Determination of sphingoid bases from hydrolyzed glucosylceramide ...

Journal of Oleo Science Copyright ©2012 by Japan Oil Chemists’ Society J. Oleo Sci. 61, (12) 681-688 (2012)

Determination of sphingoid bases from hydrolyzed glucosylceramide in rice and wheat by online postcolumn high-performance liquid chromatography with O-phthalaldehyde derivatization Hirofumi Goto1,2＊ , Keiko Nishikawa1, Noriko Shionoya1, Makoto Taniguchi2 and Tomoji Igarashi1 1 2

Japan Food Research Laboratories, Nagoya Branch, 4-5-13 Osu, Naka-ku, Nagoya, Aichi 460-0011, JAPAN Mimasaka University 50 Kitazono-cho, Tsuyama, Okayama 708-8511, JAPAN

Abstract: We developed a determination method for sphingoid bases using online post-column highperformance liquid chromatography (HPLC) with O-phthalaldehyde (OPA) derivatization. Good separation was achieved using a reversed-phase column and eluting with 50% acetonitrile containing formic acid and heptafluorobutyric acid. Using these conditions, an excellent linearity (R2 > 0.999) was achieved using standard solutions of sphinganine (d18:0), sphingosine (d18:1 4t), 4-hydroxy-sphinganine (t18:0), glucosylsphingosine (glc-d18:14t), and galactosylsphingosine (gal-d18:14t). Plant glucosylceramides were hydrolyzed with 1 M aqueous HCl in methanol for 18 h at 90℃, followed by extraction of sphingoid bases with diethyl ether in preparation for analysis using the proposed HPLC conditions. The glc-d18:14t standard was also hydrolyzed and analyzed by HPLC using the same procedure, and the d18:14t peak obtained from the hydrolyzed glc-d18:1 4t standard was used as a reference for calculation. We also confirmed the applicability of this method to the analysis of sphingoid bases in rice and wheat, obtaining relative standard deviations of 8.0% for rice and 4.6% for wheat. The recoveries of spiked rice and wheat samples were 104% and 106%, respectively. Our proposed method enables the straightforward determination of sphingoid bases without expensive facilities, employing fluorescence detection of OPA derivatives. Key words: sphingoid base, glucosylceramide, post-column HPLC, O-phthalaldehyde derivatization 1 INTRODUCTION Sphingolipids are ubiquitous in eukaryotic organisms. Ceramide, a type of sphingolipid, is well-known to support the barrier function of the skin. In addition, ceramides of animal origin and their metabolites play important roles as intracellular mediators of cell differentiation and apoptosis1, 2）. Plant glucosylceramides have also been found to induce apoptosis, and dietary plant glucosylceramides have potent physiological functions similar to those of animal ceramides3）. For instance, it has been reported that cheek trans-epidermal water loss was significantly lowered by oral intake of glucosylceramide4）. Therefore, plant glucosylceramide has attracted attention as a functional food material. Plant glycosylceramides comprise a sphingoid base backbone, an amide-linked fatty acid, and a polar head-

group such as hexose（Fig. 1）. Major constituents in rice and maize are 2-hydroxyfatty acid with 20 carbons（20h:0） and 2-amino-4t,8c-octadecadiene-1,3-diol（d18:2 4t,8c）, a sphingoid base; in wheat, 16h:0 or 20h:0 fatty acid and d18:18c sphingoid base; and in konjac, 18h:0 fatty acid and

Fig. 1　The representative structure of glucosylceramides in plants. D-Glucosyl-N-hydroxy-eicosanoyl-sphingadienine (glc-20h:0-d18:24t,8c).

＊

Correspondence to: Hirofumi Goto, Japan Food Research Laboratories, Nagoya Branch 4-5-13 Osu, Naka-ku, Nagoya, Aichi 4600011, JAPAN E-mail: [email protected] Accepted July 9, 2012 (recieved for review May 22, 2012)

Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 online

http://www.jstage.jst.go.jp/browse/jos/ http://mc.manusriptcentral.com/jjocs 681

H. Goto, K. Nishikawa, N. Shionoya et al.

d18:24t,8c sphingoid base5）. The sugar component is glucose in most plant glycosylceramides, while, in contrast, the sphingoid bases and fatty acid moieties are diverse, lending great complexity to the mixture of plant glycosylceramides. Several chromatographic analysis methods for glucosylceramide have been reported, such as high-performance liquid chromatography（HPLC）with evaporative light scat6−8） as well as HPLC in conjunctering detection （LC/ELSD） 6, 9, 10） and with tandem tion with mass spectrometry （LC/MS） 11, 12） mass spectrometry（LC/MS/MS） . The reported LC/MS/ MS method achieved high selectivity and sensitivity for the identification and quantification of glucosylceramides, making it a powerful tool for the analysis of complex sphingolipids in foods. However, it has not come into routine use in analytical laboratories because the instrument is quite expensive. A highly sensitive pre-column HPLC with Ophthalaldehyde（OPA）derivatization method for sphingoid bases has also been reported13, 14）, but since OPA derivatives are unstable, the time from derivatization to HPLC analysis must be strictly controlled. In order to resolve some of these issues, we developed a determination method for sphingoid bases using online post-column HPLC with OPA derivatization. We confirmed the applicability of this method to the analysis of the sphingoid bases obtained from aqueous HCl hydrolysis of glucosylceramide in rice and wheat, which were selected as representative foods containing glucosylceramide. The proposed method enables the straightforward determination of sphingoid bases without expensive equipment, employing fluorescence detection for OPA derivatives. Our results demonstrate that this method is applicable to the analysis of sphingolipid in rice and wheat.

2 EXPERIMENTAL 2.1 Materials 2.1.1 Reagents Sphingosine（2-amino-4t-octadecene-1,3-diol; d18:1 4t） was purchased from Wako Pure Chemical（Osaka, Japan）. Sphinganine（2-amino-octadecane-1,3-diol; d18:0）, gluco, and galactosylsphingosine （galsylsphingosine （glc-d18:14t） 4t （Alad18:1 ）were purchased from Avanti Polar Lipids, Inc. baster, AL）. 4-hydroxy-sphinganine（2-amino-octadecane1,3,4-triol; t18:0）was purchased from Enzo Life Sciences Inc.（Farmingdale, NY）. Glucosylceramide mixes from rice and wheat were purchased from Nagara Science Co., Ltd. （Gifu, Japan）. Heptafluorobutyric acid （0.5 M in water, ionpair reagent for LC-MS）was purchased from Tokyo Chemical Industry Co., Ltd. （Tokyo, Japan） . HPLC-grade acetonitrile was purchased from Wako Pure Chemical. Other reagents used for the study were of analytical grade. 2.1.2 OPA reagent for derivatization OPA（0.4 g）dissolved in 7 mL of ethanol was mixed with

1 mL of 2-mercaptoethanol and 493 mL of borate buffer. The borate buffer was prepared by dissolving 40.7 g of sodium carbonate, 18.8 g of potassium sulfate, and 13.4 g of boric acid in 1 L of water. 2.1.3 1 M aqueous HCl in methanol for hydrolysis Aqueous HCl was prepared for hydrolysis by combining 8.6 mL of HCl and 9.4 mL of water, then diluting to 100 mL with methanol. 2.1.4 Food samples Rice（rice powder）and wheat（weak flour）samples were purchased from supermarkets in Aichi, Japan, in 2011. 2.2 Methods 2.2.1 Online post-column HPLC with OPA derivatization The HPLC system （Shimadzu Corporation, Kyoto, Japan） was equipped with a pump for the mobile phase（ LC20AB）, a fluorescence detector（RF-10AXL）, an autosampler （SIL-20A） , a column oven （LTO-20AC） , and a pump for post-column reactions（LC-10ADVP） . The detector was set at excitation and emission wavelengths of 348 and 450 nm, respectively. Chromatographic separation was carried out using the following column and mobile phases: ODS JM-302 （4.6×150 mm; YMC Co., Ltd., Kyoto, Japan）with 50％ acetonitrile containing 0.1％ formic acid and 5 mM heptafluorobutyric acid. The column temperature and flow-rate were 35℃ and 1.0 mL/min, respectively. Post-column derivatization was carried out in a stainless-steel tube（0.5 mm i.d.×4 m）at 35℃. The flow-rate of the OPA reagent was 0.5 mL/min. 2.2.2 LC/MS conditions for peak identification Mass spectral measurements for peak identification were performed using a Quattro Premier XE（Waters, Milford, MA）equipped with an atmospheric pressure ionization source and an electrospray ionization interface. It was operated in positive mode with an electrospray voltage of 4.0 kV, a cone voltage of 80 V, and source temperature of 120℃. Liquid chromatography was performed on a Waters 2695 system（Waters, Milford, MA）with same condition as described in Section 2.2.1 without post-column derivatization. 2.2.3 Hydrolysis of sphingolipids and extraction of sphingoid bases Sphingolipids were hydrolyzed using the method reported by Gaver and Sweeley15）. Sphingolipids were hydrolyzed in 1 mL of 1 M aqueous HCl in methanol for 18 h at 90℃. The reaction mixture was washed twice with 2 mL of hexane, and the pH was adjusted to 10-12 with 10 M NaOH. After adding 1.5 mL of water, the sphingoid bases were extracted twice with 2 mL of diethyl ether. The diethyl ether solution was washed with 2 mL of water and dried under nitrogen flow. The residue was re-dissolved in 1 mL of methanol, and 10 μL of the solution was used for HPLC analysis.

682

J. Oleo Sci. 61, (12) 681-688 (2012)

Determination of sphingoid bases by OPA post-column HPLC

2.2.4 Extraction of lipids from food samples Total lipids were extracted using the method proposed by Folch et al.16） from rice powder or flour. Briefly, 0.5-1 g of sample was homogenized with 150 mL of 2:1 v/v chloroform-methanol and centrifuged at 3,000 rpm for 10 min. Extraction from the precipitate was repeated twice, with 100 mL and 20 mL of 2:1 v/v chloroform-methanol, using the same procedure. All supernatants were combined and washed with 93 mL of 0.88％ KCl solution. The chloroformmethanol layer was evaporated to a dry state at 40℃. 2.2.5 Determination of sphingoid bases in rice and wheat The glc-d18:14t standard was hydrolyzed with aqueous HCl in methanol, and the sphingoid base was extracted according to the procedure described in Section 2.2.3. The peak area of d18:14t from the hydrolyzed glc-d18:14t obtained by HPLC analysis was used as a reference for calculation. The areas of the sphingoid base peaks obtained from hydrolysis of plant glucosylceramides were calculated by comparison with the d18:14t reference peak from the hydrolyzed glc-d18:14t. To confirm the applicability of this method, the linearity of the calibration curve, limit of quantification（LOQ）, accuracy, and precision were determined for the analysis of sphingoid bases. The calibration curve was constructed from measurements of six hydrolyzed glc-d18:1 4t standard solutions ranging from 0.22 to 2.2 nmol. The LOQ was estimated to be ten times the signal-to-noise （S/N） ratio of the 0.22 nmol standard solution. To confirm the accuracy of the method, a recovery test of the glucosylceramide standard was carried out using rice and wheat. Glucosylceramide standard of the same origin as the test sample was spiked into

Fig. 2　Chromatograms of the sphingoid base standards. 100 ng of each sphingoid base and glycosylsphingoid base standard was injected into HPLC. The sphingoid bases were abbreviated as follows; t18:0: 4-hydroxy-sphinganine; d18:14t: sphingosine; d18:0: sphinganine; glc-d18:1 4t: glucosylsphingosine; gal-d18:1 4t: galactosylsphingosine. the sample at 80 μg per 0.5 g sample. The molar amount of sphingoid bases was converted to the mass of glucosyl-Nhydroxy-eicosanoyl-sphingadienine（ glc-20h:0-d18:2）, which is a typical plant glucosylceramide. Regardless of the origin, the molar amount of the spiked standard was calculated as glc-20h:0-d18:2. After hydrolysis and HPLC analy-

Fig. 3　Effects of the coil length and reaction temperature on hydrolysis. The sphingoid bases were abbreviated as follows; t18:0: 4-hydroxysphinganine; d18:14t: sphingosine; d18:0: sphinganine; glc-d18:14t: glucosylsphingosine; gal-d18:14t: galactosylsphingosine. The y-axis is shown as the relative peak area ratio against the peak area of d18:14t. 683

J. Oleo Sci. 61, (12) 681-688 (2012)


sis, the recovery was calculated by subtracting the original molar amount in the food sample from the analyzed amount. To confirm the precision, the sphingoid bases in rice and wheat were determined seven times, and the relative standard deviation （RSDr） was calculated.

3 RESULTS AND DISCUSSION 3.1 HPLC conditions HPLC conditions were optimized as follows. Using a reversed-phase HPLC column eluted with 50％ acetonitrile containing 0.1％ formic acid and 5 mM heptafluorobutyric acid, good separation between sphingoid bases with and without the sugar moiety was achieved （Fig. 2） . For detection of the sphingoid base peaks, online OPA post-column derivatization was optimized. As shown in Fig. 3, the coil length and temperature for the reaction were found to affect the relative sensitivity of each peak. With a 4 m coil maintained at a reaction temperature of 35℃, the relative sensitivity ratio of d18:0 and t18:0 against d18:14t was within 5％ and that for gal-d18:14t against glc-d18:14t was within 10％.

Fig. 5　R ecoveries of sphingoid bases and glucosylsphingoid bases from hydrolyzed glucosylceramide from rice.

3.2 Linearity of sphingoid base detection As shown in Fig. 4, a good linearity（R2＞0.999）was observed between sphingoid base standard solutions from 10 to 500 ng and the peak areas for three types of sphingoid bases（d18:0, d18:14t, t18:0）, and two types of glycosyl. sphingoid bases （glc-d18:14t, gal-d18:14t）

Fig. 4　Standard curves of sphingoid bases and glycosylsphingoid bases. The sphingoid bases were abbreviated as follows; t18:0: 4-hydroxy-sphinganine; d18:1 4t: sphingosine; d18:0: sphinganine; glc-d18:1 4t: glucosylsphingosine; gal-d18:1 4t: galactosylsphingosine.

Fig. 6　Chromatograms of the hydrolyzed glucosylceramide standard from rice. The upper chromatogram is the standard of the t18:0, d18:14t, d18:0, and glc-d18:14t. The middle chromatogram is hydrolyzed at 70℃ for 18 h, and the lower chromatogram is hydrolyzed at 90℃ for 18 h. 50 μg of the glucosylceramide standard from rice was hydrolyzed. Peaks were identified by LC/MS (Fig. 7) and relative retention time as follows; glc-t18:1 (Peak 1), glcd18:2 (Peak 2, 3), glc-d18:1 (Peak 4), t18:1 (Peak 5, 6), d18:2 (Peak 7, 8), O-methyl-d18:2 (Peak 9, 10).

684

J. Oleo Sci. 61, (12) 681-688 (2012)


3.3 Hydrolysis conditions The time and temperature of hydrolysis were optimized. The standard glucosylceramide mixture from rice was hydrolyzed using aqueous HCl to quantitatively obtain sphingoid bases. As shown in Fig. 5, the hydrolysis depended on the reaction time and temperature. When glucosylceramide was hydrolyzed at 90℃ for 18 h, only ＜2％ of glucosylsphingoid base was detected, but some sphingoid bases without glucoside were detected（Fig. 6）. Sphingoid base peaks were identified by LC/MS（Fig. 7）and their relative retention times. At 70 and 90℃, a plateau of approximately 60％ recovery was observed. When the standard glc-d18:14t was hydrolyzed, the recovery of d18:14t was also approximately 60％. Using the optimal hydrolysis conditions, most fatty acid and sugar moieties were released. O-methyld18:2 was observed as a by-product （Fig. 6） . It was reported that fewer by-products were generated in hydrolysis with aqueous HCl compared with other hydrolysis methods17）. For quantification of sphingoid bases from hydrolyzed glucosylceramide, the standard glc-d18:14t was hydrolyzed and the obtained d18:14t was used as the reference. We compensated for the plateau of recovery using this calculation method. 3.4 Linearity and limit of quantification As shown in Fig. 8, a good linearity（R2＞0.999）of the standard curve was observed between the molar amount of hydrolyzed glc-d18:14t injected ranging from 0.02 to 2.2 nmol and the peak areas of d18:14t. The LOQ was estimated to be 0.004 nmol from the S/N ratio of the chromatogram.

Fig. 8　Standard curve of the hydrolyzed glc-d18:14t The glc-d18:1 4t standard was hydrolyzed according to “Methods” section, and the obtained d18:14t was analyzed. 3.5 Determination of sphingoid bases in rice and wheat Chromatograms of rice and wheat samples are shown in Fig. 9. Each peak was identified by LC/MS（Fig. 10）and its relative retention time. In the rice sample, t18:1, d18:2, and t18:0 were detected, and the trihydroxy sphingoid bases, t18:1 and t18:0, were the most abundant components. In the wheat sample, t18:1, d18:2, t18:0, d18:1, and d18:0 were detected, and d18:1 was a main component. In both rice and wheat, unsaturated sphingoid bases were detected as plural peaks because they were present as geometrical and/or positional isomers. The by-product peak observed in the analysis of glucosylceramide standard from rice was

Fig. 7　Mass spectra for peak identification of the hydrolyzed glucosylceramide standard from rice. A peak number corresponds to its Fig. 6. 685

J. Oleo Sci. 61, (12) 681-688 (2012)


Fig. 9　Chromatograms of the sphingoid bases from hydrolyzed rice and wheat samples. The upper chromatogram is the standard of the t18:0, d18:14t, d18:0, and glc-d18:14t. The middle chromatogram is the hydrolyzed rice extract, and the lower is the hydrolyzed wheat extract. Peaks were identified by LC/MS (Fig. 10) and relative retention time as follows; t18:1 (Peak 1, 2, 6, 7), d18:2 (Peak 3, 4, 8, 9), t18:0 (Peak 5, 10), d18:1 (Peak 11, 12), and d18:0 (Peak 13). not detected in either rice or wheat. In this study, it was assumed that the fluorescence intensity against the molar amount of each sphingoid base peak was as same as for d18:14t. Average recoveries of the standard of the same origin were satisfactory, as exemplified by the values of 104％ in rice and 106％ in wheat. The precision as assessed by the RSDr of repeated tests was 8.0％ in rice and 4.6％ in wheat. The amount of glucosylceramide calculated from the detected sphingoid bases in rice and wheat were estimated to be 13 μmol/100g（10 mg/100 g） and 53 μmol/100g （41 mg/100 g） as glc-20h:0-d18:2, respectively. On the basis of these results, we believe that this method can be applicable to the determination of sphingoid base content in rice and wheat.

4 CONCLUSION We developed a determination method using online postcolumn HPLC with OPA derivatization for sphingoid bases obtained from glucosylceramide by aqueous HCl hydrolysis. Using the optimized HPLC conditions, good separation and linearity were achieved. We also confirmed the applicability of this method to sphingoid base determination in rice and wheat. In this study, glucosylceramide and ceramide which does not have sugar moieties were not separated before analysis. Therefore, if exact quantification of only the sphingoid bases from glucosylceramides is required, it would be necessary to separate glucosylceramides with column chromatography, etc., before hydrolysis. The proposed method is simple, inexpensive, and stable

686

J. Oleo Sci. 61, (12) 681-688 (2012)


Fig. 10　Mass spectra for peak identification of the sphingoid bases from rice and wheat. A peak number corresponds to its Fig. 9. with fluorescence detection of OPA derivatives, making it a unique and powerful technique for sphingoid base analysis in foods.

References 1）Hannun, Y. A. Functions of ceramide in coordination cellular responses to stress. Science 274, 1855-1859 （1996） . 2）Vesper, H., Schmelz, E. M., Nikolova-Karakashian, M. N., dillehay, D. L., Lynch, D. V., and Merrill, A. H. Jr. Sphingolipids in food and the emerging importance of

sphingolipids to nutrition. J. Nutr. 129, 1239-1250 （1999）. 3）Aida, K., Takakuwa, N., Minoshita, M., Sugawara, T., Imai, H., Ono, J., and Ohnishi, M. Properties and physiological effects on plant cerebroside species as functional lipids. in Advanced Research on Plant Lipids, Murata, N. ed., Kluwer Academic Publishers. Netherlands, pp.233-236（2003）. 4）Uchiyama, T., Nakano, Y., Ueda, O., Mori, H., Nakashima, M., Noda, A., Ishizaki, C., and Mizoguchi, M. Oral intake of glucosylceramide improves relatively higher level of transepidermal water loss in mice and healthy human subjects. Journal of Health Science, 54（5）, 687

J. Oleo Sci. 61, (12) 681-688 (2012)


559-566 （2008） . 5）Aida, K. Food functionality of plant glucosylceramide -Prevention of aberrant crypt foci formation by dietary plant and fungal glucosylceramides in 1,2-dimethylhydrazine-treated mice-. J. Oleo Sci. 7（ 4）, 141-149 （2007）（in Japanese） . 6）Hany Farwanah, Peter Nuhn, Reinhard Neubert, and Klaus Raith Normal-phase liquid chromatographic separation of stratum corneum ceramides with detection by evaporative light scattering and atmospehric pressure chemical ionization mass spectrometry. Analytica Chimica Acta. 492, 233-239 （2003） . 7）Kashima, M., Nakagawa, K., Sugawara, T., Miyazawa, T., Murakami, C., Miyashita, R., Ono, J., Deschamps, F. S., and Chaminade, P. Method for quantitative determination of cerebroside in“plant ceramide”foodstuffs by high performance liquid chromatography with evaporative light scattering detection. J. Oleo Sci. 51（5）, 347-354 （2002） . 8）Sugawara, T. and Miyazawa, T. Separation and determination of glycolipids from edible plant sources by high-performance liquid chromatography and evaporative light-scattering detection. Lipids 34, 1231-1237 （1999）. 9）Sugawara, T. Aida, K., Duan, J., and Hirata, T. Analysis of glucosylceramides from various sources by liquid chromatography-ion trap mass spectrometry. J. Oleo Sci. 59 （7）, 387-394 （2010） . 10）Shaner, R. L. Quantitative analysis of sphingolipids for lipidomics using triple quadrupole and quadrupole linear ion trap mass spectrometers. Lipid Res. 5, 16921707 （2009） . 11）Nana Bartke, Anne Fischbeck, and Hans-Ulrich Humpf

Analysis of sphingolipids in potates（ Solanum tuberosum L.）and sweet potates（Ipomoea batatas（L.） Lam.）by reversed phase high-performance liquid chromtography electrospry ionization tandem mass spectrometry（ HPLC-ESI-MS/MS）. Mol. Nutr. Food Res. 50, 1201-1211（2006）. 12）Jonathan E. Markham, and Jan G. Jaworski Rapid measurement of sphingolipids from Arabidopsis thaliana by reversed-phase high-performance liquid chromatography coupled to electrospry inoization tandem mass spectrometry. Rapid Communications in Mass Spectrometry 21, 1304-1314（2007）. 13）Jonathan E Markham, Jia Li, Edgar B. Cahoon, and Jan G. Jaworski Separation and identification of major plant sphingolipid classes from leaves. J. Biol. Chem. 281 （32）, 22684-22694（2006）. 14）Zama, K., Hayashi, Y., Ito, S., Hirabayashi, Y., Inoue, T., Ohno, K., Okino, N., and Ito, M. Simultaneous quantification of glucosylceramide and galactosylceramide by normal-phase HPLC using O-phtalaldehyde derivatives prepared with sphingolipid ceramide N-deacylase. Glycobiology, 19（7） , 767-775（2009）. 15）R. C. Gaver and C. C. Sweeley Method for methanolysis of sphingolipids and direct determination of longchain bases by gas chromatography. J. Am. Oil Chemists’Soc. 42, 294-298（1965）. 16）Folch, J., Lees, M., and Sloane Stanley, G. H. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226, 497-509 （1957）. 17）The Japan Biochemical Society, in Seikagaku Jikken Kouza 3 Shisitsu no Kagaku, Tokyo Kagaku Dojin, pp. 222-225（1974）（in Japanese）.

688

J. Oleo Sci. 61, (12) 681-688 (2012)