Substrate Preference and Interplay of Fucosyltransferase 8 and N

Communication pubs.acs.org/JACS

Substrate Preference and Interplay of Fucosyltransferase 8 and N‑Acetylglucosaminyltransferases Tzu-Hao Tseng,†,§ Tzu-Wen Lin,† Chien-Yu Chen,†,‡ Chein-Hung Chen,† Jung-Lee Lin,† Tsui-Ling Hsu,† and Chi-Huey Wong*,†,§ †

Genomics Research Center, Academia Sinica, 128 Academia Road, Section 2, Nankang, Taipei 115, Taiwan Institute of Microbiology and Immunology, National Yang-Ming University, 155 Linong Street, Section 2, Beitou, Taipei 112, Taiwan

§

S Supporting Information *

transduction of B cell receptor.5 Moreover, the increase of core fucosylation and overexpression of FUT8 have been observed in liver, lung, and breast cancers, among others.6−11 FUT8 is up-regulated in epithelial-mesenchymal transition (EMT), and knock-down of FUT8 decreases the metastatic ability and growth rate of non-small-cell lung cancer.11 Core fucosylation also plays an important role in postnatal development. Knockout of Fut8 in mice causes an increase of death rate to 70% 3 days after birth, growth retardation, and lung emphysema, which result from dysfunction of transforming growth factor-beta (TGF-β) receptor due to the lack of core fucosylation.12 Together these studies demonstrate that core fucosylation is pivotal in various biological functions. The acceptor specificity of FUT8 has been studied previously with fluorescence- or radio-labeled oligosaccharides. The GlcNAc-terminated bi-antennary glycan is a well-known substrate of FUT8.1,13−15 However, because of the technical limitation, the substrate specificity of FUT8 toward tri- and tetra-antennary glycans is not well-understood. In order to examine the specificity of FUT8, we generated different forms of glycopeptides from sialoglycopeptides (SGP) 16 by treatment with glycosidases and glycosyltransferases (Supporting Information, Figure S1). Desialylated SGP (A2G2-SGP), desialylated and degalactoylatedi.e., GlcNAc-terminatedSGP (A2-SGP), and SGP with only the N-glycan core (Man3GlcNAc2, abbreviated as M3-SGP) were generated by sequential treatments with sialidase, galactosidase, and hexosaminidase. Mono-GlcNAc-SGP (N-SGP) was generated by treatment with endoglycosidase (Endo) M. A2-SGP was further glycosylated by recombinant N-acetylglucosaminyltransferases to generate the glycopeptides with tri-antennary GlcNAc-terminated N-glycans, A3(2,4,2)-SGP (via GnT-IVa), A3(2,2,6)-SGP (via GnT-V), and bisecting-SGP (A2B-SGP, via GnTIII). The glycopeptide with tetra-antennary GlcNActerminated N-glycan (A4-SGP) was synthesized from A2-SGP by GnT-IVa and GnT-V. To obtain homogeneous glycopeptides, all enzyme-treated glycopeptides were purified by C18 RP-HPLC (Figure S2). The molecular weights of purified glycopeptides were further confirmed by MALDI-TOF MS, and these modified SGPs were used to examine the specificity and activity of FUT8.

ABSTRACT: The core fucosylation of N-glycans on glycoproteins is catalyzed by fucosyltransferase 8 (FUT8) in mammalian cells and is involved in various biological functions, such as protein function, cancer progression, and postnatal development. The substrate specificity of FUT8 toward bi-antennary N-glycans has been reported, but it is unclear with regard to tri-antennary and tetraantennary glycans. Here, we examined the specificity and activity of human FUT8 toward tri- and tetra-antennary Nglycans in the forms of glycopeptides. We found that the tri-antennary glycan [A3(2,4,2) type] terminated with Nacetylglucosamine (GlcNAc), which is generated by Nacetylglucosaminyltransferase (GnT)-IV, is a good substrate for FUT8, but the A3(2,2,6) type of tri-antennary glycan, generated by GnT-V, is not a substrate for FUT8. We also observed that core fucosylation reduced the activity of GnT-IV toward the bi-antennary glycan. Examining the correlation between the types of N-glycans and the expression levels of FUT8, GnT-IV, and GnT-V in cells revealed that these glycosyltransferases, particularly GnT-IV, play important roles in directing the branching and core fucosylation of N-glycans in vivo. This study thus provides insights into the interplay among FUT8, GnT-IV, and GnT-V in N-linked glycosylation during the assembly of glycoproteins.

C

ore fucosylation is the glycosylation process by which a fucose is transferred from GDP-fucose to the innermost N-acetylglucosamine (GlcNAc) residue of N-linked glycans.1 In mammalian cells, fucosyltransferase 8 (FUT8) is the only enzyme responsible for the core fucosylation via addition of an α-1,6-linked fucose to N-glycans. Many glycoproteins with N-linked glycans are corefucosylated by FUT8, but understanding the effect of core fucosylation on glycoproteins has been a challenge, mainly due to the lack of tools available to prepare homogeneous glycoprotein substrates for the study. Core fucosylation on some glycoproteins has been proven to influence protein functions. For example, depletion of core fucose on the Fc region of therapeutic IgG1 improves the binding avidity toward Fcγ receptor IIIa and enhances antibody-dependent cellmediated cytotoxicity (ADCC).2−4 Core fucosylation also regulates the ability of antigen recognition and signal © 2017 American Chemical Society

Received: April 13, 2017 Published: July 5, 2017 9431

DOI: 10.1021/jacs.7b03729 J. Am. Chem. Soc. 2017, 139, 9431−9434

Communication

Journal of the American Chemical Society The specificity and activity of FUT8 toward glycopeptides were measured by real-time GDP/NADH-coupled assay17,18 by mixing with 100 μM SGP (Figure 1), and the reaction product

Figure 2. Effect of core fucosylation on the glycosyltransferase activity of GnT-III, -IVa, and -V. The effects of core fucosylation on the enzymatic activity of GnT-III (A), GnT-IVa (B), and GnT-V (C) were examined individually by incubating each enzyme with a mixture of A2-SGP and A2F-SGP. The percentage of product generated was determined by MALDI-TOF MS and calibrated by its corresponding standard curve (Figure S4). Data are shown as mean ± SD (n = 3). P value was calculated by t test (**p < 0.01).

Figure 1. Substrate specificity of FUT8. FUT8 was mixed with each glycopeptide (100 μM), and the activity was determined by GDP/ NADH coupled assay. A2-SGP and A3(2,4,2)-SGP are excellent substrates for FUT8. Error bars represent SD of three biological replicates.

Previous studies13,21,23 and our results suggest that FUT8, GnT-IV, and GnT-V can shape the types of N-glycans in cells. We next examined the correlation between glycan structures and the gene expression of Fut8, GnT-IVa, and GnT-V in cancer cells (Figure 3). Glycans were extracted from two breast cancer

was confirmed by MALDI-TOF MS analysis (Figure S3). The result showed that FUT8 had a high activity toward A2-SGP, consistent with previous studies.13,14 Interestingly, of the triand tetra-antennary SGPs, FUT8 showed high activity toward A3(2,4,2)-SGP, but not A3(2,2,6)-SGP, A2B-SGP, or A4-SGP, indicating that FUT8 had a strict substrate specificity. This is the first time the differential specificity of FUT8 toward two types of tri-antennary glycans has been detected. Further study revealed that FUT8 had comparable enzyme kinetics toward A2-SGP (kcat = 57.63 ± 5.9 s−1, Km = 75.52 ± 14.15 μM) and A3(2,4,2)-SGP (kcat = 68.58 ± 12.99 s−1, Km = 114.4 ± 34.58 μM). As all FUT8, GnT-III, GnT-IV, and GnT-V use GlcNActerminated bi-antennary glycan as acceptor substrate,19 the interplay among these enzymes determines the forms of glycans, e.g., the bisecting modification by GnT-III makes the glycan no longer a substrate for FUT8 and GnT-V.20−22 To investigate the preferences of GnTs toward glycans, a mixture of substrates with the same amount of A2-SGP and corefucosylated A2-SGP (A2F-SGP) was incubated with recombinant GnT-III, GnT-IVa, or GnT-V, and the products were examined by MALDI-TOF MS at different time points (Figure S5) . The percentage of product conversion was calculated on the basis of the peak area of product versus the sum of peak areas of precursor and product (Figure 2). GnT-III and GnT-V showed no difference in converting A2-SGP and A2F-SGP to the bisecting and tri-antennary glycopeptides (Figure 2A,C), respectively, indicating that core fucosylation was well-tolerated by GnT-III and GnT-V. In contrast, the activity of GnT-IVa, the enzyme that mediates the biosynthesis of tri-antennary Nglycans by adding a 4-linked GlcNAc onto the 3-mannose arm, was significantly suppressed by the presence of core fucosylation (Figure 2B). Therefore, GnT-IVa preferred A2-SGP as acceptor substrate over A2F-SGP. This suggests that the triantennary glycans can be biased to A3(2,2,6) when most of the bi-antennary glycans are core-fucosylated.

Figure 3. Correlation of the glycan profile and gene expression of glycosyltransferases in MCF7, MB231, CL1−5, and CL1−5 Fut8-KD. (A) Glycomic analysis by LC-MS. The regions with darker color indicate the proportions of core fucosylation. (B) The RNA expression of glycosyltransferases is measured by real-time PCR. (C) Percentage of overall core fucosylation in cells. The percentage of individual glycan structure is shown in Supporting Information, Table S1.

cell lines, MCF7 and MDA-MB-231 (abbreviated as MB231), and lung cancer cell lines CL1−5 and CL1−5 with Fut8 knocked down (CL1−5 Fut8-KD).11 The glycan profiles were then analyzed by LC-MS after the removal of sialic acids and galactoses by acid hydrolysis and subsequent galactosidase treatment, respectively (Figures 3A and S6), while the gene expression of Fut8, GnT-IVa, and GnT-V was measured by real9432

DOI: 10.1021/jacs.7b03729 J. Am. Chem. Soc. 2017, 139, 9431−9434

Communication

Journal of the American Chemical Society

core fucosylation interferes with the catalysis by GnT-IVa. The substrate specificity of FUT8 reported previously23,24 and in this study are thus summarized in Figure 4. Here we expand and clarify the substrate specificity of FUT8 and the pathway toward tri-antennary N-glycans with core fucosylation synthesized by GnT-IV, GnTV, and FUT8 and their interplay. Given the increasing importance of protein glycosylation in disease progression and human health, understanding the interplay of various glycosyltransferases should provide new insights into the regulation of glycosylation machinery and the development of desirable glycoforms.

time PCR (Figure 3B). Our results showed that there was a significant decrease but not complete inhibiton of core fucosylation in CL1−5 Fut8-KD compared to the parent CL1−5 (60% vs 89%, Figure 3C), even though expression of the Fut8 gene was >90% suppressed in CL1−5 Fut8-KD. The observation that knock-down of Fut8 caused a shift from biantennary to A3(2,4,2) glycan could be supported by the result in Figure 2B, where it is shown that GnT-IVa preferred biantennary glycan without core fucosylation as substrate. It is noted that, in all these cell lines examined, A3(2,2,6) was 100% core-fucosylated. Since A3(2,2,6) is not a substrate for FUT8, A3(2,2,6)F could only be synthesized from A2F. Among these cells examined, MCF7 had the highest level of A3(2,4,2) and the lowest level of A2 glycan. This could be due to the high expression level of GnT-IVa (Figure 3B): GnT-IVa catalyzes the synthesis of A3(2,4,2) from A2, and it also facilitates the synthesis of A4 from A3(2,2,6). It is interesting that MCF7 had a higher level of Fut8 expression than MB231, but the level of core fucosylation in MCF7 is lower than that in MB231 (29% vs 56%, Figure 3C). Again, this could be due to the uniquely high expression of GnT-IVa in MCF7: GnT-IVa efficiently converted A2 to A3(2,4,2) and in turn lowered the concentration of A2 to limit core fucosylation. The results implied that GnT-IVa plays an essential role in directing the branching and core fucosylation of N-glycans. It has been reported that the GlcNAc on the α3-mannose (Man) arm (catalyzed by GnT-I) of bi-antennary glycan is strictly required for core fucosylation, while the modification on the α6-Man arm does not affect core fucosylation.24 The authors also found that glycan elongation or chemical modification of the GlcNAc on α3-Man suppresses core fucosylation by FUT8, while glycan elongation on the GlcNAc on α6-Man allows core fucosylation by FUT8. Here, we show that the addition of β1,4-linked GlcNAc to the α3-Man arm was tolerated for core fucosylation, but the addition of β1,6linked GlcNAc to the α6-Man was not. Together our results indicate more precise substrate specificity of FUT8. However, core fucosylation can still occur in Man5 N-glycans in cells (Figure 4).25 Moreover, we also examined the effect of core fucosylation on the branching type of N-glycans and found that

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b03729. Supplementary data, experimental procedures, Figures S1−S6, and Table S1 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Chi-Huey Wong: 0000-0002-9961-7865 Present Address ‡

C.-Y.C.: CHO Pharma Inc., Park Street, Nangang, Taipei 115, Taiwan Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Hsin-Yu Lee for providing GDP-fucose, Dr. Tsung-I Tsai for providing Endo-M enzyme, and Ya-Ping Lin in Mass Spectrometry Core Facility, Genomic Research Center, for glycan extraction. This work was supported by the Genomics Research Center, Academia Sinica, Taiwan.

■

REFERENCES

(1) Yanagidani, S.; Uozumi, N.; Ihara, Y.; Miyoshi, E.; Yamaguchi, N.; Taniguchi, N. J. Biochem. 1997, 121, 626. (2) Okazaki, A.; Shoji-Hosaka, E.; Nakamura, K.; Wakitani, M.; Uchida, K.; Kakita, S.; Tsumoto, K.; Kumagai, I.; Shitara, K. J. Mol. Biol. 2004, 336, 1239. (3) Shinkawa, T.; Nakamura, K.; Yamane, N.; Shoji-Hosaka, E.; Kanda, Y.; Sakurada, M.; Uchida, K.; Anazawa, H.; Satoh, M.; Yamasaki, M.; et al. J. Biol. Chem. 2003, 278, 3466. (4) Shields, R. L.; Lai, J.; Keck, R.; O’Connell, L. Y.; Hong, K.; Meng, Y. G.; Weikert, S. H.; Presta, L. G. J. Biol. Chem. 2002, 277, 26733. (5) Li, W.; Yu, R.; Ma, B.; Yang, Y.; Jiao, X.; Liu, Y.; Cao, H.; Dong, W.; Liu, L.; Ma, K.; et al. J. Immunol. 2015, 194, 2596. (6) Liu, Y.-C.; Yen, H.-Y.; Chen, C.-Y.; Chen, C.-H.; Cheng, P.-F.; Juan, Y.-H.; Chen, C.-H.; Khoo, K.-H.; Yu, C.-J.; Yang, P.-C.; et al. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 11332. (7) Muinelo-Romay, L.; Vázquez-Martín, C.; Villar-Portela, S.; Cuevas, E.; Gil-Martín, E.; Fernández-Briera, A. Int. J. Cancer 2008, 123, 641. (8) Takahashi, T.; Ikeda, Y.; Miyoshi, E.; Yaginuma, Y.; Ishikawa, M.; Taniguchi, N. Int. J. Cancer 2000, 88, 914. (9) Ito, Y.; Miyauchi, A.; Yoshida, H.; Uruno, T.; Nakano, K.; Takamura, Y.; Miya, A.; Kobayashi, K.; Yokozawa, T.; Matsuzuka, F.; et al. Cancer Lett. 2003, 200, 167. (10) Hutchinson, W. L.; Du, M. Q.; Johnson, P. J.; Williams, R. Hepatology 1991, 13, 683.

Figure 4. Summary of FUT8-catalyzed reaction and pathways to triantennary N-glycans containing core fucose. M5, M5G, A, and A2 glycans are known substrates for FUT8.23,24 We examined the activity of FUT8 toward tri-antennary N-glycans (green box): A3(2,4,2) glycan produced by GnT-IV was a good substrate for FUT8, while the A3(2,2,6) glycan generated by GnT-V was not. Red arrows indicate FUT8-catalyzed reaction. The asterisk indicates that the product is reported based on the detection in cells.25 9433

DOI: 10.1021/jacs.7b03729 J. Am. Chem. Soc. 2017, 139, 9431−9434

Communication

Journal of the American Chemical Society (11) Chen, C.-Y.; Jan, Y.-H.; Juan, Y.-H.; Yang, C.-J.; Huang, M.-S.; Yu, C.-J.; Yang, P.-C.; Hsiao, M.; Hsu, T.-L.; Wong, C.-H. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 630. (12) Wang, X.; Inoue, S.; Gu, J.; Miyoshi, E.; Noda, K.; Li, W.; Mizuno-Horikawa, Y.; Nakano, M.; Asahi, M.; Takahashi, M.; et al. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 15791. (13) Voynow, J.; Kaiser, R.; Scanlin, T.; Glick, M. J. Biol. Chem. 1991, 266, 21572. (14) Mollicone, R.; Moore, S. E.; Bovin, N.; Garcia-Rosasco, M.; Candelier, J.-J.; Martinez-Duncker, I.; Oriol, R. J. Biol. Chem. 2009, 284, 4723. (15) Uozumi, N.; Teshima, T.; Yamamoto, T.; Nishikawa, A.; Gao, Y.-E.; Miyoshi, E.; Gao, C.-X.; Noda, K.; Islam, K. N.; Ihara, Y.; et al. J. Biochem. 1996, 120, 385. (16) Seko, A.; Koketsu, M.; Nishizono, M.; Enoki, Y.; Ibrahim, H. R.; Juneja, L. R.; Kim, M.; Yamamoto, T. Biochim. Biophys. Acta, Gen. Subj. 1997, 1335, 23. (17) Murray, B. W.; Wittmann, V.; Burkart, M. D.; Hung, S.-C.; Wong, C.-H. Biochemistry 1997, 36, 823. (18) Gonzalo, P.; Sontag, B.; Guillot, D.; Reboud, J.-P. Anal. Biochem. 1995, 225, 178. (19) Kizuka, Y.; Taniguchi, N. Biomolecules 2016, 6, 25. (20) Schachter, H. Biochem. Cell Biol. 1986, 64, 163. (21) Gu, J.; Nishikawa, A.; Tsuruoka, N.; Ohno, M.; Yamaguchi, N.; Kangawa, K.; Taniguchi, N. J. Biochem. 1993, 113, 614. (22) Taniguchi, N.; Kizuka, Y. Adv. Cancer Res. 2015, 126, 11. (23) Oguri, S.; Minowa, M. T.; Ihara, Y.; Taniguchi, N.; Ikenaga, H.; Takeuchi, M. J. Biol. Chem. 1997, 272, 22721. (24) Calderon, A. D.; Liu, Y.; Li, X.; Wang, X.; Chen, X.; Li, L.; Wang, P. G. Org. Biomol. Chem. 2016, 14, 4027. (25) Yang, Q.; Wang, L.-X. J. Biol. Chem. 2016, 291, 11064.

9434

DOI: 10.1021/jacs.7b03729 J. Am. Chem. Soc. 2017, 139, 9431−9434