Summary

Analysis of Lipid Signaling in Drosophila Photoreceptors using Mass Spectrometry

Published: March 07, 2022
doi:

Summary

The manuscript presents versatile, robust, and sensitive mass spectrometry protocols to identify and quantify several classes of lipids from Drosophila photoreceptors.

Abstract

The activation of phospholipase Cβ (PLCβ) is an essential step during sensory transduction in Drosophila photoreceptors. PLCβ activity results in the hydrolysis of the membrane lipid phosphatidylinositol 4,5 bisphosphate [PI(4,5)P2] leading ultimately to the activation of transient receptor potential (TRP) and TRP like (TRPL) channels. The activity of PLCβ also leads subsequently to the generation of many lipid species several of which have been proposed to play a role in TRP and TRPL activation. In addition, several classes of lipids have been proposed to play key roles in organizing the cell biology of photoreceptors to optimize signaling reactions for optimal sensory transduction. Historically, these discoveries have been driven by the ability to isolate Drosophila mutants for enzymes that control the levels of specific lipids and perform analysis of photoreceptor physiology in these mutants. More recently, powerful mass spectrometry methods for isolation and quantitative analysis of lipids with high sensitivity and specificity have been developed. These are particularly suited for use in Drosophila where lipid analysis is now possible from photoreceptors without the need for radionuclide labeling. In this article, the conceptual and practical considerations in the use of lipid mass spectrometry for the robust, sensitive, and accurate quantitative assessment of various signaling lipids in Drosophila photoreceptors are covered. Along with existing methods in molecular genetics and physiological analysis such lipid is likely to enhance the power of photoreceptors as a model system for discoveries in biology.

Introduction

Phototransduction in Drosophila is mediated by a G-protein-coupled PLCβ cascade leading to the activation of the light-activated channels TRP and TRPL1. PLCβ hydrolyzes the membrane-bound phospholipid, phosphatidylinositol 4,5 bisphosphate [PI(4,5)P2] and generates diacylglycerol (DAG), and inositol 1,4,5 trisphosphate (IP3). DAG is then phosphorylated by DAG-kinase to generate phosphatidic acid (PA). Subsequently, through a series of reactions that involve the generation of lipid intermediates, PI(4,5)P2 is regenerated2. Several components of this PI(4,5)P2 cycle have functions in Drosophila photoreceptors. The mechanism by which PLC activation leads to TRP and TRPL channel gating remains unresolved. However, several lines of evidence suggest that lipid intermediates generated by PI(4,5)P2 hydrolysis may mediate this process3. Thus, it is very important to identify and quantify these lipid intermediates to shed light on the mechanism of activation of TRP and TRPL in Drosophila phototransduction. In addition to their role in phototransduction per se, lipids also play several important roles in the cellular organization of photoreceptors (reviewed in3). Understanding these functional roles of lipids is aided by the ability to detect and quantify their levels in vivo. This article provides an overview as well as protocols for the choice and implementation of methods for quantifying lipids from Drosophila photoreceptors.

Historically, lipid analysis consisted of fractionation by chemical classes followed by the analysis of individual classes. To successfully identify and quantify lipids, many analytical methods have been developed that are either targeted or non-targeted lipid analyses4,5,6,7,8,9,10. The targeted analysis focuses on known lipids and utilizes a specific method with high sensitivity for the quantitative analysis of these specific lipids. Non-targeted lipid analysis aims to detect many lipid species in a sample simultaneously. These analytical methods include thin-layer chromatography (TLC)4, gas chromatography (GC)5, liquid chromatography (LC)6, enzyme-linked immunosorbent assays (ELISA)7, nuclear magnetic resonance (NMR)8, radionuclide labeling, and mass spectrometry (MS)9,10. Although radioactive labeling is a sensitive method for the detection of lipids and can be used in the context of cultured cells, its use in the analysis of lipids in intact organisms such as Drosophila is challenging due to safety considerations of radiolabeling live flying animals. The other challenge with radioactive labeling is that it is dependent upon labeling all precursor pools to near-equilibrium and this can be difficult in the context of in vivo models.

Comprehensive lipid analysis using MS is a recent advancement made possible by the development of modern MS technologies11. MS-based analysis of lipids offers several advantages, these include small sample sizes, applicability to samples from animal models, high sensitivity, specificity as well as high throughput. In particular, the extensive use of electrospray ionization has lead to an improvement in the performance of MS for lipid analysis. The improvement of mass analyzers in mass spectrometers, including the combination of different mass analyzers, has added extra advantages and the development of a high-resolution mass analyzer has revived lipid studies11. MS-based analysis characterizes lipid molecules in two major ways: (1) Top-down lipidomics where MS experiments are aimed at rapid quantitative characterization of global changes within the lipidome and rely solely on accurate masses of intact lipid precursors12; (2) Bottom-up lipidomics which quantifies individual molecular species by detecting characteristic structural fragment ions using tandem MS13,14.

Overall, the analysis of lipids in Drosophila photoreceptors consists of two steps: first, extraction of lipids from eye/head tissue and second, analysis of extracted lipids by MS. This can be performed using one of the following methods: separation of lipids by liquid chromatography (LC) coupled to MS or without chromatographic separation by using shotgun lipidomics/direct infusion MS (DIMS). Both lipid profiling approaches are based on the use of electrospray ionization MS (ESI-MS) and have proven to be sensitive, quantitative, and efficient10,15. In DIMS analysis, the identification of lipids is based on precursor ion mass, neutral loss scans, and lipid class-specific signatures14,16,17,18. While this approach combines the speed of analysis and robustness, ion suppression of low abundance lipids due to the presence of high abundance and highly polarizable lipids in the sample cannot be avoided. Thus, low abundant lipids such as phosphoinositides and PA are often not detected or poorly detected by standard DIMS platforms, due to ion suppression among other reasons19,20. Liquid chromatography separation prior to MS (LC-MS) can help overcome ion suppression as well as focus the analysis of specific classes or species of lipid of interest21.

In this article, the steps involved in quantifying the major lipids of interest in Drosophila photoreceptors are described. In this regard, three different MS approaches have been optimized: (1) DIMS using a high-resolution mass spectrometer, (2) post-derivatization reverse-phase liquid chromatography-MS (RPLC-MS) using a triple quadruple mass spectrometer, and (3) normal phase liquid chromatography-multiple reaction monitoring-enhanced product ion scan-MS (NPLC-MRM-EPI-MS) using a triple quadruple mass spectrometer with an enhanced product ion scan function. The choice between these methods is dictated by the specific research questions under investigation. For a global description and analysis of all kinds of glycerophospholipids involved in phototransduction, DIMS should be used. However, it should be noted that in this approach low polarizable and low abundance glycerophospholipids such as phosphoinositides are not likely to be detected22. To detect these low abundance lipids, post-derivatization RPLC-MS should be performed. Using this approach, we have successfully detected and quantified phosphatidic acid (PA)23,24,25, phosphatidylinositol (PI), phosphatidylinositol 5 phosphate (PI5P)26, phosphatidylinositol 4 phosphate (PI4P), and PI(4,5)P227. DIMS and post derivatization RPLC-MS methods generate lipid class level information. For example, using these two methods one can quantify PA (34:2) whose m/z is consistent with multiple molecular species including: (i) PA (16:0/18:2), (ii) PA (18:2/16:0), (iii) PA (16:2/18:0), (iv) PA (18:0/16:2), (v) PA (16:1/18:1), (vi) PA (18:1/16:1), (vii) PA (14:2/20:0), and (viii) PA (20:0/14:2). Using these methods, one cannot get information about the fatty acyl chain composition of the PA (34:2) present in the sample. This challenge can be overcome by a hybrid MS method that couples LC-separation, multiple reactions monitoring (MRM), and enhanced product ion scan (EPI). This method is both sensitive and quantitative and allows: direct measurement, i.e., without any pre-labeling or post-processing of the sample, and establishes the molecular species with exact fatty acyl chain information25. Using this method, we have identified a large number of molecular species of PA and determined the exact composition of fatty acyl chains at SN1 and SN2 of the glycerol backbone. This approach will be useful when analyzing the function of specific molecular species of any class of lipid in photoreceptors. Detailed protocols presented here for each of these types of analyses can be adapted to other signalling lipids relevant to Drosophila photoreceptor function (for schematics see Figure 1). It should be noted that detailed methods for lipidomics analysis of several other classes of lipids (not covered in this article), have been described elsewhere. These include ceramides28,29, sphingolipids30,31, neutral lipids such as diglycerides and triglycerides32,33, and sterols15,33. In some cases, methods for analyses of these lipids have been described for Drosophila larval tissues and could be adapted for use in photoreceptors.

Protocol

1. Rearing flies and preparation of chemicals

  1. Rear flies (Drosophila melanogaster) on standard fly food in an incubator with 50% relative humidity at 25 °C without internal illumination. Prepare fly food by adding 80 g/L of corn flour, 20 g/L of D-glucose, 40 g/L of sucrose, 8 g/L of agar, 15 g/L of yeast extract, 4 mL of propionic acid, 0.7 g/L of TEGO (methyl para hydroxybenzoate), and 0.6 mL of orthophosphoric acid as described in34 and are also available at https://bangalorefly.ncbs.res.in/drosophila-media-preparation. Grow one set of flies, post-eclosions, in a cooled incubator maintained at 25 °C with continuous white light illumination of ~2,000 lux.
  2. Prepare phosphoinositide elution buffer (PEB) by adding chloroform:methanol:2.4 M hydrochloric acid in a ratio of 250:500:200 (vol/vol/vol). Prepare lower phase wash buffer (LPWS) by adding methanol:1 M hydrochloric acid:chloroform in a ratio of 235:245:15 (vol/vol/vol). Prepare post-derivatization wash solution by adding chloroform:methanol:water in an 8:4:3 (vol/vol/vol) ratio.
    NOTE: Use MS grade solvents and chemicals. Do not store the PEB and LPWS buffer for more than 3 months.
  3. Use the C4 (1.7 µm x 1 mm x 100 mm) column for the phosphoinositides, PI4P and PI(4,5)P2 and the C18 column (1.0 mm x 100 mm x 1.7 mm) for PA, phosphatidylcholine (PC), and phosphatidylinositol (PI). For NPLC-MRM-EPI method, use the silica column (1 mm x 150 mm x 3 µm).
  4. Prepare eluent A by adding hexane:isopropyl alcohol:100 mM aqueous ammonium acetate in a 68:30:2 (vol/vol/vol) ratio and eluent B by adding hexane:isopropyl alcohol:100 mM aqueous ammonium acetate in 70:20:10 (vol/vol/vol) ratio for PA, PC, and PI. Prepare solvent A for phosphoinositides by adding 0.1% formic acid in water and solvent B by adding 0.1% formic acid in acetonitrile.

2. Isolation of tissue

  1. Collect 10 flies per sample using carbon dioxide (CO2) anesthesia (flies immobilize within seconds) and decapitate the flies using a sharp blade on a CO2 anesthetizing plate.
  2. Collect dark or light adapted flies of defined age (12-24 h old) in 1.5 mL tubes and snap-freeze in liquid nitrogen. Dehydrate the flies in acetone at -80 °C for 48 h in a glass vial35. For retinal tissue, collect 100 retinae from freeze dried flies. Using a scalpel, remove the eyes from the rest of the head and scoop out the retinae.
  3. For phosphoinositides PI4P and PI(4,5)P2 analysis, when working with fresh retinae, use 25 retinae per sample from flies grown under conditions of appropriate illumination for measuring lipid levels during illumination and from flies grown without light for measuring lipid levels in dark. Store in 50 µL of 1x PBS in homogenizer tubes on dry ice until ready for extraction.

3. Lipid extraction

CAUTION: Chloroform is a toxic solvent and is carcinogenic in nature. It affects the reproductive system and is a skin and eye irritant. Precaution should be taken in handling this chemical. All the steps involving chloroform should be performed in a well ventilated chemical hood.

  1. For all glycerophospholipids other than PI4P and PI(4,5)P2
    1. For each sample, homogenize 10 fly heads or 100 retinae (collected as described in step 2.2) in 0.1 mL of 0.1 N ice-cold methanolic HCl and 30 µL of internal standard mixture (PA (17:0/14:1), PC (17:0/14:1), lysophosphatidic acid (LPA; 13:0), lysophosphatidylcholine (LPC; 13:0), LPC (17:1), PA (17:0/17:0), PA (16:0-D31/18:1), LPA (17:1), LPC (19:0), PI (12:0/13:0), PA (12:0/13:0), and PC (12:0/13:0)). Prepare standards such that the final amount of any lipid standard falls within the linear response curve of the mass spectrometer.
    2. Homogenize tissues using an automated homogenizer that allows rapid and simultaneous treatment of all the samples. Briefly spin tubes in a tabletop centrifuge and ensure no pellet is formed-this indicates complete homogenization. Transfer the methanolic homogenate into a 2 mL capped microcentrifuge tube.
    3. Add 0.2 mL of ice-cold 0.1 N methanolic HCl for recovering any residual material in the tube and combine in the 2 mL tube. Add 0.1 mL of 0.1 N ice-cold methanolic HCl followed by 0.8 mL of chloroform and mix thoroughly. Allow the mixture, containing tissue homogenate, to stand on ice for 10 min, and then add 0.4 mL of 0.88% KCl and vortex for 30 s.
    4. Centrifuge the mixture at 1,000 x g for 10 min at 4 °C to separate the aqueous and organic phases. Take out the lower organic phase containing lipids very carefully without mixing with the aqueous phase and transfer into a fresh 2 mL microcentrifuge tube.
      NOTE: During lipid extraction, transfer the lower organic phase with utmost care to avoid mixing with aqueous phase, which may hamper downstream lipid analysis.
    5. Dry this lipid solution in a vacuum evaporator at 4 °C, visually inspect the sample to ensure complete drying. Resuspend in 420 µL of 2:1 methanol:chloroform mixture for analysis. Analyze the samples immediately without storage.
  2. For phosphoinositides PI4P and PI(4,5)P2
    1. Homogenize 25 retinae in 950 µL of phosphoinositide elution buffer (PEB; 250 mL of CHCl3, 500 mL of methanol and 200 mL of 2.4 M HCl) and add internal standards (phosphatidylethanolamine (PE) 17:0/14:1; PI (17:0/14:1); PI4P (17:0/20:4); and PI(4,5)P2 (16:0/16:0)) mixture containing 50 ng of PI, 25 ng of PI4P, 50 ng of PI(4,5)P2, and 0.2 ng of PE per sample.
    2. Add 250 µL each of chloroform and 2.4 M HCl, followed by sonication for 2 min and centrifugation at 1,000 x g for 5 min at 4 °C for phase separation. Take out the lower organic phase into a fresh tube, wash with 900 µL of LPWS and centrifuge at 1,000 x g for 5 min at 4 °C.
    3. Extract lipids from the remaining aqueous phase by once again performing a phase separation as in step 3.2.2. Dry the collected organic phases in a vacuum centrifuge set at 4 °C, visually inspect the samples to ensure complete drying.

4. Organic phosphate assay

  1. Make a stock solution of 7.34 mM KH2PO4. Make the standard dilutions, using distilled water, in microcentrifuge tubes as per Table 1. Lightly vortex the tubes and transfer the standard dilutions to phosphate-free glass tubes.
  2. Prepare a separate set of glass tubes containing the lipid samples. Optimize the volume of lipid samples such that the absorbance at 630 nm (after completion of this protocol) falls within the range of standard KH2PO4 (which is 0 to 20x).
  3. Heat the glass tubes containing standard dilutions at 120 °C to complete dryness. Heat the glass tubes containing the lipid samples (take 1/8th of total sample volume required to have absorbance within the range of standard KH2PO4) at 90 °C to complete dryness.
  4. Add 50 μL of 70% perchloric acid to all the tubes and heat at 180 °C for 30 min. Cool the tubes to room temperature. Add 250 μL of water, 50 μL of 2.5% ammonium molybdate and 50 μL of 10% ascorbic acid to each tube.
    NOTE: 2.5% ammonium molybdate and 10% ascorbic acid are weight/vol concentrations. They need to be freshly made and can be stored at 4 °C for up to 1 week.
  5. Keep the tubes in a shaking incubator at 37 °C for 1 h. Aliquot 130 μL of sample into a 96-well plate and measure the absorbance at 630 nm using a spectrophotometer.

5. Derivatization

CAUTION: Trimethylsilyldiazomethane (TMSD) is reported to have many toxicological effects in humans. TMSD in solution targets kidney, liver, gastrointestinal tract, skeletal muscles, central nervous system, and respiratory and reproductive systems. Extreme precaution should be taken while handling this chemical. The entire process should be performed in a well ventilated chemical hood.

  1. To the lower organic phase of the samples obtained from the end of step 3.2.3., add 50 µL of 2 M TMSD. Allow the reaction to proceed at room temperature for 10 min with constant shaking at 250 rpm. After 10 min, add 10 µL of glacial acetic acid to quench the reaction, indicated by the disappearance of yellow color in the solution.
  2. Tap the tubes and cautiously open them to remove the N2 formed in the quenching reaction. After N2 gas has escaped, close the tubes and spin them down. Then, add 600 µL of post-derivatization wash solution and vortex for 2 min in a mixer at 250 rpm.
  3. Discard ~400 µL from the upper phase that will form. Repeat step 5.2. Discard the entire upper phase and add 50 µL of 90% methanol to the lower phase and mix it.
  4. Dry the samples for 2 h in a centrifugal concentrator at 800 x g operating in vacuum. After drying, the tubes should contain ~20 µL of the remaining sample. Add 180 µL of 100% methanol, mix it well, and store at 4 °C for up to 2-3 days prior to LC-MS/MS.

6. Data acquisition and analysis

  1. Data acquisition in direct infusion MS (DIMS)
    NOTE: MS can be done by either direct infusion or liquid chromatography MS (LC-MS) method. In this section, we describe direct infusion-based MS.
    1. Before starting the experiment, calibrate the mass spectrometer as per the manufacturer's instructions.
    2. Generate a linear response curve for the dilution series of synthetic standards and based on the linear response of the instrument, dilute the sample so that the intensity of the lipid analyte falls within the linearity of the instrument. Dilute total lipid extracts or lipid standards with a mixture of 2:1 methanol:chloroform (vol/vol). Select dilution of the total lipid extracts and the synthetic standards individually for each experiment.
    3. Transfer the extracts and standards to individual vials. Avoid air bubbles during the transfer of samples into sample vials of MS. Air bubbles will create high pressure in the column and hamper the lipid analysis.
    4. Prior to the analysis, centrifuge the samples for 9 min at 6,440 x g and load into a 96-well plate and seal with aluminum foil.
    5. Perform MS analyses on a high-resolution mass spectrometer using direct infusion method. Achieve stable ESI-based ionization of glycerophospholipids using a robotic nanoflow ion source using chips with spraying nozzles of diameter 4.1 µm.
    6. Control the ion source using a custom mass spectrometer software and set ionization voltages at +1.2 kV and -1 kV in positive and negative modes, respectively; back pressure at 1 psi in both modes; and temperature of ion transfer capillary at 180 °C. Perform acquisition at mass resolution, Rm/z400 = 100,000. See Figure 2A for the software interface to set up mass spectrometer parameters.
    7. Re-dissolve dried total lipid extracts in 400 µL of chloroform:methanol (1:2). For the analysis, load 60 µL of samples onto a 96-well plate ion source and seal with aluminum foil. Analyze each sample for 20 min in positive ion mode to detect PC, phosphatidylserine (PS), phosphatidylethanolamine (PE), PE-O (ether linked phosphatidylethanolamine), ceramide (Cer), and ceramide phosphate (Cer-P).
    8. Perform an independent acquisition in negative ion mode for 20 min where PA and PI was detected. See Figure 2B for specific details of instrument setup for data acquisition for high resolution MS and a targeted list of PA species that has been included in data dependent acquisition (DDA) set up.
      NOTE: The software interface for setting up of DDA method is shown in Figure 2C. The list of PA molecules targeted in this approach is listed in Table 2. A screenshot of the experimental outcome in DDA approach is shown in Figure 2D.
  2. Data analysis in direct infusion MS (DIMS)
    NOTE: The analysis of mass data generated by all types of mass spectrometers requires an automated lipid analysis platform. LipidXplorer36 is a non-commercial software that supports all types of DIMS lipid experiments. This software can be found here: https://www.mpi-cbg.de/research-groups/current-groups/andrej-shevchenko/projects/lipidxplorer/
    1. Once the data has been acquired using the methods described in step 6.1, identify lipid species using a lipid analysis platform by matching m/z of their monoisotopic peaks to the corresponding elemental composition constraints. The data import example is shown in Figure 3A. Set mass tolerance to 10 ppm and intensity threshold according to the noise level reported by the mass spectrometer software.
    2. After importing, compile molecular fragmentation query language (MFQL) queries for PA based on the chemical structure of PA, which is shown in Figure 3B. See Figure 3C for the MFQL set up in the lipid analysis platform.
    3. Using a similar approach, compile MFQL for other glycerophospholipids in the lipid analysis platform software. Each query targets one lipid class and many queries can be used in a single execution.
    4. Run all the MFQLs by clicking on the Run button. All the identified lipid species are reported in a single results file in .csv file format with the abundances of corresponding precursor and/or fragment ions for the subsequent quantification of lipids. An example of the output is given in Table 3.

7. Liquid chromatography and tandem MS of derivatized samples

  1. Separate the samples obtained from step 5.4 by liquid chromatography using an ultra-performance liquid chromatography system. While choosing this system, ensure that its software can integrate with that of the mass spectrometer used for the analysis.
  2. Connect the system to a triple quadrupole mass spectrometer. Choose a separation column. For the separation of lipids other than PI4P and PI(4,5)P2 choose a C18 column (1.0 mm x 100 mm x 1.7 mm). For PI4P and PI(4,5)P2, choose a C4 column of dimensions 300 Å (1.0 mm x100 mm x 1.7 µm).
  3. Prepare the mobile phase that contains ammonium formate by dissolving ammonium formate in mass spec grade water first, and then in organic solvents. Sonicate all the solvents that are used in mass spectrometry for 20 min to remove air bubbles.
  4. Equilibrate the column by infusing solution A containing water + 0.1% formic acid and solution B having acetonitrile + 0.1% formic acid.
  5. Inject the eluate from the liquid chromatography system (injection volume in the range of 1-20 µL) into the mass spectrometer for analysis. Set the flow rate at 0.1 mL/min and temperature of the column at room temperature. Inject the entire elute volume coming out of the column into mass spectrometer.
  6. In the sample injection sequence, always start with a blank solvent injection (methanol) and keep neat standard samples intermittently in between biological samples for mass spectrometry run quality control checks (every sixth run).
  7. For the hybrid triple quadrupole ion-trap mass spectrometer experimental set-up, before starting the experiment, calibrate the mass spectrometer as per the manufacturer's instructions. For PA, PC, and PI, use electrospray ionization (ESI) to generate the ions and operate in positive mode to detect the positively charged lipid species. Acquire the data and analyze it using the installed data analysis software with the system.
  8. Optimize the parameters for analysis according to the corresponding internal standard used. Mass spectrometer parameters are as follows: dwell time = 30 ms; CAD (collision activated dissociation) = 3 psi; GS1 (source gas 1) = 24 psi and GS2 (source gas 2) = 21 psi; CUR (curtain gas) = 30 psi; IS (ESI voltage) = 4.5 kV; and TEM (source temperature) = 450 °C.
  9. For PI4P and PI(4,5)P2, use mass spectrometer in the positive mode. Mass spectrometer parameters are as follows: dwell time = 65 ms; CAD (collision activated dissociation) = 2 psi; GS1 (source gas 1) and GS2 (source gas 2) =20 psi; CUR (curtain gas) = 37 psi; IS (ESI voltage) = 5.2 kV; and TEM (source temperature) = 350 °C.

8. Normal phase liquid chromatography-multiple reaction monitoring-enhanced product ion scan MS (NPLC-MRM-EPI MS)

  1. Chromatographic conditions
    1. Use a normal phase LC method using a silica column, which is able to separate PA from other phospholipids. Use hexane:isopropyl alcohol:100 mM aqueous NH4COOH in the ratio 68:30:2 as mobile phase A and isopropyl alcohol:hexane:100 mM aqueous NH4COOH in a 70:20:10 ratio as mobile phase B.
      NOTE: This combination provided the best separation and peak selectivity of different molecular species of PA.
    2. Use the chromatographic behavior of reference compounds (internal standards), in terms of resolution and peak shape, to choose the optimal conditions.
    3. Perform chromatographic separation on a normal phase silica column (1 mm x 150 mm x 3 µm) at room temperature on an ultra-performance liquid chromatography column. Set the autosampler injection volume to 6 µL and the eluent flow rate to 210 µL/min.
    4. After 5 min of equilibration with 100% of mobile phase A, linearly increase the mobile phase B to 30% over 5 min, further to 80% over 5 min, then to 100% over 5 min, and hold it constant at 100% for 5 min. Lastly, re-equilibrate the column for 9 min.
  2. Mass spectrometry
    1. Use a hybrid triple quadrupole ion-trap mass spectrometer operating in negative ESI mode. Control the system operation and data acquisition using the analysis software provided. Before starting the experiment, calibrate the mass spectrometer as per the manufacturer's instructions.
    2. Optimize the source parameters using flow injection analysis of the internal standard mixture. Accordingly, set the ion spray voltage = -4.5 kV, source temperature (TEM) = 450 °C, collision activated dissociation gas (CAD) = 3 psi. Use nitrogen gas as the collision gas and set nebulizer gas (GS1) = 24 psi, the auxiliary gas (GS2) = 21 psi, and the curtain gas (CUR) = 30 psi.
    3. Set the compound dependent ion path parameters as declustering potential (DP) = -42 V, entrance potential (EP) = -6 V, and collision cell exit potential (CXP) = -12 V, optimized using continuous infusion of internal standard mixture solution. Record full product spectra along with precursor to product MRM transitions with varying collision energy (CE) starting from 12 eV to 40 eV for fragmentation analysis using EPI scanning function available in the mass spectrometer.
      NOTE: The MRM triggered IDA based EPI simultaneously records precursor ion-product, ion scanning and on the fly MS/MS acquisition. The MRM narrowed the ion scan range in quadrupole 1 (Q1) and the ion trap enhanced the ion fragments passing through Q2 thus improving the qualitative capability of quadrupole MS/MS greatly, especially for capture of all the fragments arising from the precursor ion. In EPI mode, multiple fragment ions arising from the precursor ions are detected in Q3 with better signal-to-noise ratio.
    4. Perform MRM experiments with CE of 39 eV to gain high sensitivity. Limit the maximum number of MRM to 75 and dwell time to 30 ms to detect and record the MRM of any specific molecule which elutes from the chromatographic column at any time during the run. This increases the duty cycle of the machine.
    5. Perform experimental tuning to decide the best ionization parameters for PA, as described above. For this experiment, set the ion spray voltage = -4.5 kV, source temperature (TEM) = 450 °C, collision activated dissociation gas (CAD) = 3 psi. Use nitrogen gas as the collision gas. Set nebulizer gas (GS1) = 24 psi, the auxiliary gas (GS2) = 21 psi, and the curtain gas (CUR) = 30 psi.
    6. Manually examine all tuning data to ensure proper selection of ionization parameters and product ions. Take into consideration the minimizing potential interference between MRM channels when selecting the product ion. The experimental parameters used for the analysis of all the PA molecular species are shown in Table 4.
      NOTE: The hybrid triple quadrupole linear ion trap mass spectrometer allows us to combine MRM scan mode with the ion trap scanning function, thus, enabling fast and high scanning by utilizing methods such as EPI scan for recording useful tandem mass spectra of each detected precursor. In this study, to identify distinct molecular species of PA, we have exploited this MRM-EPI based MS/MS approach based on the conventional triple quadrupole ion path with the EPI scanning property of the ion-trap, controlled by the analysis software.

Representative Results

Determination of linearity of measurement in MS. Linearity is the MS method's ability to provide results which are directly proportional to the concentration of the lipid analyte. Linearity depends on (a) ionization efficiency of the lipid analyte and (b) ionization behavior of lipid analyte at different concentrations depends on the used ion source. In electrospray ionization (ESI) that is used in this study, linearity holds at lower concentrations depending on (a) ion transport from ESI source to the mass analyzer and (b) the mass analyzer design and on the linearity of the detector's signal. A linearity measurement using internal standards for the molecule being analyzed has to be estimated for each experiment. As an example data set, Figure 4 shows the linearity response of PA (12:0/13:0) standard for a DIMS experiment over a concentration range spanning two orders of magnitude.

DIMS approach. Figure 5 shows all the lipid classes detected and quantified from wild type Drosophila using the global DIMS approach. Y-axis represents the mole percentage calculated on summation of all molecular species detected in each class of lipid and the percentage of total lipids detected is shown in Figure 5A. Figure 5B shows all the molecular species of PI detected and quantified and Figure 5C shows the molecular species of PC detected and quantified. PA was detected in this approach but with a smaller number of molecular species because of ion suppression. Hence, PA was detected and quantified using the DDA approach. Figure 6 shows all the molecular species of PA detected and quantified using this approach from wild type Drosophila. In all of the analyses described, the level of any specific lipid class is normalized to the total amount of lipid extracted from that sample. The total amount of lipid in a sample is estimated from the total amount of organic phosphate estimated using organic phosphate assay37.

Reverse phase liquid chromatography and tandem MS of derivatized samples. A schematic of TMS-diazomethane derivatization is shown in Figure 7A. After the derivatization of PA, ESI produces a protonated ion which upon collision induced dissociation inside the mass spectrometer methylated protonated PA molecule produces two kinds of species, which are shown in Figure 7B. First one is a neutral molecule of mass 126 Da arising from the PA headgroup and the second one is the charged diacylglycerol fragment arising from the remaining of the derivatized PA precursor ion; the mass of this fragment depends upon the constituent fatty acyl chains. Such a fragmentation pattern is true for all kinds of glycerophospholipid molecules, and one can calculate the exact mass of the derivatized precursor along with their fragment masses; such calculations for PA are shown in Table 5. Experimentally, there are two ways of setting up the MS method for detecting PA. One can detect PA by scanning for neutral losses, which is the headgroup specific neutral mass arising from the fragmentation process or one can set up multiple reaction monitoring by setting up precursors – product ion masses. In the second approach the product mass is charged diacylglycerol fragment. Figure 8 summarizes the first approach where Figure 8A shows the retention time chromatogram and Figure 8B shows the mass spectra of neutral loss scanning of 126 Da. Figure 9 shows the major PA species detected from wild type Drosophilausing the second approach.

Normal phase liquid chromatography-multiple reaction monitoring-enhanced product ion scan MS (NPLC-MRM-EPI MS). In order to establish the precise composition of the acyl chains in an individual molecule of glycerophospholipid, it is necessary to detect all the fragment ions generated from a precursor ion in a triple quadrupole mass spectrometer. To extend the process of MRM beyond precursor ion-product ion pair detection, the enhanced product ion scanning function of the machine was used. In order to obtain the precise structure of each acyl chain of an individual molecular species, i.e., regioisomers (SN1 and SN2) of PA, information dependent (IDA) auto-MS/MS experiments at different collision energy starting from -12 eV to -45 eV was performed to determine the collision energy for optimal fragmentation. IDA enables on the fly acquisition of MS/MS spectra during an MS experiment. In this experiment, MRM of known synthetic standards were used as survey scan to trigger the IDA followed by enhanced product ion (EPI) scans. As shown in Figure 10, the EPI spectrum of the m/z 549.27 [M-H] ion of the chemically synthesized 1-dodecanoyl-2-tridecanoyl-sn-glycero-3-phosphatidic acid (12:0/13:0-PA) contains ions at m/z 199.19 ([R1COO]) and 213.25 ([R2COO]), corresponding to carboxylate anions arising from SN1 and SN2, respectively. Consistent with previous reports38 the intensity of the former ion is higher than the latter. The spectra also contain ions at m/z 349.21 ([M-H-R1COOH]) and 335.16 ([M-H-R2COOH]) corresponding to the neutral loss of the fatty acid moiety at SN1 and SN2. The intensity of the latter ion is two times more abundant than the former. The spectrum also contains ions at m/z 367.22 ([M-H-R'1CH=C=O]) and 353.17 ([M-H-R'2CH=C=O]) representing neutral loss of fatty acyl moiety as ketene at SN1 and at SN2, respectively. Again, the intensity of the m/z 353.17 ion is two times more than the m/z 367.22 ion. These results suggest that loss of the free fatty acid and loss of the fatty acyl ketene at SN2 is sterically more favorable than the analogous loss at SN1 and the abundance of the fragment ions m/z 335.16 ([M-H-R2COOH]) > 353.17 ([M-H-R'2CH=C=O]) suggest that neutral loss of the acid is a more facile process than the corresponding ketene loss. A similar pattern was observed for analogous SN1 fragments, i.e., intensity of m/z 349.21 ([M-H-R1COOH]) > m/z 367.22 ([M-H-R'1CH=C=O]). Therefore, the observation of the abundance of m/z 199.19 ([R1COO]) > 213.25 ([R2COO]) is attributed to the fact that the neutral loss of free fatty acid ([M-H-RxCOOH]) and the neutral loss of ketene ([M-H-R'xCH=C=O]) ions may further undergo fragmentation under the applied collision energy, CE = -39 eV, after they were formed. To validate this fragmentation pattern, a series of synthetic PA standards were characterized by EPI tandem MS to compare the types of fragment ions produced by each activation at different collision energy and evaluate the utility for determining the location of the fatty acyl chains. It was observed that for all precursor ions of PA, the fatty carboxylate anion at SN1 gives rise to the highest signal intensity at all CE other than the lowest CE of -12 eV (Figure 11). The MS/MS spectra resulting from m/z 421.24 (LPA (17:1); Figure 11A), contains fragment ions at m/z 153.02 corresponding to glycerol-3-phosphate with loss of water and m/z 267.24 corresponding to RCOO ion. When the synthetic standard PA (17:0/14:1) was analyzed, it was observed that the highest intensity fragment ion at m/z 269.32 was accompanied by m/z 153.06, 225.28 ([R2COO]), 361.21 ([M-H-R1COOH]), 405.18([M-H-R2COOH]), and 423.19([M-H-R2'CH=C=O]) ions as shown in Figure 11B. Interestingly, the mass difference between the two highest peaks arising from fatty carboxylate ions in these two MS/MS spectra (Figure 11A,B) is of 2 Da, which corresponds to a single double bond. That means m/z 267.24 corresponding to RCOO ion in LPA (17:1) plus 2 Da will give rise to a RCOO ion (m/z 269.32) without the double bond, which is the case in PA (17:0/14:1). In addition to these two standards, we also analyzed the MS/MS spectra of synthetic PA (17:0/17:0; data not shown) and observed a single intense peak at m/z 269.30. To further validate this strategy for identifying the acyl chain at SN1, a similar EPI experiment was performed with a deuterated synthetic standard PA (16:0-D31/18:1), where the fatty acyl chain at SN1 alone is deuterated allowing the unambiguous identification of the peak arising from it during fragmentation. Using this standard, it was found (Figure 11C) that the two most intense peaks are at m/z 281.37 and m/z 286.36 and the latter one is at least twice more abundant than the former. Analyzing these masses clearly tells that m/z 281.37 is derived from the 18:1 carboxylate ion and since all the SN1 fatty acyl hydrogens are replaced by the heavier deuterium isotope, there is an extra mass of 31 added to the m/z of 16:0 carboxylate ion (255.36 + 31) giving rise to the peak at m/z 286.36 (Figure 11C). Thus, the potential fragment masses arising from IDA based EPI scans could be used in addition to the MRM to identify and quantify individual molecular species of PA and CE = -39 eV can be used as the optimal fragmentation energy.

We performed similar experiments with Drosophila head lipid extracts. A representative molecule of precursor mass 699.24, which is a PA (36:2), is shown in Figure 11D. As shown in Figure 11D, the tandem spectrum of the m/z 699.32 ion in EPI scan contains ions at m/z 281.30 (18:1 carboxylate anion), 283.35 (18:0 carboxylate anion), and 279.30 (18:2 carboxylate anion). The relative intensity profile of fatty acid fragments (carboxylate anion) as observed from the spectra is as follows: I281.30 > I283.35 > I279.30. From this intensity profile, it was concluded that there are two individual acyl chain specific molecules within PA (36:2) and the exact position of these fatty acyl components will be 18:1 in SN1 with 18:0 in SN2 (I281.30 > I283.35) and 18:0 in SN1 with 18:2 in SN2 (I283.35 > I279.30).

The individual extracted ion chromatogram (XIC) of all the PA molecules is shown in Figure 12. To set up these MRM-based transitions in experimental setup, a user-friendly database can be found at 31. Using this approach, we analyzed head extracts from wild type and a loss-of-function mutant of Drosophila phospholipase D (dPld3.1)23. The results are summarized in Figure 13. Using direct infusion MS, we have previously identified and confirmed the existence of 15 molecules of PA from Drosophila head extracts and seven from retinal extract23. One can quantify the level of each PA species by comparing the monoisotopic MS1 intensity of the corresponding peaks to that of the internal standard. While this represents a substantial increase over the number of PA species previously reported from Drosophila tissues15, this type of analysis is limited in its ability to precisely identify the composition of each acyl chain of an individual molecular species, i.e., regioisomers (SN1 and SN2). Using the method described here, we revisited the analysis of the levels and composition of PA in head extracts from wild type and a loss-of-function mutant of Drosophila phospholipase D (dPld3.1). Consistent with our previous analysis using high resolution mass spectrometry (HRMS)23, we found that the total amount of PA is significantly lower in dPld3.1 compared to wildtype as shown in Figure 13D. Further, using this new method, we can accurately identify individual species of PA, establishing the unique acyl chain composition (the total number of carbon atoms and double bonds in each fatty acyl chain) at SN1 and SN2 (Figure 13B). We were able to establish that in Drosophila heads, PA is a complex mixture of 36 individual molecular species with unique acyl chain compositions (Figure 13B). Palmitic (16:0), palmitoleic (16:1), oleic (18:1), linoleic (18:2), and linolenic (18:3) fatty acyl chains are found to be populated in PA from Drosophila heads. The most abundant PA species are PA (16:1/18:1), PA (16:0/18:2), PA (18:2/20:0), PA (18:1/18:1), PA (18:2/18:2), PA (18:2/18:3), PA (18:1/18:2), and PA (18:3/20:0; Figure 13B). We also determined the distributions of combined fatty acid chain length (Figure 13A) and total number of double bonds in the two fatty acyl chains (Figure 13C) and found that the C36 series is the most populated as shown in Figure 13A and the combined fatty acid chain in SN1 and SN2 containing two double bonds are most populated as shown in Figure 13C.

Figure 1
Figure 1: Schematic showing a typical workflow of MS based lipid analysis of Drosophila tissue samples. The section number in the article where each step is described is shown in the figure. Abbreviations: MS = Mass spectrometry; NPLC-MRM-EPI-MS = normal phase liquid chromatography-multiple reaction monitoring-enhanced product ion scan-MS. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Mass spectrometry instrument set up for DIMS method. (A) An example of the software interface showing set up of ESI parameters on the MS used in the DIMS method. The ESI parameters such as gas pressure, voltage, and the polarity are shown in the screenshot. (B) A screenshot of the software interface of DIMS – DDA set-up on the MS software for high-resolution MS is shown. This screenshot shows the window for MS1, which is a high resolution of the Fourier transform mass spectrometry (FTMS) mode. Parameters such as run time, FTMS resolution, mass range, and polarity are set up using this interface. (C) A screenshot of the software interface of DIMS – DDA set-up on the mass spectrometer for PA analysis is shown. In this section, the analyzer should be selected as ion trap. In a step-wise manner the precursor masses of different MS1 based PA molecular ions should be given. The collision induced dissociation (CID) parameters should be given. CID parameters is given for the dissociation of lipid analyte inside mass spectrometer. Mass range should start from 50 to capture all the fragment ions. (D) An example of initial data acquired in a DIMS-DDA based experimental outcome is shown on the software interface of the mass spectrometer. The upper panel shows the chromatogram; x-axis is the elution time and y-axis is the relative abundance of the ion signal. The x-axis shows the m/z and the y-axis shows the relative abundance of an individual PA molecule. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Data analysis of data obtained from the DIMS method. (A) A Lipidxplorer software analysis window depicting the data import settings is shown. (B) A representative chemical structure of PA (16:1/18:1) is shown. This chemical structure helps in formulating molecular fragmentation query language (MFQL) for the detection of PA from experimental runs. (C) An example MFQL for a PA molecule is shown. Please click here to view a larger version of this figure.

Figure 4
Figure 4: A linear dose-response curve observed for PA (12:0/13:0) by the DIMS approach. The y-axis represents the observed intensity per scan and the x-axis represents the concentration in nM. Effect of concentration of PA (12:0/13:0) on instrument response. A 200 nM concentration of PA (12:0/13:0) was prepared and diluted to different concentrations, and then analyzed in direct infusion mass spectrometry. Several scans were obtained. Y-axis in the plot is showing the intensity per scan. The data points were fitted linearly and a R2 of 0.9662 was obtained. Please click here to view a larger version of this figure.

Figure 5
Figure 5: Major lipid classes identified and quantified using DIMS. (A) Major lipid classes measured using DIMS from wild type Drosophila head lipid extracts. X-axis represents individual lipid classes. Total lipid class abundance is presented as mol% of total lipids in y-axis. Error bars indicate the mean ± standard error of mean (SEM) from three separate analyses (n = 3). (B) PI molecules identified using DIMS and quantified using internal standard from 10 wild type Drosophila heads. X-axis represents individual PI molecular species detected. Y-axis represents the amount of each individual PI species in pmole. Error bars indicate the mean ± SEM from three separate analyses. (C) PC molecules identified using DIMS and quantified using internal standard in wild type Drosophila head extract. X-axis represents individual molecular species of PC. Y-axis represents the amount of each individual PC species from 10 heads in pmole in log10 scale. Error bars represent mean ± SEM from three separate analyses. Please click here to view a larger version of this figure.

Figure 6
Figure 6: PA molecules identified using DIMS-DDA method and quantified using internal standard in wild type Drosophila head extract. X-axis represents different PA molecular species. Y-axis represents the amount of each individual PA molecules in pmole in log10 scale. Error bars indicate the mean ± SEM from three separate analyses (n = 3). Each individual molecular species of PA was quantified based on the intensity and amount of internal standard PA (12:0/13:0). In this approach, the exact composition of fatty acyl chain cannot be determined and hence the molecular species of PA is represented as (total number of carbon in two fatty acyl chain:total number of double bonds in two fatty acyl chain). As can be seen from the plot, PA (36:4) is the most abundant and PA (32:1) is the least abundant. Please click here to view a larger version of this figure.

Figure 7
Figure 7: Chemical modification of glycerophospholipids using trimethylsilyl diazomethane (TMSD). (A) A representative PA species (16:1/18:1) with acyl chains in red and green color. After derivatization with TMSD, PA gets methylation of the hydroxyl groups attached to the phosphate groups occurs. The mass of PA before and after derivatization is shown. (B) The fragmentation of the methylated parent mass generates a neutral methylated head group and a charged diacylglycerol fragment. The mass of the precursor molecule and the fragments induced by CID is shown. The mass of the neutral methylated headgroup can be used for neutral loss scanning MS. The charged fragment mass can be used to quantify the red and green acyl chain-containing PA species by the MRM method described in the main text. Please click here to view a larger version of this figure.

Figure 8
Figure 8: Detection of different PA molecular species by scanning for neutral loss of 126 Da in UPLC-MS of derivatized sample. (A) Retention time chromatogram of neutral loss of 126 Da. X-axis shows retention time (min) and y-axis shows the intensity of the ion current. The box shows the region in retention time chromatogram where PA has been detected in the mass spectrometer. (B) Mass spectra of neutral loss scanning of 126 Da. The detected masses of DAG fragment after neutral loss of 126 Da of individual PA molecules are shown on top of each peak. X-axis show mass (m/z). Y-axis depicts the intensity of the ion current in counts per second unit. Each number in this plot correspond to different molecular species of PA molecules. To emphasize this, three representative arrows indicating individual PA species are shown in the figure. Please click here to view a larger version of this figure.

Figure 9
Figure 9: Major PA molecules identified using post TMSD derivatisation RPLC-MRM and quantified using internal standard in wild type Drosophila head extract. X-axis represents different PA molecular species. Y-axis represents the amount of each individual PA molecules in pmole per nmole total lipid phosphate. The values are mean of the samples with the error bars indicating ± SEM from three separate analyses (n = 3). Each individual molecular species of PA was quantified based on the intensity and amount of internal standard PA (17:0/14:1). In this approach, the exact composition of fatty acyl chain cannot be determined and hence the molecular species of PA is represented as (total number of carbon in two fatty acyl chain:total number of double bonds in two fatty acyl chain). As can be seen from the plot, PA (36:4), PA (36:5), PA (38:5) are the most abundant and PA containing the total number of carbon in two fatty acyl chain less than 30 are less abundant. Please click here to view a larger version of this figure.

Figure 10
Figure 10: Enhanced product ion (EPI) spectra of [M-H] ion of synthetic PA (12:0/13:0) standard. The fragments are listed here: (i) PO3 ion (from phosphate) at m/z 78.84 Th, (ii) glycerol-3-phosphate ion with loss of H2O at m/z 153.04 Th, (iii) SN1 RCOO ion at m/z 199.19 Th, (iv) SN2 RCOO ion at m/z 213.25 Th, (v) neutral loss of SN2 RCOOH group from [M-H] at m/z 335.16 Th, (vi) neutral loss of SN1 RCOOH group from [M-H] at m/z 349.21 Th, (vii) loss of SN2 acyl chain as ketene (RCH=C=O) from [M-H] at m/z 353.17 Th, (viii) loss of SN1 acyl chain as ketene (RCH=C=O) from [M-H] at m/z 367.22 Th, (ix) Precursor ion [M-H] at m/z 549.27 The x-axis shows the m/z values and the y-axis shows intensity of the ion current in counts per second (cps). Please click here to view a larger version of this figure.

Figure 11
Figure 11: Enhanced product ion (EPI) spectra. EPI spectra of (A) precursor of 421.24 Th of synthetic LPA (17:1), (B) precursor of 631.29 Th of synthetic PA (17:0/14:1), (C) precursor of 704.40 Th of synthetic PA (16:0-D31/18:1), (D) precursor of 699.32 Th of tissue sample PA (18:1/18:1). Fatty acid carboxylate ions observed are as follows in (A) m/z 267.24 (SN1), (B) m/z 269.33 (SN1) and 225.28 (SN2), (C) m/z 286.36 (SN1) and 281.37 (SN2), (D) m/z 281.30 (SN1 and SN2) and 283.35 (SN1) and 279.30 (SN2). The prediction of SN1 fatty acid carboxylate ions is based on higher intensity, compared to their SN2 counterpart. Please click here to view a larger version of this figure.

Figure 12
Figure 12: Detection of PA by NPLC-MRM-EPI method. Extracted ion chromatograms (XICs) of detected PA molecules using specific MRM set up for precursor mass and their corresponding SN1 fragment mass are shown in the plot. The x-axis represents time in min and y-axis represents ion current, i.e., intensity in cps (counts per second). Please click here to view a larger version of this figure.

Figure 13
Figure 13: Results from NPLC-MRM-EPI method for PA. (A) The distributions of combined fatty acid (SN1 + SN2) chain length. The abundance of each is presented as a percent relative to total PA in its category. (B) Radar chart of individual molecular species of PA quantified using NPLC-MRM-EPI. The value of each PA species (pmole/nmole of total lipid phosphate) is depicted by the node (anchor) on the spoke (axis). A line is then drawn which connects the data values for each spoke. All data are plotted in log10 scale. N = 9. (C) The distribution of combined unsaturation (SN1 + SN2) in PA. The abundance of each is presented as a percentage relative to the total PA in its category. (D) Total phosphatidic acid levels in head extract of wildtype control and dPLD3.1 obtained from NPLC-MRM-EPI experiments. The x-axis represents the genotypes, and the y-axis shows the level of PA normalized to total lipid phosphate content of the lipid extract. N = 9. Error bars indicate standard error of mean (SEM). Student's t-test was performed. ** = P-value < 0.01. Please click here to view a larger version of this figure.

Table 1: Table showing the set-up of standards for the organic phosphate assay. The amount of stock solution (concentration = 7.34 mM) and water to be mixed to make each dilution is shown. Please click here to download this Table.

Table 2: List of PA in DIMS targeted approach. The first two columns show the MASS = mass of species and CHEMSEC = the chemical formula. The third column NAME = name of the PA molecule. Please click here to download this Table.

Table 3: A screenshot of the output in .csv format from the Lipidxplorer software. Column 1 – mass of PA molecule; column 2 – the chemical formula of PA molecule; column 3 – error is in ppm; column 4 – name of the PA molecule, columns 5 to 15- intensity of individual test samples. Please click here to download this Table.

Table 4: The experimental parameters used for the analysis of all the PA molecular species in NPLC-MRM-EPI method. MOLECULAR ID – molecular name of PA, Q1 MASS (Da) – mass selected in quadrupole 1, Q3 MASS (Da) – mass selected in quadrupole 3, Retention time – retention time of the PA molecules, Dwell Time – dwell time in msec, CE – collision energy used for MS/MS fragmentation, DP – declustering potential, EP – exit potential, CXP – collision cell exit potential. Please click here to download this Table.

Table 5: Screenshot of neutral mass. TMSD derivatized mass and the DAG fragments of some representative PA molecules. Column A shows the lipid class, column B shows the total number of carbon in two fatty acyl chains, column C shows the total number of double bonds in two fatty acyl chains, column D shows the chemical formula of the lipid molecules, column E shows the neutral mass of the lipid molecule, column F shows the methylated mass after derivatization using TMSD, column G shows the protonated mass of the methylated lipid molecule, and column H shows the DAG fragment arise after collision induced dissociation of the protonated methylated lipid molecule. Please click here to download this Table.

Discussion

A number of lines of evidence converge on multiple roles of signaling lipids in regulating the organization and function of Drosophila photoreceptors. In addition to the well-studied role of lipids in regulating phototransduction3, signaling lipids have also been implicated in protein trafficking and sub-cellular organization23,30,39,40,41. These studies have been enabled by well-developed methods for the physiological42 and genetic analysis43 of photoreceptor function in the Drosophila model system. However, in order to fully understand the function of lipid molecules in Drosophila photoreceptors, it is also necessary to use methods that detect and quantify the level of lipids in photoreceptors under various physiological conditions (e.g., dark versus illumination) and also in relevant Drosophila mutants that are predicted to impact the metabolism of signaling lipids and photoreceptor physiology.

The quantification of signaling lipids in Drosophila photoreceptors represents a number of challenges: (i) the small size of retinal tissue requires the use of large number of flies for each analysis; (ii) since most analysis is done in the context of intact animals, the use of radionuclide labeling is challenging; (iii) the complex nature of lipid molecules, manifested by the diversity of constituent acyl chains needs to be captured by the analytical method. The use of modern mass spectrometry44 offers a solution to all of these challenges. Lipid analysis using MS is highly sensitive and does not require the use of radionuclides thus facilitating analysis of small amounts of tissue from photoreceptor tissue. Coupled with method development for isolation of retinal tissue from fly heads27,35, MS has allowed the functional analysis of signaling lipids in Drosophila photoreceptors23,27,29,30,41,45,46.

The lipidome of any cell, including Drosophila photoreceptors is both complex and heterogeneous. The cellular lipidome includes both structural and functional lipids and in many instances, the structural lipids (e.g., PC and PE) are far more abundant than other classes such as PA and phosphoinositides. When choosing the most suitable MS experiment to perform, it is important to take into consideration the scientific question at hand. While the use of DIMS has facilitated a global description of the lipidome15,33 low abundance lipids are often not detected or are underestimated. Therefore, DIMS should be chosen when the role of lipids is implied by other experimental evidence but the exact lipid classes that are likely to be involved need to be identified. DIMS experiments provide an overview of most of the major classes of lipids; this approach is sometimes referred to as lipid profiling. Given that many studies related to Drosophila photoreceptors often involve the analysis of signaling lipids, it is important to choose methods that allow these to be detected and quantified. Due to ion suppression of low abundant lipids by the more abundant ones, the detection of low abundance signaling lipids is challenging. The use of derivatization coupled with the LC separation of a complex lipid extract, as described in this article, allows signaling lipids to be detected much more effectively by mass spectrometry. The detailed methods described for the detection and quantification of PA23,46 are applicable for other low abundance lipids such as phosphoinositides, ceramide, and sphingosine phosphates. Thus, this method of separating derivatized lipids by chromatography followed by mass spectrometry is most appropriate for those low abundance lipids and would be best chosen to quantify levels of specific signaling lipids that need to be quantified in the context of a specific scientific question.

To solve the challenges in detecting PA due to ion suppression, several chromatographic methods were used such as reverse phase LC, HILIC, and normal phase LC in combination with various column chemistry. Reverse phase columns such as C18 and C30 could not achieve complete separation of all the molecular species of different phospholipids (PLs), presumably because the separation mechanism is likely driven by the hydrophobic fatty acyl chains rather than the polar head groups. An issue frequently encountered while performing PA chromatography is the peak tailing effect occurring due to the charged phosphate group at neutral pH. The addition of phosphoric acid diminishes this problem47, but ot occurs at the cost of rapid contamination of the ion source and transfer optics. Hence, we developed a normal phase LC method using a silica column, which can separate PA from other phospholipids, presumably because the chemical nature of PA is more polar and acidic.

Many lipid molecules are composed of a head group (e.g., inositol head group in phosphoinositides), that define the lipid class but also fatty acyl chains in the backbone that anchor it to the lipid membrane. It has generally been assumed that signaling functions reside within the headgroup of the lipid molecule. However, recent studies have indicated that the acyl chain length and saturation of signaling lipids may also contribute to their functional properties46,48,49,50,51,52. The analysis of this question in the context of in vivo model systems such as Drosophila photoreceptors requires the ability to identify signaling lipids with unique acyl chain composition at SN1 and SN2. In the context of such scientific questions, the use of NPLC-MRM-EPI MS, described in this study, to determine the precise composition of acyl chain at SN1 and SN2 will be helpful especially when the role of distinctive acyl chain composition in lipid signaling function is anticipated.

Some studies have already described a role for diet in the composition of the lipidome in Drosophila15. Thus when performing comparative analysis of the lipidome from two different Drosophila strains, it is important to rear the animals on the same batch of culture medium or use a holidic medium for the culture of the two strains53. In addition, like all other organisms, Drosophila too show variations in the composition in their genome, including single nucleotide polymorphisms, deletions, and insertions54. Such variations could impact lipid metabolism and, therefore, contribute to variability in measurements of lipid levels obtained by MS. A systematic study of the effect of such variations on the lipidome in Drosophila has not been performed. Therefore, caution should be exercised when attributing functional significance to altered levels of any lipid class or species in a Drosophila specific strain. Presently, the best approach would be to perform comparative analysis with closely matched genetic backgrounds or in the case of a mutant, test whether the observed changes in the lipidome can be rescued by reconstitution with a wildtype transgene.

The methods described here for lipidomics analysis require that eye tissue be disrupted to extract lipids for analytical processes. Inevitably, this results in the lack of spatial information, for example, the subcellular location at which signaling lipids are produced and exert their functions. To some extent, this problem can be resolved by the use of protein probes that allow the reporting of signaling lipids with spatial information (for a review55), an approach that has also been adapted for use in Drosophila photoreceptors27,56,57,58. However, this method has not been extended to lipid classes other than phosphoinositides and also does not report the acyl chain composition of the lipids being detected since the protein probes used are typically head-group specific. The recent rapid developments in quantitative imaging mass spectrometry59 that combine MS with spatial resolution at the tissue level offer an opportunity to obtain spatially resolved lipid signals in intact eye tissue. However, the best resolution achieved to date is around 20 mm and future improvements in this will facilitate such analysis in photoreceptors. Overall, the methods described in this manuscript offer approaches for the analysis of lipid signaling using Drosophila photoreceptors as a model system.

Disclosures

The authors have nothing to disclose.

Acknowledgements

The work described in this manuscript was supported by Department of Atomic Energy, Government of India (Project Identification No. RTI 4006), the Department of Biotechnology, Government of India (BT/PR4833/MED/30/744/2012) and an India Alliance Senior Fellowship (IA/S/14/2/501540) to PR. We thank the NCBS Mass Spectrometry Facility, especially Dr. Dhananjay Shinde and members of the PR lab for their contributions to developing these methods.

Materials

0.1 N methanolic HCL For total lipid isolation
0.88% KCl Sigma Aldrich P9541 For total lipid isolation
1.5 ml / 2ml LoBind Eppendorf tubes Eppendorf, 022431081/022431102 For total lipid isolation
2.3.18 16:0/18:1 Diether PE Avanti polar lipids 999974 Lipid Internal Standard
37% pure HCl Sigma Aldrich 320331 For total lipid isolation
96-well plate Total Organic Phosphate assay
Acetone Fisher Scientific 32005 For dissections
Ammonium molybdate Total Organic Phosphate assay
Ascorbic Acid Total Organic Phosphate assay
Bath sonicator
BEH300 C18 column [1.0 mm x 100mm x 1.7 mm] Waters India Pvt. Ltd. 186002352 LC
Blade holder Fine Scientific Tools 10052-11 For dissections
BOD incubator Total Organic Phosphate assay
Breakable blades Fine Scientific tools 10050-00 For dissections
Butter paper GE healthcare 10347671 For dissections
C4, 300 A0, [1.7 μm x1 mm x 100 mm] column Waters India Pvt. Ltd. 186004623 LC
Chromatography amber color glass vials with inserts Merck 27083-U
d18:1/17:0) Avanti polar lipids 860517 Lipid Internal Standard
d5-Phosphatidylinositol 3,5-bisphosphate [PI(3,5)P2]-16:0/16:0 Avanti polar lipids 850172 Lipid Internal Standard
Dissecting microscopes Olympus SZ51 For dissections
Dry heat bath.
Eluent A Hexane:Isopropyl alcohol:100 mM aqueous ammonium acetate (68:30:2) , for LC
Eluent B Hexane:Isopropyl alcohol:100 mM aqueous ammonium acetate (70:20:10), for LC
Filter paper Indica-HM2 74039 For dissections
Flasks Borosil For dissections
Flies NA NA Raghu Padinjat lab
Fly food NA NA NCBS lab kitchen, composition: corn flour 80 g/L, D-glucose 20 g/L, sucrose 40 g/L, agar 8 g/L, yeast extract 15 g/L, propionic acid 4 mL, TEGO (methyl para hydroxybenzoate) 0.7 g/L, orthophosphoric acid 0.6 mL)
Forceps Fine Scientific Tools 11254-20 For dissections
Fume hood
Funnel Borosil For dissections
Glacial acetic acid Fisher Scientific A35-500 For derivatization
Glass bottles: transparent and amber color For total lipid isolation
High-temperature-resistant phosphate-free glass tubes. Total Organic Phosphate assay
Homogenization tubes with zirconium oxide beads For total lipid isolation
Homogenizer instrument Precellys
Humidified CO2 connected to fly pads For fly pushing
Illumination controlled incubators Panasonic Sanyo MIR-553 For fly rearing
Initial organic mixture methanol:chloroform (2:1), For total lipid isolation
LC-MS grade Chloroform Sigma Aldrich 650498 For total lipid isolation
LC-MS grade Methanol Sigma Aldrich 34860 For total lipid isolation
LC-MS grade water Sigma Aldrich 34877 For total lipid isolation
Light meter HTC instruments LX-103
Low retention tips Eppendorf 0030072006/72014/72022/72030 For total lipid isolation
LTQ Orbitrap XL instrument Thermo Fisher Scientific, Bremen, Germany
Lysophosphatidic acid (LPA)- 13:0 Avanti polar lipids LM-1700 Lipid Internal Standard
Lysophosphatidic acid (LPA)- 17:1 Avanti polar lipids LM 1701 Lipid Internal Standard
Lysophosphatidylcholine (LPC) -13:0 Avanti polar lipids LM-1600 Lipid Internal Standard
Lysophosphatidylcholine (LPC) -17:1 Avanti polar lipids 855677 Lipid Internal Standard
Lysophosphatidylcholine (LPC)- 19:0 Avanti polar lipids 855776 Lipid Internal Standard
Perchloric acid. Total Organic Phosphate assay
Phosphate standard potassium dihydrogen phosphate Total Organic Phosphate assay
Phosphate-buffered saline (PBS) NA NA Composition: 137mMNaCl, 2.7mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4, pH 7.4
Phosphatidic acid (PA)- 12:0/13:0 Avanti polar lipids , LM-1400  Lipid Internal Standard
Phosphatidic acid (PA)- 17:0/14:1 Avanti polar lipids LM-1404 Lipid Internal Standard
Phosphatidic acid (PA)-(17:0/17:0) Avanti polar lipids 830856 Lipid Internal Standard
Phosphatidic acid (PA)-16:0-D31/18:1 Avanti polar lipids 860453 Lipid Internal Standard
Phosphatidylcholine (PC) -12:0/13:0 Avanti polar lipids LM-1000 Lipid Internal Standard
Phosphatidylcholine (PC)- 17:0/14:1 Avanti polar lipids LM-1004 Lipid Internal Standard
Phosphatidylethanolamine (PE) – 17:0/14:1 Avanti polar lipids LM-110 Lipid Internal Standard
Phosphatidylinositol (PI) – 17:0/14:1 Avanti polar lipids LM-1504 Lipid Internal Standard
Phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2)]-17:0/20:4 Avanti polar lipids LM-1904 Lipid Internal Standard
Phosphatidylinositol 4-phosphate (PI4P) – 17:0/20:4 Avanti polar lipids LM-1901 Lipid Internal Standard
Robotic nanoflow ion source TriVersa NanoMate (Advion BioSciences, Ithaca, NY, USA)
Rotospin instrument Tarsons 3090X
Silicone pads For dissections
solvent A 0.1% formic acid in water, for LC
solvent B 0.1% formic acid in acetonitrile, for LC
Table-top centrifuge
Thermo-mixer
TMS-diazomethane Acros AC385330050 For derivatization
Triple quadrupole mass spectrometer AB Sciex QTRAP 6500
UPLC system Waters Acquity
Vacuum centrifugal concentrator Scanvac , Labogene
Vortex machine

References

  1. Hardie, R. C. C., Raghu, P. Visual transduction in Drosophila. Nature. 413 (6852), 186-193 (2001).
  2. Raghu, P., Yadav, S., Mallampati, N. B. N. Lipid signaling in Drosophila photoreceptors. Biochimica et Biophysica Acta. 1821 (8), 1154-1165 (2012).
  3. Hardie, R. C. TRP channels and lipids: from Drosophila to mammalian physiology. The Journal of Physiology. 578 (1), 9-24 (2007).
  4. Fuchs, B., Süss, R., Teuber, K., Eibisch, M., Schiller, J. Lipid analysis by thin-layer chromatography–a review of the current state. Journal of Chromatography. A. 1218 (19), 2754-2774 (2011).
  5. Tang, B., Row, K. Development of gas chromatography analysis of fatty acids in marine organisms. Journal of Chromatographic Science. 51 (7), 599-607 (2013).
  6. Kinsey, C., et al. Determination of lipid content and stability in lipid nanoparticles using ultra high-performance liquid chromatography in combination with a Corona Charged Aerosol Detector. Electrophoresis. , (2021).
  7. Gandhi, A. S., Budac, D., Khayrullina, T., Staal, R., Chandrasena, G. Quantitative analysis of lipids: a higher-throughput LC-MS/MS-based method and its comparison to ELISA. Future Science OA. 3 (1), 157 (2017).
  8. Ouldamer, L., et al. NMR-based lipidomic approach to evaluate controlled dietary intake of lipids in adipose tissue of a rat mammary tumor model. Journal of Proteome Research. 15 (3), 868-878 (2016).
  9. Li, L., et al. Mass spectrometry methodology in lipid analysis. International Journal of Molecular Sciences. 15 (6), 10492-10507 (2014).
  10. Welti, R., Wang, X. Lipid species profiling: a high-throughput approach to identify lipid compositional changes and determine the function of genes involved in lipid metabolism and signaling. Current Opinion in Plant Biology. 7 (3), 337-344 (2004).
  11. Köfeler, H. C., Fauland, A., Rechberger, G. N., Trötzmüller, M. Mass spectrometry based lipidomics: an overview of technological platforms. Metabolites. 2 (1), 19-38 (2012).
  12. Schwudke, D., Liebisch, G., Herzog, R., Schmitz, F., Shevchenko, A. Shotgun lipidomics by tandem mass spectrometry under data-dependent acquisition control. Methods in Enzymology. 433, 175-191 (2007).
  13. Ejsing, C., et al. Automated identification and quantification of glycerophospholipid molecular species by multiple precursor ion scanning. Analytical Chemistry. 78 (17), 6202-6214 (2006).
  14. Schwudke, D., et al. Top-down lipidomic screens by multivariate analysis of high-resolution survey mass spectra. Analytical Chemistry. 79 (11), 4083-4093 (2007).
  15. Carvalho, M., et al. Effects of diet and development on the Drosophila lipidome. Molecular Systems Biology. 8, 600 (2012).
  16. Han, X., Gross, R. W. Global analyses of cellular lipidomes directly from crude extracts of biological samples by ESI mass spectrometry a bridge to lipidomics. Journal of Lipid Research. 44 (6), 1071-1079 (2003).
  17. Schwudke, D., et al. Lipid profiling by multiple precursor and neutral loss scanning driven by the data-dependent acquisition. Analytical Chemistry. 78 (2), 585-595 (2006).
  18. Ejsing, C. S., et al. Global analysis of the yeast lipidome by quantitative shotgun mass spectrometry. Proceedings of the National Academy of Sciences of the United States of America. 106 (7), 2136-2141 (2009).
  19. Fauland, A., et al. A comprehensive method for lipid profiling by liquid chromatography-ion cyclotron resonance mass spectrometry. The Journal of Lipid Research. 52 (12), 2314-2322 (2011).
  20. Schuhmann, K., et al. Shotgun lipidomics on a LTQ Orbitrap mass spectrometer by successive switching between acquisition polarity modes. Journal of Mass Spectrometry. 47 (1), 96-104 (2012).
  21. Pitt, J. J. Principles and applications of liquid chromatography-mass spectrometry in clinical biochemistry. The Clinical Biochemist. Reviews. 30 (1), 19-34 (2009).
  22. Blanksby, S. J., Mitchell, T. W. Advances in mass spectrometry for lipidomics. Annual Review of Analytical Chemistry. 3 (1), 433-465 (2010).
  23. Thakur, R., et al. Phospholipase D activity couples plasma membrane endocytosis with retromer dependent recycling. eLife. 5, 18515 (2016).
  24. Donaldson, K. The inhalation toxicology of p-aramid fibrils. Critical Reviews in Toxicology. 39 (6), 487-500 (2009).
  25. Panda, A., et al. Functional analysis of mammalian phospholipase D enzymes. Bioscience Reports. 38 (6), (2018).
  26. Ghosh, A., Sharma, S., Shinde, D., Ramya, V., Raghu, P. A novel mass assay to measure phosphatidylinositol-5-phosphate from cells and tissues. Bioscience Reports. 39 (10), (2019).
  27. Balakrishnan, S. S., et al. Regulation of PI4P levels by PI4KIIIα during G-protein-coupled PLC signaling in Drosophila photoreceptors. Journal of Cell Science. 131 (15), (2018).
  28. Masood, M. A., Yuan, C., Acharya, J. K., Veenstra, T. D., Blonder, J. Quantitation of ceramide phosphorylethanolamines containing saturated and unsaturated sphingoid base cores. Analytical Biochemistry. 400 (2), 259-269 (2010).
  29. Acharya, U., et al. Modulating sphingolipid biosynthetic pathway rescues photoreceptor degeneration. Science. 299 (5613), 1740-1743 (2003).
  30. Hebbar, S., Schuhmann, K., Shevchenko, A., Knust, E. Hydroxylated sphingolipid biosynthesis regulates photoreceptor apical domain morphogenesis. The Journal of Cell Biology. 219 (12), (2020).
  31. Abhyankar, V., Kaduskar, B., Kamat, S. S., Deobagkar, D., Ratnaparkhi, G. S. Drosophila DNA/RNA methyltransferase contributes to robust host defense in aging animals by regulating sphingolipid metabolism. The Journal of Experimental Biology. 221, (2018).
  32. Subramanian, M., et al. Altered lipid homeostasis in Drosophila InsP3 receptor mutants leads to obesity and hyperphagia). DMM Disease Models and Mechanisms. 6 (3), 734-744 (2013).
  33. Guan, X. L., et al. Biochemical membrane lipidomics during Drosophila development. Developmental Cell. 24 (1), 98-111 (2013).
  34. Caravaca, J. M., Lei, E. P. Maintenance of a Drosophila melanogaster Population Cage. Journal of Visualized Experiments: JoVE. (109), e53756 (2016).
  35. Fujita, S. C., Inoue, H., Yoshioka, T., Hotta, Y. Quantitative tissue isolation from Drosophila freeze-dried in acetone. TheBiochemical Journal. 243 (1), 97-104 (1987).
  36. Herzog, R., et al. LipidXplorer: a software for consensual cross-platform lipidomics. PloS One. 7 (1), 29851 (2012).
  37. Itoh, Y. H., Itoh, T., Kaneko, H. Modified Bartlett assay for microscale lipid phosphorus analysis. Analytical Biochemistry. 154 (1), 200-204 (1986).
  38. Hsu, F. F., Turk, J. Charge-driven fragmentation processes in diacyl glycerophosphatidic acids upon low-energy collisional activation. A mechanistic proposal. Journal of the American Society for Mass Spectrometry. 11 (9), 797-803 (2000).
  39. Acharya, J. K., et al. Cell-nonautonomous function of ceramidase in photoreceptor homeostasis. Neuron. 57 (1), 69-79 (2008).
  40. Dasgupta, U., et al. Ceramide kinase regulates phospholipase C and phosphatidylinositol 4, 5, bisphosphate in phototransduction. Proceeding of the National Academy of Science of the United States of America. 106 (47), 20063-20068 (2009).
  41. Yonamine, I., et al. Sphingosine kinases and their metabolites modulate endolysosomal trafficking in photoreceptors. The Journal of Cell Biology. 192 (4), 557-567 (2011).
  42. Katz, B., Gutorov, R., Rhodes-Mordov, E., Hardie, R., Minke, B. Electrophysiological Method for Whole-cell Voltage Clamp Recordings from Drosophila Photoreceptors. Journal of Visualized Experiments: JoVE. (124), e55627 (2017).
  43. Pak, W. L. Drosophila in vision research. The Friedenwald Lecture. Investigative Ophthalmology & Visual Science. 36 (12), 2340-2357 (1995).
  44. Shevchenko, A., Simons, K. Lipidomics: coming to grips with lipid diversity. Nature Reviews. Molecular Cell Biology. 11 (8), 593-598 (2010).
  45. Garcia-Murillas, I., et al. lazaro encodes a lipid phosphate phosphohydrolase that regulates phosphatidylinositol turnover during Drosophila phototransduction. Neuron. 49 (4), 533-546 (2006).
  46. Raghu, P., et al. Rhabdomere biogenesis in Drosophila photoreceptors is acutely sensitive to phosphatidic acid levels. TheJournal of Cell Biology. 185 (1), 129-145 (2009).
  47. Ogiso, H., Suzuki, T., Taguchi, R. Development of a reverse-phase liquid chromatography electrospray ionization mass spectrometry method for lipidomics, improving detection of phosphatidic acid and phosphatidylserine. Analytical Biochemistry. 375 (1), 124-131 (2008).
  48. Raghupathy, R., et al. Transbilayer lipid interactions mediate nanoclustering of lipid-anchored proteins. Cell. 161 (3), 581-594 (2015).
  49. Randall, A. S., et al. Speed and sensitivity of phototransduction in Drosophila depend on degree of saturation of membrane phospholipids. The Journal of Neuroscience. 35 (6), 2731-2746 (2015).
  50. Schuhmacher, M., et al. Live-cell lipid biochemistry reveals a role of diacylglycerol side-chain composition for cellular lipid dynamics and protein affinities. Proceedings of the National Academy of Sciences of the United States of America. 117 (14), 7729-7738 (2020).
  51. Raghu, P. Functional diversity in a lipidome. Proceedings of the National Academy of Sciences of the United States of America. 117 (21), 11191-11193 (2020).
  52. Khandelwal, N., et al. Fatty acid chain length drives lysophosphatidylserine-dependent immunological outputs. Cell Chemical Biology. 28 (8), 1169-1179 (2021).
  53. Piper, M. D. W., et al. A holidic medium for Drosophila melanogaster. Nature Methods. 11 (1), 100-105 (2014).
  54. Wang, F., et al. FlyVar: a database for genetic variation in Drosophila melanogaster. Database: The Journal of Biological Databases and Curation. 2015, (2015).
  55. Hammond, G. R. V., Balla, T. Polyphosphoinositide binding domains: Key to inositol lipid biology. Biochimica et Biophysica Acta. 1851 (6), 746-758 (2015).
  56. Chakrabarti, P., et al. A dPIP5K dependent pool of phosphatidylinositol 4,5 bisphosphate (PIP2) is required for G-protein coupled signal transduction in Drosophila photoreceptors. PLoS Genetics. 11 (1), 1004948 (2015).
  57. Hardie, R. C., Liu, C. -. H., Randall, A. S., Sengupta, S. In vivo tracking of phosphoinositides in Drosophila photoreceptors. Journal of Cell Science. 128 (23), 4328-4340 (2015).
  58. Yadav, S., Garner, K., Georgiev, P., Li, M., Gomez-espinosa, E. RDGB, a PI-PA transfer protein regulates G-protein coupled PtdIns (4, 5)P2 signalling during Drosophila phototransduction. Journal of Cell Science. 128 (17), 3330-3344 (2015).
  59. Unsihuay, D., Mesa Sanchez, D., Laskin, L. Quantitative Mass Spectrometry Imaging of Biological Systems. Annual Review of Physical Chemistry. 72, 307-329 (2021).
This article has been published
Video Coming Soon
Keep me updated:

.

Cite This Article
Panda, A., Thakur, R., Kumari, A., Raghu, P. Analysis of Lipid Signaling in Drosophila Photoreceptors using Mass Spectrometry. J. Vis. Exp. (181), e63516, doi:10.3791/63516 (2022).

View Video