Synthesis of gem-difluorocarba-d-glucose: A step further in sugar mimesis

Article abstract of DOI:10.1002/anie.200461244

A rearrangement strategy was used for the synthesis of α- and β-gem-difluorocarba-D-glucose, which are close congeners of α- and β-D-glucose, in which the endocyclic oxygen atom has been replaced by a gem-difluoromethylene group (see scheme). The two anom

Full text of DOI:10.1002/anie.200461244

Communications
Carbohydrate Chemistry
Synthesis of gem-Difluorocarba-d-glucose: A Step
Further in Sugar Mimesis**
AurØlie Deleuze, Candice Menozzi,
Matthieu Sollogoub,* and Pierre Sinay*
¨
Chemical modifications of carbohydrates have been used
extensively to understand the origin of the specificity in the
recognition processes by sugar-binding proteins in aqueous
solution. A classical example is the work by Lemieux, wherein
chemoselective removal of hydroxyl groups from oligosac-
charides led to the concept of key polar interactions.[1] Along
the same lines, the replacement of the endocyclic oxygen
atom by a methylene group has provided hydrolytically stable
glycoside mimetics called carbasugars, whose biochemical
properties have been studied (Figure 1).[2–8] Such a replace-
Figure 1. Glucopyranose and its carbocyclic congeners.
ment has the inherent disadvantage to suppress any possible
hydrogen bond formation that involved this electronegative
atom. This drawback clearly appeared in the case of
carbalactoside, a close mimic of lactoside, which is no
longer recognised by a b-galactosidase.[9] A model has
shown that the endocyclic oxygen atom (O-5) of the d-
galactose moiety is involved in the interaction with hydrogen-
donating groups of the active site of the enzyme.[10] One way
to circumvent this problem and to enlarge the probing
spectrum of carbasugars is to replace the oxygen atom by
other hydrogen-bond acceptors. Ideally, to fully understand
the role of the endocyclic oxygen requires the two lone pairs
to be discriminated so that they can be probed to see which of
them, if any, is involved in hydrogen bonding in the active site
of the enzyme. Fluorine atoms are not as strong hydrogen-
bond acceptors as oxygen, but they still exhibit this prop-
erty,[11–13] and more interestingly they have been used as
reporter groups of the active site of enzymes through
19F NMR spectroscopy.[14] It is therefore foreseeable that
¨
[*] A. Deleuze, C. Menozzi, Dr. M. Sollogoub, Prof. P. Sinay
DØpartement de Chimie
Ecole Normale SupØrieure
UMR 8642: CNRS-ENS-UPMC Paris 6
24 rue Lhomond
75231 Paris Cedex 05 (France)
Fax: (+33)1-4432-3397
E-mail: matthieu.sollogoub@ens.fr
pierre.sinay@ens.fr
[**] The authors would like to thank Dr. Corinne Aubert for useful
discussions about cobalt.
6680 ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200461244
Angew. Chem. Int. Ed. 2004, 43, 6680 –6683

Angewandte
Chemie
replacement of the endocyclic CH2 group by a CF2 moiety in a
carbasugar would constitute a step further in the utilization of
this family of sugar mimetics by enhancing their ability to be
accepted by enzymes involving O-5 and allowing the study of
their complexation by NMR spectroscopy.
We recently reported a novel sugar-to-carbocycle rear-
rangement promoted by triisobutylaluminum (TIBAL)[15] or
Cl3TiOiPr.[16] The proposed key step in this transformation is
the opening of the ring to give a carbocationic intermediate.
Indeed, this reaction is favored when this putative carbenium
species can be stabilized by an electron-donating group
(EDG).[17] This is generally applicable to acyclic non-sugar-
derived enol ethers (Scheme 1).[18]
Scheme 3. Synthesis of the gem-difluoro-exo-glucal 6. Reagents and
conditions: 1) a) nBuLi (2.1 equiv), THF, ꢀ788C, 2 h; b) MeI (6 equiv),
THF, ꢀ788C, 30 min; c) HMPT (4 equiv), ꢀ788C!RT, overnight, 75%
over three steps; 2) H2SO4 (1 equiv), Ac2O, 08C, 3 min, 75%;
3) MeONa (4 equiv), MeOH, 08C!RT, 2 h, 90%; 4) PCC (3 equiv), 4-
Š molecular sieves (3 equiv), CH2Cl2, room temperature, 4 h, 74%;
5) a) HMPT (5 equiv), THF, ꢀ408C!RT; b) CBr2F2 (5 equiv), HMPT
(5 equiv), THF, RT!reflux, 1 h, 95%. THF=tetrahydrofuran,
HMPT=hexamethylphosphoric triamide, PCC=pyridinium chlorochro-
mate.
Scheme 1. General rearrangement of EDG-substituted vinyl ethers. EDG=
electron-donating group, LA=Lewis acid, TIBAL=triisobutylaluminum.
As an example, we recently synthesized a carbasugar with
furanyl as the EDG which was easily introduced at the
anomeric centre of d-glucose through a glycosylation process.
In this synthesis, the anomeric center of the carba congener is
derived from the C-5 carbon atom of d-glucose.[19] Given the
As anticipated and much to our delight, the key rear-
rangement was performed through a two-step one-pot
procedure that involved first, the complexation of the
alkyne 6 by dicobalt octacarbonyl,[25] followed by the reduc-
tive TIBAL-induced rearrangement of 2.[17] Decomplexation
with CAN (ceric ammonium nitrate)[26] of the
crude compound 7 afforded alcohol 8 in 75%
yield over three steps and only one purification
step by chromatography. The 5a-gem-difluoro-
carba-a-d-glucopyranose 1[27] was then easily
obtained by controlled reduction of the triple
bond, reductive ozonolysis, and debenzylation
(Scheme 4). As previously observed,[15] the
rearrangement occurs with retention of config-
uration. The reduction of the putative transient
Scheme 2. Synthetic strategies for carbasugars through TIBAL-promoted
rearrangements. Bn=benzyl.
accessibility of ynopyranosides[20] and the recent results of
Harrity and co-workers[21] who showed that dicobalt hexa-
carbonyl clusters[22] could act as EDGs in a similar situation,
we anticipated that our TIBAL-mediated reductive rear-
rangement applied to the difluoroalkene 2 might provide a
convenient entry to our target molecule 1 (Scheme 2).
Synthesis of the ynopyranose 6, the precursor of com-
pound 2, is depicted in Scheme 3. The known[20] gem-
dibromoalkene 3 was first converted into alkyne 4 through
methylation of the acetylenic anion generated in situ. Ace-
tolysis,[23] followed by hydrolysis of the generated anomeric
acetate, and oxidation of the hemiacetal afforded the
corresponding lactone 5. Difluoromethylenation according
to Motherwell et al.[24] provided the desired gem-difluoroal-
kene precursor 6.
Scheme 4. Synthesis of the 5a-gem-difluorocarba-a-d-glucopyranose 1.
Reagents and conditions: 1) a) [Co2(CO)8] (1.5 equiv), CH2Cl2, room
temperature, 2 h; b) TIBAL (5 equiv), toluene, room temperature,
2.5 h; c) CAN (5 equiv), NEt3 (1 equiv), acetone, 30 min, 65% over
three steps; 2) a) Pd/CaCO3, H2, MeOH, room temperature, 4 h;
b) O3, CH2Cl2, ꢀ788C, 5 min; c) NaBH4, CH2Cl2, room temperature,
1 h, 76% over three steps; 3) Pd/C, H2, MeOH, room temperature,
1 h, 90%. CAN=ceric ammonium nitrate.
Angew. Chem. Int. Ed. 2004, 43, 6680 –6683
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6681

Communications
ketone is stereoselective and is classically explained by the
The observed stereoselectivity of this reduction process is
in sharp contrast to the case in which the fluorine atoms are
absent from the molecule. Indeed, when the unsaturated
sugar 15[32] was submitted to the same reaction conditions,
only the axial alcohol 16[15] was obtained in 75% yield
(Scheme 7). This result supports the directing role of fluorine
in the reduction of the ketone.
attack of the hydride ion, which is attached to the isobutyl
group, on the less-hindered face of the molecule to provide
the axial hydroxyl group (Scheme 5).
Scheme 7. Rearrangement and reduction of 15. Reagents and condi-
tions: 1) a) TiCl3OiPr (1 equiv), CH2Cl2, ꢀ788C; b) THF (5 equiv),
ꢀ788C, 15 min; c) Et3BHLi (2 equiv), CH2Cl2, ꢀ788C, 30 min, 73%
over three steps.
Scheme 5. Stereoselectivity of the reduction of the transient ketone by
TIBAL.
The reduction step described proceeds under steric
control, whereas the electronic control admitted in such a
case, the Ahn–Eisenstein effect,[28] gives the opposite stereo-
selectivity. According to this model, the most stable transi-
In summary, we have synthesized, for the first time, two
gem-difluorinated carba analogues of a- and b-d-glucopyra-
noses by using a rearrangement strategy and by taking
advantage of the Anh–Eisenstein effect to obtain the b
analogue. They are attractive candidates for probing the role
of endocyclic oxygen atoms in carbohydrates in sugar–protein
interactions. This strategy is currently being extended to other
carbohydrates.
[29,30]
ꢀ
tion-state structure has the axial C F
bond in the
ꢀ
antiperiplanar position with respect to the forming H C
bond. This model implies, as shown in
Figure 2, which depicts the late transition
state of the reduction step, a selective
attack of the hydride ion to form the
equatorial hydroxyl group.
Received: July 8, 2004
Once again, a one-pot procedure was
Figure 2. Anh–
Eisenstein effect to
obtain the axial
reduction.
developed: After reaction of the alkyne 6
with dicobalt octacarbonyl to give cluster
Keywords: carbohydrates · cyclization · fluorine · reduction ·
.
synthetic methods
2, the rearrangement was, this time,
induced by Cl3TiOiPr[16] to yield cyclo-
hexanone 10. The Lewis acid was then
quenched with THF, and the ketone was reduced by means of
the super hydride. The triple bond in the crude product 11 was
decomplexed, and the expected equatorial alcohol 12 was
obtained in 73% yield over five steps and only one
purification step by column chromatography. The 5a-gem-
difluorocarba-b-d-glucopyranose 14[31] was then obtained as
described above (Scheme 6).
[1] a) R. U. Lemieux, Chem. Soc. Rev. 1989, 18, 347 – 374; b) R. U.
Lemieux, Acc. Chem. Res. 1996, 29, 373 – 380.
[2] T. Suami, S. Ogawa, Adv. Carbohydr. Chem. Biochem. 1990, 48,
21– 90.
[3] S. Ogawa, N. Matsunaga, H. Li, M. M. Palcic, Eur. J. Org. Chem.
1999, 3, 631– 642.
[4] M. Carpintero, A. Fernandez-Mayoralas, J. Jimenez-Barbero,
Eur. J. Org. Chem. 2001, 4, 681– 689.
[5] S. Ogawa, Y. Uematsu, S. Yoshida, N. Sasaki, T. Suami, J.
Carbohydr. Chem. 1987, 6, 471– 478.
[6] S. Ogawa, A. Maruyama, T. Odagiri, H. Yuasa, H. Hashimoto,
Eur. J. Org. Chem. 2001, 5, 967 – 974.
[7] M. Carpintero, A. Bastida, E. Garcia-Junceda, J. Jimenez-
Barbero, A. Fernandez-Mayoralas, Eur. J. Org. Chem. 2001, 4,
4127 – 4135.
[8] M. Kitaoka, S. Ogawa, H. Taniguchi, Carbohydr. Res. 1993, 247,
355 – 359.
[9] E. Montero, A. Garcia-Herrero, J. L. Asensio, K. Hirai, S.
Ogawa, F. Santoyo-Gonzalez, F. J. Canada, J. Jimenez-Barbero,
Eur. J. Org. Chem. 2000, 1945 – 1952.
[10] J. F. Espinosa, E. Montero, A. Vian, J. L. Garcia, H. Dietrich, M.
Martin-Lomas, R. R. Schmidt, A. Imberty, F. J. Canada, J.
Jimenez-Barbero, J. Am. Chem. Soc. 1998, 120, 10862 – 10871.
[11] L. A. Curtiss, D. J. Frurip, M. Blander, J. Am. Chem. Soc. 1978,
100, 79 – 86.
[12] D. A. Dixon, B. E. Smart, J. Phys. Chem. 1991, 95, 1609 – 1612.
[13] J. A. K. Howard, V. J. Hoy, D. OꢀHagan, G. T. Smith, Tetrahe-
dron 1996, 52, 12613 – 12622.
Scheme 6. Synthesis of the 5a-gem-difluorocarba-b-d-glucopyranose 14. Reagents
and conditions: 1) a) [Co2(CO)8] (1.5 equiv), CH2Cl2, room temperature, 2 h;
b) TiCl3OiPr (2 equiv), CH2Cl2, ꢀ788C, 1 h; c) THF (5 equiv), ꢀ788C, 15 min;
d) Et3BHLi (2 equiv), CH2Cl2, ꢀ788C, 30 min; e) CAN (5 equiv), NEt3 (1 equiv),
acetone, 1 h, 73% over five steps; 2) a) Pd/CaCO3, H2, MeOH, room tempera-
ture, 1.25 h; b) O3, CH2Cl2, ꢀ788C, 5 min; c) NaBH4, CH2Cl2, room temperature,
1 h, 78% over three steps; 3) Pd/C, H2, MeOH, room temperature, 1 h, 90%.
[14] J. Feeney, J. E. McCormick, C. J. Bauer, B. Birdsall, C. M.
Moody, B. A. Starkmann, D. W. Young, P. Francis, R. H. Havlin,
6682
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 6680 –6683

Angewandte
Chemie
W. D. Arnold, E. Oldfield, J. Am. Chem. Soc. 1996, 118, 8700 –
8706.
[15] S. K. Das, J.-M. Mallet, P. Sinay¨, Angew. Chem. 1997, 109, 51 3 –
516; Angew. Chem. Int. Ed. Engl. 1997, 36, 493 – 496.
[16] M. Sollogoub, J.-M. Mallet, P. Sinay¨, Tetrahedron Lett. 1998, 39,
3471– 3472.
[17] M. Sollogoub, J.-M. Mallet, P. Sinay¨, Angew. Chem. 2000, 112,
370 – 372; Angew. Chem. Int. Ed. 2000, 39, 362 – 364.
[18] B. du Roizel, M. Sollogoub, A. J. Pearce, P. Sinay¨, Chem.
Commun. 2000, 1507 – 1508.
[19] M. Sollogoub, A. J. Pearce, A. Herault, P. Sinay¨, Tetrahedron:
Asymmetry 2000, 11, 283 – 294.
[20] D. Rouzaud, P. Sinay¨, J. Chem. Soc. Chem. Commun. 1983,
1353 – 1354.
[21] D. R. Carbury, S. Reigner, J. W. Myatt, N. D. Miller, J. P. A.
Harrity, Angew. Chem. 2002, 114, 2699 – 2701; Angew. Chem. Int.
Ed. 2002, 41, 2584 – 2587.
[22] K. M. Nicholas, Acc. Chem. Res. 1987, 20, 207 – 214.
[23] J. M. Aurrecoechea, B. Lopez, M. Arrate, J. Org. Chem. 2000, 65,
6493 – 6501.
[24] W. B. Motherwell, M. J. Tozer, B. C. Ross, J. Chem. Soc. Chem.
Commun. 1989, 1437 – 1439.
[25] K. M. Nicholas, M. Mulvaney, M. Bayer, J. Am. Chem. Soc. 1980,
102, 2508 – 2510.
[26] D. Seyferth, A. T. Wehman, J. Am. Chem. Soc. 1970, 92, 5520 –
5522.
1
[27] 1: [a]2D0 = + 2.4 (c = 0.9, CH3OH); H NMR (400 MHz, CD3OD,
258C, TMS): d = 4.18 (dd, 3J(5,6) = 4.8, 3J(6,6’) = 11.6 Hz, 1H;
3
3
H-6), 4.14 (dd, J(5,6’) = 4.4, J(6,6’) = 11.6 Hz, 1H; H-6’), 4.05
(ddd, J(1,2) = 3J(1,Feq) = 3.9, J(1,Fax) = 7.8 Hz, 1H; H-1), 3.83
3
3
(t, J(3,4) = 3J(3,2) = 9.4 Hz, 1H; H-3), 3.64–3.59 (m, 1H; H-2),
3
3.61(t, 3J(3,4) = 3J(4,5) = 10.1 Hz, 1H; H-4), 2.50 ppm (ddq,
3J(5,6) = 3J(5,Feq) = 4.2, 3J(5,Fax) = 30.7, 3J(5,4) = 11.0 Hz, 1H;
H-5); 13C NMR (100 MHz, CD3OD, 258C): d = 124.1 (dd,
1J(C,F) = 240.8, 1J(C,F) = 251.8 Hz; CF2), 75.6 (s; C-3), 73.5
(dd, 2J(C,F) = 22.1, 2J(C,F) = 32.3 Hz; C-1), 72.1 (d, 3J(C,F) =
11.0 Hz; C-2 or C-4), 72.0 (d, 3J(C,F) = 9.0 Hz; C-2 or C-4), 58.7
3
2
(t, J(C,F) = 2.9 Hz; C-6), 46.5 ppm (t, J(C,F) = 20.0 Hz; C-5).
19F NMR (235 MHz, CD3OD, 258C): d = ꢀ110.3 (d, 2J(F,F) =
256.5 Hz; Feq), ꢀ115.7 ppm (ddd, 2J(F,F) = 256.5, 3J(F,H-5) =
30.6, 3J(F,H-1) = 3.5 Hz; Fax); DCI-MS (desorption chemical
ionization; NH3): m/z: 232 [M+NH3+H]+, 21 5 [M+H]+; HRMS
(DCI, NH3): calcd for C7H13O5F2: 215.0731; found: 215.0726.
[28] N. T. Anh, O. Eisenstein, Nouv. J. Chim. 1977, 1, 61– 70.
[29] S. S. Wong, M. N. Paddon-Row, J. Chem. Soc. Chem. Commun.
1990, 456 – 458.
[30] C. F. Bridge, D. OꢀHagan, K. A. Reynolds, K. K. Wallace, J.
Chem. Soc. Chem. Commun. 1995, 2329 – 2330.
[31] 14: [a]2D0 = + 1.0 (c = 1.2, CH3OH); 1H NMR (400 MHz,
CD3OD, 258C, TMS): d = 4.18 (dd, 3J(5,6) = 4.6, 3J(6,6’) =
12.0 Hz, 1H; H-6), 4.15 (dd, 3J(5,6’) = 4.6, 3J(6,6’) = 12.0 Hz,
1H; H-6’), 3.80–3.71 (m, 1H; H-2), 3.68–3.62 (m, 1H; H-4),
3.53–3.48 (m, 2H; H-1, H-3), 2.12 ppm (ddq, 3J(5,6) = 3J(5,Feq) =
4.2, 3J(5,Fax) = 28.6, 3J(5,4) = 11.6 Hz, 1H; H-5); 13C NMR
(100 MHz, CD3OD, 258C): d = 122.8 (dd, 1J(C,F) = 245.4;
1J(C,F) = 248.9 Hz; CF2), 78.2 (d, 4J(C,F) = 1.2 Hz; C-3), 75.2
(t, 2J(C,F) = 20.3 Hz; C-1), 73.2 (d, 3J(C,F) = 9.5 Hz; C-2 or C-4),
71.7 (d, 3J(C,F) = 10.2 Hz; C-2 or C-4), 58.6 (t, 3J(C,F) = 3.1Hz;
2
C-6), 49.0 ppm (t, J(C,F) = 20.0 Hz; C-5). 19F NMR (235 MHz,
CD3OD, 258C): d = ꢀ110.3 (d, 2J(F,F) = 260.0 Hz; Feq),
ꢀ126.1 ppm (ddd, 2J(F,F) = 260.0, 3J(F,H-5) = 28.2, 3J(F,H-1) =
21.2 Hz; Fax); DCI-MS (NH3): m/z: 232 [M+NH3+H]+; HRMS
(DCI, NH3): calcd for C7H16O5NF2: 232.0997; found 232.0999.
[32] T. Iimori, H. Takahashi, S. Ikegami, Tetrahedron Lett. 1996, 37,
649 – 652.
Angew. Chem. Int. Ed. 2004, 43, 6680 –6683
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6683