Isolation and Lipidomic Profiling of Neuronal Lipid Droplets: Unveiling the Lipid Landscape for insights into Neurodegenerative Disorders

Abstract

Recent advances have expanded the role of lipid droplets (LDs) beyond passive lipid storage, implicating their involvement in various metabolic processes across mammalian tissues. Neuronal LDs, long debated in existence, have been identified in several neural structures, raising questions about their contribution to neurodegenerative disorders. Elucidating the specific chemical makeup of these organelles within neurons is critical for understanding their implication in neural pathologies. This study outlines an improved methodology to stimulate and isolate mature LDs from cultured primary neurons, offering insights into their unique lipid-protein composition. Integrating this method with high-throughput techniques may unveil disease-specific alterations in lipid metabolism, providing avenues for potential therapeutic interventions.

INTRODUCTION

Lipid droplets (LDs) have been traditionally viewed as a passive reservoir of the esterified lipids in adipocytes. Recent advancements in imaging and biochemical techniques have, however, detected the presence of LDs in most mammalian tissues and confirmed their contribution to various metabolic processes ranging from energy production, and membrane synthesis to signaling pathways (see Farese et. al.⁷ and Olzmann et. al.⁴ for a comprehensive review). Neurons, with their complex polarized morphology and high energy demand, are not exempt from the importance of lipid metabolism, and lipids are recently gaining attention as a potential key player in the functioning and malfunctioning of neurons. Contrary to the prevalent belief that LDs are non-existent in neurons, it has been detected in motor neurons of drosophila⁸, axons of Aplysia⁹, cerebral cortex of mice brain^10,11, and neuroblastoma-derived cell lines¹².

Neurodegenerative diseases, like Alzheimer’s disease, Parkinson’s disease, and Hereditary spastic paraplegia are multifactorial disorder that shares a common theme of progressive degeneration of neurons, but the precise etiology of these disorders remains elusive. Interestingly, abnormal accumulation of LDs in neurodegenerative disorders^10,13,14 has raised the question of whether the disability of triglyceride mobilization from LDs contributes to the cellular dysfunction and deterioration of neurons in specific brain regions of patients. A large body of evidence suggests that neuronal LDs might serve as the site for accumulating both neuroprotective and toxic lipid, and protein species^15,16. Moreover, we recently discovered that electrical activity triggers DDHD2 to facilitate transfer of fatty acids (FAs) from the triglyceride-rich LDs to mitochondria for ATP synthesis in dissociated hippocampal neurons. Consequently, comprehending the exact contributions of the lipids and proteins associated with LDs in neuronal function demands their isolation without any contamination of other cellular organelles. Here, we present a refined and highly reproducible method to facilitate the formation of mature LDs in cultured primary neurons and their purification using the sucrose density-gradient method.

Advantages

To the best of our knowledge, this is the first study where we successfully induced LD formation in dissociated neurons and purified them using a density-based floatation method. We carried out lipidomic studies on purified neuronal LDs to identify previously unknown lipids associated with them. The present study provides a snapshot of lipid-protein milieu linked to neuronal LDs. A successful integration of the method with high-throughput omics techniques provides an excellent opportunity to identify targets of disease-specific alterations in lipid and protein composition of neuronal LDs.

Applications

Neurons are polarized cells, whose architecture can be divided into soma, dendrites, and axons. Axons extend further away from the soma and support the release of neurotransmitter, that in turn requires a constant supply of ATP at nerve terminals. We recently discovered that the triglyceride lipase DDHD2 is highly enriched at the synaptic terminals and blocking its activity leads to local LD accumulation. These data suggests that neurons constantly use TGs to maintain synaptic functions. Pharmacological inhibition of the TG lipase results in the formation of near micron size LDs in dissociated cortical neurons, which in turn allow the purification of organelle suitable for analysis of lipid and protein composition.

Limitations

Lipids on the monolayer membrane of LDs are synthesized at the ER leaflets or on-site of maturing LDs. Proteins, on the other hand are trafficked from the ER membrane via membrane bridge or from the cytosol by amphipathic α-helices. Certain lipids, and cytosolic proteins in particular, are susceptible to degradation or dissociation during the physical purification process and might not reflect in the downstream omics studies. Furthermore, the detailed composition of LDs depends on their size, and therefore analytical approaches that report population averages, such as typical mass spectrometry, necessarily miss size-based variation in LD content. Therefore, further biochemical and immunocytochemical methods applied on either intact neurons or isolated LDs are warranted to discern the native and size dependent composition and functions of neuronal LDs.

MATERIALS

• Biological

  Animals: Wild-type rats and mice used for the experiments were of the Sprague-Dawley strain (Charles River code 400, RRID: RGD_734476), and CD1 backgrounds respectively. !CAUTION All animal-related experiments were performed in accordance with protocols approved by the Weill Cornell Medicine and Yale University School of Medicine IACUC.

• Chemical

  1X PBS: (ThermoFisher, Cat. # 10010031)

  KLH45: Sigma, Cat. # SML1998

  PIPES: Sigma, Cat. # P6757

  EGTA: Sigma, Cat. # E3889

  MgSO₄: Sigma, Cat. # M3409

  KOH: Sigma, Cat. # 221473

  Complete Protease Inhibitor Cocktail: Roche, Cat. # 45-11697498001 ΔCritical Prepare fresh before use.

  PMSF: Sigma, Cat. # P7626 !CAUTION PMSF is acutely toxic. Use disposable gloves while handling.

  Sucrose: Sigma, Cat. # S0389

  Glyceryl trioleate: Sigma, Cat. # T7140

  CellMask^™ plasma membrane stain: Invitrogen, Cat. # C10045

  BODIPY: Thermo Scientific, Cat. # D3922

  TLC plates: Millipore, Cat. # 1.05721.0001

  CuSO₄: Supelco, Cat. # 102791

  H₃PO₄: Sigma, Cat. # 345245

  Organic solvents: Chloroform (Sigma, Cat. # 288306); Methanol (Sigma, Cat. # 34860); n-Hexane (Sigma, Cat. # 34859); Diethyl ether (Sigma, Cat. # 309966); Acetic acid (Sigma, Cat. # 320099) !CAUTION The organic solvents are highly toxic and unstable explosive. Avoid contact with flames and always handle them in the fume hood. Dispose them in accordance with the institutional and local guidelines.

  Lipids: Glyceryl trioleate (Sigma, Cat. # T7140); SPLASH^® LIPIDOMIX Mass Spec standard (Avanti, Cat. # 330707-1EA) ΔCritical Lipids are unstable and prone to oxidation under atmospheric conditions. So parge the lipids with N₂ gas before the storage. Avoid using plastic or polymer tubes for storage of the lipids. Use glass vials for the storage of lipids dissolved in chloroform or any other organic solvent.

• Equipments

  Cell disruption nitrogen chamber: Parr Instrument, Cat. # 4639

  Ultracentrifuge: Beckman Coulter, Optima L-100 XP ultracentrifuge

  SW40 Ti rotor: Beckman Coulter, Cat. # 331302)

  SW40 14-ml ultracentrifuge tube: Beckman Coulter, Cat. # 331374

  Needles and plastic syringes

  Hamilton syringes: 1-10 μl 10-100 μl and 0.1-1 ml

  Microcapillary pipets: Kimble Glass, Cat. # 71900-25

  Particle analyzer: Anton Paar Litesizer 500 equipped with Kalliope 2.8.0 software

  Microscope: Zeiss LSM880 laser scanning confocal microscope

  ImageJ software: https://imagej.nih.gov/ij/

  Agilent Poroshell 120 (EC-C18, 2.1 x 100 mm, 2.7 μm, 1000 bar) HPLC column

  Agilent 1290 Infiniti II LC system

  Agilent 6546 LC/Q-TOF mass spectrometer

PROCEDURE

• Primary culture of Cortical neurons

  • 1
    Coat 10 nos. culture dishes of 150 mm size with 50 μg/ml poly-d-lysine for 16 hours before starting the procedure of cortical culture.
  • 2
    Dissect out both the cortices of brain from postnatal day 0 (P0) mouse or rat pups and place in ice cold HBSS.
    Δ Critical step: Dissect out and remove the parts of the brain other than cortex to ensure the purity of the cortical neurons. Other regions of the brain such as hippocampus, cerebellum etc. can also be cultured separately using specific protocols for a comparative study.
  • 3
    Slice the cortices into small pieces (^~1 mm) using a dissection knife and allow digestion in HBSS solution containing papain (20 u/ml) and DNase (20 μg/ml) for 15-20 minutes at 37 °C.
  • 4
    Triturate the digested tissue slices by pipetting back and forth with 1 ml tip and filter the debris out using a 40 μm cell strainer.
    Δ Critical step: Cell strainer with wider mesh size may allow cell debris to pass through, while ones with narrow mesh size will restrict the passage of dissociated cortical cells. Wet the strainer with 5 ml HBSS before pouring cell suspension to prevent clogging.
  • 5
    Wash the dissociated cortical cells with HBSS and plate in the coated petri plates at 12,000 - 20,000 cells/cm² density.
  • 6
    Exchange the plating media after 3 hours with neurobasal media containing 2% B-27 and 0.5 mM L-glutamine.
  • 7
    Maintain the cortical neurons at 37 °C in humidified incubator containing 5% CO₂.
  • 8
    Replenish 30% of the total volume with prewarmed complete neurobasal media at DIV4, 7 and 14.
• Induction of Lipid droplets in cortical neurons

  • 9
    Administer KLH45 (5 μM) into the cortical neurons cultured in complete neurobasal media on DIV-14 to DIV-21 and return to the incubator for 24 hours.
    Δ Critical step: To confirm the action of KLH45 and subsequent induction of LDs, stain the neurons parallelly cultured in chamber slide with BODIPY 493/503 (follow step 25) and image with fluorescence or confocal microscope.
• Disruption of cortical neurons

  • 10
    Rinse the neurons twice with ice cooled 1X PBS.
  • 11
    Pour 400 μl 0.9M MEPS buffer (ice cooled supplemented with 1X protease inhibitor cocktail and 0.2 mM PMSF) to the layer of cortical neurons and scrap using a sterile plastic cell-scrapper.
  • 12
    Transfer the cells to a pre-cooled cell disruption vessel. Dissolve oxygen-free nitrogen gas to the cytosol for 20 minutes under high pressure (750 pounds per square inch, psi).
  • 13
    Connect the release port to a collection chamber on ice by silicon pipe and release the pressure slowly to collect the disrupted neurons.
    Δ Critical step: Mount 30 μl lysed neurons between two glass slides (thickness: 0.18 mm) and observe onto light microscope at 40X magnification to confirm the lysis of cortical neurons. Higher pressure leads to complete homogenization of the cells as well as membranous organelles, therefore optimize the pressure and time for N₂ cavitation is required to achieve near complete lysis of neurons.
• Separation of intact neurons, ruptured cell membrane and nucleus

  • 14
    Transfer the lysed neurons into a fresh 15 ml falcon tube and centrifuge at 1800 g for 10 minutes at 4 °C.
  • 15
    Collect the supernatant (called post-nuclear supernatant or PNS) in a fresh tube and place on ice until the gradient is set-up. An aliquot of 500 μl PNS can be retained on ice or snap frozen in liquid nitrogen to be use as control for purification efficiency.
• Purification of LDs by density-based floatation

  • 16
    Transfer 2.5 M MEPS at the bottom of a clear SW-41 centrifuge tube and overlay PNS over it. Gently mix the two layers with cut-tip pipette.
    Δ Critical step: The ratio of MEPS (2.5 M) and PNS (prepared in 0.9 M MEPS) should be such that the final molarity of the mixture is ≥ 1.5. Split the PNS and set up gradient in two SW-41 tubes that can be easily balanced against each other during ultracentrifugation.
  • 17
    Overlay the bottom layer with 2 ml each of 1.2 M and 0.5 M MEPS from the bottom to top carefully without disturbing the interfaces.
    Δ Critical step: Use a smooth pipette with cut-tips to layer the sucrose gradients. A messy layering of the sucrose solutions will result in the incomplete isolation of LDs and contamination of other organelles.
  • 18
    Overlay 2-3 ml of 0 M MEPS over the layer of 0.5 M MEPS.
    Δ Critical step: The volume of the top layer of 0 M MEPS should be up to the maximum permissible height of the SW41 tube. Larger the volume of the top layer, purer will the isolated LDs. Salts like 100 mM sodium carbonate can be added to the top layer to remove the loosely bound non-specific proteins to the LDs³⁵.
  • 19
    Transfer the SW-41 tubes to pre-cooled (at 4°C) SW-41-Ti swinging buckets and centrifuge at 1,74,000 g for 1.5 hours at 4 °C.
  • 20
    Transfer the buckets on the ice and collect the floating top layer of LDs using 25 gauge slanted needle attached to a 1 ml syringe.
    Δ Critical step: Collect the LD sample with the minimum possible volume of 0 M MEPS buffer without disturbing the underlaying layers containing undesired membranes and cytosolic proteins. If the LDs appear dilute, centrifuge the sample at 14,000 g at 4 °C for 10 minutes and carefully remove the buffer at the bottom using a fine needle without disturbing the floating LD layer on the top.
  • 21
    Snap freeze aliquots of the isolated LDs samples in liquid nitrogen and store at −80 °C to preserve the lipids and LD associated proteins until further use.
    Δ Critical step: The LDs samples can be used within 6 months after preparation. Always thaw the LD samples on the ice.
• Dynamic light scattering

  • 22
    Dilute the LD samples in ultrapure water (1:50) and load onto dynamic light scatter particle analyzer (Anton Paar Litesizer 500).
  • 23
    Use the Kalliope 2.8.0 software to acquire the size distribution of LDs following manufacturer’s guidelines.
• Confocal micrography

  • 24
    Prepare a staining solution containing 0.5X CellMask^™ PM Stain (from 1000X stock) and 1X BODIPY 493/503 (from 1 mg/ml, i.e. 1000X stock) in 1X PBS.
  • 25
    Transfer 10 μl of LDs sample or 1:100 diluted membrane samples to two separate tubes containing freshly prepared staining solutions and incubate for 10 minutes at room temperature.
  • 26
    Mount the stained samples on clean glass slides under coverslips and seal the edges with clear nail polish.
    Δ Critical step: Washing of the unbound CellMask^™ and BODIPY493/503 stains are not required. The optimum concentration of the stains may vary between 0.5X to 2X depending on the size and density of the LDs.
  • 27
    Image the samples on confocal microscope equipped with appropriate lasers for CellMask^™ and BODIPY493/503 stains and analyze the images using ImageJ software.
• Thin layer chromatography

  • 28
    Thaw an aliquot (100-150 μl) of purified LD sample on ice, adjust the volume to 500 μl with ultrapure water and transfer it to a clean glass tube.
  • 29
    Add 2 ml methanol and 1 ml chloroform to the tube, vortexed the mixture for 1 minute and place it at 4 °C for 14-16 hours.
  • 30
    Add 1 ml each of chloroform and water to the tube, vortex it again for 1 minute and allow it to sit undisturbed at room temperature for 30 minutes, or until the organic phase (lipid containing lower layer) is distinctly separated from aqueous phase (protein containing upper layer).
    Δ Critical step: If the protein content of sample is too high, the phases are not separated well under gravity. To separate such samples, prepare the sample in glass centrifuge tube and spin at 100-200 g for 10 minutes at 4 °C.
  • 31
    Transfer the organic phase to another clean glass tube and evaporate the solvent under stream of oxygen-free nitrogen gas.
  • 32
    Resuspended the dried lipids in 30 μl chloroform and spot them onto a silica TLC plate that has been pre-cleaned with chloroform.
  • 33
    Separate the spotted lipids by two-solvent systems A and B, first half in solvent system A [n-hexane/diethyl ether/acetic acid (60:40:4)] followed by solvent system B [n-hexane/diethyl ether (59:1)]. Air-dry the solvent from TLC plate in a chemical hood.
  • 34
    Visualize the lipids by spraying the TLC plate with a solution of 10% CuSO₄ and 8% H₃PO₄ followed by charring the plate for 15-20 minutes in a preheated oven at 180 °C.
    Δ Critical step: To monitor specific lipids in the sample, spot standard lipids on separate spots on the level of other samples on the same plate. In case of any suspicion of lipid contamination in solvents, include a spot of solvent blank.
  • 35
    Capture the image of the TLC plate using a grayscale scanner and quantify the band intensities along the lipid migration lengths using ImageJ software.
• Extraction of lipids for LCMS/QToF

  • 36
    Thaw LD samples on ice and transfer to 2 ml click-seal microcentrifuge tube.
  • 37
    Add MTBE (methyl-tert-butyl ether) and spike the samples with Splashmix as the internal standards.
  • 38
    Sonicate the samples for 15 minutes at ice-cooled sonication bath.
  • 39
    Centrifuge the sample at 1000 g for 10 minutes at 4 °C to separate the lipid-containing phase from the precipitated pellet of proteins.
  • 40
    Transfer the lipid-containing organic phase to 1.5 ml microcentrifuge tube and evaporate the solvents using a vacuum concentrator without heating.
  • 41
    Resuspend the extracted lipids in n-butanol.
    Δ Critical step: This step should be performed in cold room maintained at 4 °C. Any bubble in the LC vials should be removed by gently flickering the tube with your fingers.
• LCMS/QToF

  • 42
    Load 3μl of this mixture onto an Agilent Poroshell 120 (EC-C18, 2.1 x 100 mm, 2.7 μm, 1000 bar) column using an Agilent 1290 Infiniti II LC.
  • 43
    Use Mobile phase A (60% Acetonitrile, 40% H₂O, 7.5 mM Ammonium Acetate) and Mobile phase B (90% IPA, 10% Acetonitrile, 7.5 mM Ammonium Acetate) to separate peaks for detection on an Agilent Quadrupole Time-Of-Flight 6546 mass spectrometer.
  • 44
    The gradient started at 85% (15% B) and decreased to 70% A over 2 minutes and then 52% A over 30 seconds. The gradient then slowly decreased to 18% A over 12.5 minutes and then 1% A in one minute and these concentrations held for 4 minutes.
  • 45
    Allow the gradient to restore at 85% A and wash the column for next 5 minutes.
  • 46
    Identify the lipids using MS-DIAL 4.7 (Ref. ³⁶) and quantify each lipid species using the moles of corresponding class-specific lipid standard present in Splashmix.

Recipe

• KLH45: Add 2.18 ml DMSo to a vial of 5 mg KLH45 to prepare a 5 mM stock solution. Vortex for 5 min and store the aliquots at −20 °C. The solution is stable up to 6 months under strictly maintained temperature.

• MEPS buffers: Prepare sucrose solutions of 0.9 M, 2.5 M, 1.2 M, 0.5 M & 0 M (M indicates molarity of sucrose) in ultrapure water. Add PIPES (35 mM, pH 7.2), EGTA (5 mM) and MgSO₄ (5 mM) to the sucrose solutions and maintain the pH to 7.2. The filtered stocks solutions can be used up to 6 months after preparation, if stored at 4°C. Add freshly prepared protease inhibitor cocktail (1X) immediately before starting the experiment. Δ Critical PIPES is not readily soluble in water at pH ≤ 7. Prepare a 1 M stock solution of PIPES by dissolving it in water and, while stirring, adjust the pH to 7.2 using KOH pellets. To facilitate the solubilization of 2.5 M sucrose, immerse the tube in a water-bath set at temperature 70 °C.

• TLC solvent systems: Prepare 208 ml TLC ‘solvent A’ by mixing 120 ml n-hexane, 80 ml diethyl ether and 8 ml acetic acid (ratio 60:40:4); and 240 ml TLC ‘solvent B’ by mixing 236 ml n-hexane and 4 ml diethyl ether (ratio 59:1). Δ Critical TLC ‘solvent A’ and ‘solvent B’ should be prepared and stored in narrow-mouth glass containers with glass stopper. It can be used within for 6 months unless exposed to air that may change the composition of solvents.

• Extraction solvent: Add 5 ml methanol to 15 ml methyl tert-butyl ether (MTBE) to prepare 20 ml extraction solvent. Δ Critical The solvent is highly volatile and therefore should not left open in air.

• Mobile phase A: Prepare the required volume of the phase A by mixing the following: 60% LCMS-Grade Acetonitrile, 40% MilliQ Water and 7.5 mM Ammonium Acetate.

• Mobile phase B: Prepare the required volume of the phase B by mixing the following: 90% LCMS-Grade IPA, 10% LCMS-Grade Acetonitrile and 7.5 mM Ammonium Acetate.

Troubleshooting

Problem	Possible reasons	Potential Solutions
Underdeveloped neuronal processes	1. Neurons were not healthy 2. Culture used earlier than DIV-14	1. Plate neurons at adequate density and ensure the quality of culture media 2. Allow neurons in culture for DIV-14 or more
Failure of LD induction	KLH45 was degraded	Reconstitute fresh KLH45, store at appropriate temperature in small aliquots and avoid freeze-thaw cycle
Incomplete lysis of neurons	Pressure of nitrogen in cell disruption vessel was low	Increase the pressure in nitrogen vessel
Low count of LDs	1. Incomplete lysis of cells 2. Gradients layers of high volume 3. Loss during LD collection	1. Ensure complete lysis of neurons 2. Use smaller tube for small samples 3. Tube slicer can be used
Aggregation of LDs	Electrolyte in the solution was low	1. Check the concentration and pH of buffers 2. Additional floatation step in low concentration of glycerol, 0.1-1 M NaCl or 100 mM Na2CO3 (pH 11.5)35
Disrupted LD membrane	Homogenization was too harsh	Reduce the nitrogen pressure
Contamination of membrane	1. Wide diameter tube used 2. Density gradient was messy	1. Preferably use tubes of narrow diameter for prep. with small number of neurons 2. Use smooth pipette with wide orifice tips on a sturdy platform
High phospholipid in TLC	Membrane contamination	Follow the above steps and ensure the quality of LDs before lipid extraction
Smearing of protein bands in western blot	1. Proteins were degraded 2. Incomplete extraction of lipids during protein precipitation	1. Add protease inhibitors to all MEPS buffers and set-up the gradient on ice 2. Increase the centrifugation speed to separate the lipid phase from proteins

Discussion

Our knowledge about lipid droplets has evolved substantially in past few decades, transitioning from a passive depot of lipids to the dynamic organelle involved in diverse cellular functions. Recent studies establish dysregulation of lipids and excessive accumulation LDs as hallmark of metabolic disorders such as obesity³⁷, diabetes³⁸, atherosclerosis³⁹ and fatty liver disease⁴⁰. However, despite the multiple reports of lipid accumulation in neurodegenerative disorders, the proteo-lipid identity of the organelle in neurons remained completely unknown. The establishment of a methodology facilitating the isolation of lipid droplets from neurons stands as a pivotal advancement, allowing focused exploration into the role of these organelles in neural functions. This method offers an opportunity to study lipid dynamics in neuronal metabolism, shedding light on their contributions to energy production and signaling, critical for comprehending neurodegenerative conditions. The successful deployment of the isolation technique also holds the promise in identifying novel biomarkers and disease-specific alterations, potentially elucidating the mechanisms underlying neurodegenerative disorders characterized by aberrant LD accumulation. However, acknowledging potential constrains such as size-dependent variations and alteration in lipid-protein compositions during purification process, complementary approaches like FACS, SILAC, and proximity-labelling may be indispensable in addressing size-dependent variations in LD protein complexes. Overall, this methodology marks a significant stride in neurobiology and lipid research, offering an entry point to unravel the complexities of LDs in neural contexts and their implications in neurological disorders.

Fig.1|.

Induction of lipid droplets in dissociated cortical or hippocampal neurons. a, Schematic illustration of protocol steps to induce formation of LDs in the cortical neurons and confirmation by microscopy and thin layer chromatography (TLC). The cortical neurons from rodent brain (coronal section shown) cultured on poly-d-lysin coated dishes are treated with KLH45 at DIV 14-21. b, Fluorescence micrographs of control and KLH45 treated cortical neurons after BODIPY staining. Lipids droplets are apparent as globular structures both in the soma (red arrowheads) and neuronal processes after KLH45 induction. scale bar: 20 μm. c, TLC showing the dose dependent accumilation of triglyceride (TGs) in the cortical neurons after inhibition of neuron-specific triglyceride lipase with KLH45. Standard TG (glyceryl trioleate) and free fatty acid (oleic acid) are shown on the right side of yellow line as references. Line intensity profile of control and two KLH45 tretaed (2.5 μM and 5 μM) neurons is shown in the right panel. d, Electron micrograph showing accumulation of LDs (red arrowhead) in the vicinity of synaptic vesicles at a synaptic terminal of a hippocampal neuron. scale bar: 500 nm.

Fig.2|. Isolation of cortical lipid droplets and their physical characterisation.

a, Schematic illustration for the steps for the isolation of lipid droplets from KLH45 induced cortical neurons using sucrose step-density gradient. The LDs are collected from the top layer of sucrose density gradient as shown in step 4. b, Laser scanning confocal micrographs of the purified LDs stained with BODIPY (triglyceride rich neutral core) and CellMask (phospholipid rich monolayer), scale bar: 5 μm. c, Size distribution of the cortical lipid droplets analyzed by dynamic light scattering instrument, Anton Paar Lifesizer 500. The size of the LDs varies from 200 nm to 5 μm. d, Thin layer chromatography with the lipid droplets isolated from cortical neurons. Triglyceride (Glyceryl trioleate) and cholesterol are shown on the right side for reference. d, Line intensity profile plotted for the triglyceride and cholesterol (shown by red and blue lines in d). The intensity profile indicates that the LDs are highly enriched with triglycerides.

Fig.3|. Lipidomic analysis purified LDs from cortical neurons and comparision with whole neuron lipids.

a, Distribution of lipid classes identified by LCMS performed with the lipids extracted from LDs and whole neuron. The lipid classes include LPG: Lysophosphatidylglycerol, EtherPG: Ether-linked phosphatidylglycerol, PG: Phosphatidylglycerol, LPI: Lysophosphatidylinositol, EtherPI: Ether-linked phosphatidylinositol, PI: Phosphatidylinositol, OxPS: Oxidized phosphatidylserine, LPS: Lysophosphatidylserine, EtherPS: Ether-linked phosphatidylserine, PS: Phosphatidylserine, OxPE: Oxidized phosphatidylethanolamine, EtherPE: Ether-linked phosphatidylethanolamine, PE: Phosphatidylethanolamine, LPC: Lysophosphatidylcholine, EtherPC: Ether-linked phosphatidylcholine, PC: Phosphatidylcholine, CE: Cholesteryl ester, Cer: Ceramide, SM: Sphingomyelin, DG: Diacylglycerol, OxTG: Oxidized triglyceride, TG: Triglyceride. The graph reprensents mole % (of total lipid) of indicated lipid classes in the two samples. b-d, Bar graphs represent the top three enriched lipids species (TG, PC and PI) identified in neuronal lipid droplets compared with whole neuron lipids. TG and PC represents the two major lipids in the LD fraction. Data is represented as mean ± SE. Complete lipid profile is attached as supplementary information with the manuscript.

Table 1:

Methods for lipid droplets purification

Method	Principle	Biological source	Size of LDs	Ref.
Differential Centrifugation	Size and density	CHO K2, Liver, Brown adipose tissue	0.5-10 μm	17,18
Density Gradient Centrifugation	Buoyant density	Liver, 3T3-L1 adipocytes	0.5-5 μm	19,20
OptiPrep™ Density Gradient Medium	ER kit based	Liver	0.27-6.37 μm	21

Table 2:

Purification of lipid droplets from animal tissues

Tissues	Species	Buffers	Size of LDs	Ref.
Liver	Rat	KCl; NaCl, Sucrose	0.5-2 μm	18,20
Brown adipose tissue	Mouse	Tricine, Sucrose	0.5-20 μm	22
White adipose tissue	Mouse	Tris, Na3VO4 NaF, EDTA	NA	23
Skeletal muscle	Mouse	Buffer A: Tricine, Sucrose Buffer B: HEPES, KCl, MgCl2	< 1 μm	24
Cardiac muscle	Mouse	Tris-EDTA-Sucrose	< 1 μm	25
Mammary tissue	Mouse	KH2PO4, MgCl2, Sucrose	0.5-6 μm	26,27
Intestinal mucosa	Mouse	EDTA	0.8 ± 0.4 μm	28
Kidney (Renal papillae)	Pig, Rabbit, Dog, Rat	Sucrose	0.4-0.8 㮼m	29,30
Kidney (Renal medulla)	Rat, Rabbit	Sucrose	~0.5 μm	31,32
Placenta	Human	Tris-EDTA-Sucrose; Buffer A: Tricine, Sucrose Buffer B: HEPES, KCl, MgCl2	< 1 μm	33,34

NA: not analyzed

Historic overview.

Lipid droplets as an organelle remained ignored for almost a century after its discovery in 19^th century by Richard Altmann and E.B Wilson^1,2. A series of discoveries in 1990s outlined the importance of their role in cellular functioning and metabolic disorders reignited interest and prompted the development of techniques for LD isolation from live tissues^3,4. Irrespective of their source, LDs mainly constitute a core of triglycerides and sterol esters that makes them highly buoyant in aqueous medium enabling their separation from other organelles by floatation. Initial attempts to systematically isolate LDs from plant-based sources date back to 1970s^5,6. Subsequent refinement in the methodology (See table 1) extraction of the purest form of LDs from animal sources for lipidomic and proteomic studies devoid of any membrane contamination (See table 2). Isolation of pure LDs requires careful disruption of cell membrane without rupturing internal membranous organelles. Moreover, the buoyancy of the LDs varies with the size of LDs (100 nm to more than 20 μm in some cell types) and their association of organelles (ER, mitochondria, endosomes etc.) prompted modifications in LD purification protocols to avoid co-isolation of undesirable membrane fragments.

Data availability

The source data related to the findings of the manuscript are provided as supplementary information with this manuscript and can be used for scientific purposes with permission of corresponding authors.