Method for Lipidomic Analysis: p53 Expression Modulation of Sulfatide

Figure 6 Sulfatide profiles (see also Table 2) for the polar lipid fraction of extracts of U87 cells following different treatments. Treatment of U87 ...
1 downloads 0 Views 354KB Size
Anal. Chem. 2007, 79, 8423-8430

Accelerated Articles

Method for Lipidomic Analysis: p53 Expression Modulation of Sulfatide, Ganglioside, and Phospholipid Composition of U87 MG Glioblastoma Cells Huan He,†,‡ Charles A. Conrad,§ Carol L. Nilsson,† Yongjie Ji,§ Tanner M. Schaub,† Alan G. Marshall,*,†,‡ and Mark R. Emmett*,†,‡

National High Magnetic Field Laboratory, Florida State University, 1800 East Paul Dirac Drive, Tallahassee, Florida 32310-4005, Department of Chemistry and Biochemistry, Florida State University, Tallahassee, Florida, 32306-4390, and M.D. Anderson Cancer Center, Department of Neuro-Oncology, 1515 Holcombe Boulevard, Houston, Texas 77030

Lipidomics can complement genomics and proteomics by providing new insight into dynamic changes in biomembranes; however, few reports in the literature have explored, on an organism-wide scale, the functional link between nonenzymatic proteins and cellular lipids. Here, we report changes induced by adenovirus-delivered wildtype p53 gene and chemotherapy of U87 MG glioblastoma cells, a treatment known to trigger apoptosis and cell cycle arrest. We compare polar lipid changes in treated cells and control cells by use of a novel, sensitive method that employs lipid extraction, one-step liquid chromatography separation, high-resolution mass analysis, and Kendrick mass defect analysis. Nano-LC FT-ICR MS and quadrupole linear ion trap MS/MS analysis of polar lipids yields hundreds of unique assignments of glyco- and phospholipids at sub-ppm mass accuracy and high resolving power (m/∆m50% ) 200 000 at m/z 400) at 1 s/scan. MS/MS data confirm molecular structures in many instances. Sulfatides are most highly modulated by wild-type p53 treatment. The treatment also leads to an increase in phospholipids such as phosphatidyl inositols, phosphatidyl serines, phosphatidyl glycerols, and phosphatidyl ethanolamines. An increase in hydroxylated phospholipids is especially noteworthy. Also, a decrease in the longer chain gangliosides, GD1 and GM1b, is observed in wildtype p53 (treated) cells.

10.1021/ac071413m CCC: $37.00 Published on Web 10/12/2007

© 2007 American Chemical Society

Although several LC-MS methods for cellular lipids have been published previously,1-7 a new approach is needed for polar lipids to include phospholipids and complex glycolipids such as gangliosides. Most prior analyses focus on a subset of lipids, such as phospholipids and smaller glycolipids (i.e., glucosyl ceramides)5,6,8,9 or larger glycolipids (gangliosides).10 Concomitant measurements of both types of lipid can present a broader picture of how cell membranes (e.g., in tumor cells) change in response to stimulation such as gene therapy or pharmacological agents. * To whom correspondence should be addressed. Telephone: 1-850-644-0648. Fax: 1-850-644-1366. E-mail: [email protected]. Telephone: 1-850-6440529. Fax: 1-850-644-1366. E-mail: [email protected]. † National High Magnetic Field Laboratory, Florida State University. ‡ Department of Chemistry and Biochemistry, Florida State University. § M.D. Anderson Cancer Center. (1) Sommer, U.; Herscovitz, H.; Welty, F. K.; Costello, C. E. J. Lipid Res. 2006, 47, 804-814. (2) Mauri, L.; Valsecchi, M.; Casellato, R.; Li, S.-C.; Li, Y.-T.; Sonnino, S. J. Chromatogr., B 2003, 796, 1-10. (3) Merrill, J. A. H.; Sullards, M. C.; Allegood, J. C.; Kelly, S.; Wang, E. Methods 2005, 36, 207-224. (4) Dasgupta, S.; E. L., H. J. Lipid Res. 2001, 42, 301-308. (5) Kim, H.-Y.; Wang, T.-C. L.; Ma, Y.-C. Anal. Chem. 1994, 66, 3977-3982. (6) Bielawski, J.; Szulc, Z. M.; Hannun, Y. A.; Bielawska, A. Methods 2006, 39, 82-91. (7) Ishida, M.; Yamazaki, T.; Houjou, T.; Imagawa, M.; Harada, A.; Inoue, K.; Taguchi, R. Rapid Commun. Mass Spectrom. 2004, 18, 2486-2494. (8) Schwudke, D.; Hannich, J. T.; Surendranath, V.; Grimard, V.; Moehring, T.; Burton, L.; Kurzchalia, T.; Shevchenko, A. Anal. Chem. In press. (9) Forrester, J.; Milne, S. B.; Ivanova, P.; Brown, H. A. Mol. Pharmacol. 2004, 65, 813-821. (10) Hoffman, R. C.; Jennings, L. L.; Tsigelny, I.; Comoletti, D.; Flynn, R. E.; Sudhof, T. C.; Taylor, P. Biochemistry 2004, 43, 1496-1506.

Analytical Chemistry, Vol. 79, No. 22, November 15, 2007 8423

Intact lipid analysis benefits from the high-resolution and highmass accuracy offered by Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS),11 supplemented by quadrupole linear ion trap MS/MS. Recently, glycosphingolipids (GSLs) from gliosarcoma and normal brain have been analyzed by thin-layer chromatography (TLC) and nanoelectrospray (nanoESI) quadrupole time-of-flight or FT-ICR mass spectrometry,12 to yield a distinctive GSL pattern in the tumor material in which GD3 dominated and GM1 was depleted. Unusual minor species and a novel O-acetylated GD3 component were also characterized. For decades, a complex lipid mixture was class-separated by TLC followed by hydrolysis and derivatization to generate volatile analytes for GC/MS analysis. Ionization techniques such as fast atom bombardment,13,14 field desorption,15 and chemical ionization16,17 allowed for direct analysis of intact polar lipid but lacked popularity due to inherent limitations. Both nano-ESI5,18-27 and matrix-assisted laser desorption/ionization28-30 ionization techniques were extensively used for lipid characterization and quantification because of their high sensitivity, high reproducibility, and less complex procedure. Nano-ESI has the additional advantages of easy coupling to high-performance liquid chromatography and reduction of oxidation of unsaturated lipids because lipids avoid direct contact with atmospheric oxygen before being electrosprayed from the LC effluent. Phospholipids are key components of the cell membrane. They have highly diversified structures31 based on variation of the polar head groups: e.g., inositol in phosphatidylinositol (PI), serine in phosphatidylserine (PS), glycerol in phosphatidylglycerol (PG), (11) McFarland, M. A.; Marshall, A. G.; Hendrickson, C. L.; Fredman, P.; Mansson, J. E.; Nilsson, C. L. J. Mass Spectrom. 2005, 16, 752-762. (12) Vukelic, Z.; Kalanj-Bognar, S.; Froesch, M.; Bindila, L.; Radic, B.; Allen, M.; Peter-Katalinic, J.; Zamfir, A. D. Glycobiology 2007. (13) Lehmann, W. D.; Kessler, M. Chem. Phys. Lipids 1983, 32, 123-135. (14) Chilton, F. H.; Murphy, R. C. Biomed. Environ. Mass Spectrom. 1986, 13, 71-76. (15) Lehmann, W. D.; Kessler, M. Biomed. Mass Spectrom. 1983, 10, 220226. (16) Haroldsen, P. E.; Murphy, R. C. Environ. Mass Spectrom. 1987, 14, 573578. (17) Jungalwala, F. B.; Evans, J. E.; McCluer, R. H. J. Lipid Res. 1984, 25, 738749. (18) Ivleva, V. B.; Elkin, Y. N.; Budnik, B. A.; Moyer, S. C.; O’Connor, P. B.; Costello, C. E. Anal. Chem. 2004, 76, 6484-6491. (19) Wenk, M. R.; Lucast, L.; Paolo, G. D.; Romanelli, A. J.; Suchy, S. F.; Nussbaum, R. L.; Cline, G. W.; Shulman, G. I.; McMurray, W.; De Camilli, P. Nat. Biotechnol. 2003, 21, 813-817. (20) Murphy, R. C. The Handbook of Lipid Research; Plenum Press: New York, 1993. (21) Brugger, B.; Erben, G.; Sandhoff, R.; Wieland, F. T.; Lehmann, W. D. Proc. Natl. Acad. Sci U.S.A. 1997, 94, 2339-2344. (22) Koivusalo, M.; Haimi, P.; Heikinheimo, L.; Kostiainen, R.; Somerharju, P. J. Lipid Res. 2001, 663-672. (23) Kevela, J. H.; Kim, H.-Y. Anal. Biochem. 2001, 292, 130-138. (24) Han, X.; Gross, R. W. Proc. Natl. Acad. Sci U.S.A. 1994, 91, 10635-10639. (25) Kerwin, J. L.; Tuininga, A. R.; Ericsson, L. H. J. Lipid Res. 1994, 35, 11021114. (26) Pulfer, M.; Murphy, R. C. Mass Spectrom. Rev. 2003, 22, 332-364. (27) Kwon, G.; Bohrer, A.; Han, X.; Corbett, J. A.; Ma, Z.; Gross, R. W.; McDaniel, M. L.; Turk, J. J. Biochim. Biophys. Acta 1996, 1300, 63-72. (28) Juhasz, P.; Costello, C. E. J. Am. Soc. Mass Spectrom. 1992, 3, 785. (29) Schiller, J.; Arnhold, J.; Benard, M.; Muller, M.; Reichl, S.; Arnold, K. Anal. Biochem. 1999, 267, 46-56. (30) Marto, J. A.; White, F. M.; Rietschel, E. T.; Ben-Menachem, G.; Deutsch, J.; Rottem, S. J. Biol. Chem. 1997, 272, 26262-26270. (31) Fahy, E.; Subramaniam, S.; Brown, H. A.; Glass, C. K.; Merrill, J. A. H.; Murphy, R. C.; Raetz, C. R. H.; Russell, D. W.; Seyama, Y.; Shaw, W.; Shimizu, T.; Spener, F.; van Meer, G.; VanNieuwenhze, M. S.; White, S. H.; Witztum, J. L.; Dennis, E. A. J. Lipid Res. 2005, 46, 839-861.

8424 Analytical Chemistry, Vol. 79, No. 22, November 15, 2007

Figure 1. Selected ion chromatograms generated by nano-LC ESI negative-ion FT-ICR MS of a mixture of standard gangliosides. 1 pmol of GM1, GD1a, and GT1b ((d18:1/18:0) and (d20:1/18:0)) was loaded on the column.

Figure 2. Sum of ∼1100 scans of broadband nano-LC-ESI negative-ion FT-ICR MS of the cell extract described in Experimental Methods. Inset: 1400 < m/z < 1650 spectral segment resolves lipids of more than 250 different elemental compositions, including endogenous GM1.

etc. The polar head group is attached to the terminal hydroxyl group of the glycerol backbone and fatty acyl or alkyl groups are attached to the glycerol backbone by ester or ether bonds, respectively. The nonpolar tail varies in total carbon, double bond and oxidation (peroxidation or hydroxylation) site numbers, location of the double bond and oxidation site, etc. Phospholipids can be analyzed as either positive or negative ions, according to the possible charge states of the phospholipids in solution.21,32,33 Six major phospholipid subclasses (including PI, PS, PG, phosphatidic acids, phosphatidylethanolamine (PE), and adducts of phosphatidylcholine (PC)) as well as the lyso forms of these six subclasses can be detected as negative ions. Four major subclasses (including PC, PS, PE and sphingomyelins ) can be detected as positive ions.32 The fragmentation pattern generated by CID tandem mass spectrometry of the phospholipid (especially for negative ions) can provide valuable information on the polar head groups, backbone, and fatty acyl structures. Observation of neutral loss of fatty acyl chains and inositol groups and fatty acid anions can help reconstruct the PI structures. Sub-ppm mass (32) Milne, S.; Ivanova, P.; Forrester, J.; Brown, H. A. Methods 2006, 39, 92103. (33) Murphy, R. C. Mass Spectrometry of Phospholipids: Tables of Molecular and Product Ions; Illuminati Press: Denver, NV, 2002.

Figure 3. Linear quadrupole ion trap product ion mass spectrum for precursor ions of m/z 1354.783. The fragmentation pattern is most consistent with the structure of GM2R.

accuracy can unambiguously assign the total carbon, double bond and oxidation site numbers for the fatty acyl chains. It may also be possible to locate the exact esterification site of each fatty acyl groups based on the relative abundances of fatty acid anions.33 The locations of double bond and hydroxylation site can be elucidated by applying high-energy collisional activation.34 Ozone ESI-MS can also be applied to elucidate the position of the double bonds.35 Glycosphingolipids are a structurally diverse class of lipids containing glycosylated ceramide (N-acylsphingoid bases).31 Sulfatides and gangliosides belong to the acidic GSL subclasses. Both classes can be detected as positive or negative ions. However, due to the presence of acidic groups (e.g., sulfate and sialic acid), the S/N for molecular ions is much higher for negative ions.36,37 The origin of the complexity of the GSL structures comes from isomeric variations of the saccharide head, as well as chain length, degree of unsaturation, and oxidation (hydroxylation and peroxidation) of the fatty acyl chain and sphingoid base. Tandem mass spectrometry by CID provides fragmentation data which allow the assignment of glycan connectivity in the GSL precursor ion. Lowenergy CID provides mainly glycosidic cleavages, whereas crossring cleavages resulting from high-energy CID provide additional information about glycan structures and double bond location of the ceramides.36 Sub-ppm mass accuracy allows for the confident assignment of chain length and degree of unsaturation and oxidation of the ceramides. We have developed a method of solvent/solvent extraction of cell cultures coupled to a novel, highly sensitive, single-step nanoLC separation with microelectrospray (micro-ESI) ionization. We then combine accurate mass and MS/MS to investigate changes (34) Cheng, C.; Gross, M. L. Mass Spectrom. Rev. 2000, 19, 398-420. (35) Thomas, M. C.; Mitchell, T. W.; Harman, D. G.; Deeley, J. M.; Murphy, R. C.; Blanksby, S. J. Anal. Chem. 2007, 79, 5013-5022. (36) Levery, S. B. Methods Enzymol. 2005, 405, 300-369. (37) Kushi, Y.; Handa, S.; Ishizuka, I. J. BioChem. (Tokyo) 1985, 97, 419-428.

Figure 4. Kendrick mass defect vs mass-to-charge ratio provides simple visual classification of lipid component classes that differ in heteroatom elemental composition. For each heteroatom class (NnOoSs), points lying on the same horizontal line correspond to species differing in the number of CH2 groups.

in polar lipid expression in U87 tumor cells in response to adenovirus therapy in which the vector carries either the wildtype p53 gene or no functional gene, followed by treatment with SN-38, a topoisomerase inhibitor and the active metabolite of Irinotecan (CPT-11). This treatment is known to decrease expression of galectin-1, a galactose-binding protein, in U87 cells38 and induce moderate apoptosis and G2 cell cycle arrest. Reverse order of treatment produces almost complete G2 arrest and apoptosis of 90% of the cells. Cell extracts are separated by nanoliquid chromatography and analyzed in a hybrid linear ion trap FT-ICR mass spectrometer. Separation of glyco- and phospholipids by LC is a challenging, but necessary step to obtain the most complete molecular coverage of the lipid components in a complex mixture. (38) Puchades, M.; Nilsson, C. L.; Emmett, M. R.; Aldape, K. D.; Ji, Y.; Lang, F. F.; Liu, T. J.; Conrad, C. A. J. Proteome Res. 2007, 6, 869-875.

Analytical Chemistry, Vol. 79, No. 22, November 15, 2007

8425

Figure 5. Ganglioside profiles (see also Table 1) for the polar lipid fraction of extracts of U87 cells following different treatments. The listed ceramides could be isomer(s) of the assigned structure. The effect of treatment with wild-type p53, which lowers the abundance of longer chain ceramides, is most pronounced for the GM1b and GD1 gangliosides. The data shown represent the average of duplicates. The error bars depict the actual “range of error” (i.e., the high and low values of the data). Data without range of error bars mean that the duplicates were the same.

EXPERIMENTAL METHODS Cell Culture and Treatment. The glioma cell line U87 MG (ATCC HTB-14) was grown in the presence of DMEM-F12 media (Gibco/Invitrogen, Carlsbad, CA) supplemented with 10% FBS (Gibco/Invitrogen) in a humidified CO2 incubator at 5% CO2. Cell cultures U87 MG were treated with adenoviruses (therapeutic Adp53 or Dl-312 control adenovirus vector) for 24 h and subsequent cytotoxic chemotherapy for 24 h with SN-38 (7-ethyl-10-hydroxycamptothecin) at a final concentration of 0.1 µM (stock solution of 10 mM). Cell cultures were also treated with either controlled virus Dl-312 at 1:100 multiplicity of infection (MOI) from a stock virus titered at 2.8 × 1011 plaque-forming units (pfu)/mL or test virus, which contained wild-type p53 gene, inserted within the E1 region of the adenovirus vector (Ad-p53). This was similarly used at 1:100 MOI obtained from a stock of preparation of 2.2 × 1010 pfu/mL. Cell cultures that were treated with a combination of drug and virus included a total incubation period of 48 h allowing 24 h for each agent. Cells were washed three times with roomtemperature PBS between treatments. Prior to viral infection, the cells were placed in serum-free media for 1 h to ensure adequate absorption of virus to the cells. Polar Lipid Extraction. Cells (∼2 × 106) were extracted by the addition of methanol/chloroform 1:1 and sonicated for 30 min. The extract was incubated overnight at 48 °C to optimize GSL yields. After centrifugation, the supernatant was collected and partitioned with additional chloroform and H2O (chloroform/H2O 4:11). The upper layer was collected and dried. Approximately 1/50 of the total extract was consumed per LC-MS experiment. LC-MS. The lipids were reconstituted in 80% methanol (aq) with the addition of 10 mM NH4OAc. The liquid chromatography was first optimized in micro-LC columns (50 × 1 mm) with ganglioside standards and modified to fit a nano-LC format. Data were collected by separation with nanoliquid chromatography 8426

Analytical Chemistry, Vol. 79, No. 22, November 15, 2007

(Eksigent 1D system, Livermore, CA) in a self-packed 80 mm × 50 µm phenylhexyl (Phenomenex, Torrance, CA) column. Separation is mainly achieved by dual selectivity of π-π interactions and hydrophobic interactions between the phenylhexyl resin and the analytes. The separation method was optimized for the separation of the bovine ganglioside standards GD1a, GM1, and GT1b (Sigma, St. Louis, MO). The gradient was 15%/85% to 2%/98% A/B during 4 min (Solvent A: 98% H2O, 2% methanol, and 10 mM NH4OAc. Solvent B: 98% methanol, 2% H2O, 10 mM NH4OAc) at a flow rate of 400 nL/min. LC effluent was analyzed on-line by negative-ion micro-ESI39,40 and a modified hybrid linear ion trap FT-ICR MS equipped with a 14.5 T magnet.41 Precursor ion mass spectra were collected at high mass resolving power (m/∆m50% ) 200 000 at m/z 400) and scan rate of >1 Hz. Typical broadband external calibration mass accuracy was better than 500 ppb. Datadependent MS/MS was performed in the linear ion trap (collisional dissociation) during collection of the ICR time domain data. In a separate experiment, multiplexed data-dependent LC MS/ MS (based on ion signals observed in ICR precursor ion spectra) was performed in the linear ion trap (collisional dissociation) at 4 s per composite scan. Data were analyzed by use of a Kendrick mass analysis algorithm developed in-house.42 RESULTS AND DISCUSSION Separation and Identification of Lipidome Components. LC-MS is necessary to separate and identify polar lipids in a (39) Emmett, M. R.; White, F. M.; Hendrickson, C. L.; Shi, S. D.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 1998, 9, 333-340. (40) Emmett, M. R.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 1994, 5, 605613. (41) Schaub, T. M.; Blakney, G. T.; Hendrickson, C. L.; Quinn, J. P.; Senko, M. W.; Marshall, A. G. 55th ASMS Conference, Indianapolis, IN, June 3-7, 2007; Elsevier. (42) Hughey, C. A.; Hendrickson, C. L.; Rodgers, R. P.; Marshall, A. G.; Qian, K. Anal. Chem. 2001, 73, 4676-4681.

Table 1. MS- and MS/MS-Based (See Figure 5) Assignments of Gangliosides from the Polar Lipid Fraction of Extracts of U87 Cells Treated with Wild-Type p53 or Empty Vector Followed by SN-38 assignment

measured m/z

theoretical m/z

error (ppm)

molecular formula

S/N (I) wt p53

S/N (II) empt vect

ratio II/I

GM3 (d18:1/14:0) GM3 (d18:1/15:0) GM3 (d18:1/16:0) GM3 (d18:1/16:0)+O GM3 (d18:1/16:1) GM3 (d18:1/17:0) GM3 (d18:1/18:0) GM3 (d18:1/18:1) GM3 (d18:1/22:0) GM3 (d18:1/22:1) GM3 (d18:1/23:0) GM3 (d18:1/24:0) GM3 (d18:1/24:1) GM2R (d18:1/16:0) GM2R (d18:1/22:0) GM2R (d18:1/24:0) GM2R (d18:1/24:1) GM1b (d18:1/16:0) GM1b (d18:1/16:0)+O GM1b (d18:1/17:0) GM1b (d18:1/22:0) GM1b (d18:1/23:0) GM1b (d18:1/24:0) GM1b (d18:1/24:1) GM1b (d18:1/24:1)+O asialo-GM1 (d18:1/16:0) GD1 (d18:1/16:0) GD1 (d18:1/22:0) GD1 (d18:1/23:0) GD1 (d18:1/23:1) GD1 (d18:1/24:0) GD1 (d18:1/24:1)

1123.6745 1137.6899 1151.7054 1167.7008 1149.6899 1165.7206 1179.7365 1177.7204 1235.7986 1233.7839 1249.8148 1263.831 1261.8145 1354.7847 1438.8783 1466.909 1464.8934 1516.8369 1532.8326 1530.853 1600.9307 1614.9466 1628.9626 1626.9473 1642.9434 1225.7426 1807.9329 1892.0273 1906.0421 1904.0267 1920.0587 1918.0423

1123.6746 1137.6902 1151.7059 1167.7008 1149.6902 1165.7215 1179.7372 1177.7215 1235.7998 1233.7841 1249.8154 1263.8311 1261.8154 1354.7853 1438.8792 1466.9105 1464.8948 1516.8381 1532.833 1530.8537 1600.932 1614.9476 1628.9633 1626.9476 1642.9426 1225.7427 1807.9335 1892.0274 1906.0431 1904.0274 1920.0587 1918.0431

-0.09 -0.26 -0.43 0 -0.26 -0.77 -0.59 -0.93 -0.97 -0.16 -0.48 -0.08 -0.71 -0.44 -0.63 -1.02 -0.96 -0.79 -0.26 -0.46 -0.81 -0.62 -0.43 -0.18 0.49 -0.08 -0.34 -0.07 -0.54 -0.38 -0.01 -0.43

C55H100N2O21 C56H102N2O21 C57H104N2O21 C57H104N2O22 C57H102N2O21 C58H106N2O21 C59H108N2O21 C59H106N2O21 C63H116N2O21 C63H114N2O21 C64H118N2O21 C65H120N2O21 C65H118N2O21 C65H117N3O26 C71H129N3O26 C73H133N3O26 C73H131N3O26 C71H127N3O31 C71H127N3O32 C72H129N3O31 C77H139N3O31 C78H141N3O31 C79H143N3O31 C79H141N3O31 C79H141N3O32 C60H110N2O23 C82H144N4O39 C88H156N4O39 C89H158N4O39 C89H156N4O39 C90H160N4O39 C90H158N4O39

60 90 4000 500 200 200 100 30 60 20 30 400 300 1000 30 70 100 900 60 40 9 1 100 100 20 30 100 40 200 2 3 40

50 100 4000 500 100 200 200 50 300 80 70 600 500 1000 90 200 200 1000 60 60 100 30 400 400 30 200 200 300 200 40 700 700

1 1 1 1 1 1 2 2 5 4 2 2 2 1 3 3 2 1 1 2 11 30 4 4 2 7 2 8 1 20 233 18

Table 2. MS and MS/MS Based (See Figure 6) Sulfatides from the Polar Lipid Fraction of Extracts of U87 Cells Treated with Wild-Type p53 or Empty Vector Followed by SN-38 sulfatide assignt

measured m/z

theoretical m/z

error (ppm)

molecular formula

S/N (I) wt p53

S/N (II) empt vect

ratio I/II

(34:1) (34:1)+O (34:2) (34:2)+O (42:2)

778.5144 794.5094 776.4987 792.4932 888.624

778.5145 794.5094 776.4988 792.4937 888.624

-0.13 0 -0.13 -0.63 0

C40H77N1O11S1 C40H77N1O12S1 C40H75N1O11S1 C40H75N1O12S1 C48H91N1O11S1

500 100 200 90 100

1 5 0 0 70

500 20 infinity infinity 1

complex mixture. Direct infusion of the polar lipid fraction yields a small number of peaks, and none of the larger gangliosides (GM1 and GD1) could be detected (data not shown). Conventional reversed-phase chromatography (C4, C8, or C18) did not elute the longer ceramide GSLs. The application of highly organic mobile phases and the phenylhexyl chromatographic medium permits single-step separation of a mixture of bovine ganglioside standards with high sensitivity (Figure 1). In contrast, nano-LCMS of the much more complex cell extract yielded several hundred well-resolved peaks, 400 < m/z < 2000 (Figure 2), while consuming only a small fraction of the total extract. For many of those signals, MS/MS analysis confirmed molecular assignments based on precursor accurate mass alone. Glycosphingolipids can have several isomeric forms with regard to both the carbohydrate structure and the composition of the ceramide. For the m/z 1354.783 species assigned as GM2, the MS/MS fragmentation pattern is most consistent with the carbohydrate structure of GM2R (Figure 3).

Lipidomic Profiles. All biological extracts consist of highly complex mixtures. A small aliquot (1/50 of each sample) was loaded onto the nano-LC column for each analysis. Each sample was compared to its corresponding control. The MS signal magnitude of ions of various m/z provides a semiquantitative relative abundance profile of the polar lipids. Duplicate measurements for each sample yielded consistent results in all cases. All histograms include “range of error” bars (i.e., the high and low values of the data) to demonstrate reproducibility. A 2-fold or higher change was considered to be significant for a given lipid. To simplify data analysis, lipid homologous series were sorted and identified according to their Kendrick mass defects (KMD, Figure 4).11,42,43 (Species with the same Kendrick mass defect comprise a homologous series differing only in the number of CH2 groups.) Combination of KMD, high mass accuracy (