Exogenous Mitochondrial Transfer in Differentiating Brown Adipocytes and AGPAT2-Deficient Preadipocytes

Summary

Exogenous mitochondria of human and murine origin can be internalized by differentiating preadipocytes and tracked throughout adipogenic differentiation using imaging and genetic tools. While these mitochondria physically interact with the endogenous mitochondrial network, they do not increase the expression of markers associated with mature brown adipocytes.

Abstract

Mitochondria are key signaling hubs; however, whether mitochondrial mass expansion is mechanistically required for differentiation remains an open question. AGPAT2 catalyzes the conversion of lysophosphatidic acid into phosphatidic acid, and its deficiency leads to adipose tissue deficiency and impaired adipogenesis associated with reduced mitochondrial mass. The impact of mitochondrial mass expansion on adipogenesis was assessed by transferring exogenous mitochondria into differentiating brown adipocytes. Whether mitochondrial transfer could rescue the impaired adipogenesis of AGPAT2-deficient cells was also investigated. Human and murine mitochondria were successfully transferred and incorporated into the endogenous mitochondrial network of differentiating mouse preadipocytes and persisted throughout brown adipogenesis. Adipogenic differentiation was required for the retention of transferred mitochondria. Mitochondrial transfer did not modify the expression of molecular markers of mature brown adipocytes or lipid droplet content, although it affected the relative distribution of lipid droplet size in a species-dependent manner. In Agpat2^-/- preadipocytes, mitochondrial transfer failed to rescue adipogenesis, indicating that mitochondrial mass expansion alone is insufficient to reverse the mechanisms leading to lipodystrophy in this model. These results indicate that, although exogenous human and murine mitochondria can be incorporated into the mitochondrial network of differentiating adipocytes, they do not directly influence the adipogenic program.

Introduction

Beyond their primary bioenergetic roles, mitochondria serve as signaling hubs in multiple cellular processes, including cell differentiation^1,2,3,4. Brown adipose tissue (BAT) possesses a notably high mitochondrial mass to support its thermogenic activity, highlighting its role in whole-body energy balance and in the pathogenesis of obesity⁵. While mitochondrial mass is regulated by the balance between biogenesis and mitophagy, it remains unclear whether mitochondrial mass expansion during brown adipocyte differentiation is functionally required for adipogenesis or merely a consequence of the process.

Physiological mitochondrial transfer between cells has been documented both in vitro and in vivo, where it contributes to tissue adaptation under pathological conditions, including adipose tissue dysfunction associated with obesity and insulin resistance^6,7. Furthermore, artificial mitochondrial transfer in vitro has been shown to enhance the respiratory capacity of recipient cells⁸. Together, these findings support the concept that mitochondrial transfer can modify cellular metabolic function and may contribute to tissue adaptation under conditions of bioenergetic stress^9,10,11. Compared with pharmacological or genetic strategies that stimulate endogenous mitochondrial biogenesis by activating nuclear and mitochondrial transcriptional programs, mitochondrial transfer directly delivers organelles to recipient cells, enabling rapid modification of mitochondrial mass independently of transcriptional manipulations^12,13. This approach is critical for correcting defects in mitochondrial diseases but also allows to study mitochondrial actions without the potential off-target effects of pharmacological or genetic approaches¹⁴.

In the present study, we investigated the effect of artificially increasing the mitochondrial mass by exogenous mitochondrial transfer in differentiating brown adipocytes. This approach enabled direct assessment of the impact of mitochondrial mass on the adipogenic program. To achieve this, mitochondrial transfer protocols originally developed for mesenchymal stem cells were adapted¹⁵ using both human hepatic cells and murine preadipocytes as mitochondrial donors. A method for tracking human mitochondria in differentiating mouse cells was developed using genetic tools combined with imaging techniques, demonstrating that mitochondria from both origins are incorporated into the endogenous mitochondrial network and persist throughout adipogenic differentiation.

To model defective adipogenesis associated with low mitochondrial mass, differentiating brown adipocytes derived from 1-acylglycerol-3-phosphate-O-acyltransferase 2 (AGPAT2)-deficient (Agpat2^-/-) mice were used. AGPAT2 is highly expressed in adipose tissue, where it catalyzes the formation of phosphatidic acid in the glycerolipid biosynthesis pathway. AGPAT2 deficiency causes severe lipodystrophy in both humans and mice and impairs adipogenesis in white and brown adipocyte models in association with reduced mitochondrial mass^16,17,18. Incorporation of exogenous mitochondria into differentiating preadipocytes slightly altered lipid droplet size distribution but did not increase the expression of mature brown adipocyte markers or rescue the adipogenic impairment of Agpat2^-/- preadipocytes.

Protocol

Animal procedures were approved by the Institutional Animal Care and Use Committee at Pontificia Universidad Católica de Chile, protocol 210609003. Wild-type and Agpat2^-/- P0.5 neonatal mice of both sexes, derived from a mixed C57BL/6J and 129J genetic background, were used in this study. Personal Protective Equipment (PPE): Wear a laboratory coat, safety glasses, and nitrile gloves. See Supplemental File 1 for additional protocol details and Table 1 for the buffers required for this protocol.

All toxic chemical waste generated during electron microscopy preparation, paraformaldehyde fixation, and nucleic acid extraction must be strictly segregated into designated, clearly labeled streams in compliance with institutional Environmental Health and Safety (EHS) regulations. Under no circumstances should any of these reagents be discharged into the laboratory sink or public sewage system. All hazardous chemical waste generated must be segregated according to institutional EHS classifications.

1. Cell culture

1. Isolation, differentiation, and maintenance of primary adipocytes (iBAT)
   NOTE: See Supplemental File 1 for a brief description of the procedures and Table 1 for the composition of the required solutions. A complete and detailed explanation of our in vitro brown adipogenic differentiation protocol can be found in Figueroa et al.¹⁹.
   1. Isolate preadipocytes from the interscapular brown adipose tissue (iBAT) of neonatal (P0.5) mice following the surgical and enzymatic digestion procedures described by Figueroa et al.¹⁹.
   2. Differentiate and maintain the brown adipocyte culture according to the differentiation cocktail and media exchange schedule.
      NOTE: For a successful adipogenic differentiation, it is critically important that cultures reach 90–100% confluence before initiating the adipogenic induction.
2. HepG2 culture
   1. Culture HepG2 cells in DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 µg/mL streptomycin, and 0.25 µg/mL amphotericin B. Maintain the cells at 37 °C in a humidified incubator with 5% CO₂.
      NOTE: These cells grow in clusters and may appear subconfluent despite high density due to multilayered growth.

2. Mitochondrial isolation and transference

1. Mitochondrial isolation
   NOTE: Perform all steps in a biosafety cabinet at 4 °C or on ice. Use the preparation within 2 h.
   1. Seed 2 × 10⁶ donor cells (HepG2 or preadipocytes) in 100 mm culture dishes. Grow cells until they reach 80% (HepG2) or 100% (preadipocytes) confluence.
   2. Stain donor cells with 100 nM MitoTracker Deep Red FM in FBS-free medium for 30 min at 37 °C in a 5% CO₂ incubator.
   3. Wash the cells 2x with growth medium to remove excess dye.
   4. Detach cells using trypsin, collect the suspension, and centrifuge at 300 × g for 5 min at 4 °C.
   5. Resuspend the pellet in 1 mL of mitochondrial isolation buffer (MIB) supplemented with BSA, protease inhibitors, and phosphatase inhibitors. Disrupt the plasma membrane by passing the suspension through a tuberculin syringe 5x. Homogenize the suspension by passing it through an insulin syringe at least 10x until the solution is light-brown and lacks visible clumps.
      NOTE: For western blot analysis, aliquot 50 µL of the total lysate and store at -20 °C.
2. Centrifuge the homogenate at 900 × g for 11 min at 4 °C.
   NOTE: If the supernatant after the 900 × g spin is not cloudy, cellular lysis is likely incomplete.
   1. Collect the supernatant and centrifuge at 10,000 × g for 10 min at 4 °C to pellet the mitochondria. Wash the mitochondrial pellet 3x with MIB supplemented with BSA.
   2. Perform a final resuspension in 100 µL of BSA-free MIB. Quantify protein concentration using a BCA protein assay.
3. Mitochondrial transfer assay (MTA) in preadipocytes
   1. Seed recipient preadipocytes at 5 × 10⁵ cells/well in 12-well plates (with 18 mm coverslips for microscopy) 24 h prior to the assay.
   2. Add isolated donor mitochondria directly to the recipient medium at a ratio of 10 µg of mitochondrial protein per 10⁵ cells. Rock the plates gently to distribute the mitochondria. Incubate cells for 24–120 h (corresponding to days 1, 3, and 5 of differentiation) at 37 °C in a humidified incubator with 5% CO₂ prior to analysis.

3. Mitochondrial oxygen consumption and MTA efficiency

1. Mitochondrial respirometry
   1. Set the external circulating water bath to 25 °C and ensure stable thermal equilibrium in the chamber jacket before initiating any measurements.
   2. Perform a two-point liquid phase calibration using the instrument's acquisition software.
      1. Air-saturation calibration (High Point): Add 500 µL of air-saturated demineralized water to the chamber. Set the magnetic stirrer to 60 rpm and record the signal until it stabilizes at a plateau (>1,500 mV).
      2. Zero-oxygen calibration (Low Point): Add a small (~3 mm diameter) amount of sodium sulfite to the chamber to displace any remaining air. Record until the signal drops to a stable baseline plateau (~1% of the basal signal, ~20 mV).
   3. Initiate real-time recording (total run time: ~10 min).
   4. Record for 3 min until a stable linear reading is achieved.
   5. Resuspend isolated mitochondria in 200 µL of mitochondrial assay buffer (MAS)²⁰ supplemented with 5 mM glutamate and 2.5 mM malate.
      NOTE: For the present work, the pellet of total mitochondria was extracted from two 100 mm cell-plates.
   6. Transfer the suspension to the Oxygraph chamber and close the chamber sealing plunger tightly, ensuring no air bubbles remain.
   7. Record oxygen consumption for 3 min.
   8. Add 150 µM ADP to trigger mitochondrial coupled respiration and measure ATP synthesis-coupled respiration for 3 min.
   9. Manually define the stable, linear segments corresponding to each respiratory phase. Extract the automated slope values calculated by the software, expressed as oxygen consumption rates in nmol∙mL^-1∙min^-1.
      NOTE: For interplate chamber cleaning, rinse the chamber 2x with 1 mL of demineralized water and ensure the interior is completely dry. Add 100 µL of PBS to the chamber before proceeding to the next plate.
2. Flow cytometry
   NOTE: Keep all samples on ice throughout the procedure to preserve cellular and bioenergetic integrity.
   1. Isolate mitochondria and perform mitochondrial transfer assay (MTA), as described in Sections 2.1 and 2.2, respectively, in 35 mm culture dishes, using donor cells previously labeled with MitoTracker Deep Red FM.
   2. Aspirate the culture medium and wash the cell monolayer once with 1 mL of sterile PBS.
   3. Add 500 µL of 0.25% Trypsin-EDTA and incubate the dishes at 37 °C until complete cell detachment is observed under a microscope.
   4. Neutralize trypsin by adding 1 mL of complete growth medium (DMEM with 10% FBS). Transfer the entire volume to a 15 mL conical tube and centrifuge at 300 × g for 5 min at 4 °C.
   5. Discard the supernatant by aspiration and retain the cell pellet. Resuspend the pellet in 500 µL of ice-cold PBS by gentle pipetting.
   6. Analyze the samples using a flow cytometer equipped with a 640 nm laser (see Supplemental File 1 for details on flow cytometry acquisition and analysis).
      1. Define the cell population using Forward Scatter (FSC) vs. Side Scatter (SSC) plots and apply a gating strategy to exclude debris and cell doublets.
      2. Quantify the fluorescence intensity in the far-red channel (670/30 nm emission filter) to detect internalized MitoTracker-labeled mitochondria.
      3. Perform analysis using Floreada.io (see Supplemental File 1 for gating protocol details).
         NOTE: Include a negative control (recipient cells without MTA) to set the baseline for background auto-fluorescence.
3. Confocal and electron microscopy
   1. Live cell imaging assay
      1. Seed primary preadipocytes derived from P0.5 neonatal mice into 12-well plates containing 18 mm coverslips, following the specific cell density and media requirements detailed in Section 1.
      2. Isolate mitochondria and execute the mitochondrial transfer assay (MTA) using donor cells previously labeled with MitoTracker Deep Red FM, as specified in Section 2.
      3. Maintain the recipient cultures in a humidified incubator at 37 °C with 5% CO₂ for 24 h to allow for mitochondrial uptake.
      4. Prepare a 70 nM MitoTracker Green working solution by diluting the stock in FBS-free culture medium.
      5. Aspirate the growth medium from the 12-well plates and replace it with the 70 nM MitoTracker Green solution. Incubate the plates for 25 min at 37 °C in a humidified incubator with 5% CO₂ to counterstain the total mitochondrial network.
      6. Remove the staining solution and wash the coverslips 2x with 1 mL of warm, complete culture medium.
      7. Replace the final volume with phenol red-free complete medium to minimize background auto-fluorescence during acquisition.
      8. Transfer the coverslips or the plate to the confocal microscope stage. Set the microscope’s incubation chamber to maintain a constant environment of 37 °C and 5% CO₂.
      9. Capture images immediately using appropriate excitation/emission settings for both far-red (MitoTracker Deep Red FM) and green (MitoTracker Green) channels. See Supplemental File 1 for additional details.
         NOTE: Perform all staining and handling steps under low-light conditions to prevent photobleaching of the fluorophores. Use a high-numerical aperture (NA) oil-immersion objective (60x or 100x) for optimal resolution of mitochondrial morphology.
   2. Confocal immunofluorescence assay with fixed cells
      NOTE: Following cell fixation, toxic paraformaldehyde and glutaraldehyde solutions are classified as non-halogenated organics. Nucleic acid extraction byproducts require strict segregation: phenol/chloroform mixtures must be disposed of as halogenated organics, while guanidinium-based buffers are classified as corrosive organics.
      1. Seed primary preadipocytes derived from P0.5 neonatal mice into 12-well plates containing 12 mm coverslips, as described in Section 1.
      2. Isolate mitochondria and perform the mitochondrial transfer assay (MTA) following the procedures specified in Section 2.
      3. Incubate the cultures for 24 h to allow for mitochondrial uptake.
      4. Induce brown adipogenic differentiation by following the protocol described in Supplemental File 1 and refer to Table 1 for the required buffers’ composition.
      5. Fix the cells on day 7 of differentiation (168 h). Aspirate the culture medium and add 500 µL of 4% paraformaldehyde (PFA) per well. Incubate for 20 min at room temperature.
      6. Wash the coverslips 2x by adding 1 mL of ice-cold PBS per well. Permeabilize and block the cells by incubating the coverslips in PBS containing 0.05% saponin and 0.2% BSA for 1 h at room temperature. Place the plate on an orbital shaker under gentle agitation.
      7. Wash the cells 2x with 1 mL of PBS. Stain the nuclei and lipid droplets simultaneously. Apply 1 µg/mL Hoechst 33342 and 1 µg/mL BODIPY prepared in the blocking/permeabilization buffer. Incubate for 30 min at room temperature, ensuring the plate is protected from light.
      8. Wash the coverslips 2x with 1 mL of ice-cold PBS. Mount the coverslips onto clean glass slides. Dispense 10 µL of mountant onto the slide and place the coverslip cell-side down onto the mounting medium. Allow the samples to dry overnight in a dark environment at room temperature before imaging.
      9. Acquire images using a confocal microscope. Use two fluorescence channels calibrated for the emission spectra of Hoechst and BODIPY. See Supplemental File 1 for additional imaging details.
      10. Perform lipid droplet analysis using ImageJ v1.54s. Open the lipid droplet channel, identify individual droplets manually using the Freehand selections and ROI Manager tools, and calculate the droplet area using the Analyze | Measure command.
      11. Process and analyze the resulting data using the R scripts provided in Supplemental File 2 within the RStudio v4.5.2 environment.
          NOTE: Acquire exactly five images per coverslip at 10x magnification to ensure representative sampling. Use R version 4.5.2 or higher for compatibility with the provided scripts.
   3. Transmission electron microscopy (TEM)
      NOTE: For electron microscopy byproducts, highly toxic osmium tetroxide and arsenic-containing sodium cacodylate buffers must be collected as toxic heavy metal waste, whereas uranyl acetate is managed separately as radioactive heavy metal waste under institutional radiation safety protocols.
      1. Centrifuge the isolated mitochondria, from Section 2, at 10,000 × g for 10 min at 4 °C to obtain a compact pellet.
      2. Fix the pellet by adding 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.0). Incubate for 2 h at 4 °C.
         NOTE: Always handle glutaraldehyde stock solutions and perform initial fixation steps inside a chemical fume hood. Wear nitrile gloves and goggles for eye protection. Avoid any contact with heating sources, as heated glutaraldehyde releases highly irritating vapors.
      3. Wash the mitochondrial preparation for 3 × 10 min with 0.05 M sodium cacodylate buffer, and centrifuge briefly to retain the pellet between buffer changes.
      4. Post-fix the sample by adding 1% aqueous osmium tetroxide (OsO₄). Incubate for 15 min at room temperature.
         NOTE: Osmium tetroxide must only be opened and handled inside a certified, fully functional chemical fume hood. Wear a lab coat, safety goggles or a full-face shield, and thick or double-nitrile gloves (change gloves immediately if any splash occurs, as OsO₄ can penetrate thin nitrile over time).
      5. Wash the sample for 3 × 5 min with double-distilled water. Stain the pellet with 1% aqueous uranyl acetate for 15 min.
         NOTE: Work on disposable plastic-backed bench paper. Dedicate a specific area of the lab for radioactive staining. Wear double gloves, a lab coat, and safety goggles. Wash hands thoroughly after completing the procedure.
      6. Dehydrate the preparation using a graded ethanol series. Incubate the pellet in 30%, 50%, 70%, 95%, and 100% ethanol for 5 min at each concentration.
      7. Embed the mitochondrial fractions. Incubate in a 1:1 Epon/ethanol solution for 30 min, followed by pure Epon resin for 1 h.
      8. Transfer the samples to embedding molds and polymerize in a 60 °C oven for 48 h.
      9. Attach the resin blocks to the ultramicrotome and cut ultrathin sections (80 nm).
      10. Collect sections onto grids. Stain the grids with 1% aqueous uranyl acetate for 4 min. Rinse with double-distilled water.
      11. Stain with lead citrate for 5 min following Reynolds’ method²¹. Rinse thoroughly. Visualize the grids using the TEM. See Supplemental File 1 for details on the Reynolds’ method and TEM image acquisition.

4. Molecular quantification

1. Mitochondrial DNA (mtDNA) standard curve
   1. Seed HepG2 cells into five 100 mm culture dishes. Incubate the cultures until they reach 80% of confluence.
   2. Isolate the mitochondria from the harvested cells by following the technical procedure detailed in Section 2.
   3. Extract total DNA from the resulting mitochondrial pellet using the acid-guanidinium-phenol-chloroform method as described in Section 4.2.6. Measure the DNA concentration and evaluate purity ((A₂₆₀/A₂₈₀ and (A₂₆₀/A₂₃₀ ratios) using a microvolume UV-Vis spectrophotometer.
   4. Prepare a serial dilution series using nuclease-free water as the diluent. Dilute the DNA stock to achieve final amounts of 1, 10^-1, 10^-2, 10^-3, 10^-4, and 10^-5 ng per reaction volume. Vortex each dilution for 5 s to ensure complete homogeneity before proceeding to the next step in the series.
      NOTE: Adjust the concentration range of the standard curve based on the specific sample concentration and the final PCR reaction volume.
2. Detection of Human mtDNA in recipient cells
   1. Seed primary preadipocytes derived from P0.5 neonatal mice into 60 mm culture dishes, following the density and media conditions specified in Section 1.
   2. Isolate mitochondria from donor HepG2 cells and execute the mitochondrial transfer assay (MTA) as detailed in Section 2.
   3. Incubate the cultures for 24 h at 37 °C and 5% CO₂.
   4. Induce brown adipogenic differentiation for 24–120 h (at days 1, 3, and 5) by following the media exchange protocol described in Section 1.
   5. Aspirate the culture medium and wash the cells 2x with 2 mL of ice-cold PBS.
   6. Extract total DNA using the acid-guanidinium-phenol-chloroform method²².
      NOTE: Always aliquot and perform phase separation inside the fume hood to avoid inhaling toxic phenol vapors. Use double-nitrile gloves or heavy-duty chemical-resistant gloves.
   7. Perform phase separation and discard the entire aqueous phase located above the interphase.
      CRITICAL STEP: Remove the aqueous phase completely to prevent contamination and ensure high DNA purity for downstream qPCR.
   8. Add 0.3 mL of 100% ethanol per 1 mL of the initial extraction reagent. Cap the tube and mix vigorously by inversion.
   9. Let it stand for 3 min at room temperature and centrifuge at 2,000 × g for 5 min at 4 °C to pellet the DNA.
   10. Transfer the phenol–ethanol supernatant to a separate tube for potential protein isolation. Resuspend the DNA pellet in 1 mL of 0.1 M sodium citrate (in 10% ethanol, pH 8.5) per 1 mL of extraction reagent used.
   11. Let it stand for 30 min at room temperature, gently inverting it a few times, and centrifuge at 2,000 × g for 5 min at 4 °C. Discard the supernatant.
   12. Repeat the sodium citrate wash steps two additional times.
   13. Wash the resulting pellet with 2 mL of 75% ethanol per 1 mL of initial extraction reagent.
   14. Let it stand for 20 min at room temperature and centrifuge at 2,000 × g for 5 min at 4 °C.
   15. Discard the supernatant and air-dry the pellet for 10 min. Solubilize the DNA by adding 0.3–0.6 mL of 8 mM NaOH and pipetting up and down repeatedly until the pellet disappears.
   16. Centrifuge at 12,000 × g for 10 min at 4 °C to remove any insoluble debris. Transfer the supernatant to a sterile tube and adjust the pH to 7–8 using HEPES.
   17. Measure DNA concentration and purity ((A(260)/A(280) and (A(260)/A(230) ratios) using a microvolume UV-Vis spectrophotometer.
   18. Dilute the template to a final amount of 15 ng of total DNA in nuclease-free water. Perform quantitative real-time PCR (qPCR) for human mtDNA using optical 96-well plates as specified in Section 4.3.
   19. Analyze the amplification results by applying a logarithmic regression model.
       NOTE: Conduct the logarithmic regression analysis using DNA concentration and corresponding threshold cycle (Ct) values as variables.
3. RT-qPCR for gene expression
   1. Cell preparation and differentiation
      1. Seed primary preadipocytes derived from P0.5 neonatal mice in 60 mm culture dishes as detailed in Section 1.1.
      2. Isolate mitochondria and perform the mitochondrial transfer assay (MTA) following the procedures in Section 2. Incubate the cultures for 24 h.
      3. Induce brown adipogenesis for 24–120 h (sampling at days 1, 3, and 5 of differentiation) according to the protocol in Section 1.
   2. Total RNA isolation
      1. Remove the culture medium and add 1 mL of Acid-guanidinium-phenol-chloroform extractant reagent per 1 × 10⁶ cells directly to the dish. Homogenize the samples by repeated pipetting until the lysate is uniform.
         NOTE: Always aliquot and perform phase separation inside the fume hood to avoid inhaling toxic phenol vapors. Use double-nitrile gloves or heavy-duty chemical-resistant gloves.
      2. Add 0.2 mL of chloroform per 1 mL of extractant reagent used. Cap the tubes securely and shake vigorously by hand for 15 s. Let it stand for 3 min at room temperature and centrifuge at 12,000 × g for 15 min at 4 °C.
      3. Transfer the upper aqueous phase carefully to a new nuclease-free tube. Avoid contact with the interphase. Keep the samples on ice.
      4. Add 0.5 mL of isopropanol per 1 mL of extractant reagent used. Mix gently by inversion and incubate for 10 min at room temperature.
      5. Centrifuge at 12,000 × g for 10 min at 4 °C to pellet the RNA. Discard the supernatant by inversion.
      6. Wash the RNA pellet with 1 mL of 75% ethanol. Vortex briefly and centrifuge at 7,500 × g for 5 min at 4 °C.
      7. Discard the supernatant and air-dry the RNA pellet for 10 min at room temperature. Resuspend the pellet in 30 µL of nuclease-free water. Incubate the samples at 60 °C for 15 min to ensure complete solubilization.
      8. Determine RNA concentration and purity (A₂₆₀/A₂₈₀ ratios) using a microvolume UV-Vis spectrophotometer.
      9. Verify RNA integrity via agarose gel electrophoresis.
   3. DNase treatment
      1. Add 0.1 volumes of 10× DNase buffer and 1 µL of DNase enzyme to each RNA sample. Mix gently by pipetting. Incubate the reactions at 37 °C for 30 min.
      2. Resuspend the DNase Inactivation Reagent by flicking or vortexing until the slurry is homogeneous. Add the Inactivation Reagent (0.1 volumes of the RNA sample or 2 µL, whichever is greater) and mix thoroughly. Incubate for 5 min at room temperature. Flick the tubes 3x during this period to maintain the reagent in suspension.
      3. Centrifuge at 10,000 × g for 1.5 min. Transfer the RNA-containing supernatant to a new tube.
         CRITICAL STEP: Ensure no DNase Inactivation Reagent is carried over, as it inhibits subsequent cDNA synthesis.
   4. Reverse transcription (RT)
      1. Quantify the treated RNA and dilute in nuclease-free water to use up to 2 µg of total RNA per 20 µL reaction.
      2. Prepare the RT master mix on ice following the manufacturer’s instructions. Dispense the mix into reaction tubes and add the RNA samples.
      3. Load into a thermal cycler and run the reverse transcription program as specified by the kit manufacturer.
   5. Quantitative real-time PCR (qRT-PCR)
      1. Use the cDNA generated in Section 4.3.4 as the PCR template. Dilute the cDNA in nuclease-free water to a working concentration of 20 ng per 10 µL reaction. Keep all the reagents on ice.
      2. Prepare a master mix containing SYBR Green, primers, and nuclease-free water. Include excess volume for pipetting loss. Dispense 10 µL of the reaction mixture into each well of an optical qPCR plate.
      3. Seal the plate with optical adhesive film and centrifuge briefly to collect the liquid at the bottom. Load the plate into a real-time PCR instrument and perform amplification using fast cycling mode.
      4. Determine the threshold cycle (Ct) values and perform relative quantification using the ΔΔCt method. Include a melting curve analysis at the end of the run to verify amplification specificity.
         NOTE: Normalize target mRNA abundance to 36B4 mRNA levels as the endogenous housekeeping gene. Run the qPCR plates within 2 h of setup; otherwise, store the sealed plates at 4 °C.
4. Western blot
   1. Quantify protein concentration from whole-cell and mitochondrial extracts (refer to Sections 2.1.8 and 2.1.12) using a BCA assay.
   2. Prepare 30 µg of protein per sample and add the appropriate volume of SDS-PAGE sample loading buffer (e.g., Laemmli buffer).
   3. Denature the samples by heating at 95 °C for 5 min (or follow specific temperature requirements for membrane proteins.
   4. Load the samples into the wells of a polyacrylamide gel. Separate proteins by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) at a constant voltage (e.g., 80–120 V) until the dye front reaches the bottom of the gel.
   5. Activate a PVDF membrane by immersing it in 100% methanol for 1 min, followed by equilibration in transfer buffer.
   6. Assemble the transfer sandwich, ensuring the absence of air bubbles between the gel and the membrane.
   7. Transfer the proteins onto the PVDF membrane by electroblotting at 400 mA for 2 h. Maintain the system at 4 °C throughout the process.
   8. Block non-specific binding sites by incubating the membrane in 5% non-fat milk prepared in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) for 1 h at room temperature.
   9. Incubate the membrane with primary antibodies (diluted 1:1,000 in blocking solution) overnight at 4 °C under gentle agitation.
   10. Wash the membrane 3 × 15 min with TBS-T.
   11. Incubate with horseradish peroxidase (HRP)-conjugated secondary antibodies for 2 h at room temperature.
   12. Wash the membrane 3x with TBS-T, followed by a final 5-min wash with 1× TBS.
   13. Visualize the protein bands using an enhanced chemiluminescence (ECL) substrate. Capture the signal using a digital imaging system.
       NOTE: Adjust the polyacrylamide gel percentage (e.g., 8%, 10%, or 12%) based on the molecular weight of the target proteins. If the electrotransfer apparatus lacks an active cooling system, replace ice containers every hour and keep the chamber submerged in an ice bath. Store membranes in 1× TBS at 4 °C if visualization is delayed.

Results

Mitochondrial extraction and structural and respiratory integrity

To evaluate the impact of increased mitochondrial mass on brown adipogenesis, a mitochondrial transfer assay (MTA) was performed in undifferentiated preadipocytes (Figure 1A). Mitochondria were isolated from primary murine preadipocytes (mMito) and from the human hepatic HepG2 cell line (hMito). Western blot analysis demonstrated that both preparations were highly enriched in the mitochondrial marker TOMM20 compared with whole-cell extracts (Figure 1B). Transmission electron microscopy (TEM) analysis of isolated fractions revealed well-defined double-membrane structures with preserved cristae, consistent with intact mitochondrial morphology (Figure 1C). Oximetry assays showed a significant increase in oxygen consumption, as indicated by a negative slope in the oxygen concentration curve after the addition of mitochondrial preparations into the medium, with a further increase after ADP supplementation, confirming that both hMito and mMito preparations retained coupled respiration (Figure 1D).

Efficacy and subcellular localization of transferred mitochondria

To evaluate transfer efficiency, donor mitochondria were labeled with MitoTracker Deep Red and added to recipient preadipocytes at a ratio of 10 µg of mitochondrial protein per 10⁵ recipient cells. After 24 h, flow cytometry analysis showed that approximately 60% of preadipocytes were positive for the MitoTracker Deep Red signal (Figure 1E), indicating efficient uptake of exogenous mitochondria. To determine the subcellular localization of incorporated mitochondria, recipient cells were counterstained with MitoTracker Green. Live-cell confocal microscopy revealed colocalization of the exogenous MitoTracker Deep Red-stained mitochondria with endogenous MitoTracker Green-stained mitochondrial network (Figure 1F,G). As ~50% of cells remained negative for MitoTracker Deep Red signal, the overall efficacy of exogenous mitochondria incorporation was ~50% (Figure 1G). Importantly, although these findings indicate that extracellular mitochondria are incorporated into the cellular mitochondrial network, they do not indicate functional incorporation into recipient cells’ bioenergetic or signaling pathways.

Persistence of transferred mitochondria during adipogenesis

In vitro differentiation of murine preadipocytes into brown adipocytes requires approximately 5–7 days. Therefore, the long-term persistence of transferred hMito was evaluated by quantifying human mitochondrial DNA (H-mtDNA) using species-specific qPCR. H-mtDNA remained detectable after 1, 3, and 5 days of differentiation (Figure 1H). In contrast, H-mtDNA levels rapidly declined in cells that were not induced to differentiate, becoming undetectable at day 3 of differentiation, suggesting that metabolic reprogramming or structural adaptations associated with adipogenic differentiation are required for the maintenance of exogenous mitochondria.

Impact of mitochondrial transfer assay (MTA) on brown adipogenic potential

The effect of supplementing mitochondrial mass on brown adipogenesis was evaluated. MTA did not change the mRNA abundance of canonical brown adipocyte markers (Figure 2A) but induced a mild but significant shift in lipid droplet (LD) size distribution. Specifically, transfer of mMito increased the proportion of small LDs (0–5 µm) while reducing the prevalence of larger LDs (>10 µm). In contrast, hMito transfer produced the opposite effect, favoring the accumulation of larger LDs (Figure 2B).

The lack of effects of mitochondrial transfer on brown adipogenic transcriptional markers might indicate that differentiating adipocytes have reached a mitochondrial mass sufficient for optimal adipogenic signaling. To test this possibility, MTA was performed in Agpat2^-/- preadipocytes, which exhibit reduced mitochondrial mass and impaired adipogenesis¹⁶. Although Agpat2^-/- cells successfully incorporated both hMito and mMito (Figure 2C), transferred mitochondria were rapidly lost by day 3 of differentiation, as determined by H-mtDNA quantification (Figure 2D). Finally, mitochondrial supplementation by MTA failed to increase the abundance of mature brown adipocyte markers at the mRNA level (Figure 2E), indicating that the adipogenic impairment caused by AGPAT2 deficiency cannot be rescued solely by increasing mitochondrial mass through artificial mitochondrial transfer.

Figure 1. Incorporation of transferred mitochondria of murine and human origin into differentiating brown adipocytes. (A) Workflow of the MTA. (B) Representative western blot of mitochondrial extraction purity, highlighting enrichment in TOMM20 and deficiency of other markers. (C) Representative images from transmission electron microscopy of mouse and human mitochondrial extracts, highlighting electron-dense double-membrane structures. Magnification 17,500x. N = 1. (D) Oximetry assay of mouse and human mitochondrial extract. Mean only. N = 3. (E) Flow cytometry of preadipocytes treated with mouse or human mitochondria stained with Mitotracker Deep Red for 24 h. A large population positive for the staining is highlighted. (F) Confocal microscopy of preadipocytes treated with mouse or human mitochondria stained with Mitotracker Deep Red for 24 h. Endogenous mitochondria were stained with Mitotracker Green. Scale bars = 50 µm. (G) Manders' colocalization coefficient indicating the fraction of exogenous mitochondria (MitoTracker Deep Red) overlapping with the endogenous mitochondrial network (MitoTracker Green). (H) H-mtDNA was detected in adipocytes during days 1, 3, and 5 of brown differentiation, * indicates differences relative to the undifferentiated, mitochondrial-free group, and # indicates differences within the same group relative to day 1. Mean ± SD. N = 3. Please click here to view a larger version of this figure.

Figure 2. Lack of effects of MTA on adipogenic differentiation markers and Agpat2^-/- adipocyte phenotype. (A) mRNA levels of adipogenic differentiation markers after 7 days of differentiation. Mean ± SD. N = 6. (B) MTA generates a slight but significant change in the size distribution of LDs, being smaller with mouse mitochondria and larger with human mitochondria. Scale bars = 100 µm. (C) Mitotracker Deep Red-stained mitochondria transferred to Agpat2^-/- preadipocytes treated for 24 h were detected by confocal microscopy. (D) H-mtDNA was detected in adipocytes during days 1, 3, and 5 of brown differentiation, * indicates differences relative to the undifferentiated, mitochondrial-free group, and # indicates differences within the same group relative to day 1. Mean ± SD. N = 3. (E) mRNA levels of mature adipocyte markers after 5 days of differentiation in Agpat2^-/- cells. No changes were observed. Mean ± SD. N = 3. Please click here to view a larger version of this figure.

Table 1: Buffers used in this protocol. Please click here to download this Table

Supplemental File 1: Supplemental methods and equipment specifications. Please click here to download this file.

Supplemental File 2: Lipid droplet distribution analysis. Please click here to download this file.

Discussion

Biological implications of mitochondrial transfer in adipogenesis

Our results show that although exogenous mitochondria from both human and murine sources can be incorporated into the endogenous mitochondrial network of differentiating preadipocytes, an artificial increase in mitochondrial mass is insufficient to drive the brown adipogenic program. These findings suggest that the mitochondrial expansion characteristic of brown adipogenesis is a consequence of cellular differentiation rather than a mechanistically limiting factor or a primary driver. These observations contrast with previous studies in pluripotent stem cells, where MTA significantly enhanced differentiation or induced metabolic reprogramming^23,24. In the present study, murine mitochondria were isolated from the stromal vascular fraction (SVF) of interscapular BAT, previously characterized as committed preadipocytes²⁵. Our data suggest that once cells are committed to the brown adipocyte lineage, the differentiation program is not substantially altered by increasing mitochondrial mass alone.

If mitochondria provide essential retrograde signals regulating the nuclear transcriptional program (e.g., MOTS-c, Humanin, long non-coding RNAs, ROS, or TCA intermediates)^{26,27,28,29,30}, the absence of enhanced adipogenesis following MTA may indicate that recipient cells already receive optimal signaling from their endogenous mitochondrial network. Alternatively, despite physically interacting with the host network, exogenous mitochondria may lack the specific signaling capacity to modulate nuclear transcriptional programs independently.

Cross-species transfer and metabolic priming

The divergent effects of human (hMito) and murine (mMito) mitochondria on lipid droplet (LD) size distribution highlight the relevance of mitochondrial source and metabolic priming. While these differences could theoretically stem from mito-nuclear incompatibility³¹, this is unlikely given the physical incorporation of both mitochondrial types into the recipient cells' network and the existing evidence that human mitochondria can functionally rescue murine models in vivo³². Instead, the cellular phenotype of the donor likely influences the metabolic priming of the isolated mitochondria³³. In this study, hMito were derived from a hepatocyte cell line, whereas mMito originated from preadipocytes, potentially rendering them differentially predisposed to lipid handling. The biological relevance of such metabolic priming warrants systematic investigation into future MTA-based therapeutic strategies.

Implications for the Agpat2^-/- phenotype

The inability of the MTA to rescue the Agpat2^-/- preadipocytes phenotype suggests that the lipodystrophic defect associated with AGPAT2 deficiency is not directly caused by reduced mitochondrial mass or impaired function, but rather by primary defects in lipid biosynthetic pathways. Notably, Agpat2^-/- adipocytes failed to maintain transferred mitochondria beyond day 1 of differentiation. This clearance is likely a secondary consequence of their severely impaired adipogenic program, mirroring the rapid mitochondrial loss observed in undifferentiated wild-type cells, further reinforcing the need for a permissive, differentiating environment to sustain mitochondrial retention.

Methodological advancements over existing approaches

To robustly evaluate these dynamics, the mitochondrial transfer workflow presented here was designed to overcome the limitations of previous methodologies. Unlike existing approaches that focus primarily on qualitative or short-term uptake³⁴, our workflow introduces quantitative, mitochondrial delivery with defined donor-to-recipient ratios. This enables controlled and reproducible mitochondrial input across independent experiments³⁵. Furthermore, donor mitochondria are pre-validated for bioenergetic competence prior to transfer³⁶ and are tracked longitudinally using a dual system of fluorescence labeling and species-specific mitochondrial DNA quantification³⁷. This dual approach allows for precise discrimination between donor and endogenous mitochondrial pools during long-term differentiation, enabling assessment of donor mitochondrial uptake and persistence rather than relying only on short-term physical internalization.

Protocol-critical steps for success

The success of this workflow depends on several protocol-critical steps that dictate transfer efficiency, intracellular retention, and downstream differentiation outcomes. First, maintaining the bioenergetic coupling and structural integrity of isolated mitochondria is absolute; this must be verified via respirometry (oxygen consumption) and structural preservation by TEM prior to MTA^36,38 . Second, adhering to a precise protein-to-cell ratio during transfer is critical to maximize uptake efficiency without inducing cytotoxic stress⁸. Third, the post-transfer metabolic status of the recipient cells is the primary determinant of mitochondrial persistence. In the present work, adipogenic differentiation was found to be required for the retention of exogenous mitochondria within differentiating adipocytes. Although mitochondrial mass expansion is known to occur during adipogenesis³⁹, these results suggest that cellular reprogramming towards brown adipocytes is strictly necessary for retention of mitochondrial mass, including exogenous mitochondria. The underlying mechanism for this phenomenon requires further research into the reciprocal signaling between the nuclear and mitochondrial genomes.

Applicability and methodological reproducibility

A key strength of this approach is its applicability for researchers seeking to implement MTA across different experimental systems. By relying on standardized, widely available reagents and clearly defined quantitative parameters (e.g., cell density, precise protein input, and defined post-transfer incubation windows), this protocol provides a highly reproducible scaling platform^8,40. Reproducibility is further strengthened by the incorporation of a multimodal validation framework. By combining structural visualization (confocal microscopy), molecular tracking (qPCR), and functional readouts (respirometry)⁸, researchers can cross-validate transfer success and ensure that structural, molecular, and functional outcomes are properly correlated. When recipient cell confluence, metabolic state, and processing times are standardized, transfer efficiency remains highly consistent across independent biological preparations.

Troubleshooting technical issues and mitochondrial loss

Despite these standardized parameters, researchers must account for technical variables that can compromise efficiency. A common issue is the reduced uptake of exogenous mitochondria, which typically results from prolonged isolation times that degrade mitochondrial integrity⁴¹, or the use of sub-optimal protein-to-cell ratios. In such cases, executing fresh isolations with minimal handling time is imperative. Additionally, variability in uptake efficiency often arises from differences in recipient cell confluence; both subconfluent and overconfluent cultures exhibit heterogeneous and reduced mitochondrial internalization⁴². Another frequent limitation is rapid mitochondrial loss post-uptake. In our study, human mtDNA signals rapidly decayed when transferred into non-differentiating cells. This loss likely reflects the activation of selective quality-control mechanisms, such as mitophagy, which clear exogenous organelles unless a sustained metabolic or thermogenic demand exists^39,43. Thus, the initiation of differentiation is a non-negotiable step to prevent experimental failure due to mitochondrial clearance.

Limitations

Although isolated mitochondria were confirmed to retain ATP-coupled respiration prior to transfer and to be incorporated into the recipient cells' endogenous mitochondrial network, several important questions remain unresolved. First, the specific contribution of exogenous versus endogenous mitochondria to total oxygen consumption rate (OCR) following mitochondrial transfer was not directly assessed. Second, the absolute contribution of transferred mitochondria to total mitochondrial mass was not quantified. Third, it remains unclear whether transferred mitochondria fully participate in retrograde signaling pathways regulating nuclear gene expression. Fourth, mitochondrial structural labeling with MitoTracker has the potential issue of leakage of free dye, risking artifactual cross-staining of endogenous mitochondria in recipient preadipocytes^44,45,46.

To mitigate this limitation, we evaluated the incorporation of exogenous mitochondria by simultaneously tracking and quantifying exogenous mitochondrial DNA, which is a highly sensitive and specific methodology not affected by dye leakage⁴⁴. An alternative method to circumvent this limitation is to establish stable cell lines expressing protein tags (such as Mito-GFP or Mito-mCherry) via lentiviral transduction. Finally, determining the extent of functional contribution of exogenous mitochondria to endogenous mitochondrial networks, including their bioenergetic, metabolic, and signaling roles, as well as establishing the efficacy and safety of cross-species mitochondrial transfer assays (MTA), remains a major scientific and technical challenge for future investigations. Addressing these questions will be essential for the development of effective MTA-based therapeutic strategies.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
100 mm plate	SPL Life Science	10101SPL
10x Tris/Glycine Buffer	BioRad	1610734
12 mm glass coverslips	Marienfeld	111520
12-well plate	SPL Life Science	30012SPL
16% Paraformaldehyde Aqueous Solution	Invitrogen	28908
18 mm glass coverslips	HART	-
24-well plate	SPL Life Science	30024SPL
3,3′,5-Triiodo-L-thyronine sodium salt (T3)	Sigma	T6397
3-Isobutyl-1-methylxanthine (IBMX)	Calbiochem	410957
30% Acrylamide/Bis Solution, 37.5:1	BioRad	1610158
35 mm plate	SPL Life Science	20035SPL
6-well plate	SPL Life Science	30006SPL
60 mm culture dish	SPL Life Science	20060SPL	Used for primary preadipocyte and DNA/qPCR experiments; supplier/catalog not provided
AccuRuler RGB Plus Pre-stained Protein Ladder	Maestrogen	02102-250
ACK lysing buffer	Gibco	A10492-01
Actrapid 100 UI/mL (Insulin)	Novo Nordisk	A10AB01
ADP	Merck	117105
Agarose	FERMELO BIOTEC	FER00AL200G	Used for RNA integrity verification; supplier/catalog not provided
Ammonium Persulfate	BioRad	1610700
Anti-rabbit IgG, HRP-linked Antibody	Cell Signaling	7074
Anti-rat IgG, HRP-linked Antibody	Cell Signaling	7077
Antibiotic-antimycotic	Gibco	15240062
Axio Observer inverted microscope stand	Carl Zeiss	431007-9901-000	Added from Supplemental Methods
BCA Protein Assay Kit	Thermo Scientific	D49328
BD Cytometer Setup and Tracking beads	BD Biosciences	-	Added from Supplemental Methods; catalog not provided
BD FACSDiva Software v8.0.3	BD Biosciences	-	Added from Supplemental Methods
BD LSRFortessa X-20 cell analyzer	BD Biosciences	-	Existing entry standardized to match supplemental file
BioRender	BioRender	-	Used to create Figure 1 workflow
Blotting-grade blocker	BioRad	170-6404
BODIPY 493/503	Invitrogen	D3922
Bovine Serum Albumin (BSA)	Sigma	A1470
CaCl2	Calbiochem	208291
cDNA Reverse Transcription Kit	Invitrogen	4374966
Cell Strainer 100 μm, nylon	Falcon	352360
Cell Strainer 40 μm, nylon	Falcon	352340
Collagenase type II	Gibco	17101-015
Dexamethasone	Sigma	D4902
DMEM Powder, High Glucose, Pyruvate	Thermo Scientific	12800017
DMEM/F-12, no phenol red	Thermo Scientific	21041025
DMEM/F-12, powder	Gibco	12500062
DMSO	Sigma	D8418	Solvent for induction medium; supplier/catalog not provided
EMBed-812 EMBEDDING KIT (Epon)	Electron Microscopy Sciences	14120
Ethanol absolute	Merck	100983
Ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid tetrasodium salt (EGTA)	Merck	324626
FAST SYBR MASTER MIX	Thermo Scientific	4385612
FCCP	Sigma	C2920
Fetal bovine serum	Capricorn Scientific	BS-16A
Floreada.io	Floreada.io	-	Flow cytometry analysis software
Fluoromount-G	Thermo Scientific	7984-25
GAPDH Antibody	Santa Cruz	sc-32233
Glucose	Gibco	15023-021
Glutamate	Sigma	G1626
Glutaraldehyde 25% Aqueous Solution	Electron Microscopy Sciences	16210
Goat anti-Mouse IgG (H+L) Secondary Antibody, HRP	Invitrogen	62-6520
GraphPad Prism v10.6.1	GraphPad Software	-	Statistical/logarithmic regression analysis software
Halt Protease and Phosphatase Inhibitor Cocktail 100X	Thermo Scientific	78442
HEPES	Biological Industries	41-122-100
Histone H3 (1G1)	Santa Cruz	sc-517576
Hoechst 33342	Life Technologies	H3570
ImageJ v1.54s	NIH	-	Lipid droplet image analysis software
Immun-Blot PVDF Membrane	BioRad	1620177
Indomethacin	Sigma	I7378
Insulin syringe	Cranberry	AAJECIN3	Used for cell homogenization; supplier/catalog not provided
KCl	Calbiochem	529552
KH2PO4	Calbiochem	529568
KHCO3	Sigma	60339
LAMP-1 Antibody	Abcam	ab25245
Lane Marker Reducing Sample Buffer	Thermo Scientific	39000
Malate	Sigma	M1000
MgCl2	Calbiochem	208291	Component of mitochondrial assay solution; supplier/catalog not provided
MgSO4	Sigma	M2643
Microvolume UV-Vis spectrophotometer	NanoDrop	ND-1000	Used for DNA/RNA quantification; supplier/catalog not provided
MilliQ water sterile	-	-
MitoTracker Deep Red FM	Thermo Scientific	M22426
MitoTracker Green	Thermo Scientific	M7514
NaCl	Merck	1064041000
Nikon C2plus confocal laser scanning microscope	Nikon	C2-SCHJ-1	Added from Supplemental Methods
Nikon Ti2 inverted microscope	Nikon	MEA54010	Added from Supplemental Methods
NIS-Elements AR software v5.11.03	Nikon	MQS31000	Added from Supplemental Methods
Nuclease-free water	Winkler	BM-0140
Optical 96-well qPCR plate	Thermo Scientific	4346907	Used for qPCR; supplier/catalog not provided
Optical adhesive film	Thermo Scientific	4311971	Used for qPCR plate sealing; supplier/catalog not provided
Orbital shaker	Lab-Line	4630-1	Used during blocking/permeabilization; supplier/catalog not provided
Osmium Tetroxide	Electron Microscopy Sciences	19100
Oxygraph+	Hansatech	-
PBS 10X Biologia Molecular	Winkler	BM-1340
Plan Apo 60x oil immersion objective, NA 1.4	Nikon	MRD01605	Added from Supplemental Methods
Plan-Apochromat 10x/0.45 M27 air objective	Carl Zeiss	420640-9900-000	Added from Supplemental Methods
R software v4.5.2	R Foundation	-	Used for data analysis
Real-time PCR instrument	Thermo Scientific	Quantum Studio 3	Used for reverse transcription and qPCR; supplier/catalog not provided
Reynolds Stain (Lead Citrate 3%)	Electron Microscopy Sciences	22410-1
RIPA lysis buffer	Thermo Scientific	89901
Rosiglitazone	Merck	557366
RStudio	Posit	-	Used for R analysis; manuscript/supplement cites RStudio
Saponin	Sigma	47036
Sodium bicarbonate	Sigma	S5761
Sodium Cacodylate	Electron Microscopy Sciences	12300
Sodium Dodecyl Sulfate	BioRad	1610301
Sodium hydroxide (NaOH)	Merck	B0547933	Solvent for T3; supplier/catalog not provided
Sucrose	Merck	573113	Component of mitochondrial isolation buffer; supplier/catalog not provided
Tetramethylethylenediamine (TEMED)	BioRad	1610800
Thermo Scientific F200C TEM	Thermo Scientific	-
TOMM20 antibody	Cell Signaling	42406
Tris Buffered Saline (TBS-10X)	Cell Signaling	12498
TRIzol reagent	Invitrogen	15596018
Trypsin-EDTA (0.25%)	Gibco	25200056
Tuberculin syringe	NIPRO	JD-01T2516/IB	Used for cell disruption; supplier/catalog not provided
TURBO DNA-free Kit	Thermo Scientific	AM1907
Tween 20, Molecular Biology Grade	Promega	H5152
Ultracut R	Leica	-
Uranyl Acetate	Electron Microscopy Sciences	22400
VAP-A Antibody	Santa Cruz	sc-293278
Vinculin Antibody	Santa Cruz	sc-73614
Westar Supernova	Cyanagen	XLS3
Zeiss LSM 880 confocal laser scanning microscope with Airyscan detector	Carl Zeiss	421102-9010-000
ZEN software	Carl Zeiss	-	Airyscan/deconvolution software
Gene	Forward (5’→3’)	Reverse (5’→3’)
H-mtDNA	CACTTTCCACACAGACATCA	CACTTTCCACACAGACATCA
PPARy	CACAATGCCATCAGGTTTGG	GCTGGTCGATATCACTGGAGATC
PPARα	ACAAGGCCTCAGGGTACCA	GCCGAAAGAAGCCCTTACAG
PCGα-1	AACCACACCCACAGGATC	TCTTCGCTTTATTGCTCCATGA
UCP1	GAGGTGTGGCAGTGTTCATTG	GGCTTGCATTCTGACCTTCA
PRDM16	GACATTCCAATCCCAGA	CACCTCTGTATCCGTCAGCA
PLIN1	GGGCTGTCTGAGACTGAGGT	GCAGAACTCTCTGGAGCACG
FATP4	ACCCAAAGCTGCCATTGTG	GCATGCGGAATCCAAGTACCA
36b4	CACTGGTCTAGGACCCGAGAAG	GGTGCCTCTGGAGATTTTCG