- Email: [email protected]
Enzymatic Assay of Galactosyltransferase by Capillary Electrophoresis
ANALYTICAL BIOCHEMISTRY ARTICLE NO.
263, 240 –245 (1998)
AB982762
Enzymatic Assay of Galactosyltransferase by Capillary Electrophoresis Yoshimi Kanie,* Annette Kirsch,* Osamu Kanie,*,1 and Chi-Huey Wong*,†,1 *Frontier Research Program, The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako-shi, Saitama 351-01 Japan; and †Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
Received February 20, 1998
The kinetic parameters of a galactosyltransferase-catalyzed reaction were determined for the first time using capillary zone electrophoresis (CZE) using the methylumbelliferyl (MU) glycoside of N-acetylglucosamine as the acceptor molecule. The CZE was performed using borate buffer and the enzymatic transformations were monitored at 214 nm. The kinetic parameters obtained for MU-GlcNAc were Km 5 35.9 mM and Vmax 5 7.5 mmol/ min/mg, and those for UDP-Gal were Km 5 115.3 mM and Vmax 5 12.4 mmol/min/mg. A representative inhibition assay was also carried out using UDP as an inhibitor to give the Ki value of 83.9 mM against MU-GlcNAc. The structure of the synthetic product was also confirmed using 1H NMR spectroscopies after isolation by simple chromatography. © 1998 Academic Press
Carbohydrate structures in glycoproteins, glycolipids, and polysaccharides are involved in many important biological recognition phenomena (1). The activities of these oligosaccharides are regulated by the sequential actions of glycosidases and glycosyltransferases in the Golgi apparatus. The mechanistic studies of glycosyltransferases are usually more complicated than those of glycosidases. They are traditionally carried out using (i) radiolabeled sugar nucleotides when the enzymes have weak activities (2), (ii) spectrophotometric assays based upon the released nucleoside phosphate coupled with pyruvate kinase and NADH-dependent lactate dehydrogenase (2) and the action of glycosidases to liberate fluorophore (3), or (iii) HPLC (high-performance liquid chromatography) (4) or TLC (thin-layer chromatography) (5, 6) using a fluo-
rescently labeled sugar acceptor. Other approaches for the study of glycosyltransferase activity using capillary zone electrophoresis (7–10) and electrospray mass spectrometry (11) have been examined recently in a qualitative or semiquantitative manner. Of these methods, the first one is considered to be the most sensitive, as it needs only a very small amount of material. It is, however, very inconvenient to use radiolabeled substrates, and in some cases it requires lengthy synthesis of radiolabeled substrates. Capillary zone electrophoresis (CZE)2 is an attractive alternative for the kinetic studies of this class of enzymes because it does not rely on radio isotopes, and it can be used to analyze enzyme activity of very small quantities without purification. However, since the basis of the method is the attraction of charged molecules to electrodes, differentiating the migration times of neutral carbohydrates is the most challenging task (12). Our strategy to utilize CZE in the assay of glycoenzymes is based on the decrease or increase in size of carbohydrate molecules during the reaction. In this instance, it is easier to obtain a different migration time for each compound using the borate complex compared to separation of neutral stereoisomers because the number of charges to be introduced depends on the number of hydroxyl groups (13). Studies of glycosidases involving CZE analysis have been reported to quantitate the reaction product; however, analyses of kinetic parameters have not been reported (7–10). In a representative study of galactosyltransferase (EC 2.4.1.22, GalTase) using borate buffer as electrolyte, N-acetylglucosamine carrying 4-methylumbelliferyl (MU) group (14, 8) as the aglycon was used as acceptor because it is active at ultraviolet (uv) absorptions of 206, 286, and 317 nm in addition to its
1
To whom correspondence should be addressed. Dr. Kanie: Fax 48-467-9620. E-mail:[email protected]. Dr. Wong: Fax (619) 784-2409. 240
2 Abbreviations used: CZE, capillary zone electrophoresis; MU, methylumbelliferyl; Mes, 2-(N-morpholino)ethanesulfonic acid.
0003-2697/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
241
FIG. 1. The standard curve using MU-GlcNAc was created. The migration times were reproduced within a slight error (60.2 min). The numbers on top of the peaks indicate the concentration of MU-GlcNAc in pmol.
fluorescent activity (emission at 375 nm, excitation at 317 nm). Although the assay may use the fluorescence of either the liberated 4-methylumbelliferone or the product separated by HPLC or TLC, the uv-based assay was used in this study. In addition, it is worth noting that the methylumbelliferyl group was used not only for the detection of the acceptor and the product but also for isolation of the product by simple reverse-phase column chromatography (SepPak). MATERIALS AND METHODS
Materials
b-Galactosyltransferase (EC 2.4.1.22 from bovine milk), UDP-galactose, and uridine 59-diphosphonate
(UDP) were from Sigma Chemical Co. (St. Louis, MO). Cacodylic acid sodium salt, Hepes [2-{4-(2-hydroxyethyl)1-piperazinyl} ethanesulfonic acid], Mes [2-(N-morpholino)ethanesulfonic acid], MnCl2, sodium tetraborate, and potassium hydroxide were from Nakalai Tesque Inc. (Kyoto, Japan). 4-Methylumbelliferyl 2-acetamido-2-deoxy-b-D-glucopyranoside (MU-GlcNAc) was from Wako Pure Chemical Ltd. (Osaka, Japan). Double-deionized water was prepared from a Milli-Q system from Millipore Corp. (Milford, MA). SepPak C-18 reverse-phase cartridges were from Waters Corp. (Milford, MA). Millex-GV syringe filters (0.22 mm 3 4 mm i.d.) were purchased from Nihon Millipore Ltd. (Yonezawa, Japan). Iatro Beads (60 m) was from Dia-Iatron Laboratories, Inc. (Tokyo, Japan). Assays were performed on a Waters
242
KANIE ET AL.
SCHEME 1
Quanta 4000E capillary electrophoresis system, which was equipped with a 50-cm fused silica capillary of i.d. 75 mm. Detection was carried out by on-column measurement of uv absorption at 214 nm at 7.5 cm from the cathode. Pherograms were recorded on a Millennium 2010 system from Millipore Corp. Methods Condition of capillary zone electrophoresis. The capillary used was pretreated or regenerated with 0.1
M KOH and washed with double deionized water after each injection. Samples were loaded by means of hydrostatic pressure at 10 cm height for 30 s (38.4 nl). Electrophoresis was performed at 15 kV using 50 mM sodium borate as electrolyte at a constant temperature of 35°C. Kinetic analysis of galactosyltransferase. Incubations were performed in a total volume of 250 ml. Unless otherwise stated, reaction mixtures contained 0.1 M cacodylate buffer (pH 8.3), 10 mM MnCl2, various
FIG. 2. Typical electropherograms are shown. (A) The acceptor methylumbelliferyl N-acetylglucosaminide (MU-GlcNAc); (B) the reaction mixture quenched after 30 min.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
FIG. 3. Time course study of transfer of galactose from UDP-Gal to MU-GlcNAc. The conversion rate is calculated based on the amount of UDP-Gal used. Small aliquots (50 ml) of the ice-cold reaction mixture were removed every 5 to 10 min, added to 5 ml of 100 mM borate buffer, and heated at 80°C for 10 min. The mixtures were then filtered through a Millex-GV membrane filter and finally analyzed on CZE.
amounts of UDP-Gal (10 ; 100 mM), and MU-GlcNAc (90 ; 230 mM) (and 0 ; 60 mM UDP for inhibition assay) with 5 mU / 250 ml (total volume) GalTase. After incubation for 10 min at 37°C, the reaction was diluted with 25 ml of 100 mM borate and terminated by heating at 80°C for 10 min. The resulting mixture was finally filtered with Millex-GV filter to remove the precipitate. Isolation of reaction product. To confirm the structure of the product, the combined quenched reaction mixture was loaded onto SepPak C-18 (15), prewashed with MeOH (20 ml) and water (20 ml), and washed with water (10 ml), and MU glycosides were eluted with MeOH (10 ml); the resulting mixture was concentrated under vacuum. The MU glycosides thus obtained were further purified on a column of Iatro Beads using 10:2 and 10:3 CHCl3-MeOH. Concentration of the eluent containing the product, filtration using Millex GV filter, and lyophilization gave methylumbelliferyl N-acetyllactosaminide (MU-LacNAc). Analysis by 1H NMR spectroscopy. 1H NMR (400 MHz) spectroscopies were performed using a JEOL 400 spectrometer in D2O with DOH (4.80 ppm) as an internal standard at 25°C. The isolated MU-LacNAc (ca 1 mg) was relyophilized from D2O, dissolved in D2O, and used for the NMR analysis. RESULTS AND DISCUSSION
The CZE conditions for the quantitative analysis were first examined. When the MU-GlcNAc (50 ; 200 mM; 1.92 ; 7.68 pmol actual injection amounts) was analyzed using sodium borate as the electrolyte, a lin-
243
ear correlation was observed in a range of concentrations suitable for the enzyme assay (Fig. 1). The migration times obtained were within the error of 0.02 min. Based on the result that CZE can be used for quantitative analysis, we decided to analyze GalTase reaction (Scheme 1) and started with the selection of buffer to be used in the enzyme reaction. Initial studies using Tris, cacodylate, Hepes, and Mes buffers in the GalTase reaction indicated that either cacodylate or Hepes buffer may be used in the assay. Further examination of these two buffers revealed that the former was better in the assay in terms of reaction rate despite the fact that it was reported to inhibit the GalTase reaction (16). A typical electropherogram showing the separation of MU-GlcNAc and MU-LacNAc is shown in Fig. 2. The migration times for acceptor MU-GlcNAc and product MU-LacNAc under these conditions were 8 6 0.2 and 8.5 6 0.2 min, respectively. The time course study was then carried out in the presence of 10 mM MnCl2. The concentration used was based on the conditions that give good conversion yields when 0.7 mM acceptor was used in the assay. As a result, up to 50% conversion was observed after 100 min (Fig. 3). It is known that the reaction can be improved by the addition of alkaline phosphatase to avoid product inhibition (17). Finally, estimation of the kinetic parameters was carried out using the conditions thus determined.
FIG. 4. Double-reciprocal plots were carried out to obtain Km and Vmax values of MU-GlcNAc for the galactosyltransfer reaction in the presence or absence of UDP as an inhibitor. The concentrations of UDP were (■) 0 mM, (h) 20 mM, (F) 40 mM, (E) 60 mM. The Km value calculated from the plot (■) was 35.9 mM. (Inset) Replot of 1/v ; 1/[MU-GlcNAc] over [UDP]. Ki for UDP 5 83.9 mM. The concentrations of UDP-Gal and Mn21 used in these assays are 100 mM and 10 mM, respectively.
244
KANIE ET AL.
FIG. 5. Double-reciprocal plots for the galactosyltransfer reaction were carried out to obtain the Km and Vmax values of UDP-Gal. Km 5 115.3 mM.
Yields calculated based on the relative ratio of MULacNAc to MU-GlcNAc were used for the assay to eliminate the potential experimental error arising from work-up. The apparent Km and Vmax values for acceptor were calculated (18) from the double-reciprocal plot of standard 1/v-1/[MU-GlcNAc] curve (Fig. 4) to be 35.9 mM and 7.5 mmol/min/mg, respectively. The Km value shown here is comparable to the previously re-
ported value (14, 15, 19). In a similar manner, the kinetic parameters for UDP-Gal (determined with different concentrations of UDP-Gal) were Km 5 115.3 mM and Vmax 5 12.4 mmol/min/mg (Fig. 5). Since UDP is known to inhibit the GalTase reaction (19, 20), the inhibitory activity was next evaluated. It is shown in a replot (Fig. 4, inset) that UDP is a mixed type inhibitor with respect to the acceptor MU-GlcNAc with Ki of 83.9 mM. To confirm that the peak that appeared at 8.5 min was truly the peak of MU-LacNAc, we have isolated the compound using a SepPak C-18 cartridge (15) and silica gel column chromatography. Basically, simple filtration operations were performed for both columns. The detection of the eluted compounds was carried out using CZE. Assignments of signals in 1H NMR of the product (Fig. 6) were as follows: d7.75 [d, 1H, J 5 9.3Hz, H-6 (MU)], 7.07–7.10 [2H, H-5,8 (MU)], 6.29 [s, 1H, H-3 (MU)], 5.31 [d, 1H, J1,2 5 8.3Hz, H-1 (GlcNAc)], 4.53 [d, 1H, J1,2 5 7.6Hz, H-1(Gal)], 2.46 (s, 3H, CH3), 2.04 (s, 3H, CH3CO). In conclusion, the kinetic parameters of a glycosyltransferase reaction were determined for the first time using CZE. The advantages of this method over those currently used most are the elimination of handling radiolabeled materials and only a relatively small amount (pmol order) of materials are needed for detection. Utilization of CZE for the kinetic and inhibition assays is therefore quite useful. Work is in progress to apply this method to the analysis of other enzymes.
FIG. 6. Expansion of a selected region (3.4 ; 5.5 ppm) of 1H NMR spectrum was shown. The signals assigned to the H-1 protons of GlcNAc and Gal residues were observed at 5.31 and 4.53 ppm, respectively. The coupling constants of H-1–H-2 indicated in the figure are supportive of the b-linkage for both glycosidic bonds.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
ACKNOWLEDGMENTS The authors thank Dr. Katsuhiko Suzuki for critical discussions. We thank the NMR laboratory staff, and the Institute of Physical and Chemical Research (RIKEN) for the 400 MHz NMR. We are also grateful to Dr. Yoshitaka Nagai, Director of the Glycobiology Research Group, and Dr. Tomoya Ogawa, Coordinator of the Frontier Research Program of the Institute of Physical and Chemical Research (RIKEN), for their continued support and encouragement of our research. This research was supported in part by the Science and Technology Agency of the Japanese government.
REFERENCES 1. Varki, A. (1993) Glycobiology 3, 97–130. 2. Fitzgerald, D. K., Colvin, B., Mawal, R., and Ebner, K. E. (1970) Anal. Biochem. 36, 43– 61. 3. Creme, S., and Leaback, D. H. (1980) Anal. Biochem. 103, 258 – 263. 4. Morita, N., Hase, S., Ikenaka, K., Mikoshiba, K., and Ikenaka, T. (1988) J. Biochem. 103, 332–335. 5. Creme, S., and Leaback, D. H. (1980) Biochem. Soc. Trans. 8, 558 –559. 6. Leigh, A. G. W., and Leaback, D. H. (1988) Anal. Chim. Acta 212, 213–221. 7. Chan, N. W. C., Stangier, K., Sherburne, R., Taylor, D. E., Zhang, Y., Dovichi, N. J., and Palcic, M. M. (1995) Glycobiology 5, 683– 688. 8. Lee, K. B., Desai, U. R., Palcic, M. M., Hindsgaul, O., and Linhardt, R. J. (1992) Anal. Biochem. 205, 108 –114.
245
9. Lee, K.-B., Kim, Y.-S., and Linhardt, R. J. (1991) Electrophoresis 12, 636 – 640. 10. Zeleny, R., Altmann, F., and Praznik, W. (1997) Anal. Biochem. 246, 96 –101. 11. (a) Takayama, S., Martin, R., Wu, J., Laslo, K., Siuzdak, G., and Wong, C-H. (1997) J. Am. Chem. Soc. 119, 8146 – 8151. (b) Wu, J., Takayama, S., Wong, C.-H., and Siuzdak, G. (1997) Chem. Biol. 4, 653– 657. 12. Linhardt, R. J., and Pervin, A. (1996) J. Chromatogr. A 720, 323–335. 13. Hoffstetter-Kuhn, S., Paulus, A., Gassmann, E., and Widmer, H. M. (1991) Anal. Chem. 63, 1541–1547; Honda, S., Iwase, S., and Makino, A., Fujiwara, S. (1989) Anal. Biochem. 176, 72–77; Honda, S. (1996) J. Chromatogr. 720, 337–351. 14. Geren, C. R., Magee, S. C., and Ebner, K. E. (1976) Arch. Biochem. Biophys. 172, 149 –155. 15. Palcic, M. M., Heerze, L. D., Pierce, M., and Hindsgaul, O. (1988) Glycoconj. J. 5, 49 – 63. 16. Boyle, F. A., Cook, N. D., and Peters, T. J. (1988) Clin. Chim. Acta 172, 291–296. 17. Unverzagt, C., Kunz, H., and Paulson, J. C. (1990) J. Am. Chem. Soc. 112, 9308 –9309. 18. Wilkinson, G. N. (1961) Biochem. J. 80, 324 –332. 19. Ichikawa, Y., Lin, Y.-C., Dumas, D. P., Shen, G.-J., GarciaJunceda, E., Williams, M. K., Bayer, R., Ketcham, C., Walker, L. E., Paulson, J. C., and Wong, C.-H. (1992) J. Am. Chem. Soc. 114, 9283–9298. 20. Magee, S. C., and Ebner, K. E. (1974) J. Biol. Chem. 249, 6992– 6998.
263, 240 –245 (1998)
AB982762
Enzymatic Assay of Galactosyltransferase by Capillary Electrophoresis Yoshimi Kanie,* Annette Kirsch,* Osamu Kanie,*,1 and Chi-Huey Wong*,†,1 *Frontier Research Program, The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako-shi, Saitama 351-01 Japan; and †Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
Received February 20, 1998
The kinetic parameters of a galactosyltransferase-catalyzed reaction were determined for the first time using capillary zone electrophoresis (CZE) using the methylumbelliferyl (MU) glycoside of N-acetylglucosamine as the acceptor molecule. The CZE was performed using borate buffer and the enzymatic transformations were monitored at 214 nm. The kinetic parameters obtained for MU-GlcNAc were Km 5 35.9 mM and Vmax 5 7.5 mmol/ min/mg, and those for UDP-Gal were Km 5 115.3 mM and Vmax 5 12.4 mmol/min/mg. A representative inhibition assay was also carried out using UDP as an inhibitor to give the Ki value of 83.9 mM against MU-GlcNAc. The structure of the synthetic product was also confirmed using 1H NMR spectroscopies after isolation by simple chromatography. © 1998 Academic Press
Carbohydrate structures in glycoproteins, glycolipids, and polysaccharides are involved in many important biological recognition phenomena (1). The activities of these oligosaccharides are regulated by the sequential actions of glycosidases and glycosyltransferases in the Golgi apparatus. The mechanistic studies of glycosyltransferases are usually more complicated than those of glycosidases. They are traditionally carried out using (i) radiolabeled sugar nucleotides when the enzymes have weak activities (2), (ii) spectrophotometric assays based upon the released nucleoside phosphate coupled with pyruvate kinase and NADH-dependent lactate dehydrogenase (2) and the action of glycosidases to liberate fluorophore (3), or (iii) HPLC (high-performance liquid chromatography) (4) or TLC (thin-layer chromatography) (5, 6) using a fluo-
rescently labeled sugar acceptor. Other approaches for the study of glycosyltransferase activity using capillary zone electrophoresis (7–10) and electrospray mass spectrometry (11) have been examined recently in a qualitative or semiquantitative manner. Of these methods, the first one is considered to be the most sensitive, as it needs only a very small amount of material. It is, however, very inconvenient to use radiolabeled substrates, and in some cases it requires lengthy synthesis of radiolabeled substrates. Capillary zone electrophoresis (CZE)2 is an attractive alternative for the kinetic studies of this class of enzymes because it does not rely on radio isotopes, and it can be used to analyze enzyme activity of very small quantities without purification. However, since the basis of the method is the attraction of charged molecules to electrodes, differentiating the migration times of neutral carbohydrates is the most challenging task (12). Our strategy to utilize CZE in the assay of glycoenzymes is based on the decrease or increase in size of carbohydrate molecules during the reaction. In this instance, it is easier to obtain a different migration time for each compound using the borate complex compared to separation of neutral stereoisomers because the number of charges to be introduced depends on the number of hydroxyl groups (13). Studies of glycosidases involving CZE analysis have been reported to quantitate the reaction product; however, analyses of kinetic parameters have not been reported (7–10). In a representative study of galactosyltransferase (EC 2.4.1.22, GalTase) using borate buffer as electrolyte, N-acetylglucosamine carrying 4-methylumbelliferyl (MU) group (14, 8) as the aglycon was used as acceptor because it is active at ultraviolet (uv) absorptions of 206, 286, and 317 nm in addition to its
1
To whom correspondence should be addressed. Dr. Kanie: Fax 48-467-9620. E-mail:[email protected]. Dr. Wong: Fax (619) 784-2409. 240
2 Abbreviations used: CZE, capillary zone electrophoresis; MU, methylumbelliferyl; Mes, 2-(N-morpholino)ethanesulfonic acid.
0003-2697/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
241
FIG. 1. The standard curve using MU-GlcNAc was created. The migration times were reproduced within a slight error (60.2 min). The numbers on top of the peaks indicate the concentration of MU-GlcNAc in pmol.
fluorescent activity (emission at 375 nm, excitation at 317 nm). Although the assay may use the fluorescence of either the liberated 4-methylumbelliferone or the product separated by HPLC or TLC, the uv-based assay was used in this study. In addition, it is worth noting that the methylumbelliferyl group was used not only for the detection of the acceptor and the product but also for isolation of the product by simple reverse-phase column chromatography (SepPak). MATERIALS AND METHODS
Materials
b-Galactosyltransferase (EC 2.4.1.22 from bovine milk), UDP-galactose, and uridine 59-diphosphonate
(UDP) were from Sigma Chemical Co. (St. Louis, MO). Cacodylic acid sodium salt, Hepes [2-{4-(2-hydroxyethyl)1-piperazinyl} ethanesulfonic acid], Mes [2-(N-morpholino)ethanesulfonic acid], MnCl2, sodium tetraborate, and potassium hydroxide were from Nakalai Tesque Inc. (Kyoto, Japan). 4-Methylumbelliferyl 2-acetamido-2-deoxy-b-D-glucopyranoside (MU-GlcNAc) was from Wako Pure Chemical Ltd. (Osaka, Japan). Double-deionized water was prepared from a Milli-Q system from Millipore Corp. (Milford, MA). SepPak C-18 reverse-phase cartridges were from Waters Corp. (Milford, MA). Millex-GV syringe filters (0.22 mm 3 4 mm i.d.) were purchased from Nihon Millipore Ltd. (Yonezawa, Japan). Iatro Beads (60 m) was from Dia-Iatron Laboratories, Inc. (Tokyo, Japan). Assays were performed on a Waters
242
KANIE ET AL.
SCHEME 1
Quanta 4000E capillary electrophoresis system, which was equipped with a 50-cm fused silica capillary of i.d. 75 mm. Detection was carried out by on-column measurement of uv absorption at 214 nm at 7.5 cm from the cathode. Pherograms were recorded on a Millennium 2010 system from Millipore Corp. Methods Condition of capillary zone electrophoresis. The capillary used was pretreated or regenerated with 0.1
M KOH and washed with double deionized water after each injection. Samples were loaded by means of hydrostatic pressure at 10 cm height for 30 s (38.4 nl). Electrophoresis was performed at 15 kV using 50 mM sodium borate as electrolyte at a constant temperature of 35°C. Kinetic analysis of galactosyltransferase. Incubations were performed in a total volume of 250 ml. Unless otherwise stated, reaction mixtures contained 0.1 M cacodylate buffer (pH 8.3), 10 mM MnCl2, various
FIG. 2. Typical electropherograms are shown. (A) The acceptor methylumbelliferyl N-acetylglucosaminide (MU-GlcNAc); (B) the reaction mixture quenched after 30 min.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
FIG. 3. Time course study of transfer of galactose from UDP-Gal to MU-GlcNAc. The conversion rate is calculated based on the amount of UDP-Gal used. Small aliquots (50 ml) of the ice-cold reaction mixture were removed every 5 to 10 min, added to 5 ml of 100 mM borate buffer, and heated at 80°C for 10 min. The mixtures were then filtered through a Millex-GV membrane filter and finally analyzed on CZE.
amounts of UDP-Gal (10 ; 100 mM), and MU-GlcNAc (90 ; 230 mM) (and 0 ; 60 mM UDP for inhibition assay) with 5 mU / 250 ml (total volume) GalTase. After incubation for 10 min at 37°C, the reaction was diluted with 25 ml of 100 mM borate and terminated by heating at 80°C for 10 min. The resulting mixture was finally filtered with Millex-GV filter to remove the precipitate. Isolation of reaction product. To confirm the structure of the product, the combined quenched reaction mixture was loaded onto SepPak C-18 (15), prewashed with MeOH (20 ml) and water (20 ml), and washed with water (10 ml), and MU glycosides were eluted with MeOH (10 ml); the resulting mixture was concentrated under vacuum. The MU glycosides thus obtained were further purified on a column of Iatro Beads using 10:2 and 10:3 CHCl3-MeOH. Concentration of the eluent containing the product, filtration using Millex GV filter, and lyophilization gave methylumbelliferyl N-acetyllactosaminide (MU-LacNAc). Analysis by 1H NMR spectroscopy. 1H NMR (400 MHz) spectroscopies were performed using a JEOL 400 spectrometer in D2O with DOH (4.80 ppm) as an internal standard at 25°C. The isolated MU-LacNAc (ca 1 mg) was relyophilized from D2O, dissolved in D2O, and used for the NMR analysis. RESULTS AND DISCUSSION
The CZE conditions for the quantitative analysis were first examined. When the MU-GlcNAc (50 ; 200 mM; 1.92 ; 7.68 pmol actual injection amounts) was analyzed using sodium borate as the electrolyte, a lin-
243
ear correlation was observed in a range of concentrations suitable for the enzyme assay (Fig. 1). The migration times obtained were within the error of 0.02 min. Based on the result that CZE can be used for quantitative analysis, we decided to analyze GalTase reaction (Scheme 1) and started with the selection of buffer to be used in the enzyme reaction. Initial studies using Tris, cacodylate, Hepes, and Mes buffers in the GalTase reaction indicated that either cacodylate or Hepes buffer may be used in the assay. Further examination of these two buffers revealed that the former was better in the assay in terms of reaction rate despite the fact that it was reported to inhibit the GalTase reaction (16). A typical electropherogram showing the separation of MU-GlcNAc and MU-LacNAc is shown in Fig. 2. The migration times for acceptor MU-GlcNAc and product MU-LacNAc under these conditions were 8 6 0.2 and 8.5 6 0.2 min, respectively. The time course study was then carried out in the presence of 10 mM MnCl2. The concentration used was based on the conditions that give good conversion yields when 0.7 mM acceptor was used in the assay. As a result, up to 50% conversion was observed after 100 min (Fig. 3). It is known that the reaction can be improved by the addition of alkaline phosphatase to avoid product inhibition (17). Finally, estimation of the kinetic parameters was carried out using the conditions thus determined.
FIG. 4. Double-reciprocal plots were carried out to obtain Km and Vmax values of MU-GlcNAc for the galactosyltransfer reaction in the presence or absence of UDP as an inhibitor. The concentrations of UDP were (■) 0 mM, (h) 20 mM, (F) 40 mM, (E) 60 mM. The Km value calculated from the plot (■) was 35.9 mM. (Inset) Replot of 1/v ; 1/[MU-GlcNAc] over [UDP]. Ki for UDP 5 83.9 mM. The concentrations of UDP-Gal and Mn21 used in these assays are 100 mM and 10 mM, respectively.
244
KANIE ET AL.
FIG. 5. Double-reciprocal plots for the galactosyltransfer reaction were carried out to obtain the Km and Vmax values of UDP-Gal. Km 5 115.3 mM.
Yields calculated based on the relative ratio of MULacNAc to MU-GlcNAc were used for the assay to eliminate the potential experimental error arising from work-up. The apparent Km and Vmax values for acceptor were calculated (18) from the double-reciprocal plot of standard 1/v-1/[MU-GlcNAc] curve (Fig. 4) to be 35.9 mM and 7.5 mmol/min/mg, respectively. The Km value shown here is comparable to the previously re-
ported value (14, 15, 19). In a similar manner, the kinetic parameters for UDP-Gal (determined with different concentrations of UDP-Gal) were Km 5 115.3 mM and Vmax 5 12.4 mmol/min/mg (Fig. 5). Since UDP is known to inhibit the GalTase reaction (19, 20), the inhibitory activity was next evaluated. It is shown in a replot (Fig. 4, inset) that UDP is a mixed type inhibitor with respect to the acceptor MU-GlcNAc with Ki of 83.9 mM. To confirm that the peak that appeared at 8.5 min was truly the peak of MU-LacNAc, we have isolated the compound using a SepPak C-18 cartridge (15) and silica gel column chromatography. Basically, simple filtration operations were performed for both columns. The detection of the eluted compounds was carried out using CZE. Assignments of signals in 1H NMR of the product (Fig. 6) were as follows: d7.75 [d, 1H, J 5 9.3Hz, H-6 (MU)], 7.07–7.10 [2H, H-5,8 (MU)], 6.29 [s, 1H, H-3 (MU)], 5.31 [d, 1H, J1,2 5 8.3Hz, H-1 (GlcNAc)], 4.53 [d, 1H, J1,2 5 7.6Hz, H-1(Gal)], 2.46 (s, 3H, CH3), 2.04 (s, 3H, CH3CO). In conclusion, the kinetic parameters of a glycosyltransferase reaction were determined for the first time using CZE. The advantages of this method over those currently used most are the elimination of handling radiolabeled materials and only a relatively small amount (pmol order) of materials are needed for detection. Utilization of CZE for the kinetic and inhibition assays is therefore quite useful. Work is in progress to apply this method to the analysis of other enzymes.
FIG. 6. Expansion of a selected region (3.4 ; 5.5 ppm) of 1H NMR spectrum was shown. The signals assigned to the H-1 protons of GlcNAc and Gal residues were observed at 5.31 and 4.53 ppm, respectively. The coupling constants of H-1–H-2 indicated in the figure are supportive of the b-linkage for both glycosidic bonds.
ELECTROPHORETIC GALACTOSYLTRANSFERASE ASSAY
ACKNOWLEDGMENTS The authors thank Dr. Katsuhiko Suzuki for critical discussions. We thank the NMR laboratory staff, and the Institute of Physical and Chemical Research (RIKEN) for the 400 MHz NMR. We are also grateful to Dr. Yoshitaka Nagai, Director of the Glycobiology Research Group, and Dr. Tomoya Ogawa, Coordinator of the Frontier Research Program of the Institute of Physical and Chemical Research (RIKEN), for their continued support and encouragement of our research. This research was supported in part by the Science and Technology Agency of the Japanese government.
REFERENCES 1. Varki, A. (1993) Glycobiology 3, 97–130. 2. Fitzgerald, D. K., Colvin, B., Mawal, R., and Ebner, K. E. (1970) Anal. Biochem. 36, 43– 61. 3. Creme, S., and Leaback, D. H. (1980) Anal. Biochem. 103, 258 – 263. 4. Morita, N., Hase, S., Ikenaka, K., Mikoshiba, K., and Ikenaka, T. (1988) J. Biochem. 103, 332–335. 5. Creme, S., and Leaback, D. H. (1980) Biochem. Soc. Trans. 8, 558 –559. 6. Leigh, A. G. W., and Leaback, D. H. (1988) Anal. Chim. Acta 212, 213–221. 7. Chan, N. W. C., Stangier, K., Sherburne, R., Taylor, D. E., Zhang, Y., Dovichi, N. J., and Palcic, M. M. (1995) Glycobiology 5, 683– 688. 8. Lee, K. B., Desai, U. R., Palcic, M. M., Hindsgaul, O., and Linhardt, R. J. (1992) Anal. Biochem. 205, 108 –114.
245
9. Lee, K.-B., Kim, Y.-S., and Linhardt, R. J. (1991) Electrophoresis 12, 636 – 640. 10. Zeleny, R., Altmann, F., and Praznik, W. (1997) Anal. Biochem. 246, 96 –101. 11. (a) Takayama, S., Martin, R., Wu, J., Laslo, K., Siuzdak, G., and Wong, C-H. (1997) J. Am. Chem. Soc. 119, 8146 – 8151. (b) Wu, J., Takayama, S., Wong, C.-H., and Siuzdak, G. (1997) Chem. Biol. 4, 653– 657. 12. Linhardt, R. J., and Pervin, A. (1996) J. Chromatogr. A 720, 323–335. 13. Hoffstetter-Kuhn, S., Paulus, A., Gassmann, E., and Widmer, H. M. (1991) Anal. Chem. 63, 1541–1547; Honda, S., Iwase, S., and Makino, A., Fujiwara, S. (1989) Anal. Biochem. 176, 72–77; Honda, S. (1996) J. Chromatogr. 720, 337–351. 14. Geren, C. R., Magee, S. C., and Ebner, K. E. (1976) Arch. Biochem. Biophys. 172, 149 –155. 15. Palcic, M. M., Heerze, L. D., Pierce, M., and Hindsgaul, O. (1988) Glycoconj. J. 5, 49 – 63. 16. Boyle, F. A., Cook, N. D., and Peters, T. J. (1988) Clin. Chim. Acta 172, 291–296. 17. Unverzagt, C., Kunz, H., and Paulson, J. C. (1990) J. Am. Chem. Soc. 112, 9308 –9309. 18. Wilkinson, G. N. (1961) Biochem. J. 80, 324 –332. 19. Ichikawa, Y., Lin, Y.-C., Dumas, D. P., Shen, G.-J., GarciaJunceda, E., Williams, M. K., Bayer, R., Ketcham, C., Walker, L. E., Paulson, J. C., and Wong, C.-H. (1992) J. Am. Chem. Soc. 114, 9283–9298. 20. Magee, S. C., and Ebner, K. E. (1974) J. Biol. Chem. 249, 6992– 6998.