Protocol for isolation of lipid droplets and mitochondria from acute myeloid leukemia cells for lipidomic analysis Authors: Ran Zhao, Liang Zhang, Shuxin Dong, Kazuharu Kamachi, Saurabh Kumar Gupta, Hiroki Akiyama, Jalen Nguyen, G Lavender Hackman, Michael Andreeff, Alessia Lodi, Jo Ishizawa Published: 2025 ABSTRACT SummaryLipid droplets (LDs) are dynamic organelles critical for lipid and energy homeostasis. LD dysregulation is implicated in various diseases, including acute myeloid leukemia (AML), which exhibits aberrant lipid metabolism and chemoresistance. Here, we describe an effective protocol for the co-isolation of LDs and mitochondria from lipid-deficient cell types, e.g., AML cells. We detail procedures for the isolation, quality verification, and results from mass spectrometry-based lipidomic analysis. This protocol enables the functional study of organellar lipidomic regulation in AML.Subject areas: cell biology, cell isolation, cell separation/fractionation, mass cytometry, cancer Before you begin Lipid droplets (LDs) are dynamic organelles that actively regulate lipid and energy homeostasis.1,2 LDs, which are derived from endoplasmic reticulum and are driven by triacylglycerol (TAG) synthesis, consist of hydrophobic neutral lipids, mainly triacylglycerols (TAG) and sterol esters, surrounded by a phospholipid monolayer embedded with peripheral proteins.3 This protocol is for the extraction and isolation of lipid droplets (LDs) from AML cell lines (e.g., OCI-AML3, MOLM13), with optional co-isolation of mitochondria. AML cells are cultured in suspension and collected by centrifugation. A large culture is required to effectively isolate LDs due to the inherently lower LD load in AML cells compared with adipose or liver tissues. A snap-freezing step is employed to promote the release of intracellular contents during cellular homogenization. Following homogenization, low-speed (2,000 × g) centrifugation is first used to remove the nuclear fraction. Then, owing to the low density (<1 g/mL) of neutral lipids in LDs, mostly TAGs, high-speed (16,000 × g) centrifugation is used to separate LDs from the post-nuclear supernatant (PNS). Samples must be maintained at ∼4°C throughout the extraction steps to preserve LD integrity. Our protocol allows the simultaneous isolation of LDs and mitochondria from the same sample and eliminates the need for ultracentrifugation and specialized equipment. While this protocol was developed and used in AML cell models, this workflow can be applied to the study of other cell models and is an accessible tool to investigate the role of LDs in AML and the interplay between LDs and mitochondria (Figure 1). Schematic overview of LD and mitochondria isolation from AML cells Cell culture Timing: Variable It is recommended to culture a large volume of AML cells (see note below) for collection before starting the isolation steps. 1.For suspension culture AML cells, such as OCI-AML3 or MOLM13 (origins in the key resources table), cells are expanded in RPMI-1640 medium containing 10% fetal bovine serum (FBS) at 37°C with 5% CO2. Large-format tissue culture flasks, such as 175cm2 flasks, can be used. For suspension culture AML cells, such as OCI-AML3 or MOLM13 (origins in the key resources table), cells are expanded in RPMI-1640 medium containing 10% fetal bovine serum (FBS) at 37°C with 5% CO2. Large-format tissue culture flasks, such as 175cm2 flasks, can be used. 2.Passage cells every 2 days with fresh medium and keep the culture healthy in log phase growth and prevent over-confluence. Passage cells every 2 days with fresh medium and keep the culture healthy in log phase growth and prevent over-confluence. Note: It is possible to use a smaller sample size, but visual identification of the LD layer by eye during isolation becomes difficult (Figure 2B). It is highly recommended to use freshly cultured cells. Using Cells previously cryopreserved at −80°C or liquid nitrogen may result in lower LD yield and quality. Centrifugation and floatation of LDs (A) HIB is layered onto the cloudy PNS fraction in step 14. (B) After step 16, LDs form an off-white opaque layer floating at the top. There is an angled ring of LDs on the tube wall due to the fixed-angle rotor. (C) In steps 17 and 18, the LD fraction is collected and transferred to a 1.7 mL tube with a conical bottom and re-floated by centrifugation. (D) LDs are washed and concentrated. A clear layer of off-white LDs can be observed. Some LDs are angled on the tube wall due to the fixed-angle rotor. The blue arrow represents the 20–30 μL safety volume. Reagent or equipment preparation 3.Buffer preparation.a.Prepare Isolation Buffer (IB) and Hypotonic Isolation Buffer (HIB) as per recipes in the materials and equipment section and keep at 4°C.b.Add protease/phosphatase inhibitors fresh to IB/HIB immediately before use. Buffer preparation.a.Prepare Isolation Buffer (IB) and Hypotonic Isolation Buffer (HIB) as per recipes in the materials and equipment section and keep at 4°C.b.Add protease/phosphatase inhibitors fresh to IB/HIB immediately before use. a.Prepare Isolation Buffer (IB) and Hypotonic Isolation Buffer (HIB) as per recipes in the materials and equipment section and keep at 4°C. Prepare Isolation Buffer (IB) and Hypotonic Isolation Buffer (HIB) as per recipes in the materials and equipment section and keep at 4°C. b.Add protease/phosphatase inhibitors fresh to IB/HIB immediately before use. Add protease/phosphatase inhibitors fresh to IB/HIB immediately before use. 4.Equipment setup.a.Pre-cool the benchtop refrigerated microcentrifuge to 4°C. Confirm the centrifuge can reach 16,000 × g or higher without temperature rise.b.Prepare crushed ice in an insulated container as workspace. Place buffers and dounce homogenizer (Bellco glass #1984-10002, 2 mL, pestle B 0.06 mm clearance) on ice.c.Pre-label 1.5 mL microfuge tubes and 2 mL round-bottom tubes. Keep on crushed ice.d.Have an accessible −80°C freezer for the snap-freeze step, or use dry ice as a substitute. Equipment setup.a.Pre-cool the benchtop refrigerated microcentrifuge to 4°C. Confirm the centrifuge can reach 16,000 × g or higher without temperature rise.b.Prepare crushed ice in an insulated container as workspace. Place buffers and dounce homogenizer (Bellco glass #1984-10002, 2 mL, pestle B 0.06 mm clearance) on ice.c.Pre-label 1.5 mL microfuge tubes and 2 mL round-bottom tubes. Keep on crushed ice.d.Have an accessible −80°C freezer for the snap-freeze step, or use dry ice as a substitute. a.Pre-cool the benchtop refrigerated microcentrifuge to 4°C. Confirm the centrifuge can reach 16,000 × g or higher without temperature rise. Pre-cool the benchtop refrigerated microcentrifuge to 4°C. Confirm the centrifuge can reach 16,000 × g or higher without temperature rise. b.Prepare crushed ice in an insulated container as workspace. Place buffers and dounce homogenizer (Bellco glass #1984-10002, 2 mL, pestle B 0.06 mm clearance) on ice. Prepare crushed ice in an insulated container as workspace. Place buffers and dounce homogenizer (Bellco glass #1984-10002, 2 mL, pestle B 0.06 mm clearance) on ice. c.Pre-label 1.5 mL microfuge tubes and 2 mL round-bottom tubes. Keep on crushed ice. Pre-label 1.5 mL microfuge tubes and 2 mL round-bottom tubes. Keep on crushed ice. d.Have an accessible −80°C freezer for the snap-freeze step, or use dry ice as a substitute. Have an accessible −80°C freezer for the snap-freeze step, or use dry ice as a substitute. CRITICAL: Keep all centrifugation steps and buffers at 4°C. Higher temperatures may cause the disruption of LD integrity and emulsification. CRITICAL: Pestle B (0.06 mm clearance) must be used for proper shearing, minimizing disruption to organelles while facilitating release. Note: Use proper personal protective equipment when handling ultra-low temperature equipment and materials. Innovation LDs’ lipid composition is fundamental to their roles in lipid flux and homeostasis regulation. Mass spectrometry (MS)-based lipidomics, an essential tool for characterizing lipid species in LDs and these species’ roles in cell physiology,4,5 requires the effective isolation of LDs from cellular components. Despite emerging interest in the role of lipid metabolism in hematologic malignancies, the role of LDs in AML remains underexplored, partly due to the lack of established methods for isolating LDs from AML cells. Most existing protocols for isolating LDs focus on lipid-rich tissues or solid tumor cells, and require specialized instruments or ultracentrifugation.6,7,8,9 Importantly, isolating LDs from lipid-deficient cells, such as AML cells,10 presents a significant challenge. Meanwhile, no established protocol for isolating LDs from AML cells has been reported. Our protocol allows for the simple and efficient isolation of LDs from AML cells for lipidomic analysis, as we have found that such high speeds may not be essential for extracting LDs from human AML cells. Uniquely, our protocol also establishes a branching workflow to simultaneously isolate intact mitochondria from the same sample. As mitochondria are a known primary therapeutic target in AML, and given the close interplay between LDs and mitochondria, this dual approach facilitates investigation of the interlinking functions between LDs and mitochondria in AML disease progression and therapeutic response. Key resources table REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesβ-actin antibody (working dilution 1:10,000)Sigma-AldrichA5316Perilipin-2 (PLIN2) antibody (working dilution 1:500)AbceptaAP5118cCalreticulin antibody (working dilution 1:1,000)InvitrogenPA3-900TOMM20 antibody (working dilution 1:500)Santa Cruz Biotechnologysc-17764Secondary IRDye antibodies (working dilution 1:15,000)LI-COR Biosciences92668070, 92632211Chemicals, peptides, and recombinant proteinsTris-HCl, pH 7.4Sigma-Aldrich648317EDTASigma-AldrichE4884SucroseSigma-AldrichS7903PBSCorning21-040-CMProtease/Phosphatase Inhibitor Cocktail (100×)Cell Signaling TechnologyS5872SDSSigma-AldrichL3771GlycerolSigma-AldrichG5516β-mercaptoethanolSigma-AldrichM3148BODIPY 493/503InvitrogenD3922CF568 antibody labeling kitBiotium92275RIPA buffer (10×)Cell Signaling Technology9806SSPLASH Lipidomix Mass Spec StandardAvanti Polar Lipids33070714:0 Cardiolipin (sodium salt)Avanti Polar Lipids750332PMethanol (Optima LC/MS)Fisher ScientificA456-4Chloroform (for HPLC)MilliporeSigma34854Isopropanol (Optima LC/MS)Fisher ScientificA461-4Acetonitrile Optima (LC/MS Grade)Fisher ScientificA955-4Ammonium bicarbonateSigma-AldrichA61412,6-Di-tert-butyl-4- methylphenol (BHT)Fisher ScientificAC11299Ammonium formateFisher ScientificA11550Critical commercial assaysTriglyceride Colorimetric Assay KitCayman Chemical10010303Pierce BCA Protein Assay KitThermo Scientific23227Experimental models: Cell linesOCI-AML3Ontario Cancer Institute (provider: M. Minden)DSMZ-ACC 582MOLM13DSMZ (Braunschweig, Germany)DSMZ-ACC 554OtherDounce homogenizer, pestle B (0.06 mm clearance)Bellco Glass1984–100021.5 mL microfuge tubesAxygenMCT-175-C-S2 mL round-bottom tubesUSA Scientific1620–2700Kinetex 2.6-μm C18 100 Å, 150×2.1 mm columnPhenomenex00F-4462-ANQSertVial glass vial with a fused-in insertMilliporeSigma29392-URefrigerated microcentrifuge, 16,000 × g, with fixed angle rotor.Eppendorf5415R or equiv.37°C water bathFisher ScientificIsotemp 202−80°C freezerThermo Fisher ScientificTSX ULTFluorescence Imaging MicroscopeThermo Fisher ScientificEVOS FL AutoFluorescent scanner for western blotsLI-COROdyssey ClassicVanquish UHPLCThermo ScientificN/AOrbitrap IQ-X Tribrid Mass SpectrometerThermo ScientificN/A Materials and equipment Hypotonic Isolation BufferReagentStock concentrationFinal concentrationAdd to 100 mlTris-HCl, pH 7.41 M20 mM2 mLEDTA0.5 M10 mM2 mLMilliQ H2OUp to 100 mL Hypotonic Isolation Buffer Isolation BufferReagentStock concentrationFinal concentrationAdd to 100 mLTris-HCl, pH 7.41 M10 mM1 mLEDTA0.5 M1 mM0.2 mLSucrose1 M200 mM20 mLMilliQ H2OUp to 100 mL Isolation Buffer 2× SDS sample bufferReagentStock concentrationFinal concentrationAdd to 10mLTris-HCl, pH 6.81 M250 mM2.5 mLSDS20%8%4 mL2 mL of glycerol2 mL2 mL of ddH2OUp to 10 mL 2× SDS sample buffer CRITICAL: Add protease and phosphatase inhibitors to all buffers freshly before use, and then keep buffer solutions at 4°C. Note: Prepare buffers and store them at 4°C before use for up to 1 month. Make sure all solutions are fully dissolved and sterilized with a 0.22 μm filter. IB should be inspected before every use for contamination due to high sucrose content. Abbreviations: IB, isolation buffer; Tris-HCl, tris(hydroxymethyl)aminomethane hydrochloride; EDTA, ethylenediaminetetraacetic acid; MilliQ H2O, MilliQ-filtered water; HIB, hypotonic isolation buffer; SDS, sodium dodecyl sulfate. Step-by-step method details AML cell harvest and preparation Timing: 1 h Note: A sample size of 3 × 107 or more cells is recommended. Cell numbers vary depending on cell types. Collect, wash, and snap-freeze AML cell suspension for efficient lysis and organelle release. 1.Suspension cells are transferred in 50 mL conical tubes using serological pipettes. Pellet cultured AML cells at 400 × g, 4°C for 5 min. Suspension cells are transferred in 50 mL conical tubes using serological pipettes. Pellet cultured AML cells at 400 × g, 4°C for 5 min. 2.Resuspend the pellet in 1 mL PBS to wash and then transfer the cell suspension into a 1.5 mL microfuge tube. Resuspend the pellet in 1 mL PBS to wash and then transfer the cell suspension into a 1.5 mL microfuge tube. Optional: Save a small aliquot of cell suspension for immunoblotting. 3.Centrifuge at 400 × g, 4°C for 5 min and aspirate the supernatant without disturbing the pellet. Centrifuge at 400 × g, 4°C for 5 min and aspirate the supernatant without disturbing the pellet. 4.Snap-freeze the pellet at −80°C for 10 min. Snap-freeze the pellet at −80°C for 10 min. Note: Do not freeze the sample for long. 5.Briefly thaw the pellet in a 37°C water bath for 30 s, and then place it immediately on ice. Briefly thaw the pellet in a 37°C water bath for 30 s, and then place it immediately on ice. CRITICAL: Ensure protease and phosphatase inhibitors have been added to all downstream buffers (e.g. IB/HIB) before step 6. This prevents enzymic degradation of intracellular organelles. 6.Gently resuspend the pellet in 100 μL IB buffer per 1 × 107 cells. Incubate it on ice for 30 min. Gently resuspend the pellet in 100 μL IB buffer per 1 × 107 cells. Incubate it on ice for 30 min. Note: If working with 3 × 107 cells, you will resuspend in 300 μL IB in this step. Dounce homogenization Timing: 15–30 min This step uses mechanical shear to disrupt cellular membranes and release intracellular organelles. 7.Transfer the cell suspension to a 2 mL dounce homogenizer. Transfer the cell suspension to a 2 mL dounce homogenizer. 8.Using pestle B (0.06 mm clearance), perform 50 up-down strokes to homogenize the cells on ice. Using pestle B (0.06 mm clearance), perform 50 up-down strokes to homogenize the cells on ice. Note: Ensure consistent strokes, and do not press with force at the bottom. Cells are to be disrupted by shear force on the side wall, not crushed. 9.Transfer homogenate to a 1.5 mL microcentrifuge tube. Transfer homogenate to a 1.5 mL microcentrifuge tube. 10.Rinse the homogenizer with ∼200 μL IB and add the rinse to the same tube in step 9. Rinse the homogenizer with ∼200 μL IB and add the rinse to the same tube in step 9. Optional: For multiple samples, rinse the homogenizer and pestle with Milli-Q H2O before proceeding to the next sample to avoid cross-contamination. Centrifugal isolation and floating of LDs Timing: 3–4 h 11.Centrifuge the homogenate at 2,000 × g, 4°C for 10 min to pellet nuclei and large cellular debris. Centrifuge the homogenate at 2,000 × g, 4°C for 10 min to pellet nuclei and large cellular debris. 12.Observe a cloudy post-nuclear supernatant (PNS) and a nuclear pellet. Observe a cloudy post-nuclear supernatant (PNS) and a nuclear pellet. 13.Without disturbing the nuclear pellet, transfer the PNS to a 2 mL tube. Without disturbing the nuclear pellet, transfer the PNS to a 2 mL tube. Note: A faint opaque-white ring may be visible near the top of the tube. This contains a portion of the larger LDs and should be retained. A 2 mL tube is used here due to the straight tube wall. Optional: Save a small PNS aliquot for immunoblotting. Refer to step 22 for details. 14.Carefully layer 200 μL HIB on the top of the PNS. Due to the density difference, the clear HIB should float on the cloudy PNS (Figure 2A). Carefully layer 200 μL HIB on the top of the PNS. Due to the density difference, the clear HIB should float on the cloudy PNS (Figure 2A). 15.Centrifuge at 16,000 × g, 4°C for 60 min to float the LDs. Centrifuge at 16,000 × g, 4°C for 60 min to float the LDs. 16.Observe the tube (Figure 2B):a.There should be a floating faint opaque-white layer at the top as the LD layer.b.There may also be a white ring on the tube walls. These are also LDs.c.The supernatant should be clear, and there is a small pellet at the bottom, which contains mitochondria and other organelles. Observe the tube (Figure 2B):a.There should be a floating faint opaque-white layer at the top as the LD layer.b.There may also be a white ring on the tube walls. These are also LDs.c.The supernatant should be clear, and there is a small pellet at the bottom, which contains mitochondria and other organelles. a.There should be a floating faint opaque-white layer at the top as the LD layer. There should be a floating faint opaque-white layer at the top as the LD layer. b.There may also be a white ring on the tube walls. These are also LDs. There may also be a white ring on the tube walls. These are also LDs. c.The supernatant should be clear, and there is a small pellet at the bottom, which contains mitochondria and other organelles. The supernatant should be clear, and there is a small pellet at the bottom, which contains mitochondria and other organelles. 17.Transfer the LD layer to a new 1.5 mL microfuge tube.a.Pre-wet a 200 μL tip with HIB by pipetting up and down a few times.b.Using the pre-wetted tip, transfer the LD layer from the very top to a new 1.5 mL microfuge tube, ∼50 μL at a time.c.Repeating 5–10× or until all visible LDs are collected. Transfer the LD layer to a new 1.5 mL microfuge tube.a.Pre-wet a 200 μL tip with HIB by pipetting up and down a few times.b.Using the pre-wetted tip, transfer the LD layer from the very top to a new 1.5 mL microfuge tube, ∼50 μL at a time.c.Repeating 5–10× or until all visible LDs are collected. a.Pre-wet a 200 μL tip with HIB by pipetting up and down a few times. Pre-wet a 200 μL tip with HIB by pipetting up and down a few times. b.Using the pre-wetted tip, transfer the LD layer from the very top to a new 1.5 mL microfuge tube, ∼50 μL at a time. Using the pre-wetted tip, transfer the LD layer from the very top to a new 1.5 mL microfuge tube, ∼50 μL at a time. c.Repeating 5–10× or until all visible LDs are collected. Repeating 5–10× or until all visible LDs are collected. Note: This should result in 250 μL or higher volume of dilute, suspended LDs. CRITICAL: A 1.5 mL microfuge tube is used here due to the sharp conical bottom for downstream washing steps. Optional: Reserve the remaining cytosolic fraction as well as the pellet on ice for optional mitochondria purification. See major step “co-Isolation of mitochondria from the same preparation”). Optional: Reserve a small cytosolic fraction aliquot for immunoblotting. Refer to step 22 for details. 18.Centrifuge the resulting suspension from step 17 (LD fraction) at 16,000 × g, 4°C for 10 min. Centrifuge the resulting suspension from step 17 (LD fraction) at 16,000 × g, 4°C for 10 min. 19.Using a pre-wetted gel-loading tip, carefully remove the bottom clear layer, leaving the ∼100 μL volume and LD layer intact.a.Observe any disruption of LD layer by the aspiration.b.Repeat the centrifugation step 18 if necessary to re-float the LDs until bottom clear layer removal is complete (Figure 2C). Using a pre-wetted gel-loading tip, carefully remove the bottom clear layer, leaving the ∼100 μL volume and LD layer intact.a.Observe any disruption of LD layer by the aspiration.b.Repeat the centrifugation step 18 if necessary to re-float the LDs until bottom clear layer removal is complete (Figure 2C). a.Observe any disruption of LD layer by the aspiration. Observe any disruption of LD layer by the aspiration. b.Repeat the centrifugation step 18 if necessary to re-float the LDs until bottom clear layer removal is complete (Figure 2C). Repeat the centrifugation step 18 if necessary to re-float the LDs until bottom clear layer removal is complete (Figure 2C). 20.Wash the LDs.a.Add 200 μL of HIB to the LD suspension.b.Centrifuge at 16,000 × g, 4°C for 10 min.c.Repeat step 19 and carefully remove the bottom layer of HIB. Wash the LDs.a.Add 200 μL of HIB to the LD suspension.b.Centrifuge at 16,000 × g, 4°C for 10 min.c.Repeat step 19 and carefully remove the bottom layer of HIB. a.Add 200 μL of HIB to the LD suspension. Add 200 μL of HIB to the LD suspension. b.Centrifuge at 16,000 × g, 4°C for 10 min. Centrifuge at 16,000 × g, 4°C for 10 min. c.Repeat step 19 and carefully remove the bottom layer of HIB. Repeat step 19 and carefully remove the bottom layer of HIB. 21.Concentrate the LDs.a.Centrifuge the LD suspension at 16,000 × g, 4°C for 10 min.b.Remove the bottom layer without disturbing the floating LDs until approximately 50 μL volume is left.c.Centrifuge the LDs at 16,000 × g, 4°C for 10 min.d.Remove as much of the bottom layer as possible, leaving ∼20–30 μL volume to minimize losses (Figure 2D). Concentrate the LDs.a.Centrifuge the LD suspension at 16,000 × g, 4°C for 10 min.b.Remove the bottom layer without disturbing the floating LDs until approximately 50 μL volume is left.c.Centrifuge the LDs at 16,000 × g, 4°C for 10 min.d.Remove as much of the bottom layer as possible, leaving ∼20–30 μL volume to minimize losses (Figure 2D). a.Centrifuge the LD suspension at 16,000 × g, 4°C for 10 min. Centrifuge the LD suspension at 16,000 × g, 4°C for 10 min. b.Remove the bottom layer without disturbing the floating LDs until approximately 50 μL volume is left. Remove the bottom layer without disturbing the floating LDs until approximately 50 μL volume is left. c.Centrifuge the LDs at 16,000 × g, 4°C for 10 min. Centrifuge the LDs at 16,000 × g, 4°C for 10 min. d.Remove as much of the bottom layer as possible, leaving ∼20–30 μL volume to minimize losses (Figure 2D). Remove as much of the bottom layer as possible, leaving ∼20–30 μL volume to minimize losses (Figure 2D). Preparation of LDs for downstream verification Timing: 2+ h Verification of isolated LDs may be accomplished by immunoblotting, fluorescence imaging, and LC-MS lipidomic analysis. Below describes the preparation procedures for downstream analysis. Pause point: The final concentrated LD suspension can be held at 4°C overnight for non-disruptive verification assays, such as fluorescence imaging, and frozen at −80°C for LC-MS lipidomic analysis. 22.For immunoblotting, add 2× SDS sample buffer into the lysate at a 1:1 ratio by volume.a.Perform Pierce BCA assay according to the manufacturer’s instructions to normalize protein concentration before immunoblotting. For immunoblotting, add 2× SDS sample buffer into the lysate at a 1:1 ratio by volume.a.Perform Pierce BCA assay according to the manufacturer’s instructions to normalize protein concentration before immunoblotting. a.Perform Pierce BCA assay according to the manufacturer’s instructions to normalize protein concentration before immunoblotting. Perform Pierce BCA assay according to the manufacturer’s instructions to normalize protein concentration before immunoblotting. Note: This step may require up to half of the collected LDs, as LDs do not contain large amounts of proteins. Note: A higher concentration of SDS (final 4%) is needed due to the high lipid content of LDs. 23.For fluorescence imaging of isolated LDs, freshly collected, non-disrupted and non-fixed LDs should be used.a.Add 10 μM BODIPY 493/503 (final concentration 10 μM) and PLIN2-CF568 conjugated fluorescent antibody (final concentration 100 ng/μL).b.Incubate for 1 hour at room temperature.c.Wash twice with HIB at 16,000 × g at 4°C for 10 min. Remove the bottom layer.d.Keep on ice until ready for imaging. For fluorescence imaging of isolated LDs, freshly collected, non-disrupted and non-fixed LDs should be used.a.Add 10 μM BODIPY 493/503 (final concentration 10 μM) and PLIN2-CF568 conjugated fluorescent antibody (final concentration 100 ng/μL).b.Incubate for 1 hour at room temperature.c.Wash twice with HIB at 16,000 × g at 4°C for 10 min. Remove the bottom layer.d.Keep on ice until ready for imaging. a.Add 10 μM BODIPY 493/503 (final concentration 10 μM) and PLIN2-CF568 conjugated fluorescent antibody (final concentration 100 ng/μL). Add 10 μM BODIPY 493/503 (final concentration 10 μM) and PLIN2-CF568 conjugated fluorescent antibody (final concentration 100 ng/μL). b.Incubate for 1 hour at room temperature. Incubate for 1 hour at room temperature. c.Wash twice with HIB at 16,000 × g at 4°C for 10 min. Remove the bottom layer. Wash twice with HIB at 16,000 × g at 4°C for 10 min. Remove the bottom layer. d.Keep on ice until ready for imaging. Keep on ice until ready for imaging. Note: Other markers or suitable dye combinations are also possible. LDs can be imaged live without fixing by loading a small volume (∼10 μL) onto a cover glass and slide. 24.LD extract can be further prepared for LC-MS lipidomics or frozen until use. LD suspension can be frozen at −80°C for up to 1 month.a.Prior to freezing, triglyceride assay kits (Cayman Chemicals, #10010303) can be used to normalize multiple LD samples before LC-MS lipidomic analysis (Figure 3A).b.Extract lipids for LC-MS lipidomic analysis using a modified Bligh-Dyer extraction as previously described.11,12,13,14c.Spike each sample with diluted SPLASH Lipidomix Mass Spec Standard (Avanti Polar Lipids, 330707; 1:20 dilution) and 14:0 Cardiolipin (Avanti Polar Lipids, 750332P; final concentration 5 μM) as lipid internal standards.d.Perform UHPLC-MS/MS lipidomic analysis (procedure detailed below in “quantification and statistical analysis”). LD extract can be further prepared for LC-MS lipidomics or frozen until use. LD suspension can be frozen at −80°C for up to 1 month.a.Prior to freezing, triglyceride assay kits (Cayman Chemicals, #10010303) can be used to normalize multiple LD samples before LC-MS lipidomic analysis (Figure 3A).b.Extract lipids for LC-MS lipidomic analysis using a modified Bligh-Dyer extraction as previously described.11,12,13,14c.Spike each sample with diluted SPLASH Lipidomix Mass Spec Standard (Avanti Polar Lipids, 330707; 1:20 dilution) and 14:0 Cardiolipin (Avanti Polar Lipids, 750332P; final concentration 5 μM) as lipid internal standards.d.Perform UHPLC-MS/MS lipidomic analysis (procedure detailed below in “quantification and statistical analysis”). a.Prior to freezing, triglyceride assay kits (Cayman Chemicals, #10010303) can be used to normalize multiple LD samples before LC-MS lipidomic analysis (Figure 3A). Prior to freezing, triglyceride assay kits (Cayman Chemicals, #10010303) can be used to normalize multiple LD samples before LC-MS lipidomic analysis (Figure 3A). b.Extract lipids for LC-MS lipidomic analysis using a modified Bligh-Dyer extraction as previously described.11,12,13,14 Extract lipids for LC-MS lipidomic analysis using a modified Bligh-Dyer extraction as previously described.11,12,13,14 c.Spike each sample with diluted SPLASH Lipidomix Mass Spec Standard (Avanti Polar Lipids, 330707; 1:20 dilution) and 14:0 Cardiolipin (Avanti Polar Lipids, 750332P; final concentration 5 μM) as lipid internal standards. Spike each sample with diluted SPLASH Lipidomix Mass Spec Standard (Avanti Polar Lipids, 330707; 1:20 dilution) and 14:0 Cardiolipin (Avanti Polar Lipids, 750332P; final concentration 5 μM) as lipid internal standards. d.Perform UHPLC-MS/MS lipidomic analysis (procedure detailed below in “quantification and statistical analysis”). Perform UHPLC-MS/MS lipidomic analysis (procedure detailed below in “quantification and statistical analysis”). Note: Required materials are listed in the key resources table. Verification of the isolated lipid droplets (LDs) from acute myeloid leukemia (AML) cells (A) Triglyceride quantification of LDs extracted from 3 × 107 OCI-AML3 cells (n=8 or 4); results are shown as mean ± standard deviation. Quantification was performed using a triglyceride colorimetric assay kit. (B) Immunoblotting of AML cellular fractions extracted during LD isolation. 15 μg of protein was loaded for all wells, except for LD at 4 μg. WCL, whole cell lysate; PNS, post-nuclear supernatant; CYT, cytosolic; MITO, mitochondrial. (C) Fluorescence images of isolated LDs from OCI-AML3 cells. Isolated LDs were stained with 10 μM BODIPY 493/503 and with 100 ng/μL CF568-conjubated anti-PLIN2 antibody for 1 hour at room temperature. Co-isolation of mitochondria from the same preparation Timing: 1–2 h Co-isolation workflow of the organelle from the reserved cytosolic fraction and the organelle pellet from step 17. 25.From the tube and material reserved from step 17, add 500 μL IB to resuspend the pellet. From the tube and material reserved from step 17, add 500 μL IB to resuspend the pellet. Note: Organelle pellet may appear difficult to resuspend, but can be properly broken apart by vigorous pipetting with a P1000 tip. Note: The high concentration of sucrose in IB acts as an organelle protectant for mitochondria and is often used in many established mitochondria isolation workflows. 26.Centrifuge at 2,000 × g, 4°C for 5 min to pellet residual cellular debris and larger structures. Centrifuge at 2,000 × g, 4°C for 5 min to pellet residual cellular debris and larger structures. 27.Transfer the supernatant to a new 1.5 mL tube and centrifuge at 12,000 × g, 4°C for 10 min to pellet.a.The pellet should appear off-white and consist of mitochondria. Transfer the supernatant to a new 1.5 mL tube and centrifuge at 12,000 × g, 4°C for 10 min to pellet.a.The pellet should appear off-white and consist of mitochondria. a.The pellet should appear off-white and consist of mitochondria. The pellet should appear off-white and consist of mitochondria. 28.Discard supernatant; wash the mitochondria pellet with 500 μL IB, then centrifuge at 12,000 × g, 4°C for 10 min. Discard supernatant; wash the mitochondria pellet with 500 μL IB, then centrifuge at 12,000 × g, 4°C for 10 min. 29.Repeat step 28. Repeat step 28. 30.Resuspend the mitochondria pellet in ∼50 μL IB (or appropriate volume). Resuspend the mitochondria pellet in ∼50 μL IB (or appropriate volume). Optional: Reserve a small aliquot (∼10 μL) and lyse with 1× RIPA buffer to quantify mitochondrial protein (e.g., BCA assay) for normalization. 31.Prepare mitochondria suspension for immunoblotting.a.Additionally, mitochondria can be re-pelleted and frozen at −80°C before extraction of mitochondrial lipids (using a modified Bligh-Dyer extraction) and acquisition of the LC-MS lipidomic data. Prepare mitochondria suspension for immunoblotting.a.Additionally, mitochondria can be re-pelleted and frozen at −80°C before extraction of mitochondrial lipids (using a modified Bligh-Dyer extraction) and acquisition of the LC-MS lipidomic data. a.Additionally, mitochondria can be re-pelleted and frozen at −80°C before extraction of mitochondrial lipids (using a modified Bligh-Dyer extraction) and acquisition of the LC-MS lipidomic data. Additionally, mitochondria can be re-pelleted and frozen at −80°C before extraction of mitochondrial lipids (using a modified Bligh-Dyer extraction) and acquisition of the LC-MS lipidomic data. CRITICAL: Mitochondria extracted from this step will contain mitochondrial-associated membranes (MAM), often from the ER, and should be considered as a crude mitochondria fraction. Further washing and gradient centrifugation procedures may be used to obtain pure mitochondria if desired. Expected outcomes Successful LD isolation yielded a final suspension of opaque-white LDs in HIB. A neutral lipid assay, such as Colorimetric Triglyceride Assay Kit (Cayman Chemicals 10010303) can be used to quantify TAGs to verify LD yield. The LDs isolated from 3 × 107 cells OCI-AML3 and MOLM13 cells contain ∼2.2 μg TAGs and ∼4.0 μg TAGs, respectively (Figure 3A). Immunoblotting revealed a high signal of PLIN2 exclusively in the LD fraction, and a high TOMM20 signal only in the mitochondria fraction but not in PNS, cytosolic, or whole cell samples (Figure 3B), supporting the successful isolation of LDs and crude mitochondria. Fluorescence imaging of isolated LDs stained with BODIPY 493/503 (green) and anti-PLIN2-CF568 (red) showed co-localization of the signals of LDs and PLIN2 in the ∼1 μm round spherical structures. Proper co-localization of both colors represents an intact LD phospholipid monolayer with contained neutral lipids (Figure 3C). MS-lipidomic analysis of OCI-AML3 derived samples was performed as a downstream and separate workflow, showing highly enriched (∼90%) TAGs in the LDs, consistent with LD’s main function of neutral lipid storage. We found that TAGs account for 45% and 31% of the total number of identified lipid species in LDs and whole cell lysates, respectively (Figure 4A), while accounting for 90% (LD) and 18% (WCL) of the lipid composition (Figure 4B), confirming the purity of the isolated LDs. In addition to glycerolipids (including both TAGs and diacylglycerols, DAGs), sphingolipids (including sphingomyelin, SM; hexosylceramide HexCer; and ceramide, Cer) and glycerophospholipids (phosphatidylcholine, PC; phosphatidylethanolamine, PE; phosphatidylinositol, PI; phosphatidylserine, PS, lysophosphatidylcholine, LPC) account for 32% and 18% of the total number of lipid species identified in LDs, respectively. Mass spectrometric lipidomic analysis of isolated lipid droplets (LDs) (A) Prevalence and number of lipid species identified in each lipid class in whole OCI-AML3 cells (WCL) and in mitochondria (MITO) and LDs isolated from OCI-AML3 cells (n=4). HexCer, hexosylceramide; Cer, ceramide; CL, cardiolipin; LPC, lysophosphatidylcholine; LPE, lysophosphatidylethanolamine; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidylserine; SM, sphingomyelin; DAG, diacylglycerol; TAG, triacylglycerol. (B) Lipid composition (as percent of the total amount of lipids for each lipid class) in WCL, MITO and LD samples (n=4). “Others” includes LPC, LPE and PG. HexCer and Cer were not quantified and are not included in this analysis. (C) Cardiolipin levels (as percent of the amount of all lipid classes) in WCL, MITO and LD samples (n=4); n.d., not detected; results are shown as mean±standard deviation; p-value was calculated based on student t test. In the isolated mitochondria, we found an enrichment of phosphatidylethanolamine (PE) to approximately 30%, consistent with previous reports.15 Cardiolipin, a specific glycerophospholipid component of mitochondrial membranes, was also found to be significantly enriched in the mitochondria compared to whole cell lysates (Figures 4B and 4C). The composition of the main lipid subclasses (at least 1% quantified in at least one of the sample types) in whole cells, mitochondria, and lipid droplets, respectively, was as follows (shown as mean±standard deviation; Figure 4B). Phosphatidylcholine: 45.9%±2.4%; 43.7%±1.5%; 5.2%±2.2%. Phosphatidylcholine: 45.9%±2.4%; 43.7%±1.5%; 5.2%±2.2%. Phosphatidylethanolamine: 20.7%±1.2%; 33.7%±1.7%; 0.9%±0.4%. Phosphatidylethanolamine: 20.7%±1.2%; 33.7%±1.7%; 0.9%±0.4%. Triacylglycerol: 18.2%±1.3%; 10.2%±1.0%; 89.7%±3.1%. Triacylglycerol: 18.2%±1.3%; 10.2%±1.0%; 89.7%±3.1%. Phosphatidylinositol: 6.4%±0.3%; 3.2%±0.2%; 0.9%±0.5%. Phosphatidylinositol: 6.4%±0.3%; 3.2%±0.2%; 0.9%±0.5%. Phosphatidylserine: 3.3%±0.2%; 3.0%±0.2%; 0.6%±0.3%. Phosphatidylserine: 3.3%±0.2%; 3.0%±0.2%; 0.6%±0.3%. Sphingomyelin: 1.9%±0.2%; 2.1%±0.2%; 0.7%±0.1%. Sphingomyelin: 1.9%±0.2%; 2.1%±0.2%; 0.7%±0.1%. Cardiolipin: 1.6%±0.1%; 2.2%±0.2%; Not Detected. Cardiolipin: 1.6%±0.1%; 2.2%±0.2%; Not Detected. Diacylglycerol: 1.4%±0.1%; 1.2%±0.1%; 1.9%±0.1. Diacylglycerol: 1.4%±0.1%; 1.2%±0.1%; 1.9%±0.1. Quantification and statistical analysis Whole-cell pellets, and mitochondrial and LD fractions were extracted using a modified Bligh-Dyer procedure, as described previously.11,12,13 Each sample was spiked with diluted SPLASH Lipidomix Mass Spec Standard (Avanti Polar Lipids, #330707) and tetramyristoyl cardiolipin (Avanti Polar Lipids, #710332), as lipid internal standards. Untargeted lipidomics was performed using a Vanquish ultra-high-performance liquid chromatography system coupled to an Orbitrap IQ-X Tribrid mass spectrometer (Thermo Scientific). A Kinetex 2.6 μm C18 100 Å LC column (150 × 2.1 mm, #00F-4462-AN, Phenomenex) was used for chromatographic separation. The column temperature was maintained at 45°C. Mobile phase A was composed of 60:40 acetonitrile:water with 10 mM ammonium formate and 0.1% formic acid; mobile phase B consisted of 90:10 isopropanol:acetonitrile with 10mM ammonium formate and 0.1% formic acid. The LC gradient was set to 30% B at 0 minutes, 43% B at 2 minutes, 55% B at 2.1 minutes, 65% B at 12 minutes, 85% B at 18 minutes, 100% B at 20 minutes, and 30% B at 25 minutes and held until 31 minutes. The flow rate was 260 μL/minute with a 5 μL injection volume. Lipidomics data were acquired in both positive and negative modes. Data were acquired using the following source parameters: sheath gas: 40 arb; auxiliary gas: 8 arb; sweep gas: 0.5 arb; spray voltage, 3.5 kV for positive mode and 2.5 kV for negative mode acquisition. For MS/MS analysis, high-energy collisional dissociation (HCD) was performed using a stepped collision energy of 25, 30, and 35eV. Collision-induced dissociation (CID) was triggered upon detection of phospholipid polar fragments within a 10 ppm mass tolerance. A subsequent MS3 CID event was triggered when a fatty acid neutral loss was detected within a 5 ppm mass tolerance. Quality control (QC) samples were generated separately for each group (whole cell, mitochondria, and LDs; 3 separate QC samples) and for all sample types (1 additional total QC sample containing all sample types) by combining equal volumes of all replicates within each group or all samples. Repeated measurements of the QC sample were run every 5 samples with a randomized order to monitor the signal stability of the internal standards. Coefficients of variation were calculated for the QC samples, and signals exceeding 25% coefficient of variation were omitted from the analysis. Lipid species assignments were performed based on accurate mass values, retention times, and MS/MS fragmentation. For lipid quantification, signal intensities were normalized to that of the corresponding internal standard. The lipid classification was based on the LIPID MAPS nomenclature.16 The p-value for cardiolipin percentage in WCL vs mitochondria (Figure 4C) was calculated with a two-tailed Student’s t test. Limitations This protocol was optimized for AML suspension cell lines. Isolating LDs from primary AML samples may be challenging due to limited sample size and lower lipid load. Previous reports state that snap freezing of cells may lead to disruption of larger LDs of adipocytes, and samples should not be frozen to avoid shearing and breaking of LDs in fatty tissues, such as adipocytes that contain supersized LDs.6 While the shearing of LDs caused by snap freezing remains a concern, AML cells are not known to contain supersized LDs and should be less prone to LD damage by freezing. The snap-freezing step enhances LD isolation yield from AML cells, and fluorescence imaging results suggest that the integrity of isolated LDs is mostly maintained. The mitochondrial fraction will contain MAMs, therefore, ER markers (e.g., calreticulin) are supposed to be detected in immunoblotting of the mitochondrial fraction (Figure 3B). Additional clean-up, such as Percoll gradients centrifugation, may be required for applications needing highly purified mitochondria. Minor cellular membrane contamination may not be fully excluded despite repeated washing of LDs due to density. Given the dynamic nature of LD isolation and variability of lipid species across cell types, and even the same cell type under different states, it is difficult to directly compare our LD extraction protocol with previous studies. Troubleshooting Problem 1 No LDs observed after centrifugation of PNS in step 17. Potential solution •Increase sample size (i.e., 5 × 107 or higher number of cells). Increase sample size (i.e., 5 × 107 or higher number of cells). •Ensure the snap-freeze step is properly followed. Skipping this step will greatly reduce LD yield. Ensure the snap-freeze step is properly followed. Skipping this step will greatly reduce LD yield. •Check if cells contain LDs through BODIPY 493/503 staining and fluorescence imaging. Certain AML cell lines may not express much or if any LDs. Check if cells contain LDs through BODIPY 493/503 staining and fluorescence imaging. Certain AML cell lines may not express much or if any LDs. Problem 2 ER marker signal in mitochondria and MAM contamination. Potential solution •Follow the established Percoll density purification of mitochondria to isolate pure mitochondria.17 Follow the established Percoll density purification of mitochondria to isolate pure mitochondria.17 •Increase washing steps. Increase washing steps. Problem 3 LD marker PLIN2 not appearing in SDS-PAGE and immunoblotting. Potential solution •Increase the sample size input of LDs. For verification, use all the LDs collected from the workflow. Increase the sample size input of LDs. For verification, use all the LDs collected from the workflow. •Increase the concentration of SDS in the sample buffer to solubilize proteins in the LD sample. Increase the concentration of SDS in the sample buffer to solubilize proteins in the LD sample. Resource availability Lead contact Further information and requests for resources should be directed to the lead contact, Jo Ishizawa (jishizawa@mdanderson.org). Technical contact Technical questions on executing this protocol should be directed to and will be answered by the technical contact, Ran Zhao (rzhao1@mdanderson.org). Materials availability No newly generated material is available for this protocol. Data and code availability The dataset supporting this current study has not been deposited in a public repository but is available upon request. Acknowledgments This study was supported by a grant from the US Department of Defense (CA220581; J.I.), an MD Anderson Institutional Research Grant (J.I.), an MD Anderson Start-Up Funds (J.I.), a New Investigator Research Grant from the Leukemia Research Foundation (J.I.), a grant from the US Food and Drug Administration (R01 FD006118; M.A. and J.I.), Leukemia SPORE Development Research Program (P50 CA100632; M.A. and J.I.), MD Anderson’s Multidisciplinary Research Program (M.A. and J.I.), and the University of Texas Austin Start-Up Funds (A.L.). J.N. is supported by a CPRIT Research Training Award (RP210028). H.A. is now also affiliated with Department of Hematology, Institute of Science Tokyo, Tokyo, Japan. Part of the study was performed in MD Anderson’s Flow Cytometry & Cellular Imaging Core Facility, which is supported in part by the NIH through MD Anderson’s Cancer Center Support Grant (P30 CA016672). We thank Dr. Stefano Tiziani for supporting the lipidomic analysis. We thank Joseph A. Munch in MD Anderson’s Research Medical Library for editing the manuscript. Author contributions R.Z., L.Z., and J.I. designed and conceptualized the study. R.Z., L.Z., A.L., and J.I. designed the experiments. R.Z., L.Z., and S.D. conducted the experiments. R.Z., L.Z., S.D., A.L., and J.I. analyzed and interpreted the data. K.K., S.K.G., H.A., J.N., and G.L.H. provided technical and material support. R.Z. and L.Z. drafted the manuscript. A.L. and J.I. reviewed and revised the manuscript. All authors read and approved the final manuscript. J.I., A.L., and M.A. supervised the study. Declaration of interests J.I. and M.A. have filed invention disclosure forms related to the use of imipridones in cancers. Contributor Information Alessia Lodi, Email: alessialodi@utexas.edu. Jo Ishizawa, Email: jishizawa@mdanderson.org. References Associated Data Data Availability Statement The dataset supporting this current study has not been deposited in a public repository but is available upon request.