analysis of lipid-composition changes in plasma membrane ... · scm, silica-coated membrane; sms,...

1594 Journal of Lipid Research Volume 56, 2015

Copyright © 2015 by the American Society for Biochemistry and Molecular Biology, Inc.

This article is available online at http://www.jlr.org

and cholesterol (Chol) ( 1–5 ). These sphingolipid-rich microdomains are considered to be important sites for signal transduction and death receptor functions ( 6, 7 ). Depending on the local microenvironment, certain mem-brane proteins are preferentially localized inside (or out-side) of the microdomains and serve as signaling platforms ( 8–17 ). Recent data has indicated that membrane micro-domains are dynamic nanometer-sized domains ( 18, 19 ). Results from image analysis and single-molecule track-ing studies have shown that these metastable membrane assemblies can be stabilized locally by lipid-lipid and lipid-protein interactions to coalesce and form functional domains and facilitate cell surface receptor signal transduction. A commonly held view is that microdomains regulate mem-brane protein functions by recruiting signaling molecules to the spatially limited region ( 7, 20, 21 ).

Tight interactions between Chol and SM result in the formation of domains that are resistant to solubilization with detergents ( 22 ). This property is often used to pre-pare membrane microdomains. The typical preparation method involves the treatment of postnuclear lysates, gen-erated by the removal of nuclei from cell lysates, with a nonionic detergent at low temperature, followed by sepa-ration of the low-density lipid fraction in a sucrose gradi-ent ( 1 ). However, this fraction may contain a broad range of detergent-resistant membranes other than plasma membrane microdomains. Therefore, concerns have been raised that results obtained using this conventional method are not limited to events occurring on plasma membrane

Abstract Sphingolipids accumulate in plasma membrane microdomain sites, such as caveolae or lipid rafts. Such mi-crodomains are considered to be important nexuses for sig-nal transduction, although changes in the microdomain lipid components brought about by signaling are poorly un-derstood. Here, we applied a cationic colloidal silica bead method to analyze plasma membrane lipids from mono-layer cells cultured in a 10 cm dish. The detergent-resistant fraction from the silica bead-coated membrane was ana-lyzed by LC-MS/MS to evaluate the microdomain lipids. This method revealed that glycosphingolipids composed the microdomains as a substitute for sphingomyelin (SM) in mouse embryonic fi broblasts (tMEFs) from an SM synthase 1/2 double KO (DKO) mouse. The rate of formation of the detergent-resistant region was unchanged compared with that of WT-tMEFs. C2-ceramide (Cer) stimulation caused greater elevations in diacylglycerol and phosphatidic acid levels than in Cer levels within the microdomains of WT-tMEFs. We also found that lipid changes in the microdomains of SM-defi cient DKO-tMEFs caused by serum stimulation occurred in the same manner as that of WT-tMEFs. This practical method for analyzing membrane lipids will facili-tate future comprehensive analyses of membrane micro-domain-associated responses. —Ogiso, H., M. Taniguchi, and T. Okazaki. Analysis of lipid-composition changes in plasma membrane microdomains. J. Lipid Res. 2015. 56: 1594–1605.

Supplementary key words cationic colloidal silica beads • liquid chromatography-tandem mass spectrometry • lipid raft • lipidomics • diacylglycerol • phosphatidic acid • sphingomyelin • ceramide

Membrane microdomains, including caveolae or lipid rafts, are plasma membrane domains that contain enhanced levels of sphingolipids, saturated fatty-acyl glycerophospholipids,

This work was partly supported by the Takeda Science Foundation and was performed as a joint research project with Shalom Co., Ltd. This study was also supported in part by grants from the Strategic Research Foundation Grant-Aided Project for Private Universities (grant number S1201004) and from the Applica-tion Procedures for Grant-in-Aid for Research Activity Start-up (grant number 25893269) from the Ministry of Education, Culture, Sport, Science, and Tech-nology (MEXT) in Japan, and from Kanazawa Medical University (H2012-15; 2012–2016).

Manuscript received 9 April 2015 and in revised form 23 June 2015.

Published, JLR Papers in Press, June 26, 2015 DOI 10.1194/jlr.M059972

Analysis of lipid-composition changes in plasma membrane microdomains

Hideo Ogiso , * Makoto Taniguchi , † and Toshiro Okazaki 1, * ,†

Department of Hematology/Immunology* and Medical Research Institute, † Kanazawa Medical University , 1-1 Daigaku, Uchinada, Ishikawa 920-0293, Japan

Abbreviations: CB, coating buffer; C2-Cer, N -acetylsphingosine; Cer, ceramide; Chol, cholesterol; DAG, diacylglycerol; dhCer, dihydrocer-amide; DKO, double knockout; drFF, detergent-resistant fl oating-membrane fraction; drSCM, detergent-resistant silica-coated membrane; FIPI, 5-fl uoro-2-indolyl des-chlorohalopemide; HexCer, monohexosyl-ceramide, LacCer, lactosylceramide; tMEF, immortalized mouse em-bryonic fi broblast; PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphati-dylinositol, pPE, plasmalogen-phosphatidylethanolamine; PPI, poly-phosphoinositide; PS, phosphatidylserine, S1P, sphingosine-1-phosphate; SCM, silica-coated membrane; SMS, SM synthase; WC, whole-cell .

1 To whom correspondence should be addressed. e-mail: [email protected]

The online version of this article (available at http://www.jlr.org) contains supplementary data in the form of two fi gures and one table.

by guest, on June 25, 2018w

ww

.jlr.orgD

ownloaded from

.html http://www.jlr.org/content/suppl/2015/06/26/jlr.M059972.DC1Supplemental Material can be found at:

http://www.jlr.org/

http://www.jlr.org/content/suppl/2015/06/26/jlr.M059972.DC1.html


Lipid analysis of membrane microdomains 1595

Biowest (Nuaillé, France). Other chemical reagents were of ana-lytical grade and were purchased from Nacalai Tesque (Kyoto, Japan). Ultrapure water was obtained from an RFU554CA ultra-pure water system (Advantech Co. Ltd., Tokyo, Japan).

Cell culture WT and SM synthase (SMS) 1/2 double KO (DKO) mouse

embryonic fi broblasts (tMEFs), which were immortalized by the simian virus 40 large T antigen in our laboratory ( 35 ), were cul-tured in DMEM supplemented with 10% FBS, 100 units/ml penicillin, and 100 � g/ml streptomycin at 37°C in a humidifi ed atmosphere with 5% CO 2 . Cells were grown on 10 cm dishes to a confl uence of 50–80%, and stimulants were then added for the indicated periods. Incubations with stimulants were terminated by replacing the culture medium with ice-cold PBS.

Preparation of plasma membranes with cationic colloidal silica beads

The cationic colloidal silica beads method was used for prepar-ing plasma membranes ( 26–34 ). Briefl y, this step was performed on ice using ice-cold buffers. Subconfl uent cell monolayers grown in 10 cm dishes were washed twice with PBS and then with coating buffer (CB) [20 mM MES, 150 mM NaCl (pH 5.3)]. The cells were coated with 1% (w/v) Ludox CL colloidal silica in CB. After a 10 min incubation on ice, the cells were washed twice with CB and then coated with 0.1% (w/v) polyacrylic acid in CB. After a 10 min incubation on ice, the cells with washed three times with CB. Next, the coated cells were scraped with 1 ml of hypotonic lysis buffer [20 mM HEPES buffer (pH 8.0), 1 mM EDTA], sup-plemented with protease inhibitors and 1 mg/ml BSA. The lysate was passed 20 times through an 18-gauge needle to disrupt the coated cells. The homogenized samples were fi ltered through an 80 � m nylon net. The lysate was centrifuged at 1,000 g for 20 min and the pellet was collected. To remove intracellular contami-nants, the disrupted cells were passed 10 times through a 27-gauge needle with 1 ml of washing buffer [20 mM HEPES buf-fer (pH 7.0), 150 mM NaCl, 1 mM EDTA, 8 mM DTT, 7% su-crose, 0.01% Brij 98, and protease inhibitors]. Complete cell lysis was confi rmed by observations under a light microscope. A Ny-codenz step gradient was set in centrifuge tubes containing 1 ml of 70% (w/v) and 1 ml of 50% Nycodenz solutions supplemented with 150 mM NaCl. The washed materials were centrifuged again at 1,000 g for 20 min, and the pellet was resuspended in 0.4 ml of washing buffer. Then the suspension, mixed with 0.4 ml of the 50% Nycodenz solution, was added on top of the Nycodenz gra-dient and centrifuged at 100,000 g for 60 min (Optima MAX-XP; Beckman Coulter, Brea, CA). The silica-coated membranes (SCMs) at the bottom of the centrifuge tube were resuspended in 0.3 ml of SCM buffer [5 mM imidazole buffer (pH 7.0), 150 mM NaCl, 8 mM DTT, and 7% sucrose], and centrifuged at 18,000 g for 10 min to remove excess Nycodenz. To increase the recovery rate of the membranes, plasticwares with hydrophilic surfaces, such as ProteoSave 1.5 ml microtubes (Sumitomo Bakelite, Tokyo, Japan) and MPC polymer-coating tips (Hi-Tech, Tokyo, Japan), were used as appropriate.

Preparation of detergent-resistant SCMs One percent Triton X-100 and 1% Brij 98 were added to

SCM fractions. Suspensions were homogenized by passaging 20 times through a 27-gauge needle and incubated on ice for 30 min with occasional mixing. Then, the suspensions were centrifuged at 18,000 g for 10 min at 4°C. After supernatant removal, the insoluble materials were washed with 0.3 ml SCM buffer, and then detergent-resistant SCMs (drSCMs) were obtained.

microdomains, but also refl ect those occurring in intra-cellular vesicles, such as multivesicular bodies including sphingolipid-rich membranes ( 23–25 ).

Here, we developed a comprehensive approach for de-termining the constituent lipids that mainly comprise plasma membrane microdomains. First, we modifi ed a cat-ionic colloidal silica bead method to separate the plasma membranes ( 26–34 ). Next, we designed a protocol for ob-taining detergent-resistant lipid regions using silica-coated membranes. Then, these preparations were analyzed by LC-MS/MS to determine the molecular species levels of each lipid class. Using this approach, we examined the percentages of detergent-resistant lipids by comparing the lipid levels in these subcellular fractions. In addition, we examined whether levels of specifi c lipids changed in the microdomains after stimulation. Our results indicated that this practical method may be useful for performing com-prehensive studies of component lipids in membrane mi-crodomains related to cellular responses.

EXPERIMENTAL PROCEDURES

Materials C17-sphingosine [Sph(d17:1)], C17-sphingosine-1-phosphate

[S1P(d17:1)], C17-sphinganine-1-phosphate [S1P(d17:0)], 12:0-ceramide [Cer(d18:1/12:0)], 12:0-ceramide-1-phosphate [Cer1P(d18:1/12:0)], 12:0-glucosyl( � )ceramide [GlcCer(d18:1/12:0)], 12:0-lactosyl( � )ceramide [LacCer(d18:1/12:0)], 12:0-SM [SM(d18:1/12:0)], 14:0/14:0-diacylglycerol [DAG(14:0/14:0)], 14:0/14:0-phosphatidylcholine [PC(28:0)], 14:0/14:0-phosphatidylserine [PS(28:0)], 14:0/14:0-phosphatidylglycerol [PG(28:0)], 16:0/16:0-phosphatidylinositol [PI(32:0)], Chol(D7), C17-sphingosylphosphorylcholine [lysoSM(d17:1)], monosialogan-glioside GM1 (from ovine brain), C18(plasm)/18:1-PC [PC(p18:0/18:1)], and C18(plasm)/18:1-phosphatidylethanolamine [PE(p18:0/18:1)] were purchased from Avanti Polar Lipids (Alabaster, AL). The reagents 14:0/14:0-PE [PE(28:0)] and disialoganglioside GD1a (from bovine brain) were obtained from Wako Pure Chemicals (Osaka, Japan). Monosialoganglioside 12:0-GM3 was purchased from Nagara Sciences (Gifu, Japan). N -acetylsphingosine (C2-Cer) and N -acetyl dihydrosphingosine [C2-dihydroceramide (dhCer)] were obtained from Matreya (Pleasant Gap, PA). Meth-anol, 2-propanol, and formic acid were of LC-MS grade and were purchased from Thermo-Fisher Scientifi c (Rockford, IL). Anti-calnexin (sc-11396) and anti-GAPDH (sc-20357) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti- � -actin (#4967), anti-lamin A/C (#2032), anti-hexokinase I (#2024), and anti-LAMP1 (#3243) antibodies were purchased from Cell Signaling Technology (Beverly, MA). An anti-fl otillin 1 (#610820) antibody was purchased from BD Biosciences (Bedford, MA). An anti-Na + /K + -ATPase (ab7671) antibody was obtained from Abcam (Cambridge, MA). An anti-Rab7 (R8779) antibody, Tri-ton X-100, Brij 98, Ludox CL colloidal silica, polyacrylic acid (#181285), 5-fl uoro-2-indolyl des-chlorohalopemide (FIPI), and DMEM were purchased from Sigma-Aldrich (St Louis, MO). Sec-ondary antibodies conjugated with horseradish peroxidase were obtained from Promega (Madison, WI). Protease inhibitor cocktail (cOmplete, EDTA free) was obtained from Roche (Mannheim, Germany). Nycodenz AG was purchase from Axis-Shield (Oslo, Norway). DEAE-cellulose was purchased from Wako Pure Chemi-cals or Santa Cruz Biotechnology. FBS was purchased from


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




Preparation of detergent-resistant membranes by a centrifugal fl otation method

An alternative fractionation method of isolating detergent-resistant membranes was performed by centrifugal flotation through a sucrose gradient. The cells were washed with ice-cold PBS, scraped with 1 ml of the hypotonic lysis buffer, and passed 20 times through an 18-gauge needle to disrupt the cells. Nuclei and cellular debris were pelleted by centrifugation at 1,000 g for 5 min at 4°C. After the addition of 1% Triton X-100 and 1% Brij 98, the postnuclear supernatant was homogenized by passaging 10 times through a 25-gauge needle and then incubated on ice for 30 min with occasional mixing. The lysate was mixed with an equal volume of 84% sucrose, transferred to the bottom of a cen-trifuge tube, and overlaid with 2 ml of 33% sucrose containing 150 mM NaCl and fi nally with 0.5 ml of 5% sucrose. The sam-ples were centrifuged at 150,000 g for 18 h at 4°C, and then 1.5 ml of supernatant from the top of the gradient was collected as a detergent-resistant fl oating-membrane fraction (drFF).

Western blot and protein analyses Western blot analysis was performed as follows. Proteins were

denatured at 50°C for 30 min in standard sample buffer contain-ing 2% SDS and 0.2 M DTT and resolved by 12% PAGE. When analyzing SCM or drSCM samples, 0.1 M sodium phosphate buf-fer (pH 6.0) was added to the sample buffer to detach proteins from the silica beads. Following electrophoresis, proteins were transferred to a polyvinylidene fl uoride membrane (Millipore, Bedford, MA) and subsequently incubated with appropriate pri-mary and secondary antibodies. Detection was achieved using an ECL-peroxidase detection system (Amersham Biosciences, Pisca-taway, NJ) and an LAS-4000 digital imaging system (GE Health-care, Little Chalfont, UK). SDS-PAGE was performed using a standard protocol and precast 5 � 20% gradient gels (SuperSep Ace, Wako Pure Chemicals). The proteins were visualized with a SYPRO Ruby protein gel stain kit (Lonza, Rockland, ME). Pro-tein concentrations were determined using a BCA protein assay kit (Thermo-Fisher Scientifi c).

Lipid standard solutions Internal standard solutions were prepared as described previ-

ously ( 36 ), except that 50 pmol of LacCer(d18:1/12:0) and 5,000 pmol of Chol(D7) were added to the internal standard solution. For measurements of plasmalogen PE (pPE), GM1, and GD1, an external standard solution containing 50 pmol each of PE(28:0), PE(p18:0/18:1), GM3(d18:1/12:0), and GD1a (from bovine brain) was prepared. Pretreatment and measurement of the external standard solution was performed simultaneously with the samples, and derived [peak area of PE(p18:0/18:1)/peak area of PE(28:0)] and [peak area of GM3 fragment ions from GD1a (from bovine brain)]/[peak area of GM3(d18:1/12:0)] values were used as the corrective coeffi cients for the quantita-tion of pPE and GD1, respectively. Unfortunately, we could not clearly discriminate GD1 from the other GM3 derivatives, such as GM1 and GT1, using our LC-MS/MS conditions. Therefore, these compounds were collectively represented as GM3-X.

Lipid extractions Lipid extractions were performed as described previously ( 36 ),

with several modifi cations. Briefl y, cell pellets, SCM fractions, or drSCM fractions were sonicated for 10 s with 0.1 ml methanol/butanol (1:1) to inactivate the associated enzymes using an ultra-sonic bath. After the addition of 0.05 ml standard lipid mixture, 0.05 ml of 0.5 M phosphate buffer (pH 6.0), and 0.2 ml of water, the samples were shaken with 0.7 ml of butanol and sonicated for 3 min in an ultrasonic bath. After centrifugation, the upper layer

was collected. The original suspension was re-extracted by the addition of 0.35 ml each of ethyl acetate and hexane, followed by centrifugation. The resulting extract was combined with the fi rst butanol extract. After the addition of 0.7 ml methanol, 10% (0.21 ml) of this solution was dried under reduced pressure at 40°C, and dissolved in 20 � l of LC mobile phase B and 30 � l of mobile phase A. This sample was used to analyze Cer, SM, mono-hexosylceramide (HexCer), LacCer, S1P, PEtn, PC, and Chol levels. The remaining 90% (1.8 ml) of the extract was fraction-ated on a DEAE-cellulose column (500 � l bed volume packed in a 1 ml polypropylene pipette tip). After washing with 2 ml of methanol, the column-bound lipids were eluted with 1 ml methanol/28% aqueous ammonia/formic acid (1,000:33:22). The organic solvent was evaporated from the eluate under reduced pressure at 50°C, after which dried materials were dis-solved with 50 � l of mobile phase A. The resulting sample was used for the analyses of acidic lipids (i.e., S1P, Cer1P, GM3, GM3-X, PS, PG, PI, and phosphatidic acid (PA)].

MS analysis Lipids were measured using LC-MS/MS as described previ-

ously ( 36 ), except that the collision energy was set to 30 V for Cer, SM, HexCer, LacCer, PE, and PC. The mass transitions were additionally set to 702.5/364.2, 724.5/364.2, 728.6/390.2, 730.6/392.2, 748.5/364.2, 750.5/390.2, 752.6/392.2, 774.5/390.2, and 776.6/392.2 for pPE in the positive ion mode. Chol was mea-sured using another set of LC-atmospheric pressure chemical ionization-MS/MS conditions. Briefl y, the collision energy was set to 20 V, and the mass transitions were set to 369.3/147.1 for Chol and 376.3/147.1 for Chol(D7) in the positive ion mode. Each molecular species was identifi ed based on the MS/MS spec-trum and the LC retention times, and quantities present were calculated from the peak areas of the measured lipids, compared with those of the internal standards. Each level of measured lip-ids was normalized to the total lipid content or the total PC content.

RESULTS AND DISCUSSION

Preparation of plasma membrane-rich fractions for lipidomics studies using a cationic colloidal silica beads method

The cationic colloidal silica beads method has been widely used for separating plasma membranes since its de-velopment in 1983 ( 26 ). Using this method, the surfaces of intact cells are covered with a dense coat of cationic colloidal silica beads, followed by cross-linking with poly-acrylic acid. This coating treatment on ice prevents vesicu-lation and lateral reorientation of the plasma membrane. After mechanical cell lysis, the SCMs can be separated by centrifugation, due to their high density. However, puri-ties and yields of the membranes can vary in different labo-ratories because of the different experimental conditions used, including variations in buffer systems ( 26–34 ). In ad-dition, a relatively large number of cells were needed in previous plasma membrane lipidomics studies using this method ( 30, 31 ).

In the present study, this method was optimized to en-able the reproducible and effi cient preparation of mem-branes from adherent cells cultured in 10 cm dishes. In addition, we washed the SCM pellets with a washing buffer


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




containing 150 mM NaCl, 8 mM DTT, and 0.01% Brij 98 to remove intracellular contaminants, as much as possible. Comparative Western blot analyses of Na + /K + -ATPase recovery between the whole-cell (WC) lysates and the SCM fraction from WT-tMEFs revealed an enrichment of plasma membrane protein, while GAPDH (a cytosolic marker) and hexokinase I (a mitochondrial marker) were nearly undetectable in the SCM fractions. The re-coveries of lamin A/C (a nuclear marker) and LAMP1 (a lysosomal marker) were reduced; whereas, calnexin (an endoplasmic reticulum marker), Rab7 (an endosomal marker), and actin were not reduced in the SCM frac-tions ( Fig. 1A ). In a previous study, calnexin was reported as one of major components in the plasma membrane fraction prepared by the colloidal silica beads method ( 28 ). The yield of SCM lipids was � 5% that of WC lipids ( Fig. 1B ), a reasonable value in view of the proportion of plasma membranes to WC membranes ( 31, 37 ). These results showed that the SCMs obtained in this study com-prised a plasma membrane-rich fraction, while it also may have contained some intracellular components tightly associated with the plasma membrane via cytoskel-etal proteins. These fi ndings are important, as they indi-cate that this preparation method for plasma membranes from cells cultured in a 10 cm dish was reproducible and offered a high yield.

Next, we attempted to prepare detergent-resistant mi-crodomain fractions from the SCM fraction described above. Our primary aim was to analyze changes in mem-brane lipids. PE and PS were relatively depleted after treatment with 1% Triton X-100, which is a popular de-tergent used for preparing microdomain fractions, as re-ported previously ( 2 ). For this reason, we used a mixture of 1% Triton X-100 and 1% Brij 98 for the detergent-treatment step, so that PE and PS were relatively well re-tained in the detergent-resistant fraction. The resultant insoluble materials after detergent treatment were ana-lyzed by Western blotting, SDS-PAGE, and LC-MS/MS. Although Na + /K + -ATPase and Rab7 detected by Western blotting were reduced in the drSCM fraction in compari-son with the SCM fraction ( Fig. 1A ), around 80% of the proteins in the SCM fraction remained in the drSCM fraction after the detergent treatment, and proteins de-tected in the detergent-resistant fraction showed little change ( Fig. 1C ). These results suggested that nearly all proteins bound to the silica beads remained in the drSCM fraction even after the detergent treatment, while more than half of the lipids were removed during the treatment ( Fig. 1D ). The lipid composition of drSCMs resembled raft lipids in that sphingolipids, saturated fatty acyl PC(32:0), and Chol were enriched, and that PI and PG were reduced ( 2–4, 38–41 ). The yield of drSCM lip-ids, which was estimated to be comparable to the propor-tion of detergent-resistant lipids, was � 46% of that obtained with SCM lipids. Thus, the present method was benefi cial for analyzing lipid levels in detergent-resistant membrane microdomains.

For comparison purposes with the SCM method de-scribed above, we also analyzed lipids in a general

detergent-resistant membrane fraction. Briefl y, the postnuclear fraction was solubilized with a 1% Triton X-100 and 1% Brij 98 mixture, and then a drFF was ob-tained by ultracentrifugation through a discontinuous sucrose gradient (5% to 33% to 42%). Some technical issues were confronted during lipid analysis of the drFF (supplementary Fig. 1). The recovery rate of the deter-gent-resistant membrane lipid was over 10% of the WC lipid, suggesting that the drFF also contained mem-branes other than plasma membrane microdomains ( 31, 37 ). The values were relatively variable between in-dependent experiments. In addition, the (PS+PE)/(PE+pPE+PS+PI+PG) abundance ratios of WC or drSCM samples were nearly equal to 1, although that of the drFF sample was not ( 2 ). Coexisting detergents, which were unexpectedly concentrated together with lipids during sample pretreatment, may have been responsi-ble for some of the detrimental effects observed, such as diffi culty in solvent evaporation, loss of specifi c lipids, pollution of the ion source, and ion suppression during lipid analysis by LC-MS/MS. Therefore, we concluded that accurate, precise, and reproducible measurement was diffi cult using drFF samples and that the SCM method presented here provided more pertinent infor-mation in terms of lipid microdomain analysis.

Comparative lipid analysis between WT and SMS1/2-knockout tMEF membranes

SM, one of major components of membrane microdo-mains, is synthesized from Cer and PC by SMS1/2 ( 42 ). Thus, SMS1/2 DKO-tMEFs contain low levels of SM. How-ever, DKO-tMEFs proliferate readily under normal culture conditions, although their growth rate and morphological and adherence characteristics differ from those of WT-tMEFs ( 35 ). These observations raised the question of what kinds of differences occur in the membrane microdo-mains between these two cell types. To examine dif-ferences in their component lipids, we comparatively analyzed their levels in WT- and DKO-tMEFs. Signifi cant differences in the levels of various lipids were found in each of three fractions studied, namely the WC, SCM, and drSCM fractions. The levels of HexCer, LacCer, and GM3 were higher in each DKO-tMEF fraction ( Fig. 2A ). The yield of the detergent-resistant region in the SCM prepara-tion from DKO-tMEFs ( � 47%) was comparable to that of WT-tMEFs ( Fig. 2B ). More specifi cally, signifi cant differ-ences in the levels of sphingolipids, PC, PA, DAG, and PI, were found ( Fig. 2C ). To survey the compositions for all lipids with an acyl structure, the measured level of each species was compared between WT- and DKO-tMEFs ( Fig. 2D ). SM was defi cient in DKO-tMEFs, while the HexCer and GM3 levels drastically increased without changing their acyl-structure distribution. PC(32:0) and PC(34:1) levels were signifi cantly increased, especially in drSCMs. These observations suggested that the indicated glyco-sphingolipids, as well as PC(32:0) and PC(34:1), helped to form microdomains as a substitute for SM in SM-defi cient cells. Although Cer, PA, DAG, and PI were present as mi-nor lipids, the levels of these signaling lipids were markedly


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




Changes in microdomain lipid compositions caused by C2-Cer stimulation

As a next step, we examined changes in microdomain lipids caused by exogenous stimuli, such as C2-Cer and growth factors, using the present method. Differences in responses to these stimuli between WT-tMEFs and

higher in drSCMs from DKO-tMEFs, as compared with their corresponding levels in WT-tMEFs ( Fig. 2C ). The other molecular species levels measured in this experi-ment are shown in supplementary Fig. 2. Thus, the pres-ent method enabled robust evaluation of constituent lipid levels in membrane microdomains.

Fig. 1. Evaluation of membrane fractions by protein and lipid analyses. A: Ten micrograms each of WC, SCM, and drSCM fractions from WT-tMEF samples were analyzed in Western blots. Antibodies against Na + /K + -ATPase, fl otillin1, calnexin, actin, Rab7, lamin A/C, LAMP1, GAPDH, and hexokinase I were used as markers for the plasma membranes, plasma membrane microdomains, endoplasmic reticulum, cytoskeleton, endosome, nucleus, lysosome, cytosol, and mitochondria, respectively. B: Lipids in WT-tMEF samples were analyzed by LC-MS/MS, and lipid-class distributions are shown in the circular chart. The data shown repre-sent mean values from three independent experiments. The lipid yields from each membrane fraction are also shown under the chart. Data are expressed as the mean ± SD of three independent experiments.. C: Twenty micrograms of each membrane fraction from WT-tMEFs was resolved by SDS-PAGE, and proteins were visualized using the SYPRO Ruby protein gel stain. The protein yield from each membrane fraction is also shown under the fi gure. Data are expressed as the mean ± SD of four independent experiments. D: Schematic representation of drSCM components.


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




altered by stimulation ( Fig. 3D ; supplementary Table 1). The changes observed in the PC, PI, PA, and DAG levels suggest that the polyphosphoinositide (PPI) cycle may be activated in membrane microdomains by C2-Cer stimula-tion ( 45 ).

Changes in microdomain lipids caused by serum stimulation

To assess the responses of microdomain lipids to ago-nist-receptor signaling, we fi rst studied changes in micro-domain lipids within 2 min following TNF- � stimulation, although no changes were observed in the microdomain lipids (data not shown). In our previous study using A431 cells, the results showed that PPI and PA levels in WCs increased as early as 2 min after epidermal growth-factor treatment ( 45 ). To detect rapid changes in microdomain lipids caused by growth factor stimulation, we analyzed changes in constituent lipids of the drSCM fraction prepared from WT- and DKO-tMEFs stimulated with serum. The PA(34:1) level was transiently increased by 2-fold in drSCMs from WT-tMEFs at 2 min post-serum stimulation ( Fig. 4A ). The DAG(16:0/18:1) level within the microdomain was also transiently increased at 2–10 min post-stimulation, while the DAG(18:0/20:4) was gradually increased until 30 min post-stimulation ( Fig. 4A ). Eighty percent of the PA(34:1) production was inhibited by the phospholi-pase D-specifi c inhibitor, FIPI, at 2 min post-stimulation

DKO-tMEFs were also examined. Addition of 100 � M C2-Cer to culture medium supplemented with 10% FBS caused cell death for both tMEF types after 24 h, while the addition of 50 � M C2-Cer had little effect on cell growth. A drSCM fraction was prepared at 60 min post-incubation with 100 � M C2-Cer, and the constituent lipids were ana-lyzed. In a control experiment, 100 � M C2-dhCer, which is a low toxic C2-Cer homolog, was used instead of 100 � M C2-Cer ( 43, 44 ). The levels of some of endogenous Cer molecular species were elevated after the C2-Cer stimulation ( Fig. 3A ). Regarding lipids other than Cer, several molecular species levels whose changes were >1.5-fold are shown in Fig. 3B . Typically, elevations of DAG, PA, and PI were ob-served, along with an elevation in the Cer level, in drSCMs prepared from both WT-tMEFs and DKO-tMEFs. Some differences in microdomain lipid components between WT- and DKO-tMEFs were observed, as follows. Although the Cer(d18:1/22:0) level was little changed in the drSCM fraction of WT-tMEFs following stimulation, it was in-creased by 1.8-fold in the drSCM fraction derived from DKO-tMEFs. The PA(36:2) level was increased by 3.2-fold in drSCMs from WT-tMEFs, while it was little changed in drSCMs from DKO-tMEFs. However, decreased PC(34:1) levels occurred in drSCMs derived from both tMEF types ( Fig. 3C ). The major lipid species (other than these mo-lecular species) were little changed; thus, the proportion of detergent-resistant lipids in these tMEFs was not

Fig. 2. Differences in the levels of membrane lipids between WT-tMEFs and DKO-tMEFs. A: Lipids in each WT- and DKO-tMEF sample were analyzed by LC-MS/MS. The lipid class distribution is shown in the bar chart, which shows comparisons between WT- and DKO-tMEFs. Data are expressed as the mean ± SD (n = 3). B: Percentages of detergent-resistant lipids were compared between WT- and DKO-tMEFs. C: The level of each molecular species of lipids in WC and drSCM fractions was compared between WT- and DKO-tMEFs. D: The acyl-structure distributions were compared between drSCMs from WT- and DKO-tMEFs. Data are expressed as the mean (n = 3).


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




Fig. 2.—continued.


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




Fig. 3. Changes in lipid levels in the drSCM fraction of tMEFs caused by stimulation with C2-Cer. A: Lipid levels in the drSCM fraction, prepared at 60 min after C2-Cer or C2-dhCer addition, were determined by LC-MS/MS. Each level of Cer molecular species was compared between WT- and DKO-tMEFs. The data shown represent the mean of each value from two independent experiments. Fold-increases cal-culated from the mean values are also shown. B: Additional lipid molecular species that were markedly increased following stimulation are shown. C: Changes in the major PC(34:1) species in the drSCM fraction following stimulation are shown. D: Changes in detergent-resistant lipid components following stimulation are shown.


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




Fig. 4. Changes in lipid levels in the drSCM fraction of tMEFs caused by stimulation with serum. A: Time courses of lipid-level production during serum stimulation were monitored in drSCMs prepared from WT-tMEFs. Cells were subjected to serum starvation for 3 h in medium containing 1 mg/ml BSA, after which the culture medium was replaced with medium supplemented with 10% FBS. Lipid levels in the drSCM fraction, prepared at 0–30 min post-serum stimulation, were determined by LC-MS/MS. B: The phospholipase D-specifi c inhibitor, FIPI (750 nM), was added 30 min prior to serum stimulation. The other experimental conditions were the same as described in (A). C: Time courses of lipid-level production during serum stimulation were also monitored with drSCMs prepared from DKO-tMEFs. The data shown represents mean values from two independent experiments. Each lipid level was normalized to total PC quantities.

( 46 ), although DAG(18:0/20:4) production was not af-fected by FIPI treatment at 30 min post-stimulation ( Fig. 4B ). These results suggested that phospholipase D is transiently activated as early as 2 min after serum treatment, while DAG(18:0/20:4) was gradually generated via a different pathway, such as the PPI cycle ( 45 ). The lipid levels of the other classes were little changed in the drSCM fraction, except that the PS level exhibited a tendency to decrease at 10 min post-stimulation ( Fig. 4A ). DKO-tMEFs showed

virtually the same changes as WT-tMEFs ( Fig. 4C ). These results show that changes in PA and DAG levels within plasma membranes following growth factor-receptor acti-vation occur in the same manner, even in SM-defi cient membranes.

As discussed above, we were able to evaluate changes in lipid compositions of membrane microdomains using the method developed in this study. Therefore, this method for analyzing membrane lipids should facilitate future


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




comprehensive analyses of membrane microdomain-associated responses, from the perspective of lipid molecular species.

CONCLUDING REMARKS

In this study, we did not measure PPI levels because the mass spectrometer used had a low sensitivity in the nega-tive-detection mode for PPIs. In addition, we could not perform a detailed ganglioside analysis because of a lim-ited mass range. An alternative LC-MS system, such as the LTQ-Orbitrap, would be needed for conducting PPI and ganglioside analyses ( 45 ). Although the drSCM fraction concentrated the lipid molecular species that are charac-teristic of membrane microdomains, it still retained non-raft proteins bound to the silica beads. Therefore, we could not examine proteins that shuttled in or out of membrane microdomains by this method. Beyond these limitations, this report provides a practical and benefi cial strategy for directly measuring membrane lipids.

REFERENCES

1 . Schuck , S. , M. Honsho , K. Ekroos , A. Shevchenko , and K. Simons . 2003 . Resistance of cell membranes to different detergents. Proc. Natl. Acad. Sci. USA . 100 : 5795 – 5800 .

2 . Pike , L. J. , X. Han , and R. W. Gross . 2005 . Epidermal growth factor receptors are localized to lipid rafts that contain a balance of inner and outer leafl et lipids: a shotgun lipidomics study. J. Biol. Chem. 280 : 26796 – 26804 .

3 . Brügger , B. , B. Glass , P. Haberkant , I. Leibrecht , F. T. Wieland , and H. Krausslich . 2006 . The HIV lipidome: a raft with an unusual composition. Proc. Natl. Acad. Sci. USA . 103 : 2641 – 2646 .

4 . Zech , T. , C. S. Ejsing , K. Gaus , B. de Wet , A. Shevchenko , K. Simons , and T. Harder . 2009 . Accumulation of raft lipids in T-cell plasma membrane domains engaged in TCR signalling. EMBO J. 28 : 466 – 476 .

5 . Pike , L. J. 2009 . The challenge of lipid rafts. J. Lipid Res. 50 : S323 – S328 .

6 . Mollinedo , F. , and C. Gajate . 2010 . Lipid rafts and clusters of apop-totic signaling molecule-enriched rafts in cancer therapy. Future Oncol. 6 : 811 – 821 .

7 . Lingwood , D. , and K. Simons . 2010 . Lipid rafts as a membrane-organizing principle. Science . 327 : 46 – 50 .

8 . Nachbur , U. , D. Kassahn , S. Yousefi , D. F. Legler , and T. Brunner . 2006 . Posttranscriptional regulation of Fas (CD95) ligand killing activity by lipid rafts. Blood . 107 : 2790 – 2796 .

9 . Veldman , R. J. , N. Maestre , O. M. Aduib , J. A. Medin , R. Salvayre , and T. Levade . 2001 . A neutral sphingomyelinase resides in sphin-golipid-enrich microdomains and is inhibited by the caveolin-scaffolding domai: potential implications in tumour necrosis factor signalling. Biochem. J. 355 : 859 – 868 .

10 . Legler , D. F. , O. Micheau , M. A. Doucey , J. Tschopp , and C. Bron . 2003 . Recruitment of TNF receptor 1 to lipid rafts is essential for TNF � -mediated NF- � B activation. Immunity . 18 : 655 – 664 .

11 . Hajduch , E. , S. Turban , X. Le Liepvre , S. Le Lay , C. Lipina , N. Dimopoulos , I. Dugail , and H. S. Hundal . 2008 . Targeting of PKCzeta and PKB to caveolin-enriched microdomains represents a crucial step underpinning the disruption in PKB-directed signal-ling by ceramide. Biochem. J. 410 : 369 – 379 .

12 . Grassme , H. , A. Jekle , A. Riehle , H. Schwarz , J. Berger , K. Sandhoff , R. Kolesnick , and E. Gulbins . 2001 . CD95 signaling via ceramide-rich membrane rafts. J. Biol. Chem. 276 : 20589 – 20596 .

13 . Bryant , M. R. , C. B. Marta , F. S. Kim , and R. Bansal . 2009 . Phosphorylation and lipid raft association of fi broblast growth fac-tor receptor-2 in oligodendrocytes. Glia . 57 : 935 – 946 .

14 . Gutiérrez , J. , D. Cabrera , and E. Brandan . 2014 . Glypican-1 regu-lates myoblast response to HGF via Met in a lipid raft-dependent

mechanism: effect on migration of skeletal muscle precursor cells. Skelet. Muscle . 4 : 5 .

15 . Kajiwara , K. , T. Yamada , T. Bamba , E. Fukusaki , F. Imamoto , M. Okada , and C. Oneyama . 2014 . c-Src-induced activation of ce-ramide metabolism impairs membrane microdomains and pro-motes malignant progression by facilitating the translocation of c-Src to focal adhesions. Biochem. J. 458 : 81 – 93 .

16 . Cremesti , A. , F. Paris , H. Grassmé , N. Holler , J. Tschopp , Z. Fuks , E. Gulbins , and R. Kolesnick . 2001 . Ceramide enables fas to cap and kill. J. Biol. Chem. 276 : 23954 – 23961 .

17 . Jahn , T. , E. Leifheit , S. Gooch , S. Sindhu , and K. Weinberg . 2007 . Lipid rafts are required for Kit survival and proliferation signals. Blood . 110 : 1739 – 1747 .

18 . Suzuki , K. G. N. , R. S. Kasai , K. M. Hirosawa , Y. L. Nemoto , M. Ishibashi , Y. Miwa , T. K. Fujiwara , and A. Kusumi . 2012 . Transient GPI-anchored protein homodimers are units for raft organization and function. Nat. Chem. Biol. 8 : 774 – 783 .

19 . Day , C. A. , and A. K. Kenworthy. 2009 . Tracking microdomain dynamics in cell membranes. Biochim. Biophys. Acta . 1788 : 245 – 253 .

20 . Gao , X. , P. R. Lowry , X. Zhou , C. Depry , Z. Wei , G. W. Wong , and J. Zhang . 2011 . PI3K/Akt signaling requires spatial compartmen-talization in plasma membrane microdomains. Proc. Natl. Acad. Sci. USA . 108 : 14509 – 14514 .

21 . Kiyokawa , E. , T. Baba , N. Otsuka , A. Makino , S. Ohno , and T. Kobayashi . 2005 . Spatial and functional heterogeneity of sphingolipid-rich membrane domains. J. Biol. Chem. 280 : 24072 – 24084 .

22 . Brown , D. A. , and E. London. 2000 . Structure and function of sphingolipid- and cholesterol-rich membrane rafts. J. Biol. Chem. 275 : 17221 – 17224 .

23 . Trajkovic , K. , C. Hsu , S. Chiantia , L. Rajendran , D. Wenzel , F. Wieland , P. Schwille , B. Brügger , and M. Simons . 2008 . Ceramide triggers budding of exosome vesicles into multivesicular endo-somes. Science . 319 : 1244 – 1247 .

24 . Choi , D-S. , D-K. Kim , Y-K. Kim , and Y. S. Gho . 2013 . Proteomics, transcriptomics and lipidomics of exosomes and ectosomes. Proteomics . 13 : 1554 – 1571 .

25 . Yuyama , K. , H. Sun , S. Sakai , S. Mitsutake , M. Okada , H. Tahara , J-I. Furukawa , N. Fujitani , Y. Shinohara , and Y. Igarashi . 2014 . Decreased amyloid- � pathologies by intracerebral loading of glycosphingolipid-enriched exosomes in Alzheimer model mice. J. Biol. Chem. 289 : 24488 – 24498 .

26 . Chaney , L. K. , and B. S. Jacobson . 1983 . Coating cells with col-loidal silica for high yield isolation of plasma membrane sheets and identifi cation of transmembrane proteins. J. Biol. Chem. 258 : 10062 – 10072 .

27 . Arjunan , S. , M. Reinartz , B. Emde , K. Zanger , and J. Schrader . 2009 . Limitations of the colloidal silica method in mapping the endothelial plasma membrane proteome of the mouse heart. Cell Biochem. Biophys. 53 : 135 – 143 .

28 . Mathias , R. A. , Y-S. Chen, R. J. Goode, E. A. Kapp, S. Mathivanan, R. L. Moritz, H-J. Zhu, and R. J. Simpson. 2011 . Tandem application of cationic colloidal silica and Triton X-114 for plasma membrane protein isolation and purifi cation: towards developing an MDCK protein database. Proteomics . 11 : 1238 – 1253 .

29 . Duddempudi , P. K. , P. Nakashe , M. P. Blanton , and M. Jansen . 2013 . The monomeric state of the proton-coupled folate trans-porter represents the functional unit in the plasma membrane. FEBS J. 280 : 2900 – 2915 .

30 . Ikenouchi , J. , M. Suzuki , K. Umeda , K. Ikeda , R. Taguchi , T. Kobayashi , S. B. Sato , T. Kobayashi , D. B. Stolz , and M. Umeda . 2012 . Lipid polarity is maintained in absence of tight junctions. J. Biol. Chem. 287 : 9525 – 9533 .

31 . Lorizate , M. , T. Sachsenheimer , B. Glass , A. Habermann , M. J. Gerl , H-G. Kräusslich , and B. Brügger . 2013 . Comparative lipido-mics analysis of HIV-1 particles and their producer cell membrane in different cell lines. Cell. Microbiol. 15 : 292 – 304 .

32 . Matsushima , T. , Y. Saito , J. I. Elliott , K. Iijima-Ando , M. Nishimura , N. Kimura , S. Hata , T. Yamamoto , T. Nakaya , and T. Suzuki . 2012 . Membrane-microdomain localization of amyloid � -precursor protein (APP) C-terminal fragments is regulated by phosphorylation of the cytoplasmic Thr668 residue. J. Biol. Chem. 287 : 19715 – 19724 .

33 . Kim , Y. , S. Elschenbroich , P. Sharma , L. Sepiashvili , O. Anthony , and T. Kislinger . 2011 . Use of colloidal sillica-beads for the iso-lation of cell-surface proteins for mass spectrometry-based pro-teomics. Methods Mol. Biol. 748 : 227 – 241 .


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/




34 . Elschenbroich , S. , V. Ignatchenko , P. Sharma , G. Schmitt-ulms , A. O. Gramolini , and T. Kislinger . 2009 . Peptide separations by on-line MudPIT compared to isoelectric focusing in an Off-Gel for-mat : application to a membrane-enriched fraction from C2C12 mouse skeletal muscle cells. J. Proteome Res. 8 : 4860 – 4869 .

35 . Asano , S. , K. Kitatani , M. Taniguchi , M. Hashimoto , K. Zama , S. Mitsutake , Y. Igarashi , H. Takeya , J. Kigawa , A. Hayashi , et al . 2012 . Regulation of cell migration by sphingomyelin synthases: sphingo-myelin in lipid rafts decreases responsiveness to signaling by the CXCL12/CXCR4 pathway. Mol. Cell. Biol. 32 : 3242 – 3252 .

36 . Ogiso , H. , M. Taniguchi , S. Araya , S. Aoki , L. O. Wardhani , Y. Yamashita , Y. Ueda , and T. Okazaki . 2014 . Comparative analysis of biological sphingolipids with glycerophospholipids and diacylglyc-erol by LC-MS/MS. Metabolites . 4 : 98 – 114 .

37 . Navas , P. , D. D. Nowack , and D. J. Moirã . 1989 . Isolation of puri-fi ed plasma Membranes from cultured cells and hepatomas by two-phase partition and preparative free-fl ow electrophoresis. Cancer Res. 49 : 2147 – 2156 .

38 . Pike , L. J. , X. Han , K-N. Chung , and R. W. Gross . 2002 . Lipid rafts are enriched in arachidonic acid and plasmenylethanolamine and their composition is independent of caveolin-1 expression: a quantitative electrospray ionization/mass spectrometric analysis. Biochemistry . 41 : 2075 – 2088 .

39 . Hattersley , K. J. , L. K. Hein , and M. Fuller . 2013 . Lipid composition of membrane rafts, isolated with and without detergent, from the spleen of a mouse model of Gaucher disease. Biochem. Biophys. Res. Commun. 442 : 62 – 67 .

40 . Hein , L. K. , S. Duplock , J. J. Hopwood , and M. Fuller . 2008 . Lipid composition of microdomains is altered in a cell model of Gaucher disease. J. Lipid Res. 49 : 1725 – 1734 .

41 . Bhat , H. B. , T. Kishimoto , M. Abe , A. Makino , T. Inaba , M. Murate , N. Dohmae , A. Kurahashi , K. Nishibori , F. Fujimori , et al . 2013 . Binding of a pleurotolysin ortholog from Pleurotus eryngii to sphingomyelin and cholesterol-rich membrane domains. J. Lipid Res. 54 : 2933 – 2943 .

42 . Taniguchi , M. , and T. Okazaki . 2014 . The role of sphingomy-elin and sphingomyelin synthases in cell death, proliferation and migration - From cell and animal models to human disorders. Biochim. Biophys. Acta . 1841 : 692 – 703 .

43 . Shikata , K. , H. Niiro , H. Azuma , K. Ogino , and T. Tachibana . 2003 . Apoptotic activities of C2-ceramide and C2-dihydroceramide ho-mologues against HL-60 cells. Bioorg. Med. Chem. 11 : 2723 – 2728 .

44 . Bielawska , A. , H. M. Crane , L. M. Obeid , and A. Yusuf . 1993 . Selectivity of ceramide-mediated biology. J. Biol. Chem. 268 : 26226 – 26232 .

45 . Ogiso , H. , and R. Taguchi . 2008 . Reversed-phase LC/MS method for polyphosphoinositide analyses : changes in molecular species levels during epidermal growth factor activation in A431 cells. Anal. Chem. 80 : 9226 – 9232 .

46 . Su , W. , O. Yeku , S. Olepu , A. Genna , J. Park , H. Ren , G. Du , M. H. Gelb , A. J. Morris , and M. A. Frohman . 2009 . 5-Fluoro-2-indolyl des-chlorohalopemide (FIPI), a phospholipase D pharmacological inhibitor that alters cell spreading and inhibits chemotaxis. Mol. Pharmacol. 75 : 437 – 446 .


ww

.jlr.orgD

ownloaded from


http://www.jlr.org/



analysis of lipid-composition changes in plasma membrane ... · scm, silica-coated membrane; sms,...

Documents