93

Journal of Oleo ScienceCopyright ©2011 by Japan Oil Chemists’ SocietyJ. Oleo Sci. 60, (2) 93-98 (2011)

Selective Analysis of Lipids by Thin-Layer Chromatography Blot Matrix-Assisted Laser Desorption/Ionization Imaging Mass SpectrometryNobuhiro Zaima1＊ , Naoko Goto-Inoue1, Kohsuke Adachi2 and Mitsutoshi Setou1

1 Department of Molecular Anatomy, Hamamatsu University School of Medicine (1-20-1 Handayama, Higashi-ku, Hamamatsu, Shizuoka 431-3192, JAPAN)

2 Graduate School of Agriculture, Kochi University (2000 Otsu, Monobe, Nankoku-shi, Kochi 783-8502, JAPAN)

1 IINTRODUCTION Thin-layer chromatography（TLC）has been previously

applied for the analysis of food extracts, including lipid mixtures1－4）. When no established high-performance liquid chromatography（HPLC）method is suitable for an analysis, TLC can be used as an analytical method. For example, TLC can be used for lipid analysis, to identify different classes of phospholipids such as phosphatidylcholine（PC）, phosphatidylethanolamine（PE）, phosphatidylinositol（PI）, phosphatidylserine（PS）, and sphingomyelin（SM）5）. There-fore, TLC is a powerful method for the analysis of complex food extracts. However, this method cannot be used to an-alyze phospholipids at the molecular species level. Numer-ous molecular phospholipid species are present in food as well as human tissue, and each species is suggested to play specifi c roles6）. Thus, it is important to analyze phospholip-ids at the molecular species level. For this purpose, imag-ing mass spectrometry（IMS）, which was developed using matrix-assisted laser desorption/ionization mass spectrom-etry（MALDI-MS）, is a suitable tool7, 8）. IMS enables visual-ization of individual molecules present on or near the sur-

＊Correspondence to: Nobuhiro Zaima, Department of Molecular Anatomy, Hamamatsu University School of Medicine, 1-20-1, Handayama, Higashi-ku, Hamamatsu, Shizuoka 431-3192, JAPANE-mail: zaimanobuhiro@gmail.comAccepted July 16, 2010 (received for review June 25, 2010)Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 onlinehttp://www.jstage.jst.go.jp/browse/jos/

face of tissue sections without requiring antibodies, staining, or complicated pretreatment9）. Direct analysis of a tissue section with IMS allows for the detection of a wide range of molecules such as lipids10）, peptides11）, and nutri-ents12, 13）as well as administered pharmaceuticals14）. Al-though IMS can be used to directly analyze the TLC sur-face（TLC-IMS）, complete retrieval of the target compound from the silica coated onto the plate is diffi cult. This prob-lem was largely resolved by transfer of separated lipids from the TLC plate onto polyvinylidene difl uoride（PVDF）membranes, which is referred to as TLC-blotting15）. The spectral quality on PVDF membranes is reported to be bet-ter than that on TLC plates in terms of sensitivity, mass resolution, and background interference16）. Recently, we developed a new method that combined TLC-blotting and MALDI-IMS（TLC-Blot-MALDI-IMS）17）. This method en-abled us to visualize PC and SM molecules and analyze their structures by tandem MS（MS/MS）on PVDF mem-branes. However, it is still unknown whether TLC-Blot-MALDI-IMS can be applied to the analysis of other phos-pholipids.

NOTE

Abstract: Thin-layer chromatography (TLC) is an essential method for food composition analyses such as lipid nutrition analysis. TLC can be used to obtain information about the lipid composition of foods; however, it cannot be used for analyses at the molecular level. Recently we developed a new method that combines matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI-IMS) with TLC-blotting (TLC-Blot-MALDI-IMS). The combination of MALDI-IMS and TLC blotting enabled detailed and sensitive analyses of lipids. In this study, we applied TLC-Blot-MALDI-IMS for analysis of major phospholipids extracted from bluefi n tuna. We showed that TLC-Blot-MALDI-IMS analysis could visualize and identify major phospholipids such as phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, phosphatidylcholine, and sphingomyelin.

Key words: thin-layer chromatography, imaging mass spectrometry, phospholipids, TLC-Blot-MALDI-IMS, tuna

N. Zaima, N. Goto-Inoue, K. Adachi & M. Setou

J. Oleo Sci. 60, (2) 93-98 (2011)

94

In this study, we applied TLC-Blot-MALDI-IMS to ana-lyze lipids extracted from bluefin tuna（Thunnus thyn-nus）. We show that TLC-Blot-MALDI-IMS analysis can vi-sualize and identify major phospholipids such as PE, PI, PS, PC and SM.

2 EXPERIMENTAL PROCEDURES2.1 Reagents

High performance TLC（HPTLC）silica gel 60 was pur-chased from Merck（Darmstadt, Germany）. Methanol was purchased from Kanto Chemical Company Inc.（Tokyo, Ja-pan）and 2, 5-dihydroxybenzoic acid（DHB）was obtained from Bruker Daltonics（Bremen, Germany）. Ultra pure wa-ter was obtained using a Milli-Q water system（Millipore, Bedford, MA, USA）and was used to prepare all the buffers and solvents. All the chemicals used in this study were of the highest purity available. Slices of bluefin tuna were purchased from a local market in Hamamatsu, Japan. They were stored at －30℃ until their analysis.

2.2 TLCTotal lipids were extracted from two kinds of bluefin

tuna（one was farm-raised in the Republic of Malta, and the other was wild tuna captured from the Sea of Japan）with chloroform/methanol（2/1, v/v）18）. Equal amounts of total lipids dissolved in chloroform/methanol（2/1, v/v）were manually applied as 5-mm wide spots onto HPTLC silica gel 60 plates. The plates were developed with a methylacetate/1-propanol/chloroform/methanol/0.25％ aqueous KCl（25/25/25/10/9, v/v/v/v）solvent system. After developing the samples, the HPTLC plate was sprayed with primuline reagent until it was wet, and then, the plate was air-dried. Lipid bands were visualized under UV light at 366 nm and the visualized bands were marked with a soft red pencil.

2.3 TLC-blottingTLC-blotting was carried out as described previously15, 19）.

In brief, the HPTLC plate was dipped in the blotting sol-vent, namely, 2-propanol/0.2％ aqueous CaCl2/methanol（40/20/7, v/v/v）for 10 s and then immediately placed on a fl at glass plate. Thereafter, a PVDF membrane, followed by a Teflon membrane, and finally a glass fiber filter sheet were placed over the lipid spots on the HPTLC plate. This assembly was pressed evenly for 30 s with a commercially available iron heated to 180℃. The PVDF membrane was then removed from the HPTLC plate and air-dried. Lipids on PVDF membranes were exclusively located on the re-verse side of the TLC plate.

2.4 TLC-Blot-MALDI-IMSTLC-Blot-MALDI-IMS was carried out using a MALDI

time of fl ight（TOF）/TOF-type instrument（Ultrafl ex II TOF/TOF, Bruker Daltonics）equipped with a 355-nm Nd:YAG laser at a repetition rate of 200 Hz. IMS was carried out as described previously20, 21）. In brief, 50 mg/mL DHB in meth-anol/water（1/1, v/v）was used as the matrix. DHB solution（a total of 10-20 μL）was deposited on the lipid spots on the PVDF membrane and lipids were transferred by the heat-iron method from the TLC plates to PVDF membrane. The crystallization process was accelerated under a gentle stream of cold air. To obtain high-quality mass spectra, the PVDF membrane was fi rmly pressed with Kimwipe（Crecia, Tokyo, Japan）to fl atten the membrane surface and elimi-nate the excess matrix crystals. The PVDF membrane was attached to a MALDI sample plate with electrifi ed double-adhesive tape to reduce charges on the plate. Data were acquired in the positive-ion mode（refl ector mode）, and m/z values in the range of 400-1000 were measured. The laser diameter was set to the minimum value. The MS spectra were calibrated externally using DHB（［M＋H］＋, m/z 155.12）and a peptide calibration standard（Bruker Dalton-ics）. All the spectra were acquired automatically using the FlexImaging software（Bruker Daltonics）. Normalization was also performed using this software. The software was also used to create two-dimensional ion-density maps, and peak analyses were performed using FlexAnalysis（Bruker Daltonics）. Lipids were identifi ed by referring to a previous report22）and by tandem MS analysis.

2.5 Tandem mass spectrometryMS/MS was carried out using a MALDI linear quadrupole

ion-trap- type instrument（MALDI LTQ-XL; Thermo Fisher Scientifi c, CA, USA）equipped with a 337 nm N2 laser at a repetition rate of 60 Hz in the positive ion mode. Ionization was carried out with a 337-nm pulsed N2 laser. The precur-sor and fragment ions obtained by collision-induced disso-ciation（CID）were ejected from the ion trap and analyzed. DHB（50 mg/ml）in methanol/water（1/1, v/v）was used as the matrix, and 50 μL DHB solution was deposited onto the lipid spots.

2.6 Statistical analysisStatistical analysis was performed using Stat View 5.0

（SAS Institute, Tokyo, Japan）. The statistical difference in the ratio of lipids was determined by the Student’s t test. Differences were considered signifi cant at P＜0.05.

3 RESULTS AND DISCUSSION3.1 TLC-Blot-MALDI-IMS analysis of lipids extracted from

bluefi n tuna samplesPE, PI, PC, and SM were detected by primuline staining

（Fig. 1a）. Each lipid spot was visualized by TLC-Blot-MALDI-IMS（Fig. 1b）. We focused on the three most in-

Selective Analysis of Lipids by TLC-Blot-MALDI-IMS

J. Oleo Sci. 60, (2) 93-98 (2011)

95

tense peaks for each spots, and performed molecular iden-tification by MS/MS on the separated lipid spot on the PVDF membrane（Fig. 1c-g）. MS/MS was performed using

a MALDI linear quadrupole ion-trap- type instrument be-cause neutral loss of fatty acids could not be detected by MS/MS using a MALDI TOF/TOF-type instrument. Molecu-

Fig. 1　 (a) Representative thin-layer chromatogram of phospholipids extracted from bluefin tuna after primuline staining (n = 3). (b) The ion images of PE, PI, PS, PC, and SM. Numbers in parentheses represent the kind of fatty acid (number of carbon chain: number of double bonds). Merged images of each phospholipid are shown (red: m/z 836.6, 955.6, 806.5, 828.5, and 835.6; aqua: m/z 808.6, 931.6, 902.5, 802.5, and 725.6; and green: m/z 810.6, 929.6, 858.5, and 782.5). (c) MS/MS spectrum of PE (m/z 836.6). A neutral loss of 43 is indicative of the polar head of PE. (d) MS/MS spectrum of PI (m/z 955.6). A neutral loss of 162 is indicative of the polar head of PI. (e) MS/MS spectrum of PS (m/z 806.5). Neutral losses of 87 and 185 are indicative of the polar head of PS. (f) MS/MS spectrum of PC (m/z 828.5). Neutral losses of 59 and 183 are indicative of the polar head of PC. (g) MS/MS spectrum of SM (m/z 835.6). Neutral losses of 59 and 183 are indicative of the polar head of SM.

N. Zaima, N. Goto-Inoue, K. Adachi & M. Setou

J. Oleo Sci. 60, (2) 93-98 (2011)

96

lar peaks at m/z 836.6, 808.6, and 810.6 were detected pre-dominantly in the PE spots. These molecular peaks were designated as PE（18:0/22:6）, PE（18:1/20:5）, and PE（18:0/20:5）by MS/MS（Fig. 1c）. Molecular peaks at m/z 955.6, 931.6, and 929.6 were detected predominantly for the PI spots. These molecular peaks were designated as PI（18:0/22:6）, PI（18:0/20:4）, and PI（18:0/20:5）by MS/MS（Fig. 1d）. In the negative ion mode, these PI species were detected as deprotonated molecular ions at m/z 909.6（PI（18:0/22:6））, m/z 885.6（PI（18:0/20:4））, and m/z 883.6（PI（18:0/20:5））（data not shown）. Previous reports have indi-cated that PI is not detected in the negative ion mode by the non-blotting method, i.e., TLC-MALDI-IMS, when DHB is used as matrix23）. Our data suggested that the blotting process is essential for high-sensitivity detection of the negative ion in lipids.

Although spots of PS were not positively stained with primuline, molecular peaks of PS were detected at m/z 806.5, 902.5, and 858.5 by TLC-Blot-MALDI-IMS analysis. In the MS/MS analysis of the PS spots, neutral losses of fat-ty acids were not detected（Fig. 1e）. Considering the abun-dance of fatty acids in tuna, we designated these peaks as PS（16:0/20:4）, PS（22:6/22:6）, and PS（18:0/22:6）24）. Molec-ular peaks at m/z 828.5, 802.5, and 782.5 were detected predominantly in the case of the PC spots and were as-signed. as PC（16:0/22:6）, PC（16:0/20:5）, and PC（16:0/18:1）by MS/MS of the PC spots（Fig. 1f）. Only two molecular peaks were detected in the case of the SM spots, whereas more than three peaks were detected in the case of the glycerophospholipid spots. The molecular peaks of

SM were detected at m/z 835.6 and 725.6. In MS/MS analy-sis of the SM spots, neutral losses of fatty acids were not detected（Fig. 1g）. We therefore designated these peaks as SM（d18:1/24:1）and SM（d18:1/16:0）according to a previous report25）. The merged images of these phospholipids re-vealed that their developed positions were slightly differ-ent（Fig. 1b right）. These data was consistent with previous reports19）and indicated that phospholipids was developed mainly on the basis of their number of carbon chains. The relative-to-front（RF）values of phospholipids are shown in Table 1.

3.2 Lipid profi les of farmed and wild bluefi n tunaThe lipid profi les of farmed and wild bluefi n tuna were

summarized in Table 1; the lipid profi le of farmed bluefi n tuna was different than that of wild one. The differences between the two samples may refl ect differences in dietary lipid nutrition, because lipid profi les are generally altered by exogenous factors. However, in the case of SM, no sig-nifi cant difference was detected between farmed and wild tuna. Thus, the composition of SM in tuna might be unaf-fected by lipid nutrition. Both in farmed and wild tuna, PC（16:0/18:1）was detected intensely. PC（16:0/18:1）is report-ed to be a physiologically relevant endogenous peroxisome proliferator-activated receptor（PPAR）α ligand6）. PPARα regulates the expression of enzymes related to the β-oxidation of fatty acids. PPARα is thought to have anti-inflammatory26）or anti-hyperlipidemia properties27）. The beneficial effects of dietary tuna might be partly brought about via the PC（16:0/18:1）pathway.

Table 1　Phospholipid profi les determined with TLC-Blot-MALDI-IMS

Selective Analysis of Lipids by TLC-Blot-MALDI-IMS

J. Oleo Sci. 60, (2) 93-98 (2011)

97

4 CONCLUSIONIn this study, we applied IMS for analysis of lipids that

were separated by TLC. We successfully detected differ-ences in major phospholipid components at the molecular species level by TLC-Blot-MALDI-IMS. TLC is widely used in food analysis. This technique widens the applicability of TLC analysis; for example, unknown spots on a TLC plate can be identifi ed by direct MS/MS analysis of the samples. This approach shows great potential as a tool for future nutritional studies and lipidomics research.

ACKNOWLEDGEMENTWe are grateful to Mayumi Suzuki for providing us with

technical assistance. This work was supported by the Pro-gram for Promotion of Basic and Applied Researches for Innovations in Bio-oriented Industry（BRAIN）to N.Z.; a Grant-in-Aid for Young Scientists S（20670004）to M.S.

References1） Michalec, C.; Sulc, M.; Mestan, J. Analysis of cholester-

yl esters and triglycerides by thin-layer chromatogra-phy. Nature 193, 63-64（1962）.

2） Kuge, Y.; Kanda, A.; Hara, S. Pursuit of oxidation be-havior for conjugated polyenoyl glycerols and estab-lishment of their novel oxidation prevention method. J. Oleo Sci. 58, 295-301（2009）.

3） Utsugi, A.; Kanda, A.; Hara, S. Lipase specifi city in the transacylation of triacylglycerin. J. Oleo Sci. 58, 123-132（2009）.

4） Fukasawa, R.; Kanda, A.; Hara, S. Anti-oxidative ef-fects of rooibos tea extract on autoxidation and ther-mal oxidation of lipids. J. Oleo Sci. 58, 275-283（2009）.

5） Skipski, V.P.; Peterson, R.F.; Sanders, J.; Barclay, M. Thin-Layer Chromatography of Phospholipids Using Silica Gel without Calcium Sulfate Binder. J. Lipid Res. 4, 227-228（1963）.

6） Chakravarthy, M.V.; Lodhi, I.J.; Yin, L.; Malapaka, R.R.V.; Xu, H.E.; Turk, J.; Semenkovich, C.F. Identifi -cation of a physiologically relevant endogenous ligand for PPAR alpha in liver. Cell 138, 476-488（2009）.

7） Zaima, N.; Hayasaka, T.; Goto-Inoue, N.; Setou, M. Ma-trix-associated laser clesorption/ ionization imaging mass spectrometry. Int. J. Mol. Sci. 11, 5040-5055（2010）.

8） Cornett, D.S.; Reyzer, M.L.; Chaurand, P.; Caprioli, R.M. MALDI imaging mass spectrometry: Molecular snap-shots of biochemical systems. Nat. Methods 4, 828-833（2007）.

9） Zaima, N.; Hayasaka, T.; Goto-Inoue, N., Setou, M. Im-

aging of metabolites by MALDI mass spectrometry. J. Oleo Sci. 58, 415-419（2009）.

10） Hayasaka, T.; Goto-Inoue, N.; Sugiura, Y.; Zaima, N.; Nakanishi, H.; Ohishi, K.; Nakanishi, S.; Naito, T.; Ta-guchi, R.; Setou, M. Matrix-assisted laser desorption/ionization quadrupole ion trap time-of-fl ight（MALDI-QIT-TOF）-based imaging mass spectrometry reveals a layered distribution of phospholipid molecular species in the mouse retina. Rapid Commun. Mass Spectrom. 22, 3415-3426（2008）.

11） Yao, I.; Sugiura, Y.; Matsumoto, M.; Setou, M. In situ proteomics with imaging mass spectrometry and prin-cipal component analysis in the Scrapper-knockout mouse brain. Proteomics 8, 3692-3701（2008）.

12） Goto-Inoue, N.; Setou, M.; Zaima, N. Visualization of spatial distribution of γ -aminobutyric acid in egg-plant（Solanum melongena）by matrix-assisted laser desorption/ionization imaging mass spectrometry. Anal. Sci. 26, 821-825（2010）.

13） Zaima, N.; Goto-Inoue, N.; Hayasaka, T.; Setou, M. Ap-plication of imaging mass spectrometry for analysis of rice Oryza sativa. Rapid Commun. Mass Spectrom. 24, 2723-2729（2010）.

14） Khatib-Shahidi, S.; Andersson, M.; Herman, J.L.; Gil-lespie, T.A.; Caprioli, R.M. Direct molecular analysis of whole-body animal tissue sections by imaging MALDI mass spectrometry. Anal. Chem. 78, 6448-6456（2006）.

15） Taki, T.; Ishikawa, D. TLC blotting: application to mi-croscale analysis of lipids and as a new approach to lipid-protein interaction. Anal. Biochem. 251, 135-143（1997）.

16） Guittard, J.; Hronowski, X.L.; Costello, C.E. Direct ma-trix-assisted laser desorption/ionization mass spectro-metric analysis of glycosphingolipids on thin layer chromatographic plates and transfer membranes. Rap-id Commun. Mass Spectrom. 13, 1838-1849（1999）.

17） Goto-Inoue, N.; Hayasaka, T.; Taki, T.; Gonzalez, T.V.; Setou, M. A new lipidomics approach by thin-layer chromatography-blot-matrix-assisted laser desorption/ionization imaging mass spectrometry for analyzing detailed patterns of phospholipid molecular species. J. Chromatogr. A. 1216, 7096-7101（2009）.

18） Zaima, N.; Sugawara, T.; Goto, D.; Hirata, T. Trans geo-metric isomers of EPA decrease LXRalpha-induced cellular triacylglycerol via suppression of SREBP-1c and PGC-1beta. J. Lipid Res. 47, 2712-2717（2006）.

19） Goto-Inoue, N.; Hayasaka, T.; Sugiura, Y.; Taki, T.; Li, Y.T.; Matsumoto, M.; Setou, M. High-sensitivity analy-sis of glycosphingolipids by matrix-assisted laser de-sorption/ionization quadrupole ion trap time-of-flight imaging mass spectrometry on transfer membranes. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 870, 74-83（2008）.

N. Zaima, N. Goto-Inoue, K. Adachi & M. Setou

J. Oleo Sci. 60, (2) 93-98 (2011)

98

20） Zaima, N.; Matsuyama, Y.; Setou, M. Principal compo-nent analysis of direct matrix-assisted laser desorp-tion/ionization mass spectrometric data related to me-tabolites of fatty liver. J. Oleo Sci. 58, 267-273（2009）.

21） Goto-Inoue, N.; Hayasaka, T.; Zaima, N.; Setou, M. The specific localization of seminolipid molecular species on mouse testis during testicular maturation revealed by imaging mass spectrometry. Glycobiology 19, 950-957（2009）.

22） Sugiura, Y.; Konishi, Y.; Zaima, N.; Kajihara, S.; Nakani-shi, H.; Taguchi, R.; Setou, M. Visualization of the cell-selective distribution of PUFA-containing phosphati-dylchol ines in mouse brain by imaging mass spectrometry. J. Lipid Res. 50, 1776-1788（2009）.

23） Fuchs, B.; Schiller, J.; Suss, R.; Zscharnack, M.; Bader, A.; Muller, P.; Schurenberg, M.; Becker, M.; Suckau, D. Analysis of stem cell lipids by offl ine HPTLC-MALDI-TOF MS. Anal. Bioanal. Chem. 392, 849-860（2008）.

24） Saito, H.; Ishihara, K.; Murase, T. Effect of prey fish lipids on the docosahexaenoic acid content of total fatty acids in the lipid of Thunnus albacares yellowfi n tuna. Biosci. Biotechnol. Biochem. 60, 962-965（1996）.

25） Stubiger, G.; Pittenauer, E.; Belgacem, O.; Rehulka, P.; Widhalm, K.; Allmaier, G. Analysis of human plasma lipids and soybean lecithin by means of high-perfor-mance thin-layer chromatography and matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom 23, 2711-2723（2009）.

26） Straus, D.S.; Glass, C.K. Anti-infl ammatory actions of PPAR ligands: new insights on cellular and molecular mechanisms. Trends. Immunol. 28, 551-558（2007）.

27） Barter, P.J.; Rye, K.A. Is there a role for fi brates in the management of dyslipidemia in the metabolic syn-drome? Arterioscler Thromb. Vasc. Biol. 28, 39-46（2008）.