Editing sugars: decatungstate photocatalyzed site- and stereoselective C–H functionalization in β-fucosides

Abstract

C–H bond functionalization via (photoinduced) Hydrogen Atom Transfer (HAT) catalysis is emerging as a powerful eco-sustainable tool for sugar editing, albeit challenges in governing the reactivity patterns and site-selectivity remain. Quite surprisingly, no functionalization of the C-5 position in pyranoses via C–H activation has been described to date. We herein report the development of an efficient methodology for the site-selective and stereoselective alkylation of position 5 of β-fucosides by means of decatungstate photocatalyzed C(sp³)–C(sp³) bond formation. The experimental work is accompanied by spectroscopical studies based on laser flash photolysis and supplemented by computational simulations, offering insights into the reasons underlying the selectivity profile of the reported transformations.

---

Green foundation

1. The present work develops a sustainable photocatalytic approach for the preparation of C-5 alkylated fucosides directly from the corresponding β-fucosides, with no need for pre-activation of the starting glycosides.

2. Besides providing an unprecedented transformation of the pyranose core, this approach represents a 100% atom economy process that avoids the use of any additives and takes place via site-selective C–H cleavage in β-fucosides by adopting an easily prepared photocatalyst.

3. Future research will focus on applying this sustainable strategy for the late-stage functionalization of further glycosides.

Introduction

Chemical modification of carbohydrates plays a significant role in materials science, biotechnology, as well as in drug design and development. As for the latter field, sugar elaboration allows enhancement of drug activity, stability, and specificity, and is key in studying molecular recognition events and developing targeted therapies like antibody–drug conjugates.¹ However, modifying sugars is challenging due to their complex structure and reactivity, associated with chemo-, regio- and stereo-selectivity issues. In the case of pyranose derivatives, the modification may take place at carbons C-1–C-5 in the ring or at the exocyclic C-6 position (Fig. 1A) where metal-catalysis can be helpful in this respect.^2–4 The manipulation of sugars typically occurs by substitution at the anomeric position, where the high reactivity of the hemiacetal functionality enables selective modifications.^5,6 This strategy forces the preactivation of the anomeric carbon by the presence of appropriate leaving groups.⁷ Photoinduced glycosylation via glycosyl radicals (Fig. 1B) has attracted increasing interest in recent years due to the compatibility of such intermediates with a wide range of functional groups and the mild conditions required for promoting the process. Typically, this strategy involves the installation of a redox active moiety at the anomeric position, which is then lost in the photocatalytic event.^7–13

Fig. 1 Derivatization of monosaccharides: (A) potential derivatization sites. (B) Typical strategy for the generation of glycosyl radicals and ensuing derivatization at C-1. (C) Selected photocatalytic systems for the site-selective C–H elaboration at different positions.

The modification of carbohydrates’ skeleton at a non-anomeric position is traditionally achieved by functional group manipulation of the native hydroxy groups.¹⁴ However, direct C–H bond functionalization is currently emerging as a powerful and sustainable tool for sugar editing, including the non-enzymatic synthesis of rare carbohydrates, many of which are present in naturally occurring antibiotics and other drugs.^15–19 Particularly in the context of designing glycomimetics, modifications of monosaccharides’ skeleton via C–H activation can be transformative, because they enable functionalization of the sugar scaffold while preserving the crucial pattern of hydroxyl groups for target recognition.^20,21

However, while C–H functionalization methodologies do not need any pre-activation of substrates and thus decrease the number of synthetic steps, they can be extremely challenging in terms of reactivity and site-selectivity; therefore, despite the huge synthetic potential, such approaches are still underdeveloped for the functionalization of carbohydrates.

Photocatalyzed hydrogen atom transfer (HAT) reactions represent a valuable eco-sustainable tool to perform a clean C–H homolytic cleavage, even in complex substrates, for the generation of C-based radicals.²² As a matter of fact, this approach has been recently applied for the C–H functionalization of both protected and unprotected monosaccharides. The resulting scenario is quite complex to summarize, since hydrogen abstraction depends on several factors, which include the nature and the bulkiness of the hydrogen abstracting agent, the structure of the monosaccharides as well as the steric hindrance (or the electronic effects) exerted by protecting groups, if any, and the presence of additives (e.g. Brønsted base co-catalysts).^15–19

Fig. 1C illustrates selected examples to clarify this issue. First, protected galactopyranose I was fluorinated at C-2 to deliver product II by direct HAT employing an aromatic ketone (xanthone) as the photocatalyst (PC).²³ Notably, when changing the PC, the selectivity in the hydrogen abstraction on the same derivative I dramatically changed. A C-3 selectivity was exclusively observed by using anthraquinone,^24,25 whereas a completely selective C-4 functionalization took place when adopting the decatungstate anion, either as tetrabutylammonium (TBADT; for C–C bond formation)²⁴ or sodium (NaDT; for fluorination)²⁵ salt. Overall, such strategies offered straightforward access to Giese adducts III and IV in good yields.²⁴ Similarly, the photocatalytically generated quinuclidine radical cation operated a HAT on 6-deacetylated galactose derivative V leading to side chain modification at C-6 (delivering lactone VI).²⁶ The same reaction conditions when applied to derivative VII (α-methyl glucoside) restored the derivatization at C-3 (on the route to VIII).²⁷ These examples confirm the intricacies related to predicting the fate of the C–H functionalization on carbohydrates.^28–30 Nevertheless, the study of the factors that lower the bond dissociation energy and decrease the kinetic barrier for hydrogen atom abstraction allowed for a radical-mediated site-selective epimerization of unprotected monosaccharides by HAT,^19,31–33 where the concept of network control has been proposed to account for the observed selectivity.³⁴ As apparent from Fig. 1, the C-1 ^35,36 and C-5 positions are the most challenging to modify via direct C–H functionalization, even if the C–H elaboration at the anomeric position of C-glycosides via radical bromination (typically with NBS) has long been known.^37,38

Molecular editing at the C-5 position of pyranoses is of particular significance, as this position is often involved in enzyme–substrate interactions³⁹ and in molecular recognition of carbohydrate-binding proteins. This transformation could expand the chemical space available for carbohydrate-based drugs⁴⁰ by adding functional groups to the carbohydrate skeleton. During our research program on the fucose-specific bacterial lectin BC2L-C,^41–44 for instance, we became interested in modifying the position 5 of β-fucosides, to simultaneously reach sterically divergent secondary sites of the protein. L-Fucose is a component of blood group antigens, glycoproteins, and glycolipids involved in cell communication and pathogen recognition. On the other hand, the D-enantiomer is usually found in plants and in some microorganisms.⁴⁵ Accordingly, modifying fucose at C-5 is key to expanding the chemical space for inhibitors and probes targeting these biological processes.

A literature survey indicates that the photocatalyzed C–H manipulation in 6-deoxy-monosaccharides, such as fucopyranosides (XI) or rhamnopyranosides (IX, XIII), is quite rare (Fig. 2). In such cases, a Ph₂BOH co-catalyst was adopted to direct selectivity.⁴⁶ It is impressive to note that the photocatalyzed reaction between methyl α-L-rhamnopyranoside IX and methyl acrylate gave lactone X via exclusive C–H elaboration at C-2 (Fig. 2a).⁴⁶ In contrast, the site-selectivity diverted when the same reaction conditions were applied to methyl α-L-fucopyranoside XI, wherein a C-4/C-3 = 5.3 : 1 mixture of adducts XII resulted (Fig. 2b).⁴⁶ Moreover, in the attempt of obtaining valuable 5-C-bromosugars, a radical C-5 bromination on methyl tetra-O-acetyl-α-L-rhamnopyranoside XIII dramatically failed (Fig. 2c).⁴⁷

Fig. 2 Photoinduced C–H elaboration in (a) rhamnopyranosides or (b) fucopyranosides. (c) Attempted bromination at C-5 in rhamnopyranosides. (d) This work: functionalization at C-5 in fucopyranosides.

We then realized that, to the best of our knowledge, the direct C–H functionalization at position 5 of monosaccharides is still lacking. Based on our experience on TBADT photocatalysis,^48,49 we envisaged that this easily prepared photocatalyst could be useful for such purposes, provided that the right protecting groups/substituents are present in the starting fucopyranoside. Herein, we report the development of an efficient methodology for the selective alkylation of position 5 of β-fucosides by a TBADT photo-mediated Giese type alkylation (Fig. 2d).

Results

Preparative experiments

To achieve C-5 editing of β-fucosides, we reasoned that protection of the starting pyranoside as a β-tetra-acetate (Fig. 2d; XV, R = OAc) could be an interesting starting point on different grounds. First, we surmised that a β-OAc group at the anomeric position would not sterically hinder C-5 activation (compare the α-OAc group in XIII).^47,50,51 Second, due to the electrophilicity of the hydrogen abstractor TBADT (featuring an oxygen-centered radicaloid character),⁵² the deactivation exerted at C-1 by the presence of the acetoxy group may direct the selective formation of an α-oxy radical at C-5.

Indeed, when β-tetra-O-acetylfucose 1-βOAc was treated with 1.1 equiv. of methyl acrylate 2a in the presence of 5 mol% TBADT in CH₃CN upon 370 nm irradiation for 20 h, 55% conversion of 1-βOAc resulted, leading to 3-βOAc as the exclusive product (97% yield brsm; Fig. 3A). The structure of 3-βOAc was safely assigned based on NMR spectroscopy, showing the disappearance of the H-5 proton of fucose and the presence of a quaternary carbon at 77.3 ppm (see the SI, section 2.3, for the full spectral assignment). NOESY experiments showed clear cross peaks between the axial protons on the ring (H-1 and H-3) and the protons on the newly introduced alkyl chain (Fig. 3B), while no correlation was observed between H-6 (Me) and H-1 or H-3. The NOESY crosspeak observed between fucose H-4 and the ester alkyl chain further supports the regiochemistry assignment.

Fig. 3 (A) TBADT promoted Giese-type alkylation of β-tetra-O-acetylfucose 1-βOAc with methyl acrylate 2a. Isolated yield of 3-βOAc based on consumed 1-βOAc; (B) NOESY experiments allowed establishing the C-5 configuration of 3-βOAc.

Thus, the TBADT photocatalyzed alkylation of 1-βOAc was fully regio- and stereoselective. Encouraged by these results, we moved to extend the scope of this remarkable transformation to a series of purposely prepared (see the SI, section 2.3) tri-O-acetyl-β-fucosides 1-βR, bearing different substituents at the anomeric position (Fig. 4A). For this set of experiments, all reactions were performed again in the presence of 2a and 5 mol% TBADT, but shortening the reaction time to 7 h, since prolonged irradiation did not improve 1-βR conversion. Starting from β-methyl fucoside (1-βOMe), the ester 3-βOMe was obtained in 64% yield brsm (87% 1-βOMe conversion), together with a by-product derived from the incorporation of 2a via C–H cleavage of the OMe group (10% yield brsm). Notably, no alkylation of the anomeric position was observed, even upon prolonging the reaction time (20 h), indicating that the presence of an acetoxy group is not mandatory to avoid the functionalization at C-1. Overall, a good conversion was observed for essentially all 1-βR tested (Fig. 4A). The yields largely depended on the substituent at the anomeric position, which, in some cases, induced the formation of some oligomeric products (as in the case of 3-βN₃).

Fig. 4 Scope of (A) β-fucosides and (B) alkenes. Reaction conditions: 1-βR (1 equiv., 0.1 M), methyl acrylate 2a (1.1 equiv.), TBADT (5 mol%) under 370 nm irradiation in MeCN for 7 h. The C-5 configuration of all products was established by NOESY experiments. 3-βR isolated yield brsm and 1-βR % conversion (in brackets) are reported. Conversions were determined by ¹H-NMR of the crude reaction mixtures. ^a Alkylated by-product on OMe was also formed (10% yield); ^b 20 h irradiation; ^c Yields were estimated by ¹H-NMR analysis of the crude reaction mixture since 3-βNHCOPh was inseparable from the remaining 1-βNHCOPh; ^d 3 h irradiation. An oligomeric adduct was also formed in 12% yield.

The reaction was particularly effective with the 1-cyano-fucoside 1-βCN, the phenylthio-fucoside 1-βSPh, and 1-carbomethoxy fucoside 1-βCO₂Me, which afforded the corresponding adducts 3-βCN, 3-βSPh and 3-βCO₂Me in 90%, 70% and 80% yields, respectively. The reaction was likewise tolerant to the presence of an N-amide group at the anomeric position (1-βNHCOPh). Moreover, in the preparation of 3-βN₃, the reaction time could be shortened to 3 h. In all cases, a single isomer was isolated and NMR experiments (see data in the SI) evidenced that the reaction showed the same site selectivity and stereoselectivity observed for 1-βOAc.

Further experiments to improve the conversion of 1-βR and/or the yield of 3-βR by varying the reaction medium, the catalyst loading, etc., were unsuccessful. However, the process did not take place in the absence of either the photocatalyst or light (Table S1).

Substrates 1-βN₃, 1-βCN and 1-βCO₂Me were then selected to test the alkene scope of the reaction by using methyl vinyl ketone 2b, acrylonitrile 2c, 2-vinyl pyridine 2d and phenyl vinyl sulfone 2e, as the reaction partners (Fig. 4B). Also in this case, the reactions were consistently performed by adopting the standard conditions (5 mol% TBADT, 7 h irradiation). Notably, a fully regio- and stereoselective derivatization at C-5 was observed, as assessed by NOESY experiments (see the SI). Thus, functionalization of 2b with 1-βN₃, 1-βCN and 1-βCO₂Me proceeded with conversions similar to those observed with 2a (compounds 4 were formed in up to quantitative yield). As for 2c and 2d, although the yields of compounds 5 and 6 are in the 77–95% range, the conversion of the fucoside did not exceed 50%. Further optimization was not beneficial for the reaction outcome (Table S2). Green metric parameters (PMI, E factor, AE end EcoScale; see the SI, section 5) have been calculated for selected reactions. The values obtained range from good to excellent with 100% atom economy.

Additional control experiments were performed to elucidate the reaction mechanism and gain insights into the observed selectivity. First, to clarify the radical nature of the process, we repeated selected experiments on 1-βOMe, 1-βN₃, 1-βCN and 1-βCO₂Me in the presence of 2 equiv. of TEMPO under otherwise standard conditions (Michael acceptor omitted; see Scheme 1). In all cases, the corresponding C-5 adducts 8-βR were isolated in variable yields as the only product and NMR experiments (NOESY) demonstrated that the reaction takes place in a completely regio- and stereoselective fashion.

Scheme 1 Trapping experiments with TEMPO. ^a Isolated yields (1-βR % conversion). ^b 1 h irradiation.

We then tested the reactivity of the fucosides depicted in Chart 1 in the model reaction with 2a. Notably, no reaction occurred when the α-fucoside 1-αOAc (rather than the β-anomer 1-βOAc) was subjected to the standard photocatalytic conditions, highlighting a dramatic effect of the anomeric configuration on the reaction course. Similarly, the α-methyl fucoside 1-αOMe did not react even upon prolonging the irradiation (24 h). Notably, no TEMPO adducts were formed from substrates 1-αOAc and 1-αOMe.

Chart 1 Control reactions on substituted fucosides (0.1 M) under TBADT photocatalyzed conditions in the presence of 2a (1.1 equiv.).

The effect of the protecting groups on the photoreactivity of 1-βCN was also investigated. Thus, the unprotected cyano derivative 9 gave rise to a complex mixture of products and changing the acetoxy groups (in compound 1-βCN) to the silyloxy (compound 10) or benzyloxy (compound 11) groups led again to a complex mixture of products (Chart 1). The latter experiments safely confirmed the crucial role of the OAc protecting groups in directing the process toward a selective transformation.

Deuteration experiments

To gain further insights into the reasons underlying the observed selectivity, compound 1-βCN and TBADT (5 mol%) were irradiated in deuterated media, both in the absence and in the presence of 2a (see section 4 in the SI). In the absence of 2a, adoption of a MeCN/D₂O mixture as the solvent resulted in partial deuteration of 1-βCN, with deuterium selectively installed at C-5, as observed by ¹H-NMR analysis. No deuteration took place in CD₃CN. On the other hand, the presence of 2a (tested only in MeCN/D₂O) mainly led to monodeuterated 3-βCN-D, where deuteration occurred at the α-position with respect to the ester moiety (¹H-NMR).

LFP experiments

Laser-Flash Photolysis (LFP) experiments were also conducted on selected substrates (1-βOMe, 1-βN₃, 1-βCN and 1-αOMe), to investigate the kinetics of the C–H cleavage step: in particular, the quenching rates of the excited state of TBADT (labelled wO)⁵² by the above-mentioned fucosides were measured (detailed methodologies are provided in section 3 of the SI). For each sugar, a Stern–Volmer (SV) plot was constructed and a kinetic constant for the HAT step was determined. SV analysis revealed, for β fucosides 1-βOMe, 1-βN₃ and 1-βCN, high quenching rates, resulting in kinetic constants of 4.4 × 10⁸ M⁻¹ s⁻¹, 6.3 × 10⁸ M⁻¹ s⁻¹, and 4.9 × 10⁸ M⁻¹ s⁻¹, respectively, which are comparable to those already observed for excellent hydrogen donors (e.g. isopropanol⁵³ or various aldehydes⁵⁴). In contrast, as shown in Fig. 5, 1-αOMe is characterized by a remarkably lower kinetic constant (1.4 × 10⁷ M⁻¹ s⁻¹), in accordance with the experimental behaviour observed for this substrate.

Fig. 5 Stern–Volmer plots and quenching constants for 1-βOMe and 1-αOMe measured at 25 °C in MeCN.

Computational investigation

To get further insight into the observed reactivity patterns, we performed computational work on reference compounds 1-βOAc and 1-αOAc. In particular, we determined the reaction parameters characterizing the C–H cleavage at positions C-1 and C-5 of named fucosides by the oxygen-centered radical tBuO˙, a convenient prototype of a hydrogen abstractor species, mimicking the behaviour of the decatungstate reactive state.⁵⁴ To this end, we employed density functional theory at the ωB97xD/def2TZVP level of theory to optimize the geometries of relevant compounds, radical intermediates and transition states, with MeCN as the reaction medium modeled through an implicit approach (see section 6 of the SI for further details). Fig. 6 shows both the energy barrier (ΔG^‡; solid bars) and energy change (ΔG; dashed bars) values of the modeled hydrogen atom transfer processes, offering a measure of the associated kinetic and thermodynamic parameters, respectively.

Fig. 6 Energy barrier (ΔG^‡; solid bars) and energy change (ΔG; dashed bars) values for the modeled hydrogen atom transfer processes, as from calculations at the ωB97xD/def2TZVP level of theory in bulk acetonitrile (see section 6 in the SI for additional details).

As apparent from the graph, all the studied C–H cleavage reactions are exergonic in nature, with the radical intermediates at C-5 being similarly stabilized in 1-βOAc and 1-αOAc (ΔG around −10 kcal mol⁻¹), and in both cases to a larger extent than the corresponding radicals at C-1. As for the anomeric position (C-1), the radical formed from the β-isomer 1-βOAc benefits a larger stabilization (ΔG ∼−7.8 kcal mol⁻¹) than that formed from the α-isomer 1-αOAc (ΔG ∼−5.0 kcal mol⁻¹), as expected based on anomeric stabilization of the radical.^55,56 Thermodynamic parameters, however, merely give an indication about the feasibility of the studied processes, while kinetic parameters are typically more important to elucidate selectivity profiles. Along this line, in 1-βOAc, the C–H cleavage at C-5 (ΔG^‡ ∼+12.9 kcal mol⁻¹) has to confront a smaller energy barrier than that at C-1 (ΔG^‡ ∼+19.6 kcal mol⁻¹), with a quite large ΔΔG^‡ (>6.7 kcal mol⁻¹) that fully aligns with the observed regioselectivity. A similar situation is observed in 1-αOAc, although the ΔG^‡ associated with C–H cleavage at C-5 is larger (∼+13.8 kcal mol⁻¹), while that at C-1 is smaller (∼+17.4 kcal mol⁻¹), resulting in a reduced ΔΔG^‡ (∼3.6 kcal mol⁻¹) for regioselectivity.

Discussion

The results reported above show that C-5 functionalization of β-fucosides can be achieved by TBADT-photocatalyzed HAT activation, in a way well beyond our initial expectations. In particular, although the reaction appears to require acetylation of positions 2–4 of the sugar, it is surprisingly tolerant of various anomeric substituents, provided that they are not in the α position (Fig. 4 and Chart 1). From a preparative point of view, it is noteworthy that excellent conversions and yields were obtained for the β-phenylthiofucoside 3-βSPh, which in principle can be directly activated as a (C-5 modified) fucosyl donor.⁵⁷ Similarly remarkable is the reactivity of the β-fucosyl azide 1-βN₃, which gave a satisfactory yield of the C-5 Giese alkylation product 3-βN₃, despite the known photochemical lability of alkylazides.⁵⁸ In this regard, glycosyl azides are important intermediates in the synthesis of glycosylamides and other N-glycosyl derivatives.⁵⁹

In contrast, no reaction occurred on any of the α-substrates tested (1-αOAc and 1-αOMe, Chart 1). LFP experiments performed on 1-αOMe (Fig. 5) showed that this compound is kinetically inert vs. the excited state of TBADT, which is consistent with the strong steric effects typically observed for the bulky decatungstate anion.⁴⁸ It is perhaps more surprising that C-5 regioselectivity persists in the reaction of the 1-βOMe anomer, even if the anomeric position could now provide a radical potentially doubly stabilized by the two acetal oxygens. Indeed, it is well-established that axial radicals in anomeric positions are both energetically favored and more nucleophilic than their equatorial counterparts, due to stereo-electronic effects.^60,61 However, it is likewise well-known that anomeric interaction effects are not additive; hence, the stabilizing effect of the second oxygen atom at C-1 should not be expected to direct the regioselectivity.^62,63 Additionally, thermodynamic stability of the radical intermediate is clearly not the determining factor in this reaction, otherwise the captodative anomeric radical at C-1 in 1-βCN and 1-βCOOMe would shift regioselectivity from C-5 to C-1 – which does not occur. As it happens, C-5 selectivity persists with all the β-anomeric substituents we tested, irrespective of their electronic and steric characteristics, as also supported by the TEMPO trapping experiments (Scheme 1). Thus, the observed reactivity is more consistent with kinetic control in the C–H activation step, which is also consistent with our computational studies (Fig. 6). On the one hand, the markedly different energy barriers for C–H cleavage determined in 1-βOAc support the complete selectivity toward C-5 functionalization observed for such substrate. At the same time, the lack of reactivity shown by 1-αOAc can be traced back to the less favorable C–H cleavage at C-5, possibly due to the steric hindrance exerted by the axial acetoxy group installed at C-1; such an effect is expected to be even more pronounced in the case of the bulky decatungstate anion,^48,49 compared with the simplified hydrogen abstractor tBuO˙ employed here. In terms of energy barriers for C–H cleavage, the hereby reported values align well with those recently reported by some of us for the successful functionalization of the formyl C–H bond in α-tert-butoxy acetaldehyde (ΔG^‡ for C–H cleavage ∼+12.1 kcal mol⁻¹).⁵⁴

Overall, the above considerations are consistent with a mechanism (Scheme 2) in which the excited decatungstate anion (in its reactive state wO) abstracts the C-5 hydrogen atom from 1-βR under kinetic control. The resulting radical 1-βR˙, whose formation is validated by the TEMPO-trapping experiments, is stabilized in the axial position by anomeric-like interactions with the ring oxygen and stereoselectively reacts with the Michael acceptor, yielding intermediate 3-βR˙, which ultimately restores the decatungstate anion [W₁₀O₃₂]⁴⁻ in its original state, forming Giese-type product 3-βR. This mechanistic scenario is further corroborated by deuteration experiments. The absence of deuteration in CD₃CN rules out hydrogen atom abstraction from the solvent, while the partial deuteration patterns observed in the presence of D₂O are compatible with H/D scrambling at the reduced photocatalyst stage (H⁺[W₁₀O₃₂]⁵⁻ ⇄ D⁺[W₁₀O₃₂]⁵⁻) followed by back-H (or D) donation to 1-βR˙ or 3-βR˙ (in the presence of 2a).⁶⁴

Scheme 2 Mechanistic proposal.

Conclusions

In this paper, we showed the development of an efficient green methodology for the site-selective and stereoselective alkylation of position 5 in β-fucosides by means of decatungstate photocatalyzed Giese-type alkylation. This photocatalyst may be easily prepared in a one-step procedure starting from inexpensive starting materials.⁶⁵

Mechanistic studies supported that kinetic effects are likely to determine the C-5 regioselectivity for the TBADT-mediated hydrogen atom abstraction, which enables the overall transformation. The exclusive axial selectivity of the alkylation is attributed to stabilizing interactions of the C-5 radical with the ring oxygen, which parallel those well-established for anomeric (C-1) radicals. The reaction is efficient, versatile and tolerant of various Michael acceptors and anomeric substituents on the fucose ring, including C-, N- and S-glycosides.

The totally unprecedented reactivity we described here for the easy access to a range of C-5 alkylated fucosides via late stage functionalization of β-fucosides has a number of potential applications in the synthesis of fucose-based glycomimetics. L-Fucose is a widespread component of mammalian glycans, where it plays essential functions in cell regulation and immune homeostasis. Abnormal fucosylation is one of the hallmarks of cancer and can be therapeutically targeted. Thus, expanding the chemical space available to fucose-based ligands and inhibitors is a major contribution to the glycotool-box. The chemical reactivity and 3D conformational properties of this new class of C-5 modified fucose rings will need to be established, as well as the potential extension of the present transformation to different monosaccharides. Studies are in progress in our laboratories along these lines.

Author contributions

S. M., M. F. and A. B. conceived and supervised the research. E. C., M. P., D. R. and M. G. designed and performed the experiments. S. M., M. F., D. R. and A. B. analysed the data and wrote the manuscript. All authors discussed the results and commented on the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: experimental procedures, characterization data of the synthesized compounds, details of additional experiments, LFP data, DFT analysis, and NMR spectra (PDF). See DOI: https://doi.org/10.1039/d5gc05775b.

Acknowledgements

The project is supported by EU funding within the NextGeneration EU-MUR PNRR Extended Partnership Initiative on Emerging Infectious Diseases (project no. PE00000007, INF-ACT) and, in part, within the GREEN TECH project of Regione Lombardia (Bando collabora e Innova) and the MUSA – Multilayered Urban Sustainability Action – project, funded by the European Union – NextGenerationEU under the National Recovery and Resilience Plan (NRRP) Mission 4 Component 2 Investment Line 1.5: Strengthening of research structures and creation of R&D “innovation ecosystems”, setup of “territorial leaders in R&D”. HMRS mass spectrometry analyses were performed at the Mass Spectrometry facility of the Unitech COSPECT at Università degli Studi di Milano. The authors acknowledge the CINECA award under the ISCRA initiative, for the availability of high-performance computing resources and support. Support from the Ministero dell'Università e della Ricerca (MUR) and the University of Pavia through the program “Dipartimenti di Eccellenza 2023–2027” is gratefully acknowledged. M. F. is thankful for the financial support provided by the Project PRIN PNRR ‘LIGHT CAT’ (no. P2022RHMCM) supported by the European Commission – NextGenerationEU programme.

References

1. C. T. Campbell, S.-G. Sampathkumar and K. J. Yarema, Mol. Biosyst., 2007, 3, 187–194.
2. N. Probst, G. Grelier, S. Dahaoui, M. Alami, V. Gandon and S. Messaoudi, ACS Catal., 2018, 8, 7781–7786.
3. J. Ghouilem, C. Tran, N. Grimblat, P. Retailleau, M. Alami, V. Gandon and S. Messaoudi, ACS Catal., 2021, 11, 1818–1826.
4. J. Wu, A. Kopp and L. Ackermann, Angew. Chem., Int. Ed., 2022, 61, e202114993.
5. C. S. Bennet, Selective Glycosylations: Synthetic Methods and Catalysts, Wiley, 2017.
6. R. Das and B. Mukhopadhyay, ChemistryOpen, 2016, 5, 401–433.
7. S. Vidal, Analysis and preparation of Glycans, in Comprensive Glycoscience, ed. J. J. Barchi Jr., Elsevier, 2021.
8. Y. Jiang, Y. Zhang, B. C. Lee and M. J. Koh, Angew. Chem., Int. Ed., 2023, 62, e202305138.
9. D. J. Gorelik, S. P. Desai, S. Jdanova, J. A. Turner and M. S. Taylor, Chem. Sci., 2024, 15, 1204–1236.
10. H. Xie, S. Wang and X.-Z. Shu, J. Am. Chem. Soc., 2024, 146, 32269–32275.
11. S. Dubey, Z. Azeem and P. K. Mandal, Adv. Synth. Catal., 2024, 366, 4017–4041.
12. J. Wang, F. Zhou, Y. Xu and L. Zhang, Chem. – Asian J., 2025, 20, e202401114.
13. Y. Jiang, Y. Wei, Q. Y. Zhou, G. Q. Sun, X. P. Fu, N. Levin, Y. Zhang, W. Q. Liu, N. X. Song, S. Mohammed, B. G. Davis and M. J. Koh, Nature, 2024, 631, 319–327.
14. V. Dimakos and M. S. Taylor, Chem. Rev., 2018, 118, 11457–11517.
15. T. G. Frihed, M. Bols and C. M. Pedersen, Eur. J. Org. Chem., 2016, 2740–2756.
16. A. Shatskiy, E. V. Stepanova and M. D. Kärkäs, Nat. Rev. Chem., 2022, 6, 782–805.
17. C. C. J. Loh, Chem. Catal., 2024, 4, 100891.
18. L. Xia, Q. Huang and L. Dai, Org. Chem. Front., 2024, 11, 4926–4933.
19. C. E. Suh, H. M. Carder and A. E. Wendlandt, ACS Chem. Biol., 2021, 16, 1814–1828.
20. A. Tamburrini, C. Colombo and A. Bernardi, Med. Res., 2020, 40, 495–531.
21. A. Bernardi and S. Sattin, Eur. J. Org. Chem., 2020, 4652–4663.
22. L. Capaldo, D. Ravelli and M. Fagnoni, Chem. Rev., 2022, 122, 1875–1924.
23. J. N. Capilato, C. R. Pitts, R. Rowshanpour, T. Dudding and T. Lectka, J. Org. Chem., 2020, 85, 2855–2864.
24. Y. Li and Y. Kuninobu, Adv. Synth. Catal., 2023, 365, 2577–2587.
25. E. Kurokawa, Y. Li, S. Okuyama, K. Sekine and Y. Kuninobu, J. Org. Chem., 2025, 90, 7801–7808.
26. J. L. Jeffrey, J. A. Terrett and D. W. C. MacMillan, Science, 2015, 349, 1532–1536.
27. I. C. Wan, M. D. Witte and A. J. Minnaard, Chem. Commun., 2017, 53, 4926–4929.
28. A. Matsumoto, M. Yamamoto and K. Maruoka, ACS Catal., 2022, 12, 2045–2051.
29. M. L. M. C. Mouthaan, K. Pouwer, M. L. G. Borst, M. D. Witte and A. J. Minnaard, Synthesis, 2022, 4683–4689.
30. Y.-A. Zhang, X. Gu and A. E. Wendlandt, J. Am. Chem. Soc., 2022, 144, 599–605.
31. Y. Wang, H. M. Carder and A. E. Wendlandt, Nature, 2020, 578, 403–408.
32. C. J. Oswood and D. W. C. MacMillan, J. Am. Chem. Soc., 2022, 144, 93–98.
33. H. M. Carder, Y. Wang and A. E. Wendlandt, J. Am. Chem. Soc., 2022, 144, 11870–11877.
34. H. M. Carder, G. Occhialini, G. Bistoni, C. Riplinger, E. E. Kwan and A. E. Wendlandt, Science, 2024, 385, 456–463.
35. Y. Masuda, H. Tsuda and M. Murakami, Angew. Chem., Int. Ed., 2020, 59, 2755–2759.
36. S. Xu, Y. Ping, M. Xu, G. Wu, Y. Ke, R. Miao, X. Qi and W. Kong, Nat. Chem., 2024, 16, 2054–2065.
37. L. Somsák and K. Czifrák, Radical-mediated Brominations at Ring-positions of Carbohydrates – 35 Years Later, in Advances in Carbohydrate Chemistry and Biochemistry, ed. A. L. T. Pilar Rauter, The Royal Society of Chemistry, 2013, vol. 39, pp. 1–37.
38. S. Mazzotta, J. Gori, G. Cavazzoli, S. Sattin, G. Macetti and A. Bernardi, Eur. J. Org. Chem., 2025, e202500599.
39. J. D. McCarter and S. G. Withers, J. Am. Chem. Soc., 1996, 118, 241–242.
40. X. Cao, X. Du, H. Jiao, Q. An, R. Chen, P. Fang, J. Wang and B. Yu, Acta Pharm. Sin. B, 2022, 12, 3783–3821.
41. R. Bermeo, K. Lal, D. Ruggeri, D. Lanaro, S. Mazzotta, F. Vasile, A. Imberty, L. Belvisi, A. Varrot and A. Bernardi, ACS Chem. Biol., 2022, 17, 2899–2910.
42. S. Mazzotta, G. Antonini, F. Vasile, E. Gillon, J. Lundstrøm, A. Varrot, L. Belvisi and A. Bernardi, Molecules, 2023, 28, 1494.
43. G. Antonini, A. Bernardi, E. Gillon, A. Dal Corso, M. Civera, L. Belvisi, A. Varrot and S. Mazzotta, J. Med. Chem., 2024, 67, 19546–19560.
44. G. Antonini, M. Fares, D. Hauck, P. Mała, E. Gillon, L. Belvisi, A. Bernardi, A. Titz, A. Varrot and S. Mazzotta, J. Med. Chem., 2025, 68, 9681–9693.
45. H. M. Flowers, in Adv. Carbohydr. Chem, 1981, vol. 39, pp. 279–345.
46. V. Dimakos, H. Y. Su, G. E. Garrett and M. S. Taylor, J. Am. Chem. Soc., 2019, 141, 5149–5153.
47. G. Zhang, N. W. See, N. Wimmer, M. J. Godinez, S. A. Cameron, R. H. Furneaux and V. Ferro, Org. Lett., 2024, 26, 5956–5960.
48. (a) D. Ravelli, M. Fagnoni, T. Fukuyama, T. Nishikawa and I. Ryu, ACS Catal., 2018, 8, 701–713; (b) K. Yamada, T. Fukuyama, S. Fujii, D. Ravelli, M. Fagnoni and I. Ryu, Chem. – Eur. J., 2017, 23, 8615–8618.
49. D. Ravelli, S. Protti and M. Fagnoni, Acc. Chem. Res., 2016, 49, 2232–2242.
50. J. Hammoud, A. Joosten and T. Lecourt, Carbohydr. Res., 2019, 486, 107834.
51. M. Boultadakis-Arapinis, V. Gandon, E. Prost, L. Micouin and T. Lecourt, Adv. Synth. Catal., 2014, 356, 2493–2505.
52. V. De Waele, O. Poizat, M. Fagnoni, A. Bagno and D. Ravelli, ACS Catal., 2016, 6, 7174–7182.
53. D. Dondi, M. Fagnoni and A. Albini, Chem. – Eur. J., 2006, 12, 4153–4163.
54. E. Cassera, V. Martini, V. Morlacci, S. Abrami, N. Della Ca’, D. Ravelli, M. Fagnoni and L. Capaldo, JACS Au, 2025, 5, 3491–3499.
55. B. Giese and J. Dupuis, Angew. Chem., Int. Ed. Engl., 1983, 22, 622–623.
56. H. Abe, S. Shuto and A. Matsuda, J. Am. Chem. Soc., 2001, 123, 11870–11882.
57. J. D. C. Codée, R. E. J. N. Litjens, L. J. van den Bos, H. S. Overkleeft and G. A. van der Marel, Chem. Soc. Rev., 2005, 34, 769–782.
58. N. Gritsan and M. Platz, in Organic Azides, ed. S. Bräse and K. Bakert, Wiley, 2009, pp. 311–372.
59. Z. Györgydeák and J. Thiem, in Advances in Carbohydrate Chemistry and Biochemistry, Elsevier, 2006, pp. 103–182.
60. B. Giese, Angew. Chem., Int. Ed. Engl., 1989, 28, 969–980.
61. I. V. Alabugin, L. Kuhn, M. G. Medvedev, N. V. Krivoshchapov, V. A. Vil’, I. A. Yaremenko, P. Mehaffy, M. Yarie, A. O. Terent'ev and M. A. Zolfigol, Chem. Soc. Rev., 2021, 50, 10253–10345.
62. A. S. Menon, D. J. Henry, T. Bally and L. Radom, Org. Biomol. Chem., 2011, 9, 3636–3657.
63. Y.-R. Luo, Handbook of Bond Dissociation Energies in Organic Compounds, Taylor & Francis, 2002.
64. L. Ryu, A. Tani, T. Fukuyama, D. Ravelli, M. Fagnoni and A. Albini, Angew. Chem., Int. Ed., 2011, 50, 1869–1872.
65. S. Protti, D. Ravelli, M. Fagnoni and A. Albini, Chem. Commun., 2009, 47, 7351–7353.

---

Footnote

† Dedicated to the memory of Prof. Angelo Albini (1946–2025) for his outstanding contributions to Green Photochemistry.

---