Abstract
Existence of endocyclic cleavage reaction is now clearly shown from experimental evidence of endocyclic cleavage reaction as well as computational chemistry. Not only stereoelectronic factor, several factors could be main factors for endocyclic cleavage reaction. Endocyclic cleavage reaction is useful for 1,2-cis aminoglycoside formation, which is difficult by conventional glycosylation. By using endocyclic cleavage reaction, several glycosides with 1,2-cis aminoglycoside were prepared.
Glycosides are ubiquitous in nature and important for many biological processes, including cell–cell communication, bacterial adhesion, immunological events, fertilization, and cell proliferation [1]. To precisely analyze these roles, facile access to glycoconjugate-derived oligosaccharides is imperative [2]. Glycosylation reaction is the most important reaction for synthesizing oligosaccharides and glycoconjugates [3]. Usually, glycosides are cleaved in an exocyclic manner in both chemical glycosylation and enzymatic reactions. During exocyclic cleavage, the bond between the anomeric carbon and external heteroatom is cleaved to give cyclic cation 2. In chemical glycosylation reaction, nucleophile 3 approachs from α- or β- side to yield 1,2-trans4 or 1,2-cis5 glycosides, respectively (Scheme 1). On the other hand, the bond between the anomeric carbon and O5 oxygen can be cleaved to give acyclic cation 6. This type of cleavage is called as endocyclic cleavage reaction. To date, endocyclic cleavage reaction has not been extensively investigated both in glycochemistry and glycobiology. In this paper, we present evidence of the endocyclic cleavage reaction and its synthetic utility, mainly relying on our previous results [4], [5], [6], [7], [8].
Until early 1980s, only mechanistic aspects of the endocyclic cleavage reaction have been studied, using kinetic investigations. Endocyclic cleavage reaction of 3,6-anhydromethyl glucosides was reported by Haworth as early as in 1941 [9]. Clayton reported that hydrolysis of thioglycoside 7 was accompanied by anomerization and ring rearrangement to afford compounds 8, 10, and 11 (Scheme 2) [10]. This ring contraction must involve ring opening and recyclization of acyclic linear cation 9 [11]. Later, Lemieux and Hindsgaul analyzed the anomerization rate of carboxylic acid-containing pyranoside [12]. Compared with the anomerization rate in the β- to α- direction of compound 12a with SnCl4 (t1/2 of reaction: ~1400 min), compounds 12b and 12c underwent rapid anomerization (t1/2 of reaction: ~10 min and ~7 min, respectively) (Scheme 3). Compound 12a showed rapid anomerization (t1/2 of reaction: ~14 min) when acetic acid was added. Although these do not constitute unequivocal evidence, the authors speculated that the anomerization proceeded through endocyclic cleavage reaction.
In 1986, based on molecular dynamics calculations, Post and Karplus speculated that oligosaccharide hydrolysis proceeds via endocyclic cleavage reaction [13]. They noticed that in X-ray crystallography analyses, the conformation of β-N-acetylglucosamine moiety of the substrate was restricted in the chair form when bound to lysozyme (Scheme 4). Based on the stereoelectronic theory, they speculated that this chair conformation caused endocyclic cleavage reaction.
Stereoelectronic assistance of an antiperiplanar lone pair of electrons is proposed to facilitate both C=O bond cleavage modes [14], [15]. In acetal hydrolysis, the reactant prefers a conformation where the lone pair is antiperiplanar to the cleaved C–O bond owing to the stereoelectronic effects shown in Scheme 5i. The antiperiplanar geometry allows an optimal overlap between a non-bonding electron pair and the vacant σ* orbital of the adjacent bond. If stereoelectronic effect theory is applied to the pyranoside exocyclic cleavage hydrolysis, the α-anomer 24 is expected to be hydrolyzed more rapidly than the corresponding β-anomer 26 (Scheme 5ii and iii, respectively). This is because of a lone electron pair on the oxygen antiperiplanar to the bond being broken in the α-anomer when the pyranoside conformation is restricted in 4C1 conformation. However, in reality, the α- and β-anomers (24 and 26) are hydrolyzed at nearly equal rates. Based on the stereoelectronic effects, it is necessary for the 4C1 conformation of the β-anomer 26 to change before hydrolysis to maximize the overlap of the lone pair of the oxygen (Scheme 5iii). In fact, the hydrolysis of the β-anomer 26 occurs via the twist-boat conformation 27, and the energy for this conformational change is very small compared with the activation energy for the reaction. Based on the above theory, Post and Karplus proposed an endocyclic cleavage model when the conformation of the substrate is restricted in a 4C1 conformation.
Inspired by Karplus’s hypothesis, several investigations conducted to obtain evidence of the endocyclic cleavage have been published. E.g. Franck designed an alkyl β-tetrahydropyran acetal as a carbohydrate mimic compound and an acyclic cation produced by the endocyclic cleavage via an intramolecular aza-Diels-Alder reaction during methanolysis [16]. Anslyn used deuterium-scrambling test, and the cis-decalin-type conformationally locked β-alkyl acetal underwent the endocyclic cleavage reaction, with a maximum 30% yield at room temperature [17].
We found that when a 2,3-trans carbamate group was introduced into 2-amino-2-deoxy-glycoside, anomerization reaction in the β- to α-direction easily occurred [4]. For instance, during a reductive benzylidene opening reaction in the presence of BF3·OEt2, β-thioglycoside 28 was anomerized to α-thioglycoside 29 (Scheme 6) [5]. But, after 6 h, the expected β-glycoside 30 was not detected in the products, and α-glycoside 29 and diol 31 were obtained instead. The existence of diol 31 clearly evidences the occurrence of endocyclic cleavage reaction. The linear cation generated by the endocyclic cleavage reaction was captured by Et3SiH to give diol E (Scheme 7). When the acyclic cation was recycled without reduction, α-glycoside D was generated. A characteristic feature of this endocyclic cleavage reaction is that it can occur at lower temperature and significant amount compared with the exocyclic cleavage reaction compared to the previous reported above examples, without the carbamate group.
Further evidence of the endocyclic cleavage reaction was obtained. Compound 32 underwent an intramolecular Friedel–Crafts reaction involving the acyclic cation and the aromatic moiety at a carbamate nitrogen of the substituent in the presence of BF3·OEt2 to give 34 and 35, together with an α-glycoside 33 (Scheme 8) [5]. Pyranosides with a 2,3-trans carbonate group 36 also underwent the endocyclic cleavage reaction, and an intermolecular Friedel–Crafts reaction involving the cation and the solvent toluene 37 was clear evidence 38 (Scheme 9).
Quantum-mechanical calculations revealed origin of endocyclic cleavage reaction in this system [6]. Excellent agreement is found between the predicted TS energies and the experimental reactivity ranking (Table 1). The experimental feasibility of anomerization order is as follows: 39>41>32>44>46, and excellent agreement is found between the calculations and experiments. And calculations of the conformation of anomerized pyranosides 32, 36, 39, and 41 revealed that the torsion angles of C1–C2–C3–C4 are deformed. The deformation around the C–2 to C–3 bond exceeded 10° and was attributed to the cyclic group. In contrast, the conformations of compounds 48 and 50 are not strained, and they were not anomerized. Inner strain caused by deformation or pyranoside ring structure in fused systems raised ground state energies compared to those without the cyclic group. The endocyclic cleavage reaction is often discussed in relation with the stereoelectronic effect theory; however, we concluded that this anomerization results from the inner strain caused by the fused rings as a predominant factor in this system (Fig. 1).
The substituent group on the carbamate group nitrogen affects the anomerization reaction. When an alkyl group, such as benzyl, o-nitro benzyl, or p-methoxybenzyl, was employed, both α- and β-compounds were obtained [7]. In contrast, acetyl group and several carbamate group substituents showed complete anomerization. Specifically, the acetyl group underwent complete anomerization in disaccharide 52 as well as in a wide range of substrates (Scheme 10). The origin of substituent effect was also suggested by Density Functional Theory calculations. Conformational search in the gas phase revealed that the anti-orientation of the carbonyl group assists the endocyclic cleavage reaction. The minimum energy of the syn conformer was 36.4 kJ mol−1 higher than that of the anti conformer, and the lowest energy barrier between the anti and syn conformers was 42.0 kJ mol−1. These observations suggest that an N-acetyl carbamate substitution predominantly adopt the anti conformation, with the carbonyl oxygen oriented toward the anomeric site, resulting in a lower dipole moment. The carbonyl would stabilize the generated cation at the anomeric position.
Recently, several examples related to the endocyclic cleavage reactions have been reported. Anomerization of pyranosides with 2,3-trans carbamate and carbonate groups were reported by Crich and Oscarson [18]. Oscarson reported a high α/β ratio of the glycosylation products depending on the amount of AgOTf (Scheme 11) [19]. When α-glycoside 55 was obtained when more than 0.4 equivalents of AgOTf were used, on the other hand, β-glycoside 56 was the major product when a smaller amount of AgOTf was used.
Ye reported that the type of additive influenced the selectivity in a glycosylation reaction employing 2,3-trans carbamate- and carbonate-carrying thioglycosides (Scheme 12). Basic additives, such as pyridine derivatives, resulted mainly in β-glycosides, whereas Lewis acids, such as BF3·OEt2, TMSOTf, and SnCl4, gave α-glycosides [20], [21], [22], [23].
Nifantiev reported that sulfated pyranosides were changed to furanosides (Scheme 13) [24]. This PIF (pyranoside-into-furanoside) reaction is thought to proceed through the endocyclic cleavage reaction, which is supported by computational chemistry. The O2 sulfate group of compound 61 participates in the cleavage of the O5–C1 bond. 2-O-acetylated compounds did not yield furanosides. Usually, synthetic methods for the preparation of selectively-protected furanosides are more complicated than those for pyranosides. The sulfate groups were removed under mild conditions. Such PIF rearrangement might be a powerful tool for preparation of differentially protected furanosides.
Murphy reported a systematic investigation of the anomerization of glucuronic acid derivatives in the presence of a Lewis acid. Pilgrim and Murphy found that glucuronic acid and galacturonic acid derivatives were much more (~10 to 3000 times) easily anomerized than protected glucose and galactose derivatives in the presence of SnCl4 and TiCl4 (Scheme 14) [25]. The intermediate acyclic cation that resulted from the endocyclic cleavage reaction was captured by NaBH3CN, to give alcohol. It is thus clear that this anomerization proceeded through an endocyclic cleavage reaction.
Furanosides have been known to be anomerized more easily than pyranosides, and clear evidence of the endocyclic cleavage reaction involving furanosides was reported [26]. Protected furanoside 68 gave high yields of acyclic thioacetal 70 via acyclic bromide 69 (Scheme 15).
Having provided evidence for the occurrence of the endocyclic cleavage reaction, we will now present its synthetic utility. The 1,2-cis aminoglycoside structure is often found in biologically active oligosaccharides, including antibiotics, GPI anchors, and heparin. Although a stereoselective glycosylation reaction has been developed, nevertheless, complete 1,2-cis aminoglycoside bond formation is difficult to attain using this conventional glycosylation reaction [27]. Progress in the selective O-glycoside synthesis field has been reviewed, including 1,2-cis glycoside synthesis, often regarded as more difficult than 1,2-trans glycoside synthesis [28], [29]. We expect that the endocyclic cleavage reaction will become a powerful tool for the generation of 1,2-cis aminoglycosides.
Mycothiol 76 is found in Gram-positive bacteria, and is important for the growth and survival of Mycobacterium tuberculosis [30], [31]. It is believed that mycothiol (MTH) biosynthesis constitutes a drug target for tuberculosis [32]. Another function of MSH was recently reported; the compound is involved in the biosynthesis of lincomycin A, a sulfur-containing lincosamide antibiotic [33], so several mycothiol synthesis have been reported [34], [35], [36], [37], [38]. A 1,2-trans glycoside 73 was prepared from phthaloyl group-protected glucosamine derivative 71 and resolved inositol 72 in 90% yield (Scheme 16). After removal of the phthaloyl group, carbamate group was introduced by triphosgene, and the hydroxy and carbamate nitrogen were acetylated in the presence of DMAP to give 74. Complete anomerization of 1,2-trans glycoside 74 via the endocyclic cleavage reaction was then achieved by BF3·OEt2. A subsequent introduction of the cysteine moiety and protecting manipulation resulted in mycothiol 76.
Furthermore, the endocyclic cleavage reaction can exhibit unique reaction. In a conventional glycosylation reaction, stereochemistry of the anomeric carbon is determined when the glycosyl bond is created. However, the endocyclic cleavage reaction can change multiple stereochemistries at the anomeric position in the β- to α- direction. In fact, the four stereochemistries at the anomeric carbon at tetrasaccharide 77 were converted to the corresponding 78 in a single operation (Scheme 17) [7].
In this paper, we presented evidence for the endocyclic cleavage reaction and its mechanism and synthetic utility. Further investigation into the endocyclic cleavage reaction will open new opportunities in reaction mechanism study and synthesis of glycoconjugates [39].
Article note
A collection of invited papers based on presentations at the XXVIII International Carbohydrate Symposium (ICS-28), New Orleans, July 17–21 2016.
Acknowledgments
S. M. received support through a Grant-in-Aid for Scientific Research (C) (24590041) from the Japan Society for Promotion of Science, Yamada Science Foundation, Takeda Science Foundation, and A-STEP from Japan Science and Technology Agency. We thank Dr. Hiroko Satoh for computational calculations and Ms. Akemi Takahashi for technical assistance.
References
[1] Y. van Kooyk, G. A. Rabinovich. Nat. Immunol. 9, 593 (2008).Search in Google Scholar
[2] L.-X. Wang, B. G. Davis. Chem. Sci.4, 3381 (2013).Search in Google Scholar
[3] M. T. C. Walvoort, G. A. van der Marel, H. S. Overkleeft, J. D. C. Codée. Chem. Sci.4, 897 (2013).Search in Google Scholar
[4] S. Manabe, K. Ishii, Y. Ito. J. Am. Chem. Soc. 128, 10666 (2006).Search in Google Scholar
[5] S. Manabe, K. Ishii, D. Hashizume, H. Koshino, Y. Ito. Chem. Eur. J.15, 6894 (2009).Search in Google Scholar
[6] H. Satoh, S. Manabe, Y. Ito, H. P. Lüthi, T. Laino, J. Hutter. J. Am. Chem. Soc. 113, 5610 (2011).Search in Google Scholar
[7] S. Manabe, H. Satoh, J. Hutter, H. P. Lüthi, T. Laino, Y. Ito. Chem. Eur. J. 20, 124 (2014).Search in Google Scholar
[8] S. Manabe, Y. Ito. Beilstein J. Org. Chem.12, 328 (2016).Search in Google Scholar
[9] W. N. Haworth, L. N. Owen, F. Smith. J. Chem. Soc. 88 (1941).Search in Google Scholar
[10] C. J. Clayton, N. A. Hughes, S. A. Saeed. J. Chem. Soc. C. 644 (1967).Search in Google Scholar
[11] For review: B. Capcon. Chem. Rev.69, 407 (1969).Search in Google Scholar
[12] R. U. Lemieux, O. Hindsgaul. Carbohydr. Res. 82, 195 (1980).Search in Google Scholar
[13] C. B. Post, M. Karplus. J. Am. Chem. Soc. 108, 1317 (1986).Search in Google Scholar
[14] P. Deslongchams. Stereoelectronic Effects in Organic Chemistry, 1st ed., Pergamon Press, New York (1983).Search in Google Scholar
[15] A. J. Kirby. Acc. Chem. Res. 17, 305 (1984).Search in Google Scholar
[16] R. B. Gupta, R. W. Frank. J. Am. Chem. Soc.109, 6554 (1987).Search in Google Scholar
[17] J. L. Liras, E. V. Anslyn. J. Am. Chem. Soc. 116, 2645 (1994).Search in Google Scholar
[18] D. Crich, A. U. Vinod. J. Org. Chem.70, 1291 (2005).Search in Google Scholar
[19] M. E. Boysen, M. Gemma, Lahmann, S. Oscarson. Chem. Commun. 3044 (2005).Search in Google Scholar
[20] Y. Geng. L.-H. Zhang, X. Ye. Chem. Commun. 597 (2008).Search in Google Scholar
[21] Y. Geng, Q. Qin, X.-S. Ye. J. Org. Chem. 77, 5255 (2012).Search in Google Scholar
[22] Q. Qin, D.-C. Xing, X.-S. Ye. Carbohydr. Res. 43, 104 (2015).Search in Google Scholar
[23] Y. Geng, X-S. Ye. Synlett, 2506 (2010).Search in Google Scholar
[24] V. B. Krylov, D. A. Argunov, D. Z. Vinnitskiy, S. A. Verkhnyatskaya, A. G. Gerbst, N. E. Ustyuzhanina, A. S. Dmitrenok, J. Huebner, O. Holst, H.-C. Siebert, N. Nifantiev. Chem. Eur. J. 20, 16516 (2014).Search in Google Scholar
[25] W. Pilgrim, P. V. Murphy. J. Org. Chem.75, 6747 (2010).Search in Google Scholar
[26] O. St-Jean, M. Prévost, Y. Guindon. J. Org. Chem. 78, 2935 (2013).Search in Google Scholar
[27] K. J. Kerns, P. Wei. ACS Symposium Series, 1102, 235 (2012).Search in Google Scholar
[28] T. J. Boltje, T. Buskus, G.-J. Boons. Nature Chem. 1, 611 (2009).Search in Google Scholar
[29] S. S. Nigudkara, A. V. Demchenko. Chem. Sci. 6, 2687 (2015).Search in Google Scholar
[30] G. L. Newton, R. C. Fahey, G. Cohen, Y. Aharonowitz, Y. J. Bacteriol.175, 2734 (1993).Search in Google Scholar
[31] S. Knapp, B. Amorelli, E. Darout, C. C. Ventocilla, L. M. Goldman, R. A. Huhn, E. C. Minnihan. J. Carbohydr. Chem. 24, 103 (2005).Search in Google Scholar
[32] D. W. Gammon, R. Hunter, D. J. Steenkamp, T. T. Mudzunga. Bioorg. Med. Chem. Lett. 13, 2045 (2013).Search in Google Scholar
[33] Q. Zhao, M. Wang, D. Xu, Q. Zhang, W. Liu. Nature, 518, 115 (2015).Search in Google Scholar
[34] G. M. Nicholas, P. Kovac, C. A. Bewley. J. Am. Chem. Soc. 124, 3492 (2002).Search in Google Scholar
[35] S. Lee, J. P. N. Rosazza. Org. Lett.6, 365 (2004).Search in Google Scholar
[36] C. C. Chung, M. M. L. Zukueta, L. T. Padiyar, S.-C. Hung. Org. Lett. 13, 5496 (2011).Search in Google Scholar
[37] K. Ajayi, V. V. Thakur, R. C. Rapo, S. Knapp. Org. Lett. 12, 2630 (2010).Search in Google Scholar
[38] M. S. McConnell, F. Yu, H. M. Nguyen. Chem. Commun. 49, 4313 (2013).Search in Google Scholar
[39] P. V. Murphy. Carbohydr. Chem. 41, 90 (2016).Search in Google Scholar
©2017 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/