A novel approach to measure mitochondrial respiration in frozen biological samples

Abstract

Respirometry is the gold standard measurement of mitochondrial oxidative function, as it reflects the activity of the electron transport chain complexes working together. However, the requirement for freshly isolated mitochondria hinders the feasibility of respirometry in multi-site clinical studies and retrospective studies. Here, we describe a novel respirometry approach suited for frozen samples by restoring electron transfer components lost during freeze/thaw and correcting for variable permeabilization of mitochondrial membranes. This approach preserves 90-95% of the maximal respiratory capacity in frozen samples and can be applied to isolated mitochondria, permeabilized cells, and tissue homogenates with high sensitivity. We find that primary changes in mitochondrial function, detected in fresh tissue, are preserved in frozen samples years after collection. This approach will enable analysis of the integrated function of mitochondrial Complexes I to IV in one measurement, collected at remote sites or retrospectively in samples residing in tissue biobanks.

Keywords: frozen tissue; methodology; mitochondrial content; mitochondrial uncoupled respiration; oxygen consumption.

Figure 1. Mitochondria isolated from previously frozen liver maintain intact electron transport system

1. A

   Representative traces of oxygen consumption rate (OCR) of mouse liver mitochondria isolated from fresh or frozen tissue sustained by pyruvate + malate. Pyruvate + malate + ADP (PM + ADP), oligomycin (oligo), FCCP, and antimycin A + rotenone (AA + ROT) were sequentially injected to assess mitochondrial respiratory states.

2. B

   Pyruvate + malate‐dependent state 3 (substrate plus ADP)/state 4 (substrate without ADP) in fresh and frozen liver mitochondria.

3. C

   Quantification of maximal respiration rate (MRR) supported by pyruvate + malate in fresh and frozen liver mitochondria.

4. D

   Representative traces of OCR of liver mitochondria isolated from fresh or frozen tissue supported by the Complex II substrate succinate + rotenone + ADP (SR + ADP).

5. E

   Succinate + rotenone‐dependent state 3/state 4 in fresh and frozen liver mitochondria.

6. F

   Quantification of the different bioenergetic parameters sustained by succinate + rotenone in fresh and frozen liver mitochondria.

7. G

   Representative traces of OCR of liver mitochondria isolated from fresh or frozen tissue sustained by the Complex 1 substrate NADH + ADP.

8. H

   NADH‐dependent state 3/state 4 in fresh and frozen liver mitochondria.

9. I

   MRR driven by NADH in fresh and frozen liver mitochondria.

10. J

    Representative traces of OCR of liver mitochondria isolated from fresh or frozen tissue starting in state 1 and sustained by substrates without ADP (state 4) and by substrates with ADP (state 3). Mitochondria were tested for CAT sensitivity.

11. K

    Representative traces of OCR of liver mitochondria isolated from fresh or frozen tissue starting in state 1 and sustained by substrates with ADP (state 3). Mitochondria were tested for CAT sensitivity.

Data information: Panels (A, D, G, J, and K) are representative seahorse traces including four technical replicates. Biological replicates: (B and C), n = 4; (E and F), n = 6; and (H and I), n = 4. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM.

Figure 2. RIFS measurement of Complex I, II, and IV activity in fresh and frozen liver mitochondria

1. A

   Representative pyruvate + malate seahorse profile using RIFS the respirometry protocol in mouse liver mitochondria isolated from fresh or frozen tissue. Pyruvate + malate (Pyr + Mal), antimycin A + rotenone (AA + ROT), TMPD + ascorbate (TMPD/Asc), and azide were injected sequentially

2. B

   Pyruvate + malate‐ and TMPD/ascorbate (Complex IV, CIV)‐dependent respiration in fresh and frozen liver mitochondria.

3. C

   Representative succinate + rotenone seahorse profile using the RIFS respirometry protocol in mouse liver mitochondria isolated from fresh or frozen tissue.

4. D

   Succinate + rotenone (Succ)‐ and CIV‐dependent respiration in fresh and frozen liver mitochondria.

5. E

   Representative NADH seahorse profile using RIFS respirometry protocol in liver mitochondria isolated from fresh or frozen tissue.

6. F

   NADH‐ and CIV‐dependent respiration in fresh and frozen liver mitochondria.

Data information: Panels (A, C, and E) are representative seahorse traces including four technical replicates. Biological replicates: (B and F), n = 3–4; (D), n = 4–6. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM.

Figure 3. Lysates from fresh and frozen tissues maintain intact electron transport system

1. A

   Representative pyruvate + malate seahorse profile using the standard respirometry protocol in mouse liver homogenates obtained from fresh or frozen tissue. Pyruvate + malate using the standard respirometry protocol. Pyruvate + malate + ADP (PMA), oligomycin (oligo), FCCP, and antimycin A + rotenone (AA + ROT)

2. B

   MRR dependent by pyruvate + malate in fresh and frozen liver homogenates.

3. C

   Representative succinate + rotenone seahorse profile using the standard respirometry protocol in liver homogenates obtained from fresh or frozen tissue.

4. D

   MRR dependent by succinate + rotenone in fresh and frozen liver homogenate.

5. E

   Representative NADH seahorse profile using the standard respirometry protocol in liver homogenates obtained from fresh or frozen tissue.

6. F

   MRR dependent by NADH in fresh and frozen liver homogenate.

Data information: Panels (A, C, and E) are representative seahorse traces including four technical replicates. Biological replicates: (B), n = 6; (D), n = 8; and (F), n = 4. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM.

Figure 4. RIFS measurement of Complex I, II, and IV activity in fresh and frozen liver homogenates

1. A

   Representative pyruvate + malate seahorse profile using RIFS respirometry protocol in liver homogenates obtained from fresh or frozen tissue (top panel) and quantification (bottom panel).

2. B

   Representative succinate + rotenone seahorse profile using RIFS respirometry protocol in liver homogenates obtained from fresh or frozen tissue (top panel) and quantification (bottom panel).

3. C

   Representative NADH seahorse profile using RIFS respirometry protocol in liver homogenates obtained from fresh or frozen tissue (top panel) and quantification (bottom panel).

4. D

   NADH‐dependent respiration in frozen liver mitochondria in the presence of phenformin and metformin and the indicated concentration.

5. E

   Complex I‐, II‐, and IV‐dependent respiration in liver homogenates from WT and Ndufs4 KO mice showed impaired Complex I with normal Complex II and IV respiration in mouse Ndufs4 KO.

6. F

   Representative Western blot followed by quantification of Complex I (NDUFA9 and NDUFB8) levels in Ndufs4 KO samples. β‐actin was used as loading control.

Data information: Biological replicates: (A and C–F), n = 3–4; (B), n = 4–6. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM. For calculating statistics, the number of replicates used is the biological replicates indicated above. We used the following tests: two‐way ANOVA with Tukey's multiple comparison test in panels (D and F); unpaired t‐test in panel (E). *P < 0.05; **P < 0.01; ****P < 0.0001.Source data are available online for this figure.

Figure 5. RIFS is compatible with a variety of different tissues

1. A–E

   Representative seahorse profiles using RIFS respirometry protocol in frozen homogenates from mouse BAT (A), heart (B), kidney (C), brain (D), and lung (E).

2. F

   BNGE and Western blot of soleus muscle incubated with trypsin prior to homogenization.

3. G

   BNGE followed by Western blot of soleus muscle incubated with collagenase type II prior to homogenization.

4. H

   SDS–PAGE followed by Western blot of soleus muscle incubated with collagenase type II prior to homogenization.

5. I

   RIFS respirometry quantification in soleus muscle in the indicated conditions.

6. J

   Complex I‐, II‐, and IV‐dependent respiration in homogenates from frozen tissues from wild‐type mice using RIFS respirometry protocol per milligram of protein.

Data information: Biological replicates: (A–E, I and J), n = 3–6. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM. Western blots illustrate a representative blot of three independent experiments. Source data are available online for this figure.

Figure 6. Mitochondrial content quantification using MTDR

1. A

   Fluorescence in freshly isolated mouse liver mitochondria under the indicated treatments and stained with MTDR and MTR. Fluorescence was normalized by protein and referred as percentage of untreated samples.

2. B

   MTDR fluorescence in frozen liver mitochondria and homogenate at the indicated concentration. MTDR fluorescence correlates with protein amount but it saturates at higher protein levels.

3. C

   Linear correlation between MTDR fluorescence and protein abundance in frozen liver mitochondria and homogenate, when the system is not saturated. For this correlation, saturation values have been excluded and only linear range values have been considered.

4. D

   MTDR Operetta imaging in frozen liver mitochondria and homogenate at the indicated concentration.

5. E

   Western blot for mitochondrial protein and actin in increasing concentrations of frozen liver mitochondria and homogenate as indicated.

6. F

   Correlation between MTDR fluorescence (from B and C) and protein levels (from E) in frozen liver mitochondria and homogenate.

7. G

   MTDR Operetta imaging in frozen homogenates of different tissues.

8. H

   MTDR fluorescence in frozen tissue homogenate by microgram of sample.

9. I

   Spectrophotometric citrate synthase activity measure by milligram of lysate.

10. J

    Western blot and quantification of mitochondrial protein by actin (loading control) in tissue homogenates.

Data information: Biological replicates: (A), n = 4; (B–F), n = 5; and (G–J), n = 3. Every biological replicate represents the average of four technical replicates. Data are the mean ± SEM. Western blots illustrate a representative blot of three independent experiments. Source data are available online for this figure.

Figure 7. RIFS validation in pathophysiological relevant mouse and human disease models

1. A–D

   Complex I‐, II‐, and IV‐dependent respiration by protein (A), MTDR mitochondrial content (B), Complex I‐, II‐, and IV‐dependent respiration corrected by mitochondrial mass (C), and Western blot followed by quantification (D) in heart homogenates from WT and Ndufs4 KO mice.

2. E

   Complex I‐, II‐, and IV‐dependent respiration by protein in Polg mutant (Polg MT) and control heart homogenates.

3. F

   MTDR mitochondrial content in Polg MT and control heart homogenates.

4. G

   Complex I‐, II‐, and IV‐dependent respiration corrected by mitochondrial mass.

5. H

   Western blot followed by quantification in heart homogenates from WT (control) and Polg MT mice.

6. I

   Quantification of Complex I‐, II‐, and IV‐dependent respiration in zebrafish muscle homogenate from young and old fish.

7. J

   Quantification of Complex I‐ and IV‐dependent respiration using RIFS in zebrafish deyolked embryo homogenate acutely treated with rotenone before freezing.

8. K

   Complex I to Complex IV ratio in zebrafish deyolked embryo homogenate in the indicated conditions.

9. L

   Complex I activity measure spectrophotometrically in zebrafish deyolked embryo homogenate in the indicated conditions.

10. M

    Representative H&E sections from human donors show the presence of multilocular adipose tissue in the pheochromocytoma (Pheo) sample. UCP1‐DAB staining (central panels) and Grp75/mortalin fluorescence staining (bottom panels) from control and Pheo patients.

11. N

    Western blot followed by quantification in homogenates from control and Pheo patients.

12. O

    Complex I‐, II‐, and IV‐dependent respiration by protein in periadrenal adipose tissue homogenates from control and Pheo human patients.

13. P

    Mitochondrial content by MTDR quantification in homogenates from control and Pheo human patients.

14. Q

    Quantification of complex activities (I, II, and IV) in MAS versus MAS supplemented with Cyt c and ALA buffer in isolated and cryopreserved human monocytes, lymphocytes, and platelets.

Data information: Biological replicates: (A–I), n = 3; (J–L), n = 3 (each n represents a pool of 300 embryos); (N–P), n = 8; and (Q), n = 3–6 per group. Every biological replicate represents the average of at least four technical replicates. Data are the mean ± SEM. Western blots illustrate a representative blot of three independent experiments. The stainings are representative of eight control and four Pheo cases. Three representative fields were imaged and assessed for every biological sample. Contrast and brightness settings are the same across the samples. Scale bars represent 100 μm. For calculating statistics, the number of replicates used is the biological replicates indicated above. We used the following tests: two‐way ANOVA with Sidak's multiple comparison test in panels (A–E, G, and O); one‐way ANOVA Tukey's multiple comparison test in panels (J–L); unpaired t‐test in panels (F, H, N, and P). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Source data are available online for this figure.