Enantioselective synthesis of tetrafluorinated glucose and galactose

Article abstract of DOI:10.1021/ol801272e

(Chemical Equation Presented) Polyfluorinated carbohydrates have emerged as interesting probes to investigate polar hydrophobicity effect(s) in protein-carbohydrate interactions. A convenient enantioselective synthesis of tetrafluorinated analogues of two of the most important monosaccharides, D-glucose and D-galactose, is reported, as well as our first results regarding the glycosylation of these sugar analogues.

Full text of DOI:10.1021/ol801272e

ORGANIC
LETTERS
2008
Vol. 10, No. 17
3673-3676
Enantioselective Synthesis of
Tetrafluorinated Glucose and Galactose
Roxana S. Timofte and Bruno Linclau*
UniVersity of Southampton, School of Chemistry, Highfield, Southampton SO17 1BJ
bruno.linclau@soton.ac.uk
Received June 5, 2008
ABSTRACT
Polyfluorinated carbohydrates have emerged as interesting probes to investigate “polar hydrophobicity” effect(s) in protein-carbohydrate
interactions. A convenient enantioselective synthesis of tetrafluorinated analogues of two of the most important monosaccharides, D-glucose
and D-galactose, is reported, as well as our first results regarding the glycosylation of these sugar analogues.
Carbohydrates are central to a multitude of critical biological
functions.1 Given that carbohydrates generally do not make
good drugs, research into carbohydrate mimetics is of great
importance.2
In pioneering work, DiMagno introduced the concept of
“polar hydrophobicity”.3 By extensive replacement of car-
bohydrate CHOH groups by CF2 groups, binding was
expected to be enhanced due to the hydrophobic effect. In
addition, favorable electrostatic interactions involving the
polarized C-F bond with cationic or dipolar sites in the
receptor would be possible, with such interactions being
negligible in aqueous medium. It was shown that the
hexafluorinated pyranose 1 (Figure 1) crosses the erythrocyte
membrane at an approximately 10 times higher rate than
Figure 1. DiMagno’s heavily fluorinated hexapyranose.
glucose itself, and it was proven that this rate increase was
due to enhanced affinity to the transporter protein (and not
because of simple membrane diffusion originating from
increased lipophilicity).
(1) (a) Varki, A. Glycobiology 1993, 3, 97–130. (b) Lasky, L. A. Annu.
ReV. Biochem. 1995, 64, 113–139. (c) Dwek, R. A. Chem. ReV. 1996, 96,
683–720. (d) Lis, H.; Sharon, N. Chem. ReV. 1998, 98, 637–674. (e) Rudd,
P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A. Science 2001,
291, 2370–2376. (f) Ritchie, G. E.; Moffatt, B. E.; Sim, R. B.; Morgan,
B. P.; Dwek, R. A.; Rudd, P. M. Chem. ReV. 2002, 102, 305–319. (g)
Haltiwanger, R. S.; Lowe, J. B. Annu. ReV. Biochem. 2004, 73, 491–537.
(2) (a) Doores, K. J.; Gamblin, D. P.; Davis, B. G. Chem. Eur. J. 2006,
12, 656–665. (b) Davis, B. G. Science 2004, 303, 480–482. (c) Davis, B. G.
Chem. ReV. 2002, 102, 579–601. (d) Gruner, S. A. W.; Locardi, E.; Lohof,
E.; Kessler, H. Chem. ReV. 2002, 102, 491–514. (e) Ritter, T. K.; Wong,
C.-H. Angew Chem., Int. Ed. 2001, 40, 3508–3533. (f) Koeller, K. M.;
Wong, C.-H. Nat. Biotechnol. 2000, 18, 835–841. (g) Armstrong, G. D.
Curr. Opin. Drug DiscoVery DeV. 2000, 3, 191–202. (h) Sears, P.; Wong,
C.-H. Angew. Chem., Int. Ed. 1999, 38, 2300–2324. (i) Marcaurelle, L. A.;
Bertozzi, C. R. Chem. Eur. J. 1999, 5, 1384–1390.
In addition, glycosylation of natural products can signifi-
cantly influence their biological activity.4 Hence, investiga-
tions into the glycosylation of polyfluorinated sugars is also
of high interest.
To enable further studies of the biological implications of
these modifications, we have developed methodology for a
short, convenient, and enantioselective synthesis of tetraflu-
orinated carbohydrates.5 In this Letter, we report the synthesis
(4) (a) Weymouth-Wilson, A. C. Nat. Prod. Rep. 1997, 14, 99–110. (b)
Langenhan, J. M.; Griffith, B. R.; Thorson, J. S. J. Nat. Prod. 2005, 68,
1696–1711. (c) Thorson, J. S.; Hosted, T. J., Jr.; Jiang, J.; Biggins, J. B.;
Ahlert, J. Curr. Org. Chem. 2001, 5, 139–167.
(3) (a) Kim, H. W.; Rossi, P.; Schoemaker, R. K.; DiMagno, S. G. J. Am.
Chem. Soc. 1998, 120, 9082–9083. (b) Biffinger, J. C.; Kim, H. W.;
DiMagno, S. G. ChemBioChem 2004, 5, 622–627.
(5) Boydell, A. J.; Vinader, V.; Linclau, B. Angew. Chem., Int. Ed. 2004,
43, 5677–5679.
10.1021/ol801272e CCC: $40.75
Published on Web 08/01/2008
2008 American Chemical Society

Scheme 1. Retrosynthetic Analysis
of tetrafluorinated analogues 2 and 3 of two important sugars,
D-glucose and D-galactose, as well as our first results
regarding their glycosylation.
equiv of the relative expensive 10 had to be employed. By
using sodium dithionite in basic medium as the initiator,9
an excellent yield of 11 on a large scale was obtained despite
employing a reduced excess of 10 (1.25 equiv).
The retrosynthetic analysis is shown in Scheme 1. The key
carbon-carbon bond disconnection as indicated corresponds
to our recently established tetrafluoroethylene lithium mediated
cyclization protocol,5 featuring 4 and 5 as key intermediates.
These intermediates would be obtained from the corresponding
tetrafluoroalkylidene bromides 6 and 7 by a bromine-lithium
exchange reaction. The absolute configuration of both 6 and 7
would be established by a Sharpless asymmetric dihydroxylation
(SAD) reaction6 of alkene 9, with the synthesis of the anti-
configured 7 requiring an inversion of configuration. Given SN2
reactions in the R-position of a perfluoroalkyl group are known
to be difficult,7 it was chosen to invert the C5-OH (carbohydrate
numbering), leading to enantiomeric diol ent-8 as the target
intermediate. Alkene 9 is not commercially available and could
be obtained from bromoiodotetrafluoroethane 10 (Scheme 2).
Iodide elimination from ꢀ-iodoperfluoroalkanes10 was
investigated, with DBU outperforming other bases (KOH,
DBN) in terms of E/Z selectivity. It was found that a low
temperature (-50 °C) was necessary to achieve a very high
E/Z ratio (determined by 1H NMR). Despite the low
temperature, the elimination reaction was very fast, and even
on a large scale, complete conversion was observed after 30
min. Unfortunately, the geometric isomers were not easily
separated, and most SAD reactions were executed on a
mixture of isomers. The configuration of the E-isomer was
10c,e
1
confirmed by H and 19F NMR.
It is known that perfluoroalkyl-substituted alkenes have a
reduced reactivity toward dihydroxylation.11,12 The recom-
mended modification for SAD of unreactive alkenes is to
increase the OsO4 and ligand loading.6b Indeed, using 0.8
mol % of OsO4 and 2 mol % of chiral ligand, Qing reported
excellent enantioselectivities (ꢀ-SAD, 99.4% ee, R-SAD,
97.5% ee, 0 °C)12a on a similar E-alkene 12 (Table 1).
Scheme 2. Synthesis of the E-Alkene Substrate 9
Table 1. Sharpless Asymmetric Dihydroxylation of 9
Selective radical-mediated coupling of the bifunctional
building block 10 with alkenes has been reported using Fe/
Cp2TiCl2 as the initiation system,8 and reaction of allyl
benzyl ether with 10 under these conditions afforded 11 in
80% yield. However, it was found that the yield of large-
scale reactions was considerably lower, and in addition, 3.75
t
time
yielda eeb
entry
ligand
(°C) (days)
diol
(%)
(%)
1c
2d
3e
4e
(DHQD)2PHAL 4-6
8
8
8
10
8
8
8
77
81
67
86
92
88
96
86
(DHQD)2PYR
(DHQD)2AQN
(DHQ)2PHAL
4-6
0
5-6
(6) Selected reviews. (a) Zaitsev, A. B.; Adolfsson, H. Synthesis 2006,
1725–1756. (b) Noe, M. C.; Letavic, M. A.; Snow, S. L. Org. React. 2005,
66, 109–625. (c) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. ReV. 1994, 94, 2483–2547.
ent-8
a Isolated yield b Determined by HPLC (Chiralcel OD-H), see Supporting
Information. c Pure E-9. d E/Z ) 98:2. e E/Z ) 96:4.
(7) McBee, E. T.; Battershell, R. D.; Braendlin, H. P. J. Am. Chem.
Soc. 1962, 84, 3157–3160.
(8) Hu, C.-M.; Qiu, Y.-L. J. Fluorine Chem. 1991, 55, 109–111.
(9) (a) Huang, W.-Y. J. Fluorine Chem. 1992, 58, 1–8, and references
therein. (b) Wu, F.-H.; Huang, W.-Y. J. Fluorine Chem. 2001, 110, 59–61.
(c) Wang, Q.-F.; Hu, B.; Luo, B.-H.; Hu, C.-M. Tetrahedron Lett. 1998,
39, 2377–2380.
Using similar conditions, the enantioselectivity for the
ꢀ-SAD reaction of 9 was 92% (entry 1), which is signifi-
cantly lower compared to Qing’s result with 12. In addition,
3674
Org. Lett., Vol. 10, No. 17, 2008

eight days at 4-6 °C were required to obtain a good yield
of 8, which is much longer than the two days at 0 °C required
for 12. However, with (DHQD)2AQN, a 96% ee was
obtained (entry 3). Unfortunately, the enantioselectivity for
the corresponding R-SAD reaction was only 86% ee (entry
4). We did not isolate any dihydroxylation product arising
from the minor Z-isomer.
hydrolysis,14 concomitant formate hydrolysis could not be
avoided, leading mainly to the anti-diol 15. Hence, 7 was
then obtained by regioselective benzylation of 15 as described
above. We were surprised to observe that in this case the
desired benzyl ether 16 was isolated in only 58% yield, next
to an unexpectedly large amount of the dibenzyl ether 17.
DIC-mediated formylation then gave 7.
Next, the anionic cyclization reaction was investigated
(Scheme 4). Bromine-lithium exchange was initiated with
1 equiv of MeLi,5 leading to the lithiated derivatives 4 and
5. We were pleased to observe that, in addition to the
formation of furanoses reported earlier,5 the cyclization to
give the pyranose derivatives 18 and 22 also proceeded in
excellent yield. Scaling up of the reaction was uneventful
and allowed the isolation of a set of minor byproducts, all
arising from the lithiated intermediate. Expected byproducts
were the ꢀ-fluoride elimination products 19 and 23, assigned
via the characteristic trifluorovinyl signals (19F NMR) as well
as the characteristic CFdCF2 band (IR),15 and the protonated
The subsequent conversion to the cyclization precursors
6 and 7 is shown in Scheme 3. Regioselective benzylation
Scheme 3. Synthesis of the Cyclization Precursors
2
products 20 and 24, easily assigned by the large JF-H
coupling constant for -CF2H (≈53 Hz). To our surprise,
the corresponding methylated compounds 21 and 25 were
also obtained, which are thought to arise from reaction of 4
and 5 with the byproduct of the Br-Li exchange reaction
(CH3Br). Key spectroscopic data for 21 and 25
(-CF2CF2CH3) include a tt (δ ≈ 1.8 ppm, 1H NMR) showing
3
4
a characteristic JH-F and JH-F coupling.
Finally, hydrogenolysis of 18 and 22 led to the deprotected
tetrafluorinated D-glucose and D-galactose as an inseparable
mixture of anomers (Scheme 5).
was achieved by exploiting the increased acidity of the
hydroxyl group adjacent to the fluoroalkyl substituent.13 With
NaH as base, excellent regioselectivity was obtained leading
to 13 in 71% yield, with the dibenzylated product (not
shown) only isolated in 4.5% yield. Subsequent formylation
led to 6 in excellent yield.
The initial approach for the synthesis of the tetrafluorinated
glucose precursor 7 involved analogous benzylation of ent-8
to give ent-13, but all attempts to achieve subsequent formate
formation with inversion of C5-OH configuration failed
completely (not shown). Hence, ent-8 was converted to the
corresponding cyclic sulfate 14, followed by regioselective
opening12a,b with ammonium formate. Unfortunately, des-
pite using mild conditions for the subsequent residual sulfate
At this stage, our attention turned to the glycosylation of
these polyfluorinated sugars. Because a cation is destabilized
when adjacent to a perfluoroalkyl group,16 formation of an
oxocarbenium ion intermediate, typical for most conventional
glycosylation methods, is difficult. For the same reason,
glycosidic bonds are expected to be very stable when 2,2-
difluoro substitution is present. Glycosylation of (protected)
1 by nucleophilic substitution of a corresponding triflate was
reported to be slow, though solvolysis was successful.3b
However, glycosylation by anomeric alkylation17 has proven
to be possible for related (di)fluorinated systems,18,19 and
this method was selected for further study.
(14) Kim, B. M.; Sharpless, K. B. Tetrahedron Lett. 1989, 30, 655–
658.
(15) Examples: (a) Sauveˆtre, R.; Masure, D.; Chuit, C.; Normant, J. F.
Synthesis 1978, 128–130. (b) Burdon, J.; Coe, P. L.; Haslock, I. B.; Powell,
R. L. J. Fluorine Chem. 1999, 99, 127–131.
(10) (a) Elsheimer, S.; Foti, C. J.; Bartberger, M. D. J. Org. Chem. 1996,
61, 6252–6255. (b) Seidel, P.; Ugi, I. Z. Naturforsch. B Anorg. Chem. Org.
Chem. 1982, 37, 499–503. (c) Zhu, W.; Li, Z. J. Chem. Soc., Perkin Trans
1 2000, 1105–1108. (d) Foulard, G.; Brigaud, T.; Portella, C. Tetrahedron
1996, 52, 6187–6200. (e) Zarif, L.; Greiner, J.; Pace, S.; Riess, J. G. J. Med.
Chem. 1990, 33, 1262–1269.
(16) (a) Creary, X. Chem. ReV. 1991, 91, 1625–1678. (b) Jansen, M. P.;
Koshy, K. M.; Mangru, N. N.; Tidwell, T. T. J. Am. Chem. Soc. 1981,
103, 3863–3867. (c) Olah, G. A.; Pittman, C. U., Jr. J. Am. Chem. Soc.
1966, 88, 3310–3312.
(11) Review: Linclau, B. Chim. Oggi 2007, 25, 51–54.
(17) Reviews: (a) Schmidt, R. R. Angew. Chem., Int. Ed. 1986, 25, 212–
235. (b) Schmidt, R. R. Pure Appl. Chem. 1989, 61, 1257–1270. (c) Schmidt,
R. R.; Castro-Palomino, J. C.; Retz, O. Pure Appl. Chem. 1999, 71, 729–
744.
(12) Examples of SAD on E-fluoroalkyl-substituted alkenes: (a) Jiang,
Z.-X.; Qing, F.-L J. Org. Chem. 2004, 69, 5486–5489. (b) Jiang, Z.-X.;
Qin, Y. Y.; Qing, F.-L. J. Org. Chem. 2003, 68, 7544–7547. (c) Zhu, W.;
Li, Z. J. Chem. Soc., Perkin Trans 1 2000, 1105–1108. (d) Wang, R.-W.;
Qing, F.-L. Org. Lett. 2005, 7, 2189–2192. (e) Audouard, C.; Barsukov, I.;
Fawcett, J.; Griffith, G. A.; Percy, J. M.; Pintat, S.; Smith, C. A. Chem.
Commun. 2004, 1526–1527.
(18) (a) Hart, D. O.; He, S.; Chany, C. J., II; Withers, S. G.; Sims,
P. F. G.; Sinnott, M. L.; Brumer, H., III Biochemistry 2000, 39, 9826–
9836. (b) Zhang, R.; McCarter, J. D.; Braun, C.; Yeung, W.; Brayer, G. D.;
Withers, S. G. J. Org. Chem. 2008, 73, 3070–3077
.
(13) (a) Katagiri, T. In Enantiocontrolled Synthesis of Fluoro-organic
Compounds; Soloshonok, V. A., Eds.;Wiley: Chicester, 1999; pp 161-178.
(b) Morelli, C. F.; Fornili, A.; Sironi, M.; Duri, L.; Speranza, G.; Manitto,
P. Tetrahedron: Asymmetry 2002, 13, 2609–2618.
(19) (a) Fried, J.; Hallinan, E. A.; Szwedo, M. J., Jr. J. Am. Chem. Soc.
1984, 106, 3871–3872. (b) Fawcett, J.; Griffiths, G. A.; Percy, J. M.; Pintat,
S.; Smith, C. A.; Spencer, N. S.; Uneyama, E. Chem. Commun. 2004, 302–
303
.
Org. Lett., Vol. 10, No. 17, 2008
3675

Scheme 4. Anionic Cyclization to the Tetrafluorinated Galactose and Glucose Derivatives
Our first investigations into the glycosylation of 18 by
anomeric alkylation are shown in Scheme 6. The use of Fried
conditions19a led to 26 in high yield, but with a modest
anomeric selectivity. Interestingly, reducing the amount of
KOH (1.3 equiv) results in a slight excess of the R-anomer
(ꢀ/R ) 1:1.1). More encouraging results in terms of anomeric
selectivity were obtained with NaH as a base, while still
obtaining a good yield. Separation of R- and ꢀ-anomers of
26 and 27 was successfully achieved by chromatography.
The anomeric configuration was tentatively assigned based
on the chemical shift of the anomeric proton, with the
equatorial anomeric proton (R-anomer) downfield compared
and 17.4 Hz) confirm that the 4-OBn is in the axial position.
Similar values were found for 26ꢀ, suggesting both anomers
4
26 exist in a C1 chair conformation.
Scheme 6. Anomeric Glycosylation
2
aBased on isolated yield of the pure anomers. bDetermined by 19F NMR
on the crude reaction mixture.
to an axial anomeric proton (ꢀ-anomer), and on the JF2-C1
2
coupling constants. The magnitude of JF2-C1 is dependent
on the orientation of an electronegative substituent on the
coupled carbon atom, with an increase in magnitude going
from a gauche to a trans orientation of the substituent with
respect to the fluorine involved in the coupling.20 In this
respect, the anomeric substituent outweighs the ring oxygen
in importance. Hence, for 26, the anomer with the higher
chemical shift for the anomeric proton also displays 2JF2-C1
values of 37.7 and 25.1 Hz, which are consistent for 26r.
The large value is the coupling of Fax with the anomeric
In summary, 2,3-dideoxy-2,2,3,3-tetrafluoro-D-galactose 2
and 2,3-dideoxy-2,2,3,3-tetrafluoro-D-glucose 3 were suc-
cessully synthesized via our perfluoroalkylidene lithium
mediated cyclization strategy. Glycosylation was successfully
achieved using an anomeric alkylation protocol. Further
research into the identification of conditions to achieve highly
R- and ꢀ-selective glycosylations is in progress.
carbon atom. For 26ꢀ, both JF2-C1 values are 23.2 Hz.21
2
Acknowledgment. R.S.T. thanks the Overseas Research
Students award scheme for funding. We wish to acknowledge
the use of the EPSRC’s Chemical database service at
Daresbury.22
Interestingly, for 26r, both 2JF3-C4 coupling constants (30.0
Supporting Information Available: Experimental pro-
Scheme 5. Final Deprotection Step
1
cedures and spectral data, including copies of H and 13C
NMR spectra, and determination of the enantioselectivity of
the SAD reaction. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL801272E
(20) Wray, V. J. Chem. Soc., Perkin II 1976, 1598–1605.
(21) These sets of coupling on stants correspond with literature examples
of R anomers of 2-deoxy-2,2-difluoro pyranosides. For the ꢀ-anomers, 2JF2-
C1 values of 28 and 20 Hz were reported: (a) El-Laghdach, A.; Echarri, R.;
Matheu, M. I.; Barrena, M. I.; Castillo´n, S.; Garcia, J J. Org. Chem. 1991,
56, 4556–4559. (b) El-Laghdach, A.; Matheu, M. I.; Castillo´n, S.; Bliard,
C.; Olesker, A.; Lukacs, G. Carbohydr. Res. 1992, 233, C1–C3.
(22) Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem. Inf. Comput.
Sci. 1996, 36, 746.
3676
Org. Lett., Vol. 10, No. 17, 2008