==== Front
ACS Cent Sci
ACS Cent Sci
oc
acscii
ACS Central Science
2374-7943
2374-7951
American Chemical Society

10.1021/acscentsci.3c00201
Article
C–H Glycosylation of Native Carboxylic Acids: Discovery of Antidiabetic SGLT-2 Inhibitors
Wang Sanshan †∥
Chen Kaiqi †∥
Guo Fusheng †
Zhu Wenneng †
Liu Chendi †
Dong Haoran †
https://orcid.org/0000-0003-3560-5774
Yu Jin-Quan *‡
https://orcid.org/0000-0002-0380-8035
Lei Xiaoguang *†§
† Beijing National Laboratory for Molecular Sciences, Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education, Department of Chemical Biology, College of Chemistry and Molecular Engineering, Synthetic and Functional Biomolecules Center, and Peking-Tsinghua Center for Life Sciences, Peking University, Beijing 100871, China
‡ Department of Chemistry, The Scripps Research Institute,10550 North Torrey Pines Road, La Jolla, California 92037, United States
§ Institute for Cancer Research, Shenzhen Bay Laboratory, Shenzhen 518107, China
* E-mail: xglei@pku.edu.cn.
* E-mail: yu200@scripps.edu.
09 06 2023
28 06 2023
9 6 11291139
15 02 2023
© 2023 The Authors. Published by American Chemical Society
2023
The Authors
https://creativecommons.org/licenses/by/4.0/ Permits the broadest form of re-use including for commercial purposes, provided that author attribution and integrity are maintained (https://creativecommons.org/licenses/by/4.0/).

C-Glycosides are critical motifs embedded in many bioactive natural products. The inert C-glycosides are privileged structures for developing therapeutic agents owing to their high chemical and metabolic stability. Despite the comprehensive strategies and tactics established in the past few decades, highly efficient C-glycoside syntheses via C–C coupling with excellent regio-, chemo-, and stereoselectivity are still needed. Here, we report the efficient Pd-catalyzed glycosylation of C–H bonds promoted by weak coordination with native carboxylic acids without external directing groups to install various glycals to the structurally diverse aglycon parts. Mechanistic evidence points to the participation of a glycal radical donor in the C–H coupling reaction. The method has been applied to a wide range of substrates (over 60 examples), including many marketed drug molecules. Natural product- or drug-like scaffolds with compelling bioactivities have been constructed using a late-stage diversification strategy. Remarkably, a new potent sodium-glucose cotransporter-2 inhibitor with antidiabetic potential has been discovered, and the pharmacokinetic/pharmacodynamic profiles of drug molecules have been changed using our C–H glycosylation approach. The method developed here provides a powerful tool for efficiently synthesizing C-glycosides to facilitate drug discovery.

A direct C−H glycosylation with native substrates without external auxiliaries is established. Its synthetic utility is demonstrated with over 60 examples including the synthesis of a novel SGLT-2 inhibitor.

National Natural Science Foundation of China 10.13039/501100001809 22193073 Beijing Municipal Commission of Education 10.13039/501100002888 BJJWZYJH01201910001001 Ministry of Science and Technology of the People''s Republic of China 10.13039/501100002855 2022YFC3401500 Ministry of Science and Technology of the People''s Republic of China 10.13039/501100002855 2022YFC2502500 National Natural Science Foundation of China 10.13039/501100001809 92253305 document-id-old-9oc3c00201
document-id-new-14oc3c00201
ccc-price
==== Body
pmcIntroduction

C-Glycosides, which possess a carbohydrate moiety linked to an aglycone unit by a C–C bond as opposed to the more common C–O linkage, are essential structural motifs present in bioactive natural products, drug molecules, and glycosylated proteins of eukaryote cells.1−9C-Glycosides bound to aryl motifs are of particular interest because of their presence in a diverse range of bioactive natural products and drugs (Figure 1a). In contrast to O-linked glycosides, which can undergo enzymatic hydrolysis and other decomposition pathways at the C–O bond, C-glycosides possess increased chemical and metabolic stability, which affords them unique advantages as potential therapeutic agents, particularly as stable mimics of O-glycosides.7,10,11 This concept has been demonstrated by developing a series of effective sodium-glucose cotransporter-2 (SGLT-2) inhibitors inspired by the O-glycoside natural product Phlorizin (Figure 1a), which served as blockbuster drug molecules to treat type 2 diabetes (T2D).7 Furthermore, medicinal chemists have shown that the pharmacological properties of small molecules and peptides can be modulated by including a glycosyl group,12 thereby improving or altering the drug’s metabolic stability,9 membrane permeability,13 biodistribution,14,15 and integration with the drug target.16

Figure 1 C-Aryl glycosides and C–H glycosylation methods. a, Selected examples of C-glycoside natural products and drugs. b, Pioneering examples of C–H glycosylation methods. c, Pd-Catalyzed direct C–H glycosylation of arenes and heteroarenes with native carboxylic acid promoted by weak coordination. SGLT-2, sodium-glucose cotransporter-2; RdRP, RNA-dependent RNA polymerase; DG, Directing Group; Phth, Phthalimide; OPG, Protected OH group; BQ, 1,4-Benzoquinone.

As a result of their presence in bioactive molecules and significant therapeutic potential, the synthesis of C-glycosides has attracted broad attention. Despite the extensive effort in developing new strategies and tactics in the past few decades,17−24 highly efficient and selective (regio- and stereoselective) methods for synthesizing C-glycosides, particularly C-aryl glycosides, via C–C coupling are still rare.4,5 The Friedel–Crafts-type C-glycosylation of arenes remains one of the most commonly used methods. Due to this method’s dependence on the electronics of the aryl substrate (electron-donating groups are strongly preferred), electron-deficient arenes are inherently poorly reactive, and low regio- and stereoselectivity are obtained with substrates possessing multiple sites capable of electrophilic aromatic substitution. Other strategies for C-glycosylation, including transition-metal-catalyzed coupling, generally require extra steps to generate the precursors of both the glycosyl and aromatic partners.17−24 Over the past decade, the development of direct C–H activation/C–C coupling reactions25−39 provides a new avenue for synthesizing C-glycosides via C–C coupling. Several elegant pioneering examples40−43 have demonstrated the C–H glycosylation concept (Figure 1b). Ye and co-workers have reported the first Pd-catalyzed ortho C(sp2)–H glycosylation to install a glycal to an aromatic group with the assistance of an 8-aminoquinoline directing group.40 Chen and co-workers have developed an ortho C(sp2)–H glycosylation method directed by the same auxiliary to install an integral glycosyl part to the aglycon.41 Liu’s group42 and Ackermann’s group43 have also developed C(sp3)–H glycosylation methods with auxiliaries. When the directing groups were installed on the carbohydrate part, 2-deoxy C-glycosides were obtained by Messaoudi and co-workers.44 While these auxiliary approaches are powerful and inspirational, removing the auxiliaries from carbohydrates can be challenging. Therefore, developing C–H glycosylation with a broad range of native substrates is highly desirable for realizing compelling and diverse C-glycosylation, especially for medicinal chemistry in drug discovery.

Here, we report an efficient ligand-enabled Pd-catalyzed C–H glycosylation reaction with native substrates without external directing groups to attach various glycals to the structurally diverse aglycon parts (Figure 1c). This method has been applied to a wide range of substrates, with over 60 examples. Many drug molecules are compatible with our conditions to undergo direct late-stage glycosylation. Subsequent transformations have afforded diverse scaffolds with compelling bioactivities. Initial mechanistic investigations have indicated that the C–H coupling may involve the generation of the glycosyl radical species. The method developed here provides powerful tools for efficiently synthesizing C-glycosides for drug discovery.

Results and Discussion

Discovery of the Auxiliary Free C–H Glycosylation with Native Substrates

Our experimental design was inspired by previous reports of Pd-catalyzed C–H alkylations and vinylations capable of coupling weakly coordinating substrates, such as native carboxylic acids, with abundant alkyl- and vinyl-boron reagents.45−47 These elegant C–H cross-couplings led us to hypothesize that a direct C(sp2)–H glycosylation of free carboxylic acids without exogenous directing groups could be achieved using sugar-boron reagents. o-Tolyacetic acid 1a was selected as a model substrate to investigate the proposed C–H glycosylation. Different boron sugars and conditions developed for C–H cross couplings were examined. Only trace desired product 3a was detected when utilizing triisopropylsilyl (TIPS)-protected glycal boron-pinacol reagent 2a with the assistance of a monoprotected amino acid ligand (MPAA ligand).

Subsequent thorough optimization of the palladium source, ligand, oxidant, base, solvent, and other additives (summarized in Table 1; see Supporting Information for detailed reaction optimizations) resulted in the identification of the optimal conditions, employing 10 mol % Pd(PhCN)2Cl2, 20 mol % Ac-Ala-OH (Ligand L1), silver carbonate, potassium phosphate dibasic, and benzoquinone in ethanol at 80 °C to afford the desired C–H glycosylated product 3a in 83% yield. Ac-Ala-OH (L1) as the ligand was the key for affecting this weak-coordinated C–H glycosylation (entries 2 to 5).31 Different protecting groups for the amino group of the MPAA ligand such as acetyl, Boc, and Fmoc protecting groups were also tested, while acetyl-protected alanine exhibited the best result in different conditions (see Supporting Information Tables S2 and S2.1). Pioneering work had proven that the acetylamino (NHAc) motif of the ligand could work as an internal base during the concerted metalation deprotonation (CMD) step to accelerate C–H bond cleavage with phenyl acetic acid substrates.25,26,48,49 Different palladium sources also afforded the desired product (entries 6 and 7), albeit in slightly lower yields. Encouragingly, when the loading of the Pd(PhCN)2Cl2 was decreased to 5 mol %, the yield could still be maintained at 60% (entry 8). Ag2CO3 was a critical additive, and the yield was dramatically diminished when utilizing other silver salts (entry 9). The selection of solvent (entries 10 and 11) and base (entries 12 and 13) was also crucial for achieving a usable yield in this direct C–H glycosylation reaction. Benzoquinone was also proven to be an essential additive to generate the desired product (entries 14 and 15). It is noteworthy that the reaction does not require an inert atmosphere, and the yield is reduced without added water (entry 16).

Table 1 Reaction Optimization for the C–H Glycosylation Reaction

a Standard condition: 1a (0.1 mmol), 2a (0.2 mmol), Pd(PhCN)2Cl2 (10 mol %), Ac-Ala-OH (20 mol %), Ag2CO3 (2.0 equiv., 0.2 mmol), K2HPO4 (2.0 equiv., 0.2 mmol), BQ (0.1 equiv., 0.01 mmol), H2O (2.0 equiv., 0.2 mmol), EtOH (1 mL), 80 °C, 16 h.

b The yields were determined by 1H NMR using dibromomethane as an internal standard.

Substrate Scope

With the optimal ligand and reaction conditions in hand, the broad scope of various substituted phenylacetic acid substrates in the C(sp2)–H glycosylation was examined (Figure 2). Substrates bearing different electron-donating and electron-withdrawing groups (3a–s) at either ortho- or meta-positions of the phenylacetic acids reacted smoothly to yield the desired products. The substrates possess highly electron-withdrawing trifluoromethyl (3e, 3p) and nitro groups (3j) typically unreactive in Friedel–Crafts-type C-glycosylations, which are compatible with our method. Interestingly, several functional groups which can serve as handles for further functionalizations, such as the protected amino group (3l), the protected alkyne group (3s), and bromo functionalities (3o), were also tolerated in this reaction with moderate to good yields. Regarding the substituents of para-position, not surprisingly, several minor diglycosylated products were observed in these cases, along with the monoglycosylated products (3t-3y). These nonpolar analogues (3t′-3y′) could be quickly confirmed by proton NMR, but the practical separation of these compounds was quite challenging. Then more complex multisubstituted substrates were evaluated (3aa-3ag), and the reaction still worked smoothly in these cases, even for strongly electron-withdrawing or electron-donating systems (3ab, 3ac, and 3ad). The reaction of the heteroarenes such as thiophene (3ah, 3ai) and phenyl-thiophene (3aj) gave the ortho-glycosylated products in good regioselectivity. Other ring systems, such as pyrrole, furan, indole, and pyridine, were also evaluated, but no desired products were observed.

Figure 2 Syntheses ofC-aryl glycosides via Pd-catalyzed C–H glycosylation with native substrates. Isolated yields on a 0.1 mmol scale. (†) Yields determined by 1H NMR using dibromomethane as internal standard.

Then the substitutions at the benzylic position were carefully assessed. First, we found substituents at the benzylic position, such as methyl (3ak), dimethyl (3al), isopropyl (3am), cyclo-propyl (3an), cyclo-butyl (3ao), phenyl (3ap), the protected amino (3aq), and the protected hydroxyl group (3ar), can be tolerated in our glycosylation reaction. And only monoglycosylated products were observed in these cases due to the extra stereohindrance of the side chain. More interestingly, enantioselective C–H glycosylation was observed in many instances (3ap-3ar) with the help of the Ac-Ala-OH ligand, as similar results have been reported in another type of C–H functionalization.50

Then we set out to examine the scope of different glycals. The boron glycals with chemically stable silyl protecting groups can be obtained by a two-step transformation from the reported acetylated glycals.51 When the TIPS protecting groups were changed to less-hindered tert-butyldimethylsilyl (TBS) protecting groups (3ba) or the cyclic type of silicon protecting groups (3bb), the yield remained around 50%. It seems that glycosylation favors a bulky protecting group. Then different glycals such as the glycal derived from rhamnose (3bc), l-xylose (3bd), d-xylose (3bi), arabinose (3be), galactose (3bf), fucose (3bg), and allose (3bh) were examined. The C-glycosylation worked smoothly in most cases. However, we observed that, when the steric configurations of the sugar hydroxyl groups were oriented in the same direction (see 3be, 3bf, and 3bg), the yield was dramatically reduced compared to the similar glycal that owned opposite stereochemistries for the hydroxyl groups (see 3bd, 3a, and 3bc). The glycals derived from furanose were also tested, but only trace products were observed. Unfortunately, it is challenging to generate the boron glycal species from the corresponding di- or trisaccharides. Therefore, it is not feasible to install polysaccharides directly.

Phenylacetic acid is an active auxin for plants and the key intermediate in the pharmaceutical, pesticide, and perfume industries. The structure of phenylacetic acid is privileged in many drug molecules, especially in Ibuprofen-type analgesic and anti-inflammatory drugs. Site-selective C–H functionalization has been proven to be a powerful approach to realize late-stage modifications of complex natural products and drug molecules for diversification and optimization of new drug candidates with different bioactivities or improved drug-like properties.37,38 The lack of a direct C–H glycosylation method with native substrates limited the discovery of new glycoside drug candidates and the maneuver of profiles of drug molecules. Our C–H glycosylation methodology with native substrates provides access to the late-stage glycosylation of complex drug molecules. Several commercial drugs were subjected to our standard reaction conditions. The anti-inflammatory drug Ibuprofen was ortho-glycosylated in 66% yield (3ca). To demonstrate the viability of this late-stage functionalization method, more currently marketed pharmaceuticals, including Flurbiprofen (3cb), Naproxen (3cc), Actarit (3cd), Ketoprofen (3ce), Diclofenac (3cf), Carprofen (3cg), Loxoprofen (3ch), Isoxepac (3ci), and Fenoprofen (3cj), were all examined. The reaction worked smoothly among these complex substrates to deliver the ortho-specific glycosylated drug derivatives. Our extensive substrate scope investigations have demonstrated the power and versatility of our direct C–H glycosylation chemistry in the efficient synthesis and derivatization of functional organic molecules.

Scalable Synthesis and Late-Stage Diversifications to Generate Various Glycosylated Natural Product- or Drug-Like Molecules

After the accomplishment of substrate scope, large-scale attempts and late-stage diversifications were carried out to optimize new bioactive molecules by utilizing the obtained C-aryl glycals (Figure 3). First, a large-scale test was performed. Naproxen (1cc) was chosen as the entry candidate and subjected to our standard reaction conditions with a 1 mmol scale. Encouragingly, we were pleased to observe that the yield of 3cc was still maintained at about 90% to acquire the desired product with a 0.7 g scale (Figure 3a).

Figure 3 Scalable synthesis and late-stage structure diversifications. a, Gram-scale synthesis of the C–H glycosylation reaction under the standard condition. b, Sequential C–H glycosylation. c, Diverse transformations by utilizing active vinyl and carboxyl groups of the obtained C-glycosides. DMDO, 3,3-Dimethyldioxirane; DIBAL-H, Diisobutylaluminum hydride; TBAF, Tetra-n-butylammonium fluoride.

Then a sequential C–H glycosylation can be achieved by utilizing monoglycosylated product 3w as a proper substrate to generate a diglycosylated product 4, which bore different glycals (Figure 3b). Furthermore, the bromo group could work as a handle to install other carbohydrates and functionalities to expand the product’s complexity further.

The synthetic utility of the methodology was also estimated via different approaches (Figure 3c). First, several transformations were carried out in the carboxyl group. Methylation can be achieved efficiently by utilizing TMSCHN2 to yield the ester 5. The delicate manipulation of oxidation states allowed us to achieve remarkably different results via controlling oxidants and temperature. When single equivalent diisobutylaluminum hydride (DIBAL-H) reduced the methyl ester at a relatively low temperature, a corresponding aldehyde was first received, then underwent a 6π electrocyclic cyclization to process pericyclic structure 6 as the skeleton of the natural product mornaphthoate C (see Figure 1a). The methylated substrate 5 could be reduced to a hydroxyl group by 3 equiv of DIBAL-H or LiAlH4. Then the reduction product was treated with freshly prepared dimethyldioxirane (DMDO). An epoxidation would occur at the vinyl part first, then undergo a ring-opening process in situ by the hydroxyl group to yield the spiro complex 7, which was identical to the structure of the antidiabetic drug Tofogliflozin. Stereochemistry and conformations of the spiro scaffolds were unambiguously assigned by NMR analysis.52 When the C-aryl glycals were treated with freshly prepared DMDO, a similar epoxidation-ring opening process yielded another interesting spiro complex 8.

Finally, the generated C-aryl glycals were converted to the classical C-glycosides with highly stereocontrolled configurations via different transformations. Direct hydrogenation was performed on the methylated C-aryl glycal 3a with Pd/C to obtain the 2-deoxy glucoside 9 with excellent yield and stereoselectivity.40 When the subsequent change of C-aryl glycal 3bb was conducted under the standard condition of Brown hydroboration, C-glucoside 10 was generated smoothly as a single β-configuration isomer.51 The precise control of β-configuration has proven to be challenging by the reported C–H glycosylation methods.41 Since β-C-glucosides broadly exist in drug molecules,7 this method should provide meaningful synthetic applications in medicinal chemistry. When the silyl-protecting groups were changed to benzyl groups, further transformation with DMDO and Superhydride was conducted to afford the desired C-glycosides 11 with highly controlled α-configuration.53,54 Collectively, we have demonstrated that our newly developed method could be applied to generate structurally diverse and functionalized C-glycosides rapidly and efficiently.

Synthesis of a Novel SGLT-2 Inhibitor and a Glycosylated Drug Molecule with the Changed PK/PD Profiles

Several SGLT-2 inhibitors have been recently developed as blockbuster drug molecules for treating type 2 diabetes (T2D), affecting about 26 million people in the U.S. and more than 382 million people worldwide. Besides blood sugar control, some SGLT-2 inhibitors, such as Tofogliflozin, have been shown to provide significant cardiovascular benefits in T2D patients.55−57 To further improve the potency and safety profiles of SGLT-2 inhibitors with strong intellectual property protection, new chemical space of the potential inhibitors should be extensively explored in the medicinal chemistry efforts. Notably, compound 7, with a fascinating spiro-type C-glycoside motif, was efficiently generated by our chemistry. A thorough review of all literature and disclosed patents showed that this spiro sugar framework had not been reported. Therefore, this result inspired us to synthesize compound 13, an exciting analogue of SGLT-2 inhibitor Tofogliflozin, in only two steps from 7 (Figure 4a). Molecular docking experiments were conducted first using the recently reported cryogenic electron microscopy (cryo-EM) protein structure of SGLT-2 (PDB: 7VSI).58 The results showed that compound 13 formed a strong interaction with the nearby amino acid residues in the drug-binding pocket of SGLT-2. Stereo views of the docking model indicated several hydrogen-bonding interactions between 13 and nearby residues, such as Y290, S287, and Q457 (Figure 4b). As expected, this new analogue 13 also showed potent inhibition activity against SGLT-2 with half-maximal inhibitory concentration (IC50) = 436 ± 54 nM (Figure 4c). Compound 13 would serve as a promising lead with the new chemical entity for further drug discovery. Further molecular docking experiments indicated that, compared to Tofogliflozin, the spiro motif of compound 13 might bear a clash with the nearby amino acid residue T153 (Figure S1, Supporting Information). This new analogue 13 may help us to explore more molecular interactions and provide new mechanistic insights for small-molecule inhibition of SGLT-2. In addition, our method may help establish a compound library with different chemical skeletons to discover new potent inhibitors.

Figure 4 Synthesis of a new SGLT-2 inhibitor and alteration of PK/PD profiles for a drug molecule via direct C–H glycosylation. a, Synthesis of novel SGLT-2 inhibitor 13. b, Compound 13 docked into the pocket of SGLT-2. Stereo views of the docking model show the interactions between 13 (green) and nearby residues (pink). The hydrogen-bonding interactions are shown as dotted lines. c, Biological evaluation of 13 for SGLT-2 inhibition by NBDG uptake assay. Dapagliflozin was chosen as the positive control. The IC50 of 13 was calculated as 436 ± 54 nM. d, Significant changes have been observed in the biodistribution and elimination half-time(T1/2) of glycosylated Naproxen 14. p-values were determined by one-way ANOVA with Tukeys multiple comparison post hoc test, *** p < 0.001. Values represent the means ± SEM of three independent experiments. TAS-F, Tris(dimethylamino)sulfonium difluorotrimethylsilicate; TBAF, Tetra-n-butylammonium fluoride.

Additionally, previous studies have suggested that the glycosylated drug molecules usually show different pharmacokinetics and pharmacodynamics (PK/PD) profiles compared to the corresponding prototype molecules.12 To test this hypothesis, compound 3cc was subjected to a 2-step modification to generate the 2-deoxy glycosylated Naproxen 14. Then PK/PD animal studies were performed, and the results showed that significant changes in the biodistribution of glycosylated Naproxen 14 were observed. In particular, the concentrations of 14 in the kidney and liver are six times and eight times higher compared to the prototype drug Naproxen, respectively (Figure 4d). The pharmacokinetic data also showed that glycosylated Naproxen has a shorter drug half-life than Naproxen (Figure 4d, Supporting Information Tables S7 & S8). We expect that direct C-glycosylation of drug molecules would be a powerful tool for medicinal chemists to explore new chemical space and alter the PK/PD properties in drug development.

Mechanistic Investigations

To probe whether this C–H glycosylation proceeds via a Pd(II)/Pd(0) catalytic cycle,29,47,59 the palladacycle 15 formed by C–H activation was reacted with the boron glycal coupling partner.59 The desired product was obtained in 52% yield, supporting the intermediacy of palladacycle 15 (Figure 5a).

Figure 5 Mechanistic studies. a, Stoichiometric reaction of the palladacycle with glycal-Bpin. b, Radical homocoupling experiment. c, Reaction inhibition by TEMPO and BHT. TEMPO, 2,2,6,6-Tetramethyl-1-piperidinyloxy; BHT, 2,6-Di-tert-butyl-4-methyl phenol.

When subjecting compound 2a to the standard reaction conditions in the absence of the phenylacetic acid coupling partner, homocoupling product 16 was obtained in 64% yield (Figure 5b), suggesting that the boron-glycal species may serve as possible glycal radical precursors.

Further experiments showed that the radical trapping reagent TEMPO completely inhibits the C–H glycosylation, suggesting the involvement of a radical pathway (Figure 5c). Notably, when the radical scavenger BHT was added to the reaction mixture, the BHT-substituted glycal 18 was isolated and characterized (Figure 5c), which further supported a radical mechanism. The addition of TEMPO also retarded the C–H arylation reaction48 (Figure 5c). Electron paramagnetic resonance (EPR) experiments were further performed to investigate the radical intermediate (Supporting Information Figure S4). As a result, EPR signals were also detected during the reaction, further indicating that the reaction process may involve a radical intermediate.

Conclusion

In summary, we have developed a new method for efficient C-aryl glycoside synthesis via the Pd-catalyzed C–H cross-coupling of native carboxylic acid substrates with sugar-boron reagents without the use of specialized directing auxiliaries. The generality of the method is demonstrated by a broad substrate scope, particularly the late-stage C-glycosylation of many marketed drug molecules. The diversification of the resulting C-aryl glycals further illustrates the synthetic utility of this reaction to generate complex natural product- or drug-like scaffolds that are difficult to prepare by the existing strategies. Enabled by this methodology, we have discovered a new potent small-molecule inhibitor of SGLT-2 that could serve as a drug lead for type 2 diabetes treatment. The direct C–H glycosylation of drug molecules also provides a powerful tool for medicinal chemists to alter the PK/PD profiles with ease. Although more studies are needed to elucidate the reaction mechanism fully, initial studies such as radical inhibition and EPR experiments indicate that a glycal radical is potentially involved in the C–H glycosylation. We envision the direct C–H glycosylation of native substrates will emerge as an efficient strategy for the efficient, flexible, and scalable synthesis of C-glycosyl complex pharmaceuticals and other medicinally relevant compounds.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscentsci.3c00201.General information, experimental procedures, characterization data, and copies of 1H and 13C NMR spectra and references (PDF)

Supplementary Material

oc3c00201_si_001.pdf

Author Contributions

∥ These authors (S. Wang and K. Chen) contributed equally to this work. X. Lei, J. Yu, and S. Wang initiated the project. X. Lei, J. Yu, S. Wang, K. Chen, and F. Guo conceived and designed the experiments, analyzed the data, and prepared the manuscript with input from all the authors. F. Guo conducted the biological assays. C. Liu, W. Zhu, and H. Dong synthesized the part of the substrates. X. Lei and J. Yu managed the whole project.

The authors declare no competing financial interest.

Acknowledgments

We thank Dr. Jiang Zhou (Peking University) and Dr. Xiaoran He (Peking University) for assistance with high-resolution MS analysis; Prof. Qian Wan (Huazhong University Of Science And Technology), Prof. Gong Chen (Nankai University), Dr. Kevin Wu (TSRI), Dr. Benke Hong (Peking University), Dr. Luoyan Liu (TSRI), Dr. Zhen Wang (TSRI), and Dr. Shaoqun Qian (TSRI) for helpful discussion. This work is funded by grants from the National Key Research and Development Program of China (2022YFC3401500 and 2022YFC2502500 to X.L.), the National Natural Science Foundation of China (22193073 and 92253305 to X.L.), and the Beijing Outstanding Young Scientist Program (BJJWZYJH01201910001001 to X.L.)
==== Refs
References

Wu T. C. ; Goekjian P. G. ; Kishi Y. Preferred conformation of C-glycosides. 1. Conformational similarity of glycosides and corresponding C-glycosides. J. Org. Chem. 1987, 52 , 4819–4823. 10.1021/jo00230a038.
Bililign T. ; Griffith B. R. ; Thorson J. S. Structure, activity, synthesis and biosynthesis of aryl-C-glycosides. Nat. Prod. Rep. 2005, 22 , 742–760. 10.1039/b407364a.16311633
Hocek M. ; Stambasky J. ; Kocovsky P. C-nucleosides: synthetic strategies and biological applications. Chem. Rev. 2009, 109 , 6729–6764. 10.1021/cr9002165.19761208
Yang Y. ; Yu B. Recent Advances in the Chemical Synthesis of C-Glycosides. Chem. Rev. 2017, 117 , 12281–12356. 10.1021/acs.chemrev.7b00234.28915018
Kitamura K. ; Ando Y. ; Matsumoto T. ; Suzuki K. Total Synthesis of Aryl C-Glycoside Natural Products: Strategies and Tactics. Chem. Rev. 2018, 118 , 1495–1598. 10.1021/acs.chemrev.7b00380.29281269
Hultin P. G. Bioactive C-glycosides from bacterial secondary metabolism. Curr. Top. Med. Chem. 2005, 5 , 1299–1331. 10.2174/156802605774643015.16305533
Chao E. C. ; Henry R. R. SGLT2 inhibition--a novel strategy for diabetes treatment.. Nat. Rev. Drug Discovery 2010, 9 , 551–559. 10.1038/nrd3180.20508640
Holshue M. L. ; DeBolt C. ; Lindquist S. ; Lofy K. H. ; Wiesman J. ; Bruce H. ; Spitters C. ; Ericson K. ; Wilkerson S. ; Tural A. ; Diaz G. ; Cohn A. ; Fox L. ; Patel A. ; Gerber S. I. ; Kim L. ; Tong S. ; Lu X. ; Lindstrom S. ; Pallansch M. A. ; Weldon W. C. ; Biggs H. M. ; Uyeki T. M. ; Pillai S. K. First Case of 2019 Novel Coronavirus in the United States. N. Engl. J. Med. 2020, 382 , 929–936. 10.1056/NEJMoa2001191.32004427
Bednarska N. G. ; Wren B. W. ; Willcocks S. J. The importance of the glycosylation of antimicrobial peptides: natural and synthetic approaches. Drug Discovery Today 2017, 22 , 919–926. 10.1016/j.drudis.2017.02.001.28212948
Dondoni A. ; Marra A. Methods for anomeric carbon-linked and fused sugar amino Acid synthesis: the gateway to artificial glycopeptides. Chem. Rev. 2000, 100 , 4395–4422. 10.1021/cr9903003.11749352
Leclerc E. ; Pannecoucke X. ; Ethève-Quelquejeu M. ; Sollogoub M. Fluoro-C-glycosides and fluoro-carbasugars, hydrolytically stable and synthetically challenging glycomimetics. Chem. Soc. Rev. 2013, 42 , 4270–4283. 10.1039/C2CS35403A.23212149
Wadzinski T. J. ; Steinauer A. ; Hie L. ; Pelletier G. ; Schepartz A. ; Miller S. J. Rapid phenolic O-glycosylation of small molecules and complex unprotected peptides in aqueous solvent. Nat. Chem. 2018, 10 , 644–652. 10.1038/s41557-018-0041-8.29713033
Varamini P. ; Mansfeld F. M. ; Blanchfield J. T. ; Wyse B. D. ; Smith M. T. ; Toth I. Synthesis and biological evaluation of an orally active glycosylated endomorphin-1. J. Med. Chem. 2012, 55 , 5859–5867. 10.1021/jm300418d.22680612
Polt R. ; Dhanasekaran M. ; Keyari C. M. Glycosylated neuropeptides: a new vista for neuropsychopharmacology?. Med. Res. Rev. 2005, 25 , 557–585. 10.1002/med.20039.16075406
Bonina F. ; Puglia C. ; Rimoli M. G. ; Melisi D. ; Boatto G. ; Nieddu M. ; Calignano A. ; Rana G. L. ; Caprariis P. d. Glycosyl derivatives of dopamine and L-dopa as anti-Parkinson prodrugs: synthesis, pharmacological activity and in vitro stability studies. J. Drug Targeting 2003, 11 , 25–36. 10.1080/1061186031000086090.
Herzner H. ; Reipen T. ; Schultz M. ; Kunz H. Synthesis of glycopeptides containing carbohydrate and Peptide recognition motifs. Chem. Rev. 2000, 100 , 4495–4538. 10.1021/cr990308c.11749356
Zhu F. ; Rourke M. J. ; Yang T. ; Rodriguez J. ; Walczak M. A. Highly Stereospecific Cross-Coupling Reactions of Anomeric Stannanes for the Synthesis of C-Aryl Glycosides. J. Am. Chem. Soc. 2016, 138 , 12049–12052. 10.1021/jacs.6b07891.27612008
Badir S. O. ; Dumoulin A. ; Matsui J. K. ; Molander G. A. Synthesis of Reversed C-Acyl Glycosides through Ni/Photoredox Dual Catalysis. Angew. Chem., Int. Ed. 2018, 57 , 6610–6613. 10.1002/anie.201800701.
Liu J. ; Gong H. Stereoselective Preparation of α- C-Vinyl/Aryl Glycosides via Nickel-Catalyzed Reductive Coupling of Glycosyl Halides with Vinyl and Aryl Halides. Org. Lett. 2018, 20 , 7991–7995. 10.1021/acs.orglett.8b03567.30525666
Lv W. ; Chen Y. ; Wen S. ; Ba D. ; Cheng G. Modular and Stereoselective Synthesis of C-Aryl Glycosides via Catellani Reaction. J. Am. Chem. Soc. 2020, 142 , 14864–14870. 10.1021/jacs.0c07634.32808778
Wei Y. ; Ben-Zvi B. ; Diao T. Diastereoselective Synthesis of Aryl C-Glycosides from Glycosyl Esters via C-O Bond Homolysis. Angew. Chem., Int. Ed. 2021, 60 , 9433–9438. 10.1002/anie.202014991.
An Y. ; Zhang B. S. ; Ding Y. N. ; Zhang Z. ; Gou X. Y. ; Li X. S. ; Wang X. ; Li Y. ; Liang Y. M. Palladium-catalyzed C-H glycosylation and retro Diels-Alder tandem reaction via structurally modified norbornadienes (smNBDs). Chem. Sci. 2021, 12 , 13144–13150. 10.1039/D1SC03569J.34745545
Takeda D. ; Yoritate M. ; Yasutomi H. ; Chiba S. ; Moriyama T. ; Yokoo A. ; Usui K. ; Hirai G. β-Glycosyl Trifluoroborates as Precursors for Direct α-C-Glycosylation: Synthesis of 2-Deoxy-α-C-glycosides. Org. Lett. 2021, 23 , 1940–1944. 10.1021/acs.orglett.1c00402.33625241
Miller E. M. ; Walczak M. A. Light-Mediated Cross-Coupling of Anomeric Trifluoroborates. Org. Lett. 2021, 23 , 4289–4293. 10.1021/acs.orglett.1c01035.34029464
Wang D.-H. ; Engle K. M. ; Shi B.-F. ; Yu J.-Q. Ligand-Enabled Reactivity and Selectivity in a Synthetically Versatile Aryl C–H Olefination. Science 2010, 327 , 315–319. 10.1126/science.1182512.19965380
Engle K. M. ; Wang D.-H. ; Yu J.-Q. Ligand-Accelerated C–H Activation Reactions: Evidence for a Switch of Mechanism. J. Am. Chem. Soc. 2010, 132 , 14137–14151. 10.1021/ja105044s.20853838
Engle K. M. ; Wang D. H. ; Yu J. Q. Constructing multiply substituted arenes using sequential palladium(II)-catalyzed C-H olefination. Angew. Chem., Int. Ed. 2010, 49 , 6169–6173. 10.1002/anie.201002077.
Dai H. X. ; Stepan A. F. ; Plummer M. S. ; Zhang Y. H. ; Yu J. Q. Divergent C-H functionalizations directed by sulfonamide pharmacophores: late-stage diversification as a tool for drug discovery. J. Am. Chem. Soc. 2011, 133 , 7222–7228. 10.1021/ja201708f.21488638
Engle K. M. ; Mei T. S. ; Wasa M. ; Yu J. Q. Weak coordination as a powerful means for developing broadly useful C-H functionalization reactions. Acc. Chem. Res. 2012, 45 , 788–802. 10.1021/ar200185g.22166158
Pichette Drapeau M. ; Gooßen L. J. Carboxylic Acids as Directing Groups for C-H Bond Functionalization. Chem.-Eur. J. 2016, 22 , 18654–18677. 10.1002/chem.201603263.27730686
He J. ; Wasa M. ; Chan K. S. L. ; Shao Q. ; Yu J. Q. Palladium-Catalyzed Transformations of Alkyl C-H Bonds. Chem. Rev. 2017, 117 , 8754–8786. 10.1021/acs.chemrev.6b00622.28697604
Zhu Y. ; Chen X. ; Yuan C. ; Li G. ; Zhang J. ; Zhao Y. Pd-catalysed ligand-enabled carboxylate-directed highly regioselective arylation of aliphatic acids. Nat. Commun. 2017, 8 , 14904 10.1038/ncomms14904.28383026
Rodrigalvarez J. ; Nappi M. ; Azuma H. ; Flodén N. J. ; Burns M. E. ; Gaunt M. J. Catalytic C(sp3)–H bond activation in tertiary alkylamines. Nat. Chem. 2020, 12 , 76–81. 10.1038/s41557-019-0393-8.31863014
Ghiringhelli F. ; Uttry A. ; Ghosh K. K. ; Gemmeren M. Direct β- and γ-C(sp(3))-H Alkynylation of Free Carboxylic Acids*. Angew. Chem., Int. Ed. 2020, 59 , 23127–23131. 10.1002/anie.202010784.
Uttry A. ; van Gemmeren M. Direct C(sp3)–H Activation of Carboxylic Acids. Synthesis 2020, 52 , 479–488. 10.1055/s-0039-1690720.
Uttry A. ; Mal S. ; van Gemmeren M. Late-Stage β-C(sp(3))-H Deuteration of Carboxylic Acids. J. Am. Chem. Soc. 2021, 143 , 10895–10901. 10.1021/jacs.1c06474.34279928
Guillemard L. ; Kaplaneris N. ; Ackermann L. ; Johansson M. J. Late-stage C–H functionalization offers new opportunities in drug discovery. Nat. Rev. Chem. 2021, 5 , 522–545. 10.1038/s41570-021-00300-6.37117588
Li Z. ; Wang Z. ; Chekshin N. ; Qian S. ; Qiao J. X. ; Cheng P. T. ; Yeung K. S. ; Ewing W. R. ; Yu J. Q. A tautomeric ligand enables directed C–H hydroxylation with molecular oxygen. Science 2021, 372 , 1452–1457. 10.1126/science.abg2362.34840353
Dutta S. ; Bhattacharya T. ; Geffers F. J. ; Bürger M. ; Maiti D. ; Werz D. B. Pd-catalysed C-H functionalisation of free carboxylic acids. Chem. Sci. 2022, 13 , 2551–2573. 10.1039/D1SC05392B.35340865
Liu M. ; Niu Y. ; Wu Y. F. ; Ye X. S. Ligand-Controlled Monoselective C-Aryl Glycoside Synthesis via Palladium-Catalyzed C-H Functionalization of N-Quinolyl Benzamides with 1-Iodoglycals. Org. Lett. 2016, 18 , 1836–1839. 10.1021/acs.orglett.6b00566.27026362
Wang Q. ; An S. ; Deng Z. ; Zhu W. ; Huang Z. ; He G. ; Chen G. Palladium-catalysed C–H glycosylation for synthesis of C-aryl glycosides. Nat. Catal. 2019, 2 , 793–800. 10.1038/s41929-019-0324-5.
Liu Y. ; Wang Y. ; Dai W. ; Huang W. ; Li Y. ; Liu H. Palladium-Catalysed C(sp(3))-H Glycosylation for the Synthesis of C-Alkyl Glycoamino Acids. Angew. Chem., Int. Ed. 2020, 59 , 3491–3494. 10.1002/anie.201914184.
Wu J. ; Kaplaneris N. ; Ni S. ; Kaltenhäuser F. ; Ackermann L. Late-stage C(sp(2))-H and C(sp(3))-H glycosylation of C-aryl/alkyl glycopeptides: mechanistic insights and fluorescence labeling. Chem. Sci. 2020, 11 , 6521–6526. 10.1039/D0SC01260B.34094117
Ghouilem J. ; Tran C. ; Grimblat N. ; Retailleau P. ; Alami M. ; Gandon V. ; Messaoudi S. Diastereoselective Pd-Catalyzed Anomeric C(sp3)–H Activation: Synthesis of α-(Hetero)aryl C-Glycosides. ACS Catal. 2021, 11 , 1818–1826. 10.1021/acscatal.0c05052.
Wasa M. ; Engle K. M. ; Lin D. W. ; Yoo E. J. ; Yu J. Q. Pd(II)-catalyzed enantioselective C-H activation of cyclopropanes. J. Am. Chem. Soc. 2011, 133 , 19598–19601. 10.1021/ja207607s.22059375
Thuy-Boun P. S. ; Villa G. ; Dang D. ; Richardson P. ; Su S. ; Yu J. Q. Ligand-accelerated ortho-C-H alkylation of arylcarboxylic acids using alkyl boron reagents. J. Am. Chem. Soc. 2013, 135 , 17508–17513. 10.1021/ja409014v.24124892
Hu L. ; Shen P. X. ; Shao Q. ; Hong K. ; Qiao J. X. ; Yu J. Q. Pd(II) -Catalyzed Enantioselective C(sp(3))-H Activation/Cross-Coupling Reactions of Free Carboxylic Acids. Angew. Chem., Int. Ed. 2019, 58 , 2134–2138. 10.1002/anie.201813055.
Engle K. M. ; Thuy-Boun P. S. ; Dang M. ; Yu J.-Q. Ligand-Accelerated Cross-Coupling of C(sp2)–H Bonds with Arylboron Reagents. J. Am. Chem. Soc. 2011, 133 , 18183–18193. 10.1021/ja203978r.21913636
Li Z. ; Park H. S. ; Qiao J. X. ; Yeung K.-S. ; Yu J.-Q. Ligand-Enabled C–H Hydroxylation with Aqueous H2O2 at Room Temperature. J. Am. Chem. Soc. 2022, 144 , 18109–18116. 10.1021/jacs.2c08332.36137252
Xiao K. J. ; Chu L. ; Yu J. Q. Enantioselective C-H Olefination of α-Hydroxy and α-Amino Phenylacetic Acids by Kinetic Resolution. Angew. Chem., Int. Ed. 2016, 55 , 2856–2860. 10.1002/anie.201510808.
Parkan K. ; Pohl R. ; Kotora M. Cross-coupling reaction of saccharide-based alkenyl boronic acids with aryl halides: the synthesis of bergenin. Chem.-Eur. J. 2014, 20 , 4414–4419. 10.1002/chem.201304304.24590755
Wurst J. M. ; Liu G. ; Tan D. S. Hydrogen-bonding catalysis and inhibition by simple solvents in the stereoselective kinetic epoxide-opening spirocyclization of glycal epoxides to form spiroketals. J. Am. Chem. Soc. 2011, 133 , 7916–7925. 10.1021/ja201249c.21539313
Vaňková K. ; Rahm M. ; Choutka J. ; Pohl R. ; Parkan K. Facile Approach to C-Glucosides by Using a Protecting-Group-Free Hiyama Cross-Coupling Reaction: High-Yielding Dapagliflozin Synthesis. Chem.-Eur. J. 2021, 27 , 10583–10588. 10.1002/chem.202101052.34048112
Koester D. C. ; Kriemen E. ; Werz D. B. Flexible synthesis of 2-deoxy-C-glycosides and (1→2)-, (1→3)-, and (1→4)-linked C-glycosides. Angew. Chem., Int. Ed. 2013, 52 , 2985–2989. 10.1002/anie.201209697.
Zinman B. ; Wanner C. ; Lachin J. M. ; Fitchett D. ; Bluhmki E. ; Hantel S. ; Mattheus M. ; Devins T. ; Johansen O. E. ; Woerle H. J. ; Broedl U. C. ; Inzucchi S. E. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N. Engl. J. Med. 2015, 373 , 2117–2128. 10.1056/NEJMoa1504720.26378978
Neal B. ; Perkovic V. ; Mahaffey K. W. ; de Zeeuw D. ; Fulcher G. ; Erondu N. ; Shaw W. ; Law G. ; Desai M. ; Matthews D. R. Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes. N. Engl. J. Med. 2017, 377 , 644–657. 10.1056/NEJMoa1611925.28605608
McMurray J. J. V. ; Solomon S. D. ; Inzucchi S. E. ; Koober L. ; Kosiborod M. N. ; Martinez F. A. ; Ponikowski P. ; Sabatine M. S. ; Anand I. S. ; Bělohlávek J. ; Böhm M. ; Chiang C. E. ; Chopra V. K. ; de Boer R. A. ; Desai A. S. ; Diez M. ; Drozdz J. ; Dukát A. ; Ge J. ; Howlett J. G. ; Katova T. ; Kitakaze M. ; Ljungman C. E. A. ; Merkely B. ; Nicolau J. C. ; O’Meara E. ; Petrie M. C. ; Vinh P. N. ; Schou M. ; Tereshchenko S. ; Verma S. ; Held C. ; DeMets D. L. ; Docherty K. F. ; Jhund P. S. ; Bengtsson O. ; Sjöstrand M. ; Langkilde A. M. Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction. N. Engl. J. Med. 2019, 381 , 1995–2008. 10.1056/NEJMoa1911303.31535829
Niu Y. ; Liu R. ; Guan C. ; Zhang Y. ; Chen Z. ; Hoerer S. ; Nar H. ; Chen L. Structural basis of inhibition of the human SGLT2-MAP17 glucose transporter. Nature 2022, 601 , 280–284. 10.1038/s41586-021-04212-9.34880493
Salazar C. A. ; Flesch K. N. ; Haines B. E. ; Zhou P. S. ; Musaev D. G. ; Stahl S. S. Tailored quinones support high-turnover Pd catalysts for oxidative C-H arylation with O2. Science 2020, 370 , 1454–1460. 10.1126/science.abd1085.33214286