Studies on the inhibition of sialyl- and galactosyltransferases

Recent advances in the development of sialyltransferase inhibitors to control cancer metastasis: A comprehensive review

Metastasis accounts for the majority of cancer-associated mortalities, representing a huge health and economic burden. One of the mechanisms that enables metastasis is hypersialylation, characterized by an overabundance of sialylated glycans on the tumor surface, which leads to repulsion and detachment of cells from the original tumor. Once the tumor cells are mobilized, sialylated glycans hijack the natural killer T-cells through self-molecular mimicry and activatea downstream cascade of molecular events that result in inhibition of cytotoxicity and inflammatory responses against cancer cells, ultimately leading to immune evasion. Sialylation is mediated by a family of enzymes known as sialyltransferases (STs), which catalyse the transfer of sialic acid residue from the donor, CMP-sialic acid, onto the terminal end of an acceptor such as N-acetylgalactosamine on the cell-surface. Upregulation of STs increases tumor hypersialylation by up to 60% which is considered a distinctive hallmark of several types of cancers such as pancreatic, breast, and ovarian cancer. Therefore, inhibiting STs has emerged as a potential strategy to prevent metastasis. In this comprehensive review, we discuss the recent advances in designing novel sialyltransferase inhibitors using ligand-based drug design and high-throughput screening of natural and synthetic entities, emphasizing the most successful approaches. We analyse the limitations and challenges of designing selective, potent, and cell-permeable ST inhibitors that hindered further development of ST inhibitors into clinical trials. We conclude by analysing emerging opportunities, including advanced delivery methods which further increase the potential of these inhibitors to enrich the clinics with novel therapeutics to combat metastasis.

Exploration of the Sialic Acid World

Sialic acids are cytoprotectors, mainly localized on the surface of cell membranes with multiple and outstanding cell biological functions. The history of their structural analysis, occurrence, and functions is fascinating and described in this review. Reports from different researchers on apparently similar substances from a variety of biological materials led to the identification of a 9-carbon monosaccharide, which in 1957 was designated "sialic acid." The most frequently occurring member of the sialic acid family is N-acetylneuraminic acid, followed by N-glycolylneuraminic acid and O-acetylated derivatives, and up to now over about 80 neuraminic acid derivatives have been described. They appeared first in the animal kingdom, ranging from echinoderms up to higher animals, in many microorganisms, and are also expressed in insects, but are absent in higher plants. Sialic acids are masks and ligands and play as such dual roles in biology. Their involvement in immunology and tumor biology, as well as in hereditary diseases, cannot be underestimated. N-Glycolylneuraminic acid is very special, as this sugar cannot be expressed by humans, but is a xenoantigen with pathogenetic potential. Sialidases (neuraminidases), which liberate sialic acids from cellular compounds, had been known from very early on from studies with influenza viruses. Sialyltransferases, which are responsible for the sialylation of glycans and elongation of polysialic acids, are studied because of their significance in development and, for instance, in cancer. As more information about the functions in health and disease is acquired, the use of sialic acids in the treatment of diseases is also envisaged.

Two N-terminally truncated variants of human β-galactoside α2,6 sialyltransferase I with distinct properties for in vitro protein glycosylation

Sialic acid groups of protein N-glycans are important determinants of biological activity. Exposed at the end of the glycan chain, they are potential targets for glycan remodeling. Sialyltransferases (STs; EC 2.4.99) are the enzymes that catalyze the sialic acid transfer from a CMP-activated donor on to a carbohydrate acceptor in vivo. Recombinant expression of the full-length human β-galactoside α2,6 sialyltransferase I (ST6Gal-I) was hampered and therefore variants with truncated N-termini were investigated. We report on the distinct properties of two N-terminally truncated versions of ST6Gal-I, namely Δ89ST6Gal-I and Δ108ST6Gal-I, which were successfully expressed in human embryonic kidney cells. The different properties of these enzymes result most probably from the loss of interactions from helix α1 in the Δ108ST6Gal-I variant, which plays a role in acceptor substrate binding. The K_m for N-acetyl-d-lactosamine was 10-fold increased for Δ108ST6Gal-I (84 mM) as compared to Δ89ST6Gal-I (8.3 mM). The two enzyme variants constitute a suitable tool box for the terminal modification of N-glycans. While the enzyme Δ89ST6Gal-I exhibited both ST (di-sialylation) and sialidase activity on a monoclonal antibody, the enzyme Δ108ST6Gal-I showed only ST activity with specificity for mono-sialylation.

An efficient approach to the discovery of potent inhibitors against glycosyltransferases

We describe a standardized approach for searching potent and selective inhibitors of glycosyltransferases by high throughput quantitative MALDI-TOFMS-based screening of focused compound libraries constructed by 1,3-dipolar cycloaddition of the desired azidosugar nucleotides with various alkynes. An aminooxy-functionalized reagent with a stable isotope was conjugated with oligosaccharides to afford glycopeptides as acceptor substrates with improved ion sensitivity. Enhanced ionization potency of new substrates allowed for MALDI-TOFMS-based facile and quantitative analysis of enzymatic glycosylation in the presence of glycosyl donor substrates. A non-natural synthetic sugar nucleotide was identified to be the first highly specific inhibitor for rat recombinant alpha2,3-(N)-sialyltransferase (alpha2,3ST, IC(50) = 8.2 microM), while this compound was proved to become a favorable substrate for rat recombinant alpha2,6-(N)-sialyltransferase (alpha2,6ST, K(m) = 125 microM). Versatility of this strategy was demonstrated by identification of two selective inhibitors for human recombinant alpha1,3-fucosyltransferase V (alpha1,3-FucT, K(i) = 293 nM) and alpha1,6-fucosyltransferase VIII (alpha1,6-FucT, K(i) = 13.8 microM).

Expression of sialyltransferase activity on intact human neutrophils

Endogenous polymorphonuclear leukocyte (PMN)-associated sialidase activity enhances PMN adhesion to and migration across the endothelium through the removal of sialylated cell-surface residues. We tested the hypothesis that PMNs also express sialyltransferase (ST) activity that restores sialyl residues to the PMN surface. We developed a highly sensitive fluorometric assay to demonstrate that intact human PMNs can mediate and accept sialyl residue transfer. This ST activity is inhibited by a ST inhibitor, CMP, which also inhibits the transendothelial migration of PMNs in response to IL-8 in vitro and in vivo. We conclude that intact PMNs express sialidase and ST activities that permit rapid modulation of their surface sialylation and their ability to adhere to and migrate across the endothelium.

CMP substitutions preferentially inhibit polysialic acid synthesis

It is widely reported that derivatives of sugar moieties can be used to metabolically label cell surface carbohydrates or inhibit a particular glycosylation. However, few studies address the effect of substitution of the cytidylmonophosphate (CMP) portion on sialyltransferase activities. Here we first synthesized 2'-O-methyl CMP and 5-methyl CMP and then asked if these CMP derivatives are recognized by alpha2,3-sialyltransferases (ST3Gal-III and ST3Gal-IV), alpha2,6-sialyltransferase (ST6Gal-I), and alpha2,8-sialyltransferase (ST8Sia-II, ST8Sia-III, and ST8Sia-IV). We found that ST3Gal-III and ST3Gal-IV but not ST6Gal-I was inhibited by 2'-O-methyl CMP as potently as by CMP, while ST3Gal-III, ST3Gal-IV, and ST6Gal-I were moderately inhibited by 5-methyl CMP. Previously, it was reported that polysialyltransferase ST8Sia-II but not ST8Sia-IV was inhibited by CMP N-butylneuraminic acid. We found that ST8Sia-IV as well as ST8Sia-II and ST8Sia-III are inhibited by 2'-O-methyl CMP as robustly as by CMP and moderately by 5-methyl CMP. Moreover, the addition of CMP, 2'-O-methyl CMP, and 5-methyl CMP to the culture medium resulted in the decrease of polysialic acid expression on the cell surface and NCAM of Chinese hamster ovary cells. These results suggest that 2'-O-methyl CMP and 5-methyl CMP can be used to preferentially inhibit sialyltransferases, in particular, polysialyltransferases in vitro and in vivo. Such inhibition may be useful to determine the function of a carbohydrate synthesized by a specific sialyltransferase such as polysialyltransferase.

Abnormal glycosylation with hypersialylated O-glycans in patients with Sialuria

Sialuria is an inborn error of metabolism characterized by coarse face, hepatomegaly and recurrent respiratory tract infections. The genetic defect in this disorder results in a loss of feedback control of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine-kinase by CMP-N-acetylneuraminic acid (CMP-NeuAc) resulting in a substantial overproduction of cytoplasmic free sialic acid. This study addresses fibroblast CMP-NeuAc levels and N- and O-glycan sialylation of serum proteins from Sialuria patients. CMP-NeuAc levels were measured with HPLC in fibroblasts. Isoelectric focusing (IEF) of serum transferrin and of apolipoprotein C-III (apoC-III) was performed on serum of three Sialuria patients. Isoforms of these proteins can be used as specific markers for the biosynthesis of N- and core 1 O-glycans. Furthermore, total N- and O-linked glycans from serum proteins were analyzed by HPLC. HPLC showed a clear overproduction of CMP-NeuAc in fibroblasts of a Sialuria patient. Minor changes were found for serum N-glycans and hypersialylation was found for core 1 O-glycans on serum apoC-III and on total serum O-glycans in Sialuria patients. HPLC showed an increased ratio of disialylated over monosialylated core 1 O-glycans. The hypersialylation of core 1 O-glycans is due to the increase of NeuAcalpha2,6-containing structures (mainly NeuAcalpha2-3Galbeta1-3[NeuAcalpha2-6]GalNAc). This may relate to KM differences between GalNAc-alpha2,6-sialyltransferase and alpha2,3-sialyltransferases. This is the first study demonstrating that the genetic defect in Sialuria results in a CMP-NeuAc overproduction. Subsequently, increased amounts of alpha2,6-linked NeuAc were found on serum core 1 O-glycans from Sialuria patients. N-glycosylation of serum proteins seems largely unaffected. Sialuria is the first metabolic disorder presenting with hypersialylated O-glycans.

Soyasaponin I decreases the expression of alpha2,3-linked sialic acid on the cell surface and suppresses the metastatic potential of B16F10 melanoma cells

The transfer of sialic acids to the non-reducing terminal positions on sugar chains of glycoconjugates is catalyzed by sialyltransferases (STs). Increased sialylation is correlated with oncogenic transformation and metastatic potential. ST inhibitors may be potentially valuable as anti-cancer and anti-metastatic agents. In this study, we evaluated the effects of soyasaponin I (Ssa I), a known inhibitor of STs, on tumor metastasis through studying a highly metastatic cancer cell line B16F10. Ssa I specifically inhibited the expression of alpha2,3-linked sialic acids without affecting other glycans on the B16F10 cell surface. We also found that Ssa I decreased the migratory ability of cells, enhanced cell adhesion to extracellular matrix proteins. Finally, a pulmonary metastasis assay demonstrated that alteration of glycosylation in this way significantly reduced the ability of tumor cells to distribute to the lungs of mice. Collectively, these findings suggested that alpha2,3-linked sialic acids may play an important role in metastasis potential of B16F10 cells.

Soyasaponin-I-modified invasive behavior of cancer by changing cell surface sialic acids

OBJECTIVE

Sialylation involving tumor formation and invasive behavior goes along with altered sialyltransferase (ST) activity. A potent ST inhibitor, soyasaponin I (SsaI), was discovered to selectively inhibit the cellular alpha2,3-sialyltranserase activity. In this study, we further test the effects of SsaI on modifying the metastatic and invasive behaviors of cancer cell lines.

METHODS

Nonmetastatic breast cancer cell line, MCF-7, and highly metastastic breast cancer cell line, MDA-MB-231, were used to investigate the effects of SsaI on tumor cells.

RESULTS

SsaI did not affect cell growth cycle and also failed to inhibit cell growth in this study (the concentration of SsaI < or=100 muM). SsaI was as predicted to successfully inhibit cellular alpha2,3-ST activity and depressed the dose-dependent tumor cell surface alpha2,3-sialic acid expression. In addition, different concentrations of SsaI did stimulate MCF-7 cell adhesion to collagen type I linearly and significantly enhanced cell adhesion to the Matrigel-matrix. Furthermore, SsaI significantly decreased MDA-MB-231 cell migration. Reverse transcriptase polymerase chain reaction for evaluating mRNA expression of ST3Gal I, III and IV showed that SsaI also down-regulated the expression of ST3Gal IV but did not affect the other two.

CONCLUSIONS

The results showed that SsaI was implicated in the invasive behavior of tumor cells, suggesting that altered alpha2,3-sialylation pathway played a crucial role in the adhesion and tumor metastases. SsaI is a good candidate for studying the biological roles of ST, and might provide a new preventive strategy in tumor metastasis.

Human xylosyltransferase I: functional and biochemical characterization of cysteine residues required for enzymic activity

XT-I (xylosyltransferase I) is the initial enzyme in the post-translational biosynthesis of glycosaminoglycan chains in proteoglycans. To gain insight into the structure-function relationship of the enzyme, a soluble active form of human XT-I was expressed in High Five insect cells with an apparent molecular mass of 90 kDa. Analysis of the electrophoretic mobility of the protein under non-reducing and reducing conditions indicated that soluble XT-I does not form homodimers through disulphide bridges. In addition, the role of the cysteine residues was investigated by site-directed mutagenesis combined with chemical modifications of XT-I by N-phenylmaleimide. Replacement of Cys471 or Cys574 with alanine led to a complete loss of catalytic activity, indicating the necessity of these residues for maintaining an active conformation of soluble recombinant XT-I by forming disulphide bonds. On the other hand, N-phenylmaleimide treatment showed no effect on wild-type XT-I but strongly inactivated the cysteine mutants in a dose-dependant manner, indicating that seven intramolecular disulphide bridges are formed in wild-type XT-I. The inhibitory effect of UDP on the XT-I activity of C561A (Cys561-->Ala) mutant enzyme was significantly reduced compared with all other tested cysteine mutants. In addition, we tested for binding to UDP-agarose beads. The inactive mutants revealed no significantly different nucleotide-binding properties. Our study demonstrates that recombinant XT-I is organized as a monomer with no free thiol groups and strongly suggests that the catalytic activity does not depend on the presence of free thiol groups, furthermore, we identified five cysteine residues which are critical for enzyme activity.

Analysis of the DXD motifs in human xylosyltransferase I required for enzyme activity

Human xylosyltransferase I (XT-I) is the initial enzyme involved in the biosynthesis of the glycosaminoglycan linker region in proteoglycans. Here, we tested the importance of the DXD motifs at positions 314-316 and 745-747 for enzyme activity and the nucleotide binding capacity of human XT-I. Mutations of the 314DED316 motif did not have any effect on enzyme activity, whereas alterations of the 745DWD747 motif resulted in reduced XT-I activity. Loss of function was observed after exchange of the highly conserved aspartic acid at position 745 with glycine. However, mutation of Asp745 to glutamic acid retained full enzyme activity, indicating the importance of an acidic amino acid at this position. Reduced substrate affinity was observed for mutants D747G (Km=6.9 microm) and D747E (Km=4.4 microm) in comparison with the wild-type enzyme (Km=0.9 microm). Changing the central tryptophan to a neutral, basic, or acidic amino acid resulted in a 6-fold lower Vmax, with Km values comparable with those of the wild-type enzyme. Despite the major effect of the DWD motif on XT-I activity, nucleotide binding was not abolished in the D745G and D747G mutants, as revealed by UDP-bead binding assays. Ki values for inhibition by UDP were determined to be 1.9-24.6 microm for the XT-I mutants. The properties of binding of XT-I to heparin-beads, the Ki constants for noncompetitive inhibition by heparin, and the activation by protamine were not altered by the generated mutations.

Synthesis and evaluation of phosphoramidate amino acid-based inhibitors of sialyltransferases

Several phosphoramidate analogues of CMP-N-acetylneuraminic acid were prepared for evaluation as inhibitors of alpha-2,3- and alpha-2,6-sialyltransferase. Central to the synthesis was the oxidative coupling of an amino acid ester with an H-phosphonate to construct the phosphoramidate linkage. All compounds synthesized were weak inhibitors of both of the sialyltransferases as determined by an HPLC-based inhibition assay.

Inhibition of CMP-sialic acid transport into Golgi vesicles by nucleoside monophosphates

We examined the interactions of nucleotides with the CMP-sialic acid transporter in order to better understand which features play a role in binding and to investigate the relationship between binding and subsequent transport. With respect to the sugar, the transporter requires a complete ribose ring for tight binding, and the 2'-ara hydrogen makes an important contact. The enzyme exhibits little specificity with respect to the 2'- and 3'-hydroxyls, as it tolerated substitutions ranging from fluorine to an azido group. In the base, the C4 amine and C2 carbonyl groups make important contacts, while the N3 nitrogen does not. However, adding a methyl group to N3 dramatically reduced binding, indicating that mass at this position sterically hinders binding. Adding a group at C5 had either no effect or slightly enhanced binding. To determine if the transporter recognizes these CMP analogues as substrates, we assayed them for their ability to trans stimulate CMP-sialic acid import. These data suggest that the enzyme transports a wide variety of NMPs, and the rate of transport is inversely proportional to the K(I) of the analogue. The importance of our findings for understanding the specificities of the different nucleotide-sugar tranlocators and the design of novel glycosylation inhibitors are discussed.

Achievements and challenges of sialic acid research

Slalic acids are one of the most important molecules of life, since they occupy the terminal position on macromolecules and cell membranes and are involved in many biological and pathological phenomena. The structures of sialic acids, comprising a family of over 40 neuraminic acid derivatives, have been elucidated. However, many aspects of the regulation of their metabolism at the enzyme and gene levels, as well as of their functions remain mysterious. Sialic acids play a dual role, not only are they indispensable for the protection to and adaptation of life, but are also utilised by life-threatening infectious microorganisms. In this article the present state of knowledge in sialobiology, with an emphasis on my personal experience in this research area, is outlined including a discussion of necessary future work in this fascinating field of cell biology.

Use of an altered sugar-nucleotide to unmask the transition state for alpha(2-->6) sialyltransferase

Rat liver alpha(2-->6) sialyltransferase catalyzes the formation of a glycosidic bond between N-acetylneuraminic acid and the 6-hydroxyl group of a galactose residue at the nonreducing terminus of an oligosaccharide. This reaction has been investigated through the use of the novel sugar-nucleotide donor substrate UMP-NeuAc. A series of UMP-NeuAc radioisotopomers were prepared by chemical deamination of the corresponding CMP-NeuAc precursors. Kinetic isotope effects (KIEs) on V/K were measured using mixtures of radiolabeled UMP-NeuAc's as the donor substrate and N-acetyllactosamine as the acceptor. The secondary beta-(2)H KIE was 1.218 +/- 0.010, and the primary (14)C KIE was 1.030 +/- 0.010. A large inverse (3)H binding isotope effect of 0.944 +/- 0.010 was measured at the terminal carbon of the NeuAc glycerol side chain. These KIEs observed using UMP-NeuAc are much larger than those previously measured with CMP-NeuAc [Bruner, M., and Horenstein, B. A. (1998) Biochemistry 37, 289-297]. Solvent deuterium isotope effects of 1.3 and 2.6 on V/K and V(max) were observed with CMP-NeuAc as the donor, and it is revealing that these isotope effects vanished with use of the slow donor substrate UMP-NeuAc. Bell-shaped pH versus rate profiles were observed for V(max) (pK(a) values = 5.5, 9.0) and V/K(UMP)(-)(NeuAc) (pK(a)values = 6.2, 9.0). The results are considered in terms of a mechanism involving an isotopically sensitive conformational change which is independent of the glycosyl transfer step. The isotope effects reveal that the enzyme-bound transition state bears considerable charge on the N-acetylneuraminic acid residue, and this and other features of this mechanism provide new directions for sialyltransferase inhibitor design.

New sialyltransferase inhibitors based on CMP-quinic acid: development of a new sialyltransferase assay

Quinic acid (4) was transformed into phosphitamides 6, 14, and 15, which could be readily linked to 5'-O-unprotected cytidine derivative 7; ensuing oxidation of the obtained phosphite triesters with tert-butylhydroperoxide furnished the corresponding phosphate triesters 8, 16, and 17, respectively. Hydrogenolytic debenzylation of the phosphate moiety, base catalysed removal of acetyl protective groups, and basic hydrolysis of the methylester of the quinic acid moiety furnished CMP-Neu5Ac analogues 1-3. In order to measure their inhibition of sialyltransferases, a nonradioactive sialyltransferase assay [employed for alpha(2-6)-sialyltransferase from rat liver (EC 2.4.99.1)] based on reversed-phase HPLC separation of UV-labelled acceptor 20 (p-nitrophenyl glycoside of N-acetyllactosamine) from the UV-labelled product 21 (p-nitrophenyl glycoside of sialyl alpha(2-6')-N-acetyllactosamine) and p-nitrophenylalanine as internal standard was developed. The assay reproduced the reported K(M) values for CMP-Neu5Ac and N-acetyllactosamine and the Ki values for CDP. 1 and 2 turned out to be potent sialyltransferase inhibitors.

Isotope trapping and kinetic isotope effect studies of rat liver alpha-(2-->6)-sialyltransferase

A mechanistic study of rat liver alpha-(2-->6) sialyltransferase (ST) is presented that includes isotope trapping experiments and kinetic isotope effects on V/K for the ST-catalyzed reaction of isotopically labeled CMP-N-acetylneuraminate and N-acetyllactosamine. The isotope trapping experiments confirmed that the kinetic mechanism is steady-state random, and further analysis indicated that for this sialyltransferase the experimentally observed isotope trapping ratio (product trapped/substrate released) was equivalent to the commitment to catalysis, Cf, the quantity required to correct the kinetic isotope effects. Cf was found to range from 1.0 (at 1.6 mM LacNAc) to 1.7 (at 100 mM LacNAc). After correction for Cf, the isotope effects were as follows: secondary beta-dideuterium, 1.04-1. 05; anomeric carbon primary 14C, 1.000 +/- 0.004; a small 3H binding effect of 1.016 +/- 0.007 at C9; and a carboxylate carbon secondary 14C isotope effect of 0.998 +/- 0.004. This pattern of KIEs is quite different than observed for solvolysis of CMP-NeuAc [Horenstein, B. A., and Bruner, M. (1996) J. Am. Chem. Soc. 118, 10371-10379]. Based on the results of ab-initio modeling of isotope effects, a hypothesis is presented which reconciles the unusual pattern of KIEs on the basis of binding interactions at the carboxylate carbon.