© 2018 FCCA (Forum: Carbohydrates Coming of Age)E47

Trends in Glycoscience and GlycotechnologyVol. 30 No. 174 (May 2018) pp. E47–E53

Function and Structure Analysis of Glycolipid Microdomains

Kazuya KabayamaGraduate School of Science, Osaka University, Machikaneyama 1–1, Toyonaka, Osaka 560–0043, Japan

FAX: +81–6–6850–5419, E-mail: kaba@chem.sci.osaka-u.ac.jp

(Received on December 26, 2017, accepted on February 27, 2018)

Key Words: lipid rafts, fluorescence microscopy, live cell imaging, detergents, mass spectrometry

AbstractGlycosphingolipids are not uniformly present in cell membranes and form microdomains called glycolipid microdomains to-

gether with sphingomyelin and cholesterol. Many signaling molecules accumulate there, and glycolipid composition greatly influ-ences signal transmission efficiency. Therefore, analyzing the dynamics and structure of glycolipid microdomains is very important in understanding life phenomena. In recent glycolipid microdomain studies, along with biochemical methods, analysis of membrane molecular dynamics by fluorescence microscopy, localization analysis using an electron microscope, and lipid structure analysis by mass spectrometry have been conducted. In this review, I mainly introduce the glycolipid microdomain analysis method using fluo-rescence microscopy and mass spectrometry.

A. IntroductionPrevious studies have demonstrated that membrane hetero-

geneity plays an important role in biological membranes, and the concept of lipid rafts is fundamental to this heterogeneity. How-ever, in the field of lipid raft research, analytical methods are at a standstill, and physical methods of studying the biological mem-brane as a liquid layer dominated by the spatiotemporal space are still under development. This creates a large void between biologi-cal and physicochemical techniques in analyzing actual cells. Lipid rafts accumulate a lot of glycosphingolipids, so lipid rafts are spe-cifically referred to as glycolipid microdomains in this review. On the other hand, there are also sterol non-containing microdomain (1), or saturated fatty acid-containing glyceroglycolipids exist as a glycolipid microdomains (2, 3). Gangliosides, which are known as a constituent of glycolipid microdomains, is a generic term for the glycosphingolipid family including sialic acid, and GM3 is the first molecule in its biosynthetic pathway. Since GM3 is widely ex-pressed in various cells of mammals including humans, physiologi-cal functions and pathophysiological significance are becoming of interest. Gangliosides are expressed on the cell membrane surface and are involved in adhesion and infiltration of cancer cells, recep-tor activity, growth factor-like activity and intercellular recogni-tion. There are many reports of their involvement in relation to the malignancy of cancer. As one of the mechanisms of these actions, accumulation or dissociation of membrane receptor molecules into the glycolipid microdomains is proposed. However, at present, the analytical method for glycolipid microdomains is limited to a technique of evaluating their heterogeneous localization on the cell membrane by whether it is insoluble or soluble in the detergent as mainly determined with sucrose density gradient centrifugation. No new procedure have been developed beyond this. Furthermore,

nonionic detergents such as the commonly used Triton X-100 have the property of reducing lipid–protein interactions. Evaluating the accumulation or dissociation of membrane receptor molecules in glycolipid microdomains becomes difficult as electrostatic interac-tions must be taken into consideration. Therefore, we intended to elucidate the function and structure of the glycolipid microdomain with a methods to eliminate detergents thoroughly, or without us-ing detergents. For the former, we considered introducing molecu-lar dynamics analysis using a fluorescence microscope. In recent years, there has been remarkable development of visualization technology using fluorescent labeled molecules, and molecular dynamics analysis is gaining attention not only for its use in basic research but also for clinical applications. For the latter method, a convenient separation technique for detergent removal using an organic solvent, which is also useful for pretreatment in mass spec-trometry, has been developed. Each method will be outlined from the viewpoint of its utility to evaluate glycolipid microdomains.

B. Difficulties of Glycolipid Microdomain AnalysisAt present, several problems are encountered in analyz-

ing the function and structure of glycolipid microdomains. First, membrane lipids are not fixed in sample treatment for analysis by general electron microscopes. Nevertheless, immunoelectron microscopy of glycolipids has been carried out using samples im-mobilized with glutaraldehyde. There are many reports to discuss that accumulation of colloidal gold due to aggregation properties of anti-glycolipid antibodies (mainly IgM antibodies) is a lipid raft structure. To resolve this issue, Fujimoto et al. have adopted a method of completely fixing lipids by a rapid freezing method (4). Further, Kusumi et al. used total internal reflection fluorescence microscopy to analyze the dynamics of lipid and lipid-anchored

MINIREVIEWdoi: 10.4052/tigg.1413.2E(JSCR AWARDEE REPORT edited by JSCR)

© 2018 FCCA (Forum: Carbohydrates Coming of Age) E48

proteins in detail; they demonstrated that membrane lipids cannot be immobilized, and called for reevaluating of marking data with conventional antibodies (5). Secondly, it is difficult to comprehen-sively interpret glycolipid microdomains only via fractionation of lipid rafts by density gradient centrifugation. In other words, because the molecules accumulated depend on the concentration and type of detergent, it is not possible to obtain a unified view (6). Regarding this issue, many discussions have taken place so far, but no conclusions have been drawn. As a potential solution to the problem, in addition to biochemical data, we used live cell imaging with a fluorescence microscope to elucidate the actual glycolipid microdomains. Thirdly, the detergent in the membrane microdo-main fraction interferes with Time-of-Flight Mass Spectrometer (TOF-MS) analysis. To address the issue, this article describes the use of a triple quadrupole mass spectrometer (Triple-Q TOF-MS) that is recently becoming widely operated, and the separation of glycolipids from the detergent using dichloroethane.

C. Molecular Dynamics Analysis Using Fluorescence Microscopy

Currently, the optical microscope is shifting to coexistence of “observing” and “measuring” capabilities. That is, it is used as an analytical instrument for visualizing and analyzing the expression, dynamics, and functions of molecules. Optical-based microscopy technologies have also taken an important position as an analytical technique in recent years. As an example, various microscopy tech-nologies have been awarded at the annual “Method of the Year” (7) which the publication Nature Methods started in 2007 to honor

analytical techniques. Selected technologies include super-resolu-tion fluorescence microscopy in 2008, optogenetics in 2010, light-sheet fluorescence microscopy in 2014, and single particle cryo-electron microscopy in 2015, indicating that microscopy methods are useful and indispensable. In particular, the inventors of the super-resolution microscope and the cryo-electron microscope gar-nered the Nobel Prize Award several years later, respectively. We believe that microscopy techniques are a very useful tool for eluci-dating cell biological phenomena and have previously reported on lipid rafts using confocal laser scanning microscopes and electron microscopes (8, 9). In the following sections, we describe the func-tion and structure analysis method of glycolipid microdomains us-ing microscopy and mass spectrometry technologies.

D. Glycolipid Microdomain Analysis Using Fluores-cence Recovery after Photobleaching

Fluorescence recovery after photobleaching (FRAP) is used as a molecular dynamics analysis technique using fluorescence mi-croscopy, especially for measuring fluidity of cell membrane mol-ecules (10, 11). In FRAP, it is necessary to apply a fluorescent label to the target molecule. In a living cell in which the labeled mol-ecule is expressed, the region to be analyzed is briefly irradiated with a strong laser to fade only the fluorescence without destroying the function of the molecule. If the target molecule is immobile, the fluorescence in the faded region will not be recovered. If the target molecule is mobile, it diffuses and mixes with surrounding non-faded molecules, and the fluorescence in the faded region re-covers. By plotting the time course of fluorescence recovery rate

Fig. 1. Graph showing time course of fluorescence recovery rate and outline of fluorescence recovery method after photobleaching (FRAP):The dynamic component ratio of the target molecule can be obtained from the terminal point of the recovery curve (predictable by nonlinear curve fitting).It is also possible to calculate the diffusion rate from the rising speed of the curve (A). Example of FRAP: GFP fusion insulin receptor (IR) (IR-GFP) and caveolin 1 (Cav1) (Cav1-GFP) were expressed in HeLa cells, and the membrane surface was observed time-sequentially by confocal microscopy (Top and bottom images, respectively). ROI (measurement area) was bleached 16 s after the start of measurement, and images after 1 min and 3 min were acquired.Since IR diffuses on the cell membrane, the fluorescence intensity of the ROI recovers with time. On the other hand, since Cav1 forms a caveolae structure and is immobilized, the fluorescence luminance does not recover (B).

© 2018 FCCA (Forum: Carbohydrates Coming of Age)E49

on a graph, it is possible to calculate the dynamic component ratio and the diffusion rate of the target molecule. An example of FRAP is shown in Fig. 1. With this method, changes in membrane fluid-ity due to changes in composition of constituents of glycolipid microdomains are analyzed. By combining this information with results obtained from the biochemically isolated, detergent-resis-tant microdomain (DRM), it is possible to analyze and compare the dynamics of lipid raft environments on the actual cell membrane. Simons et al., who originally proposed lipid rafts, have performed FRAP analysis of various membrane proteins expressed on apical membranes of MDCK cells to demonstrate the existence of do-mains that phase separate in biological membranes (12). By using this method, we also succeeded in analyzing the electrostatic inter-action between acidic glycolipid and membrane receptor proteins accumulating in the glycolipid microdomain (8).

E. Glycolipid Microdomain Analysis Using Fluores-cence Correlation Spectroscopy

Fluorescence Correlation Spectroscopy (FCS) is a method that can measure the diffusion coefficient, molecular weight, and even the size of target molecule by replacing fluctuation of pho-tons emitted by fluorescent molecules passing through the focal region of the microscope with a correlation function (13). In rela-tion to carbohydrate substances, Mizuno and Inazu et al. reported the evaluation of binding between fluorescence-labeled oligosac-charides and lectin by FCS in an in vitro system (14). Recently, Hell et al. developed a method to analyze detailed membrane lipid dynamics in living cells using FCS measurements and a super-

resolution microscope (15). However, since the measurement principle is difficult to understand, and the detection system is expensive, the method is currently not widely used. In addition, in the field of biology, they may consider that there are interactions among molecules to be more important than how fast are the mol-ecules moving. Therefore, in recent years, an analysis method that has been emerging is fluorescence cross correlation spectroscopy (FCCS), which analyzes the simultaneity of fluorescent molecules of two colors. Regarding the principles and detailed methods of FCS and FCCS, there are excellent reviews (16) by experts, thus we will omit them here and briefly describe only the concepts in this article. When separate fluorescent molecules (or fluorescently-labeled molecules) do not interact, the molecules pass through the focal region of the microscope at different time points. In this case, since the fluorescence emitted from the molecules do not have si-multaneity in the time axis, the cross correlation function does not arise (Fig. 2A). On the other hand, when two fluorescent molecules interact with each other, they simultaneously pass through the focal region, and the cross correlation function increases (Fig. 2B). From the comparison of these values, the presence or absence of interac-tion is determined. An example of FCCS using ganglioside GM1 and cholera toxin B subunit is shown (Fig. 2C). Although dedi-cated instruments specialized for FCCS analysis are commercially available, measurements also can be conducted by simply adding a highly sensitive electron multiplier called APD (Avalanche photo diode) to an existing confocal microscope. We believe that this analytical method is applicable to analyze carbohydrate chain-lectin interactions in glycolipid microdomains. In addition, Gratton

Fig. 2. Outline of Fluorescence Cross Correlation Spectroscopy (FCCS): When the two fluorescent molecules do not interact, the fluores-cence emitted by the molecules do not have simultaneity, and the cross correlation function (black line) does not rise (A). In contrast, when two fluorescent molecules interact with each other, the cross correlation function rises at the same time because they simultaneously passes through the focal region (B). Example of interaction analysis by FCCS: FCCS analysis results from four types of sphingolipids (sphingomy-elin (SM), glucosylceramide (GlcCer), lactosylceramide (LacCer), and ganglioside (GM1) labeled with BODIPY and cholera toxin B subunit (CTX) labeled with Alexa Fluor® 647 are shown.Cross correlation coefficient with CTX-Alexa 647 selectively increased only with BODIPY-GM1 (C).

© 2018 FCCA (Forum: Carbohydrates Coming of Age) E50

et al. established a new method in which the number of photons is measured by a PMT (Photo multiplier tube) used in a conven-tional confocal microscope instead of the aforementioned APD. To compensate for the accuracy of APD with PMT, a method was developed that divides the acquired data and the average value is evaluated (17, 18). If more case studies are conducted in the future, we expect that this technique will develop into a standard dynamic analysis method using confocal microscopy.

F. Glycolipid Microdomain Analysis Using Single Par-ticle Tracking

Single particle tracking analysis (SPT) is also a useful tech-nique for analyzing lateral diffusion of membrane receptors. To vi-sualize one particle on the cell membrane, a total internal reflection fluorescence microscope is used (19, 20). The principles behind this technique are noted in a prior review (21). There are research reports using artificial membranes in the fields of biophysics and surface science which mentioned that this method can be applied to analysis of the lipid rafts (22, 23). Detailed analysis of lipid rafts in living cells has also been conducted in recent years (24). To ana-lyze the dynamics of the glycolipid microdomain, it is desirable to use a probe that does not fluorescently label the lipid portion so as not to affect the phase separation state of the membrane. Recently, Kiso and Kusumi et al. have reported a system that performs fluo-rescence labeling on the non-reducing terminal part of the ganglio-side glycan chain and single particle tracking analysis (25). Previ-ously, using a total internal reflection fluorescence microscope, we also conducted dynamic observations of the insulin receptor (IR-GFP) and caveolin-1 (Cav1-GFP) fused with fluorescent proteins on the cell membrane (8). However, when performing diffusion measurements using a total internal reflection microscope, fluo-rescent proteins such as GFP are prone to fade rapidly (fading is completed about 5 s). Since restrictions are imposed on the setting of conditions, it is therefore necessary to use fluorescent molecules with high photostability. Thus, we prepared IR-TMR and Cav1-TMR labeled with tetramethylrhodamine (TMR) using HaloTag® technology (Promega) in which a tag protein forms a specific cova-lent bond with a small molecule ligand, and conducted an analysis. As a result, it was possible to observe one particle (although veri-fication is still necessary to determine whether it is one molecule) of the labeled molecules for about 15 s. By measuring the diffusion coefficients of approximately 100 particles, it was possible to dis-play the dynamics of the target molecules in a histogram. That is, since lateral diffusion of membrane molecules can be individually verified, information that cannot be extracted from the ensemble average can be extracted in a heterogeneous membrane environ-ment in which a glycolipid microdomain exists. For measurements

of diffusion coefficients, luminescent spot tracking can facilitate the progression from image acquisition to analysis. Tracking can be performed by using a plugin introduced to the free analysis ap-plication Image-J or by employing a commercially available appli-cation such as G-Track (G-angstrom Co.) (Fig. 3).

G. Simple Method for Complete Detergent Removal from Lipid Raft Fractions

In addition to the dynamic analysis of cell membrane mol-ecules by fluorescence microscopy, structural analysis of glyco-lipids accumulated in lipid rafts is also an important means for understanding the principles of glycolipid microdomain construc-tion. As a biochemical technique for separating and purifying lipid raft molecules, a method of recovering detergent- resistant microdomains (DRM) by sucrose density gradient centrifugation at low temperature is generally used. Nonionic detergents such as Triton X-100, Lubrol 98, Brij 58/97, and NP-40 are commonly used in this procedure (26, 27). Since detergents are very similar in structure and physical properties to glycolipids, thorough elimina-tion of detergents is essential in mass spectrometry of glycolipids contained in DRM. Nevertheless, the fact that a convenient method for separating glycolipids from detergents has not been found has

Fig. 3. Example of particle analysis: ① AVI image of cells expressing IR-TMR. ② Trace result of one particle (bright spot where the green lines intersect in image ①). ③ Histogram display of diffusion coefficients of all tracking-possible bright spots (using G-Track, G-angstrom Co.).

© 2018 FCCA (Forum: Carbohydrates Coming of Age)E51

hindered analysis of the glycolipid microdomain components (Fig. 4A, B). Therefore, we performed method development by drying until solid, a mixture of GM3 and Triton X-100 in test tubes, then washing the solids with 18 different kinds of organic solvents. The components of the fractions in the organic solvents and that re-maining in the test tubes were compared. As a result, we succeeded in completely fractionating the two components by using dichloro-ethane (DCE) (28). Based on this finding, spectral measurements using MALDI-QIT-TOF MS were carried out with DRM after the Svennerholm procedure and DCE washing. No detergent-derived peaks were detected in the DRM and we succeeded in identifying GM3 molecular species with different fatty acid chain lengths (Fig. 4C). By this method, structural analysis of glycolipids localized in lipid rafts as well as glycolipid composition of whole cells or tissues became possible. Therefore, it is expected that reports on comparative analysis related to pathological conditions and the like will increase in the future.

H. PerspectiveRecent fluorescence microscope technologies have embarked

on a new trend to explore the principles of super-resolution and harness them into practical use. On the other hand, by improving existing equipment or applications, precise measuring instruments can be developed. In the analysis using the FRAP method, it is possible to separate dynamic molecules into different velocity components by performing nonlinear curve fitting on the fluores-cence recovery curve, from which qualitative changes in the mem-brane can be predicted. Recently, Image Correlation Spectroscopy (ICS), which extended the concept of FCS to two dimensions, has been used for analysis of membrane molecules (29, 30). We are considering whether we can elucidate the function of glycolipid microdomains by using the Raster Image Correlation Spectroscopy (RICS) (18, 31). This allows for diffusion measurement and inter-action analysis of membrane molecules, and extends the ICS that can be analyzed by confocal fluorescence microscopy.

On the other hand, in mass spectrometry, we succeeded in separating not only detergent but also triglyceride from sample

Fig. 4. Total lipid analysis of 3T3L1 preadipocytes after sucrose density gradient centrifugation. Ganglioside GM3 and detergent Triton X-100 are mixed in the detergent-resistant microdomain (DRM) (A: orcinol staining, B: primulin staining). MALDI-QIT-TOF MS spectrum of ganglioside GM3, which was analyzed after DRM of 3T3L1 preadipocytes was washed with dichloroethane (DCE). Fatty acid chains shown in red were identified by MS/MS analysis (C)

© 2018 FCCA (Forum: Carbohydrates Coming of Age) E52

of glycolipids in adipocytes using the DCE washing method, and achieved structural analysis (32). If we use the adipose tissues de-rived diabetes patients and animal models, we will be able to verify our working hypothesis of lifestyle-related diseases, which is “the insulin receptor shifts from caveolae to GM3-rich domains in insu-lin-resistant adipocytes, thereby impair the insulin signaling.” (8) Furthermore, the DCE washing method is shown to be useful not only for the removal of nonionic detergents but also for the remov-al of the anionic detergents taurocholic acid and deoxycholic acid, and the amphoteric ionic detergent, CHAPS (28). Therefore, this washing method is available and started to be used not only for the removal of the detergent after membrane microdomain fraction-

ation, but also for other applications, including the recovery of gly-colipids from glycosyltransferase, carbohydrate hydrolysis enzyme reactions, cell lysates, and removal of detergents from the reaction mix (33). In this article, analysis of glycolipid microdomains using fluorescence microscopy and mass spectrometry was described. In the future, we expect not only to improve existing instruments but also to introduce new technology, such as visualization of glycolip-id microdomains at a spacial resolution exceeding a light diffrac-tion limit using super-resolution microscope, and high resolution structure determination of biomolecules in solution using single particle cryo-electron microscopy. These will help to elucidate the principle of glycolipid microdomain construction completely.

References 1. Hakomori, S. (2004) Glycoconj. J. 21, 125–137. 2. Kina, K., Masuda, H., Nakayama, H., Nagatsuka, Y., Nabetani, T., Hirabayashi, Y., Takahashi, Y., Shimada, K., Daida, H., Ogawa, H., Takamori,

K., and Iwabuchi, K. (2011) J. Immunol. 186, 5323–5332. 3. Nagatsuka, Y., Horibata, Y., Yamazaki, Y., Kinoshita, M., Shinoda, Y., Hashikawa, T., Koshino, H., Nakamura, T., and Hirabayashi, Y. (2006)

Biochemistry 45, 8742–8750. 4. Fujita, A., Cheng, J., Hirakawa, M., Furukawa, K., Kusunoki, S., and Fujimoto, T. (2007) Mol. Biol. Cell 18, 2112–2122. 5. Tanaka, K. A., Suzuki, K. G., Shirai, Y. M., Shibutani, S. T., Miyahara, M. S., Tsuboi, H., Yahara, M., Yoshimura, A., Mayor, S., Fujiwara, T. K.,

and Kusumi, A. (2010) Nat. Methods 7, 865–866. 6. Inokuchi, J., and Kabayama, K. (2008) Trends Glycosci. Glyc. 20, 353–371. 7. Schuster, S. C. (2008) Nat. Methods 5, 1. 8. Kabayama, K., Sato, T., Saito, K., Loberto, N., Prinetti, A., Sonnino, S., Kinjo, M., Igarashi, Y., and Inokuchi, J. (2007) Proc. Natl. Acad. Sci.

U.S.A. 104, 13678–13683. 9. Sekimoto, J., Kabayama, K., Gohara, K., and Inokuchi, J. (2012) FEBS Lett. 586, 191–195.10. Axelrod, D., Koppel, D. E., Schlessinger, J., Elson, E., and Webb, W. W. (1976) Biophys. J. 16, 1055–1069.11. Lippincott-Schwartz, J., Snapp, E., and Kenworthy, A. (2001) Nat. Rev. Mol. Cell Biol. 2, 444–456.12. Meder, D., Moreno, M. J., Verkade, P., Vaz, W. L., and Simons, K. (2006) Proc. Natl. Acad. Sci. U.S.A. 103, 329–334.13. Kinjo, M., and Rigler, R. (1995) Nucleic Acids Res. 23, 1795–1799.14. Mizuno, M., Noguchi, M., Imai, M., Motoyoshi, T., and Inazu, T. (2004) Bioorg. Med. Chem. Lett. 14, 485–490.15. Eggeling, C., Ringemann, C., Medda, R., Schwarzmann, G., Sandhoff, K., Polyakova, S., Belov, V. N., Hein, B., von Middendorff, C., Schonle, A.,

and Hell, S. W. (2009) Nature 457, 1159–1162.16. Krieger, J. W., Singh, A. P., Bag, N., Garbe, C. S., Saunders, T. E., Langowski, J., and Wohland, T. (2015) Nat. Protoc. 10, 1948–1974.17. Digman, M. A., Brown, C. M., Sengupta, P., Wiseman, P. W., Horwitz, A. R., and Gratton, E. (2005) Biophys. J. 89, 1317–1327.18. Rossow, M. J., Sasaki, J. M., Digman, M. A., and Gratton, E. (2010) Nat. Protoc. 5, 1761–1774.19. Axelrod, D., Thompson, N. L., and Burghardt, T. P. (1983) J. Microsc. 129, 19–28.20. Sako, Y., Minoghchi, S., and Yanagida, T. (2000) Nat. Cell Biol. 2, 168–172.21. Axelrod, D., Burghardt, T. P., and Thompson, N. L. (1984) Annu. Rev. Biophys. Bioeng. 13, 247–268.22. Tero, R., Sazaki, G., Ujihara, T., and Urisu, T. (2011) Langmuir 27, 9662–9665.23. Wu, H. M., Lin, Y. H., Yen, T. C., and Hsieh, C. L. (2016) Sci. Rep. 6, 20542.24. Suzuki, K. G. (2016) Methods Mol. Biol. 1376, 229–238.25. Komura, N., Suzuki, K. G., Ando, H., Konishi, M., Koikeda, M., Imamura, A., Chadda, R., Fujiwara, T. K., Tsuboi, H., Sheng, R., Cho, W.,

Furukawa, K., Furukawa, K., Yamauchi, Y., Ishida, H., Kusumi, A., and Kiso, M. (2016) Nat. Chem. Biol. 12, 402–410.26. Brown, D. A., and Rose, J. K. (1992) Cell 68, 533–544.27. Fiedler, K., Kobayashi, T., Kurzchalia, T. V., and Simons, K. (1993) Biochemistry 32, 6365–6373.28. Suzuki, Y., and Kabayama, K. (2012) J. Lipid Res. 53, 599–608.29. Wiseman, P. W. (2015) Cold Spring Harb. Protoc. 2015, 336–348.30. Kolin, D. L., and Wiseman, P. W. (2007) Cell Biochem. Biophys. 49, 141–164.31. Hedde, P. N., Dorlich, R. M., Blomley, R., Gradl, D., Oppong, E., Cato, A. C., and Nienhaus, G. U. (2013) Nat. Commun. 4, 2093.32. Kojima, H., Suzuki, Y., Ito, M., and Kabayama, K. (2015) Lipids 50, 913–917.33. Sun, C. Q., Hubl, U., Hoefakker, P., Vasudevamurthy, M. K., and Johnson, K. D. (2014) PLoS One 9, e94206.

© 2018 FCCA (Forum: Carbohydrates Coming of Age)E53

Kazuya Kabayama (PhD): Department of chemistry, Graduate school of science, Osaka university. 1997 Graduated at Keio university, Faculty of science and technology (Prof. Seiichiro Ogawa). 1997–1999 Master course of graduate school of science and technology, Keio university (Prof. Seiichiro Ogawa). 1999–2002 Doctor course of graduate school of pharmaceutical sciences, Hokkaido universi-ty (Prof. Yasuyuki Igarashi). 2002–2003 Postdoctoral fellow, Graduate school of pharmaceutical sci-ences, Hokkaido university. 2004–2006 Research fellow, Core Research for Evolution Science and Technology (CREST) program. 2006–2008 Assistant professor, Institute of molecular biomembranes and glycobiology, Tohoku pharmaceutical University (Prof. Jin-ichi Inokuchi). 2009–2014 Associate professor (principal investigator), Institute of glycoscience, Tokai university. 2014–present Associate professor, Graduate school of science, Osaka university (Prof. Koichi Fukase).

Information of the Authors