Earticle

Glycosylation is highly sensitive to the biochemical environment and has been implicated in development and disease. Global glycan profiling of human serum based on mass spectrometry has already led to promising markers for several types of cancer and disease. However, there is no compatible tool to verify the accuracy of sample preparation. Current research and analysis of glycome, different expression methods were exist even same glycan composition. There is no standard reference. Because of these reasons, we have developed glycan map using mass spectrometry. In this study, we profiled a large of set of human sera (n=200) by MALDI-TOF and found 80 common glycan compositions. Glycan map was established based on the glycan correlation and structure connectivity. We observed 18 detected glycans regardless of individual health and environment effect. These glycans should be detected on profiling. Therefore, representative glycans can be applied to quality control reference of glycan profiling. A new glycome map was proposed to provide all possible glycan compositions and structure in human serum. By projecting serum glycans onto a glycome map, total serum N-glycan modifications can be easily characterized with high speed and high accuracy. This represents a new platform for the discovery of biomarkers resulting from variations in N-glycan biosynthesis during disease progression.

Sialic acids are widely distributed in nature as terminal sugars attached to glycoproteins and glycolipids. The biosynthesis of this molecule is important for activity of carbohydrate-binding proteins including lectins, antibodies, and receptors. Among sialyltransferases, β-Galactoside α2, 6-sialyltransferase (ST6Gal-I) has been responsible for α2, 6-sialylation of N-Glycan,an action that is highly correlated with colon cancer progression and metastasis. We have previously demonstrated that ST6 Gal-I-induced α2, 6 sialylation contributes to enhancement of cell migration and radio resistance of colon cancer. A number of studies have focused on the involvement of sialylation in tumorigenesis, but the mechanism underlying ST6Gal-I-induced cancer progression and the identity of enzyme substrates has received scant research attention. To provide further support for the relevance of ST6Gal-I in the malignancy of colon cancer, we prepared and characterized a ST6Gal-I-knockdown SW480 colorectal carcinoma cell line. We found that inhibition of ST6Gal-I expression increased cell proliferation and tumor growth in vitro and in vivo. An examination of the effect of sialylation on epidermal growth factor receptor (EGFR) activity and downstream signaling, which are highly correlated with cell proliferation, showed that the loss of ST6Gal-I augmented EGF-induced EGFR phosphorylation and activation of extracellular signal-regulated kinase (ERK) in colon cancer cells. Moreover, ST6Gal-I induced sialylationof both wild type and mutant EGFR. These studies provide the first demonstration that ST6Gal-I induces EGFR sialylation in human colon cancer cell lines. Importantly, the anticancer effect of the EGFR kinase inhibitor, gefitinib, was increased in ST6Gal-I-deficient colon cancer cells. In contrast, overexpression of ST6Gal I decreased the cytotoxic effect of gefitinib. These results suggest that sialylation of the EGFR affects EGF-mediated cell growth and down regulates sensitivityto gefitinib in colon cancer cells.

Free oligosaccharides in human milk (HMO) have multiple biological functions, for example, the growth of beneficial bacteria in the gut and the development of brain. And they also provides important property for the postnatal development of the immune system. Until now, HMO has studied through MALDI-TOF MS and nano-LC/Chip Q-TOF MS for profiling and qualitative analysis. On the other hand, UPLC-Triple quadrupole MS is five fold faster than nano-LC/Chip Q-TOF MS. In this study, we have profiled and quantified free oligosaccharides extracted from HMO. And UPLC-Triple quadrupole MS is well-suited quantitative analysis on HMO. Here, we were comparable to data of HMO using MALDI-TOF, nano-LC/Chip Q-TOF and UPLC-Triple Quadrupole for verifying the suitability of UPLC-Trople Quadrupole MS on HMO analysis. Extraction of free oligosaccharides was performed by following steps including the removal of lipids by Folch method and ethanol protein precipitation. To get rid of αand β isomers, free oligosaccharides were reduced by reductive amination. Extracted free oligosaccharides were further purified by solid phase extraction using a porous graphitized carbon cartridge. Purified and enriched free oligosaccharides were redissolved in water prior to mass spectrometry analysis. First, we obtained that 64 free oligosaccharides from MALDI-TOF MS and 110 free oligosaccharides from nano-LC/Chip Q-TOF MS. Top Fifteen of free oligosaccharides profiled by nano-LC/Chip Q-TOF MS were found also by UPLC-Triple quadrupole MS. This study can be applied as an analytical platform to extract, profile, and quantify free oligosaccharides by UPLC-Triple quadrupole MS in milk and dairy products.

Porous graphitized carbon (PGC) is mainly employed for effective isomer-specific separation and enrichment of the glycans based on size, polarity, and three-dimensional structure. However, PGC separates α and β anomers in an aldehyde which is produced by PNGase F treatment to release N-glycans, thus by making effective isomer separation difficult. Here, we have optimized glycan reducing conditions to get rid of α and β anomers by reducing aldehyde using a sodium borohydride. Glycans released from ribonuclease B (high–mannose type), Immunoglobulin (complex type), and fetuin (highly sialylated glycans) were observed by optimized reducing method. Since N-glycans reduction have different portion of α and β anomers depending on three types of glycans. Reduced reducing N-glycans have been analyzed by nano LC chip/Q-TOF mass spectrometry and the number and quantity of isomers before and after reducing were examined to optimize the reducing conditions. After releasing N-glycans using PNGase F, released glycans were purified by solid phase extraction (SPE) and reduced by sodium borohydride. Reduced glycans were futher purified and enriched by SPE. Reduced N-glycans were separated and quantified using nanoflow liquid chromatography on PGC chip-based device. Reducing condition was successfully optimized and accomplished on three types of glycans. Indeed, we found less peaks in chromatogram after reducing N-glycans which suggest the number of glycans isomers were decreased because of the removal of α and β anomers. For example, the ion at m/z 1462.5445 corresponding to [Hex]3[HexNAc]4[Fuc]1 which is the most abundant glycan in IgG has single peak in chromatogram after reducing glycans. Indeed, this glycan does not have any isomer, which is already reported in the literature. In the future work, detailed data process will be performed for elucidation of removing α and β anomers in ribonuclease B and fetuin. This method will be successfully applied to isomer-specific separation of the glycans on a PGC in biopharmaceutical glycoanalysis and glycosylation study.

There are two types of glycosylation, N-linked and O-linked, respectively. N-linked glycan (or N-glycan) is attached to the asparagines (Asn), which is usually released by N-glycosylation specific enzyme known as peptide-N-glycosidase F (PNGase F). O-linked glycan (or O-glycan) is attached to the either serine (Ser) or threonine (Thr), however, there is no enzyme for the universal release of O-glycans from glycoproteins. In general, O-glycans are chemically released by β-elimination, which often release N-glycans that are abundant in human fluids. Thus, O-glycan release still remains analytical challenge in the field of glycomics. Here, we have developed the method to release O-glycans selectively and specifically. Mucin are large glycoproteins that carry dense clusters of O-linked glycans and IgG are glycoproteins that attacted only N-glycans. N and O mixed glycans were prepared in the lab to exist both N-glycans and O-glycans together in a sample. We approached three methods to compare and determine the best method to release selective and specific O-glycans release. 1) Method A : O-glycoproteins were eluted in acetonitrile (ACN) with 0.1% trifluoroacetic acid (TFA) from the cartridge that N-linked glycans were already removed. Glycoproteins were further treated with sodium borohydride (NaBH4) to release O-glycans and then enriched by porous graphitized carbon (PGC). We found that O-glycans were eluted in 20% ACN/H2O. 2) Method B : We separated N-glycans released by PNGase F and glycoproteins in mixed samples using molecular weight cut off (MWCO) 10,000 Da spin-column with centrifugation because glycoproteins molecular weight is higher than N-glycans. Glycoproteins were further treated with NaBH4 to release O-glycans and then enriched by PGC. We found that O-glycans were eluted in 20% ACN/H2O. 3) Method C : Glycoproteins in mixed samples after PNGase F treatment are precipitated by cold 100% ethanol. Glycoproteins were further treated with NaBH4 to release O-glycans and then enriched by PGC. We found that O-glycans were eluted in 20% ACN/H2O. Purified O-glycans were analyzed using Matrix-assisted Laser Desorption/Ionization time-of-flight mass spectrometry (MALDI-TOF MS). Indeed, we found O-glycans such as [Hex]1[HexNAc]2, [Hex]2[HexNAc]4 and [Hex]2[HexNAc]3[Fuc]1. This suggests that O-glycans can be selectively and specifically released under the absence of high abundant N-glycans using our approach. In the future, we will analyze enriched O-linked glycans quantitatively by nano-LC/chip Q-TOF MS to compare three methods. This analytical platform can be widely used for O-glycan release and enrichment.

O-GlcNAcylation, the addition of β-N-acetylglucosamine(O-GlcNAc) to serine or threonine, is one of post-translational modification occurs on myriad proteins in nucleus and cytosol. It cycles on proteins with a time-scale similar to protein phosphorylation, and has surprisingly extensive cross talk with phosphorylation, where is serves as a nutrient/stress sensor to modulate signaling, transc ription, and cytoskeletal functions. O-GlcNAcylated proteins in Drosophila were analyzed using 2 D gel electrophoresis and MALDI/TOF-MS, and ATP synthase β was identified as a novel O-Gl cNAcylated protein in Drosophila SL2 cell. F1F0-ATPsynthase complex produces ATP from ADP in the presence of a proton gradient across mitochondrial membrane which is generated by electr on transport of the respiration, and β subunit has ATPase activity. By immunoprecipiation and i mmunoblotting, O-GlcNAcylation of ATP synthase β was confirmed. Interestingly, we found that O-GlcNAcylation of ATP synthase β is increased by nucleocytoplasmic O-GlcNAc transferase. H ence we will concentrate on demonstrating how the ATP synthase β is modified with O-GlcNAc and what the functional roles of O-GlcNAcylation on ATP synthase β are.

Most cancer cells produce energy by glycolysis followed by fermentation in the cytosol, even though they exist in the presence of oxygen. But the research about cancer metastasis has been limited to hypoxia condition until now. Interestingly it was reported that O-GlcNAc modification on protein was significantly increased under hypoglycemic condition. O-GlcNAcylation is one of the post-translational protein modification and it appears to be involved in many different cellular activities. And it was thought that O-GlcNAcylation is also related with EMT. In our study, we want to examine whether hypoglycemic condition can induce cancer metastasis and it is related to increased O-GlcNAc modification or not. So we focused on EMT to visualize cancer metastasis. We found that the amount of several EMT marker were changed under hypoglycemic condition in cancer cell. And one of them was O-GlcNAcylated. These findings tell us that hypoglycemic condition can be closely connected with cancer metastasis via O-GlcNAclation. So we suggest possibilities that hypoglycemic condition induced EMT in cancer cell and it related with increased O-GlcNAc modification.

O-GlcNAcase (OGA) is the enzyme that catalyzes the elimination of O-GlcNAc moiety on its substrates. It has two isoforms, full OGA and variant OGA that has alternative stop codon within intron10, which is not spliced-out. As a result, variant OGA lacks C-terminal HAT (histone acetyl transferase) domain and contains additional 15 amino acids. It is known that full OGA exists both in cytosol and nucleus, primarily to the cytoplasm, however, there are two opposite reports about the subcellular localization of two OGA isoforms by different research groups. The first group figured out that full OGA localizes in cytoplasm, whereas variant OGA exists within nucleus in 2001. However, recently the other group reported that full OGA localized diffusely throughout the nucleus and cytoplasm, whereas variant OGA resides with lipid droplets in cytosol. So these two finding motivates our laboratory to commence this study as to nucleocytoplasmic shuttling mechanism of two OGA isoforms. We also focus on the various post-translational modifications on OGAs. Full OGA should weigh 103 kDa, but it weighs approximately 130kDa in reality. So we speculate that various modifications might be related to OGA’s subcellular distribution. To verify and get the answers to several issues about OGAs, we first overexpressed two FLAG tagged-OGA isoforms in HeLa cells to trace the subcellular localization of them. Then we made three C-terminal domain deletion mutants to know the critical domain for its localization. And we also check modification occurred on OGAs by using immune-precipitation. Consequentially we can get schematic drawing as to OGA’s nucleocytoplasmic shuttling, and we expect that this research makes it easy to understand and approach O-GlcNAc modification.

Some portions of N-glycans are phosphorylated in Saccharomyces cerevisiae by the dual action of two proteins encoded by MNN4 (ScMNN4) and MNN6 (ScMNN6). It has been known that ScMnn4 protein (ScMnn4p) is a positive regulator of ScMnn6p, a real mannosyltransferase adding mannosylphosphate residues to glycan. On the other hand, YlMPO1, Yarrowia lypolytica homologue of ScMNN4, was reported to mediate mannosyl-phosphorylation of N- and O-glycans in Y. lypolytica without the involvement of MNN6 homologues (1). In the present study, we further tested whether YlMpo1p can display mannosyltransferase activity independently in S. cerevisiae without the help of either ScMnn4p or ScMnn6p. YlMPO1 gene was heterologously expressed in S. cerevisiae wild-type, ScMNN4 disruptant (ScMnn4Δ), ScMNN6 disruptant (ScMnn6Δ), ScMNN4 and ScMNN6 double disruptant (ScMnn4ΔMnn6Δ). N-glycan profiles of their cell wall mannoproteins, which were analyzed using HPLC and DNA sequencer, clearly showed that the expression of YlMPO1 greatly increases the amount of mannosyl-phosphorylated glycans in all these mutant strains as well as wild-type, indicating its independent mannosyltransferase activity. Furthermore, increased amounts of mannosyl-phosphorylated glycans were much higher upon the heterologous expression of YlMPO1 compared to recombinant expression of ScMNN4. The strategy of YlMPO1 expression to increase the phosphorylated glycans would be usefully employed for glyco-engineering of yeast to produce therapeutic enzymes containing mannose-6-phosphate.

The purpose of this study is to optimize ELISA protocol to quantify colorectal cancer antigen GA733 and Fc antibody complex protein (GA733-Fc) expressed in transgenic plant. To confirm whether GA733 was functionally fused to Fc, mAb CO17-1A recognizing GA733 was coated on the 96 well ELISA plate. Goat anti-human IgG conjugated to horseradish peroxidase was applied to detect the Fc fragment of human IgG. ELISA was conducted by variable conditions of coating, washing, blocking, and sample amount. In coating, the mammalian-derived mAb CO17-1A were diluted at 2.5, 5, 10 μg/ml (100 μl/well). In washing, 1X PBS as compared to 1X PBS-T (1X PBS+0.05% Tween 20). In blocking, two types of buffers, 5% skim milk and 3% BSA solutions were compared each others. The results showed that 3% BSA gived better blocking than 5% skim milk. For plant samples, we applied transgenic plant leaf extracts expressing antigen GA733-Fc with dilution factors 1/2, 1/5, 1/10 and 1/100. Taken together, the optimized ELISA conditions to quantify plant-drived GA733-Fc were 0.5 μg of mAb CO17-1A per well for coating 3% BSA for blocking buffer, and 1/5 leaf extract sample dilution factor.

Plants are considered to represent a promising expression system for production of valuable recombinant immunotherapeutic proteins. Plants as an expression system has several advantages: lack of animal pathogenic contaminants, low cost of production, and ease of agricultural scale-up compared to available systems. We have studied to enhance expression efficiency of immunotherapeutic recombinant protein GA733, colorectal cancer vaccine candidate with fusion of human IgG Fc fragment and KDEL, endoplasmic reticulum retention signal resulting in GA733-FcK. Furthermore, selectively harvesting plant tissues highly expressing GA733-FcK from entire plant biomass is essential to improve final productivity. Thus, in this study, we have established several transgenic plant lines to highly express GA733-FcK and investigated variable expression level of recombinant GA733-FcK proteins along with leaf and stem tissues. The mRNA and protein levels in top, middle, and bottom leaf or stem tissues were observed using real time PCR and immunoblot analyses. In addition, the glycan structures and DNA methylation patterns were studied with respect of aging of plant tissues.

Colorectal cancer is the third most commonly diagnosed cancer in the world. Monoclonal antibody (mAb) CO17-1A recognizes the tumor-associated antigen GA733-2, a cell surface glycoprotein highly expressed in colorectal carcinoma cells which is applicable for preventing and curing colorectal cancer. In this study, we tried to produce a new recombinant anti-colorectal cancer large single chain (lsc) mAb based on mAb CO17-1A in the baculovirus-insect cell protein expression system. Two kinds of recombinant lsc mAbs were generated where variable light chain (VL) and heavy chain (HC) of mAb CO17-1A were fused together by an interchain linker. The only difference between the two mAbs is based on fusion of an ER retention signal (KDEL) at its C-terminus of HC. Polymerase chain reaction analysis verified the presence of both recombinant genes in the bacmid for generating viral expression vectors in insect cells. Western blot confirmed the expression of lsc mAbs in baculovirus-infected insect cells. Cell enzyme linked immunosorbent assay (ELISA) showed that the mAbs from cell lysates bound to SW480 and SW620 human colorectal cancer cells. These results indicate that the baculovirus insect expression system can produce anti-colorectal lsc mAb recognizing human colorectal cancer cells.