Background and Originality Content

Carbohydrates play a crucial role in a myriad of biological processes and have emerged as key compounds in living systems and medical science.^[1] N -glycosides, a prominent class of carbohydrates, are extensively found in a wide array of pharmaceuticals, bioactive compounds, and natural products.^[2] They also play essential roles in numerous physiological and pathological processes.^[3] Although β -glycosylamino anomers are commonly encountered, a significant number of natural glycoproteins are characterized by anα -linkage at the anomeric center of their peptide motifs (Fig. 1a)^{[3j, 4]}. An exemplary case in point is trehazoline^[5], a trehalase inhibitor. Additionally, the diverse functional implications of these carbohydrate derivatives are further illustrated by compounds such as N -mannosyl tryptophan.^[6] This compound is a key agent in posttranslational protein modifications. Another example is Nephritogenoside^{[4b, c, 7]}, a glycopeptide from the glomerular basement membrane of rats, noted for its role in inducing glomerulonephritis.

Figure 1 . Stereoselective Synthesis of 2-deoxy-α-N -glycosides.

The modification of asparagine residues through 2-deoxyglycosylation has been shown to enhance the hydrolysis-resistant stability of peptide chains, thereby increasing both biological activity and selectivity. As a result, the N -(2-deoxyglycosyl)-amide motif holds significant value in the fields of biochemistry and pharmacology.^[8] Despite its significance, literature on synthetic methodologies for this structure remains sparse.^[9] Among the known synthetic strategies, the direct N -(2-deoxyglycosyl)-amide formation from nucleophiles and glycals stands out as a notably straightforward approach for generating 2-deoxysugar compounds. Particularly for the synthesis of sulfonamide and simple amide groups with the employment of various transition-metal catalysts, organocatalysts, and Brønsted acid catalysts^[10]. These methods predominantly facilitate the production of thermodynamically favored β -glycosyl amides. ^{[10b, 11]} However, the selective synthesis of thermodynamically disfavoredα -glycosyl amides, especially 2-Deoxy-α -N -glycosides, remains a formidable task.^[8d]

Transition-metal-catalyzed glycosylation reactions have emerged as highly efficient tools for synthesizing a wide range of glycosides and glycoconjugates. However, when it comes to the synthesis ofα -N -glycosides, methods utilizing transition-metal catalysis are currently limited. ^[12] Considering the significance of α -N -glycosides in medicinal chemistry, there is a strong need to develop an efficient method for their assembly. In this context, the crucial factor in achieving this transformation lies in the discovery of appropriate amidating reagents. In recent studies, Seo and Chang, Yu, and Zhu^[13]have identified 1,4,2-dioxazol-5-ones as effective electrophilic amidating reagents in Ni-catalyzed hydroamidation reactions. Building upon this knowledge, we propose the extension of Ni-catalyzed hydrofunctionalization to the hydroamidation of glycals with dioxazolones. This strategy would enable the direct synthesis of a diverse array of 2-deoxy-α-N-glycosides. As depicted in Figure 1c, we proposed a plausible mechanism: the reaction initiates with the syn-addition of glycals to NiH species, leading to the formation of anα -configured alkylnickel intermediate. Subsequently, Ni-nitrenoid is generated, and through an inner-sphere nitrenoid transfer, the C-N bond is established. Finally, the desired products are obtained upon protonation.

In this study, we have developed a highly efficient strategy for synthesizing a wide array of 2-deoxy-α -N -glycosides with exceptional α-stereoselectivity under mild reaction conditions. Notably, this reaction demonstrates a broad substrate scope and remarkable tolerance towards diverse functional groups. Furthermore, we have extensively explored the synthetic applicability of this method through gram-scale experiments.

Results and Discussion

Table 1. Reaction Development of Ni-Catalyzed Hydroamination from Glycals with Dioxazolones ^a

^a Reaction conditions: NiBr₂·DME (5 mol%), L (5 mol%), NaI (30 mol%), (MeO)₃SiH (0.8 mmol, 4.0 equiv), 1a (0.2 mmol, 1.0 equiv), 2a(0.4 mmol, 2.0 equiv), 1,4-Dioxane:THF (4:1, 0.27M), t -BuOH (0.6 mmol, 3.0 equiv), 20 °C, 24 h. ^b yield of 3awere determined by ¹⁹F NMR with 1-chloro-4-fluorobenzene as an internal standard. ^cYield of isolated product.

Table 2 Substrate Scope of Ni-Catalyzed Hydroamination from Glycals with Dioxazolones ^a

^a Reaction conditions: 1 (0.2 mmol, 1.0 equiv), 2 (0.4 mmol, 2.0 equiv), NiBr₂·DME (5 mol%), L (5 mol%), NaI (30 mol%), (MeO)₃SiH (0.8 mmol, 4.0 equiv),1,4-dioxane:THF (4:1, 0.27M), t -BuOH (0.6 mmol, 3.0 equiv), 20 °C, 24 h.

Our initial investigations focused on the α-selective N-glycosylation of tri-O-benzyl-D-glucal (1a ) using 3-(4-fluorophenyl)-1,4,2-dioxazol-5-one (2a ) as the amidating reagent. We discovered that NiBr₂∙DME, combined with the chiral pyridoxazoline ligand (L4 ) and trimethoxysilane, delivered the desired hydroamidation product 3a in an isolated yield of 82% with exceptional α -selectivity (entry 1). Subsequently, we explored the impact of ligand steric hindrance on the reaction’s yield. By increasing the steric hindrance of the ligands (L2 -L4 ) through modifications in the substituents on the pyridine, we observed significantly improved yields. However, other ligands (L5 -L6 ) resulted in notably lower yields (entries 2-6). Furthermore, when using alternative nickel sources such as NiCl₂∙DME, lower yields were obtained (entry 7). Dimethoxy(methyl) silane and di-ethoxy(methyl) silane were found to be less effective (entries 8-9). The addition of other proton sources, such as H₂O and CH₃OH, showed improvements in yield. However, t -BuOH proved to be optimal (entries 10-12). Of note, the inclusion of a catalytic amount of NaI as an additive was crucial for the reaction (entry 13), although the exact role of NaI is still being investigated.

With the optimized reaction conditions in hand, we proceeded to investigate the substrate scope of this reaction, and the summarized results can be found in Scheme 2. Initially, phenyl dioxazolones with mono- and di-substituted groups were examined. It was observed that phenyl dioxazolones substituted with electron-withdrawing groups smoothly underwent the reaction, resulting in the formation of the desired products (3a -3d ) with good yields and excellent α -selectivity. The absolute α-anomeric configuration of the products was confirmed through X-ray crystal structures of3a and 3d . Unsubstituted phenyl dioxazolones (3e ) provided the target product with good yields and excellentα -selectivity, while phenyl dioxazolones with electron-donating substituents such as methyl (3f ), methoxy (3g ), and phenoxy (3h ) were also investigated, demonstrating high yields and stereoselectivity. Furthermore, phenyl dioxazolones with di-substituted groups were converted to theirN -(2-deoxyglycosyl)-amide counterparts (3i and3j ) in moderate yields. Moreover, dioxazolones with fused-ring systems also yielded the target products (3k and 3l) in excellent yields. To further strengthen the synthetic utility of this method, theN -(2-deoxyglycosyl)-amides were also successfully prepared from dioxazolones derived from probenecid (3m ). However, alkyl dioxazolones were found to be incompatible with this transformation. The hydroamidation reaction proceeded smoothly for a diverse range of glycals with various functional groups, as demonstrated in Table 2. Most of the glycals derived from D-glucal reacted smoothly with 3-(4-fluorophenyl)-1,4,2-dioxazol-5-one (2a ) under standard conditions yielding good to high results with excellentα -selectivity (3o-3r ). Notably, high yields and good stereoselectivity were also obtained with the unprotected primary alcohol group on the glycols (3p ). Furthermore, the reaction was also successful with glycals derived from galactal. When using these substrates, the reaction proceeded smoothly, resulting in moderate yields and excellent stereoselectivity (3s -3v ).

To demonstrate the practical application of this protocol, a gram-scale reaction was conducted, resulting in the synthesis of the desired product 3a with moderate yield and excellent selectivity, as depicted in Figure 2a. Moreover, we extended the utility of this methodology for the synthesis of 2-deoxy sugar amino acids (SAAs), as illustrated in Figure 2b. Initially, readily available glycals were subjected to oxidation and methylation, yielding 6-carboxy glycals (4 ). By employing our developed method, the target compound, 1-amino-2-deoxy-α-D-galacturonic acid derivative (5 ), was obtained directly in a single step with moderate yield and high stereoselectivity. Sugar amino acids represent highly substituted polyfunctional building blocks that hold significant potential in drug discovery and materials science.^[14]

Figure 2 . Gram-Scale reaction and synthetic application.

Gram-scale reaction conditions: NiBr₂·DME (5 mol%),L (5 mol%), NaI (30 mol%), (MeO)₃SiH (12 mmol, 4.0 equiv), 1 (3 mmol, 1.0 equiv), 2 (6 mmol, 2.0 equiv), 1,4-dioxane:THF (4:1, 0.27 M), t -BuOH (9 mmol, 3.0 equiv), 20 °C, 24 h.

Conclusions

In conclusion, we have successfully developed a nickel-catalyzed hydroamination reaction for the stereoselective synthesis of 2-deoxy-α-N-glycosides from glycals and 1,4,2-dioxazol-5-ones. Notably, the reaction proceeds under mild conditions, exhibits a broad substrate scope, excellent α -stereoselectivity, and remarkable tolerance towards diverse functional groups. Furthermore, we have exemplified the versatility of this reaction through the one-pot, stereoselective synthesis of 2-deoxy sugar amino acids, which hold significant promise in applications in drug discovery and materials science.

Experimental