1.3.4 -selective O-glycosylation by additives
Moreover, additives also significantly influence the stereoselectivity.
Bogusiak et al. reported the selective 1,2-cis glycofuranoside synthesis is improved by the addition of a catalytic amount of hexamethyl phosphoramide (HMPA) as an additive.23
Few years later, Crich and his coworker have shown that the challenging -sialylation can be performed by using diphenyl sulfoxide (Ph2SO) and trifluoromethanesulfonic anhydride (Tf2O).24 The excess
couplings and to suppress the formation of elimination product (Scheme 6). They demonstrated that the diphenyl sulfoxide is not only a promoter in glycosylation but it also traps the first-formed oxocarbenium ions.
They also investigated the use of a series of sulfoxides in place of diphenyl sulfoxide (Table 1).
Scheme 7 Glycosylation of a phenyl thiosialoside donor with diphenyl sulfoxide (Ph2O) and triflic anhydride (Tf2O).
Table 1. The effective of additives and sulfoxides
Entry Activator Solvent
(temp)
% yielda ()b
1 CH Cl (-60) 82%
(2.2:1)
2 CH2Cl2/CH3CN1:1(-78) 89%
(1.8:1)
3 CH3CH2CN(-78) 77% (2:1)
4 (4-NO2-Ph)PhSO CH2Cl2(-78) 75%
(2.7:1) 5 (4-OMe-Ph)PhSO CH2Cl2(-78) 50% (2:1)
6 S
O
CH2Cl2(-78) 50% (6:1)
a Isolated yields. b Determined by 1H NMR on the crude reaction mixture.
In 2007, Boons et al. presented an excellent -selective glycosylation of 2-azido-2-deoxy-glucosyl trichloroacetimidates, when performed at a relatively high reaction temperature in the presence of PhSEt or thiophene.25 With NMR and computational studies,
-anomeric sulfonium intermediate formed due to steric hindrance. As a result, the acceptor will come from -side (Scheme 8).
Scheme 8. Boons’s method the glycosylation of 2-azido-2-deoxy glucosides using sulfonium ions.
1.3.5 -Selective O-glycosylation by amide-type molecules
presence of a quaternary mixture of 4-nitobenzenesulfonyl chloride, silver trifluoromethanesulfonate, N,N-dimethylacetamide, and triethylamine (Table 2).26 They proposed a plausible pathway that the alcohol may react with the hypothetical intermediate of -iminium ion to form the corresponding -glucoside. The intermediate -iminium ion is more stable than intermediate -iminium ion thermodynamically, but intermediate -iminium ion is more reactive to form the -glucoside. But they didn’t show any physical data to support their hypothesis.
Table 2. -glucosylation by using DMA (DMF) as an additive
Entry Acceptor additives Yield / ratio
1 I DMFa 58% 77:23
2 I DMF (2.5) 92% 88:12
3 I DMA (2.5) 86% 93:7
4 I DMA (5.0) 73% 86:14
5 II DMA (2.5) 85% 89:11
6 III DMA (2.5) 87% 90:10
7 IV DMA (2.5) 91% 47:53
8 IV DMA (5.0) 88% 73:27
9 IV DMA (10.0) 54% 72:28
a As a solvent.
In 2003, Nishida and his coworker reported a practical glycosylation by one-pot method using Appel agents in N,N-dimethylformamide.27 The role of DMF is demonstrated according to the evident 1H NMR spectra. The signals indicated that the -glycosyl bromide by using Appel agent could be transformed to -glycosyl iminium species when the solvent is DMF. (Scheme 9) Though Lemieux and co-workers19a reported a similar solvent effect, they didn’t indicate the occurrence of such DMF-glycosyl adducts.
Scheme 9 Overview of One-pot -glycosylation Using Appel agents in DMF
Two years later, they still speculate not only -glycosyl bromide but also -glycosyl imidate could be the species to induce the
stereoselectivity.28 Compare to -glycosyl imidate, the -glycosyl imidate is more reactive and it may not be accessible by NMR at the room temperature due to the rapid equilibration. Furthermore, at the lower temperature DMF will be frozen. Nowadays, there is no mean to identify the real species in this reaction.
2 Motivation
1,2-cis-glycosidic bonds are widely occurred in numerous natural oligosaccharides, glycosides, and glycoconjugates, which are widely distributed in living tissues. These compounds are also found in the human milk, in blood group compounds, in bacterial lipopolysaccharide antigens, and many other sources. Such as Lewis (Le) antigens, O-linked glycoproteins, -Gal Ceramide (KRN7000), polysulfated glycosaminoglycans, globotriaosylceramide (Gb3) and N-linked glycoproteins. But there is no general solution for 1,2-cis -glycosidic bond formations by chemical preparation.
Based on the literatures described before, DMF has been used for the stereoselective glycosylation several times. But the exact role of DMF is still out there. To date, the reported examples only use glycosyl halides as donors. Could we apply it to other glycosyl donor, for example common-used thioglycoside? According to the literature survey, DMF can participate the glycosylation to form more stable intermediate, Could we apply it to pre-activation strategy and elevate it to iterative glycosylation? We herein reported the investigation and findings based on the above context and initial finding in our laboratory.
3 Results and discussion
Based on the preliminary studies of glycosyl chlorides, we observed that residual DMF in the glycosylation mixture promoted 1,2-cis
-glycosidic bond formation. Along this line, we hypothesize that this
-glycosylation should also be applicable to thioglycosyl donors, which as a stable glycosyl donor, open access to elucidation of the reaction mechanism. To the best of our knowledge, such investigations have not been reported.
3.1 Optimize the conditions for DMF-modulating glycoslations
In this thesis, we investigated two DMF-modulating glycosylation procedures, and they were depicted in Schemes 9a and 9b.
Scheme 10. (a) DMF-modulating glycosylation procedure (procedure A).
(b) DMF-modulating glycosylation procedure (procedure B).
In procedure A, as adapted from standard glycosylation protocol, a mixture of thioglycosyl donor, glycosyl acceptor and DMF is treated with
N-iodosuccinimide (NIS) and trimethylsilyl triflate (TMSOTf) (Scheme 10a).29 In procedure B, the thioglycosyl donor is firstly pre-activated with NIS and TMSOTf in the presence of DMF. Upon completion of activation, glycosyl acceptor is added and, it reacts with a presumably glycosyl imidate to furnish desired glycosylation product (Scheme 10b).
At the outset, the procedure A was applied to couple commercially available galactosyl acceptor 3 with a perbenzyl thiogalactoside 1. After some experimentations, one molar equivalent of TMSOTf (with respect to glycosyl donor) was required for effective activation of the donor (Scheme 11). A larger amount of TMSOTf may be probably attributed due to a mild Lewis basicity nature of DMF. Nevertheless, the DMF modulator exhibits an -directing effect in glycosylations using thioglyosyl donors, which is in line with our previous findings in glycosyl chlorides.30
Scheme 11. The influence of the equivalent of TMSOTf.
In addition, we observed a quantity-selectivity dependent relationship between the stoichiometric amount of DMF addition and the degree of
glycosylation selectivity. Explicitly speaking, when the amount of DMF increased from zero to 1.5 equiv, the -anomer-ratio of the glycosylation product 4 increased from 1/1 to 3/1 (Table 3, entries 14).
However, this moderate selectivity is still inadequate for synthetic application, but further increase in amount of DMF addition (>1.5 equiv.) aiming at selectivity improvement was prohibited due to the formation of a side-product, namely the formyl transfer product 6.26 Our rationale for this moderate-selectivity in glycosylation is that the arming benzyl groups of donor 1 may promote the departure of DMF from glycosyl imidate; as a consequence, the -directing effect of DMF was attenuated.[21]
Based on such a notion, a conformational restrain benzylidene thiogalactoside 2 is used in place of 1.[22] However, replacing the donor alone did not bring about satisfactory improvement, and a 6/1
-anomer ratio of glycosylation product 5 was obtained (Table 3, entry 5). Nonetheless, adopting the pre-activation procedure B in conjunction with an increase in DMF addition (from 1.5 to 6.0 equiv) did improve the -anomer ratio of 5 to 19/1 (Table 3, entries 68). One may question about whether the ethereal type solvent (as mentioned in the introduction section) could result in similar -directing effect as implicated in previous cases.21b Thus, glycosylation of 3 with 2 was repeated in tetrahydrofuran (THF), 1/3 CH2Cl2/Et2O and 1/2 toluene/dioxane mixture using procedure A. In these experiments, the procedure B is not applicable because this procedure does not work in the absence of DMF. Donor 2 was poorly soluble in pure diethyl ether so
that a 1/3 CH2Cl2/ether mixture was employed. The 1/2 toluene/dioxane mixture was found aggregating at 10˚C so that the glycosylation in 1/2 toluene/dioxane mixture was conducted at 0˚C.No significant selectivity was observed for glycosylations irrespective of the type of ethereal solvent (Table 3, entry 5 vs 912).
In the past, dimethylacetamide (DMA) was used as an additive to promote the -selectivity of glycosylation.26 We were curious to examine if DMA could substitute for DMF in our procedure. Thus, glycosylation of 3 with 2 following the procedure B, was repeated with DMA addition, but the observed selectivity was not attractive (Table 3, entry 13).
Table 3. Investigation of DMF-modulating glycosylation procedures A and B with galactosyl acceptor 3.
Entry Donor (equiv)
DMF (equiv)
T (˚C)
Time (h)
Product, yield%,
[a]
1 1 (1.2)[b] 0 -25 0.5 4, 90, 1/1 2 1 (1.2)[b] 0.8 -10 1.0 4, 70, 3/2 3 1 (1.2)[b] 0.8 0 1.0 4, 77, 3/2 4 1 (1.2)[b] 1.5 0 1.0 4, 80, 3/1 5 2 (1.5)[b] 1.5 -10 2.0 5, 82, 6/1 6 2 (1.5)[c] 1.5 -10 2.0 5, 80, 8/1 7 2 (1.5)[c] 3.0 -10 2.0 5, 87, 15/1 8 2 (1.5)[c] 6.0 -10 2.0 5, 87, 19/1 9 2 (1.5)[b] 0[d] -10 0.3 5, 90, 1/1 10 2 (1.5)[b] 0[e] -10 0.2 5, 85, 1.5/1 11 2 (1.5)[b] 0[f] -10 0.5 5, 83, 1/1.5 12 2 (1.5)[b] 0[f] 0 4.0 5, 40, 1/1.5 13 2 (1.5)[c] -[g] -10 3.0 5, 80, 4/1 [a] ratios were determined by HPLC (conditions given in SI). [b] Procedure A was used. [c] Procedure B was applied. [d] 1/3 CH2Cl2/Et2O mixture was used as solvent. [e] THF was used as solvent. [f] 1:2 Toluene/dioxane was used as solvent. [g]
6 equiv of DMA was added.26
3.2 Test the scope of pre-activated DMF-mediated glycoslation
After confirming the effectiveness of the pre-activated DMF-modulating glycosylation (procedure B), this study next investigated its scope of application. In this regard, aglycon acceptors 10–13, and O-glycoside acceptors 14–17 were coupled with thioglycosyl donors 2 (Figure 7, Table 4). For comparison the effectiveness of this method to conventional method as well as provision of reference data for HPLC analysis, all glycosylations were performed with and without addition of DMF.
Figure 7. (a) Thioglycosyl donors: 2. (b) Acceptors: 10–17; (c) Glycosylation products: 18–25.
Generally, reaction rates were lower in the presence of DMF than with its absence; nonetheless, the time required for completion of DMF-modulating glycosylation remained acceptable (2 to 6 h).
Regarding the stereochemical control, DMF exerted a powerful
-directing effect on all glycosylations. In some cases, the selectivity was dramatically reversed (Table 4, entries 2, 4, 5, 11, and 12).
Table 4. Results of glycosylation of acceptors 10−17 using glycosylation procedure B.
Entry D[a] A[a] T (˚C) Time (h)
Product, yield%, [b]
18-25 with DMF no DMF[c]
1 2 10 -10 2 18 83, 12/1 80, 1/1
2 2 11 -10 2 19 76, 8/1 85, 2/5
3 2 12 -10 6 20 45, 19/1 50,15/1
4 2 13 0 2 21 79, 8/1 73, 2/5
5 2 14 -10 5.5 22 75, 12/1 80, 2/3
6 2 15 0 6 23 80, 49/1 50, 2/1
7 2 16 -10 2 24 82, 12/1 80, 3/2
8 2 17 0 4 25 60, 25/1 63, 5/1
[a] D referred to donor and A referred to acceptor. [b] -Anomer ratios were determined by HPLC (settings were given in experimental). [c] Routine glycosylation (without DMF addition) was applied.
3.2.1 Application of DMF-modulating glycosylation to other thioglycoside donors
Encouraged by the results of DMF-modulating glycosylations, we moved on to investigate the application of our method to prepare 1,2-cis-O-linkage with other thioglyucoside donors. As such, we decided to choose some thiofucoside 7, thiorhamnoside 8 and thioglucopyranoside 9 to evaluate their performance in glycosylations of acceptors 14, 15 and 17 (Figure 8). For comparison, conventional glycosylations in the absence of DMF modulator were also carried out in parallel.
Figure 8. The pairs of thioglycosyl donor, acceptors and their corresponding glycosylation product.
Table 5. Results of glycosylation using glycosylation procedure B
Entry D A T (˚C) Time (h)
Product, yield%,
Product with DMF no DMF 1 7 14 -10 4.5 26 75, 5/1 77, 1/1
2 8 17 -10 4 27 70, 49/1 80, 5/1
3 9 15 0 6 28[a] 76, 49/1[a] 60, 2/3 4 9 17 0 5 29[a] 75, 9/1[a] 70, 2/5
[a]The glycosylation was performed under ultra-sonification.
3.3 Application of DMF-modulating glycosylation to thioglycoside acceptors
A unique feature of the DMF-modulating glycosylation is the entrapment of oxocarbenium ions as glycosyl imidates. This feature provides an opportunity for development of a new pre-activated glycosylation procedure. In a typical oligosaccharide synthesis, introduction of different anomeric functions to glycosyl donor and acceptor is required such that the activation of the former does not affect the later. Though the reactivities of glycosyl donor and acceptor can also be tuned to create reactivity disparity that allowing their coupling by reactivity-based glycosylation, this strategy requires a long protecting
group manipulation for building block preparation.10b, 11a, 31 The merit of a pre-activated glycosylation is to allow coupling of glycosyl substrates with the same anomeric function rendering the use of different anomeric function or the tuning of chemical reactivity, unnecessary. Such an approach not only shortens the synthetic steps in oligosaccharide synthesis, but it also paves the way to iterative one-pot glycosylation method.11a To the best of our knowledge, there is no pre-activation procedure that endows with -directing capability.15 To demonstrate the applicability of the DMF-modulating procedure, thioglycoside acceptors 30–40 were glycosylated with thioglycoside donors 2, 7, 8, and 9 following procedure B (Figure 9). Preparations and references of thioglycosyl acceptors 30−40 were given in experimental section. Table 6 summarizes the yields and -anomer ratios of corresponding glycosylation products 41–55.
A known side-reaction in glycosylations of thioglycosides is the transfer of the thio-acetal function from acceptor to donor.32 Gratifyingly, such transfer reaction did not occur in the DMF-modulating procedure perhaps due to masking of the reactive oxocarbenium ion by DMF molecule. The glycosylations in this study proceeded smoothly and the corresponding -anomers were furnished in 45 to 85% yields with high to excellent -selectivities. However, the reaction yields were on average lower than those produced from glycosylations of O-glycosides. We attributed this to the activation of thioglycoside product by residual NIS and/or some side reactions stemming from the imidate intermediates. To re-validate the -directing effect of DMF, the glycosylation of 36 with 2
was repeated by using a lesser amount of DMF (1.5 equiv) and the
-anomer ratio of glycosylation product 47 decreased sharply to 4/1 (data not shown). a) Thioglycosyl acceptors 30-40
b) Glycosylation products 41-55
R =
Figure 9. (a) Thioglycosyl acceptors 30–40; (b) Glycosylation products:
41–55.
Table 6. Results of glycosylation of thioglycosyl acceptors 30–40 using glycosylation procedure B
Entry Donor Acceptor T (˚C) Time (h)
-anomer (yield%),[a] [b]
1 2 30 -10 3 41 (60), 36/1
2 2 31 0 6 42 (55), 6/1
3 2 32 0 3 43 (55), 11/1
4 2 33 -10 3 44 (45), 11/1
5 2 34 -10 3 45 (85), 49/1
6 2 35 -10 2 46 (65), 12/1
7 2 36 0 4 47 (70), 49/1[31]
8 2 37 0 2 48 (50), 13/1
9 2 38 -10 3 49 (75), 19/1
10 2 39 0 4 50 (85), 49/1
11 7 40 -10 3 51 (56), 49/1
12 7 32 -10 6 52 (61), 49/1
13 8 35 -10 3 53 (55), 6/1
14 9 36 0 5 54 (50), 49/1[c]
[c]
[a] Glycosylation procedure B was applied and the yield (%) referred to isolated
-anomer. [b] Ratios of glycosylation products were determined by HPLC analysis (HPLC conditions was given in experimental). [c] The glycosylation was performed under ultrasonification.33
3.4 Mechanistic investigations
3.4.1 Isolate the hydrolysis product of glycosyl imidate
Scheme 12 Isolate the hydrolysis product of glycosyl imidate.
Previous 1H NMR spectroscopy was employed to detect the glycosyl imidates under different reaction context.34 We reasoned that the glycosyl imidates if formed should undergo hydrolysis in work-up to give the glycosyl formates; and isolation of such formate products would indicate the existence of imidates. Thus, thiogalactosides 1 and 2 were activated in the presence of DMF, and the reaction was subsequently quenched by triethylamine (TEA) without the addition of acceptor. Upon standard workup, -glycosyl formates (56, 57) could be isolated in 5 and 10%
respectively along with ca 80% of glycosyl hemiacetals, which were presumably the hydrolyzed products of glycosyl formates. Both glycosyl
yields (Scheme 12). Chemical identities of 56 and 57 were evidenced by (1) the chemical shifts of anomeric protons (6.40 ppm for 56, 6.47 ppm for 57); (2) 3Jcoupling constants of anomeric protons (3.4 Hz for 56, 3.6 Hz for 57); and (3) the characteristic chemical shifts of formate protons at 8.14 ppm for both 56 and 57. However, we were not able to obtain the corresponding -glycosyl formate, which might be attributed to its poor stability for standard isolation.
Try to prove the presence of -glycosyl imidate, we turned to real-time monitoring of the activation process by 1H NMR spectroscopy.
3.4.2 Real-time variable temperature NMR study
Since glycosyl imidate formation is the key step in DMF-modulating glycosylation, the detection of glycosyl imidate is crucial to support the proposed mechanism. In this regard, we prepared a simpler 4,6-O-benzylidene-2,3-di-O-methyl thiogalactoside 58, which was activated with NIS and TMSOTf promoters in CDCl3 and followed by the glycosylation of acceptor 59 using procedure B (Figure 10a). 1H-, 13C-, and HSQC-NMR spectroscopy of the reaction mixture were taken at 0, 90, and 180 min time points. Figures 9b-d showed selected regions of corresponding 1H NMR spectra. Comparing the spectra of the pre-activated reaction mixture at 0 min and the TMSOTf activated mixture at 90 min (Figure 10b and 10c), a new set of 1H NMR signals are clearly identified, including an anomeric proton at 6.39 ppm (3J = 3 Hz, 60-Ha), a benzylidene proton at 5.60 ppm (60-Hb), an imidoyl proton at 8.90 ppm (60-Hc), and N,N-dimethyl protons at 3.40, and 3.32 ppm
(60-Hd). These signals are generated from the presumably -glycosyl imidate 60.34a The relative downfield positions of 60-Ha,c,d indicate their close proximity to an electron-deficient center. Following the addition of acceptor 59, the signals stemming from imidate 60 vanished, and another two sets of signals emerged. One set includes an anomeric proton at 5.13 ppm (3J = 3 Hz, 61-Ha) and a benzylidene proton at 5.59 ppm (61-Hb) corresponding to the expected -glycoside 61. Another set (indicated by asterisks in Figure 10d) was originated from a -N-galactosyl succinimide, which is a common side-product produced in NIS promoted glycosylation.31
As the real-time NMR study provided evidence for the presence of the -glycosyl imidate, it is reasonable to propose the formation of
/-glycosyl imidates in DMF-modulating glycosylations. And the
-glycosyl imidate, due to a more reactive nature, reacts preferentially with the acceptor to give the -glycosylating product. At this time, we are not able to detect the presence of -imidate.
Figure 10. (a) Glycosylation of 58 with 59 following procedure B. (b) 1H NMR spectrum taken just prior to TMSOTf addition (0 min). (c) 1H NMR spectrum taken at 90 min following TMSOTf addition (90 min). (d)
1H NMR spectrum taken at 90 min after addition of 59.
58
58-Hb 58-Ha
59-Hc
59-Ha
59
59-Hb 58-Hb
58-Ha
59-Hd
60
Residual 60-Hb
60-Ha 60-Hb
3.4.3 Temperature profile using VT-NMR
Due to the life time of the -glycosyl imidate is short via NMR analysis, we try to trap it under a lower reaction temperature. In addition, a variable NMR study may also study the stability of -glycosyl imidate intermediate. Because of the melting point and boiling point of CDCl3 and DMF, the range of temperature allowed in these experiments is ranged from 50 to 50 ˚C.
From 50 to 40 ˚C, there is no significant signal appeared. We didn’t observe the signals of -glycosyl imidate. Furthermore, among this temperature range the signals of -glycosyl imidate still appeared, it indicate the high stability of it. At 50 ˚C, it became difficult to lock the NMR signals for further analysis.
Figure 11. Temperature profile diagram by using VT-NMR from -50oC to 50oC
3.4.4 DMF-d7 substitution experiment
In order to obtain more information about the reaction mechanism, we also did the DMF-d7 exchange experiment. In the presence of DMF, the donor of thioglycoside was pre-activated in CDCl3 at -10oC by NIS/TMSOTf promoters. After 30 minutes, we added DMF-d7 and
(a) 50oC
(c) -10oC (b) 40oC
(d) -40oC Hc
Hc
Hc
Ha
Ha
Ha Hb
Hb
Hb
Hd
Hd
Hd
examined the 1H spectra at 0 and 30 minute.
From spectra, we observed the signals of glycosyl imidate decreased when the DMF-d7 was added. This is estminated by comparing the ratio of anomeric proton (non-exchangeable) to the formamide proton (exchangneable), the ratios of H-1/formamide-H decreased from 1/1 to 1/0.2 (Figure 12). This information indicate that the -glycosyl imidate have an equilibrium with DMF. Or -glycosyl imidate react with DMF-d7
first, then -glycosyl imidate become -glycosyl imidate by equilibrium.
It remains too early to exclude the possibility of other mechanism.35 For elucidation, further experimental investigations are in progress.
O
Fig
3.5 Plausible mechanism of DMF-modulated pre-activated glycosylation
Scheme 13. Proposed mechanism of DMF modulating glycosylation.
On the basis of above experiment, we hypothesized that the activation of thioglycoside generates an oxocarbenium ion pair, which after trapped by a nucleophilic DMF, gives rise to an equilibrium mixture of --glycosyl imidates. Assuming that the -imidate is more reactive than its -counterpart; subsequent coupling of the -imidate with an acceptor produces the desired -anomer as a major product (Scheme 13). Since DMF plays as a modulating function in the reaction, we coined this new glycosylation strategy as a DMF-modulating glycosylation strategy.
4 Conclusions
In summary, a novel DMF-modulating glycosylation strategy is developed, which achieves excellent -selectivity in glycosylation by simple addition of DMF. Further elaboration leads to the development of a practical pre-activated -selective glycosylation strategy.
Considering the availability of DMF, we anticipate that the synthesis concept mentioned above will find broad application in oligosaccharide synthesis. This work is accepted for publication in Angewandte Chime international edition 2011.
5 Experimental
5.1 General experiment procedure:
Reagent-grade chemicals were purchased from commercial vendors and used without further purification. Dichloromethane (CH2Cl2) was dried by Asianwong solvent purification system (AWS-1000).
N,N-Dimethylformamide (DMF) was stocked with flame-dried molecular sieves (MS) under N2. Progress of reactions was monitored by thin-layer chromatography on silica gel 60 F-254 plate and visualized under UV illumination and/or by staining with acidic ceric ammonium molybdate or p-anisaldehyde. HPLC analysis was performed over Mightysil column (Si-60 250-4.6) and eluted with EtOAc/hexane/CH2Cl2 mixture at a 0.8 mL min-1 flow rate by the gradient pump (L-2130) and UV detector (L-2400) from Hitachi. Silica gel (Geduran Si-60, 0.063-0.200 mm) for chromatography was obtained from Merck. NMR spectra were recorded at 300 MHz and 75 MHz spectrometers in Brüker console or 500 MHz and 125 MHz in Varian console as specified. Sonification was provided by standard bench top ultra-sonicator (Branson 2210R-MT). Real time NMR study of glycosylation of acceptor 59 with donor 58 was performed with 500 MHz in Varian console. The chemical shifts were calibrated against the residual proton signal and 13C signals of deuterated chloroform. Coupling constants in Hz was calculated from chemical shifts of 1H NMR spectra. Acceptors 3, 10, 11, and 13 are commercially available, glycosyl donors 1,36 2,36 7,36 8,37 9,36 and glycosyl acceptors 12,38 14,39 15,39 17,39 30,39 31,15 32,17 33,40 35,15 36,41 37,42 and 4036 are prepared on the base of literature procedures.
5.2 General pre-activated DMF-modulating glycosylation procedure (procedure B).
Mixture of 2,3-di-O-benzyl-4,6-O-benzylidene-thiogalactopyranoside 239 (166 mg, 0.3 mmol, 1.5 equiv) and flame-dried molecular sieve (AW300) was suspended in dried CH2Cl2 (4.0 mL) such that the final concentration of 2 was 75 mM. Then, DMF (93 L, 1.2 mmol, 6.0 mol equiv) was added to the mixture. The resulting mixture was stirred at RT for 10 min and at -10 ˚C cooling bath for an additional 10 min.
Mixture of 2,3-di-O-benzyl-4,6-O-benzylidene-thiogalactopyranoside 239 (166 mg, 0.3 mmol, 1.5 equiv) and flame-dried molecular sieve (AW300) was suspended in dried CH2Cl2 (4.0 mL) such that the final concentration of 2 was 75 mM. Then, DMF (93 L, 1.2 mmol, 6.0 mol equiv) was added to the mixture. The resulting mixture was stirred at RT for 10 min and at -10 ˚C cooling bath for an additional 10 min.