Synthesis of C-spiro-glycoconjugates from sugar lactones via zinc mediated Barbier reaction

Abstract

Anomeric gem-diallylation, mono-β-crotylation and mono-β-propargylation of sugar 1,5 and 1,4 lactones have been achieved under Barbier reaction conditions using Zn powder and a catalytic amount of TMSCl as an activator. Ring closing olefin metathesis of the synthesized gem-diallyl derivatives furnished C-spiro cyclopentene glycosides. Finally, the cyclopentene rings were converted into carbohydrate based tricyclic morpholine fused triazole glycoconjugates as potential SGLT2 inhibitors.

---

Introduction

C-Spiro-glycosides¹ represent structural subunits of biologically significant and structurally demanding natural products.² Because of their intrinsic stability these compounds can be used to synthesise specific biological targets.³ Despite their widespread distribution and usefulness, there are only a few scattered reports for accessing these privileged scaffolds.⁴ Synthetically these compounds are the downstream products of C,C-disubstituted glycosyl diene derivatives, which are appropriate predecessors for unsaturated spiro bicyclic sugar offshoots owing to their structural make up which allows them to undergo ring closing olefin metathesis.⁵ However, there are no general methods for the synthesis of C,C-diallyl glycosides. Among the reported methods, application of Keck’s one electron C–C-bond forming reaction to generate a diallyl system by quenching the allylic radical generated in situ with allyltri-n-butylstannane,⁶ the use of an allyl Grignard reagent for the synthesis of a gem-diallyl sugar from a gluconolactone in two steps, namely the opening of the sugar to obtain the gem-allyl derivative with an open chain diol and thereafter an oxidative cyclisation,⁷ the radical reaction of sugar dihalides^5b and allyltributyltin under UV irradiation forming a mixture of products, and the Lewis acid mediated nucleophilic substitution of sugar hemiacetals⁸ with allyltrimethylsilane are noteworthy (Fig. 1). However the excess use of reagents, multi-step reaction procedures, low reaction yields, side product formation, stringent conditions and moisture sensitivity of some of the reagents limit their use and urge the development of new reagents that are more efficient and user friendly, leading to synthetically useful procedures with improved yields.

Fig. 1 Synthetic routes to C,C-diallyl glycosides available in the literature.

C–C bond formation using the Barbier reaction has been well known for many decades. It has been successfully applied recently in the preparation of new β-lactam antibiotics,⁹ the propargylation of cyclic imides,¹⁰ the asymmetric allenylation of aliphatic aldehydes catalyzed by a chiral formamide,¹¹ and the synthesis of propargylic and allenic alcohols.¹² Although bis-allylation reactions of nitriles, anhydrides, carbonyl compounds and acid chlorides under Barbier conditions using Zn¹³ and Sm¹⁴ are well documented, their application in carbohydrate chemistry has not been reported so far. In continuation of our research interest on glycosylation,¹⁵ here we report zinc mediated gem-diallylation, mono-crotylation and mono-propargylation reactions of sugar 1,5 and 1,4 lactones and their application in the synthesis of spiro-C-glycosides as potential SGLT2 inhibitors.

Results and discussion

A typical procedure for the synthesis of sugar 1,5 lactone starts with an easily available free sugar such as D-glucose (1.1), D-galactose (2.1), D-mannose (3.1) or L-rhamnose (4.1), which was peracetylated and reacted with thiophenol in the presence of a Lewis acid in a one-pot procedure to give phenyl 2,3,4,6-tetra-O-acetyl-1-thio-β-D-glucopyranoside (1.2, 85%), phenyl 2,3,4,6-tetra-O-acetyl-1-thio-β-D-galactopyranoside (2.2, 82%), phenyl 2,3,4,6-tetra-O-acetyl-1-thio-β-D-mannopyranoside (3.2, 80%), or phenyl 2,3,4-tri-O-acetyl-1-thio-β-L-rhamnopyranoside (4.2, 78%). Deacetylation under Zemplén conditions¹⁶ followed by a subsequent perbenzylation reaction afforded phenyl 2,3,4,6-tetra-O-benzyl-1-thio-β-D-glucopyranoside (1.3, 96%), phenyl 2,3,4,6-tetra-O-benzyl-1-thio-β-D-galactopyranoside (2.3, 98%), phenyl 2,3,4,6-tetra-O-benzyl-1-thio-β-D-mannopyranoside (3.3, 95%) or phenyl 2,3,4-tri-O-benzyl-1-thio-β-L-rhamnopyranoside (5.3, 93%). The thioglycosides 1.3, 2.3, 3.3 and 4.3 thus obtained were hydrolysed using NIS in water/acetone to give the hemiacetals 2,3,4,6-tetra-O-benzyl-D-glucopyranose (1.4, 97%), 2,3,4,6-tetra-O-benzyl-D-galactopyranose (2.4, 94%), 2,3,4,6-tetra-O-benzyl-D-mannopyranose (3.4, 96%) and 2,3,4-tri-O-benzyl-1-thio-β-L-rhamnopyranoside (4.4, 95%).¹⁷

Afterwards these compounds were oxidized to the corresponding lactones 2,3,4,6-tetra-O-benzyl-D-glucono-1,5-lactone (1, 92%), 2,3,4,6-tetra-O-benzyl-D-galactno-1,5-lactone (2, 87%), 2,3,4,6-tetra-O-benzyl-D-mannono-1,5-lactone (3, 96%) and 2,3,4-tri-O-benzyl-1-thio-β-L-rhamno-1,5-lactone (4, 95%) (Scheme 1).¹⁸ A PCC mediated oxidation reaction¹⁹ of tri-O-benzyl-D-glucal afforded 3,4,6-tri-O-benzyl-2-deoxy-D-glucono-1,5-lactone (5) and 4,6-di-O-benzyl-2,3-dideoxy-D-erythro-hex-2-enono-l,5-lactone (6) in 60% and 15% yield, respectively (Scheme 2).

Scheme 1 Synthetic route to sugar 1,5 lactones (1–4).

Scheme 2 Synthesis of sugar 1,5 lactones (5 and 6).

The synthesis of 2,3,5-tri-O-benzyl-D-ribono-1,4-lactone (7) starts with D-ribose, which was reacted with methanol under Fischer glycosylation conditions. A subsequent perbenzylation reaction formed methyl-2,3,5-tri-O-benzyl-α-D-riboside (7.2, 82%), and then hydrolysis by 3 N HCl yielded 2,3,5-tri-O-benzyl-D-ribofuranose (7.3, 75%). Finally, oxidation of 7.3 with DMSO/Ac₂O¹⁸ afforded 7 in 86% yield. Compound 8 was synthesised by the direct oxidation of 2-deoxy-D-ribose with Br₂/K₂CO₃²⁰ in water with 70% yield (Scheme 3).

Scheme 3 Synthetic routes to sugar 1,4 lactones.

The synthesis of 2,3:5,6-di-O-isopropylidene-D-manno-1,4-lactone (9) starts with D-mannose (3.1), which was easily converted into the diisopropylidene derivative 9.2 (75%) in the presence of dry acetone and conc. H₂SO₄.²¹ Anomeric oxidation of 8.2 with DMSO/Ac₂O¹⁷ yielded 9 (85%) (Scheme 3).

2,3,4,6-tetra-O-benzyl-D-glucono-1,5-lactone (1) was chosen as a model substrate for an allylation reaction under Barbier conditions. The optimisation results are summarised in Table 1. Initially, treatment of 1 with 1 equiv. of allyl bromide, 4.0 equiv. of zinc powder and 0.3 equiv. of TMSCl resulted in the formation of the monoallyl sugar derivative 1b (20%) as a mixture of α : β = 1 : 1. With 2.0 equiv. of allyl bromide, the yield of 1b increased (37%) without the formation of 1a. However, increasing the molar ratio of TMSCl to 0.5 equiv. decreased the yield of 1b (30%). When the molar ratio of Zn : allyl bromide : TMSCl increased to 6 : 3 : 0.3, formation of 1b : 1a was observed in the ratio of 1.5 : 1.0. After a careful optimisation of the reaction conditions, it was found that the best result for the formation of the C,C-diallyl 1a was obtained when the relative molar ratio of zinc powder : allyl bromide : TMSCl was increased to 6 : 4 : 0.3 (entry 8). The formation of 1a was confirmed by spectroscopic analysis. The appearance of characteristic bis-allylic peaks in the ¹H NMR spectrum of 1a and also the anomeric quaternary carbon at δ 75.8 ppm in the ¹³C NMR spectrum were in full agreement with the literature data.⁸ Subjecting the reaction to refluxing conditions or sonication at 40 °C did not improve the reaction yield (entries 9 and 10).

Table 1 Optimization of the reaction conditions^a

Entry	Zn powder (equiv.)	Allyl bromide (equiv.)	TMSCl (equiv.)	Solvent	Yield^b (%) (1a : 1b)
a General reaction conditions: 1 equiv. of compound 1 was used in THF at rt for 4 h.b Isolated yield after column chromatography.c Reaction mixture was refluxed at 60 °C.d Reaction mixture was sonicated at 40 °C.
1	4	1	0.3	THF	20 (0 : 1)
2	4	2	0.3	THF	37 (0 : 1)
3	4	2	0.5	THF	30 (0 : 1)
4	6	2	0.3	THF	48 (0 : 1)
5	6	2	0.5	THF	42 (0 : 1)
6	6	3	0.3	THF	60 (1.5 : 1)
7	6	3	0.5	THF	53 (1.5 : 1)
8	6	4	0.3	THF	95 (1 : 0)
9	6	4	0.3	THF^c	82 (1 : 0)
10	6	4	0.3	THF^d	93 (1 : 0)
11	6	4	—	THF	57 (2 : 1)
12	6	5	—	THF	73 (1 : 0)
13	6	6	—	THF	86 (1 : 0)
14	6	4	0.3	DCE	0
15	6	4	0.3	ACN	0

---

Performing the reaction without TMSCl using the same molar ratio of 6 : 4 (Zn : allyl bromide) decreased the yield of the desired product. However, a further increase in the molar ratio of zinc and allyl bromide to 6 : 6 in the absence of TMSCl led to the formation of 1a as the sole product (entry 13). It is noteworthy that the reaction of gluconolactone with an excess of allyl magnesium bromide leads to an opening of the sugar ring without formation of the bis-allyl product.⁷

With these optimised reaction conditions in hand, a series of sugar lactones, including pyrano-lactones (2, 3), deoxy-lactones (4, 5), 2,3-α,β-unsaturated gluconolactone (6) and furano-lactones (7, 8, 9), were subjected to C,C-gem-diallylation. The results are presented in Table 2. It can be seen that in all cases the reaction proceeded smoothly, leading to the formation of the expected gem-allyl sugar derivatives in good to excellent yields. It was observed that the nature of the sugar hardly affected the reaction time and yield. After getting encouraging results for the bis-allylation reaction, we thought about the bis-crotylation reaction of sugar lactones using crotyl bromide. As such, compound 1 was treated with crotyl bromide under the standard reaction conditions. In this case we ended up with the monosubstituted derivative 10 (β : α = 1 : 1) in almost quantitative yield. Increasing the molar proportion of the reagents did not give successful results. The same result, i.e. monopropargylation, was obtained when we attempted a bis-propargylation reaction of 1 with propargyl bromide using the developed Barbier reaction conditions affording 11 (β : α = 9 : 1) in 85% yield. The ¹H NMR peak at 2.05 ppm (t, J = 2.5 Hz, 1H) and the ¹³C NMR anomeric carbon peak at 96.9 ppm indicated the formation of the β isomer predominantly.

Table 2 Gem-diallylation, mono-crotylation and mono-propargylation reactions of various sugar 1,5 and 1,4 lactones

Entry	Substrate	Activated halides	Product^a	Yield^b (%)
a Identified using spectroscopic analysis.b Isolated yield after column chromatography.c β : α = 1 : 1.d β : α = 9 : 1.
1		Allyl bromide		95
2		Allyl bromide		89
3		Allyl bromide		92
4		Allyl bromide		84
5		Allyl bromide		87
6		Allyl bromide		67
7		Allyl bromide		65
8		Allyl bromide		86
9		Allyl bromide		81
10	1	Crotyl bromide		80^c
11	1	Propargyl bromide		85^d

---

As discussed earlier, bis-allyl sugar derivatives are important starting materials for synthesising C-spiro cyclopentene compounds. Thus, we turned our attention to some of the applications of the bisallyl sugar derivatives.

Initially, 1a was subjected to RCM using a Grubbs 2nd generation catalyst to afford the C-spiro cyclopentenyl compound 12 in 85% yield (Scheme 4). The formation of 12 was confirmed by the disappearance of the proton signals at δ 5.90–5.80 ppm and 5.19–4.98 ppm, and the appearance of proton signals at δ 5.66–5.62 ppm in the ¹H NMR spectrum. Further, the disappearance of signals at 118.5 and 118.4 ppm in the ¹³C NMR spectrum of compound 12 was in accordance with our predicted structure. Similarly, compounds 2a and 4a were also subjected to RCM using the above conditions to afford compounds 13 (81%) and 14 (86%).

Scheme 4 Synthesis of the C-spiro cyclopentenyl glycosides.

Fused triazole compounds are of interest due to their various biological properties²² and also their clinical applications.²³ Keeping in mind the pharmaceutical importance of morpholine,²⁴ we became interested in synthesising new scaffolds containing the morpholine fused 1,2,3-triazole based on a carbohydrate core by modifying the spiro cyclopentene part of compound 12. This could pave the way for the preparation of a wide variety of different bioactive compounds. Compound 12 was subjected to a mCPBA mediated epoxidation reaction. Gratifyingly, we observed the formation of a single diastereomer 15 containing a β epoxide ring. This was confirmed by the ¹H NMR spectrum, which showed a coupling constant of J = 3.5 Hz between the two protons of the epoxy ring which is characteristic of β epoxide formation. Energy minimisation data generated from ChemDraw software was in agreement with the observed result. Treatment of epoxide 15 with a NaN₃/NH₄Cl solution at 60 °C afforded a 1 : 1 mixture of azidoalcohols 16. In order to facilitate triazole formation, compound 16 was propargylated using a propargyl bromide/NaOH solution and TBAB, and then was further treated with CuI under ‘click’ conditions to generate the triazole based glycoconjugate 17 in 87% yield as an inseparable mixture (Scheme 5).

Scheme 5 Synthesis of spiro glycoconjugates.

Dehydration of β-C-glycosides 10 and 11 using Et₃SiH and BF₃·OEt₂ led to the formation of β-crotyl glycoside 18 and β-propargyl glycoside 19 in 68% and 72% yields (Scheme 6). The spectroscopic analysis of 18 and 19 was in agreement with the literature data.^25,26

Scheme 6 Synthesis of β-C-crotyl glucoside and β-C-propargyl glucoside.

Conclusions

In conclusion, we have developed a simple procedure for the gem-diallylation of sugar 1,5 and 1,4 lactones, and the mono-crotylation and mono-propargylation of sugar 1,5 lactones using Zn powder and TMSCl as an activator at room temperature. These reactions avoid the necessary maintenance of cryogenic temperatures, have shorter reaction times and result in good yields. The application of this methodology for the synthesis of C-spiro-cyclopentenyl glycosides, C-spiro-morpholine fused triazole based glycoconjugates, β-C-crotyl and β-C-propargyl glycosides has also been highlighted.

Acknowledgements

The authors are thankful to Dr Ram A. Vishwakarma, Director IIIM Jammu and DST (IFCH-18, GAP 1145) for their generous funding. ML, AH and DKS are thankful to CSIR-UGC New Delhi for the junior research fellowship. IIIM publication no. IIIM/1641/2014.

Notes and references

1. (a) P. R. Sridhar, K. Seshadri and G. M. Reddy, Chem. Commun., 2012, 48, 756; (b) L. E. Paquette, M. J. Kinney and U. Dullweber, J. Org. Chem., 1997, 62, 1713.
2. (a) For a review: H. N. C. Wong, Eur. J. Org. Chem., 1999, 1757; (b) M. Sannigrahi, Tetrahedron, 1999, 55, 9007; (c) A. P. Krapcho, Synthesis, 1974, 383; (d) E. J. Corey and A. Guzman-Perez, Angew. Chem., Int. Ed. Engl., 1998, 37, 389.
3. (a) A. C. Araújo, A. P. Rauter, F. Nicotra, C. Airoldi, B. Costa and L. Cipolla, J. Med. Chem., 2011, 54, 1266; (b) S. L. Midland, N. T. Keen and J. J. Sims, J. Org. Chem., 1995, 60, 1118.
4. (a) N. Noguchi and M. Nakada, Org. Lett., 2006, 8, 2039; (b) N. Haddad, I. Rukhman and Z. Abramovich, J. Org. Chem., 1997, 62, 7629.
5. (a) B. Lv, Y. Feng, J. Dong, M. Xu, B. Xu, W. Zhang, Z. Sheng, A. Welihinda, B. Seed and Y. Chen, ChemMedChem, 2010, 5, 827; (b) G. R. Chen, Z. B. Fei, X. T. Huang, Y. Y. Xie, J. L. Xu, J. Gola, M. Steng and J. P. Praly, Eur. J. Org. Chem., 2001, 2939.
6. M. K. Gurjar, S. V. Ravindranadh and S. Karmakar, Chem. Commun., 2001, 241.
7. A. M. Periers, P. Laurin, Y. Benedetti, S. Lachaud, D. Ferroud, A. Iltis, J.-L. Haesslein, M. Klich, G. L'Hermite and B. Musicki, Tetrahedron Lett., 2000, 41, 867.
8. (a) Y. Takashi and O. Yoshiki, Heterocycles, 2002, 57, 229; (b) O. Yoshik and Y. Takashi, Synthesis, 2007, 3021.
9. Y. S. Cho, J. E. Lee, A. N. Pae, K. Choi and H. Y. Koh, Tetrahedron Lett., 1999, 40, 1725.
10. S. H. Kim and E. H. Han, Tetrahedron Lett., 2000, 41, 6479.
11. K. Iseki, Y. Kuroki and Y. Kobayashi, Tetrahedron: Asymmetry, 1998, 9, 2889.
12. N. Kurono, K. Sugita and M. Tokuda, Tetrahedron, 2000, 56, 847.
13. Y. J. Wei, H. Ren and J. X. Wang, Tetrahedron Lett., 2008, 49, 5697 For an acid chloride substrates using zinc, see: Y. Ishino, M. Mihara and M. Kageyama, Tetrahedron Lett., 2002, 43, 6601.
14. L. Zhifang and Z. Yongmin, Tetrahedron, 2002, 58, 5301.
15. (a) M. Tatina, A. K. Kusunuru, S. K. Yousuf and D. Mukherjee, Chem. Commun., 2013, 49, 11409; (b) A. K. Kusunuru, M. Tatina, S. K. Yousuf and D. Mukherjee, Chem. Commun., 2013, 49, 10154; (c) D. Mukherjee, S. K. Sarkar, U. S. Chowdhuryc and S. C. Taneja, Tetrahedron Lett., 2007, 48, 663; (d) D. Mukherjee, S. K. Sarkar, P. Chattopadhyay and U. S. Chowdhury, J. Carbohydr. Chem., 2005, 24, 251; (e) D. Mukherjee, P. K. Roy and U. S. Chowdhury, Tetrahedron, 2001, 57, 7701; (f) L. R. Mallikharjuna, S. K. Yousuf, D. Mukherjee and S. C. Taneja, Org. Biomol. Chem., 2012, 10, 9090; (g) D. K. Sharma, L. R. Mallikharjuna, T. Sidiq, A. Khajuria, A. K. Tripathi, S. K. Yousuf and D. Mukherjee, RSC Adv., 2013, 3, 11450.
16. (a) G. Zemplén, Ber. Dtsch. Chem. Ges. B, 1926, 59B, 1254; (b) G. Zemplén and E. Pascu, Ber. Dtsch. Chem. Ges. B, 1929, 62B, 1613.
17. M. S. Motawia, J. Marcussen and B. L. Møller, J. Carbohydr. Chem., 1995, 14, 1279.
18. J. D. Albright and L. Goldman, J. Am. Chem. Soc., 1965, 87, 4214.
19. R. Patrick and S. Pierre, Carbohydr. Res., 1981, 98, 139.
20. W. Schmidt and J. Paust, Beitr. Pflanzenschutzentwickl. BASF-Ag., 1981, 766(4), 294.
21. O. T. Schmidt, Methods Carbohydr. Chem., 1963, 2, 318.
22. (a) S. Jantova, S. Letasiova, A. Repicky, R. Ovadekova and B. Lakatos, Cell Biochem. Funct., 2006, 24, 519; (b) A. Lauria, A. Guarcello, G. Dattolo and A. M. Almerico, Tetrahedron Lett., 2008, 49, 1847; (c) J. L. Kelley, C. S. Koble, R. G. Davis, E. W. McLean, F. E. Soroko and B. R. Copper, J. Med. Chem., 1995, 38, 4131; (d) A. R. Katritzky, Y. Zhang and S. K Singh, Heterocycles, 2003, 60, 1225.
23. (a) A. W. Thomas, Bioorg. Med. Chem. Lett., 2002, 12, 1881; (b) G. Broggini, G. Molteni, A. Terraneo and G. Zecchi, Tetrahedron, 1999, 55, 14803.
24. (a) M. Hajos, J. C. Fleishaker, J. K. Filipiak-Reisner, M. T. Brown and E. H. F. Wong, CNS Drug Rev., 2004, 10, 23; (b) J. Jakubowska, M. Wasowska Lukawska and M. Czyz, Eur. J. Pharmacol., 2008, 596, 41; (c) J. J. Hale, S. G. Mills, M. MacCoss, S. K. Shah, H. Qi, D. J. Mathre, M. A. Cascieri, S. Sadowski, C. D. Strader, D. E MacIntyre and J. M. Metzger, J. Med. Chem., 1996, 39, 1760.
25. (a) H. Akira, S. Yasuyuki and S. Hideki, Carbohydr. Res., 1987, 171, 223; (b) H. Akira, S. Yasuyuki and S. Hideki, Tetrahedron Lett., 1984, 25, 2383.
26. L. C. Kit, S. C. Gregory, I. Sirajul and B. W. Peter, Tetrahedron Lett., 2005, 46, 61.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section and copies of ¹H and ¹³C NMR of all compounds. See DOI: 10.1039/c3ra46796a

---