English

• With the rapid development of peptide drugs, the modification of bioactive peptides has attracted wide attention of scientists in the field of chemical biology and synthetic chemistry [1-4]. Methods to directly and post-translationally modify peptides or proteins is difficult by conventional two-electron chemical approaches since it also needs to maintain good biocompatibility and site/chemoselectivity simultaneously under the reaction conditions [5]. As an alternative, radical chemistry, particularly by mild photoredox-catalyzed single electron transfer (SET) processes, has been proven to be a powerful tool for subsequent transformation and modification of various peptide side chains or proteins [6-10]. α,β-Dehydroalanine (DHA), as one of the most common unnatural amino acids, exist in various ribosome synthesized and post-translational modified peptides and other naturally occurring peptides [11]. Serine (Ser) or cysteine (Cys) residues used as precursors can effortlessly introduce DHA into complex peptide and protein sites through chemical or enzymatical methods [12]. Due to contain α,β-unsaturated carbonyl motif, DHA shows dramatic chemical reactivity as an acceptor of electrophilic or nucleophilic radicals [13,14]. Furthermore, the addition of a radical to DHA produces α-amino radical intermediates can be stabilized through captodative effect of amino and carbonyl groups [15]. This radical addition reactions are used to introduce various functional groups into peptides and proteins through single electron transfer (SET) process (Scheme 1a) [16-20].

  Scheme 1

  

  Scheme 1.  Modification of DHA and construction of C-glycosidic bond. (a) Late-stage peptide modification at DHA. (b) Tin-mediated C-glycosylation of DHA. (c) Our previous works of C-glycosylation. (d) Photoredox-catalyzed C-glycosylation of DHA in peptides. PC = photocatalyst.

  DownLoad: Full-Size Img PowerPoint

  Glycosylation is one of the most common and important post-translational modifications of proteins in organisms, which controls the localization, function, activity, lifespan, and diversity of proteins in tissues and cells [21,22]. Therefore, diversified glycosidic linkages are the basis for the functional regulation of glycoproteins. The relatively rare C-glycosidic bonds are generally more resistant to acid, base, as well as deglycosylase, than the prevalent N/O-glycosidic bonds. Therefore, C-glycosylation has received intensive attentions [23-26] and this strategy was also widely used in the improvement of glycopeptide drugs [27-30]. Radical glycosylation of peptides mainly focuses on aromatic amino acids (Trp and Phe) [31-33], glycine [34-36] and alkyne/alkene modified amino acids and peptides [37,38], while the glycosylation of DHA motifs is rarely reported. Kessler and Metzler-Nolte groups reported that nBu₃SnH/AIBN-mediated radical addition of glycosyl bromides to DHA derivatives for the synthesis of glycopeptides [39,40]. The Beckwith and co-workers performed C-glycosylation of DHAs under nBu₃SnCl/NaBH₃CN/UV conditions (Scheme 1b) [41]. These pioneering works have some limitations, such as the use of toxic tin reagents, the need for heating or UV to initiate the reactions, and poor chemo- and stereoselectivities.

  Recently, our group constructed the C-glycosidic bond of glycosyl bromides with aromatic heterocycles and alkynyl bromides, respectively, under the photoredox-catalyzed conditions (Scheme 1c) [42,43]. Encouraged by our previous works, as well as aforementioned radical-based C-glycosylation of peptides and proteins, we aim at C-glycosylation of DHA in peptides with glycosyl bromides as glycosyl radical precursors under photoredox catalytic conditions (Scheme 1d) [44,45].

  We initiated our studies by choosing galactose-derived bromide 1a and α,β-dehydroalanine 2 as the model substrates under various photoredox conditions. Some representative results are exhibited in Table 1 (For comprehensive reaction condition optimization, see Supporting information). Based on our previous works about C-glycosylation [42,43] and Gagné group's report on intermolecular addition of glycosyl halides to alkenes [46], the coupling of galactosyl bromide 1a and DHA 2 proceeded successfully to provide C-glycoalanine 3a in an 88% NMR yield (56% isolated yield), with acetonitrile as the solvent, [Ir(ppy)₂(dtbbpy)]PF₆ (PC 1, 2 mol%) as the photocatalyst (PC), Hantzsch ester (HE) as the reductive quencher and Cs₂CO₃ as the acid scavenger under the irradiation of 45 W blue LEDs (Table 1, entry 1). Other photocatalysts, such as [Ir(dF-CF₃-ppy)₂(dtbpy)]PF₆ (PC 2), Ru(bpy)₃(PF₆)₂ (PC 3), Eosin Y (PC 4), fluorescein (PC 5) were evaluated, but lower yields were obtained in all cases (entries 2–5). Solvents, such as 1,4-dioxane, tetrahydrofuran (THF) and dichloromethane (DCM), were also tested and no better result was achieved (entries 6–8). When Cs₂CO₃ was replaced by K₂CO₃, the yield decreased to 43% (entry 9). Delightfully, the isolated yield was improved to 80% when the 2 equiv. of HE was used (entry 10). As expected, no reaction was observed in the absence of blue light or the photocatalyst (entries 11 and 12). It is worth mentioning that C-glycoalanine 3a was produced with only α configuration at anomeric position and 2:1 diastereomeric ratio (dr) at α position of alanine in all cases.

  Table 1

  

  Table 1.  Optimization of reaction conditions.^a

  DownLoad: CSV

  Having established the optimized reaction conditions, we began to test the reactivity of various glycosyl bromides with α,β-dehydroalanine 2. As shown in Scheme 2, D-galactose-, D-glucose-, D-mannose-, D-fucose- and D-arabinose-derived bromides underwent this reaction smoothly to give C-glycopeptide 3a-3i with moderate to good yields (54%−80%) and α-stereoselectivity. In addition, disaccharide-derived bromides (such as lactose and melibiose) could also be used as the coupling couplers and the C-glycosylation products 3h and 3i were generated in satisfactory yields (72% and 65% respectively). The α-configuration was carefully established by ¹H NMR analysis based on coupling constants [47,48].

  Scheme 2

  

  Scheme 2.  Scope of glycosyl bromides. Reaction conditions: A solution of 1 (0.2 mmol), 2 (0.24 mmol), PC 1 (2.0 mol%), HE (0.4 mmol) and Cs₂CO₃ (0.3 mmol) in MeCN (2 mL) was irradiated by 45 W blue LEDs for 18 h, isolated yields. The dr values were determined by ¹H NMR analysis of crude reaction mixtures.

  DownLoad: Full-Size Img PowerPoint

  After succeeding in C-glycosylation of α,β-dehydroalanine 2 with glycosyl bromides, we turned our attention to C-glycosylation of DHA-containing peptides. First, we chose DHA-containing tripeptide 4a as the model substrate. Unfortunately, the coupling of tripeptide 4a with glycosyl bromide 1a under our newly established optimized conditions in Table 1 resulted in a complicated reaction mixture and the desired C-glycopeptide 5a could not be identified. Therefore, we had to re-optimize the reaction conditions (Table 2). When Cs₂CO₃ was replaced by Et₃N, the target C-glycopeptide 5a was obtained in 27% with only α configuration at anomeric position and 3.8:1 dr at α position of alanine (entry 1). Given that Et₃N can serve as the reductive quencher and the acid scavenger, the dosage of Et₃N increased to 3.0 equiv. without the addition of HE and the yield of 5a was improved to 56% (entry 2). Different reductants were tested (entries 3–6) and DIPEA (Red 2) gave the best result. Other solvents, such as 1,4-dioxane, THF and dimethyl sulfoxide (DMSO), were not superior to MeCN (entries 7–9). Subsequently, photocatalysts, such as [Ir(dF-CF₃-ppy)₂(dtbpy)]PF₆ (PC 2), Ru(bpy)₃(PF₆)₂ (PC 3), Eosin Y (PC 4), and 4CzIPN (PC 6), were also explored, but none of them showed better catalytic efficiency than PC 1 (entries 10–13). The control experiments showed that the blue light, photocatalyst and reductant were all indispensable for this reaction (entries 14–16).

  Table 2

  

  Table 2.  Optimization of reaction conditions.^a

  DownLoad: CSV

  After determining the optimized conditions for C-glycosylation of DHA-containing peptides, we then investigated substrate scopes of glycosyl bromides and DHA-containing peptides. As shown in Scheme 3a, a variety of glycosyl bromides were coupled with DHA-containing tripeptide 4a. Bz-protecting D-galactose-derived bromide could also under this C-glycosylation to give the C-glycopeptide 5b in a 54% isolated yield. Other monosaccharides, such as D-glucose, D-mannose, L-arabinose, D-glucuronide and D-fucose, could also be coupled with DHA 4a to provide the corresponding C-glycopeptides 5c–5g in 39%−78% yields. Polysaccharide-derived bromides (maltose, lactose, and even maltotriose) could serve as the coupling partners providing C-glycosylation products 5h-5k in moderate yields (31%−43%). We then explored C-glycosylation of various DHA-containing peptides with D-galactose-derived bromide 1a (Scheme 3b). Dipeptide DHAs gave targeted compounds 5l and 5m with acceptable yields (31% and 39%, respectively). DHA-containing tripeptides (Gly-Ser-Phe) with different protecting strategies could smoothly couple with glycosyl bromide 1a to give C-glycotripeptides 5n-5p in yields of 54%–69%. This strategy was also applicable to DHA-containing peptides with various amino acid residues, such as Phe, Trp, Cys, Glu and Met. The desired C-glycotripeptides 5q-5u could be constructed in 34%−57% yields. Notably, we attempted to use a mixed solvent of MeCN or 1,4-dioxane with phosphate pH 7.0 buffer, the products 5o and 5q were also obtained with yields of 42%, 53% and 50%, respectively. These mild and near-physiological conditions will be expected to provide valuable experience for C-glycosylation of complex biological peptides and proteins.

  Scheme 3

  

  Scheme 3.  Scope of glycosyl bromides and DHA-containing peptides. Reaction conditions: A solution of 4 (0.2 mmol), 1 (0.3 mmol), PC 1 (2 mol%), DIPEA (0.6 mmol) in MeCN (2 mL) was irradiated by 45 W blue LEDs for 18 h, isolated yield. ^a 1.4-Dioxane/phosphate pH 7.0 buffer (3:1, 2 mL) instead of MeCN. ^b MeCN/phosphate pH 7.0 buffer (3:1, 2 mL) instead of MeCN.

  DownLoad: Full-Size Img PowerPoint

  After the exploration of the substrate scopes, we then investigate the reaction mechanism. As shown in Fig. 1a, the addition of 2.0 equiv. of 2,2,6,6-tetramethyl-1-piperidyloxy (TEMPO) into the reaction mixture under standard conditions could inhibit the desired reaction completely. When DHA 4a was replaced by 1,1-diphenylene (5 equiv.), the galactosyl radical-trapping product 6 was obtained in a 63% yield (Fig. 1a). These experiments indicated that the galactosyl radicals may participate in the reaction process. The Stern-Volmer quenching experiment supported that DIPEA was the quencher of the excited photocatalyst (Fig. 1b).

  Figure 1

  

  Figure 1.  Mechanism studies. (a) Radical trap experiments. (b) Stern–Volmer quenching experiment of the excited [Ir(ppy)₂(dtbbpy)]PF₆ (5 × 10^–4 mol/L) with DIPEA. (c) Proposed mechanism.

  DownLoad: Full-Size Img PowerPoint

  On the basis of the control experiments, as well as the literature precedents, a possible mechanism was proposed (Fig. 1c). An excited photocatalyst Ir(Ⅲ)* is formed by the irradiation of photocatalyst Ir(Ⅲ) with visible light. Ir(Ⅲ)* is reductively quenched by DIPEA to give the corresponding DIPEA radical cation together with a low-valent Ir(Ⅱ) complex [42,43]. Glycosyl bromide 1 is then reduced by Ir(Ⅱ) to produce the glycosyl radical Ⅰ along with the regeneration of photocatalyst Ir(Ⅲ). Alternatively, based on Leonori's proposal [49], radical intermediate Ⅱ generates radical Ⅰ from glycosyl bromide 1 through the halogen-atom transfer (XAT) mechanism. Subsequently, the addition of radical Ⅰ to DHA 4 forms a stable α-amino radical intermediate Ⅲ [39-41]. This intermediate Ⅲ can abstract a hydrogen atom to give the C-glycopeptide product 5.

  In summary, we have proposed and implemented a strategy of synthesizing C-glycopeptides from α,β-dehydroalanine or DHA-containing peptides and glycosyl bromides under mild photocatalytic conditions. The main characteristics of this strategy include readily available the starting materials, simple experimental operation, outstanding functional group compatibility and near-physiological conditions (pH ~ 7 and temperature T ≤ 37 ℃ in aqueous media). Peptides with different amino acid residues can serve as good coupling partners, enabling the site-specific C-glycosylation modification. 30 C-glycopeptide products were synthesized with moderate to good yields.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (No. 22171133) and State Key Laboratory of Analytical Chemistry for Life Science (No. 5431ZZXM2308).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.109128.

  
  
• 1. [1]

     L.R. Malins, Curr. Opin. Chem. Biol. 46 (2018) 25–32.

  2. [2]

     O. Boutureira, G.J.L. Bernardes, Chem. Rev. 115 (2015) 2174–2195. doi: 10.1021/cr500399p

  3. [3]

     J.Q. Liu, A. Shatskiy, B.S. Matsuura, M.D. Karkas, Synthesis 51 (2019) 2759-2791. doi: 10.1055/s-0037-1611852

  4. [4]

     J.N. deGruyter, L.R. Malins, P.S. Baran, Biochemistry 56 (2017) 3863–3873. doi: 10.1021/acs.biochem.7b00536

  5. [5]

     B.G. Davis, X.X. Chen, B. Josephson, ACS Cent. Sci. 9 (2023) 614–638. doi: 10.1111/1468-5922.12917

  6. [6]

     S.J. McCarver, J.X. Qiao, J. Carpenter, et al., Angew. Chem. Int. Ed. 56 (2017) 728–732. doi: 10.1002/anie.201608207

  7. [7]

     C. Bottecchia, M. Rubens, S.B. Gunnoo, et al., Angew. Chem. Int. Ed. 56 (2017) 12702–12707. doi: 10.1002/anie.201706700

  8. [8]

     B.A. Vara, X.P. Li, S. Berritt, et al., Chem. Sci. 9 (2018) 336–344. doi: 10.1039/C7SC04292B

  9. [9]

     N. Ichiishi, J.P. Caldwell, M. Lin, et al., Chem. Sci. 9 (2018) 4168–4175. doi: 10.1039/c8sc00368h

  10. [10]

      S. Bloom, C. Liu, D.K. Kolmel, et al., Nat. Chem. 10 (2018) 205–211. doi: 10.1038/nchem.2888

  11. [11]

      P.G. Arnison, M.J. Bibb, G. Bierbaum, et al., Nat. Prod. Rep. 30 (2013) 1568. doi: 10.1039/c3np90040a

  12. [12]

      M.A. Ortega, W.A. van der Donk, Cell. Chem. Biol. 23 (2016) 31–44. doi: 10.1016/j.chembiol.2015.11.012

  13. [13]

      J.M. Chalker, S.B. Gunnoo, O. Boutureira, et al., Chem. Sci. 2 (2011) 1666–1667. doi: 10.1039/c1sc00185j

  14. [14]

      S.B. Gunnoo, A. Madder, ChemBioChem 17 (2016) 529–553. doi: 10.1002/cbic.201500667

  15. [15]

      H.G. Viehe, Z. Janousek, R. Merenyi, L. Stella, Acc. Chem. Res. 18 (1985) 148–154. doi: 10.1021/ar00113a004

  16. [16]

      T.H. Wright, B.J. Bower, J.M. Chalker, et al., Science 354 (2016) aag1465. doi: 10.1126/science.aag1465

  17. [17]

      A. Yang, S. Ha, J. Ahn, et al., Science 354 (2016) 623–626. doi: 10.1126/science.aah4428

  18. [18]

      B. Josephson, C. Fehl, P.G. Isenegger, et al., Nature 585 (2020) 530–537. doi: 10.1038/s41586-020-2733-7

  19. [19]

      A.D. de Bruijn, G. Roelfes, Chem. Eur. J. 24 (2018) 11314–11318. doi: 10.1002/chem.201803144

  20. [20]

      M. Imiolek, G. Karunanithy, W.L. Ng, et al., J. Am. Chem. Soc. 140 (2018) 1568–1571. doi: 10.1021/jacs.7b10230

  21. [21]

      A.J. Parodi, Biochim. BioPhys. Acta 1426 (1999) 287–295. doi: 10.1016/S0304-4165(98)00130-5

  22. [22]

      J.N. Arnold, M.R. Wormald, R.B. Sim, P.M. Rudd, R.A. Dwek, Annu. Rev. Immunol. 25 (2007) 21–50. doi: 10.1146/annurev.immunol.25.022106.141702

  23. [23]

      Y. Yang, B. Yu, Chem. Rev. 117 (2017) 12281–12356. doi: 10.1021/acs.chemrev.7b00234

  24. [24]

      W. Yan, M. Zheng, P. Chuang, et al., Chin. Chem. Lett. 34 (2023) 108021. doi: 10.1016/j.cclet.2022.108021

  25. [25]

      W. Zhu, Q. Sun, H. Chang, et al., Chin. J. Chem. 40 (2022) 571–576. doi: 10.1002/cjoc.202100658

  26. [26]

      K.L. Lu, Y. Ma, S. Liu, S. Guo, Y. Zhang, Chin. J. Chem. 40 (2022) 681–686. doi: 10.1002/cjoc.202100438

  27. [27]

      A. Tamburrini, C. Colombo, A. Bernardi, Med. Res. Rev. 40 (2020) 495–531. doi: 10.1002/med.21625

  28. [28]

      P.H. Seeberger, D.B. Werz, Nature 446 (2007) 1046–1105. doi: 10.1038/nature05819

  29. [29]

      D.C. Koester, A. Holkenbrink, D.B. Werz, Synthesis 19 (2010) 3217–3242.

  30. [30]

      A. Dondoni, A. Marra, Chem. Rev. 100 (2000) 4395–4422. doi: 10.1021/cr9903003

  31. [31]

      R. Mao, S. Xi, S. Shah, et al., J. Am. Chem. Soc. 143 (2021) 12699–12707. doi: 10.1021/jacs.1c05567

  32. [32]

      D. Takeda, M. Yoritate, H. Yasutomi, et al., Org. Lett. 23 (2021) 1940–1944. doi: 10.1021/acs.orglett.1c00402

  33. [33]

      E.M. Miller, M.A. Walczak, Org. Lett. 23 (2010) 4289–4293.

  34. [34]

      K. Mukai, S. Kasuya, Y. Nakagawa, D. Urabe, M. Inoue, Chem. Sci. 6 (2015) 2765–2769. doi: 10.1039/C5SC00457H

  35. [35]

      P. Ji, Y. Zhang, Y. Wei, et al., Org. Lett. 21 (2019) 3086–3092. doi: 10.1021/acs.orglett.9b00724

  36. [36]

      R. Qi, C. Wang, Z. Ma, et al., Angew. Chem. Int. Ed. 61 (2022) e202200822. doi: 10.1002/anie.202200822

  37. [37]

      W. Shang, S.N. Su, R. Shi, et al., Angew. Chem. Int. Ed. 60 (2021) 385–390. doi: 10.1002/anie.202009828

  38. [38]

      Y.H. Liu, Y.N. Xia, T. Gulzar, et al., Nat. Commun. 12 (2021) 4924–4932. doi: 10.1038/s41467-021-25127-z

  39. [39]

      H. Kessler, V. Wittmann, M. Kock, M. Kottenhahn, Angew. Chem. Int. Ed. 31 (1992) 902–904. doi: 10.1002/anie.199209021

  40. [40]

      R. Hamzavi, C. Meyer, N. Metzler-Nolte, Org. Biomol. Chem. 4 (2006) 3648–3651. doi: 10.1039/b607463d

  41. [41]

      G.A. Adamson, A.L.J. Beckwith, C.L.L. Chai, Aust. J. Chem. 57 (2004) 629–633. doi: 10.1071/CH04054

  42. [42]

      L. Xia, W. Fan, X.A. Yuan, S. Yu, ACS Catal. 11 (2021) 9397–9406. doi: 10.1021/acscatal.1c02088

  43. [43]

      L. Xia, M. Jin, Y. Jiao, S. Yu, Org. Lett. 24 (2022) 364–368. doi: 10.1021/acs.orglett.1c04041

  44. [44]

      A. Chen, S. Zhao, Y. Han, et al., Chem. Sci. 14 (2023) 7569–7580. doi: 10.1039/d3sc01995k

  45. [45]

      L. Poletti, A. Massi, G. Di Carmine, et al., Org. Lett. 25 (2023) 4862–4867. doi: 10.1021/acs.orglett.3c01660

  46. [46]

      R.S. Andrews, J.J. Becker, M.R. Gagne, Angew. Chem. Int. Ed. 49 (2010) 7274–7276. doi: 10.1002/anie.201004311

  47. [47]

      R. Persky, A. Albeck, J. Org. Chem. 65 (2000) 5632–5638. doi: 10.1021/jo0003908

  48. [48]

      E.R. van Rijssel, D.V. Filippov, J.D. Codee, et al., J. Org. Chem. 80 (2015) 4553–4565. doi: 10.1021/acs.joc.5b00419

  49. [49]

      T. Constantin, M. Zanini, D. Leonori, et al., Science 367 (2020) 1021–1026. doi: 10.1126/science.aba2419

• 

  Scheme 1  Modification of DHA and construction of C-glycosidic bond. (a) Late-stage peptide modification at DHA. (b) Tin-mediated C-glycosylation of DHA. (c) Our previous works of C-glycosylation. (d) Photoredox-catalyzed C-glycosylation of DHA in peptides. PC = photocatalyst.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Scope of glycosyl bromides. Reaction conditions: A solution of 1 (0.2 mmol), 2 (0.24 mmol), PC 1 (2.0 mol%), HE (0.4 mmol) and Cs₂CO₃ (0.3 mmol) in MeCN (2 mL) was irradiated by 45 W blue LEDs for 18 h, isolated yields. The dr values were determined by ¹H NMR analysis of crude reaction mixtures.

  下载: 全尺寸图片 幻灯片
  

  Scheme 3  Scope of glycosyl bromides and DHA-containing peptides. Reaction conditions: A solution of 4 (0.2 mmol), 1 (0.3 mmol), PC 1 (2 mol%), DIPEA (0.6 mmol) in MeCN (2 mL) was irradiated by 45 W blue LEDs for 18 h, isolated yield. ^a 1.4-Dioxane/phosphate pH 7.0 buffer (3:1, 2 mL) instead of MeCN. ^b MeCN/phosphate pH 7.0 buffer (3:1, 2 mL) instead of MeCN.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Mechanism studies. (a) Radical trap experiments. (b) Stern–Volmer quenching experiment of the excited [Ir(ppy)₂(dtbbpy)]PF₆ (5 × 10^–4 mol/L) with DIPEA. (c) Proposed mechanism.

  下载: 全尺寸图片 幻灯片

  Table 1.  Optimization of reaction conditions.^a

  下载: 导出CSV

  Table 2.  Optimization of reaction conditions.^a

  下载: 导出CSV
•