Silver-assisted gold-catalyzed formal synthesis of the anticoagulant Fondaparinux pentasaccharide

Clinically approved anti-coagulant Fondaparinux is safe since it has zero contamination problems often associated with animal based heparins. Fondaparinux is a synthetic pentasaccharide based on the antithrombin-binding domain of Heparin sulfate and contains glucosamine, glucuronic acid and iduronic acid in its sequence. Here, we show the formal synthesis of Fondaparinux pentasaccharide by performing all glycosidations in a catalytic fashion for the first time to the best of our knowledge. Designer monosaccharides were synthesized avoiding harsh reaction conditions or reagents. Further, those were subjected to reciprocal donor-acceptor selectivity studies to guide [Au]/[Ag]-catalytic glycosidations for assembling the pentasaccharide in a highly convergent [3 + 2] or [3 + 1 + 1] manner. Catalytic and mild activation during glycosidations that produce desired glycosides exclusively, scalable route to the synthesis of unnatural and expensive iduronic acid, minimal number of steps and facile purifications, shared use of functionalized building blocks and excellent process efficiency are the salient features.

S ulfated linear polysaccharides consisting of alternating disaccharide units of α-1,4-linked glucosamine and either glucuronic acid or iduronic acids such as heparin (H) and heparin sulfate (HS) are present on the surface of most animal cells, membranes, and extracellular matrices 1,2 . They play a pivotal role in diverse biological pathways including tumor metastasis, cell growth, cell adhesion, wound healing, inflammation, diseases of the central nervous system, etc. 3,4 . Both H and HS are heavily O-and N-sulfated, they belong to glycosaminoglycan polysaccharides and are extracted and isolated from natural animal sources (porcine intestine or bovine lung or sometimes from turkeys, mice, camel, whales, lobsters, etc.) 5 . H and HS are routinely used as anticoagulant drugs during major surgeries such as cardiopulmonary bypass, knee replacement, hip replacement in order to prevent the occurrence of venous thrombosis 6 .
Porcine or bovine-derived Heparin has been used in clinics for many decades as an anticoagulant drug due to its strong affinity binding with antithrombin III thereby preventing venous thrombosis 7 . Unfractionated heparin (UFH) (MW avg~1 5,000, 45 monosaccharide chains) and low molecular weight heparin (LMWH) (MW avg~6 000) are marketed for a longtime 8,9 . Several LMWHs are marketed under different trade names such as Enoxaparin, Nadroparin, Reviparin, Dalteparin, Tinzaparin, Certoparin, and Danaparoid depending on the type of depolymerization 10,11 . For example, Enoxaparin is isolated from UFH after β-eliminative cleavage employing alkali through peeling off reaction whereas Nadroparin is obtained by deaminative cleavage employing nitrous acid 12 . Mechanistic investigations revealed that the H binds to antithrombin with high affinity, brings in a conformational change thereby converting it to a rapid (1000×) inhibitor of thrombin (FIIa) 13 . Apart from thrombin, antithrombin interacts with coagulation factor Xa (FXa). LMWHs derived by chemical and/or enzymatic depolymerization procedures from UFH vary in both their relative abilities to enhance the inhibition of FXa and FIIa (anti-FIIa) and in their physicochemical properties. It has been noticed that specific FXa inhibitory activity increases as the mean molecular weight decreases. For example, UFH (MW avg~1 5,000) has an anti-FXa/Anti-FIIa activity ratio of 1.0 whereas the same ratio for Enoxaparin (MW avg~4 200) is 3.9 and Bemiparin (MW avg~3 600) was 8.0 12 . However, chemically and enzymatically extracted H and HS from animal sources suffer from microheterogeneity, presence of viral or prion contaminants; and hence, strongly influence their purity and quality from batch to batch 14,15 .
The problem manifested into a pinnacle due to the worldwide distribution of contaminated animal-sourced heparin about a decade ago.
Much before, in the 1980s, a unique pentasaccharide domain of heparin was found to be clinically effective as a specific FXa inhibitor [16][17][18] . This important discovery paved the way for the chemically synthesized pentasaccharide that later led to the launch of the first synthetic anticoagulant antithrombotic Fondaparinux (Arixtra ® ) 1 in 2004 ( Fig. 1) 19 . Fondaparinux (MW = 1725) has well-controlled pharmacokinetic and pharmacodynamics properties, is free from any viral or prion impurities, and importantly, is a specific FXa inhibitor. Extensive structureproperty relationship studies proved that essential sulfate and carboxylic acid groups shall be located at opposite sides of the pentasaccharide. Despite its predictable anticoagulant dose and long half-life, Fondaparinux (1) is very expensive compared to H and HS derived from animals as its synthesis demands a long and tedious procedure diminishing the overall efficiency 20 . Indeed, the synthesis of fondaparinux pentasaccharides and other heparin oligosaccharides is a herculean task due to the intricacies involved in the installation and unblocking of multiple orthogonal protecting groups. Since the first synthesis of the pentasaccharide by Petitou in 1987 21 , several synthetic strategies have been reported for the heparin fragments involving stepwise glycosylation invoking many protecting groups and glycosylation protocols.
Given the complexity of Fondaparinux, very few total syntheses [22][23][24][25][26][27][28][29] are reported to date and they mainly focused on the following aspects in order to improve the overall yield: (i) optimizing chemistry of individual monosaccharides; (ii) identification of right pairs of orthogonal protecting groups; (iii) stereoselective glycosylation chemistry. On the contrary, HS polymerase, sulfotransferases, and epimerases were employed for the enzymatic synthesis of the Fondaparinux. In spite of these, access to differentially substituted derivatives is very significant for structure-property relationships of the Fondaparinux. Reported synthesis of the pentasaccharide 1 till date have modeled their convergent or linear or multiple one-pot strategies either by 3 + 2 or 3 + 1 + 1 combination of modular saccharide building blocks. A rapid and facile synthetic strategy for compound 1 that also enables the creation of diverse molecular entities that differ in sulfation pattern is still in demand. Accordingly, an efficient synthetic strategy has been envisioned for the synthesis of Fondaparinux pentasaccharide standing on the silver-assisted gold-catalyzed activation of glycosyl carbonates.

Results and discussion
We targeted the synthesis of Fondaparinux pentasaccharide by employing our recently discovered silver-assisted gold-catalyzed glycosidations 30 on strategically designed monosaccharide building blocks. The synthesis of Fondaparinux pentasaccharide 1 was envisioned from the regioselectively protected pentasaccharide 2. Trisaccharides 3 was chosen as the key precursor for effecting either 3 + 2 glycosylation using disaccharide 4 or 3 + 1 + 1 elongation by coupling iduronate 5 followed by the azidoderivative 6. Trisaccharides 3 can be synthesized by stepwise glycosylation using alkynyl glycosyl carbonates and Au-and Agsalts as catalysts. Iduronates (5, 10) and glucuronate 8 are envisaged from D-Glucose whereas building blocks 6, 7, and 9 are imagined from D-Glucosamine (Fig. 1). Importantly, building blocks are designed in such a way that the scale-up and convergent synthesis from common precursors shall be accomplished in minimal steps.
Synthesis of GH disaccharide. Accordingly, our synthesis endeavor commenced with the identification of a scalable method for the iduronate 5. Chelation assisted nucleophilic addition on a C-5 aldehyde 12 that can be easily accessed from diacetoneglucofuranose derivative 11 was envisioned for the synthesis of idose derivative 13 ( Fig. 2A) 31 . The nucleophile shall be chosen in such a way that it can be converted into a carboxylate at a later stage. Hence, as a model reaction, commercially available diacetoneglucofuranose was transformed into aldehyde 14a 32 and was treated with commercially available PhMgBr to notice the formation of phenyl carbinols that are ido-configured 15a and gluco-configured 16a in a ratio of 10:1 with an overall yield of 85% which could be easily separated by silica gel column chromatography. Furanose to pyranose conversion of compound 15a was easily accomplished under acidic conditions to afford pyranosides which were treated with Ac 2 O and pyridine to get idoderivative 17a in 74% yield. Oxidation of idopyanoside 17a to iduronate 18a was accomplished with the help of RuO 4 which was generated in situ 33,34 from RuCl 3 and NaIO 4 followed by the esterification using K 2 CO 3 and CH 3 I (Fig. 2B).
Synthesis of target molecule warrants installation of benzyl ether at the C-3 position. Thus, freshly prepared aldehyde 14b was subjected to Grignard reaction using PhMgBr to obtain 12:1 ratio of ido-(15b) and gluco-(16b) derivatives in 83% yield. Furanoside 15b was converted to pyranoside 17b by the treatment of aq. 90% TFA followed by acetylation under Ac 2 O/ py conditions. At this stage, oxidation of phenyl group of idoside 17b in the presence of benzyl ether posed a serious challenge. Several oxidation conditions are tried and failed to affect the regioselective oxidation of the phenyl group and hence could not isolate the iduronate 18b (Fig. 2B).
Idose synthesis via the chelation assisted Grignard addition strategy is highly beneficial as the synthesis on large scales is feasible [35][36][37][38] . Therefore, the addition of 2-Thienylmagnesium bromide 39,40 caught our attention as the thiophene can undergo oxidation under mild conditions without disturbing the benzyloxy group at the C-3 position. Consequently, freshly prepared 2thienylmagnesium bromide was treated with aldehyde 14b to obtain the ido-derivative 19 predominantly in 85% yield (see Supplementary Fig. S13a-c). Acetylation under Ac 2 O to obtain acetate 20 followed by the oxidation of the thiophene moiety to the desired acid and subsequent esterification underwent smoothly to afford iduronate in 74% yield over two steps followed by the Zemplén deacetylation afforded compound 21 in 94% yield (Fig. 2C). Furanose to pyranose conversion of iduronate 21 under acidic conditions followed by its conversion to the isopropylidene derivative 5 occurred uneventfully and the remaining C-4 axial hydroxyl group was protected as silyl ether using TBDMSOTf/2,6-lutidine to afford compound 23 that was converted to the much-desired glycosyl donor 10 in two steps.  Firstly, careful hydrolysis of the isopropylidene group afforded a hemiacetal (see Scheme S1) that was directly treated with 1ethynylcyclohexyl (4-nitrophenyl) carbonate (24) 30 in the presence of DMAP to obtain an anomeric mixture of carbonates, and finally, the C-2-OH was protected as its acetate to obtain donor 10 in 97% yield (Fig. 2C, see Supplementary Scheme S1).
Synthesis of glucosamine building blocks. In parallel, synthesis of regioselectively protected glucosamine building blocks commenced with the preparation of Troc-protected glucosamine derivative 26 adopting known procedures 41 . Tetraol 26 was first converted to its allyl α-D-glycoside which was converted to benzylidene 27 by using benzylidenedimethylacetal and the Trocprotecting group was unmasked by Zn-mediated reaction to afford the amino alcohol 28 in 77% yield. An azide at the C-2 position strongly influences the stereochemical outcome of glycosidation in favor of desired 1,2-cis or α-glucoside. Therefore, conversion of amine to azide was easily accomplished by the use of freshly prepared azidosulfonylimidazole 42 in CH 3 OH-THF to afford compound 29 in 82%. At this point, a series of protections on the lone hydroxyl group were considered in order to explore orthogonal protections/deprotections at a subsequent stage. Accordingly, the lone hydroxyl group was protected as p-methoxybenzyl ether (30a) or benzyl ether (30b) or naphthyl ether (30c) or silyl ether (30d) or acetate (30e) using appropriate reaction conditions (Fig. 3). In continuation, benzylidene acetal of compounds 30a-30e was hydrolyzed under acidic conditions, regioselective protection of the C 6 -OH was accomplished to get desired aglycons 31a-31f, 6 in good yields (see Supplementary Scheme S2). The C 6 -silyl ethers were accomplished using silyl chloride and trimethylamine whereas C 6 -esters were realized by the treatment of diol with corresponding anhydrides at 25°C (Fig. 3). It has been foreseen that the synthesis of trisaccharide 3 can be achieved by either [2 + 1] or [1 + 2] fashion for which another set of building blocks are required. Therefore, compound 31b was protected as either naphthyl ether (32a), PMB ether (32b), or benzyl ether (32c) under suitable conditions. Cleavage of the silyl ether of naphthyl ether 32a followed by the C 6 -protection afforded other building blocks 32d and 32e. Compounds 32a-32e are quite interesting as they possess orthogonal protecting groups that can be unmasked when desired without affecting the other. In the next sequence, Pd-catalyzed cleavage 43,44 of the allyl glycoside afforded hemiacetals that were conveniently converted to the much-desired carbonate donors 33a-33e in excellent yields ( Fig. 3 and Supplementary Scheme S3).
Synthesis of glucuronic acid building blocks. Thus synthesized acetate building block 6 and the building block 10 prepared vide supra were subjected to the silver-assisted gold-catalyzed glycosidation using (2,4-( t Bu) 2 C 6 H 3 O) 3 PAuCl (25)/AgOTf to afford a 1,2-trans disaccharide in 89% that upon hydrolysis of the silyl  (Fig. 2C). Next in our journey is the synthesis of glucuronate donor 8. Here again, a diversified set of glucuronyl donors (38a-38c) were envisioned in order to perform reciprocal donor-acceptor studies with orthogonal protecting groups (Fig. 4).
The path started with the commercially available benzyl protected diacetone glucofuranose 34 which was converted to the pyranoside 35a-35c in three steps. Stirring of compound 34 in aqueous TFA at room temperature facilitated hydrolysis of isopropylidene and subsequent treatment with allyl alcohol in acidic conditions 0-80°C afforded allyl α-D-glucopyranoside whose C4 and C6 alcohols were locked as benzylidene. Further, the remaining C-2-OH was protected as either levulinoate (35a) or acetate (35b), or benzoate (35c) under standard conditions. The presence of ester at C-2 would assist in neighboring group participation to obtain 1,2-trans glucuronide.
In continuation, the benzylidene of allyl pyranosides 35a-35c was hydrolyzed using PTSA/MeOH-CH 2 Cl 2 to obtain a diol and the resulting primary hydroxyl moiety was oxidized under TEMPO/BAIB conditions 24,25 to obtain an acid that was protected as their methyl esters 36a-36c (see Scheme S3). The lone hydroxyl group of ester 36a-36c was orthogonally protected to obtain esters 37a-37c. Furthermore, allyl glucuronides 37a-37c were hydrolyzed under PdCl 2 /MeOH-CH 2 Cl 2 conditions and subsequently treated with the carbonate reagent 24 to get donors 38a-38c in excellent yields (Fig. 4). Synthesis of the trisaccharide 3 is contingent on developing protocols and synthesizing all the identified building blocks in enough quantities. Trisaccharide 3 synthesis commenced with optimization of conditions for the stereoselective synthesis of D-E disaccharide using donors 33a-33e and aglycons 36a-36c prepared to vide supra. All glycosidations between azidoglucosyl donors and acceptor were conducted employing Au-phosphite (25) and AgOTf in CH 2 Cl 2 containing 4 Å MS powder to afford the disaccharide 39a-39i ( Table 1).
Synthesis of DEF trisaccharide. Our explorations on reciprocal donor-acceptor selectivity studies commenced with the silver-assisted gold-catalyzed glycosidation with the donor 33a and acceptor 36a that have orthogonal protecting groups TBDPS-and Nap-on glycosyl donor and Lev-, Benzyl-moieties to afford D-E disaccharide 39a in 77% yield with 4:1 (α:β) ratio (Entry 1). Changing the protecting group to the acetate as in 36b did not alter the α:β ratio of disaccharides 39b (Entry 2), Next, the glycosylation between donors 33b-33e and acceptor 36b showed marginal improvement in the α:β ratio with little difference in the yield of the resulting disaccharides 39c-39f (Entries 3-6). However, a quantum jump in the α:β ratio to 8:1 was noticed when the glycosylation was carried out between the glycosyl donor 33a and 33b at −20°C with a very high yield as well (Entry 7). Though the selectivity got improved, the 1,2-cis or αlinked disaccharide and the unwanted β-isomer are noticed to have similar polarities and hence, demanded multiple column chromatographic separations thereby compromising the overall process efficiency. Therefore, the glycosylation between the donor 33a-33c and acceptor 36c was performed at 25°C to obtain the disaccharide 39g-39i with a similar α:β ratio of D-E disaccharides (Entry 8-10). To our satisfaction, a good Rf difference was noticed in compounds 39g and 39i thereby facilitating the easy purification. Further, the temperature dependence of the glycosylation between donor 33a, 33c, and acceptor 36c at −20 and −40°C was performed to notice that the best stereoselectivity and yield were obtained with the donor possessing C6-OTBDPS and C4-OBn ethers (33c) and acceptor having C2-OBz (Entries 11-14). Thus the reciprocal donor-acceptor studies showed that glycosylation shall be carried out at −40°C using glycosyl donor 33c and acceptor 36c (Entry 14, Table 1). The next milestone is the synthesis of D-E-F trisaccharide. In this direction, allyl moiety of the disaccharide 39i was smoothly cleaved off by Pd-catalyzed conditions to afford hemiacetals (see Supplementary Scheme S4) which were treated with the carbonate reagent 24 in the presence of DMAP in CH 2 Cl 2 to afford the desired D-E disaccharide donor 40 that can be used for the synthesis of D-E-F trisaccharide (Fig. 5).
Next, attempts to synthesize trisaccharide from the disaccharide 40 and 31c-31f failed to give the desired trisaccharide; instead, resulted in the isolation of 1,2-eliminated compound 41 as a major product (Entries 1-4, Table 2). Discouraged by these results while executing the DE + F strategy for the DEF trisaccharide prompted us to explore its synthesis by D + EF strategy. Accordingly, 38a + 31e glycosylation was performed and noticed the formation of the 1,2-orthoester 42 (Entry 5, Table 2). However, 38b + 31c produced the desired EF disaccharide 43a albeit in poor yield due to the loss while separating the desired EF disaccharide from a myriad of uncharacterized products that formed (Entry 6, Table 2). Although 38c + 31c afforded 43b in 80% but the polarity of the acceptor 31c and the product 43b were almost similar thereby warranting multiple flash purifications (Entry 7, Table 2). In addition, hydrolysis of the TBDMS-ether in the presence of TBDPS-ether also diminished the net yield in the subsequent step. Finally, 38c + 31a resulted in the desired EF disaccharide 44 with 1,2-trans interglycosidic linkage in excellent 91% yield (Entry 8, Table 2).
Synthesis of Fondaparinux pentasaccharide. Continuing our synthesis, cleavage of silyl ether of the E-F disaccharide 44 was smoothly achieved under HF•py conditions to obtain alcohol 45 which was subjected to the silver-assisted gold-catalyzed glycosidation with glycosyl donor 33c to afford the D-E-F trisaccharide 46 in 90% yield. Gratifyingly, trisaccharide 46 was noticed that glycosidation happened in stereoselective fashion resulting in a single 1,2-cis or α-anomer only at −40°C; an anomeric mixture of trisaccharides was noticed above −20°C. Trisaccharide 46 was extrapolated to the carbonate donor 47 in two aforementioned steps by Pd-catalyzed hydrolysis of the allyl ether and subsequent treatment with the carbonate reagent 24 (Fig. 6).
Trisaccharide donor 47 was split into portions and the first portion was treated with the iduronate 5 under Au/Ag-catalysis conditions to afford the tetrasaccharide 48 in 91% within 15 min at 25°C. Gratifyingly, the glycosidation between 47 and 5 resulted in complete α-selectivity at 25°C presumably due to the presence of the isopropylidene of the aglycone. The remaining portion of the trisaccharide donor 47 was treated with the disaccharide 4 that was prepared vide supra to obtain the desired pentasaccharide 50a reminiscent of the Fondaparinux in 86% yield (Fig. 6, see Supplementary Fig. 98a-p) 14 .
In parallel, hydrolysis of the isopropylidene and PMB-ether present in the tetrasaccharide 48 was effected under acidic conditions and C-1 position was transformed into the alkynyl  carbonate and the remaining diol was protected as diacetate to afford compound 49. Finally, donor 49 was coupled with the acceptor 6 affording the desired pentasaccharide 50b in 75% yield (see Supplementary Fig. 101a-p). Pentasaccharide 50a was split into two portions and subjected to selective deprotections. Unmasking of the p-methoxybenzyl group of the F-unit was successfully accomplished by the action of DDQ in CH 2 Cl 2 -H 2 O at 25°C in 30 min to afford compound 51 in 85% yield. The monosulfated derivative of compound 51, which is otherwise very difficult to synthesize, shows a very significant improvement in the anticoagulant activity. Further, the fluoride ion mediated cleaving of the silyl ether afforded the diol 53 in 81% yield. In parallel, the second portion was subjected to the deprotection of silyl ether first to afford pentasaccharide 52 followed by the deprotection of the PMB-ether afforded the compound 53 (see Supplementary Fig. 104a-p). These regioisomeric hydroxyls will be highly useful for the sulfation to study their biological properties.

Conclusions
In summary, a flexible, modular, and highly efficient synthetic strategy has been developed for the synthesis of Fondaparinux pentasaccharide that stands on the gold-silver catalyzed glycosidation chemistry. This unique route is constructed upon the coupling of thoughtfully identified monosaccharide building blocks which can be synthesized from shared precursors. A scalable chelation assisted method for the synthesis of protected iduronic acid was accomplished using thiophene as a surrogate for the carboxylic acid. All glycosidations are conducted under silver-assisted gold-catalyzed conditions utilizing recently discovered alkynylcyclohexyl carbonate donor chemistry. All glycosidations were optimized to give exclusive stereoselectivity thereby minimizing the tedious purifications. This strategy offers a new catalytic route to the synthesis of mono 3-O-sulfation at Fring of fondaparinux pentasaccharide which is hitherto very difficult to obtain 45 . This significantly improved route for fondaparinux pentasaccharide illustrates a new set of building blocks, the utility of gold-catalyzed glycosidations, and promised to yield orthogonally protected hydroxyl groups that facilitate installation of sulfates in a regiodefined fashion. The 3 + 2 route is preferable for the synthesis of fondaparinux pentasaccharide; however, 3 + 1 + 1 is suitable for diversification of the core structure for structure-function studies. Further research on the exploitation of this strategy for the synthesis of other glycosaminoglycan derivatives of therapeutic significance is currently underway.

Methods
General methods. Unless otherwise noted, materials were obtained from commercial suppliers and were used without further purification. All metal salts were purchased from Sigma-Aldrich. Unless otherwise reported all reactions were performed under Argon atmosphere. Removal of the solvent in vacuo refers to the distillation using a rotary evaporator attached to an efficient vacuum pump. Products obtained as solids or syrups were dried under a high vacuum. Analytical thinlayer chromatography was performed on pre-coated silica plates (F 254 , 0.25 mm thickness); compounds were visualized by UV light or by staining with anisaldehyde spray. Optical rotation was measured on a digital polarimeter. IR spectra were recorded on a Fourier-transform infrared spectrometer. Nuclear magnetic resonance (NMR) spectra were recorded either on a 400 or a 500 or a 600 MHz with CDCl 3 or CD 3 OD as the solvent and TMS as the internal standard. High-resolution mass spectroscopy was performed using an electrospray ionization time-of-flight mass analyzer. Low-resolution mass spectroscopy was performed on ultraperformance liquid chromatography-mass spectrometry with SWADESI-TLC interface. For NMR analysis and high-resolution mass spectrometry of the compounds in this article, see Supplementary  Combined organic phases were washed with brine solution, dried over Na 2 SO 4 , and concentrated in vacuo. The resulting crude residue was redissolved in anhydrous DMF (15 mL) was added 1.5 equivalent of K 2 CO 3 . After stirring for 15 min, iodomethane (2 mmol) was added under an argon atmosphere and the mixture was stirred at 25°C for 8 h in a dark place. After complete consumption, the reaction was quenched by the addition of an excess amount of saturated solution of Na 2 SO 3 , followed by the addition of water and extracted with EtOAc. The combined organic phases were washed with brine, dried over anhydrous Na 2 SO 4 , and concentrated under reduced pressure. The crude residue was purified by silica gel column chromatography. This method was utilized for the preparation of compounds 18a and 21.
Selective oxidation of 1°alcohol to esters. To a biphasic solution of the 4,6-diol (1 mmol) in 2:1 CH 2 Cl 2 -water (10 mL) was added (diacetoxyiodo)benzene (2.5 mmol) and TEMPO (0.2 mmol) simultaneously, stirred vigorously at 25°C. After 3 h, the reaction was quenched by the addition of a saturated aqueous solution of Na 2 SO 3 and extracted with CH 2 Cl 2 , combined organic phases were washed with brine (10 mL), dried over Na 2 SO 4, and concentrated in vacuo. The crude product was redissolved in anhydrous DMF (5 mL) and added 1.5 equivalents of K 2 CO 3 . After stirring for 15 min, iodomethane (2 eq) was added dropwise under an argon atmosphere and stirred for 8 h at 25°C in a dark place. After complete consumption, the reaction was arrested by adding a saturated aqueous solution of Na 2 SO 3 . Extracted with EtOAc, combined organic layers were washed with brine, dried over anhydrous Na 2 SO 4 , concentrated under reduced pressure. The crude residue was purified by silica gel column chromatography using ethyl acetate and n-hexane as mobile phase to obtain the desired product. This method was utilized for the preparation of compounds 36a-36c.
Grignard reaction. Aldehyde 14a, 14b (1 mmol) was dissolved in 5 mL of the anhydrous THF and added slowly to the freshly prepared Grignard reagent (1.5 mmol) at 0°C under argon atmosphere. The reaction mixture was allowed to stir at 25°C for 3 h. After completion, the reaction was quenched by the addition of saturated aqueous ammonium chloride solution, water (20 mL) and extracted with EtOAc (3 × 10 mL). The combined organic phases were washed with brine solution (10 mL), dried over anhydrous Na 2 SO 4 , and concentrated in vacuo. The crude residue was purified by silica gel column chromatography using EtOAc and hexane (30-35%) as a mobile phase to afford aryl carbinols as Ido-or Glc-isomers. Grignard reagent preparation: To a freshly activated Mg-metal (1.5 mmol) and 10 mL of the anhydrous THF in two necks round bottom flask equipped with a condenser was added aryl/heteroaryl bromide (1.0 mmol) slowly under argon atmosphere and the mixture was stirred at 70°C for 1 h. This method was utilized for the preparation of compounds 15, 16, and 19.
Preparation of glycosyl carbonate donors from allyl glycosides. Hemiacetals from allyl glycosides. To a biphasic solution of the allyl glycoside (1.0 mmol) in 3:1 CH 3 OH: CH 2 Cl 2 (20 mL) was added 0.15 equivalent of PdCl 2 and the reaction mixture was stirred for 4-8 h at 25°C, the reaction was quenched by adding an excess of Et 3 N and filtered through a bed of Celite ® . The filtrate was concentrated in vacuo and the crude residue was subjected to silica gel column chromatography using ethyl acetate and n-hexane as mobile phase to obtain the desired hemiacetal. This method was utilized for the preparation of compounds.
Synthesis of glycosyl carbonates. To a solution of glycosyl hemiacetal (1.0 mmol) in anhydrous CH 2 Cl 2 (5 mL) was added DMAP (1.5 eq) and ethynylcyclohexyl (4nitrophenyl) carbonate 24 (1.2 eq), the reaction mixture was stirred at 25°C for 3 h. After complete consumption of hemiacetals, the reaction mixture was concentrated in vacuo and subjected to silica gel column chromatography using EtOAc and hexane as mobile phase.
Silver-assisted gold-catalyzed glycosidation. To a solution of glycosyl donor (1.0 mmol) and acceptor (0.9 mmol) in anhydrous CH 2 Cl 2 (5 mL) was added freshly activated 4 Å MS powder (0.4 g) at 25°C under argon atmosphere. After 15 min of vigorous stirring at 25°C {−40°C for compound 46), chloro[tris(2,4-di-t-butyl phenyl)phosphite]gold(I) (8 mol%) (25) and AgOTf (8 mol%) were added simultaneously to the reaction mixture and stirred for 15 min. After completion, the reaction mixture was quenched by adding an excess of Et 3 N and filtered through a bed of Celite ® , the filtrate was concentrated in vacuo and the crude residue was purified by silica gel column chromatography using ethyl acetate and hexane as mobile phase.