Chemical Fucosylation of a Polysaccharide: A ... - ACS Publications

Article pubs.acs.org/Biomac

Chemical Fucosylation of a Polysaccharide: A Semisynthetic Access to Fucosylated Chondroitin Sulfate Antonio Laezza,† Alfonso Iadonisi,† Cristina De Castro,‡ Mario De Rosa,§ Chiara Schiraldi,§ Michelangelo Parrilli,∥ and Emiliano Bedini*,† †

Department of Chemical Sciences, University of Naples Federico II, Complesso Universitario Monte S.Angelo, via Cintia 4, I-80126 Napoli, Italy ‡ Department of Soil, Plant, Environmental, and Animal Production Sciences, University of Naples Federico II, via Università 100, I-80055 Portici, Italy § Department of Experimental Medicine, Second University of Naples, via de Crecchio 7, I-80138 Napoli, Italy ∥ Department of Biology, University of Naples Federico II, Complesso Universitario Monte S.Angelo, via Cintia 4, I-80126 Napoli, Italy S Supporting Information *

ABSTRACT: Chemical O-glycosylation of polysaccharides is an almost unexplored reaction. This is mainly due to the difficulties in derivatizing such complex biomacromolecules in a quantitative manner and with a fine control of the obtained structural parameters. In this work, chondroitin raw material from a microbial source was chemo- and regioselectively protected to give two polysaccharide intermediates, that acted in turn as glycosyl acceptors in fucosylation reactions. Further manipulations on the fucosylated polysaccharides, including multiple de-O-benzylation and sulfation, furnished for the first time nonanimal sourced fucosylated chondroitin sulfates (fCSs)−polysaccharides obtained so far exclusively from sea cucumbers (Echinoidea, Holothuroidea) and showing several very interesting biological activities. A semisynthetic fCS was characterized from a structural point of view by means of 2D-NMR techniques, and preliminarily assayed in an anticoagulant test.

■

INTRODUCTION

fucosylated chondroitin sulfates (fCSs) with different Fuc branches. In particular, a variation in Fuc and/or GalNAc sulfation pattern has been observed.3 fCSs have been shown to possess several interesting activities in biological events related to atherosclerosis, cellular growth, angiogenesis, inflammation, fibrosis, cancer metastasis, virus infection, hyperglycemia, and, above all, coagulation and thrombosis.2 It is noteworthy that most of these biological activities strictly require the presence of sulfated Fuc branches on CS backbone.4 The anticoagulant and antithrombotic activity of fCS seems to be driven by a serpin-dependent mechanism in which thrombin inhibition is mediated by both antithrombin (AT) and heparin cofactor II (HC-II).4c Nonetheless, a serpinindependent mechanism was proposed too, as fCS showed anticoagulant activity also when tested on AT- and HCII-free plasmas.5 Furthermore, the antithrombotic action of fCS was discovered to be retained even when it was orally taken up.6 These unique activities greatly support fCS as a strong candidate for a new anticoagulant and antithrombotic drug. Nonetheless, two drawbacks must be taken into account: the severe

Chondroitin sulfate (CS) is a biomacromolecule with a linear backbone constituted of glucuronic acid (GlcA) and N-acetylgalactosamine (GalNAc), linked together through alternating β1 → 3 and β-1 → 4 glycosidic bonds. The resulting 4)-β-GlcA-(1 → 3)-β-GalNAc-(1→ disaccharide repeating unit could be decorated with one or more sulfate groups to give different sulfation patterns.1 Sulfated L-fucose (Fuc) branches have been found α-glycosidically linked at position O-3 of GlcA units in CS polysaccharides extracted from sea cucumbers (Echinoidea, Holothuroidea) (Figure 1).2 Different varieties of sea cucumbers from the seawater of different geographical zones show

Received: May 13, 2015 Revised: June 12, 2015 Published: June 17, 2015

Figure 1. General structure of natural fCSs. © 2015 American Chemical Society

2237

DOI: 10.1021/acs.biomac.5b00640 Biomacromolecules 2015, 16, 2237−2245

Article

Biomacromolecules

were tested for glycosylation couplings with fucosyl donors under different reaction conditions, and then subjected to further chemical manipulations to give semisynthetic fCS, that was structurally characterized by 2D-NMR techniques and subjected to a preliminary test for anticoagulant activity.

regulations for animal-derived drugs and the high variability of the sulfation patterns in animal tissues, requiring a strict control of the sulfation level in fCS products. Indeed, the fraudulent addition of per-O-sulfated CS as a contaminant in some heparin lots induced strong allergic-type responses7 and caused a cluster of serious adverse events, including 149 deaths in US and Europe.8 In order to open a route toward the large-scale production of a potent, safe and highly controllable anticoagulant and antithrombotic fCS-based drug, the synthesis of some fCS diand trisaccharide fragments was very recently accomplished,9 but, to the best of our knowledge, their biological activity has not yet been reported. On the other hand, the anticoagulant activity of partially depolymerized fCS polysaccharides has been shown to be impaired with the decrease in molecular size10 and a polysaccharide composed of at least 6−8 repeating units seems to be necessary for the antithrombotic action.11 Since the total synthesis of such species would require a tremendous effort, a semisynthetic approach based on chemical modification of a natural, nonanimal sourced polysaccharide would be highly desirable. The production of highly pure chondroitin polysaccharide from microbial sources has recently been reported.12 Furthermore, the regioselective modification of chondroitin from large-scale, fed-batch fermentation of Escherichia coli K4 to give CSs with several different sulfation patterns was achieved in few, high yielding steps.13 Starting from the same chondroitin polysaccharide, a tailored sequence of regioselective steps including fucosylation and sulfation as key steps would open for the first time an access to semisynthetic, nonanimal sourced fCS polysaccharides. The chemical O-glycosylation of polysaccharides is an almost unexplored reaction. This is mainly due to their generally poor solubility in the aprotic solvents required for the glycosylation reaction. This could be circumvented by derivatizing the polysaccharide with suitable, apolar protecting groups regioselectively placed on hydroxyls not involved in the glycosylation. A plethora of protecting groups has been developed for the total synthesis of oligosaccharides and glycoconjugates; nonetheless their application on polysaccharides is rather limited, mainly because it is often not possible to accomplish the protection with both a quantitative degree of substitution (DS) and regioselectivity.14 To the best of our knowledge, the only reports on polysaccharide O-glycosylation concerns the reaction between suitably protected polysaccharide derivatives with glycosyl orthoester or glycosyl oxazoline donors under rather harsh conditions to give products possessing mono- or disaccharide branches 1,2-trans-O-linked exclusively to the primary positions of the natural polysaccharide backbone.15 Unfortunately, such reactions cannot be applied in our case, because α-Fuc branches are 1,2-cis-O-linked to secondary positions in fCSs. Furthermore, even if several chemical16 and enzymatic17 methods for fucosylating mono- and oligosaccharide derivatives have been developed, glycosylation reactions involving fucosyl and, more generally, deoxyhexosyl donors are known to be challenging, due to the lower electron-withdrawing effect caused by the absence of one (or more) hydroxyls.18 Indeed, this can sometimes increase too much the rate of leaving group release at the anomeric position, thus lowering the yield of conjugation, especially with poorly reactive glycosyl acceptors, as polysaccharides are. Here we report for the first time a study of the chemical fucosylation of a polysaccharide. In particular, suitably protected derivatives were prepared starting from chondroitin obtained by E. coli K4 fed-batch fermentation.12h Chondroitin derivatives

■

EXPERIMENTAL SECTION

General Methods. Commercial-grade reagents and solvents were used without further purification, except where otherwise indicated. The term “pure water” refers to water purified by a Millipore Milli-Q Gradient system. Centrifugations were performed with an Eppendorf Centrifuge 5804R instrument at 4 °C (4600g, 10 min). Dialyses were conducted on Spectra/Por 3.5 kDa cutoff membranes at 4 °C. Sizeexclusion chromatography was performed on a Bio-Gel P2 column (0.75 × 67.5 cm, Bio-Rad) using 50 mM ammonium acetate as a buffer at a flow rate of 0.2 mL/min. The column eluate was monitored continuously with a Knauer K-2310 refractive index refractometer. Freeze-drying was performed with a 5 Pascal Lio 5P 4K freeze-dryer. NMR spectra were recorded on a Bruker DRX-400 (1H: 400 MHz, 13C: 100 MHz) instrument or on a Bruker DRX-600 (1H: 600 MHz, 13C: 150 MHz) instrument equipped with a cryo probe, in D2O (acetone as internal standard, 1H: (CH3)2CO at δ 2.22 ppm; 13C:(CH3)2CO at δ 31.5 ppm) or DMSO-d6 (1H: CHD2SOCD3 at δ 2.49 ppm; 13C: CD3SOCD3 at δ 39.5 ppm). Double quantum-filtered phase-sensitive correlation spectroscopy (COSY), phase-sensitive nuclear Overhauser effect spectroscopy (NOESY), and total correlation spectroscopy (TOCSY) experiments were performed using spectral widths of either 6000 Hz in both dimensions, using data sets of 4096 × 256 points. Quadrature indirect dimensions were achieved through the States-TPPI method; spectra were processed applying an unshifted Qsine function to both dimensions, and the data matrix was zero-filled by a factor of 2 before Fourier transformation. TOCSY and NOESY mixing times were set to 120 and 200 ms, respectively. Heteronuclear single quantum correlation-distortionless enhancement by polarisation transfer (HSQC-DEPT) experiments were measured in the 1H-detected mode via single quantum coherence with proton decoupling in the 13C domain, using data sets of 2048 × 256 points and typically 32 increments. As for HSQC-TOCSY, data sets of 2048 × 256 points were used, with 100 increments; mixing time was set to 100 ms and spectra were transformed as indicated for homonuclear spectra; data matrix were doubled and linear prediction applied the new points by using 30 coefficients; Qsine functions were used as window function and values selected depended by the best resolution obtained. A Viscotek instrument (Malvern) was used to determine molecular mass data. The degree of substitution (DS) of polysaccharide intermediates and fCSs is attributed to disaccharide repeating units. Chondroitin Methyl Ester 1. Chondroitin sodium salt from E. coli O5:K4:H4 fed-batch fermentation (2.183 g, 5.445 mmol repeating unit, 89−94% purity as evaluated by NMR and capillary electrophoresis; weight-averaged molecular mass = 45.0 kDa and polydispersity = 1.40, as evaluated by high-performance size-exclusion chromatography combined with a triple detector array)12h,19 was dissolved in pure water (140 mL) and passed through a short Dowex 50 WX8 column (H+ form, 20− 50 mesh, approximately 450 cm3). Elution with pure water was continued until the pH of the eluate was neutral. Freeze-drying of the collected eluate gave chondroitin (2.010 g, 5.303 mmol), which was manually chopped and then suspended under Ar atmosphere in dry methanol (250 mL). After overnight stirring at room temperature (rt), a 0.58 M methanolic solution of acetyl chloride (124 mL, 72.4 mmol) was added via cannula, and stirring was continued overnight. The reaction was then quenched by neutralization with 1 M aqueous NaHCO3. The mixture was concentrated by rotoevaporation and then freeze-dried. The protocol was repeated to give, after dialysis and freeze-drying, chondroitin methyl ester 1 (1.567 g, 73%, DS = 0.92) as a white waxy solid. Chondroitin n-Dodecyl Ester 3. Polysaccharide 2 (0.962 g, 1.551 mmol repeating unit) was suspended in N,N-dimethylformamide (DMF; 32 mL) that was freshly dried over 4 Å MS. Tetrabutylammonium fluoride (TBAF; 3.59 mL, 12.40 mmol) was added to give a clear 2238


Article

Biomacromolecules solution that was then treated with n-dodecyl iodide (6.3 mL, 25.5 mmol). The solution was stirred at 80 °C overnight and then treated with diisopropyl ether (130 mL) to give a gummy precipitate, that was collected by centrifugation and then dried under vacuum overnight to afford an approximately 30:70 w/w mixture of 3 and TBAI/TBAF (2.456 g, DS = 1). Chondroitin Methyl Ester 4,6-Benzylidene 4. Chondroitin methyl ester 1 (0.598 g, 1.519 mmol repeating unit) was suspended in DMF (30 mL), that was freshly dried over 4 Å MS, and stirred at 80 °C for 2 h. α,α-Dimethoxytoluene (2.3 mL, 15.2 mmol) freshly dried over 4 Å MS, and then (+)-camphor-10-sulfonic acid (CSA; 88.3 mg, 0.380 mmol) were added. The mixture was stirred at 80 °C overnight; after that a clear, yellowish solution was obtained. It was cooled to rt and treated with diisopropyl ether (72 mL). The obtained white precipitate was collected by centrifugation and then dried under vacuum overnight. Derivative 4 was obtained as a yellowish amorphous solid, as a 74% w/w mixture with CSA and DMF (0.878 g, 89%, DS = 0.99). Chondroitin n-Dodecyl Ester 4,6-Benzylidene 5. Chondroitin n-dodecyl ester 3 (0.237 g approximately 30% w/w pure) was dissolved in DMF (12 mL) that was freshly dried over 4 Å MS and stirred at 80 °C for 2 h. α,α-Dimethoxytoluene (0.65 mL, 4.34 mmol) freshly dried over 4 Å MS and then a 0.2 M solution of CSA in freshly dried DMF (0.52 mL, 0.104 mmol) were added. The solution was stirred at 80 °C overnight. It was then cooled to rt and treated with water (40 mL) to give a white precipitate that was collected by centrifugation and then dried under vacuum overnight. A 86% w/w mixture of derivative 5 with TBAF and DMF (97.4 mg, 97% over two steps from 3, DS = 0.96) was obtained as a white amorphous solid. Chondroitin Methyl Ester 4,6-Benzylidene 2,3-Bis-trimethylsilylether 6. Polysaccharide 4 (63.2 mg, 74% w/w pure) was suspended in pyridine (4 mL). [(CH3)3Si]2NH (274 μL, 1.31 mmol) and (CH3)3SiCl (166 μL, 1.31 mmol) were then added. After overnight stirring at 50 °C, a clear solution was obtained. It was cooled to rt and treated with diisopropyl ether (16 mL) to give a yellowish precipitate, that was collected by centrifugation and then dried under vacuum overnight. A 52:48 w/w mixture of derivative 6 and pyridine (103 mg, 88%, DS = 2) was obtained. Alternatively, after overnight stirring at 50 °C, the crude reaction mixture was treated with water (16 mL) to afford, after centrifugation and overnight drying, pure polysaccharide 6 (50.3 mg, 93%, DS = 1.03) as a white precipitate. Glycosylation with Ethyl 2,3,4-Tri-O-benzyl-β-L-fucopyranosyl Thioglycoside 7. A mixture of polysaccharide 4 (or 5 or 6) (48.4 mg, 0.101 mmol) and fucosyl thioglycoside 7 (241 mg, 0.505 mmol) was coevaporated three times with dry toluene (2.5 mL). AW-300 4 Å molecular sieves and then CH2Cl2 (3.4 mL) and DMF (2.5 mL) (or tetrahydrofuran (THF)), which were freshly dried over 4 Å molecular sieves, were added to the mixture under argon atmosphere. NIS (125 mg, 0.550 mmol) and then TMSOTf (99 μL, 0.550 mmol) were added. The mixture was stirred at rt (or −20 °C) under argon atmosphere for 4 h, then the molecular sieves were eliminated by decantation. A few drops of triethylamine and then diisopropyl ether (30 mL) were added, and the mixture was stored at −28 °C overnight. The obtained yellowish precipitate was collected by centrifugation and then dried under vacuum overnight to afford derivative 9 (or 10) (94.6 mg). Glycosylation with 2,3,4-Tri-O-benzyl-L-fucopyranosyl NPhenyl-trifluoroacetimidate 8. Polysaccharide 4 (or 5) (98.1 mg, 0.204 mmol) and fucosyl donor 8 (617 mg, 1.02 mmol) were coevaporated three times with dry toluene (5.0 mL) in two separated round-bottomed flasks. AW-300 4 Å molecular sieves and then DMF (6.0 mL) (or THF), which was freshly dried over 4 Å molecular sieves, were added to the polysaccharide, whereas 4 Å molecular sieves and then CH2Cl2 (10.0 mL), that was freshly dried over 4 Å molecular sieves, were added to the donor. The mixture containing the polysaccharide was then treated with a 0.49 M solution of TMSOTf in CH2Cl2 (42 μL, 20.6 μmol), and then the mixture containing the donor was immediately added via cannula under argon atmosphere. The mixture was stirred at rt for 4 h, then the molecular sieves were eliminated by decantation. A few drops of triethylamine and then diisopropyl ether (30 mL) were added, and the mixture was stored at −28 °C overnight. The obtained white

precipitate was collected by centrifugation and then dried under vacuum overnight to afford derivative 9 (or 10) (153.9 mg). Acetylation. Derivative 9 (or 10) (88.4 mg) was suspended in CH3CN (2.0 mL), and then Ac2O (650 μL), triethylamine (275 μL), and 4-(dimethylamino)pyridine (DMAP; 4.8 mg) were added. After overnight stirring at rt, a clear solution was obtained. In the case of methyl ester derivatives (9 → 11), diisopropyl ether (15 mL) was added to give a white precipitate that was collected by centrifugation and then dried under vacuum overnight to afford derivative 11 (72.5 mg). In the case of dodecyl ester derivatives (10 → 12), no polysaccharide precipitation could be obtained with any solvent. Therefore, the crude mixture was concentrated to give a residue that was coevaporated several times with toluene. A brownish oil (381 mg) containing polysaccharide 13 was obtained. It was not further purified, but used in the following synthetic step as it was. Oxidative Cleavage of Benzyl and Benzylidene Protecting Groups. A suspension of polysaccharide 11 (or 12) (67.7 mg) in ethyl acetate (800 μL) was treated with a 0.27 M solution of NaBrO3 in pure water (800 μL). A 0.24 M solution of Na2S2O4 in pure water (750 μL) was added portionwise over a period of 10 min. The triphasic mixture was vigorously stirred at rt overnight under visible light irradiation. The yellowish solid was then collected by centrifugation to afford derivative 13 (or 14) (36.8 mg). Sulfation and Deprotection. Derivative 13 (or 14) (33.3 mg) was suspended in DMF (570 μL) that was freshly dried over 4 Å molecular sieves and then treated with a 1.17 M solution of pyridine−sulfur trioxide complex in freshly dried DMF (840 μL). After 2 h of stirring at 50 °C, the suspension turned into a clear solution. Stirring at 50 °C was continued overnight. A saturated NaCl solution in acetone (5 mL) was added. The obtained yellowish precipitate was collected by centrifugation and then dissolved in pure water (7.0 mL). The solution was treated with a 15% w/v NaOH solution to adjust pH to 13. The solution was stirred for 6 h at rt and then 1 M HCl was added until neutralization. Dialysis and subsequent freeze-drying yielded a slightly yellow solid, that was further purified by filtration through a Sep-pak C18 cartridge and then by size-exclusion chromatography. Freeze-drying of the fractions afforded fCS polysaccharide (35.2 mg) as a white waxy solid. Determination of Molecular Mass. The gel permeation chromatographic (GPC) system was equipped with an autosampler (consisting of an isocratic specific pump for GPC and an online solvent degaser) and a TDA 302 module (triple detector array) that includes a column oven and a triple detector consisting of an RI detector, a fourbridge viscometer and a LS detector. The latter consisted of a right-angle light scattering (LALS) detector that performed measurements of the incident beam with an optimum signal/noise ratio. The OmniSEC software program was used for the acquisition and analysis of the Viscotek data. The size exclusion chromatographic method employed two TSK-GEL GMPWXL columns (Tosoh Bioscience, Italy. Hydroxylated polymethacrylated base material, 100−1000 Å pore size, 13 μm mean particle size, 7.8 × 30.0 cm) in series that were preceded by a TSK-GEL guard column GMPWXL (Tosoh Bioscience, 12 μm mean particle size, 6.0 × 4.0 cm) and a mobile phase consisting of 0.1 M NaNO3 in bidistilled water; the mobile phase (pH 7.0) was deaerated during the use by the in-line degasser. The flow rate was 0.6 mL/min and the concentration of the samples analyzed ranged from 0.5 to 4.0 g/L to have a column load for each analysis (injection volume × sample concentration × intrinsic viscosity) of approximately 0.2−0.3 dL. The analyses were performed at 40 °C with a running time of 50 min. Universal calibration for the determination of instrumental costants (KRI, KV e KLS) was performed by using a poly(ethylene oxide) (PEO) standard (22 kDa PolyCAL, Viscotek). The acquisition and analysis of the Viscotek data were made using an advanced detector software package (OmniSEC). Anticoagulant Activity. AT-dependent antifactor Xa assay was performed by a kinetic colorimetric method using a commercial kit (Stachrom Heparin, Stago). Briefly, 100 μL of the sample were added to 100 μL of purified bovine antithrombin diluted in 1 μM Tris EDTA buffer (pH 8.4). The determination was carried out by adding 200 μL of factor Xa that was incubated for exactly 2 min at 37 °C. Then 200 μL of 2239


Article

Biomacromolecules Scheme 1. Transformation of Chondroitin into Polysaccharide Acceptors 4, 5, and 6

factor Xa chromogenic substrate were added, and the color obtained was stable for several hours. The absorbance was read at 405 nm against a blank and was inversely proportional to the concentration of the polysaccharide. A standard curve was constructed (r2 = 0.996) using dilutions of heparin standard from 0.1 to 0.5 IU/mL.

for 4 and 5, respectively) and benzylidene methine signals (δ 5.51 and 5.48 ppm for 4 and 5, respectively) was possible only for derivative 4 (DS = 0.99), because acetyl signal of derivative 5 was overlapped by other signals. HSQC-DEPT 2D-NMR integration (see Figure S4 in Supporting Information) of acetyl and benzylidene methine densities gave a DS of 0.96 for 5. Dodecyl ester derivative 5 showed a satisfying solubility (approximately 10 mg/mL) in CH 2Cl2 -ethereal solvent mixtures, that are typically used for α-stereoselective fucosylations. On the contrary, methyl ester 4 was not soluble at all in such solvents. For this reason, a further derivatization of polysaccharide 4 was attempted, by converting the 2,3-diol on GlcA units into its bis(trimethylsilyl) derivative. Under reaction conditions reported for trimethylsilylation of polysaccharides ((CH3)3SiCl, [(CH3)3Si]2NH in pyridine at rt),15d no reaction was observed, and derivative 4 was recovered unreacted. By performing the reaction at 50 °C overnight, followed by precipitation with diisopropyl ether, polysaccharide 6 could be obtained with a quantitative DS (2), as estimated from 1H NMR integration (see Figure S5 in the Supporting Information). Nonetheless, 6 was recovered not in pure form, but mixed with a non-negligible amount of pyridine, that could be detrimental for fucosylation reaction. Attempts to purify 6 by further precipitation or pyridine azeotropic distillation failed. Alternatively, polysaccharide 6 could be precipitated from its reaction mixture with water, but in this case a lower DS (1.03) was obtained (see Figure S6 in the Supporting Information), as expected due to trimethylsilyl (TMS) ether lability in aqueous mixtures. Chondroitin derivatives 4, 5, and 6 (DS = 1.03) were used as polysaccharide acceptors for fucosylation reactions. Known perO-benzylated L-fucosyl thioglycoside 723 and N-phenyl trifluoroacetimidate 816c (Figure 2) were selected as suitable glycosyl donors possessing a nonparticipating ether-type protecting

■

RESULTS AND DISCUSSION A fermentation broth of E. coli K4 obtained by a fed-batch process was subjected to microfiltration, protease treatment, diafiltration, and mild hydrolysis,12h to give 89−94% pure chondroitin, free of any toxic lipopolysaccharide contaminant, as evaluated by NMR analysis and capillary electrophoresis.19 Derivatization of chondroitin to obtain polysaccharide products soluble in aprotic solvents started with chemoselective esterification of the GlcA carboxylic acid. Methylation was accomplished by stirring a suspension of the polysaccharide in methanol in the presence of acetyl chloride (Scheme 1). The degree of substitution (DS) of derivative 1 could not be evaluated by comparison of the 1H NMR acetyl and methoxy signal (δ = 2.02 and 3.85 ppm, respectively) integrations, because the OCH3 signal overlapped with some carbinolic ones (see Table S1 and Figure S1 in the Supporting Information). This was circumvented by HSQC-DEPT 2D-NMR integration (see Figure S1 in the Supporting Information), assuming that the signals to be compared displayed similar 1JCH coupling constants and that a difference of around 5−8 Hz from the experimental set value did not cause a substantial variation of the integrated peak volumes. Furthermore, within a specific density set, the measured volumes scale with the 1JCH value used in the experiment, but their increase (or decrease) is affected to the same extent, so that the proportion between them is maintained even if their 1JCH is distant (i.e., 20 Hz) from the set experimental value.20 To obtain a DS close to 1 (0.92), the methylation reaction had to be conducted at rt overnight twice. Alternatively, chondroitin esterification with a long alkyl chain (n-dodecyl) was performed on tetrabutylammonium salt 221 under the reaction conditions recently optimized for alginate esterification.22 Thus, a solution of 2 in DMF was treated with n-dodecyl iodide in the presence of TBAF at 80 °C to afford chondroitin ester 3 with DS = 1. Both ester derivatives 1 and 3 were subjected to regioselective protection of the diol at O-4,6 positions of GalNAc units with a benzylidene ring, by treating them with α,α-dimethoxytoluene in DMF at 80 °C in the presence of CSA as catalyst. An evaluation of DS by integration of the 1H NMR acetyl (δ 1.77 and 1.73 ppm

Figure 2. Glycosyl donors used in the fucosylation reactions. 2240


Article

Biomacromolecules group at position O-2, as required for a 1,2-cis-stereoselectivity in the glycosylation reaction. Since DMF has been recently shown to actsimilarly to ethereal solventsas an excellent α-stereodirecting modulator in glycosylations (fucosylation included) with thioglycoside donors,24 the fucosylation reactions of dodecyl ester derivative 5 were conducted under homogeneous conditions in 5:3 v/v CH2Cl2-DMF or in 3:1 v/v THF-CH2Cl2 (Table 1, entries i−iii).

spectra as well as the coprecipitation of part of Fuc byproducts with the polysaccharide did not allow an exact evaluation of DS of fucosylation reactions (see Figures S7−S8 in Supporting Information). Therefore, glycosylation products 9-i, 9-iii, 10iv, 10-v, 10-vi, and 10-vii were subjected to further reactions in order to transform them into fCS polysaccharides and investigate their structural features. Thus, the six polysaccharides 9 and 10 were acetylated under homogeneous conditions with Ac2O in the presence of triethylamine and DMAP in CH3CN to protect GlcA hydroxyls that did not couple with fucosyl donors during the glycosylation step (Scheme 2). In order to have free hydroxyls that could be then sulfated, benzyl groups were then cleaved by oxidation with NaBrO3 and Na2S2O4 in an H2O-ethyl acetate mixture.25 This reaction caused also the oxidative opening of the benzylidene ring on GalNAc to either 4-O- or 6-O-benzoylated units,13e,25 thus giving derivatives 13 and 14 with three free hydroxyls on Fuc branches and a further one at position either 4 or 6 of GalNAc units randomly distributed on the polymer chain. Benzylidene and benzyl cleavage could be confirmed by the splitting and downfield shift of aromatic signals (δ 7.32 ppm for benzylidene and benzyl, and δ 7.50, 7.65, 7.94 ppm for the benzoate resulting from benzylidene, respectively) in the 1H NMR spectra of polysaccharides 13 and 14 (see Figures S11− S12 in the Supporting Information). fCS-i,-iii,-iv,-v,-vi were finally obtained by sulfation of the free alcohols with SO3· pyridine complex in DMF, followed by global acyl deprotection under alkaline hydrolytic conditions and purification of the obtained water-soluble polysaccharides by dialysis, filtration on a C-18 silica gel cartridge, and size exclusion chromatography in order to eliminate all the low molecular mass contaminants accumulated during the synthetic steps. The degree of fucosylation (DF) of the semisynthetic fCSs was evaluated by integrating the 1H NMR or HSQC-DEPT Fuc methyl and GalNAc acetyl signals (δH/C = 1.3/17.2 and 2.0/23.9 ppm, respectively) (Figures S13, S14, S18, S22, S26, S30; see Supporting Information). fCS-iv, fCS-v, and fCS-vi obtained from methyl ester acceptor 4 had a DF of 0.77, 0.32, and 1.15, respectively, whereas fCS-vi derived from silylated acceptor 6

Table 1. Glycosylation Reactions of Chondroitin Acceptors 4−6 with Fucosyl Donors 7 and 8 entry c

i iid iiie ivc vf vid viic

acceptor

donora

solvent

productb

5 5 5 4 4 4 6 (DS = 1.03)

7 8 8 7 7 8 7

3:1 v/v THF-CH2Cl2 3:1 v/v THF-CH2Cl2 5:3 v/v CH2Cl2-DMF 5:3 v/v CH2Cl2-DMF 5:3 v/v CH2Cl2-DMF 5:3 v/v CH2Cl2-DMF 1:1:1 v/v/v CH2Cl2-DMEDMF

9-i -9-iii 10-iv 10-v 10-vi 10-vii

a

Five equivalents with respect to the acceptor. bPostulated structures depicted in Scheme 2. cConditions: NIS (5.5 equiv), TMSOTf (5.5 equiv), AW-300 4 Å-MS, rt, 4 h. dConditions: TMSOTf (0.1 equiv), AW-300 4 Å-MS, rt, 4 h (inverse procedure). eConditions: TMSOTf (0.5 equiv), AW-300 4 Å-MS, rt, 4 h (inverse procedure). fConditions: NIS (5.5 equiv), TMSOTf (1.7 equiv), AW-300 4 Å-MS, −20 °C, 4 h.

Methyl ester 4 was glycosylated in 5:3 v/v CH2Cl2-DMF (entries iv−vi), whereas a ternary solvent mixture (1:1:1 v/v/v CH2Cl2DME-DMF) ensured homogeneous conditions for glycosylation of silylated derivative 6 (DS = 1.03) (entry vii). Polysaccharide products obtained by precipitation with suitable solvents from the crude glycosylation mixtures showed very complex 1H NMR spectra, except for the product of entry ii. In this case, chondroitin acceptor 5 was recovered unreacted, probably due to quenching of the low quantity of the catalyst by residual TBAF as impurity in the polysaccharide (see Experimental Section for details). In the other cases, the complexity of the 1H NMR

Scheme 2. Transformation of Glycosylation Products of Table 1 into fCSsa

a

Derivatives with the same Latin number were obtained during the same semisynthetic sequence (e.g., 9i → 11i → 13i → fCS-i) 2241


Article

Biomacromolecules

with residual n-dodecyl moieties. Since their β-(ω-1) methylene signals overlap with those of the Fuc methyls, the DF of fCS-i and fCS-iii had to be calculated by applying the following equation:

showed a very low DF (Table 2). The 1H NMR spectrum of fCS polysaccharides (fCS-i and fCS-iii) semisynthesized from Table 2. Yield and Structural Data of fCS-i,iii−vii fCS i iii iv v vi vii

yielda 56% 58% 98% 76% 76% 89%

DFb

α/βc

0.60 0.87 0.77 0.32f 1.15 0.14

e

n.d. n.d.e 2.9 (2.6) 2.3 (4.0) 2.6 (2.7) n.d.e

O-3-linked-: O-2-linked-:(O-4,6)linked-α-Fuc ratiod

DF =

e

n.d. n.d.e 40%: 22%: 38% 38%: 62%: 0% 33%: 24%: 43% n.d.e

I(CH3Fuc) −

20 I (CH3dodecy1) 3

I(CH3NHAc)

(1)

fCS-i and fCS-iii DF values were similar to those of fCS-iv and fCS-v ones; nonetheless the former polysaccharides had to be discarded due not only to the incomplete replacement of the free carboxylic acid moiety on GlcA units, but also to the lower mass recovery calculated from the polysaccharide acceptor over five steps (Table 2). A detailed 2D-NMR (COSY, TOCSY, NOESY, HSQCDEPT, and HSQC-TOCSY) analysis on fCS-iv, fCS-v, and fCSvi was then conducted, in order to investigate the regio- and stereochemistry of the Fuc branches as well as the sulfation pattern of these semisynthetic fCSs. Very similar 2D-NMR spectra for the three semisynthetic polysaccharides were obtained (Figures 3 and S19−S30; see Supporting Information). Here spectra of fCS-iv are presented. HSQC-DEPT spectrum of fCS-iv (Figure 3a) showed three densities (δH/C 5.64/98.1, 5.50/ 98.4, and 5.39/96.7 ppm, respectively) in the typical region for anomeric CH of α-linked sulfated Fuc units.3b,10a Thus, three different α-linked Fuc branches (indicated as α-FucI, α-FucII, and

a

Mass yield determined with respect to starting glycosyl acceptor (4− 6) over five steps. bDetermined by 1H NMR integration of Fuc methyl and GalNAc acetyl signals, except where differently indicated. c Estimated by 1H NMR integration according to eq 2 (in parentheses the values estimated by HSQC-DEPT integration). dDetermined by HSQC-DEPT integration of related anomeric signals. e Not determined. fDetermined by HSQC-DEPT integration of Fuc methyl and GalNAc acetyl signals.

dodecyl ester acceptor 5 showed a signal at δ 0.88 ppm attributable to methyl group of residual n-dodecyl chain (Figures S13 and S14; see Supporting Information). Reiteration of the hydrolysis reaction with LiOOH26 still gave a polysaccharide

Figure 3. Zoom of (a) HSQC-DEPT, (b) COSY, (c) TOCSY, and (d) NOESY NMR spectra (600 MHz, D2O, 298 K) of fCS-iv (densities enclosed in the dotted circles of HSQC-DEPT spectrum were integrated for Fuc branching site regio- and stereochemistry as well as GalNAc-4,6-sulfation assessment). 2242


Article

Biomacromolecules Table 3. 1H (Roman) and 13C NMR (Italic) Chemical Shift Attribution of fCS-iv (600 MHz, 298 K, D2O)a residueb α-Fuc2,3,4S -(1→3)-GlcA I

α-Fuc2,3,4SII-(1→2)-GlcA α-Fuc2,3,4S III-(1→4 or 6)-GalNAc GalNAc6S GalNAc4S α-Fuc2,3,4S III-(1→6)-GalNAc GlcA α-Fuc2,3,4SI-(1→3)-GlcA α-Fuc2,3,4SII-(1→2)-GlcA

1

2

3

4

5

6

other signals

5.64 98.1 5.50 98.4 5.39 96.7 4.55 102.2 4.55 102.2 n.d.c

4.53 73.6 4.51 73.6 4.51 73.6 4.02 52.2 4.02 52.2 4.02 52.2 3.33 73.6 3.57 73.1 3.54 80.3

4.61 73.4 4.51 73.6 4.51 73.6 3.82 81.4 3.59 76.9 n.d.c

4.94 80.3 4.87 80.3 4.91 80.3 4.08−4.14 67.9−68.9 4.75 77.6 n.d.c

4.94 67.4 4.22 67.7 n.d.c

1.33 17.2 1.31 17.2 n.d.c

--

3.53 74.9 3.73 82.3 3.69 76.1

3.73 81.3 n.d.c

3.84−3.96 73.3−73.6 3.81 75.8 3.84−3.96 73.3−73.6 3.71 77.5 3.71 77.5 3.71 77.5

4.21−4.32 68.2−68.7 3.79 62.4 4.15 67.4 --

Ac 1.96−2.05 23.9 Ac 1.96−2.05 23.9 Ac 1.96−2.05 23.9 --

--

--

--

--

4.48 105.6 4.48 105.6 4.70 102.4

3.85 81.4

---

Chemical shifts expressed in δ relative to internal acetone (1H: (CH3)2CO at δ = 2.22 ppm; 13C: (CH3)2CO at δ = 30.9 ppm). bIndicated by italic characters. cNot determinable.

a

α-FucIII, respectively) could be hypothesized to be present on the chondroitin backbone. Their sulfation pattern was evaluated by COSY and TOCSY spectra (Figure 3b,c), which allowed the assignment of the chemical shifts for H-2, H-3 and H-4 signals of α-Fuc branches. The values were very similar for the three units (δ 4.51−4.53, 4.51−4.61, and 4.87−4.94 ppm for H-2, H-3 and H-4, respectively; see Table 3). Their downfield shifts suggested that the three different α-linked Fuc branches were all 2,3,4-tri-Osulfated. This finding was in agreement with a quantitative cleavage of benzyl protecting groups on Fuc branches by oxidation with NaBrO3 and Na2S2O4 (12-iv → 14-iv, Scheme 2) to give a 2,3,4-triol, that was then quantitatively sulfated in the subsequent step. Since FucI, FucII, and FucIII presented the same sulfation pattern and stereochemistry of linkage, distinct sites of branching on the chondroitin backbone could be hypothesized to determine their differentiation. This could be investigated by means of NOESY spectrum (Figure 3d), that revealed that the FucI signal (δH 5.64 ppm) correlated with a density at δH 3.73 ppm, attributable to H-3 of some GlcA units by means of the other 2D-NMR spectra and literature data.3b,10a Similarly, FucII signal (δH 5.50 ppm) could be assigned to the anomeric CH of a branching unit linked at O-2 (δH/C 3.54/80.3 ppm) of some GlcA units. The presence of both GlcA-O-2 and GlcA-O-3 positions substituted with Fuc units is confirmed also by the correlation between Fuc methyl density (δH/C 1.31−1.33/17.2 ppm) and Fuc-CH-5 signals (δH/C 4.94/67.4 and 4.22/67.7 ppm) in COSY, TOCSY and HSQC-TOCSY spectra (see Figures S19, S20, and S23 in the Supporting Information). Indeed, a marked difference between the chemical shift of H-5 atom of O-2- and O-3-linked αFuc units has been very recently highlighted10a (δH 4.79−4.89 ppm for O-3 linked α-Fuc in natural fCSs or Lewis X;10a δH 4.16− 4.35 ppm for O-2 linked α-Fuc linked in A, B, and H(O) histoblood group antigens10a,27). NOESY spectrum showed two interresidue correlations for FucIII anomeric signal (δH 5.39 ppm), a major one at δ 4.21 ppm and a minor one at δ 3.95 ppm. The former could be associated with O-6 (δC 67.4 ppm) of some GalNAc units by means of HSQC-DEPT spectrum. Since the DS for benzylidenation reaction of acceptor 4 was nearly quantitative (Scheme 1), a partial TMSOTf-mediated cleavage of the

benzylidene protecting group during the fucosylation step may explain this finding. The minor NOESY correlation from FucIII anomeric signal could be not unambiguously assigned; nonetheless the existence of a benzylidene cleavage process during fucosylation suggested that Fuc branching at position O-4 of some GalNAc units could be also present. Relative integration of FucI, FucII, and FucIII HSQC-DEPT densities could return an estimate of relative Fuc branching degree at positions GlcA-O-3, GlcA-O-2, and GalNAc-(O-4,6), respectively (Table 2). Interestingly, in fCS-v spectra, no FucIII anomeric signal could be detected (see Figures S24−S26 in the Supporting Information). Indeed, this polysaccharide showed Fuc branching exclusively at GlcA-O-2 and GlcA-O-3 positions, due to the glycosylation reaction conducted at lower temperature (−20 °C), which prevented the TMSOTf-mediated cleavage of the benzylidene protecting group on GalNAc 4,6-diols. The α/β stereochemical ratio of Fuc branching could be not evaluated by integrating α- and β-Fuc anomeric signals, because the latter could be unambiguously detected in neither 1H NMR nor HSQC-DEPT spectra due to low intensity and overlapping with signals of α-Fuc28 or GlcA and GalNAc units. Therefore, an indirect, semiquantitative estimation of α/β Fuc ratio was done by evaluating β-Fuc branches amount as difference of Fuc methyl signal (comprising both α- and β-linked units) and α-anomeric signals integrations. The following equation was applied: α /β =

I(CH‐1 α‐Fuc) I(CH‐6 Fuc) 3

− I(CH‐1 α‐Fuc)

(2)

The predominance of α-linked units (Table 2) confirmed the α-stereodirecting effect of DMF, already known for oligosaccharide glycosylations.24 As to the sulfation pattern in residues other than Fuc, the concurrent presence in the HSQC-DEPT spectrum (Figure 3a) of signals of both sulfated (δH/C 4.75/77.6 ppm) and nonsulfated (δH/C 4.08−4.14/67.9−68.9 ppm) CH at position 4 as well as sulfated (δH/C 4.21−4.32/68.2−68.7 ppm) and nonsulfated (δH/C 3.79/62.4 ppm) CH at position 6 of GalNAc units,13e,29 suggested an A,C sulfation pattern for the chondroitin backbone. 2243


Article

Biomacromolecules

fCS was obtained through tailored chemical reactions on microbial-sourced chondroitin. The key step of the synthetic strategy was the fucosylation with per-O-benzylated thioglycoside or N-phenyl-trifluoroacetimidate Fuc donor of suitably protected chondroitin acceptors. Further protecting group manipulations on the fucosylated chondroitin intermediate gave free hydroxyls at selected positions, which were then sulfated. To the best of our knowledge, this is the first work demonstrating the possibility of glycosylating secondary hydroxyls of a polysaccharide, moreover with a labile sugar such as Fuc. Different protection patterns on Fuc donors as well as alternative manipulations of the protecting groups on fucosylated chondroitin intermediates could be planned, in order to obtain a family of semisynthetic fCSs, possessing different Fuc branching and sulfation patterns. This would open an avenue to a wide structure−activity relationship investigation of the anticoagulant properties of fCSs that would be not limited to the natural structures found in sea cucumbers. Work to this aim is currently in progress and will be published in due time.

This is in agreement with the known nonregioselective oxidative opening of the benzylidene ring to give a free hydroxyl at either 4O- or 6-O-position of GalNAc units13e,24 before the sulfation step (12-iv → 14-iv, Scheme 2). A 26% sulfation degree at position O4 was estimated by HSQC-DEPT relative integration of the GalNAc4S CH-4 signal (δH/C 4.75/77.6 ppm) with respect to the CH-2 signal of whichever GalNAc unit (δH/C 4.02/52.2 ppm). Evaluation of O-6 sulfation degree by integration of sulfated GalNAc6S CH-6 signal (δH/C 4.21−4.32/68.2−68.7 ppm) was complicated by overlapping with other ones in the HSQC-DEPT spectrum. In the fCS-v case, the absence of Fuc branching on GalNAc units allowed an evaluation of O-6 sulfation degree (75%) by relative integration of GalNAc6S CH-5 signal (δH/C 3.84−3.96/73.3−73.6 ppm) with respect to the CH-2 one. A structure of fCS-iv and fCS-v summarizing the information inferred from 2D-NMR analysis is depicted in Figure 4. Interestingly, fCS-iv showed features similar to the novel structure very recently suggested for fCS polysaccharide from Apostichopus japonicus.30

■

ASSOCIATED CONTENT

S Supporting Information *

1

H and 13C NMR chemical shift attribution of polysaccharides 1, 3−5, copies of 1D- and 2D-NMR spectra of semisynthetic fCSs and intermediates thereof, and HP-SEC-TDA profiles. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biomac.5b00640.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Figure 4. Structural features of fCS-iv and fCS-v.

High-performance size exclusion chromatography combined with a triple detector array (HP-SEC-TDA)31 allowed the evaluation of weight-averaged molecular mass (Mm) for fCS-iv (8.5 kDa), that was lower with respect to starting chondroitin (23.7 kDa), presumably due to polysaccharide chain shortening during acid-mediated reactions (glycosylation and sulfation). Nonetheless, polydispersity was only poorly increased (1.22 and 1.35 for starting chondroitin and fCS-iv, respectively). Interestingly, a Mm value of 8.5 kDa is indicated in the literature to be in an optimal range to retain anticoagulant activity (at least 6−8 repeating units are required), while minimizing undesirable effectssuch as platelet aggregation, bleeding, hypotension exhibited by natural high molecular mass fCSs.10b,11,32 Semisynthetic fCS-iv was preliminarly assayed for anticoagulant activity. AT-dependent activity against factor Xa (0.38 IU/ mg) was more than 20-fold higher than CS from shark (0.017 IU/mg), thus confirming that the anticoagulant activity could be mostly attributed to sulfated Fuc branching units,4 and more than 500 times less active than unfractionated heparin (198 IU/mg). This activity profile is very similar to that reported for low molecular mass fCS polysaccharides obtained from natural sources.4c,10a,11a However, more biological investigations are necessary in order to better characterize the anticoagulant activity of fCS-iv, and they are scheduled for the near future.

Funding

Regione Campania (Progetto di Ricerca POR-BIP BioIndustrial Processes) and Ministero Italiano dell’Istruzione dell’Università e della Ricerca (MIUR, Progetto di Ricerca Progetto di Ricerca di Interesse Nazionale E61J12000210001) are acknowledged for financial support. Consorzio Regionale di Competenza in Biotecnologie Industriali (BioTekNet) is also acknowledged. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully thank Dr. Paola Diana (Department of Experimental Medicine, Second University of Naples) for molecular mass analyses and Miss Anna Virginia Adriana Pirozzi (Department of Experimental Medicine, Second University of Naples) for anticoagulant activity assay.

■

REFERENCES

(1) Chondroitin Sulfate: Structure, Use and Health Implications; Pomin, V. H., Ed.; Nova Science Publishers, Inc.: Hauppauge, NY, 2013. (2) Pomin, V. H. Mar. Drugs 2014, 12, 232−254. (3) (a) Myron, P.; Siddiquee, S.; Al Azad, S. Carbohydr. Polym. 2014, 112, 173−178. (b) Chen, S.; Xue, C.; Yin, L.; Tang, Q.; Yu, G.; Chai, W. Carbohydr. Polym. 2011, 83, 688−696. (4) (a) Monteiro-Machado, M.; Tomaz, M. A.; Fonseca, R. J. C.; Strauch, M. A.; Cons, B. L.; Borges, P. A.; Patrão-Neto, F. C.; TavaresHenriques, M. S.; Teixeira-Cruz, J. M.; Calil-Elias, S.; Cintra, A. C. O.; Martinez, A. M. B.; Mourão, P. A. S.; Melo, P. A. Toxicon 2015, 98, 20−

■

CONCLUSIONS The production of fCS polysaccharide has been for the first time accomplished using a nonanimal source. Indeed, semisynthetic 2244


Article

Biomacromolecules

(16) (a) Daly, R.; McCabe, T.; Scanlan, E. M. J. Org. Chem. 2013, 78, 1080−1090. (b) Fujiwara, R.; Horito, S. Carbohydr. Res. 2011, 346, 2098−2103. (c) Adinolfi, M.; Iadonisi, A.; Ravidà, A.; Schiattarella, M. Synlett 2004, 275−278. (d) Périon, R.; Lemée, L.; Ferrières, V.; Duval, R.; Plusquellec, D. Carbohydr. Res. 2003, 338, 2779−2792. (e) Vermeer, H. J.; van Dijk, C. M.; Kamerling, J. P.; Vliegenthart, J. F. G. Eur. J. Org. Chem. 2001, 193−203. (f) Schmid, U.; Waldmann, H. Chem.Eur. J. 1998, 4, 494−501. (17) (a) Benesova, E.; Lipovova, P.; Dvorakova, H.; Kralova, B. Glycobiology 2013, 23, 1052−1065. (b) Sakurama, H.; Fushinobu, S.; Masafumi, H.; Yoshida, E.; Honda, Y.; Ashida, H.; Kitaoka, M.; Kumagai, H.; Yamamoto, K.; Katayama, T. J. Biol. Chem. 2012, 287, 16709− 16719. (c) Serna, S.; Yan, S.; Martin-Lomas, M.; Wilson, I. B. H.; Reichardt, N.-C. J. Am. Chem. Soc. 2011, 133, 16495−16503. (d) Cobucci-Ponzano, B.; Conte, F.; Bedini, E.; Corsaro, M. M.; Parrilli, M.; Sulzenbacher, G.; Lipski, A.; Dal Piaz, F.; Lepore, L.; Rossi, M.; Moracci, M. Chem. Biol. 2009, 16, 1097−1108. (e) Wada, J.; Honda, Y.; Nagae, M.; Kato, R.; Wakatsuki, S.; Katayama, T.; Taniguchi, H.; Kumagai, H.; Kitaoka, M.; Yamamoto, K. FEBS Lett. 2008, 582, 3739− 3743. (f) Cobucci-Ponzano, B.; Conte, F.; Mazzone, M.; Bedini, E.; Corsaro, M. M.; Rossi, M.; Moracci, M. Biocat. Biotrans 2008, 26, 18− 24. (g) Farkas, E.; Thiem, J.; Ajisaka, K. Carbohydr. Res. 2000, 328, 293− 299. (18) Comegna, D.; Bedini, E.; Di Nola, A.; Iadonisi, A.; Parrilli, M. Carbohydr. Res. 2007, 342, 1021−1029. (19) Restaino, O. F.; Cimini, D.; De Rosa, M.; De Castro, C.; Parrilli, M.; Schiraldi, C. Electrophoresis 2009, 30, 3877−3883. (20) (a) Gargiulo, V.; Lanzetta, R.; Parrilli, M.; De Castro, C. Glycobiology 2009, 19, 1485−1491. (b) Guerrini, M.; Naggi, A.; Guglieri, S.; Santarsiero, R.; Torri, G. Anal. Biochem. 2005, 337, 35−47. (21) Valoti, E.; Miraglia, N.; Bianchi, D.; Valetti, M.; Bazza, P. US Patent Application Publication 2012/0295865. (22) Pawar, S. N.; Edgar, K. J. Carbohydr. Polym. 2013, 98, 1288−1296. (23) Lonn, H. Carbohydr. Res. 1985, 139, 105−113. (24) Lu, S.-R.; Lai, Y.-H.; Chen, J.-H.; Liu, C.-Y.; Mong, K.-K. T. Angew. Chem., Int. Ed. 2011, 50, 7315−7320. (25) Adinolfi, M.; Barone, G.; Guariniello, L.; Iadonisi, A. Tetrahedron Lett. 1999, 40, 8439−8441. (26) Lucas, H.; Hasten, J. E. M.; van Dinther, T. G.; Meuleman, D. G.; van Aelst, S. F.; van Boeckel, C. A. A. Tetrahedron 1990, 46, 8207−8228. (27) Meloncelli, P. J.; West, L. J.; Lowary, T. L. Carbohydr. Res. 2011, 346, 1406−1426. (28) (a) Hua, Y.; Gu, G.; Du, Y. Carbohydr. Res. 2004, 339, 867−872. (b) Mulloy, B.; Mourão, P. A. S.; Gray, E. J. Bacteriol. 2000, 77, 123−135. (29) Mucci, A.; Schenetti, L.; Volpi, N. Carbohydr. Polym. 2000, 41, 37−45. (30) Yang, J.; Wang, Y.; Jiang, T.; Lv, Z. Int. J. Biol. Macromol. 2015, 72, 911−918. (31) (a) La Gatta, A.; De Rosa, M.; Marzaioli, I.; Busico, T.; Schiraldi, C. Anal. Biochem. 2010, 404, 21−29. (b) Bertini, S.; Bisio, A.; Torri, G.; Benzi, D.; Terbojevich, M. Biomacromolecules 2005, 6, 168−173. (32) Sheehan, J. P.; Walke, E. N. Blood 2006, 107, 3876−3882.

33. (b) Mourão, P. A. S.; Guimarães, M. A. M.; Mulloy, B.; Thomas, S.; Gray, E. Br. J. Hamaetol. 1998, 101, 647−652. (c) Mourão, P. A. S.; Pereira, M. S.; Pavão, M. S.; Mulloy, B.; Tollefsen, D. M.; Mowinckel, M. C.; Abildgaard, U. J. Biol. Chem. 1996, 271, 23973−23984. (5) Glauser, B. F.; Pereira, M. S.; Monteiro, R. Q.; Mourão, P. A. S. Thromb. Haemost. 2008, 100, 420−428. (6) Fonseca, R. J.; Mourão, P. A. S. Thromb. Haemost. 2006, 96, 822− 829. (7) Greinacher, A.; Michels, J.; Schäfer, M.; Kiefel, V.; MuellerEckhardt, C. Br. J. Hamaetol. 1992, 81, 252−254. (8) (a) Guerrini, M.; Beccati, D.; Shriver, Z.; Naggi, A.; Viswanathan, K.; Bisio, A.; Capila, I.; Lansing, J. C.; Guglieri, S.; Fraser, B.; Al-Hakim, A.; Gunay, N. S.; Zhang, Z.; Robinson, L.; Buhse, L.; Nasr, M.; Woodcock, J.; Langer, R.; Venkataraman, G.; Linhardt, R. J.; Casu, B.; Torri, G.; Sasisekharan, R. Nat. Biotechnol. 2008, 6, 669−675. (b) Kishimoto, T. K.; Viswanathan, K.; Ganguly, T.; Elankumaran, S.; Smith, S.; Pelzer, K.; Lansing, J. C.; Sriranganathan, N.; Zhao, G.; Galcheva-Gargova, Z.; Al-Hakim, A.; Bailey, G. S.; Fraser, B.; Roy, S.; Rogers-Controne, T.; Buhse, L.; Whary, M.; Fox, J.; Nasr, M.; Dal Pan, G. J.; Shriver, Z.; Langer, R. S.; Venkataraman, G.; Austen, F.; Woodcock, J.; Sasisekharan, R. New Engl. J. Med. 2008, 358, 2457−2467. (9) (a) Ustyuzhanina, N. E.; Fomitskaya, P. A.; Gerbst, A. G.; Dmitrenok, A. S.; Nifantiev, N. E. Mar. Drugs 2015, 13, 770−787. (b) Tamura, J.-I.; Tanaka, H.; Nakamura, A.; Takeda, N. Tetrahedron Lett. 2013, 54, 3940−3943. (10) (a) Panagos, C.; Thomson, D.; Moss, C.; Hughes, A. D.; Kelly, M. S.; Liu, Y.; Chai, W.; Venkatasamy, R.; Spina, D.; Page, C. P.; Hogwood, J.; Woods, R. J.; Mulloy, B.; Bavington, C.; Uhrin, D. J. Biol. Chem. 2014, 289, 28284−28298. (b) Zhao, L.; Lai, S.; Huang, R.; Wu, M.; Gao, N.; Xu, L.; Qin, H.; Peng, W.; Zhao, J. Carbohydr. Polym. 2013, 98, 1514− 1523. (11) (a) Wu, M.; Wen, D.; Gao, N.; Xiao, C.; Yang, L.; Xu, L.; Lian, W.; Peng, W.; Jiang, J.; Zhao, J. Eur. J. Med. Chem. 2015, 92, 257−269. (b) Wu, M.; Xu, S.; Zhao, J.; Kang, J.; Ding, H. Food Chem. 2010, 122, 716−723. (12) (a) He, W.; Fu, L.; Li, G.; Jones, J. A.; Linhardt, R. J.; Koffas, M. Metab. Eng. 2015, 27, 92−100. (b) Cimini, D.; Fantaccione, S.; Volpe, F.; De Rosa, M.; Restaino, O. F.; Aquino, G.; Schiraldi, C. Appl. Microbiol. Biotechnol. 2014, 98, 3955−3964. (c) Liu, J.; Yang, A. H.; Liu, J.; Ding, X. F.; Liu, L. M.; Shi, Z. P. Biotechnol. Lett. 2014, 36, 1469− 1477. (d) Restaino, O. F.; Di Lauro, I.; Cimini, D.; Carlino, E.; De Rosa, M.; Schiraldi, C. Appl. Microbiol. Biotechnol. 2013, 97, 1699−1709. (e) Cimini, D.; De Rosa, M.; Carlino, E.; Ruggiero, A.; Schiraldi, C. Microb. Cell Fact. 2013, 12, 46. (f) Wu, Q. L.; Yang, A. H.; Zou, W.; Duan, Z. Y.; Liu, J.; Chen, J.; Liu, L. M. Biotechnol. Prog. 2013, 29, 1140− 1149. (g) DeAngelis, P. L. Appl. Microbiol. Biotechnol. 2012, 94, 295− 305. (h) Cimini, D.; Restaino, O. F.; Catapano, A.; De Rosa, M.; Schiraldi, C. Appl. Microbiol. Biotechnol. 2010, 87, 1779−1787. (i) Schiraldi, C.; Cimini, D.; De Rosa, M. Appl. Microbiol. Biotechnol. 2010, 87, 1209−1220. (13) (a) Laezza, A.; De Castro, C.; Parrilli, M.; Bedini, E. Carbohydr. Polym. 2014, 112, 546−555. (b) Bianchi, D.; Valetti, M.; Bazza, P.; Miraglia, N.; Valoti, E. World Patent WIPO PCT 2012/152872 A1. (c) Bedini, E.; Parrilli, M. Carbohydr. Res. 2012, 356, 75−85. (d) Bedini, E.; De Castro, C.; De Rosa, M.; Di Nola, A.; Restaino, O. F.; Schiraldi, C.; Parrilli, M. Chem.Eur. J. 2012, 18, 2123−2130. (e) Bedini, E.; De Castro, C.; De Rosa, M.; Di Nola, A.; Iadonisi, A.; Restaino, O. F.; Schiraldi, C.; Parrilli, M. Angew. Chem., Int. Ed. 2011, 50, 6160−6163. (f) Zoppetti, G.; Oreste, P. US Patent US 2004/6777398. (14) Fox, S. C.; Li, B.; Xu, D.; Edgar, K. J. Biomacromolecules. 2012, 12, 1956−1972. (15) (a) Ishimaru, M.; Nagatsuka, M.; Masubuchi, A.; Okazaki, J.; Kurita, K. Polym. Bull. 2014, 71, 301−313. (b) Kurita, K.; Matsumura, Y.; Takahara, H.; Hatta, K.; Shimojoh, M. Biomacromolecules 2011, 12, 2267−2274. (c) Kurita, K.; Akao, H.; Yang, J.; Shimojoh, M. Biomacromolecules 2003, 4, 1264−1268. (d) Kurita, K.; Shimada, K.; Nishiyama, Y.; Shimojoh, M.; Nishimura, S. Macromolecules 1998, 31, 4764−4769. 2245


Chemical Fucosylation of a Polysaccharide: A ... - ACS Publications

Recommend Documents