins are pore-forming toxins secreted by Gram-positive bacteria. In some embodiments, cholesterol-dependent cytolysins have a characteristic -barrel structure. In some embodiments, cholesterol-dependent cytolysins are monomeric proteins that oligomerize on the membrane surface of target cells. In some embodiments, cholesterol-dependent cytolysins form a ring-like pre-pore complex at the membrane surface of target cells, and insert a large -barrel into the membrane. In some embodiments, the presence of cholesterol in the target membrane is required for pore-formation. In some embodiments, cholesterol-dependent cytolysins selectively permeabilize cellular plasma membranes without damaging mitochondrial membranes. Non-limiting examples of cholesterol-dependent cytolysins are provided in Table 1. Other examples will be apparent to the skilled artisan. In particular embodiments, the cholesterol-dependent cytolysin is a perfringolysin O (PFO).

(36) According to aspects of the invention, the selective permeabilization of cellular membranes by cholesterol-dependent cytolysins (e.g., PFOs) without damaging mitochondrial membranes can be used in assays to measure mitochondrial metabolites. Many mitochondrial metabolites readily cross the mitochondrial membrane to and from the cytosol. However, these metabolites do not readily cross the cellular membrane. This makes it difficult to assay these metabolites without disrupting the cells and in the process disrupting the natural physiological environment of the mitochondria. Surprisingly, cholesterol-dependent cytolysins (e.g., PFOs) have been found to permeabilize cellular membrane sufficiently to allow mitochondrial metabolites to enter and exit the cell. In some embodiments, this allows mitochondrial activity to be evaluated by measuring the extracellular levels of one or more mitochondrial metabolites. According to aspects of the invention, the selectivity of cholesterol-dependent cytolysins (e.g., PFOs) is useful, because it allows the cellular membrane to be permeabilized with respect to mitochondrial metabolites without disrupting the mitochondrial membrane. In some embodiments, this allows the activity of the mitochondria to be evaluated in their natural cellular environment.

(37) According to some aspects of the invention, PFO-based cell permeabilization methods are provided for mitochondrial function assays. The methods are generally applicable across different cell types (e.g., established cell lines and primary cells). Kits that provide reagents for the methods are also provided herein. In some embodiments, these kits are useful for assays of Complexes I, II, III, IV and/or V, Oxidation Phosphorylation (OxPhos) capacity, and/or Respiratory (ETC/RC) capacity. Accordingly, in some embodiments, simple and reproducible assays for mitochondrial function are provided. In some aspects, the assays and kits provide diagnostic tools for mitochondrial dysfunction. In some embodiments, aspects of the invention are useful to understand the role of mitochondrial metabolism in pathophysiology. In some embodiments, aspects of the invention provide diagnostic tools for mitochondrial disorders.

(38) In some embodiments, cholesterol-dependent cytolysin-based (e.g., PFO-based) cell permeabilization eliminates the need for isolating mitochondria, and maintains the cellular microenvironment around them. This permits mitochondrial function assays to be performed at close to physiological conditions using assays techniques that involve measuring the uptake, release, consumption, and/or production of cellular metabolites (e.g., mitochondrial metabolites), for example, using microplate-based respirometry. This also permits examination of the effects of membrane impermeable agents on mitochondrial function.

(39) TABLE-US-00001 TABLE 1 Non-Limiting Examples of Cholesterol-Dependent Cytolysins PLYLUM Firmicutes CLASS Bacilli ORDER Bacillales FAMILY Bacillaceae ID % Identity % Similarity Length Accession # GENUS Bacillus SPECIES B. anthracis ALO 72 (68) 88 (83) 462 (512) ZP_03017964.1 B. thurigiensis TLO 74 (69) 88 (83) 462 (512) YP_037419 B. cereus CLO 74 (69) 88 (84) 462 (512) YP_002369889.1 B. weihenstephanensis WLO 74 (69) 87 (83) 462 (512) ABY46062 Listeriaceae Listeria L. monocytogenes LLO 43 (40) 66 (62) 469 (529) DQ838568.1 L. seeligeri LSO 45 (41) 67 (63) 469 (530) P31830.1 L. ivanovii ILO 46 (43) 66 (62) 469 (528) AAR97343.1 Planococcaceae Lysinibacillus L. sphaericus SPH 76 (72) 90 (87) 463 (506) YP_001699692.1 Paenibacillaceae Paenibacillus P. alvei ALV 75 (71) 87 (84) 462 (501) P23564 Brevibacillus B. brevis BVL 73 (69) 88 (84) 464 (511) YP_002770211.1 Lactobacillales Streptococcaceae Streptococcus S. dysgalactiae* SLOe 67 (56) 83 (70) 463 (571) BAD77791 S. pyogenes SLO 67 (56) 83 (70) 463 (571) NP_268546.1 S. canis SLOc 66 (55) 82 (69) 463 (574) Q53957 S. pseudonemoniae PSY 46 (43) 67 (63) 466 (471) ACI76900 S. pneumonia PLY 46 (43) 67 (64) 466 (471) ABO21366.1 S. mitis MLY 46 (43) 67 (63) 466 (471) ABK58695 S. suis SLY 41 (40) 65 (63) 465 (497) ABE66337.1 S. intermedius ILY 41 (37) 65 (59) 469 (532) B212797.1 S. mitis (Lectinolysin) LLY 39 (29) 62 (47) 463 (665) BAE72438.1 Clostridia Clostridiales Clostridiaceae Clostridium C. perfringens PFO 463 (500) NP_561079 C. butyricum BRY 69 (65) 85 (82) 462 (513) ZP_02950902.1 C. tetani TLY 60 (55) 78 (72) 464 (527) NP_782466.1 C. botulinumB BUYb 60 (49) 78 (63) 464 (602) YP_001886995.1 C. botulinum E3 BLYe 60 (48) 77 (60) 464 (602) YP_001921918.1 C. botulinumC BLYc 60 (56) 79 (74) 463 (518) ZP_02620972.1 C. novyi NVL 58 (54) 78 (73) 463 (514) YP_878174.1 Actinobacteria Actinobacteria Bifidobacteriales Bifidobacteriaceae Gardenella G. vaginallis VLY 40 (39) 65 (60) 466 (516) EU522488.1 Actinomycetales Actinomycetaceae Arcanobacterium A. pyogenes PLO 41 (38) 60 (56) 469 (534) U84782.2

(40) TABLE-US-00002 PerfringolysinO[Clostridiumperfringensstr.13] gi|18309145|ref|NP_561079.1| (SEQIDNO:1) MIRFKKTKLIASIAMALCLFSQPVISFSKDITDKNQSIDSGISSLSYNRN EVTLASNGDKIESFVPKEGKKTGNKFIVVERQKRSLTTSPVDISIIDSVN DRTYPGALQLADKAFVENRPTILMVKRKPININIDLPGLKGENSIKVDDP TYGKVSGAIDELVSKWNEKYSSTHTLPARTQYSESMVYSKSQISSALNVN AKVLENSLGVDFNAVANNEKKVMILAYKQIFYTVSADLPKNPSDLFDDSV TFNDLKQKGVSNEAPPLMVSNVAYGRTIYVKLETTSSSKDVQAAFKALIK NTDIKNSQQYKDIYENSSFTAVVLGGDAQEHNKVVTKDFDEIRKVIKDNA TFSTKNPAYPISYTSVFLKDNSVAAVHNKTDYIETTSTEYSKGKINLDHS GAYVAQFEVAWDEVSYDKEGNEVLTHKTWDGNYQDKTAHYSTVIPLEANA RNIRIKARECTGLAWEWWRDVISEYDVPLTNNINVSIWGTTLYPGSSITY N
Oxidative Phosphorylation (OxPhos)

(41) OxPhos, one of the key functions of mitochondria, is carried out by five multimeric enzyme complexes (I-V, see below) with the help of electron donors (NADH, & FADH.sub.2) and electron carriers (ubiquinone, Cytochrome c). The NADH and FADH.sub.2 feed electrons to the electron transport/respiratory chain (ETC/RC) that establishes an electrochemical gradient (p=.sub.m+pH) across the inner mitochondrial membrane called proton motive force (p). The p is the driving force for ATP synthesis using ADP and Pi. Four enzyme complexes constitute the ETC/RC. Complex I is NADH-ubiquinone oxidoreductase; Complex II is succinate-ubiquinone oxidoreductase; Complex III is ubiquinol-Cytochrome c oxidoreductase; and Complex IV is Cytochrome c-ubiquinone oxidoreductase. Coupled electron transfer with proton translocation across the mitochondrial membrane by Complexes I, III, and IV establishes p, which drives ATP synthesis using the ATP synthase (Complex V) (see FIGS. 1 and 15). Primarily, the ETC/RC function is supported by mitochondrially generated NADH and FADH2 from the oxidative metabolism of pyruvate via the tricarboxylic acid (TCA) cycle. Under certain circumstances, 13-oxidation of fatty acids and amino acids' metabolism also contribute to mitochondrial NADH/FADH.sub.2 pool. Often, the pyruvate oxidative metabolism predominates. Thus, the oxidation of other substrates is tightly regulated based on the physiological context of a given cell. NADH is also generated within cytosol from the reaction of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The metabolic flux through GAPDH is dependent on continuous NADH oxidation to regenerate NAD+ that supports GAPDH function. Within the cytosol the NADH is regenerated by lactate dehydrogenase (LDH) and by NADH redox shuttling using the malate-aspartate (MAS) and glycerol-3-phosphate (GPS) shuttles (see FIG. 1). The relative contribution of LDH and redox shuttles in regenerating NAD+ is a consideration in different cell types and tissues under normal and pathophysiological conditions.

(42) Mitochondrial Dysfunction and Disease:

(43) Impairments in OxPhos are often referred to as mitochondrial dysfunction (and are associated with mitochondrial disorders), and can result from hereditary and somatic mutations in nuclear genes or mtDNA, or functional impairments by drugs or toxins. Mutations in over 100 genes constituting the oxidative phosphorylation machinery are linked with mitochondrial encephalopathies in humans, which are the most common metabolic diseases with an incidence of over 1/5000 in live births. Respiratory chain Complex I deficiency is a cause of mitochondrial diseases in many cases. Twenty five of at least fifty known genes implicated in Complex I biogenesis are found associated with mitochondrial diseases. Pathogenic mutations in structural subunits (e.g., NDUFA1, 2, 11; NDUFS1-4, 6-8; NDUFV1,2) and assembly factors (e.g., NDUFAF1-6) have been identified. Neurodegenerative diseases such as Parkinson's disease, Alzheimer's disease, and Huntington's disease are also associated with mitochondrial dysfunction. Further, mtDNA mutations are found associated with almost all types of cancers. Type 2 diabetes is also linked with declining mitochondrial function in relevant tissues such as -cells and muscles. Type 2 diabetes represents a major clinical challenge due to the sharp rise in obesity-induced disease. Thus, in some embodiments, methods are provided for accurate assessment of mitochondrial function in the context of pathophysiology.

(44) Mitochondrial Function Assays:

(45) Plasma membrane permeabilization removes the permeability barrier of cells and allows accurate estimation of mitochondrial function in intact cells. The selective permeabilization of the plasma membrane has been achieved by exploiting the differential distribution of the cholesterol in membranes. Because most cholesterol is present in the plasma membrane, intracellular membranes are expected to remain largely untouched by cholesterol dependent pore forming agents. Thus, it is possible to selectively permeabilize the plasma membrane in the presence of cholesterol-dependent pore forming agents. Disclosed herein are mitochondrial function assay methods based on a method of cell permeabilization, using a cholesterol dependent pore forming protein (e.g., perfringolysin O (PFO) from Clostridium perfringens). The mechanism of pore formation by PFOs has been characterized.

(46) Applications:

(47) In some embodiments, methods and kits disclosed herein are useful with systems for analyzing extracellular flux (XF) (e.g., commercially available extracellular flux (XF) analyzers, e.g., available from Seahorse Bioscience) to assess mitochondrial function in cells. In some embodiments, users of such analyzers who investigate bioenergetic pathways isolate cells or tissue, adhere them to a culture plate, and perform bioenergetic assessments. These techniques may be used to provide valuable information and insight into mitochondrial biology, under conditions that allow for experimental control of the substrate supply and demand. According to aspects of the invention, whole cells may be selectively permeabilized directly in the plates (e.g., XF plates), allowing for control over substrate supply and demand, access to both oxidative phosphorylation and respiratory chain components, as well as eliminating the need to go through complicated and potentially damaging mitochondrial isolation techniques.

(48) Kits:

(49) In some embodiments, cholesterol-dependent cytolysins and assay-specific reagents are packaged into a kit format. In some embodiments, the kit design may vary with respect to specific assays rather than cells types. In some embodiments, these kits are designed to address specific aspects of mitochondrial metabolism such as the OxPhos, TCA cycle, and cell-specific OxPhos/TCA cycle features. Kits are provided for Complex I-IV assays, OxPhos capacity and ETC/RC capacity assays.

(50) Assays for Complex I, II, and OxPhos & ETC/RC capacities are disclosed herein and corresponding kits are provided in some embodiments. In some embodiments, ADP-stimulated respiration is a measure of the Complex V activity, e.g., when the substrates are not limiting. In some embodiments, the use of succinate with glycerol-3-phosphate together ensures that the ETC/RC activity is not limiting factor, and thus the ADP-stimulated, oligomycin-sensitive respiration is the output of Complex V function. Similarly, in some embodiments, the carboxyatractiloside-sensitive respiration gives the functional output of the ATP/ADP nucleotide translocator (ANT). In some embodiments, the activity of Complex III is assayed using glycerol-3-phosphate+succinate supported, Antimycin A-sensitive respiration. Likewise, the activity of Complex IV is assayed in some embodiments using ascorbate+TMPD supported, KCN-sensitive respiration as described. In some embodiments, a kit has components selected from: PFO, ADP, FCCP and assay buffer.

(51) The actual concentration of reagents will typically vary with the experimental design (e.g., 24-well vs. 96-well assay format) and the number of assays. PFO can typically be used in a range of 1-100 nM. ADP can typically be used in a range of 1-2 mM. And FCCP can typically be used in a range of 2-4 M. The components are typically provided as 100-1000-fold concentrated stocks, which can be used at desired concentrations within the recommended range by the user to get the maximal mitochondrial performance within the assays.

(52) It should be appreciated that a kit may include one or more components (e.g., one or more substrates or inhibitors) for a specific Complex assay (e.g., one of Complex I-V) along with a cholesterol-dependent cytolysin (e.g., PFO), and optionally one or more reducing agents (e.g., DTT or other suitable reducing agent). The different components may be provided in separate containers in a kit. However, in some embodiments, two or more different components may be combined in a single container (e.g., sample tube, well, etc.). Table 2 provides non-limiting examples of components of kits for mitochondrial complexes.

(53) TABLE-US-00003 TABLE 2 Non-Limiting Examples of Components of Kits for Mitochondrial Complexes Complex Substrate Inhibitor Complex glutamate (G); pyruvate (P), malate (M), Rotenone I: isocitrate (I) and alpha-ketoglutarate (K) in different combinations; commonly used are glutamate + malate (GM), or pyruvate + malate (PM); PM; GM, PGMIK; GMIK combinations). Complex Succinate (S) malonate II: Complex Glycerol 3-phosphate Antimycin A III: Complex TMPD + Ascorbate KCN IV: Complex Succinate (with or without glycerol-3- Oligomycin V: phosphate) + ADP
Microplate-Based Respirometry:

(54) In some embodiments, functional assays will be based on a microplate-based system or device, for example using the extracellular flux (XF) analyzer from Seahorse Biosciences. In some embodiments, wild type PFO and/or its variants will be used for selective plasma membrane permeabilization to eliminate the substrate transport barrier, and characteristics of mitochondrial performance (e.g., maximal mitochondrial performance) can be determined. Methods for assaying specific mitochondrial functions that work across different cell types can be implemented. In some embodiments, one or more microplates may be preloaded (and/or provided) with one or more assay components (e.g., substrates, inhibitors, etc.) and one or more cholesterol-dependent cytolysins (e.g., PFO), and optionally one or more reducing agents (e.g., DTT or other suitable reducing agent).

(55) As used herein, the terms approximately or about in reference to a number are generally taken to include numbers that fall within a range of 1%, 5%, 10%, 15%, or 20% in either direction (greater than or less than) of the number unless otherwise stated or otherwise evident from the context (except where such number would be less than 0% or exceed 100% of a possible value).

(56) All references described herein are incorporated by reference for the purposes described herein.

(57) Exemplary embodiments of the invention will be described in more detail by the following examples. These embodiments are exemplary of the invention, which one skilled in the art will recognize is not limited to the exemplary embodiments.

EXAMPLES

Example 1: Experimental Design and Methodology

(58) PFO-Based Cell Permeabilization Methods for Mitochondrial Function Assays:

(59) Wild type PFOs and variants have been tested for maximal mitochondrial performance (see Table 1 for examples of PFOs) using methods disclosed herein. Specific assays for different OxPhos components applicable to a wide variety of were developed based on cell permeabilization methods described herein. Both established and primary cells were utilized to test the general applicability of PFO-based assays.

(60) It has been recognized that a reducing agent may be used to increase shelf life of wild-type PFO. PFO contains only one Cys residue at position 459. To avoid the need of a reducing agent in permeabilization reactions, a Cys free derivative, PFO.sup.C459A, has been used. This variant has activity comparable to wild type PFO. A mutant rPFO.sup.T319C-V334C provides for conditional cell permeabilization, as it does not form pore in the membrane following insertion until a reducing agent, such as DTT is added. After treatment with rPFO.sup.T319C-V334C the cells may be washed to remove the protein before the experiment.

(61) TABLE-US-00004 TABLE 3 PFO variants and their properties: To simplify the mutation notation, all derivatives that include C459A mutation will he called rPFO. PFO derivative Characteristic Pore formation SEQ ID NO: PFO (nPFO) Wild type/native protein Yes 8 rPFO Recombinant, Cys less Yes 9 derivative of PFO (PFO.sup.C459A); cholesterol binding may be reduced slightly rPFO.sup.T319C-V334C Monomeric binding Yes, triggered 11 by DTT

(62) a) Preparation of Functional PFO:

(63) A non-limiting purification scheme that has been used for obtaining recombinant PFO (e.g., wild-type or variant) is illustrated in FIG. 9. Briefly, the scheme involves over-expressing a PFO protein and/or its variants with a His-Tag for affinity purification. In some embodiments, E. coli BL21 (DE3) (Invitrogen) cells expressing PFO and rPFOs conditionally are grown in 2 L cultures at 37 C. with constant agitation. Expression of the PFO/rPFO are induced by the addition of isopropyl -D-thiogalactopyranoside (IPTG, Gold Biochemicals, St. Louis, Mo.) to a final concentration of 1 mM when the turbidity of the culture at 600 nm reaches 0.5-0.6. After 3 h of induction, cells are harvested by centrifugation (4,400 g, 15 min. 4 C.). The cell pellets are suspended in a total of 25 mL of buffer B [10 mM MES (2-(N-morpholino)-ethanesulfonic acid) (pH 6.5), 150 mM NaCl] in the presence of protease inhibitors, PMSF and benzamidine at a concentration of 50 mg/mL. The cells are lysed by one passage through a French pressure cell (1 in. diameter piston, Aminco, Silverspring, Md.) at 15,000 psi.

(64) After clearing the cell lysate by centrifugation at 31,000 g for 15 min. 4 C., the supernatant is loaded onto a column (1.5 cm I.D.10 cm) containing chelating Sepharose Fast Flow (GE Healthcare, Piscataway, N.J.) that had been preloaded with Ca.sup.2+ and equilibrated with buffer B at room temperature. The column is washed with 115 mL of buffer B (2 mL/min), and a linear gradient 0 to 50 mM imidazole pH 6.5 to remove additional contaminating proteins. The bound PFO/rPFO is eluted with 55 mL of buffer B containing 300 mM imidazole.

(65) The pooled fractions containing the bulk of the protein are dialyzed overnight at 4 C. against 4 L of buffer C [10 mM MES (pH 6.5), 1 mM EDTA], and loaded directly onto a SP Sepharose lip (GE Healthcare, Piscataway, N.J.) cation exchange column (1.5 cm I.D.10 cm) equilibrated with buffer C. The column is washed with 60 mL, of buffer B (3 mL/min), and 30 mL of 0.1 M NaCl in buffer C before the elution of the PFO/rPFO with a 100 mL linear gradient (3 mL/min) from 0.1 to 0.9 M NaCl in buffer C. PFO eluted at 0.5 M NaCl and the pooled fractions containing the PFO are dialyzed against buffer A made 10% (v/v) in glycerol, aliquoted into cryovials, quick-frozen in liquid nitrogen, and stored at 80 C. Dithiothreitol (5 mM) is included as a reducing agent when purifying PFO derivatives containing Cys residues. Cys-less derivatives (rPFO) do not need DTT additions.

(66) b) Optimization of the Conditions for Different Mitochondrial Function Assays:

(67) A general methodology for Complex I, II, V, OxPhos capacity, and ETC/RC capacity has been developed. These assays can be implemented with a variety of cell types such as peripheral blood monocytes, myoblasts, neurons, astrocytes, and synaptosomes. In some embodiments, cell-specific modifications may be utilized for certain proteins. The modifications may depend upon the buffer conditions and supplements used. In some embodiments, low K.sup.+ containing buffer was found to be advantageous for mitochondrial functional assays, e.g., in assays using INS1E cells. Similar buffers may be suitable for use with other cell types in certain contexts. However, the relative level of different ions (including K.sup.+) may be optimized for different cell types in some instances.

(68) To evaluate conditions for primary cells in the context of pathophysiology, a mouse model (Ndufa1.sup.S55A) of the partial Complex I deficiency (50%) may be used. As disclosed in Example 11, primary cells (neurons, mouse embryonic fibroblasts, blood monocytes, thymocytes and splenocytes) and synaptosomes may be derived from the Ndufa1.sup.K1 mice to determine the physiological effects of partial Complex I assembly. Partial Complex I deficiencies are the most common cause of mitochondrial diseases in humans. However, other deficiencies also may be studied using these methods.

Example 2: Experimental Evaluations Using Different Cell Types

(69) Similar experimental conditions may be used for assays of Complexes I-V, and OxPhos and ETC/RC capacity and across different cells (see Table 4 for a non-limiting list of cells).

(70) TABLE-US-00005 TABLE 4 Cells for PFO-based mitochondrial function assays. Category Subcategory Cells Cell Adherent cells HEK293, C2C12, HepG2, A549, H460, lines MCF7 Adherent cells INS1, SHSY-5Y with coating Non- Cell seeding, spin down, and measurements adherent on same day with above cells cells Cells may be grouped with respect to their OxPhos capacity vs. ETC/RC capacity to provide a reference for each enlisted cell line. Primary Adherent -cells, astrocytes, Note: primary cells cells cells neurons, fibroblasts, may require coating hepatocytes, mammary of plates (e.g., V7 epithelial cells, plates) with PEI or myocytes/blasts similar reagents e.g. (cardiac, skeletal), PDL, CellTak. Adipocytes (WAT/BAT), adult stem cells Non- Peripheral blood May involve coating adherent monocytes of plates with PEI or cells with (PBMCs)/Lymphocytes similar reagents e.g. coatings (T & B cells) PDT, CellTak Others Synaptosomes Spun down in PIE- coated plates (e.g., V7 plates) Cells may be grouped with respect to their OxPhos capacity vs. respiratory (ETC/RC) capacity to provide a reference for each enlisted cell line. For example excitable cells, such as cells, neurons and muscle cells have comparable OxPhos vs. respiratory (ETC/RC) capacity under low inorganic phosphate (Pi) medium/ buffer. Under the same conditions non excitable cell such as fibroblasts have lower OxPhos capacity compared to respiratory capacity. Increasing the Pi concentration to 10 mM significantly increases the OxPhos capacity in these cells, but it is still relatively lower than respiratory capacity.

(71) The conditions for XF-based respirometry have been evaluated with mammary epithelial cells and adult stem cells. Similar experiments may be performed with other cells. For example, different primary cells, such as mammary epithelial cells, mouse embryonic fibroblasts, can be used. In some embodiments, based on experiments with primary neurons and astrocytes, these primary cells can be derived from the Ndufa1.sup.S55A mice. In some embodiments, it may be advantageous to optimize the concentrations of different assay components for different cell types. For example, it may be advantageous to grow excitable cells such as muscle cells, and neurons in the presence of different amounts K.sup.+, Ca.sup.2+ and/or camitine than other cells. However, assays with -cells, which are also excitable cells, indicate such cells may be evaluated effectively in a wide range of assay conditions. Testing these parameters for a given cell type is possible while using ADP-stimulated and succinate supported respiration as output response for the mitochondrial performance.

Example 3: Assay Using Digitonin as a Permeabilizing Reagent

(72) Digitonin concentration and respiration buffers were optimized for maximal ADP-stimulated respiration using succinate as substrate in the presence and absence of Cytochrome c. FIG. 2 shows representative data for conditions for INS1E cells that involve the addition of exogenous Cytochrome c. A Ca.sup.2+-free low K.sup.+ respiration buffer [20 mM TES pH7.4, 3.5 mM KCL, 120 mM NaCl, 0.4 mM KH.sub.2PO.sub.4, 1.2 mM Na.sub.2SO.sub.4, 2 mM MgSO.sub.4, 1 mM EGTA with 0.4% fatty acid free BSA] was used unless otherwise specified. Since, the intracellular K.sup.+ ion concentration is higher (120 mM), Ca.sup.2+-free high K.sup.+ buffers (25-120 mM NaCl replaced with equimolar KCL in the buffer) were tested to determine if mitochondria performed better. It was found that in certain instances high K.sup.+ buffers underperformed compared to the low K.sup.+ buffer using INS1E cells. Under the same conditions, the performance of two other commonly used permeabilizing agents, saponin, and alamethicin, was evaluated; performance results were compared with digitonin. Although a comparable level of performance to digitonin was achieved using saponin and alamethicin, the assays were not very reproducible. These studies suggested that for each cell type, the digitonin concentration had to be optimized, and this was more variable across different species than within one particular species. Like other detergents, there were reproducibility issues with digitonin, which was believed to be due to the heterogeneity in pore formation, which cannot be effectively controlled with detergents. The requirement of exogenous Cytochrome c with detergent-based assays suggests that mitochondrial integrity is compromised, which results in the loss of Cytochrome c.

Example 4: PFO-Based Assays

(73) A PFO-based assay was developed and found to overcome problems associated with digitonin-based assays. Initially, a wild type PFO was used for selective permeabilization of the plasma membrane. The performance of PFO was compared to digitonin in assays using INS1E cells and Chinese hamster lung fibroblasts, (V79-G3). Surprisingly, PFO outperformed digitonin in both cell types (see FIGS. 3A-3B). The need for exogenous Cytochrome c in PFO-based assays was evaluated. This was accomplished by comparing the ADP-stimulated respiration in the presence and absence of Cytochrome c. The tests showed that even lower concentrations of PFO could be used without the addition of Cytochrome c (FIG. 4).

(74) The lack of exogenous Cytochrome c requirement in assays with PFO indicates that the mitochondrial integrity is better preserved with PFO relative to digitonin. Thus, further studies using PFO were carried out in the absence of Cytochrome c. Although the data shown here are for INS1E cells only, the conditions are applicable to different types of cells.

Example 5: Determination of the Spare Oxidative Phosphorylation (OxPhos) and Respiratory (ETC/RC) Capacities

(75) To compare the OxPhos and ETC/RC capacities of cells, cells were pre-incubated in Ca2+-free respiration buffer with substrates (e.g., 15 mM glucose and 10 mM succinate) and then respiration rates were measured. After measuring basal respiration rates, cells were permeabilized with PFO and then ADP- and FCCP-stimulated respiration rates were measured successively. In this way, the spare capacities of the OxPhos and ETC/RC were determined (FIG. 5A). Without the addition of ADP to PFO permeabilized cells, the respiration declined to a steady-state, and remaining respiration was supported only by the proton (H.sup.+) leak across the mitochondrial inner membrane (FIGS. 5B & C). Addition of ADP stimulated respiration close to the basal respiration level, and the presence or absence of Cytochrome c did not appear to make a substantial difference (FIG. 5C). The OxPhos and ETC/RC capacities are measured by the ADP-stimulated and the protonophore FCCP-stimulated respiration, respectively. Further increase in the respiration over ADP-stimulation by FCCP suggests that the OxPhos capacity is lower compared to ETC/RC capacity (FIG. 5A). This may be a common feature of certain cells such as fibroblasts, which may not face an acute ATP shortage that requires a larger spare OxPhos capacity (FIGS. 5A and 6A). However, in other cells (e.g., -cells) that primarily rely on the mitochondrial OxPhos for their bioenergetic needs, both the OxPhos and ETC/RC capacities must be equal. This point is underscored by a similar level of respiratory stimulation by ADP and FCCP in pancreatic -cells suggesting that their OxPhos and ETC/RC capacities are almost equal (FIG. 6). Taken together these data underscore the usefulness of PFO-based assays for assessing the relative capacities of OxPhos and ETC/RC in a single experiment. This is a very powerful approach as the information gained is physiologically relevant and reproducible. Further, this experimental system also permits the basis of respiration decline in a given cell to be determined following drug treatments that can impair cellular bioenergetics as described herein.

Example 6: Assessment of ETC/RC and Substrate Supply

(76) With the reproducibility of PFO-based cell permeabilization, functional screening of mutants with defects in ETC/RC complexes can be performed. The data in FIGS. 7A-7B illustrate an example of Complex I mutants. Complex I mutants lack respiration on substrates that generate NADH (e.g., glutamate+malate), while they show respiration comparable to control cells on succinate that generates FADH2. As described herein, the NADH & FADH2 feed electrons to Complexes I & II respectively. The activities of other complexes also can be assayed using specific substrates and inhibitors (see description of Complex III & IV assays provided herein).

(77) Apart from the determinations of the functional impairments in ETC/RC, the PFO-based assays can be tailored to determine the limitations of substrate supply to the ETC/RC, which can cause respiratory decline under certain conditions. To address this issue, the respiratory decline caused by oligomycin in cells was evaluated. Data in FIG. 8 show that this oligomycin-induced respiratory decline in cells is due to limitations in substrate supply. Respiration could be restored only in the PFO-permeabilized cells in the presence of both succinate and FCCP. Further, the addition of Ca.sup.2+ dropped the respiration significantly, which is believed to be due at least in part to the opening of the permeability transition pore.

(78) In summary, the data presented demonstrate the feasibility of the PFO-based functional assays of mitochondrial function. These assays can be performed using any suitable technique. In some embodiments, microplate-based respirometry using the XF analyzer from Seahorse Biosciences can be used. The PFO can be used to permeabilize cells in a range of settings to study the mitochondrial metabolism apart from the XF analyzer based studies.

Example 7: Evaluation of Mitochondrial NADH Metabolism in Pancreatic -Cells

(79) FIGS. 10-14 show results from assays described below involving PFO-based permeabilization. In association with the glucokinase, the role of mitochondrial metabolism is widely recognized in glucose sensing. It is thought that the low affinity glucokinase, which is insensitive to feedback inhibition by ATP, increases the flux through glycolysis and enhances pyruvate production. Pyruvate is suggested to be metabolized by TCA cycle to generate electron donors (NADH & FADH2 for respiratory Complexes I & II, respectively) for ATP production, and provide additional signals for sustained release of insulin. Although imaging based studies have suggested that pyruvate metabolism within mitochondria may not be efficient in generating NADH, its implication on mitochondrial function does not appear to have been explored in detail elsewhere.

(80) Using an in situ respirometry assay with permeabilized INS1E and dispersed islet cells, the ability of -cells to generate NADH on different substrates was explored. The data show that -cell mitochondria do not display a detectable level of respiration on NADH-generating substrates, while the mitochondria from lung fibroblasts, astrocytes and neurons display robust respirations under the same experimental conditions. Furthermore, the respiration rates on succinate and -glycerophosphate, the substrates for Complexes II & III respectively, were very robust. These data indicate a unique regulation of NADH metabolism within -cell mitochondria, which may be associated with limited NADH production and/or consumption by enzymes other than Complex I. NADH levels were lower compared to that found in fibroblast and astrocyte mitochondria. It is proposed that -cells regulate NADH output per glucose within mitochondria by negatively regulating key steps to favor the reliance on redox shuttles, and help export of malate and citrate to cytosol for tight coupling of glucose metabolism with insulin secretion.

(81) -cell bioenergetics is dependent on oxidative phosphorylation. It plays a role in insulin secretion by -cells. The reliance of insulin secretion on cytosolic NADH oxidation via redox shuttles suggests that the Respiratory Chain function in -cells is primarily dependent on the cytosolic electron donors. Therefore, whether NADH production within -cell mitochondria is lower compared with other cells (such as astrocytes and fibroblasts) that are not glucose sensitive has been tested. The differences in Complex I-dependent respiration using various NADH generating substrates were monitored. Furthermore, in order to probe the role of oxidative phosphorylation in pyruvate cycling, the respiratory decline in the presence of oligomycin that blocks ATP synthase (Complex V) activity was monitored.

(82) The results of PFO-based assays in Example 7 indicate that permeabilized -cells do not show Complex-I dependent respiration due to limited NADH production within mitochondria. In intact -cells respiration could be supported solely by the redox shuttles. The respiratory and oxidative phosphorylation capacities in -cells are comparable. In these experiments, oxidative phosphorylation in -cells was important to maintain respiratory activity irrespective of the substrate being used. The respiratory decline in the presence of oligomycin observed in this example is associated with substrate limitations, which occurs due to block in pyruvate cycling within -cells.

Example 8: Determination of Spare Oxidative Phosphorylation Capacity and Cell-Specific Features of Mitochondrial Metabolism Using Perfringolysin O and Variants Thereof

(83) Overview:

(84) In this Example, the effectiveness of using detergents for cell permeabilization in mitochondrial function assays was compared with cholesterol-dependent cytolysin (CDC) perfringolysin O (PFO) and certain variants thereof. CDCs are secreted as water soluble monomers of 50-70 kDa that form large ring- and are shaped homooligomeric pores (35-50 monomers/oligomer) in cholesterol containing membranes. The formed pores are approximately 250 in diameter and allow the passage of large molecules (e.g. antibodies, -amylase, and thyroglobulin). Results described below indicate that mitochondrial integrity is better preserved when cells are permeabilized with PFO compared to detergents.

(85) In addition, PFO-based methods have been develop that permit determination of spare OxPhos capacity and other bioenergetic features in small samples using microplate-based respirometry. Using these methods it was found that in bioenergetically demanding cells, such as cells, the spare OxPhos capacity is comparable to the spare respiratory capacity, while in other cells such as fibroblasts, it is lower than the respiratory capacity. These data indicate that in the presence of adequate respiratory substrate such as succinate and ADP, the level of inorganic phosphate (Pi) has a significant effect on the OxPhos capacity.

(86) Material and Methods:

(87) Reagents:

(88) Rotenone was from Calbiochem. Other reagents were from Sigma unless otherwise stated.

(89) Preparation of the Functional PFO:

(90) Native (nPFO), the Cysteine-free rPFO (nPFO containing the C459A mutation), and the engineered disulphide-bond containing mutant dbPFO (rPFO containing the double T319C-V334C mutation) were purified using art known methods. These derivatives contain the polyhistidine tag from the pRSET-B vector (Invitrogen). No significant functional or structural differences were detected among PFO derivatives bearing or lacking the polyhistidine tag.

(91) Both nPFO and rPFO are cytolytically active, but rPFO utilizes higher cholesterol concentrations when tested using model membranes. In contrast, dbPFO binds to membranes but does not form pores because one of the transmembrane -hairpins is covalently linked to domain 2 via a disulphide bond. Reduction of the disulphide bond by the addition of (2S,3S)-1,4-bis(sulfanyl)butane-2,3-diol (DTT) releases the locked transmembrane -hairpin triggering the insertion of a large transmembrane barrel.

(92) Following purification the Cysteine-free rPFO and the dbPFO were stored in buffer A [50 mM HEPES pH 7.5, 100 mM NaCl, and 10% (v/v) glycerol] while the nPFO was stored in buffer A supplemented with 5 mM DTT to retain its cytolytic activity. Proteins were kept at 80 C. until used. The protein concentration was calculated using a molar absorptivity (.sup.208) of 74260 cm.sup.1M.sup.1.

(93) Cells and Culture Conditions:

(94) Rat insulinoma INS1E cells were grown in RPMI1640 medium (Mediatech Inc, Manassas, Va.) which was supplemented with 11.1 mM of glucose, 10% fetal bovine serum (FBS), 1 mM HEPES (Invitrogen), and 50 M of -mercaptoethanol (2-Sulfanylethan-1-ol). Starvation medium contained 4 mM glucose instead of 11.1 mM glucose in the presence of 1 mM Na-pyruvate. Other cells such as Chinese hamster lung fibroblasts (V79-G3, CCL16-B2, CCL16-B2-MWFE), human embryonic kidney (HEK293) cells, and mouse C2C12 myoblasts were grown in Dulbecco's Modified Eagle's Medium (DMEM, Invitrogen, Grand Island, N.Y.) supplemented with 10% FBS (Invitrogen), 1% nonessential amino acids (Mediatech, Inc, Manassas. Va.), and 1% antibiotic mix (PenStrep, Invitrogen) at 37 C. in a humidified atmosphere of 5% CO2/95% air. The human neuroblastoma SHSY5Y cells were cultured in DMEM/F12 media with 10% FBS and 1% antibiotics mix. Cells were harvested after washing once with Ca2+ and Mg2+-free phosphate buffered saline (PBS: pH 7.4) using 0.05% trypsin-EDTA (Invitrogen).

(95) Isolation of Rat Primary Pancreatic -Cells:

(96) Pancreatic islets were isolated from Wistar rats using methods known in the art. Collagenase P enzyme solution (1.2-1.4 mg/ml, Roche Diagnostics Corporation, Indianapolis, Ind.) was injected into the distal end of the donor pancreas. After digestion, islets were gradient purified and then handpicked and cultured in RPMI-1640 medium supplemented with 10% heat-inactivated FBS and 1% PenStrep (Mediatech, Manassas, Va.) in a 5% CO2 incubator. Single cells from islets were prepared using art known methods. After 24-48 hr post culture, islets were collected at 1100 rpm for 5 min at 4 C. followed by washing in PBS twice. Collected islets were treated with 0.05% trypsin supplemented with EDTA (Invitrogen) for 3 min and then triturated gently to separate cells. Trypsin activity was stopped by adding RPMI medium containing serum before triturating. 100,000 cells were seeded per well in polyethyleneimine (PEI)-coated V7 PS XF24 culture plates from Seahorse Bioscience (Billerica, Mass.). Cells were used for experiments 72 hr after seeding unless otherwise noted. In rat islets, -cells are the predominant cell types (80%). Unless otherwise indicated dispersed rat islet cells are referred to herein as primary -cells.

(97) Respirometry and Cell Permeabilization:

(98) Cells grown 80% confluence in V7 tissue culture plates were used for in situ microplate-based respirometry using the XF24 Flux Analyzer (Seahorse Biosciences). All assays were performed with V7 PS plates unless otherwise specified. Cells were seeded at the following densities/well: 10-20,000 (lung fibroblasts-V79-G3, B2-MWFE; HT1080); 25,000 (primary astrocytes); 30,000 (HEK293); 50,000/well (INS1E): and 100,000 (primary cells). Cells were grown for 24-72 hr after seeding, unless otherwise indicated, washed twice with 500 l of the indicated respiration buffer (see Table 6) and then incubated in a non-CO.sub.2 incubator at 37 C. for 30-60 min. XF24 cartridges pre-hydrated for 24 hr were calibrated according to the manufacturer's instructions after loading injection ports with the indicated compounds. After calibration of the sensor cartridge per manufacturer's instructions, the V7 culture plate with cells was loaded into the XF24 analyzer. Respiratory activity of cells was measured using cycles of mixing, waiting and measuring at 0.5-2, 1.5-2 and 3-5 min. respectively, depending upon the cell type used. After 3-4 respiration rate measurements, cells were permeabilized with the indicated reagent (e.g., digitonin, saponin, PFO, rPFO, or dbPFO). The indicated respiratory substrates (e.g. succinate, or glutamate+malate) were either present in the respiration buffer before permeabilization of cells or added with the permeabilizing agent or afterward. Optimum concentrations of ADP (1-2 mM), and carbonylcyanide p-trifluoromethoxy phenylhydrazone (FCCP) were determined by titration assays for different types of cells used. FIGS. 23A-23B show a typical FCCP titration assay using HEK293 cells. Oligomycin was not used in FCCP titration assays. Same FCCP concentrations that gave maximal respiration in intact cells were applied to permeabilized cells. Successive (1st, 2nd, 3rd) additions of compounds were made using injection ports a, b, & c respectively. All respiratory substrates (succinate, malate, glutamate, pyruvate etc.) were used at 10 mM concentration unless otherwise noted. Cyt C was used at 10 M concentration. Actual or normalized (basal respiration set to 100% or 0%) oxygen consumption rates (ORC) are presented as measures of respiration. Increase in the respiratory activity over basal respiration rate by ADP and FCCP were taken as spare (reserve) OxPhos capacity and spare respiratory capacity.

(99) NAD(P)H Assays:

(100) Mitochondria from INS1E and B2-MWFE cells were isolated using art known methods. Mitochondrial content was measured using a microplate-based BCA protein assay kit (Thermo Fisher). 20 g of mitochondria in 250 l Ca2+ free LKB (without glucose) were incubated with glutamate+malate (10 mM each) in the presence of 2 M rotenone for 115 min at 37 C. FLUOstar Omega fluorimeter (BMG Labtech) was used to measure the relative levels of NAD(P)H at 355 nm excitation and 460 nm emission wavelengths. An NADH standard curve was used to determine concentrations of NAD(P)H.

(101) Results:

(102) Mitochondrial Integrity is Compromised in Detergent-Permeabilized Cells:

(103) A rat insulinoma cell line, INS1E, was used to assess permeation conditions for mitochondrial function assays. INS1E cells are an experimental model for rat pancreatic cells. Digitonin has been used for selective plasma membrane permeabilization to assess mitochondria function. A digitonin-based assay was evaluated to determine whether it would be suitable for a microplate based respirometry that employs a limited number of cells (e.g., 5000 to 100,000 cells).

(104) First, an appropriate digitonin concentration was determined for maximal ADP-stimulated respiration using succinate as the substrate. Maximal and sustained respiration was observed with 0.01% digitonin in the presence of 10 M Cyt C (FIG. 16A). In the absence of Cyt C, maximal respiration was not achieved even at lower digitonin concentrations. These data indicate that digitonin impairs outer mitochondrial membrane integrity resulting in Cyt C release. With a different batch of digitonin obtained from the same source decreasing respiration rates were observed with increasing concentrations within 0.0025-0.01% range. The type of respiration buffer used had significant effect on mitochondrial function when cells were permeabilized with digitonin (FIG. 29). Thus even with careful titrations, a digitonin concentration that would work reproducibly in different assay conditions for a given cell type was not identified.

(105) To test the coupling of respiratory activity with ATP synthesis, the extent to which ADP-stimulated respiration is sensitive to oligomycin was evaluated. Oligomycin is an inhibitor of the ATP synthase (Complex V) that uses p to synthesize ATP from ADP and Pi (see FIG. 15). The majority of the ADP stimulated respiration (75%) was oligomycin sensitive when oligomycin was added along with ADP. The remaining respiration is expected to be supported by the H+ leak across inner mitochondrial membrane.

(106) To determine the respiratory capacity, FCCP was used. FCCP, which is a protonophore, induces maximal respiration by dissipating the H+ gradient across the mitochondrial inner membrane. Thus, a suitable FCCP concentration as determined by titration was used to determine the respiratory capacity (maximal respiration) of INS1E cells. Both ADP and FCCP gave comparable respiratory stimulations indicating that spare OxPhos capacity and respiratory capacity, respectively, were comparable in INS1E cells (FIG. 16B, 16C). Significant differences were not observed between the spare OxPhos and respiratory capacities using succinate as the respiratory substrate. Furthermore, FCCP-stimulated respiration in the presence of oligomycin was not significantly different from that observed without oligomycin, which indicates that respiratory capacity was primarily determined by the functional capacity of ETC/RC.

(107) To assess the extent to which the same experimental conditions can be employed to assay mitochondrial function in other cells, ADP-stimulated, succinate-supported respiration was evaluated in various cell types, including primary rat astrocytes, Chinese hamster fibroblasts and human cells. Similar conditions were applicable to rat astrocytes (FIG. 16D), and digitonin titrations were utilized for hamster (B2-MWFE) and humans (HT1080) cells (FIG. 16E, F).

(108) The relative performances of other agents such as saponin (another detergent), and alamethicin (a pore forming peptide) were compared to digitonin. Both saponin and alamethicin have been used in assays of mitochondrial function in different experimental settings. Results with saponin, but not alamethicin, were comparable to digitonin in INS1E cells (FIG. 24A, 24B). Compared to digitonin results with saponin were less reproducible.

(109) Mitochondrial Integrity Remains Intact in PFO-Permeabilized Cells:

(110) Cholesterol-dependent cytolysins (CDCs) were used to preserve mitochondrial integrity following cell permeabilization. CDCs (e.g., perfringolysin O (PFO)) bind to membranes containing relatively high amounts of cholesterol. Pores formed by the insertion of transmembrane -barrels of a CDC were found to be better controlled and more homogenous than those obtained with detergents. Intracellular organelles therefore remain structurally and functionally intact in the presence CDCs.

(111) Several perfringolysin O (PFO) derivatives were developed for permeabilization assays. It was found that the respiratory response was more robust in P cells permeabilized with nPFO compared with digitonin (FIG. 17A). Such an improvement in mitochondrial function with nPFO was also observed in Chinese hamster lung fibroblasts as well (V79-G3). nPFO dose-response analyses were performed to determine a minimal concentration required for maximal mitochondrial function. Spare OxPhos and respiratory capacities of INS1E and V79-G3 cells were measured at different doses. It was found that, in some embodiments, 1 nM PFO was adequate for maximal mitochondrial performance (FIGS. 17B-D). A further increase in respiration was not observed by FCCP addition beyond ADP-stimulation at all nPFO doses tested (FIG. 17B).

(112) Using an nPFO-based assay it was found that OxPhos capacity of INS1E cells was comparable to respiratory capacity (FIGS. 16B, C and FIGS. 17B, C). Unlike in INS1E cells, the spare OxPhos capacity in V79-G3 cells was lower compared to respiratory capacity (FIGS. 17E, F). There was no change in spare OxPhos and respiratory capacities against a wide range of nPFO concentration in V79-G3 cells. Even for Chinese hamster V79-G3 cells as little as 1 nM nPFO was sufficient for maximal mitochondrial performance (FIG. 17F).

(113) In contrast to digitonin based-assays, exogenous Cyt C was not required when mitochondrial function was assayed in nPFO-permeabilized cells. There were no differences observed in mitochondrial performance in the absence or presence of added Cyt C across different cell types (FIGS. 18A, D, G, H). In cells respiratory activity is generally proportional to external glucose concentration, while in other cell types it is not. When glucose is not limiting, cell permeabilization also results in respiratory decline (FIG. 19F). In the presence of succinate alone, respiration declined to a steady-state level in permeabilized V79-G3 cells (FIG. 18D). The remaining respiration was supported by H+ leak as it was comparable to oligomycin insensitive respiration in intact cells (FIG. 18E). Addition of ADP alone restored the respiration close to the basal level (pre-permeabilization state, FIG. 18D). When ADP was added simultaneously with nPFO, there was no respiratory decline, and instead there was 20% increase in respiration over the basal level (FIGS. 17E, 18F).

(114) ADP-stimulated respiration was sensitive to oligomycin, indicating respiratory coupling with ATP synthesis (FIG. 18F). ADP alone result in respiratory stimulation indicating that plasma membrane was impermeable to ADP. The respiratory decline in the presence of oligomycin overlapped with that of nPFO-permeabilized cells (FIG. 18E). These data indicate that ETC/RC function was coupled with ATP synthesis in both INS1E and V79-G3 cells and mitochondrial integrity was preserved.

(115) Determination of the Spare OxPhos Capacity Using PFO-Based Assays:

(116) Respiratory stimulation with appropriate amounts of ADP in the presence of saturating substrate concentration at the time of permeabilization permits measurement of spare OxPhos capacity. Data in FIG. 19A show a typical experiment for determining spare OxPhos and respiratory capacities. Human HEK293 cells were used for these assays. cells were incubated in Ca2+-free respiration buffer containing 15 mM glucose, and then succinate and ADP were added with nPFO simultaneously. The ADP-stimulated respiration over the basal respiration gives the estimate of spare OxPhos capacity. To determine whether spare OxPhos capacity (SOC) matched the spare respiratory capacity (SRC), FCCP was added to the same cells. By monitoring the ADP- and FCCP-stimulated respiration within the same cells, the fraction of spare respiratory capacity supporting ATP synthesis (SOC/SRC) was determined. Comparison of the spare respiratory capacity of intact and PFO permeabilized cells (SRCi vs. SRCp) indicated that SRC could be underestimated by 60% in intact cells using glucose as the respiratory substrate in some cases (FIG. 19A).

(117) FIG. 19B show data from which an estimation of total OxPhos capacity may be made. It shows that by adding spare OxPhos capacity (SOC) to the oligomycin-sensitive respiration (OSR) total OxPhos capacity (TOC) of a given cell type can be determined under a given experimental condition. Since the oligomycin effect is relatively instantaneous, the sum of spare OxPhos capacity (SOC) and oligomycin-sensitive respiration (OSR) in intact cells provides an accurate estimate of total OxPhos capacity (TOC=SOC+OSR). Further coupling efficiency (CE) can be determined by monitoring oligomycin-sensitivity of the ADP-stimulated respiration (FIG. 19C). The presence of oligomycin did not appear to affect spare and total respiratory capacities (FIG. 19B).

(118) In the presence of substrate respiratory decline in PFO-permeabilized cells is associated with ADP leaking from the cytoplasm, (FIG. 18D, G, H). An alternative assay design was used to determine total OxPhos capacity (FIG. 19D). This permits an analysis of the relationship between OxPhos and respiratory capacities on a given substrate. Experimental conditions, such as the time taken to achieve steady-state respiration rate following PFO-permeabilization may be optimized for a given cell type.

(119) Assessment of OxPhos Capacity Using PFO-Based Assays:

(120) OxPhos capacity can be influenced by certain factors, including, for example, substrate(s) and experimental conditions. Having determined that the PFO-based assay was robust and reproducible, factors that could influence OxPhos capacity were assessed. Spare OxPhos capacity was found to vary with different substrates such as glutamate+malate, succinate, and glycerol-3-phosphate, which support ETC/RC function at Complex I, II and III respectively (FIG. 20A). OxPhos was significantly increased in the presence of free Ca2+ (220 nM) even with succinate, which oxidation is considered insensitive to Ca2+ (FIG. 20B). Experiments were performed using LKB buffer to facilitate a comparison of mitochondrial bioenergetics in intact and permeabilized cells under the same conditions.

(121) Considering that intracellular K+ concentration is higher in cytoplasm of cells, and PFO-based method is robust and consistent, the extent to which OxPhos capacity was affected by the choice of the respiration medium was evaluated. Table 6 shows a list of respiration buffers that were compared based on effects on OxPhos capacity of HEK293 cells. HEK293 cells were used because they show lower OxPhos capacity than respiratory capacity (FIG. 19A, B). ADP-stimulated respiration was minimal in the LKB buffer. The maximal respiration was observed in MAS buffer (FIG. 20C). Compared to LKB and HKB buffers, the coupling efficiency was significantly higher in the MAS buffer (p<0.05, FIG. 20D).

(122) Further analyses indicated that the concentration of KH.sub.2PO.sub.4 was a factor that influenced OxPhos capacity (FIG. 20E). Other components of the respiration medium such as BSA (0.2% and 0.4%) and the concentration of buffering agent (e.g. TES or HEPES) had minimal effects on the FCCP-stimulated respiration and did not produce significant effects on ADP-stimulated respiration. When the concentration of KH.sub.2PO.sub.4 was increased from 0.4 mM to 10 mM, the spare OxPhos capacity was comparable to that observed in MAS buffer (FIG. 20E). Under these conditions, the OxPhos capacity was lower than the respiratory capacity. These data indicate that, apart from substrate and ADP, Pi availability can limit OxPhos capacity.

(123) The choice of Na+vs. K+ buffer may have relatively minor effects on the determination of spare Phos/respiratory capacities in non-excitable cells such as HEK293 (FIG. 20C, E, F) compared to excitable cells such as INS1E (FIG. 20G, H). In INS1E cells, while the spare OxPhos capacity was significantly higher with succinate, a Complex II substrate, significant difference were not observed on Complex I substrates glutamate+malate (FIG. 20I).

(124) Assessment of ETC/RC Using PFO-Based Assays:

(125) To demonstrate that PFO-based assays can identify specific defects in ETC/RC assembly, Complex I-deficient CCL16-B2 cells were used. The CCL16-B2 cells are impaired in Complex I assembly due to the absence of MWFE subunit encoded by the Ndufa1 gene. First, rotenone sensitivity of glutamate+malate supported respiration was tested in respiration competent V79-G3 cells as shown in FIG. 21A. Glutamate+malate-supported respiration was sensitive to rotenone, a specific inhibitor of Complex I. No glutamate+malate supported respiration was detected in CCL16-B2 cells, while robust respiration was observed using succinate (FIG. 21B). In contrast, B2-MWFE cells, which are CCL16-B2 cells complemented with wild type MWFE protein, showed both glutamate+malate- and succinate-supported respirations under the same conditions (FIG. 21C). These data indicate that, using the PFO-based assays disclosed herein, specific defects in ETC/RC function can be identified using specific substrate and inhibitor combinations from a limited amount of sample, without isolating mitochondria.

(126) Assessing TCA Cycle Metabolism by Respirometry Using PFO-Based Assays:

(127) Since the TCA cycle is the main source of NADH within mitochondria, Complex I-dependent respiration can also provide information about the functional status of individual NADH generating steps, and the factors affecting transport of metabolites into mitochondria. Malate is a TCA cycle metabolite that regulates transport of citrate/isocitrate, -ketoglutarate and Pi across the mitochondrial inner membrane.

(128) To determine whether and to what extent malate would affect the utilization of pyruvate, isocitrate and -ketoglutarate to support respiratory activity, respiration rates were measured first in the absence of malate and then in its presence within same assay. This was done in B2-MWFE cells that showed Complex I-dependent respiration (FIG. 21C). Maximal respiratory response with isocitrate and -ketoglutarate was achieved after addition of malate (FIG. 21D). A similar response was also observed for pyruvate and glutamate, both of which showed even lower respiratory response in the absence of malate. This was unexpected as transport of pyruvate and glutamate is not expected to be directly affected by the presence of malate.

(129) Individually, pyruvate, glutamate, malate, isocitrate and .-ketoglutarate were not observed to support respiration in primary -cells, while succinate supported respiration was normal in the same cells (FIG. 21E). These data indicate that malate supports maximal respiratory response with NADH generating substrates such as pyruvate, glutamate, isocitrate, and -ketoglutarate.

(130) Under conditions that showed Complex I-dependent respiration in primary rat astrocytes, no significant respiration in INS1E was observed (FIG. 21F). This may be due at least in part to -cell specific negative regulation of NADH metabolism within mitochondria. Such a regulation would permit cells to rely mostly on cytosolic NADH for their bioenergetic needs in insulin secretion.

(131) When mitochondria were treated with glutamate+malate for 15 min at 37 C. in the presence of 2 M rotenone to inhibit NADH oxidation, the NAD(P)H level was lower 40% (4.050.65 M compared with 6.830.11 M, p<0.002 by students t test) in INS1E mitochondria compared to B2-MWFE mitochondria. B2-MWFE cells, whose mitochondria were used as positive controls, show Complex I dependent respiration (FIG. 21C).

(132) Assessment of OxPhos System:

(133) PFO-based assays were used to distinguish the effect of substrate limitations from the direct effects on OxPhos system. INS1E cells, a -cell model, showed robust glucose-stimulated respiration (FIG. 22A) and progressive respiratory decline when treated with oligomycin (FIG. 22A). The respiratory decline in the presence of oligomycin was observed in both low and high glucose-treated cells (FIG. 22A). These data indicate that in low glucose, respiration is supported by the H+ leak across the inner mitochondrial membrane; whereas in high glucose substantially all glucose-induced respiration supports ATP synthesis (FIG. 22A). In both cases oligomycin-insensitive respiration declined progressively with time. A faster decline in low glucose medium indicated that substrate supply to ETC/RC could become limiting over time. A similar effect was observed in the presence of pyruvate suggesting that the effect was mostly downstream of glycolysis, and was related to pyruvate metabolism (FIG. 22A).

(134) To assess limitations in substrate supply to ETC/RC, respiration rescue experiments were performed in the same cells after permeabilization with nPFO. First INS1E cells were treated with oligomycin in LKB respiration medium (without CaCl.sub.2, no EGTA added). Approximately 30 min after oligomycin addition, respiration started to decline gradually (FIG. 22B). When intact cells were treated with FCCP there was no observed increase in respiration. The FCCP could restore respiration only in permeabilized cells when a substrate (succinate) was added. These data indicate that substrate limitation effects a respiratory decline in INS1E cells when OxPhos is completely blocked. This is consistent with the notion that shuttling between mitochondria and cytoplasm of metabolites that influence cell physiology will be halted in the absence of continuous ATP synthesis. This may result in substrate limitation to the ETC/RC where glycerol-3-phosphate generation by metabolic cycling may be important for -cell bioenergetics. These experiments illustrate that the PFO-based assays disclosed herein may be used to evaluate direct and indirect effects on ETC/RC function.

Example 9: Analysis of PFO Variants

(135) The cholesterol dependent binding of different PFO variants was tested using artificial membranes (liposomes), as shown in FIG. 27. rPFO binding utilized 4 mol % more cholesterol than the native PFO (nPFO). Introduction of the D434S mutation into the rPFO derivative shifted the membrane binding properties of the mutant protein such that it could be used in the context of similar cholesterol levels as for nPFO. Similarly, when the same mutation was introduced into a PFO derivative containing a disulfide bond (i.e., rPFO.sup.T319C-V334C) the binding properties of resulting protein dbPFO.sup.D434S-C459A were similar to the native PFO. We have therefore restored the cholesterol binding properties of the native toxin in the rPFO and its disulfide bond containing derivative rPFO.sup.T319C-V334C. These mutants may be used in the absence of reducing agents such as DTT in binding assays to avoid Cys oxidation or disulfide-bond formation between proteins. When using the rPFO.sup.T319C-V334C or dbPFO.sup.D434S-C459A mutants, the reducing agent may be added after the protein is bound to the target membranes to induce pore formation. Furthermore, the unbound protein may be removed before triggering pore formation using the rPFO.sup.T319C-V334C or dbPFO.sup.D434S-C459A mutants. The fraction of bound PFO derivatives (0.1 M final concentration) to liposomes of varying cholesterol content and POPC, POPE and SM in a constant 1:1:1 ratio (0.2 mM total lipid final concentration) was determined using intrinsic Trp fluorescence as described in experimental procedures. The FIG. 27 shows the cholesterol dependent binding isotherms for various derivatives: native PFO (squares), PFO.sub.C459A (circles), PFO.sup.D434S-C459A (upward triangles) and PFO.sup.T319C-V334C-D434S-C459A (or dbPFO.sup.D434S-C459A, downward triangles). In FIG. 27, data points are the average of at least two measurements and standard deviations.

(136) TABLE-US-00006 TABLE 5 PFO variants and their properties: Pore PEO derivative Comments formation SEQ ID NO PFO (nPFO) Native or wild type protein, Yes 8 utilizes DTT for maintaining activity and storage rPFO Recombinant, Cys less Yes 9 (Cysteine Free) derivative of PFO (PFO.sup.C459A); it does not require DTT; cholesterol sensitivity is reduced slightly compared to PFO rPFO.sup.D434S rPFO with D434S mutation Yes 10 which restores cholesterol sensitivity comparable to that in wild type PFO; it does not require DTT rPFO.sup.T319C-V334C Disulfide bond introduced in Yes, triggered 11 rPFO by double mutation by DTT T319C-V334C; monomeric binding and gets inserted in the membrane; also referred as dbPFO; cholesterol sensitivity lower than PFO dbPFO.sup.D434S-C459A rPFO.sup.T319C-V334C with D434S Yes, triggered 12 (may be referred to as be mutation; cholesterol sensitivity by DTT rPFO.sup.T319C-V334C-D434S) comparable to PFO Note: The amino acid numbering of the substitutions set forth in Tables 3 and 5 are based on the sequence set forth in SEQ ID NO: 1 or 2. Analogous substitutions may be made in PFOs having different sequence lengths. For example, in some embodiments, a PFO may have a truncated N-terminus (e.g., a PFO without an N-terminal signal sequence, e.g., without the first 28 amino acids of SEQ ID NO: 1 or 2) or an extended N-terminus (e.g., a PFO having an N-terminal peptide tag). In such embodiments, the same amino acids may be substituted although the relative position of those amino acids from the N-terminus may be different than in the context of the sequence set forth in SEQ ID NO: 1 or 2.

(137) Preparation of PFO Derivatives.

(138) PFO derivatives were express and purified using appropriate methods known in the art for protein expression and purification. The PFO derivative containing a native PFO sequence (amino acids 29-500 of SEQ ID NO: 2) plus the polyhistidine tag that came from the pRSETB vector (Invitrogen) is named nPFO. The PFO Cys-less derivative (nPFO.sup.C459A, where Cysteine 459 is replaced by Alanine) is named rPFO. Mutagenesis of PFO was done using the QuickChange (Stratagene) procedure as described previously. Table 5 provides additional information regarding PFO derivatives.

(139) Assay for Binding.

(140) Binding to liposomes was performed using the change in the Trp emission intensity produced by the binding of PFO to cholesterol containing membranes as described previously. Briefly, emission for Trp fluorescence was recorded at 348 nm (4 nm bandpass) with the excitation wavelength fixed at 295 nm (2 nm bandpass). The signal of monomeric PFO derivatives were obtained with samples containing 200 nM protein in buffer A (HEPES 50 mM, NaCl 100 mM, DTT 1 mM. EDTA 0.5 mM, pH 7.5) using 4 mm4 mm quartz cuvettes. The net emission intensity (F.sub.0) for monomers was obtained after subtracting the signal of the sample before the protein was added. Liposomes were added (200 M total lipids) and the samples were incubated 20 min at 37 C. Trp emission after membrane incubation was measured after re-equilibration of the sample at 25 C., and the signal from an equivalent sample lacking the protein was subtracted (F). Fraction of protein bound was determined as (FF.sub.0)/(F.sub.fF.sub.0), where F.sub.r is the emission intensity when all the protein is bound. Binding of PFO derivatives to cholesterol dispersions in aqueous solutions was done as describe previously.

(141) Preparation of Lipids and Liposomes.

(142) Nonsterol lipids were obtained from Avanti Polar Lipids (Alabaster, Ala.), and cholesterol was from Steraloids (Newport, R.I.). Large unilamellar vesicles were generated as described previously. Briefly, equimolar mixtures of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), and sphingomyelin (SM, porcine brain), were combined with the indicated amount of cholesterol (5-cholesten-3-ol) in chloroform. The thin film of lipids formed after chloroform evaporation was resuspended in buffer A and passed through an extruder equipped with 0.1 m filter 21 times. Liposomes were stored in ice and discarded after three weeks.

Example 10: PFO Derivatives Show Comparable or Better Performance to Native PFO in Mitochondrial Function Assays

(143) Several cysteine-free derivatives of PFO were produced and evaluated. Like native PFO, these derivatives can permeabilize mammalian cells. In some embodiments, it was found that the derivatives can permeabilize cells as efficiently as the native PFO. The cholesterol sensitivity of these derivatives was assessed using liposomes as artificial membranes. Side-by side comparisons of cysteine-free PFO derivatives show comparable or relatively better performance than native PFO.

(144) FIGS. 28A-28D show the relative performance of nPFO, rPFO and rPFO.sup.D434S in mitochondrial function assays using human HEK293 cells, rPFO and rPFO.sup.D434S were comparable, and maintained their activity in the absence of added reducing agents. A visual inspection under microscope indicated that the overall cellular morphology following permeabilization with rPFO.sup.D434S was better preserved compared to other derivatives under the conditions tested. Swelling of cytoplasm/cell ghosts was relatively low after permeabilization with rPFO.sup.D434S. In some embodiments, therefore, rPFO.sup.D434S is advantageous for general use.

(145) In some embodiments, rPFO is advantageous for permeabilization of cells because it has a relatively lower cholesterol sensitivity than other PFO derivatives. For example, cholesterol content in certain mammalian cells may rise under certain pathophysiological conditions (e.g., metabolic syndrome), in such cases cholesterol can accumulate in mitochondria, which may compromise mitochondrial function if the PFO is left in the assay medium. Because rPFO has a relatively low cholesterol sensitivity it may be advantageous for cell permeablization of such cells. Derivatives of PFO that are conditionally active have been developed. In some embodiments, the derivatives of PFO (dbPFO.sup.D434S-C459A and rPFO.sup.T319C-V334C) are triggered to form pores in plasma membranes by 50 nM DTT. The use of DTT at concentrations of approximately 50 nM or less does not significantly interfere with cell function when used with these derivatives. These derivatives are useful in cells having high levels of cholesterol. In such cases, removing excess PFO before permeabilization can preserve mitochondrial function after permeabilization. In some embodiments, rPFO.sup.T319C-V334C and dbPFO.sup.D434S-C459 are useful for cell permeabilization because they can be removed before inducing pore formation with 50 nM DT. These are concentrations of DTT that are not expected to affect mitochondrial function significantly. In some embodiments, dbPFO.sup.D434S-C459A is functionally comparable rPFO.sup.T319C-V334C.

Example 11: Comparison of PFO Derivatives with Digitonin

(146) Assays for assessing mitochondrial function that utilize digitonin for cell permeabilization are, in some embodiments, limited because (i) observed effects of digitonin on cell permeabilization are concentration dependent, (ii) observed effects of digitonin are influenced by buffer composition, (iii) stable respiration is often not obtained even after careful titration of digitonin, and (iv) observed effects of digitonin have low reproducibility due to limited dynamic range of the digitonin concentration (see FIGS. 29A-29F). In some embodiments, relatively high concentrations of digitonin can damage mitochondrial membranes and release Cytochrome c which is important for respiratory chain function.

(147) As disclosed herein. PFO derivatives overcome these limitations of digitonin. PFO is useful as a permeabilizing agent, in part, because it can be handled with high precision facilitating reliable and reproducible assays. In comparison with digitonin, PFO derivatives produce reproducible results within a concentration range of 0.1-20 nM without significant difference in the experimental output (compare FIGS. 28A-28D and FIGS. 29A-29F). In addition, respiratory activity of cells permeabilized with PFO (in contrast with digitonin) is stable in all the buffers tested (See Example 8).

(148) Choice of Buffers Affects Aspects of Mitochondrial Bioenergetics in Intact and Permeabilized Cells:

(149) Respiratory buffers differing in K+ and inorganic phosphate contents (Pi) (see Table 6) were used to evaluate their impact on mitochondrial function. FIG. 20E-20F shows that buffer choice can affect the oxidative phosphorylation (OxPhos) capacity of cells, which was maximal in MAS buffer. By changing the phosphate (Pi) concentration in LKB from 0.4 mM to 10 mM, the OxPhos capacity in LKB could be increased to the level observed in MAS buffer (FIG. 20E, 20F). Given that different cells (INS1E, and others) look morphologically healthy in low K+ buffers (LKB, LPBT; see Table 6), these buffers have been used in many assays. An example of data showing relatively faster respiratory decline in INS1E cells in the presence of oligomycin (an ATP synthase inhibitor) in HPBT (high K+ buffer) is shown in FIG. 30.

(150) Assessing Respiratory Inhibition in Intact and Permeabilized Cells Simultaneously:

(151) To determine how a drug-induced respiratory inhibition in intact cells translates in to maximal inhibition of a respiratory chain complex, rotenone was used as an example for inhibiting Complex I. FIG. 31A shows the dose-dependent inhibition of the cellular respiration. While in low phosphate LKB buffer (0.4 mM Pi), the ADP-stimulated respiration was not apparent (FIG. 31A), ADP-stimulated respiration was observed in high phosphate LBPT buffer (10 mM Pi). ADP-stimulated respiration was inhibited by rotenone dose-dependently (FIG. 31D). Complex I inhibition was measured in the presence of 3 M FCCP using the respiration supported by glutamate+malate (FIG. 31A, 31D). It was recognized that with increasing Complex I inhibition by rotenone, Complex II-dependent respiration was also significantly inhibited in LBPT buffer (FIG. 31D, 31E) and was minimal in LKB buffer (FIG. 31A, 31B).

(152) This observation could be due to the build-up of oxaloacetate, which is a physiological inhibitor of Complex II. In the presence of significant Complex I inhibition by the rotenone oxaloacetate may not be converted into citrate due to NADH build up. When Complex II-dependent respiration was measured in rotenone-treated cells in the absence of glutamate+malate, there was no significant Complex II inhibition detected. Thus, Complex II function may be measured in the absence of glutamate+malate when Complex I is significantly inhibited. A higher level of phosphate can also result in succinate exit from mitochondria in exchange for its entry. Thus, maintaining a succinate/phosphate ration >1 for Complex II assay in the presence of Complex I inhibition may be desirable to avoid limitations in succinate supply to Complex II.

(153) Mouse embryonic fibroblasts (MEFs) derived from a Complex I deficient mouse model (Ndufa1S55A) were used to assess the extent to which genetically encoded partial respiratory chain deficiencies could be detected using PFO-based assays. Intact MEFs from the mutant mice showed reduced cellular respiration on glucose+ pyruvate or pyruvate alone (FIG. 32A, 32B). Observed differences in pyruvate containing respiration buffer confirmed the deficiency in oxidative metabolism, and indicated lack of impaired glycolysis.

(154) In rPFO permeabilized MEFs, there was 50% reduced Complex I function, while Complex II function was normal. To minimize confounding effects of glutamate+malate in the presence of Complex I deficiency, Complex II function was measured in parallel separately. In summary, these assays provide a framework that can be used for detecting partial respiratory chain deficiencies in human patients with different pathological conditions including mitochondrial diseases.

(155) TABLE-US-00007 TABLE 6 Compositions of the respiration buffers. The Ca2+-free buffers contained no added CaCl.sub.2, whereas the regular LKB contained 1.3 mM CaCl2 and no added EGTA. Glucose was added at indicated concentrations in the respiration buffers as described in the text and/or figure legends for specific assays. Component LKB.sup.1 HKB.sup.2 KHE.sup.3 MAS.sup.4 LPBT.sup.5 HPBT.sup.6 KCl (mM) 3.5 120 115 3.5 120 NaCl (mM) 120 3.5 120 3.5 KH.sub.2PO.sub.4 0.4 0.4 10 10 10 10 (mM) Na.sub.2SO.sub.4 (mM) 1.2 1.2 1.2 1.2 MgCl.sub.2 (mM) 2 2 2 5 2 2 EGTA (mM) 1 1 1 1 1 1 TES (mM).sup.a 20 20 3 3 HEPES 3 2 (mM).sup.b Mannitol 220 (mM) Sucrose (mM) 70 BSA (%)c 0.4 0.4 0.2 0.2 0.2 0.2 pH 7.4 7.4 7.2 7.2 7.4 7.4 .sup.aNaN-Tris-(hydroxymethyl)-methyl-2-amino-ethanesulphonic acid; .sup.b4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; cBovine serum albumin (fatty-acid free) .sup.1Low K+ buffer; .sup.2High K+ buffer; .sup.3K+-HEPES-EGTA buffer; .sup.4Mannitol and sucrose buffer; .sup.5,.sup.6Phosphate, BSA and TES concentrations changed in LKB and HKB respectively.

(156) Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only and the invention is described in detail by the claims that follow.

(157) Use of ordinal terms such as first, second, third. etc., in the claims to modify a claim element does not by itself connote any priority, precedence, or order of one claim element over another or the temporal order in which acts of a method are performed, but are used merely as labels to distinguish one claim element having a certain name from another element having a same name (but for use of the ordinal term) to distinguish the claim elements.