Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Feb 21.
Published in final edited form as: Org Lett. 2020 Feb 6;22(4):1495–1498. doi: 10.1021/acs.orglett.0c00078

Synthesis of Glycosyl Chlorides and Bromides by Chelation Assisted Activation of Picolinic Esters under Mild Neutral Conditions

Peng Wen 1,, Christopher J Simmons 2,, Zhi-xiong Ma 1, Stephanie A Blaszczyk 2, Paul G Balzer 1, Wenjing Ye 1,, Xiyan Duan 1,, Hao-Yuan Wang 1, Dan Yin 2, Christopher M Stevens 1, Weiping Tang 1,2,*
PMCID: PMC7050992  NIHMSID: NIHMS1557274  PMID: 32026682

Abstract

A general method has been developed for the formation of glycosyl chlorides and bromides from picolinic esters under mild and neutral conditions. Benchtop stable picolinic esters are activated by a copper (II) halide species to afford the corresponding products in high yields with a traceless leaving group. Rare β glycosyl chlorides are accessible via this route through neighboring group participation. Additionally, glycosyl chlorides with labile protecting groups previously not easily accessible can be prepared.

Graphical Abstract

graphic file with name nihms-1557274-f0001.jpg


Carbohydrates have garnered significant interest from the scientific community due to their heavy prevalence in biological systems and indispensable functionality as mediators of intra- and intercellular interactions.1 Given the heterogeneous nature of carbohydrates in biological sources, the structurally defined synthetic glycans have proved as invaluable tools for studies in glycobiology.2 In synthetic carbohydrate chemistry, glycosylation is arguably one of the most important and challenging reactions,3 and the need for a more efficient and general glycosylation method has stimulated tremendous efforts to develop glycosyl donors with novel anomeric leaving groups.4 Interestingly, glycosyl halides, which are among the earliest known activated glycosyl donors, have always been one of the most versatile glycosyl donors for complex carbohydrate synthesis since its discovery over a century ago.5 On the basis of Koenigs’ and Knorr’s conditions for activating glycosyl bromides with silver salt, a wide variety of methods have been developed by employing different silver or mercury salts, or even metal-free reagents.6 Meanwhile, the toolbox of glycosyl halides as glycosyl donors has expanded from traditionally used glycosyl bromides and chlorides to include more stable glycosyl fluorides7 and more reactive glycosyl iodides,8 rendering the opportunity of fine-tuning the reactivities of glycosyl donors for particular circumstances. Glycosyl halides also display unique properties in controlling the anomeric stereochemistry of glycosylation products by invoking classical neighboring group participation,9 halide exchange,10 or invertive SN2 mechanisms, making 1,2-trans, 1,2-cis, and 2-deoxy-β-glycosides11 accessible from one set of glycosyl donors.

Glycosyl halides, particularly acetohalogenoses have been prepared from their anomeric acetate or hemiacetal precursors under strong Lewis or Brønsted acidic conditions.12 However, toxic halogenating reagents are frequently employed and these forcing conditions are usually incompatible with acid-labile protecting groups or interglycosydic linkages and only allow the preparation of glycosyl halides with limited level of structural diversity.13 Therefore, methods that can expand the functional group compatibility of halogenation have been developed over the past several years.14 Nevertheless, in view of the increasing popularity of glycosyl halides in carbohydrate synthesis, a set of general, mild and neutral conditions for their preparation from readily available benchtop stable precursors is still highly demanded.

Glycosyl picolinic esters (Pico) have been developed as stable and easily accessible glycosyl donors, which can be activated by non-toxic copper(II) salts under mild conditions,15 through remote activation.16 In response to the potential transesterification side reactions, we recently reported isoquinoline-1-carboxylate (ISQ) as a readily available, traceless, and bulkier anomeric leaving group, which successfully blocks the attack of hydroxylic acceptors on the carbonyl carbon center (Scheme 1).17 While screening potential copper (II) sources for the activation of isoquinolinic esters, we discovered that using CuCl2 produced no O-glycosides, but instead led to the formation of the corresponding glycosyl chloride in high yield. The glycosyl halide was formed under conditions that were mild enough to facilitate inclusion of previously unavailable protecting groups that have low stability in acidic solutions. No strong Lewis acid is required for the reaction to proceed and the pH of the solution remains neutral. Herein we report the formation of a range of glycosyl chlorides and bromides in high yields under mild and neutral conditions using copper (II) salts as both the activator and halogen source. The picolinic ester acts as a traceless leaving group and facilitates quick separation of the products from the reaction mixtures with no workup required.

Scheme 1.

Scheme 1.

Copper (II) Activated Leaving Groups for Glycosylations of Sugars

The isoquinolinic ester was advantageous over picolinic ester for glycosylation reaction, because the carbonyl group in the latter could be attacked by the OH group in the glycosyl acceptor to form a transesterification product while no transesterification was observed for the fomer.17 We began our investigation by comparing the isoquinolinic esters and picolinic esters and their ability to produce the corresponding glycosyl chloride in order to examine potential differences in reactivity. We determined that both carboxylate species produced the desired products in essentially the same yield with similar rates of reaction (Scheme 2). While both picolinic acid and isoquinolinic acid are commercially available and can be easily coupled to a pyranose moiety, the lower cost of the picolinic acid made it the preferred candidate for further study. Thus, we carried forward using picolinic esters as substrates for further examination of the reaction potential and scope.

Scheme 2. Comparison of Picolinic and Isoquinolinic Esters for the Formations of Glycosyl Halidesa.

Scheme 2.

a. Reactions run on 0.1 mmol scale with 0.3 mL of solvent and 150 mg of molecular sieves

An array of carbohydrate picolinic esters with armed ether-type protecting groups were synthesized and treated with CuCl2 in dichloromethane in the presence of 4 Å molecular sieves as the only additive (Scheme 3). The reactions proceeded smoothly at room temperature with good yields and high α-selectivity observed for all of the per-benzylated monosaccharides screened (2a-2e). While most of the glycosyl chloride products were isolable by flash column chromatography, the crude reaction products were generally pure for further use without purification, as determined by NMR of the crude products. With the mild and neutral reaction conditions, the benzylidene protected glycosyl chlorides (2f and 2g) and silyl ether protected glycosyl chloride (2h) were prepared in good yields with no degradation of the protecting groups observed. PMB group, which can be more readily cleaved oxidatively than its unsubstituted counterpart, was also well tolerated with 2i isolated in 95% yield. A second picolinic ester at C2 position did not change the stereochemical outcome of the reaction and the α-chloride 2j was isolated in 81% yield. Satisfactory result was also obtained when reaction was run on 1 mmol scale, as exemplified by the isolation of 2a in essentially quantitative yield.

Scheme 3. Scope of Benzylated Glycosyl Chlorides.

Scheme 3.

a Yield of reaction run on 1 mmol scale. b. Yield of crude product. c. 2.2 Equivalents of CuCl2 were used.

The high α-selectivity in all cases, which was independent on the anomeric configurations of the starting picolinic esters, suggests that an oxocarbenium intermediate is involved and the thermodynamic products is likely formed through halide exchange via a SN1 pathway. Although the current methodology is applicable to the synthesis of a wide variety of α-glycosyl chlorides with different configurations and deoxygenation patterns, it is worthwhile to mention that the preparation of the per-benzylated 2-deoxy glucopyranosyl chloride was hampered by the concomitant formation of significant amount of glucal byproduct due to the elimination side reaction.

Carbohydrate picolinic esters with electron-withdrawing protecting groups were then examined for the preparation of disarmed glycosyl chlorides (Scheme 4). In the presence of ester-type protecting groups, the reactions required increased amounts of CuCl2 (2.2 equiv.) and longer reaction times (48–72 h) to ensure the completion of the reaction. Nevertheless, the corresponding glycosyl chlorides were isolated in good to excellent yields. Carbohydrate picolinic esters with gluco-configurations led to the formation of thermodynamically unstable β-glucosyl chlorides18 as exclusive products (2k-2m). This revealed that the neighboring group participation from the ester at the C2-position was the major contributor to the stereoselectivity at anomeric position during the chlorination. The low isolation yield of 2m was a result of the low conversion of the α-glycosyl picolinic ester. With a strong electron-withdrawing carboxylate group at 6-position, the departure of the picolinate from the α-anomer is reduced without the anchimeric assistance of a trans-ester group at C2-position.

Scheme 4. Scope of Benzoylated and Acetylated Glycosyl Chlorides.

Scheme 4.

a. Yield of crude product.

In comparison, per-benzoylated galactopyranosyl, fucopyranosyl and xylopyranosyl picolinic esters afforded glycosyl chlorides as mixtures of anomers (2p-2r), as indicated by NMR of the crude products. Such a deviation in anomeric selectivity can be interpreted as a function of the stabilization (or destabilization) effect of substituents on sugar rings on the putative glycosyl oxocarbenium ion. With all electron-withdrawing substituents on equatorial position, or an axial electron-withdrawing group at C2-position, their destabilization effect on the electron-deficient oxocarbenium ion intermediate was maximized.19 This resulted in the attack of a chloride ion on the relatively more stable dioxalenium ion formed by neighboring group participation, leading to the formation of 1,2-trans glycosyl chloride products 2k-2o. Conversely, an axially-oriented substituent further away from the oxocarbenium carbon center, or a deoxygenation modification reduces the inductive effect,20 therefore, shifting the reaction towards a mechanism involving a more stabilized oxocarbenium ion intermediate.

Not only were we successful in obtaining glycosyl chlorides, but using copper (II) bromide as the activator we could also obtain glycosyl bromides in high yields (Scheme 5). While both armed and disarmed glycosyl bromides were obtained in high yields, unlike the previous examples where a β-glycosyl chlorides could be obtained from neighboring group participation from a C2 ester, using copper (II) bromide with a C2-benzoylated or acetylated monosaccharide only gave the α-glycosyl bromide products (3f-3j). These results are in accordance with bromide ion’s better leaving group ability and the high reactivity of β-glycosyl bromides. Again, the reaction can be scaled up to 1 mmol without any loss of reaction efficiency (3a). Attempts to prepare glycosyl fluorides using copper(II) fluoride salt under the same conditions were not successful. Because copper (II) oxidizes iodide to iodine, we tried to prepare glycosyl iodides using copper(I) iodide salt under the same conditions, but no reaction occurred.

Scheme 5. Scope of Glycosyl Bromidesa.

Scheme 5.

a. Yield of reaction run on 1 mmol scale. b. Yield of crude product. c. 2.2 Equivalents of CuBr2 were used.

To further demonstrate the utility of glycosyl halides obtained from this method, we synthesized two disaccharides from benzoylated glycosyl bromides (Scheme 6). Using silver triflate as the activator and taking advantage of neighboring group participation of a C2 ester, we obtained the disaccharides (4a and 4b) in good yields with high stereoselectivity.

Scheme 6. Glycosylations of Glycosyl Bromidesa.

Scheme 6.

a. Reactions run on 0.05 mmol scale with 0.3 mL of solvent and 180 mg of molecular sieves.

In summary, we have successfully developed a new method to obtain glycosyl chlorides and bromides without the use of strong Brønsted or Lewis acids, and thus can accommodate a wider variety of protecting groups in target compounds. This method utilizes readily available benchtop stable precursors and can access various β-glycosyl chlorides through neighboring group participation using benzoyl and acetyl protected sugars.

Supplementary Material

Supporting Information

ACKNOWLEDGMENT

We thank the University of Wisconsin-Madison and NIH (U01GM125290) and S. A. B. thanks the NIH Chemistry-Biology Interface Training Grant (T32 GM008505) for financial support. We would also like to thank the Analytical Instrumentation Center and the Medicinal Chemistry Center within the UW-Madison School of Pharmacy for their assistance.

Footnotes

The authors declare no competing financial interest.

ASSOCIATED CONTENT

Supporting Information

Detailed experimental procedures, characterization data (IR, 1H, 13C NMR and HRMS), and spectra (1H and 13C NMR). This material is available free of charge via the Internet at http://pubs.acs.org.

REFERENCES

  • (1).(a) Wang L-X; Davis BG, Chem. Sci 2013, 4, 3381–3394; [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Bertozzi CR; Kiessling LL, Science 2001, 291, 2357–2364. [DOI] [PubMed] [Google Scholar]
  • (2).Lepenies B; Yin J; Seeberger PH, Curr. Opin. Chem. Biol 2010, 14, 404–411. [DOI] [PubMed] [Google Scholar]
  • (3).Adero PO; Amarasekara H; Wen P; Bohé L; Crich D, Chem. Rev 2018, 118, 8242–8284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (4).Ranade SC; Demchenko AV, J. Carbohydr. Chem 2013, 32, 1–43. [Google Scholar]
  • (5).(a) Koenigs W; Knorr E, Ber. Dtsch. Chem. Ges 1901, 34, 957–981; [Google Scholar]; (b) Fischer E; Armstrong E, Ber. Dtsch. Chem. Ges 1901, 34, 2885–2900. [Google Scholar]
  • (6).Nitz M; Bundle DR, Glycosyl halides in oligosaccharide synthesis In Glycoscience: Chemistry and Chemical Biology I–III, Springer: 2001; pp 1497–1542. [Google Scholar]
  • (7).(a) Wadzinski TJ; Steinauer A; Hie L; Pelletier G; Schepartz A; Miller SJ, Nat. Chem 2018, 10, 644–652; [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Nicolaou K; Dolle R; Papahatjis D, J. Am. Chem. Soc 1984, 106, 4189–4192; [Google Scholar]; (c) Hashimoto S; Hayashi M; Noyori R, Tetrahedron Lett. 1984, 25, 1379–1382; [Google Scholar]; (d) Mukaiyama T; Hashimoto Y; Shoda S. i., Chem. Lett 1983, 12, 935–938; [Google Scholar]; (e) Mukaiyama T; Murai Y; Shoda S. i., Chem. Lett 1981, 10, 431–432. [Google Scholar]
  • (8).(a) Lowary TL, Carbohydr. Res 2009, 344, 1110–1122; [DOI] [PubMed] [Google Scholar]; (b) Kronzer FJ; Schuerch C, Carbohydr. Res 1974, 34, 71–78; [Google Scholar]; (c) Gervay J; Nguyen TN; Hadd MJ, Carbohydr. Res 1997, 300, 119–125; [Google Scholar]; (d) Bhat AS; Gervay-Hague J, Org. Lett 2001, 3, 2081–2084. [DOI] [PubMed] [Google Scholar]
  • (9).Capon B; McManus SP, Neighboring group participation. Plenum Publishing Corporation: 1976; Vol. 1. [Google Scholar]
  • (10).Lemieux R; Hendriks K; Stick R; James K, J. Am. Chem. Soc 1975, 97, 4056–4062. [Google Scholar]
  • (11).Kaneko M; Herzon SB, Org. Lett 2014, 16, 2776–2779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (12).Haynes L; Newth F, The glycosyl halides and their derivatives In Advances in carbohydrate chemistry, Elsevier: 1955; Vol. 10, pp 207–256. [DOI] [PubMed] [Google Scholar]
  • (13).(a) Egan L; Squires T; Vercellotti J, Carbohydr. Res 1970, 14, 263–266; [Google Scholar]; (b) Polat T; Linhardt RJ, Carbohydr. Res 2003, 338, 447–449; [DOI] [PubMed] [Google Scholar]; (c) Giri SK; Ravindranathan Kartha K, Synth. Commun 2010, 40, 3378–3383; [Google Scholar]; (d) Kováč P; Palovčik R, Carbohydr. Res 1977, 56, 399–403; [Google Scholar]; (e) Cicchillo RM; Norris P, Carbohydr. Res 2000, 328, 431–434. [DOI] [PubMed] [Google Scholar]
  • (14).(a) Ernst B; Winkler T, Tetrahedron Lett. 1989, 30, 3081–3084; [Google Scholar]; (b) Sugiyama S; Diakur JM, Org. Lett 2000, 2, 2713–2715; [DOI] [PubMed] [Google Scholar]; (c) Chang C-W; Chang S-S; Chao C-S; Mong K-KT, Tetrahedron Lett. 2009, 50, 4536–4540; [Google Scholar]; (d) Montero J-L; Winum J-Y; Leydet A; Kamal M; Pavia AA; Roque J-P, Carbohydr. Res 1997, 297, 175–180. [Google Scholar]
  • (15).(a) Koide K; Ohno M; Kobayashi S, Tetrahedron Lett. 1991, 32, 7065–7068; [Google Scholar]; (b) Furukawa H; Koide K; Takao K.-i.; Kobayashi S, Chem. Pharm. Bull 1998, 46, 1244–1247. [Google Scholar]
  • (16).(a) Hanessian S; Bacquet C; Lehong N, Carbohydr. Res 1980, 80, C17–C22; [Google Scholar]; (b) Hanessian S; Lou B, Chem. Rev 2000, 100, 4443–4464. [DOI] [PubMed] [Google Scholar]
  • (17).Wang H-Y; Simmons CJ; Blaszczyk SA; Balzer PG; Luo R; Duan X; Tang W, Angew. Chem. Int. Ed 2017, 129, 15904–15908. [DOI] [PubMed] [Google Scholar]
  • (18).Ibatullin FM; Selivanov SI, Tetrahedron Lett. 2002, 43, 9577–9580. [Google Scholar]
  • (19).Jensen HH; Bols M, Acc. Chem. Res 2006, 39, 259–265. [DOI] [PubMed] [Google Scholar]
  • (20).(a) Namchuk MN; McCarter JD; Becalski A; Andrews T; Withers SG, J. Am. Chem. Soc 2000, 122, 1270–1277; [Google Scholar]; (b) Withers SG; Percival MD; Street IP, Carbohydr. Res 1989, 187, 43–66. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

RESOURCES