Glycosylation is a common tool in drug development. The introduction of sugar units can often change the water solubility, targeting, metabolic stability and other properties of the target molecule. However, the chemical method of introducing sugar units into the target molecule still faces many challenges. Most glycosylation reactions go through oxonium intermediates or their equivalents, and the mechanism often wanders between the two extremes of SN1 and SN2. Affected by multiple factors such as temperature, substrate reactivity, and additives, the stereoselectivity of the reaction is difficult to predict. In addition, if sensitive reagents or unstable substrates are required for glycosylation, it will also bring many operational inconveniences.
Recently, the team of Dawen Niu from Sichuan University published an article entitled "Palladium catalysis enables cross-coupling–like SN2-glycosylation of phenols" in Science.
Palladium-catalyzed cross-coupling is highly robust and has become an indispensable tool in organic synthesis. If glycosylation reactions can be as modular as palladium-catalyzed cross-coupling reactions, the Synthesis of sugar compounds will be greatly simplified and their downstream functions can be studied more. However, palladium catalysis is generally effective in sp2-hybridized carbon centers, while glycosidic bonds are mostly sp3-hybridized C-O bonds. Based on this, the team developed a new mode of activating glycosyl donors.
The authors used glycosyl sulfides containing aryl iodides as donors and initiated Pd(0)-mediated oxidative addition (OA), where the key Pd(II) OA intermediate was transformed from an arylating agent (Csp2 electrophile) to a glycosylation agent (Csp3 electrophile). This approach inherits many advantages of cross-coupling reactions, including operational simplicity and strong functional group tolerance. At the same time, the SN2 mechanism is retained, which is the fundamental reason for the high stereoselectivity of the reaction. The single crystal structure of the key intermediate shows that the coordination of the sulfur atom to the palladium center polarizes the C-S, making it sufficiently electrophilic to attack the phenol compound.
Fig. 1 Glycosylated phenols: background and synthetic methods. (Deng, et al., 2023)
Phenols are abundant in both natural and artificial compounds. Glycosylation of phenols is complicated because they exhibit moderate nucleophilicity compared to alcohols under acidic conditions. Moreover, phenols are environmental nucleophiles that may give rise to either O-glycosylated or C-glycosylated products. Pd-catalyzed cross-coupling is usually initiated by Pd(0)-mediated OA, followed by ligand exchange and reductive elimination to afford the desired products. Considering the limitations of this cycle in activating/building Csp3 centers, the author designed a strategy that uses o-iodobiphenyl-substituted sulfides as Glycosyl Donors. Its aryl iodide unit readily undergoes OA with the Pd(0) catalyst to form an OA complex that acts as an efficient glycosyl (Csp3) electrophile, likely due to its propensity to undergo Csp2-S reductive elimination. The nucleophilic attack of phenolate 10 on 12 proceeds via a clean and general SN2 mechanism, leading to the inversion of the glycosyl center and the generation of 13.
Starting from a model reaction between sulfide α-S-glycoside or β-S-glycoside and 4-methoxyphenol to generate O-glycoside, conditions were established to prepare β-O-glycoside/α-O-glycoside from α-S-glycoside/β-S-glycoside in high yield with the formation of dibenzothiophene as a byproduct. β-O-glycoside was obtained from α-S-glycoside, while α-O-glycoside was formed if β-S-glycoside was used. The complete inversion of the glycosidic center in α-S-glycoside or β-S-glycoside indicated that this Pd-catalyzed Glycosylation proceeded via an SN2-type mechanism. Control experiments were also performed to determine the factors that affect the performance of this method.
Glycosylation methods rarely maintain high efficiency for a wide range of substrates because the reaction mechanism (and outcome) often varies with the reactivity of the donor and acceptor. This method is adaptable to different glycosyl units, providing two possible stereoisomers with high purity. The experimental results demonstrate the excellent functional group tolerance of this OA-initiated glycosylation method. Obtaining the corresponding O-glycosides with high stereochemical purity by conventional methods would be tedious due to the lack of suitable donors, the absence of stereospecific auxiliary agents, and the sensitivity of the products to acid-promoted hydrolysis.
To demonstrate the utility of this approach, it was applied to the modification of natural products and commercial drugs. Pd-catalyzed SN2 glycosylation can be performed in tandem with other Pd-catalyzed cross-coupling reactions to provide carbohydrate-containing compounds via a one-pot, multistep process, in which a single Pd(0) complex catalyzes two distinct steps.
Fig. 2 Mechanism study. (Deng, et al., 2023)
The acetyl-protected sulfide donor 44 reacted smoothly with Pd(PPh3)4, and the OA complex 45 could be separated from the mixture by column chromatography. The convenience of the OA step may be attributed to the sulfur atom in 44, which can be pre-coordinated to Pd(0). Similar to other classical Pd(II)OA complexes (47), the palladium center in 45 adopts a square planar configuration with two neutral ligands occupying the para positions. Solid state structural solution of 45 shows the formation of a positive charge at the C1 position. Two potential pathways were investigated by DFT calculations: in principle, reductive elimination of the C-S bond in 45 to produce the sulfonium Int-I, followed by phenolate attack from 46 to afford the glycoside 47. Alternatively, direct attack on the phenolate to convert 46 to 45 via TS-II, followed by reductive elimination of the C-S bond in Int-II (45), would also afford 47, yielding dibenzothiophene as a byproduct and regenerating the Pd(0) catalyst. By comparing the free energies of species TS-I and TS-II, the authors observed that the pathway via TS-II exhibited a lower barrier. Presumably, the coordination of the sulfur atom to the Pd center polarizes the C-S bond in 45 and renders it electrophilic enough to withstand phenolate attack. The relative reactivity of various phenols bearing different para-substituents (48) in the glycosylation reaction was also compared. In internal competition experiments, those with electron-withdrawing substituents (49) reacted faster, presumably because they are more easily deprotonated under alkaline conditions. Through external competition experiments, it was found that the turnover frequency inferred from the conversion rate of β-S-glycosides increased with the increasing electron-withdrawing ability of the para-substituent. In the absence of phenol 48, essentially no reaction occurred. These results indicate that the phenolate nucleophile is involved in the step that determines the conversion rate, further supporting the pathway through TS-II. Although more experiments are needed to elucidate the details of the mechanism, the ability of Pd-containing OA complexes to act as glycosyl (Csp3) electrophiles is quite general.
In summary, we report a novel Pd-catalyzed SN2 glycosylation approach starting from OA. The utility of this approach is demonstrated in a general and simple SN2 glycosylation of phenols, which is an outstanding challenge in the synthesis of O-glycosides. The versatility and mildness of this approach are further demonstrated in several one-pot, multi-step, and multi-component reactions. This study will bring opportunities for Pd-mediated glycosylation and advance Carbohydrate synthesis and its applications in various fields.
Reference
