Jour

nal o

f Cel

l Sci

ence

COMMENTARY ARTICLE SERIES: IMAGING

Molecular probes to visualize the location, organization anddynamics of lipids

Masashi Maekawa1 and Gregory D. Fairn1,2,3,4,5,*

ABSTRACT

Cellular lipids play crucial roles in the cell, including in energy

storage, the formation of cellular membranes, and in signaling and

vesicular trafficking. To understand the functions and characteristics

of lipids within cells, various methods to image lipids have been

established. In this Commentary, we discuss the four main types

of molecular probes that have significantly contributed to our

understanding of the cell biology of lipids. In particular, genetically

encoded biosensors and antibodies will be discussed, and how they

have been used extensively with traditional light and electron

microscopy to determine the subcellular localization of lipids and

their spatial and temporal regulation. We highlight some of the

recent studies that have investigated the distribution of lipids

and their ability to cluster using super-resolution and electron

microscopy. We also examine methods for analyzing the

movement and dynamics of lipids, including single-particle

tracking (SPT), fluorescence recovery after photobleaching

(FRAP) and fluorescence correlation spectroscopy (FCS).

Although the combination of these lipid probes and the various

microscopic techniques is very powerful, we also point out several

potential caveats and limitations. Finally, we discuss the need for

new probes for a variety of phospholipids and cholesterol.

KEY WORDS: Electron microscopy, Imaging, Lipid, Super-

resolution

IntroductionWith more than1000 lipid species in the typical mammalian cell,understanding their functions is a daunting task and requires an

armamentarium of techniques and probes. Although some classesand species of lipids serve major structural roles within the cell,others are potent signaling molecules kept at low concentrations.

With so many individual molecules and species of lipids withinthe cell, the level of organization is truly remarkable. Not onlydoes lipid composition vary between the specific organelles

within the cell, but there also exists an asymmetric distributionbetween leaflets of the same bilayer and inhomogeneousdistribution within the same leaflet. This organization is the

result of a multitude of lipid–lipid and lipid–protein interactions,

metabolic pathways, lipid transporters and vesicular transportpathways.

Lipids perform three general functions within eukaryotic cells:they (1) operate as energy stores in the form of triacylglyceridesand cholesterol esters, (2) act as building blocks of cellularmembranes owing to their amphiphilic nature, and (3) serve as

secondary messenger and/or regulators in signal transduction andtransport processes (van Meer et al., 2008). Owing to the array offunctions and their biological importance, a great deal of time and

effort has gone into understanding, not only these lipids, but alsothe enzymes and proteins that regulate their metabolism andsubcellular distribution and function. Collectively, this research

constitutes the field of lipidomics. Lipidomics has been greatlybolstered since the early 2000s owing to advances and availabilityin mass spectrometry, computational methods and microscopy.

The ability to quantify lipids using mass spectrometry andto determine their localization in a dynamic fashion representpowerful complementary techniques.

Mass spectrometry enables us to not only perform accuratequantitative analysis of lipid classes but also allows us todistinguish molecular species of lipids due to variations in their

acyl chains (Han and Gross, 2003; Watson, 2006). Whencombined with subcellular fractionation, mass spectrometry cancatalog the lipidome of specific organelles. This approach,

although very powerful, does have a variety of limitations. Forinstance, in many of these experiments a large number of cells arerequired for robust data generation, and it takes several hours

to isolate the organelles or membranes of interest. This has ledresearchers to develop techniques in which spatial information isadded to the mass spectrometry data. Advances in this area haveled to mass-spectrometry-based imaging approaches. One

such technique uses matrix-assisted laser desorption ionization(MALDI) mass spectrometry together with controlled two-dimensional movement of the sample during the data collection

(Cornett et al., 2007). Thus, it enables the results for specific lipidspecies to be represented as an image. The major drawback of thistechnique is that, owing to limited resolution, it is more suitable

for tissue slices than individual cells. Alternative approacheshave been developed that achieve sub-micron resolution andtherefore are more desirable for cell biologists. Recent technical

developments in time-of-flight, secondary ion mass spectroscopy(SIMS) have enabled the analysis of both the composition andorganization of lipids with a lateral resolution of ,500 nm(Touboul et al., 2011). The use of next-generation SIMS and

incorporating the use of stable-isotope-labeled lipids has allowedresearchers to obtain a lateral resolution of 100 nm for a fewclasses of lipids (for more information, see Kraft and

Klitzing, 2014). Although these are great tools, it remainsdifficult to examine dynamic and asynchronous processes,such as cell migration or macropinocytosis in cell populations.

This is especially true for low-abundance lipids, such as

1Keenan Research Centre for Biomedical Science, St. Michael’s Hospital, 209Victoria Street, Toronto, ON M5S 1T8, Canada. 2Department of Surgery,University of Toronto, Toronto, ON M5T 1P5, Canada. 3Department ofBiochemistry, University of Toronto, Toronto, ON M5S 1A8, Canada. 4Institute ofMedical Science, Faculty of Medicine, University of Toronto, Toronto, ON M5S1A8, Canada. 5Institute for Biomedical Engineering and Science Technology(iBEST) at Ryerson University and St. Michael’s Hospital, Toronto, ON M5B 2K3,Canada.

*Author for correspondence (fairng@smh.ca)

� 2014. Published by The Company of Biologists Ltd | Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

1

JCS Advance Online Article. Posted on 1 September 2014 JCS Advance Online Article. Posted on 1 September 2014

Jour

nal o

f Cel

l Sci

ence

phosphoinositides and diacylglycerol (DAG), that can act aspotent signaling molecules but have short lifetimes and precise

sites of action. Furthermore, mass spectrometry has limitedability to elucidate the transbilayer distribution of phospholipidsand cholesterol in the inner (or cytosolic) and the outer (orluminal) leaflets of cellular membranes or other membrane areas

with spatial inhomogeneities.Microscopic visualization of lipids represents a powerful

technique that is complementary to mass-spectrometry-based

analysis. However, with the exception of a few lipid molecules,such as dehydroergosterol and cholestatrienol, the vast majorityof lipids are not intrinsically fluorescent (Rogers et al., 1979).

Therefore a variety of surrogates and probes have been identifiedor generated to facilitate the microscopic analysis of lipids. Theseprobes can be combined with other lipid probes of varying

spectral properties or with fluorescent proteins for colocalizationstudies. In addition, many of these probes are suitable for live-cellimaging and are very useful for monitoring the subcellularlocalization and dynamics of specific lipids. Notably, this method

enables the monitoring of specific lipids in processes such asphagocytosis and cell polarization (Yeung et al., 2008; Fairnet al., 2011a; Sarantis and Grinstein, 2012). However, one

limitation of this approach is that, at the current time, theseprobes cannot distinguish the acyl chains of glycerolipids. Thus toprovide unambiguous conclusions, we recommend performing

complementary studies (i.e. using both mass-spectrometry andvisualization of lipids) whenever possible.

This Commentary will review approaches for imaging lipids

using specific lipid probes, including a discussion of the types ofavailable probes and their use with a variety of conventional and

super-resolution imaging techniques. First, we discuss generalapproaches for analysis of intracellular ‘localization’ of lipids

using fluorescent microscopy, before moving on to advancedmethods for visualization of lipids in clusters with electronmicroscopy and sub-diffraction-limited microscopy. Finally,we highlight the probes and techniques that are being used to

analyze the dynamics of lipids by fluorescence recovery afterphotobleaching (FRAP), fluorescence correlation spectrometry(FCS) and single-particle tracking (SPT).

Lipid probes and their subcellular localizationFor imaging of lipids with standard epifluorescence or confocal

microscopy, the selection of suitable probes is very important andcan vary owing to experimental design. A variety of establishedlipid probes and surrogates (Table 1) have been used to generate

a map of many of the lipids in the cell (Fig. 1). In general, probesused for the visualization of lipids can be divided into fourcategories: (1) fluorophore-labeled lipids, (2) antibodies, (3) toxindomains, and (4) genetically encoded protein domains. Each of

them has advantages and disadvantages associated with their use,and investigators must keep in mind that antibodies and proteindomains must compete with endogenous lipid binding proteins to

localize to the membrane. In this regard, many of these proteinsonly detect the ‘free’ or available lipids. That said, these modulardomains have been used extensively to determine the distribution

of lipids between organelles and membrane leaflets.

Fluorophore-conjugated lipidsFluorophore-conjugated lipids are a conceptually attractive wayto examine the movement and localization of the particular lipids

Table 1. Established lipid probes

Lipid Probes (Kd) Intracellular localization References

PtdSer Anti-PtdSer antibody PM; nascent phagosomes Yeung at al., 2006Lact-C2 (190 nM) PM (inner); endosomes Yeung et al., 2008Annexin Va PM (outer) Andree et al., 1990TopFluor-PS PM; endosomes Kay et al., 2012

Phosphatidic acid Sos1-PH (470 nM) PM (inner)b Zhao et al., 20072PABD Membrane rufflesc Bohdanowicz et al., 2013

DAG PKCd-C1 Golgi; PM (inner); phagosomes Colon-Gonzalez and Kazanietz, 2006Sphingomyelin Lysenin (5.3 nM) PM (outer) Yamaji et al., 1998

BODIPY–sphingomyelin PM; endosomes; Golgi Puri et al., 2001Ceramide Anti-ceramide antibody PM; Golgi Krishnamurthy et al., 2007

BODIPY–ceramide Golgi Pagano et al., 1991Lactosylceramide BODIPY–lactosylceramide PM; endosomes; Golgi Puri et al., 1999LBPA Anti-LBPA antibody Late endosomes Kobayashi et al., 1998Cholesterol Filipin PM; endosomes Bornig and Geyer, 1974

TopFluor–cholesterol Endosomes; Golgi Kleusch et al., 2012Perfringolysin O-D4 PM (outer) Shimada et al., 2002

PtdIns3P 26EEA1-FYVE (38 nMd) Early endosomes Gillooly et al., 2000p40phox-PX (0.71 nM) Ellson et al., 2001; Stahelin et al., 2003

PtdIns4P Osh2-PH (1.3 mM) PM (inner) Roy and Levine, 2004; Yu et al., 2004FAPP1-PH Golgi Godi et al., 2004

PtdIns(4,5)P2 PLCd-PH (210 nMe) PM (inner) Stauffer et al., 1998PtdIns(3,4)P2 TAPP-1-PH (5.4 nM) Membrane rufflesc Dowler et al., 2000PtdIns(3,4,5)P3 GRP-1-PH (27.3 nMe) Membrane rufflesc Gray et al., 1999

Btk1-PH Varnai et al., 1999PtdIns(3,4)P2 and PtdIns(3,4,5)P3 Akt-PH [570 nM for PtdIns(3,4)P2;

400 nM for PtdIns(3,4,5)P3]Membrane rufflesc Frech et al., 1997; Gray et al., 1999

aCa2+ is required;bduring serum stimulation;cduring phagocytosis and macropinocytosis;dKd of GST–FYVE;emeasured using free inositol phosphates (Lemmon et al., 1995; Kavran et al., 1998). LBPA, lysobisphosphatidic acid.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

2

Jour

nal o

f Cel

l Sci

ence

within the cell. They are easily delivered to cells by pre-loadingthem onto carrier molecules, such as bovine serum albumin (BSA),and can be added to the growth medium. In general, thefluorophores are conjugated to either the polar head group (i.e.

Rhodamine-labeled phosphatidylethanolamine) or to one of theacyl chains or hydrophobic components (i.e. BODIPY–cholesterol)(Axelrod et al., 1976; Li et al., 2006). One concern is that once thehead group of the phospholipid is modified with a fluorescent

His–GFP–PFO-D4(Cholesterol)

GFP–PLCδδ-PH [PtdIns(4,5)P2]

A B C D

Bodipy–LacCer Anti-LBPA

E

Cytosol

Medium or lumen

PM

Membrane ruffles

SM, Cholesterol

Medium

cis-Golgi

Early endosomes

Late endosomes

PtdSer, PtdIns(4,5)P2

PtdSerPtdIns3P

PtdSerLBPA

PtdIns(3,5)P2

DAGPtdIns4PLacCer

Recycling endosomes

PtdSer

PtdSerPA

PtdIns(3,4,5)P3PtdIns(3,4)P2

Domain of genetically encoded protein

Domain of toxins Antibody for lipids Fluorophore-labeled lipids

Fluorophore-labeled second antibody

Fluorophore Fluorophore

trans-GolgiCytosol

Fig. 1. Subcellular distribution of lipids visualized by four types of lipid probes. (A–D) The schematics at the top illustrate the different types of commonlyused lipid probes and include an example below. Note that the size of lipids and probes (especially, antibodies and toxin domains) are not to scale. Additionally,antibodies and some toxins have multivalent binding sites which enable them to bind to more than one lipid at a time. (A) BODIPY–lactosylceramide (LacCer) isused as a probe for lactosylceramide, which localizes to the trans-Golgi network (TGN). Representative image of a cell after BODIPY–LacCer was addedto the medium, followed by a 60-minute chase period in label-free medium. The yellow dotted line outlines the cell. (B) LBPA (green in the schematic) can bevisualized with anti-LBPA antibodies. Here, fixed CHO cells were incubated with mouse monoclonal anti-LBPA antibody (6C4) in the presence of 0.05% saponin,followed by incubation with fluorophore-labeled secondary antibody. As shown in the image, it localizes to late endosomes. The yellow dotted line indicates theoutline of the cell. (C) Cholesterol (orange in the schematic) localizes in the outer leaflets of PM as visualized by using the cholesterol-binding toxin PerfringolysinO (PFO). For the image, domain 4 of the PFO protein was purified (66His-tagged PFO-D4), added to cells and imaged using confocal microscopy.(D) Phosphoinositides (blue in the schematic) can be visualized by using genetically encoded protein domains. In the example shown here, PtdIns(4,5)P2 in theinner leaflets of PM was visualized by GFP–PLCd-PH in CHO cells using confocal microcopy. Scale bars: 10 mm. (E) Schematic diagram of the intracellulardistribution of lipids in the cytosolic (inner) leaflets and the luminal (outer) leaflets of cellular membranes. The subcellular distribution of lipid in different cellularmembranes varies throughout the cells. Presumably, the luminal leaflets of intracellular compartments illustrated contain a certain degree of cholesterol andsphingomyelin (SM, highlighted in orange). PA, phosphatidic acid.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

3

Jour

nal o

f Cel

l Sci

ence

molecule, the resulting phospholipid no longer retains theproperties of the parent molecule. This is also a concern when

the hydrophobic acyl chains are modified with molecules, such asthe low-polarity boron-dipyrromethene (BODIPY) or the moremoderately polar nitro-benzoxadiazolyl (NBD). As a result, thefluorophore-conjugated lipids likely have different biophysical

properties than the natural lipids and this could impact on theircharacteristics, such as the rate of intramembrane transbilayermovement (so called flip-flop), spontaneous intervesicle or

organelle transfer, and lateral packing (Chattopadhyay andLondon, 1987; Martin and Pagano, 1987; Abrams and London,1993; Martin and Pagano, 1994; Kaiser and London, 1998;

Baumgart et al., 2007; Elvington and Nichols, 2007; Kay et al.,2012; Sezgin et al., 2012). Although these concerns might not becrucial for some experiments, it should also be noted that not all of

these molecules are metabolically inert, and it is impossible todistinguish between the original fluorophore-labeled lipids andtheir metabolites that are generated by various enzymes (e.g.phospholipases) in vivo. Thus, although fluorescently labeled lipids

are still useful for in vitro assays, their in vivo use is somewhatmore limited.

AntibodiesAntibodies have been widely used for determining thelocalization and abundance of proteins in immunocytochemistry

and immunohistology. Similarly, there are a few antibodiesthat recognize endogenous unlabeled lipids, with a classicexample being the anti-lysobisphosphatidic acid (LBPA)

antibody (Kobayashi et al., 1998). However, the visualizationof lipids with antibodies often requires the cells to be fixed andpermeabilized for the antibody to gain access to intracellularcompartments. Care must be taken, because the fixation and

permeabilization procedure might not only result in unintendedremoval and redistribution of lipids but also in differencesin accessibility between different organelles. This concept is

illustrated in a study by Irvine and co-workers that usedformaldehyde and glutaraldehyde to fix and saponin topermeabilize cells for immunofluorescence detection of

phosphoinositides in the plasma membrane (Hammond et al.,2009a). Alternatively, to examine phosphoinositides in the Golgi,the authors used a lower concentration of formaldehyde,no glutaraldehyde and permeabilized cells using digitonin

(Hammond et al., 2009a). Yet despite this attention to detail, ithas been demonstrated that integral membrane proteins and lipidsare very difficult, if not impossible, to truly fix (Tanaka et al.,

2010). Beyond the issues with fixation, it must be noted thatantibodies are significantly larger than phospholipids andtypically recognize two or more lipid molecules. Thus, it is

unclear whether antibodies only recognize a particular pool oflipid or whether they are able to recognize all of the lipid present.One final limitation is that, in general lipids, are poor antigens

owing to their presence in the immunized animal (Alving andRichards, 1977; Schuster et al., 1979), and currently only a fewantibodies are suitable for the identification of specific lipids,such as ceramide and LBPA. However, until alternative probes

for these lipids become available, these antibodies constitute aviable approach for their imaging.

Toxin domainsMany secreted toxins and pore-forming molecules recognizelipids in the exofacial leaflet of the plasma membrane (PM),

including the cholera toxin B subunit, lysenin and various

cholesterol-dependent cytolysins (Mizuno et al., 2011). Many ofthese toxins – or for safety and cell viability reasons, non-toxic

domains thereof – can be generated in recombinant form,either as fusion proteins with fluorescent proteins, such as GFP,or chemically conjugated to molecules such as Alexa Fluorsuccinimidyl ester. These probes have been used extensively to

study the outer leaflet of the plasma membrane, as well asthe lumen of endocytic pathway vesicles (Yachi et al., 2012).Typically, these types of probes are very specific and can be used

for live-cell imaging. However, as with antibodies, their mainlimitation is that only a few such toxins have been identified, and,moreover, they typically recognize lipids that reside within

exofacial leaflet of the PM.

Domains of genetically encoded proteinsAnalogous to toxin domains is the use of modular proteindomains as probes for lipids, and several examples are included inTable 1. These modular domains are genetically encoded proteinsthat are often expressed as fusions with GFP or mCherry. This

type of probe has proven very useful for the analysis of not onlythe intracellular localization of lipids but also the dynamicchanges in these lipids in response to stimuli (Varnai and Balla,

2006; Schlam et al., 2013; Steinberg et al., 2014). These lipid-binding domains are produced in the cytosol and translocate tomembranes where the ligands are located. Thus quantification of

membrane lipid changes should be measured as the ratio offluorescence intensity in the membrane of interest and thecytosol. Several such probes (i.e. PLCd-PH, Akt-PH and EEA1-

FYVE) have been used to visualize the distribution andmetabolism of various phosphoinositides in vivo (Varnai andBalla, 2006). In many of these experiments, binding-deficientmutant versions of the probe can be used as a negative control.

Alternatively, the levels of the lipid in question can be depletedby targeted knockdown of the biosynthetic enzyme or bytreatment with pharmacological inhibitors to ensure the probe is

responding properly to the changes. However, caution must beused when interpreting the results obtained as their affinity andspecificity can vary greatly. Furthermore, highly expressed

probes can potentially bind to a significant fraction of the lipid,thereby preventing its normal function (Gillooly et al., 2000;Fairn et al., 2007; Lemmon, 2008).

Collectively, these four types of lipid probes have allowed lipid

researchers and other cell biologists to examine the subcellularlocalization of specific lipid classes and determine how lipids andelectrostatics contribute to organelle identity. Beyond the

subcellular distribution of these probes using standard lightmicroscopy, some of these have been used for high-resolutionelectron and super-resolution microscopy. With the increasing

availability of super-resolution instruments and the ever-growinglist of lipid probes, it should be possible to obtain considerablymore detailed information regarding lipid clustering and

their inhomogeneity in the future. In the remainder of thisCommentary, we will highlight a number of important findingsthat have been achieved using such probes in combination withhigh-resolution and dynamic live-cell imaging techniques to

determine the lateral movement and confinement of lipids.

Electron microscopic imaging of lipidsElectron microscopy (EM) has been used extensively in the fieldof cell biology and was a breakthrough technique used in manypioneering studies. Although EM is not compatible with live-cell

imaging, its high resolution (,0.2 nm) enables the observation of

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

4

Jour

nal o

f Cel

l Sci

ence

intracellular structures and localization of molecules in greatdetail (Palade, 1952; Palade and Porter, 1954; Knott and Genoud,

2013). However, to date, EM has not been extensively used toexamine the localization of lipids. The major limitation is thatmany of the chemical fixation procedures and processingprotocols used to perform EM of proteins do not allow to

sufficient preservation of membrane lipids. However, more recentprotocols that involve the use of rapid freezing followed bydehydration (freeze substitution) or freeze fracturing make it

possible to maintain phospholipids in their native (or near-native)state (van Genderen et al., 1991; Voorhout et al., 1991; Mobiuset al., 2002) and they have been used to visualize antibodies,

toxins and genetically encoded probes of lipids in order to obtaina high-resolution map of the lipids under investigation in cellularmembranes.

In general, two types of methods can be used to monitor lipidsusing EM, pre-fixation or post-fixation labeling of lipids.Although the addition of the probes that recognize the specificlipids occur at different times, in both cases, the probes are

detected after fixation by using an antibody or protein Aconjugated to colloidal gold. Pre-fixation detection of lipids isaccomplished by using a plasmid-based biosensor that is

expressed in the cytosol as an epitope-tagged fusion to allowfor visualization. Examples of this strategy include the use of atandem FYVE domain to visualize phosphatidylinositol 3-

phosphate (PtdIns3P), the pleckstrin homology (PH) domainfrom phospholipase C for phosphatidylinositol 4,5-bisphosphate[PtdIns(4,5)P2] and the Lact-C2 domain for phosphatidylserine

(PtdSer) (Fairn et al., 2011b; Zhou et al., 2014). During thechemical fixation process, the protein-based probes arecrosslinked to proteins in their vicinity, and thus maintain theiroriginal distribution. Therefore, even if the lipids of interest are

not maintained during the rapid fixing process, these proteinprobes will provide an accurate representation of the lipids. Forexample, using the Lact-C2 and AKT-PH to visualize PtdSer

and phosphatidylinositol 3,4,5-trisphosphate [PtdIns(3,4,5)P3],respectively, EM analysis has demonstrated that both lipids areenriched in the plasma membrane over other organelles (Fairn

et al., 2011b; Zhou et al., 2014). However, the enhancedresolution of EM revealed that both of these probes formclusters in the PM, and that PtdSer is enriched in caveolae (Fairnet al., 2011b; Zhou et al., 2014). Taken together, these studies

have provided details regarding the localization of lipids thatcannot be obtained with conventional light microscopy. However,one limitation of this approach is that because the probes are

expressed in the cytosol they only have access to the cytosolicleaflet of organelles.

Post-fixation labeling and detection offers the advantage that

the probe has access to both leaflets of the membrane bilayer. Ingeneral, the procedures involve rapid freezing, either freezesubstitution and embedding in a resin, or freeze-fracture replica

formation. Following either of these preparations, the cellsections are labeled with the probe of interest. During thisprocessing, care must be taken to maintain the lipids in theirproper location while preserving their accessibility to the

recombinant probe (van Genderen et al., 1991; Voorhout et al.,1991; Mobius et al., 2002; Fujita et al., 2009; Tanaka et al., 2010;Fairn et al., 2011b). Another advantage of this approach is that

other probes, such as antibodies and potentially toxins, can alsobe used to detect the lipid under investigation. For instance, use ofan anti-LBPA antibody has demonstrated that there is an

enrichment of LBPA in late endosomes over early endosomes

(Kobayashi et al., 1998). This approach has also been used toexamine the localization of PtdIns3P and PtdSer using purified

recombinant GST–26EEA1-FYVE and GST–Lact-C2 probes,respectively (Gillooly et al., 2000; Fairn et al., 2011b). In thePtdIns3P study, it was shown that a large portion of the PtdIns3P

formed on the limiting membrane of endosomes is internalized

into intraluminal vesicles during multi-vesicular body formation(Gillooly et al., 2000; Fig. 2A). Owing to the small size of thisendocytic structure, fluorescence microscopy is unable to

distinguish a signal emanating from the limiting membrane orthe lumen of this compartment. Thus, post-fixation labeling of thePtdIns3P suggests that a large proportion of PtdIns3P could be

consumed during intraluminal body formation (Gillooly et al.,2000).

Recently, a freeze-fracture labeling approach was used to

examine PtdIns3P distribution during autophagosome formationin yeast and mammalian cells (Cheng et al., 2014; Fujimoto et al.,2014). Using the phox homology (PX) domain of p40phox (alsoknown as NCF4) in yeast, it has been shown that PtdIns3P is

enriched on the luminal leaflet of the autophagosomal isolationmembrane. Surprisingly, when the experiment was performedwith mammalian cells, the PX domain detected PtdIns3P in the

cytosol-facing leaflet (Cheng et al., 2014; Fujimoto et al., 2014).These observations suggest that although many of the regulatorsinvolved in autophagy are conserved between yeast and humans

the precise molecular mechanisms are possibly different.

Super-resolution microscopy of lipidsRecent developments in fluorescence microscopy techniqueshave allowed researchers to overcome the limitations that areimposed by the diffraction boundary of conventional fluorescencemicroscopy. The series of recent advancements in super-

resolution microscopy has resulted in a decrease in the limits ofoptical resolution from ,250 nm to ,10 nm (Galbraith andGalbraith, 2011). For the imaging of lipids at this resolution, a

variety of ensemble or single-molecule techniques have beenused, For instance, by using photoactivated localizationmicroscopy (PALM), a recent study has shown that there are

clusters of cholesterol and sphingomyelin in the outer leaflets ofPM by labeling with recombinant domains of the toxins Dronpa–PFO-D4 and Dronpa–lysenin, respectively (Fig. 2B) (Mizunoet al., 2011). Making use of stimulated emission depletion

(STED), it could be shown that PtdIns(4,5)P2 is highly enrichedwithin segregated domains in the PM (Fig. 2C) (van den Bogaartet al., 2011). In this particular study, PtdIns(4,5)P2 was visualized

using both a recombinant citrine-tagged GST–PLCd-PH(citrine is a YFP analog) protein and anti-PtdIns(4,5)P2

antibody. Furthermore, a similar study used stochastic optical

reconstruction microscopy (STORM) imaging to demonstrate thepresence of PtdIns(4,5)P2 and PtdIns(3,4,5)P3 in the PM by usinganti-phospholipid antibodies that are directly conjugated to Alexa

Fluor 647 (Wang and Richards, 2012). Collectively, the ability tovisualize lipids using EM and super-resolution microscopyconstitutes a powerful tool to help to better understand theorganization of signaling hubs these lipids are part of.

Approaches to visualize lipid dynamicsCellular membranes, with the exception of lipid droplets,

are typically lipid bilayers formed by the amphiphilicity ofglycerolipids and cholesterol. The weak hydrophobic interactionbetween lipids in membranes enables the lateral diffusion of

lipids and proteins in membranes (Singer and Nicolson, 1972;

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

5

Jour

nal o

f Cel

l Sci

ence

Holthuis and Levine, 2005; van Meer et al., 2008). As mentionedabove, some lipids form clusters in the PM, and these clusters are

likely to be important for signal transduction and other functionsof membrane proteins. It has been suggested that lateral diffusionof lipids in cellular membranes regulates the formation anddispersion of these lipids nanodomains and thus the functions of

membrane proteins (Simons and Toomre, 2000; Sprong et al.,2001; Lingwood and Simons, 2010). Therefore, characterizingthe dynamics of lipids in membranes and elucidation of the

mechanisms that regulate this motion are crucial. In addition,recent evidence has demonstrated that the plasma membrane iscompartmentalized by actin-dependent membrane protein fences

(Kusumi et al., 2011). When proteins or lipids are imaged at highacquisition rates, they display a non-Brownian diffusion patternowing to the obstacles that temporarily restrict their movements

(Ritchie et al., 2005). A variety of optical techniques (andvariations of the theme) have been used to determine themovement of lipids and proteins in the plane of the membrane,including FRAP, FCS and SPT. FRAP and FCS approaches

are frequently used to measure the long-range diffusion andfluctuations of a population of fluorescent molecules,respectively, whereas SPT is a robust method used to analyze

the mobility of individual lipids (or proteins) in the PM of livingcells and in model membranes. For these techniques, fluorophore-labeled lipids, toxin domains and genetically encoded proteins are

suitable as lipid probes. Below, we will highlight examples of theuse of each of these techniques and describe how they cangenerate complementary data sets.

FRAPFRAP analysis with lipid probes can be performed by standard laserscanning or spinning-disk confocal microscopy to irreversibly bleach

the fluorescence of probes in a region of interest (Axelrod et al., 1976;Corbett-Nelson et al., 2006; Hammond et al., 2009b; Kay et al., 2012).In some cases, total internal reflection fluorescence (TIRF)

microscopy is also used for FRAP analysis (Axelrod, 2001;Hammond et al., 2009b). The region that has been photo-bleachedis then monitored by time-lapse microscopy to determine the kinetics

of the recovery of the fluorescent signal (Lippincott-Schwartz et al.,2003; Lippincott-Schwartz and Patterson, 2003). As several of thelipid probes used for FRAP have a high rate of association anddisassociation, additional post-acquisition analysis must be performed

to elucidate the actual diffusion rate of the lipid under investigation(Fig. 3A; for more details, see Hammond et al., 2009b).

Recent studies have investigated the lateral diffusability of

PtdIns(4,5)P2 in the plasma membrane by FRAP analysisusing both protein-based probes and fluorophore-labeledlipids (Hammond et al., 2009b; Golebiewska et al., 2011).

Plasmalemmal PtdIns(4,5)P2 is a spatial signal and contributes tothe identity of the PM through the recruitment of effectors. Bydetermining the diffusion rates of PtdIns(4,5)P2 in the membrane,

researchers can estimate the area an effector protein can surveywhile it is bound to PtdIns(4,5)P2 if the rate of dissociation isknown. Using the GFP–PLCd-PH and BODIPY–PtdIns(4,5)P2,the lateral diffusion (D) was found to be ,1 mm2/s. The

disassociation time (t) of the GFP–PLCd-PH from this lipid isestimated to be 2.4 s (Hammond et al., 2009b). Therefore, thedistance traveled by the protein while bound to the lipid can be

estimated by the equation !(26D6t) (Teruel and Meyer, 2000).This suggests that PLCd or other effectors with comparableaffinity for this lipid can potentially travel 1–3 mm in the plane of

the membrane before a significant fraction of the protein

B

A

BDronpa−Lysenin

CAnti−PtdIns(4,5)P2 antibody

Confocal STED

PALM

EMGST−2xFYVE

Fig. 2. High-resolution imaging of lipids using EM, PALM and STED.(A) EM images of localization of PtdIns3P labeled with GST–26FYVE areshown. An ultrathin frozen section of fixed baby hamster kidney cells wasincubated with GST–26FYVE. After washing and fixation, samples wereincubated with anti-GST antibody and then 10-nm gold particles conjugatedto protein A (shown by arrows). Clear cytoplasmic coats of early endosomesare shown by arrowheads (left). The multi-vesicular bodies show labeling ofGST–26FYVE concentrated on the internal membranes (right). Scale bars:200 nm. Images reproduced from Gillooly et al., 2000 with permission fromEMBO J. (B) SM-enriched domains in the outer leaflets of PM werevisualized by PALM. HeLa cells were incubated with recombinant Dronpa–N-terminal truncated Lysenin at room temperature for 10–60 min, washedquickly, fixed with 4% formaldehyde in PBS and observed with PALMmicroscopy. Expansion of the region indicated by the white box is shown onthe right side. Scale bars: 500 nm. Images reproduced from Mizuno et al.,2011 with permission from The Royal Society of Chemistry. (C) Confocal(left) and corresponding STED (right) images of PtdIns(4,5)P2 in the cytosolicleaflets of PM are shown. PtdIns(4,5)P2-enriched domains in the cytosolicleaflets of PM are visualized with high resolution by STED. The resolution isincreased in the STED image compared to the corresponding confocalimage. Membrane sheets of PC12 cells were prepared by rupturing the cellswith probe sonication as described previously (Sieber et al., 2006). Then, themembrane sheets were immunostained with a monoclonal anti-PtdIns(4,5)P2

antibody (2C11) and a secondary antibody labeled with Alexa Fluor 488, andobserved with STED microscopy. Scale bars: 5 mm. Images reproduced fromvan den Bogaart et al., 2011 with permission from Nature.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

6

Jour

nal o

f Cel

l Sci

ence

disassociates (Hammond et al., 2009b) This ability to diffuse inthe plane of the membrane is likely an important contributor to

the formation of protein–protein interactions and macromolecularcomplexes (Fig. 3B; Golebiewska et al., 2011). The study ofHammond et al., did not characterize the presence of a diffusionbarrier for PtdIns(4,5)P2, but the study using the BODIPY–

PtdIns(4,5)P2 demonstrated that this lipid does not diffuse out ofthe forming phagosome, suggesting that the there is a diffusionbarrier at the tips of the pseudopods that drives phagosome

formation (Golebiewska et al., 2011).A similar study examined the dynamics of PtdSer using

TopFluor–PtdSer and GFP–Lact-C2 using FRAP in HeLa cells

(Kay et al., 2012). The mean diffusion coefficient of expressedGFP–Lact-C2 in the PM of HeLa cells was determined to be0.33 mm2/s compared to 0.49 mm2/s for the TopFluor–PtdSer

(Kay et al., 2012). This value is substantially lower than thediffusion coefficients for the PtdIns(4,5)P2 probe (GFP–PLCd-

PH), suggesting that PtdSer might interact with other lipids orproteins to limit its diffusion. Support of this concept is providedby the fact that only 43% of the TopFluor–PtdSer is mobile in theFRAP experiments. Taken together, the results suggest that there

is a fraction of PtdSer that is either immobile or greatly restrictedin its lateral mobility. One possibility is that the PtdSer clustersobserved by EM might in fact be sequestered PtdSer molecules

that are confined by interactions with cholesterol and/or proteincomplexes (Fairn et al., 2011b; Kay et al., 2012).

FCSFCS encompasses many related techniques to analyzefluctuations in fluorescence for populations or individual

A

C

(iii)

(i) (ii) (iii)

Diffusion

Association/dissociation

Pre-bleaching Bleaching Recovery

Membrane

Medium orcytosol

Lipid probe (lipid−binding domain)

Bleached lipid probe

B(i) (ii) (iii)

Diffusion

Pre-bleaching Bleaching Recovery

Membrane

Medium orcytosol

Lipid probe (fluorophore labeled lipids)

Bleached lipid probe

GFP−Lact-C2 Raw data

Classified tracks

(iii)1.0

0.2

0.4

0.8

0.6

Confined

Free

Frac

tion

of tr

acks

Fig. 3. Dynamics of lipids analyzed byFRAP and SPT. (A,B) Schematic diagramof the FRAP analysis. The fundamentaldifferences of the mode of recovery forthese types of lipid probes are shown(Hammond et al., 2009b; Kay et al., 2012).(A) Following the photobleaching of thelipid-binding domains that are produced inthe cytosol, recovery will occur by twomeans; diffusion of probes while they arebound to their lipid ligand (a red two-wayarrow in iii) and dissociation of bleachedprobes and association of unbleachedprobe from the cytosolic pool (a blue two-way arrow in iii). (B) Using fluorophore-labeled lipids, probes in both leaflets arephoto-bleached (ii) before fluorescence isrecovered through the diffusion of the lipidsonly (a red two-way arrow in iii). (C) SPTanalysis of GFP–Lact-C2 in the cytosolicleaflets of PM. GFP–Lact-C2 wastransfected to HeLa cells and SPT usingTIRF microscopy was performed. Rawimages and data are reproduced from Kayet al., 2012 with permission from Molecular

Biology of the Cell. Representative imagesequences (i) and track classification (ii) ofGFP–LactC2 on the inner leaflet of the PMof HeLa cells are shown. Scale bar: 2 mm.(iii) The graph shows the breakdown of thetracks illustrated in (i).

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

7

Jour

nal o

f Cel

l Sci

ence

labeled molecules diffusing into and out of an excitation region.This excitation region can be a focal-spot or line-scan that is

monitored over time using either single- or two-photon excitation.(Fahey et al., 1977; Korlach et al., 1999; Schwille et al., 1999;Haustein and Schwille, 2007). Line-scan FCS has been used toexamine both the diffusion of BODIPY–cholesterol in liquid

ordered and disordered domains in phase-separating modelmembranes. Using a variety of phosphatidylcholine (PtdCho),sphingomyelin and cholesterol model membranes, a study found

that the diffusion coefficient was approximately an order ofmagnitude higher in liquid-disordered regions compared to theliquid-ordered regions (Ries et al., 2009). When the same

experiment was conducted with HEK cells, the BODIPY–cholesterol had a diffusion coefficient of 0.33 mm2/s, which iscomparable to that of PtdSer. The dynamics of BODIPY–

PtdIns(4,5)P2 in the inner leaflets of PM and in formingphagosomes has also been analyzed with FCS and was found tohave a diffusion coefficient of ,1 mm2/s, in both formingphagosomes and the PM of J774 macrophages (Golebiewska

et al., 2008; Golebiewska et al., 2011).FCS analysis of lipids can be influenced by a number of

factors, including the relative amount of cytosol in the region of

interest. For example, FCS analysis detected both a fast and slowpopulation of molecules in the study using the TopFluor–PtdSerand GFP–Lact-C2 probes in the PM of HeLa cells (Kay et al.,

2012). The diffusion coefficient of the slow fraction likelyreflects the population of TopFluor–PtdSer in the membrane andthe GFP–Lact-C2 bound to PtdSer, respectively. By contrast,

the fast-moving fraction of these probes is likely a solubledegradation product of the TopFluor–PtdSer and the unbound(cytosolic) GFP–Lact-C2. Thus, understanding the nature of theprobe used can assist in the interpretation of the FCS results.

SPTSPT techniques have been used largely in the field of membrane

biophysics to determine the lateral diffusion of lipids and proteinsin both PM and supported bilayers (Saxton and Jacobson, 1997;Martin et al., 2002). TIRF microscopy can be combined with

SPT, and this system is particularly sensitive for measuring thedynamics of lipid probes at the surface of either model

membranes (in vitro) or living cells (in vivo) (Knight andFalke, 2009; Kay et al., 2012). Although FRAP is able to

determine the average dynamics of hundreds or thousands ofmolecules, SPT measures the dynamics of individual moleculesin considerably shorter time intervals and with greater precision.Thus, lipid subpopulations that cannot be identified using FRAP

can be categorized by SPT analysis (Saxton and Jacobson, 1997).In SPT, time-lapse sequences of images of probes are taken withfluorescence microscopy and analyzed with a variety of

algorisms. Typically, the position and intensity of a particle isdetermined by detecting local maximum intensities and thenfitting Gaussian curves that approximate the two-dimensional

point spread function of the microscope (Jaqaman et al., 2008;Flannagan et al., 2010; Jaqaman et al., 2011). Using SPT, a studycompared the mobility of two PtdSer probes, TopFluor–PtdSer

and GFP–Lact-C2, in the PM (Kay et al., 2012). Interestingly,only 29% of the TopFluor–PtdSer is freely diffusible with theremaining 71% being confined to smaller areas that have anaverage radius of 111 nm. Whereas the analysis of the GFP–Lact-

C2 probe showed that 78% of the GFP–Lact-C2 displays freediffusion and the remaining 22% are confined within an averagearea of 360 nm (Fig. 3C) (Kay et al., 2012). Clearly the two

different PtdSer probes are generating opposing data, but thequestion is why? Like all translocation-based protein domainprobes, Lact-C2 can only bind to the unengaged head group and

therefore ‘available’ PtdSer. Therefore, the results would suggestthat the majority (,80%) of the PtdSer that is unbound orunshielded by proteins is freely diffusible. By contrast, the acyl-

chain-modified TopFluor–PtdSer can be identified and trackedregardless of it binding to proteins or being in complexes such ascaveolae. The TopFluor–PtdSer result is consistent with a largepercentage of the plasmalemmal PtdSer being bound or occupied

by proteins or membrane domains. This reinforces the notion thatresearchers must be aware of the nature of the probe being used tohave a full appreciation of the results. These differences also

demonstrate the utility of using multiple probes for a specificlipid to gain a better understanding of its behavior in vivo.Similar experiments have been used to follow the dynamics

of PtdIns(3,4,5)P3 on supported lipid bilayers (Knight andFalke, 2009). Here, PtdIns(3,4,5)P3 was monitored using an

FilipinA B

Topfluor−Cholesterol DehydroergosterolC

Fig. 4. Intracellular distribution of cholesterol visualized by established cholesterol probes. (A) Filipin allows visualization of cholesterol in the PM,recycling endosomes and other intracellular vesicles. For filipin staining, CHO cells were fixed with 3.7% formaldehyde in PBS, then cells were treated with0.5 mg/ml filipin at 4˚C for 16 h. Images were acquired by laser scanning confocal with a 405-nm laser. (B) TopFluor–cholesterol accumulates in intracellularvesicles after incubation at 37˚C. A total of 10 nmol/ml TopFluor–cholesterol in PBS containing 1% BSA was loaded to CHO cells for 10 min on ice. After washingwith PBS, cells were cultured at 37˚C for 10 min, then cells were fixed with 3.7% formaldehyde in PBS. Images were acquired using confocal microscopy.(C) DHE accumulates in PM and recycling endosomes. TRVb1 cells, a CHO cell line that expresses human transferrin receptor, were labeled with a DHE–methyl-b-cyclodextrin complex for 1 min, washed and chased at 37˚C for 30 min, then images were obtained by live-cell imaging (Mondal et al., 2009). Images inA and B were obtained with laser scanning confocal microscopy. The image in C was obtained with a non-confocal inverted fluorescence microscopy optimizedfor DHE imaging as described by Mukherjee et al., 1998. Reproduced from Mondal et al., 2009 with permission from Molecular Biology of the Cell. Scale bars:10 mm.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

8

Jour

nal o

f Cel

l Sci

ence

Alexa-Fluor-555-conjugated GST–GRP1-PH in order todetermine both the diffusion coefficient for the PtdIns(3,4,5)P3

and dissociation rate of the probe (Knight and Falke, 2009). Thistype of in vitro analysis will allow researchers to move to in vivo

systems and gain a better understanding of PtdIns(3,4,5)P3

signaling. Overall, SPT analysis is a powerful tool to observe

lipid (and protein) dynamics and to obtain new insights intothe molecular mechanisms of lipid–lipid and/or lipid–proteininteractions in membranes.

Future perspectivesIn this Commentary, we have highlighted established probes for

a variety of lipids (Table 1); however, lipids probes are notcurrently available for all classes of lipids. For instance, protein-based probes for visualizing intracellular phosphatidylcholine and

phsophatidylethanolamine (PtdEtn) have not yet been identified.It has been reported that the short polypeptides duramycin andcinnamycin specifically bind to PtdEtn (Navarro et al., 1985;Choung et al., 1988), and cinnamycin has been used visualize

PtdEtn in the outer leaflets of PM during cytokinesis andapoptosis (Emoto et al., 1996; Makino et al., 2003). However,these molecules cause cell lysis at high concentrations.

Additionally, fixation and/or permeabilization are necessary forthe compounds to gain access to intracellular compartmentsmaking them less than ideal. In addition, although there are

many well-established probes for other phosphoinositides,specific probes for phosphatidylinositol 3,5-bisphosphate[PtdIns(3,5)P2] and phosphatidylinositol 5-phosphate [PtdIns5P]

have been lacking. However, the cytosolic N-terminal polybasicdomain of TRPML1 (ML1N) has recently been reported tospecifically bind to PtdIns(3,5)P2 (Li et al., 2013). This domainawaits further characterization but appears to be suitable to

analyze the intracellular localization of PtdIns(3,5)P2 usinglight microscopy. Clearly, the isolation of probes with greaterspecificity and/or affinity will provide more accurate information

about lipids or specific pools of lipids. A prime example of this isthe characterization of a new phosphatidylinositol 4-phosphate[PtdIns4P] probe, P4M that demonstrates the presence of this

lipid in late endosomes/lysosomes (Hammond et al., 2014).Cholesterol is one of the most abundant lipid molecules in the

PM, the cell and within the body. Recent studies have establisheda concept that cholesterol forms membrane nanodomains (lipid

rafts) in the outer leaflets of PM (Lindwood and Simons, 2010).Lipid rafts represent a scaffold where specific proteins assembleand play important roles in many cells functions such as signal

transduction and virus infection (Simons and Toomre, 2000;Simons and Ehehalt, 2002). Several probes for cholesterol havebeen established to visualize intracellular cholesterol (Fig. 4, see

also Box 1). Although these cholesterol probes can labelcholesterol in cellular membranes, the big problem is that theseprobes cannot distinguish cholesterol in the outer leaflets and in

the inner leaflets of PM (Ohno-Iwashita et al., 2010a; see alsoBox 1). Thus, design of new cholesterol probes that canspecifically visualize cholesterol in the inner leaflets of PMwould be an important future work.

Finally, recent technological advances have produced a varietyof super-resolution microscopic techniques, for example,structured illumination microscopy (SIM), STED, PALM and

STORM. However, these still have limitations that need to beovercome for their widespread use in the imaging of lipids(Galbraith and Galbraith, 2011). For instance, high image

resolution often trades off with acquisition speed, which is

important as lipids and proteins can potentially move faster thanthe speed with which super-resolution microscopes capture

images. To overcome these issues, we will need to use photo-switchable probes and small-molecule fluorophores for super-resolution microscopes, in addition to technical improvement for

temporal-spatial resolution (Fernandez-Suarez and Ting, 2008).As illustrated throughout this Commentary, BODIPY has beenused extensively to generate fluorophore conjugated lipids.

However, BODIPY has relatively poor photo-stability comparedto several other fluorescent dyes. An alternative dye that couldbecome more popular is ATTO647N, which is routinely used in

STED microscopy. ATTO647N has been used to label either thehead group or acyl chain of a variety of lipids, such as PtdEtn,sphingomyelin and the ganglioside GM1 (Eggeling et al., 2009).Indeed, the dynamic behavior of various lipids in the PM of living

cells has already been measured in greater detail with thecombined use of FCS and STED (Eggeling et al., 2009; Muelleret al., 2011), which enhanced resolution. We therefore believe

Box 1. Technical challenges for the visualization ofintracellular cholesterol

Cholesterol is an essential structural component of cellularmembranes and has pivotal roles in signal transductions andmembrane trafficking (Ikonen, 2008). Established methods forvisualization of intracellular cholesterol are discussed below.

FilipinThe antibiotic filipin is an intrinsically fluorescent polyenecompound, which specifically binds to free cholesterol (Bornigand Geyer, 1974). Filipin is routinely used for visualization of freecholesterol in fixed cells. In CHO cells, PM and endosomes arestained by filipin in CHO cells (Fig. 4A). Notably, filipin can stainfree cholesterol in all of the intracellular membrane compartmentsbecause of its membrane permeability (Miller, 1984; Ohno-Iwashitaet al., 2010b). Thus, filipin staining is not an appropriate method forlive-cell imaging. In addition, filipin photobleaches very quickly.Thus, it can be challenging to obtain high-quality images with astandard confocal microscopy that typically use 405 nm lasers.

TopFluor–cholesterol and BODIPY–cholesterolTopFluor–cholesterol is a fluorophore-labeled cholesterol (Li et al.,2006; Holtta-Vuori et al., 2008). TopFluor–cholesterol can be easilyloaded to cells with BSA and mainly localizes in endosomes in CHOcells (Fig. 4B). One concern is that the addition of the fluorophoremight alter the properties and dynamics of cholesterol and thus mightnot accurately mimic endogenous cholesterol (Solanko et al., 2013).In addition, TopFluor–cholesterol cannot distinguish the cholesterol inthe cytosolic leaflets and luminal leaflets cellular membranes.

DehydroergosterolDehydroergosterol (DHE) is a naturally fluorescent sterol and acholesterol analog. In cells loaded with DHE, most of the fluorescentsignal accumulates in the recycling endosomes and to a lesser extentin the PM in CHO cells (Fig. 4C). One advantage is that DHE issuitable for live-cell imaging experiments, but overall it is weaklyfluorescent and is typically used with a modified wide-field microscope.Previous studies using fluorescence quenchers have demonstratedthat the majority of the DHE is localized within the inner leaflet of thePM (Mondal et al., 2009). This result was surprising considering thebelief that most cholesterol in the PM is localized to the exofacialleaflet. It is unclear if the observations using DHE are true forcholesterol (Hyslop et al., 1990). However, it should be noted that thevast majority of cholesterol could be replaced by DHE without the lossof cell viability (Hao et al., 2002; Wustner, 2007; Mondal et al., 2009).

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

9

Jour

nal o

f Cel

l Sci

ence

that in the near future analysis of lipid dynamics by FCS and SPTcombined with super-resolution microscopy will provide new

insights into the cell biology of lipids and membranes.

AcknowledgementsWe thank Sergio Grinstein for helpful discussions and Takashi Hirama (bothHospital for Sick Children, Toronto, Canada) for technical advice regarding theuse of the BODIPY–LacCer. We apologize to all colleagues whose work was notcited due to space restrictions.

Competing interestsThe authors declare no competing interests.

FundingG.D.F. is a recipient of a New Investigator Salary Award from Canadian Institutesof Health Research (CIHR) and an Ontario Early Research Award. Work in ourlaboratory is supported by St. Michael’s Hospital New Investigator Start-up Fund,an operating grant from CIHR and a Discovery Grant from the Natural Sciencesand Engineering Research Council of Canada.

ReferencesAbrams, F. S. and London, E. (1993). Extension of the parallax analysis ofmembrane penetration depth to the polar region of model membranes: use offluorescence quenching by a spin-label attached to the phospholipid polarheadgroup. Biochemistry 32, 10826-10831.

Alving, C. R. and Richards, R. L. (1977). Immune reactivities of antibodies againstglycolipids – II. Comparative properties, using liposomes, of purified antibodiesagainst mono-, di- and trihexosyl ceramide haptens. Immunochemistry 14, 383-389.

Andree, H. A., Reutelingsperger, C. P., Hauptmann, R., Hemker, H. C.,Hermens, W. T. and Willems, G. M. (1990). Binding of vascular anticoagulantalpha (VAC alpha) to planar phospholipid bilayers. J. Biol. Chem. 265, 4923-4928.

Axelrod, D. (2001). Total internal reflection fluorescence microscopy in cellbiology. Traffic 2, 764-774.

Axelrod, D., Koppel, D. E., Schlessinger, J., Elson, E. and Webb, W. W. (1976).Mobility measurement by analysis of fluorescence photobleaching recoverykinetics. Biophys. J. 16, 1055-1069.

Baumgart, T., Hunt, G., Farkas, E. R., Webb, W. W. and Feigenson, G. W.(2007). Fluorescence probe partitioning between Lo/Ld phases in lipidmembranes. Biochim. Biophys. Acta 1768, 2182-2194.

Bohdanowicz, M., Schlam, D., Hermansson, M., Rizzuti, D., Fairn, G. D.,Ueyama, T., Somerharju, P., Du, G. and Grinstein, S. (2013). Phosphatidicacid is required for the constitutive ruffling and macropinocytosis of phagocytes.Mol. Biol. Cell 24, 1700-1712, , S1-S7.

Bornig, H. and Geyer, G. (1974). Staining of cholesterol with the fluorescentantibiotic ‘‘filipin’’. Acta Histochem. 50, 110-115.

Chattopadhyay, A. and London, E. (1987). Parallax method for directmeasurement of membrane penetration depth utilizing fluorescencequenching by spin-labeled phospholipids. Biochemistry 26, 39-45.

Cheng, J., Fujita, A., Yamamoto, H., Tatematsu, T., Kakuta, S., Obara, K.,Ohsumi, Y. and Fujimoto, T. (2014). Yeast and mammalian autophagosomesexhibit distinct phosphatidylinositol 3-phosphate asymmetries. Nat. Commun. 5,3207.

Choung, S. Y., Kobayashi, T., Takemoto, K., Ishitsuka, H. and Inoue, K. (1988).Interaction of a cyclic peptide, Ro09-0198, with phosphatidylethanolamine inliposomal membranes. Biochim. Biophys. Acta 940, 180-187.

Colon-Gonzalez, F. and Kazanietz, M. G. (2006). C1 domains exposed: fromdiacylglycerol binding to protein-protein interactions. Biochim. Biophys. Acta1761, 827-837.

Corbett-Nelson, E. F., Mason, D., Marshall, J. G., Collette, Y. and Grinstein, S.(2006). Signaling-dependent immobilization of acylated proteins in the innermonolayer of the plasma membrane. J. Cell Biol. 174, 255-265.

Cornett, D. S., Reyzer, M. L., Chaurand, P. and Caprioli, R. M. (2007). MALDIimaging mass spectrometry: molecular snapshots of biochemical systems. Nat.Methods 4, 828-833.

Dowler, S., Currie, R. A., Campbell, D. G., Deak, M., Kular, G., Downes, C. P.and Alessi, D. R. (2000). Identification of pleckstrin-homology-domain-containing proteins with novel phosphoinositide-binding specificities. Biochem.J. 351, 19-31.

Eggeling, C., Ringemann, C., Medda, R., Schwarzmann, G., Sandhoff, K.,Polyakova, S., Belov, V. N., Hein, B., von Middendorff, C., Schonle, A. et al.(2009). Direct observation of the nanoscale dynamics of membrane lipids in aliving cell. Nature 457, 1159-1162.

Ellson, C. D., Gobert-Gosse, S., Anderson, K. E., Davidson, K., Erdjument-Bromage, H., Tempst, P., Thuring, J. W., Cooper, M. A., Lim, Z. Y., Holmes,A. B. et al. (2001). PtdIns(3)P regulates the neutrophil oxidase complex bybinding to the PX domain of p40(phox). Nat. Cell Biol. 3, 679-682.

Elvington, S. M. and Nichols, J. W. (2007). Spontaneous, intervesicular transferrates of fluorescent, acyl chain-labeled phosphatidylcholine analogs. Biochim.Biophys. Acta 1768, 502-508.

Emoto, K., Kobayashi, T., Yamaji, A., Aizawa, H., Yahara, I., Inoue, K. andUmeda, M. (1996). Redistribution of phosphatidylethanolamine at the cleavagefurrow of dividing cells during cytokinesis. Proc. Natl. Acad. Sci. USA 93, 12867-12872.

Fahey, P. F., Koppel, D. E., Barak, L. S., Wolf, D. E., Elson, E. L. and Webb,W. W. (1977). Lateral diffusion in planar lipid bilayers. Science 195, 305-306.

Fairn, G. D., Curwin, A. J., Stefan, C. J. and McMaster, C. R. (2007). Theoxysterol binding protein Kes1p regulates Golgi apparatus phosphatidylinositol-4-phosphate function. Proc. Natl. Acad. Sci. USA 104, 15352-15357.

Fairn, G. D., Hermansson, M., Somerharju, P. and Grinstein, S. (2011a).Phosphatidylserine is polarized and required for proper Cdc42 localization andfor development of cell polarity. Nat. Cell Biol. 13, 1424-1430.

Fairn, G. D., Schieber, N. L., Ariotti, N., Murphy, S., Kuerschner, L., Webb, R. I.,Grinstein, S. and Parton, R. G. (2011b). High-resolution mapping revealstopologically distinct cellular pools of phosphatidylserine. J. Cell Biol. 194, 257-275.

Fernandez-Suarez, M. and Ting, A. Y. (2008). Fluorescent probes for super-resolution imaging in living cells. Nat. Rev. Mol. Cell Biol. 9, 929-943.

Flannagan, R. S., Harrison, R. E., Yip, C. M., Jaqaman, K. and Grinstein, S.(2010). Dynamic macrophage ‘‘probing’’ is required for the efficient capture ofphagocytic targets. J. Cell Biol. 191, 1205-1218.

Fujimoto, T., Yamamoto, H. and Ohsumi, Y. (2014). Differentphosphatidylinositol 3-phosphate asymmetries in yeast and mammalianautophagosomes revealed by a new electron microscopy technique.Autophagy 10, 933-935.

Fujita, A., Cheng, J., Tauchi-Sato, K., Takenawa, T. and Fujimoto, T. (2009). Adistinct pool of phosphatidylinositol 4,5-bisphosphate in caveolae revealed by ananoscale labeling technique. Proc. Natl. Acad. Sci. USA 106, 9256-9261.

Galbraith, C. G. and Galbraith, J. A. (2011). Super-resolution microscopy at aglance. J. Cell Sci. 124, 1607-1611.

Gillooly, D. J., Morrow, I. C., Lindsay, M., Gould, R., Bryant, N. J., Gaullier,J. M., Parton, R. G. and Stenmark, H. (2000). Localization ofphosphatidylinositol 3-phosphate in yeast and mammalian cells. EMBO J. 19,4577-4588.

Godi, A., Di Campli, A., Konstantakopoulos, A., Di Tullio, G., Alessi, D. R.,Kular, G. S., Daniele, T., Marra, P., Lucocq, J. M. and De Matteis, M. A.(2004). FAPPs control Golgi-to-cell-surface membrane traffic by binding to ARFand PtdIns(4)P. Nat. Cell Biol. 6, 393-404.

Golebiewska, U., Nyako, M., Woturski, W., Zaitseva, I. and McLaughlin, S.(2008). Diffusion coefficient of fluorescent phosphatidylinositol 4,5-bisphosphatein the plasma membrane of cells. Mol. Biol. Cell 19, 1663-1669.

Golebiewska, U., Kay, J. G., Masters, T., Grinstein, S., Im, W., Pastor, R. W.,Scarlata, S. and McLaughlin, S. (2011). Evidence for a fence that impedes thediffusion of phosphatidylinositol 4,5-bisphosphate out of the formingphagosomes of macrophages. Mol. Biol. Cell 22, 3498-3507.

Gray, A., Van Der Kaay, J. and Downes, C. P. (1999). The pleckstrin homologydomains of protein kinase B and GRP1 (general receptor for phosphoinositides-1) are sensitive and selective probes for the cellular detection ofphosphatidylinositol 3,4-bisphosphate and/or phosphatidylinositol 3,4,5-trisphosphate in vivo. Biochem. J. 344, 929-936.

Hammond, G. R., Schiavo, G. and Irvine, R. F. (2009a). Immunocytochemicaltechniques reveal multiple, distinct cellular pools of PtdIns4P andPtdIns(4,5)P(2). Biochem. J. 422, 23-35.

Hammond, G. R., Sim, Y., Lagnado, L. and Irvine, R. F. (2009b). Reversiblebinding and rapid diffusion of proteins in complex with inositol lipids serves tocoordinate free movement with spatial information. J. Cell Biol. 184, 297-308.

Hammond, G. R., Machner, M. P. and Balla, T. (2014). A novel probe forphosphatidylinositol 4-phosphate reveals multiple pools beyond the Golgi.J. Cell Biol. 205, 113-126.

Han, X. and Gross, R. W. (2003). Global analyses of cellular lipidomes directlyfrom crude extracts of biological samples by ESI mass spectrometry: a bridge tolipidomics. J. Lipid Res. 44, 1071-1079.

Hao, M., Lin, S. X., Karylowski, O. J., Wustner, D., McGraw, T. E. and Maxfield,F. R. (2002). Vesicular and non-vesicular sterol transport in living cells. Theendocytic recycling compartment is a major sterol storage organelle. J. Biol.Chem. 277, 609-617.

Haustein, E. and Schwille, P. (2007). Fluorescence correlation spectroscopy:novel variations of an established technique. Annu. Rev. Biophys. Biomol.Struct. 36, 151-169.

Holthuis, J. C. and Levine, T. P. (2005). Lipid traffic: floppy drives and asuperhighway. Nat. Rev. Mol. Cell Biol. 6, 209-220.

Holtta-Vuori, M., Uronen, R. L., Repakova, J., Salonen, E., Vattulainen, I.,Panula, P., Li, Z., Bittman, R. and Ikonen, E. (2008). BODIPY-cholesterol: anew tool to visualize sterol trafficking in living cells and organisms. Traffic 9,1839-1849.

Hyslop, P. A., Morel, B. and Sauerheber, R. D. (1990). Organization andinteraction of cholesterol and phosphatidylcholine in model bilayer membranes.Biochemistry 29, 1025-1038.

Ikonen, E. (2008). Cellular cholesterol trafficking and compartmentalization. Nat.Rev. Mol. Cell Biol. 9, 125-138.

Jaqaman, K., Loerke, D., Mettlen, M., Kuwata, H., Grinstein, S., Schmid, S. L.and Danuser, G. (2008). Robust single-particle tracking in live-cell time-lapsesequences. Nat. Methods 5, 695-702.

Jaqaman, K., Kuwata, H., Touret, N., Collins, R., Trimble, W. S., Danuser, G.and Grinstein, S. (2011). Cytoskeletal control of CD36 diffusion promotes itsreceptor and signaling function. Cell 146, 593-606.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

10

Jour

nal o

f Cel

l Sci

ence

Kaiser, R. D. and London, E. (1998). Determination of the depth of BODIPYprobes in model membranes by parallax analysis of fluorescence quenching.Biochim. Biophys. Acta 1375, 13-22.

Kavran, J. M., Klein, D. E., Lee, A., Falasca, M., Isakoff, S. J., Skolnik, E. Y. andLemmon, M. A. (1998). Specificity and promiscuity in phosphoinositide bindingby pleckstrin homology domains. J. Biol. Chem. 273, 30497-30508.

Kay, J. G., Koivusalo, M., Ma, X., Wohland, T. and Grinstein, S. (2012).Phosphatidylserine dynamics in cellular membranes. Mol. Biol. Cell 23, 2198-2212.

Kleusch, C., Hersch, N., Hoffmann, B., Merkel, R. and Csiszar, A. (2012).Fluorescent lipids: functional parts of fusogenic liposomes and tools for cellmembrane labeling and visualization. Molecules 17, 1055-1073.

Knight, J. D. and Falke, J. J. (2009). Single-molecule fluorescence studies of aPH domain: new insights into the membrane docking reaction. Biophys. J. 96,566-582.

Knott, G. and Genoud, C. (2013). Is EM dead? J. Cell Sci. 126, 4545-4552.Kobayashi, T., Stang, E., Fang, K. S., de Moerloose, P., Parton, R. G. andGruenberg, J. (1998). A lipid associated with the antiphospholipid syndromeregulates endosome structure and function. Nature 392, 193-197.

Korlach, J., Schwille, P., Webb, W. W. and Feigenson, G. W. (1999).Characterization of lipid bilayer phases by confocal microscopy andfluorescence correlation spectroscopy. Proc. Natl. Acad. Sci. USA 96, 8461-8466.

Kraft, M. L. and Klitzing, H. A. (2014). Imaging lipids with secondary ion massspectrometry. Biochim. Biophys. Acta 1841, 1108-1119.

Krishnamurthy, K., Dasgupta, S. and Bieberich, E. (2007). Development andcharacterization of a novel anti-ceramide antibody. J. Lipid Res. 48, 968-975.

Kusumi, A., Suzuki, K. G., Kasai, R. S., Ritchie, K. and Fujiwara, T. K. (2011).Hierarchical mesoscale domain organization of the plasma membrane. TrendsBiochem. Sci. 36, 604-615.

Lemmon, M. A., Ferguson, K. M., O’Brien, R., Sigler, P. B. and Schlessinger,J. (1995). Specific and high-affinity binding of inositol phosphates to an isolatedpleckstrin homology domain. Proc. Natl. Acad. Sci. USA. 92, 10472-10476.

Lemmon, M. A. (2008). Membrane recognition by phospholipid-binding domains.Nat. Rev. Mol. Cell Biol. 9, 99-111.

Li, Z., Mintzer, E. and Bittman, R. (2006). First synthesis of free cholesterol-BODIPY conjugates. J. Org. Chem. 71, 1718-1721.

Li, X., Wang, X., Zhang, X., Zhao, M., Tsang, W. L., Zhang, Y., Yau, R. G.,Weisman, L. S. and Xu, H. (2013). Genetically encoded fluorescent probe tovisualize intracellular phosphatidylinositol 3,5-bisphosphate localization anddynamics. Proc. Natl. Acad. Sci. USA 110, 21165-21170.

Lingwood, D. and Simons, K. (2010). Lipid rafts as a membrane-organizingprinciple. Science 327, 46-50.

Lippincott-Schwartz, J. and Patterson, G. H. (2003). Development and use offluorescent protein markers in living cells. Science 300, 87-91.

Lippincott-Schwartz, J., Altan-Bonnet, N. and Patterson, G. H. (2003).Photobleaching and photoactivation: following protein dynamics in living cells.Nat. Cell Biol. Suppl., S7-S14.

Makino, A., Baba, T., Fujimoto, K., Iwamoto, K., Yano, Y., Terada, N., Ohno, S.,Sato, S. B., Ohta, A., Umeda, M. et al. (2003). Cinnamycin (Ro 09-0198)promotes cell binding and toxicity by inducing transbilayer lipid movement.J. Biol. Chem. 278, 3204-3209.

Martin, O. C. and Pagano, R. E. (1987). Transbilayer movement of fluorescentanalogs of phosphatidylserine and phosphatidylethanolamine at the plasmamembrane of cultured cells. Evidence for a protein-mediated and ATP-dependent process(es). J. Biol. Chem. 262, 5890-5898.

Martin, O. C. and Pagano, R. E. (1994). Internalization and sorting of afluorescent analogue of glucosylceramide to the Golgi apparatus of human skinfibroblasts: utilization of endocytic and nonendocytic transport mechanisms.J. Cell Biol. 125, 769-781.

Martin, D. S., Forstner, M. B. and Kas, J. A. (2002). Apparent subdiffusioninherent to single particle tracking. Biophys. J. 83, 2109-2117.

Miller, R. G. (1984). The use and abuse of filipin to localize cholesterol inmembranes. Cell Biol. Int. Rep. 8, 519-535.

Mizuno, H., Abe, M., Dedecker, P., Makino, A., Rocha, S., Ohno-Iwashita, Y.,Hofkens, J., Kobayashi, T. and Miyawaki, A. (2011). Fluorescent probes forsuperresolution imaging of lipid domains on the plasma membrane. Chem. Sci.2, 1548-1553.

Mobius, W., Ohno-Iwashita, Y., van Donselaar, E. G., Oorschot, V. M.,Shimada, Y., Fujimoto, T., Heijnen, H. F., Geuze, H. J. and Slot, J. W. (2002).Immunoelectron microscopic localization of cholesterol using biotinylated andnon-cytolytic perfringolysin O. J. Histochem. Cytochem. 50, 43-55.

Mondal, M., Mesmin, B., Mukherjee, S. and Maxfield, F. R. (2009). Sterols aremainly in the cytoplasmic leaflet of the plasma membrane and the endocyticrecycling compartment in CHO cells. Mol. Biol. Cell 20, 581-588.

Mueller, V., Ringemann, C., Honigmann, A., Schwarzmann, G., Medda, R.,Leutenegger, M., Polyakova, S., Belov, V. N., Hell, S. W. and Eggeling,C. (2011). STED nanoscopy reveals molecular details of cholesterol- andcytoskeleton-modulated lipid interactions in living cells.Biophys. J. 101, 1651-1660.

Mukherjee, S., Zha, X., Tabas, I. and Maxfield, F. R. (1998). Cholesteroldistribution in living cells: fluorescence imaging using dehydroergosterol as afluorescent cholesterol analog. Biophys. J. 75, 1915-1925.

Navarro, J., Chabot, J., Sherrill, K., Aneja, R., Zahler, S. A. and Racker, E.(1985). Interaction of duramycin with artificial and natural membranes.Biochemistry 24, 4645-4650.

Ohno-Iwashita, Y., Shimada, Y., Hayashi, M. and Inomata, M. (2010a). Plasmamembrane microdomains in aging and disease.Geriatr. Gerontol. Int. 10 Suppl.,S41-S52.

Ohno-Iwashita, Y., Shimada, Y., Hayashi, M., Iwamoto, M., Iwashita, S. andInomata, M. (2010b). Cholesterol-binding toxins and anti-cholesterol antibodiesas structural probes for cholesterol localization. Subcell. Biochem. 51, 597-621.

Pagano, R. E., Martin, O. C., Kang, H. C. and Haugland, R. P. (1991). A novelfluorescent ceramide analogue for studying membrane traffic in animal cells:accumulation at the Golgi apparatus results in altered spectral properties of thesphingolipid precursor. J. Cell Biol. 113, 1267-1279.

Palade, G. E. (1952). A study of fixation for electron microscopy. J. Exp. Med. 95,285-298.

Palade, G. E. and Porter, K. R. (1954). Studies on the endoplasmic reticulum. I.Its identification in cells in situ. J. Exp. Med. 100, 641-656.

Puri, V., Watanabe, R., Dominguez, M., Sun, X., Wheatley, C. L., Marks, D. L.and Pagano, R. E. (1999). Cholesterol modulates membrane traffic along theendocytic pathway in sphingolipid-storage diseases. Nat. Cell Biol. 1, 386-388.

Puri, V., Watanabe, R., Singh, R. D., Dominguez, M., Brown, J. C., Wheatley,C. L., Marks, D. L. and Pagano, R. E. (2001). Clathrin-dependent and -independent internalization of plasma membrane sphingolipids initiates twoGolgi targeting pathways. J. Cell Biol. 154, 535-548.

Ries, J., Chiantia, S. and Schwille, P. (2009). Accurate determination ofmembrane dynamics with line-scan FCS. Biophys. J. 96, 1999-2008.

Ritchie, K., Shan, X. Y., Kondo, J., Iwasawa, K., Fujiwara, T. and Kusumi, A.(2005). Detection of non-Brownian diffusion in the cell membrane in singlemolecule tracking. Biophys. J. 88, 2266-2277.

Rogers, J., Lee, A. G. and Wilton, D. C. (1979). The organisation of cholesteroland ergosterol in lipid bilayers based on studies using non-perturbingfluorescent sterol probes. Biochim. Biophys. Acta 552, 23-37.

Roy, A. and Levine, T. P. (2004). Multiple pools of phosphatidylinositol 4-phosphate detected using the pleckstrin homology domain of Osh2p. J. Biol.Chem. 279, 44683-44689.

Sarantis, H. and Grinstein, S. (2012). Subversion of phagocytosis for pathogensurvival. Cell Host Microbe 12, 419-431.

Saxton, M. J. and Jacobson, K. (1997). Single-particle tracking: applications tomembrane dynamics. Annu. Rev. Biophys. Biomol. Struct. 26, 373-399.

Schlam, D., Bohdanowicz, M., Chatgilialoglu, A., Steinberg, B. E., Ueyama, T.,Du, G., Grinstein, S. and Fairn, G. D. (2013). Diacylglycerol kinases terminatediacylglycerol signaling during the respiratory burst leading to heterogeneousphagosomal NADPH oxidase activation. J. Biol. Chem. 288, 23090-23104.

Schuster, B. G., Neidig, M., Alving, B. M. and Alving, C. R. (1979). Productionof antibodies against phosphocholine, phosphatidylcholine, sphingomyelin,and lipid A by injection of liposomes containing lipid A. J. Immunol. 122, 900-905.

Schwille, P., Haupts, U., Maiti, S. and Webb, W. W. (1999). Molecular dynamicsin living cells observed by fluorescence correlation spectroscopy with one- andtwo-photon excitation. Biophys. J. 77, 2251-2265.

Sezgin, E., Levental, I., Grzybek, M., Schwarzmann, G., Mueller, V.,Honigmann, A., Belov, V. N., Eggeling, C., Coskun, U., Simons, K. et al.(2012). Partitioning, diffusion, and ligand binding of raft lipid analogs in modeland cellular plasma membranes. Biochim. Biophys. Acta 1818, 1777-1784.

Shimada, Y., Maruya, M., Iwashita, S. and Ohno-Iwashita, Y. (2002). The C-terminal domain of perfringolysin O is an essential cholesterol-binding unittargeting to cholesterol-rich microdomains. Eur. J. Biochem. 269, 6195-6203.

Sieber, J. J., Willig, K. I., Heintzmann, R., Hell, S. W. and Lang, T. (2006). TheSNARE motif is essential for the formation of syntaxin clusters in the plasmamembrane. Biophys. J. 90, 2843-2851.

Simons, K. and Ehehalt, R. (2002). Cholesterol, lipid rafts, and disease. J. Clin.Invest. 110, 597-603.

Simons, K. and Toomre, D. (2000). Lipid rafts and signal transduction. Nat. Rev.Mol. Cell Biol. 1, 31-39.

Singer, S. J. and Nicolson, G. L. (1972). The fluid mosaic model of the structureof cell membranes. Science 175, 720-731.

Solanko, L. M., Honigmann, A., Midtiby, H. S., Lund, F. W., Brewer, J. R.,Dekaris, V., Bittman, R., Eggeling, C. and Wustner, D. (2013). Membraneorientation and lateral diffusion of BODIPY-cholesterol as a function of probestructure. Biophys. J. 105, 2082-2092.

Sprong, H., van der Sluijs, P. and van Meer, G. (2001). How proteins move lipidsand lipids move proteins. Nat. Rev. Mol. Cell Biol. 2, 504-513.

Stauffer, T. P., Ahn, S. and Meyer, T. (1998). Receptor-induced transientreduction in plasma membrane PtdIns(4,5)P2 concentration monitored in livingcells. Curr. Biol. 8, 343-346.

Steinberg, B. E., Magalhaes, M. A. and Grinstein, S. (2014). Expression ofgenetically encoded fluorescent probes to monitor phospholipid dynamics in liveneutrophils. Methods Mol. Biol. 1124, 269-277.

Tanaka, K. A., Suzuki, K. G., Shirai, Y. M., Shibutani, S. T., Miyahara, M. S.,Tsuboi, H., Yahara, M., Yoshimura, A., Mayor, S., Fujiwara, T. K. et al. (2010).Membrane molecules mobile even after chemical fixation. Nat. Methods 7, 865-866.

Teruel, M. N. and Meyer, T. (2000). Translocation and reversible localization ofsignaling proteins: a dynamic future for signal transduction. Cell 103, 181-184.

Touboul, D., Laprevote, O. and Brunelle, A. (2011). Micrometric molecularhistology of lipids by mass spectrometry imaging. Curr. Opin. Chem. Biol. 15,725-732.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

11

Jour

nal o

f Cel

l Sci

ence

van den Bogaart, G., Meyenberg, K., Risselada, H. J., Amin, H., Willig, K. I.,Hubrich, B. E., Dier, M., Hell, S. W., Grubmuller, H., Diederichsen, U. et al.(2011). Membrane protein sequestering by ionic protein-lipid interactions.Nature 479, 552-555.

van Genderen, I. L., van Meer, G., Slot, J. W., Geuze, H. J. and Voorhout, W. F.(1991). Subcellular localization of Forssman glycolipid in epithelial MDCK cellsby immuno-electronmicroscopy after freeze-substitution. J. Cell Biol. 115, 1009-1019.

van Meer, G., Voelker, D. R. and Feigenson, G. W. (2008). Membranelipids: where they are and how they behave. Nat. Rev. Mol. Cell Biol. 9, 112-124.

Varnai, P. and Balla, T. (2006). Live cell imaging of phosphoinositide dynamicswith fluorescent protein domains. Biochim. Biophys. Acta 1761, 957-967.

Varnai, P., Rother, K. I. and Balla, T. (1999). Phosphatidylinositol 3-kinase-dependent membrane association of the Bruton’s tyrosine kinase pleckstrinhomology domain visualized in single living cells. J. Biol. Chem. 274, 10983-10989.

Voorhout, W., van Genderen, I., van Meer, G. and Geuze, H. (1991).Preservation and immunogold localization of lipids by freeze-substitution andlow temperature embedding. Scanning Microsc. Suppl. 5, S17-S24; discussionS24-S25.

Wang, J. and Richards, D. A. (2012). Segregation of PIP2 and PIP3 intodistinct nanoscale regions within the plasma membrane. Biol. Open 1, 857-862.

Watson, A. D. (2006). Thematic review series: systems biology approaches tometabolic and cardiovascular disorders. Lipidomics: a global approach to lipidanalysis in biological systems. J. Lipid Res. 47, 2101-2111.

Wustner, D. (2007). Fluorescent sterols as tools in membrane biophysics and cellbiology. Chem. Phys. Lipids 146, 1-25.

Yachi, R., Uchida, Y., Balakrishna, B. H., Anderluh, G., Kobayashi, T., Taguchi,T. and Arai, H. (2012). Subcellular localization of sphingomyelin revealed by twotoxin-based probes in mammalian cells. Genes Cells 17, 720-727.

Yamaji, A., Sekizawa, Y., Emoto, K., Sakuraba, H., Inoue, K., Kobayashi, H.and Umeda, M. (1998). Lysenin, a novel sphingomyelin-specific binding protein.J. Biol. Chem. 273, 5300-5306.

Yeung, T., Terebiznik, M., Yu, L., Silvius, J., Abidi, W. M., Philips, M., Levine, T.,Kapus, A. and Grinstein, S. (2006). Receptor activation alters inner surfacepotential during phagocytosis. Science 313, 347-351.

Yeung, T., Gilbert, G. E., Shi, J., Silvius, J., Kapus, A. and Grinstein, S. (2008).Membrane phosphatidylserine regulates surface charge and proteinlocalization. Science 319, 210-213.

Zhao, C., Du, G., Skowronek, K., Frohman, M. A. and Bar-Sagi, D. (2007).Phospholipase D2-generated phosphatidic acid couples EGFR stimulation toRas activation by Sos. Nat. Cell Biol. 9, 707-712.

Zhou, Y., Liang, H., Rodkey, T., Ariotti, N., Parton, R. G. and Hancock, J. F.(2014). Signal integration by lipid-mediated spatial cross talk between Rasnanoclusters. Mol. Cell. Biol. 34, 862-876.

COMMENTARY Journal of Cell Science (2014) 127, 1–12 doi:10.1242/jcs.150524

12