Synthesis of sugar-derived phostones by activation of γ-hydroxyphosphonic acids

Article abstract of DOI:10.1016/S0040-4039(03)00240-5

Treatment of carbohydrate-derived γ-hydroxyphosphonic acids under usual acetylation conditions affords the corresponding phostones in good yield. The method could be used for the preparation of free phostones as well as α- and β-phosphonomethylarabinose.

Full text of DOI:10.1016/S0040-4039(03)00240-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 2347–2349
Synthesis of sugar-derived phostones by activation of
g-hydroxyphosphonic acids
Michae¨l Bosco, Philippe Bisseret and Jacques Eustache*
Laboratoire de Chimie Organique et Bioorganique associe´ au CNRS, Universite´ de Haute-Alsace, Ecole Nationale Supe´rieure de
Chimie de Mulhouse 3, rue Alfred Werner, F-68093 Mulhouse Cedex, France
Received 27 November 2002; revised 22 January 2003; accepted 24 January 2003
Abstract—Treatment of carbohydrate-derived g-hydroxyphosphonic acids under usual acetylation conditions affords the corre-
sponding phostones in good yield. The method could be used for the preparation of free phostones as well as a- and
b-phosphonomethylarabinose. © 2003 Elsevier Science Ltd. All rights reserved.
For some time, we have been interested in phosphorus-
containing sugar analogs as surrogates of sugar 1-phos-
phates. The latter compounds are involved in a
multitude of biosynthetic pathways. In particular,
decaprenylphosphoarabinose 1 is the natural substrate
of mycobacterial arabinosyl transferases1 and 5-phos-
of the expected b- -triacetate, phostones 9 were
D
obtained. Chromatographic separation of 8 and 9 then
became obvious and alkaline hydrolysis of both acetate
and phostone groups gave phosphonates 7a and 7b.
In carbohydrate chemistry, there are several reports of
unexpected five-membered phostone formation; for
example during hydrogenolysis of g-benzyloxy-phos-
phonates (-phosphinates)6 or while performing an
Arbusov reaction on a g-iodoalcohol.7,8 We could find
pho- -arabinosyl pyrophosphate 2 is an intermediate in
D
the biosynthesis of 1.2 Both compounds may play a key
role in mycobacterial cell-wall assembly and are poten-
tial targets for new types of antibiotics.
In the course of our studies, we needed to prepare
several monoalkyl esters of a- and b-D-arabinofura-
only one example related to our work, in which forma-
tion of a phostonic acid as a minor side product was
observed during acetylation of a modified nucleoside
featuring a g-hydroxyphosphonic acid.9 In the above
examples, phostone (phostonic acid) formation was an
unwanted side reaction. In our case, it proved to be an
excellent method not only for the preparation of pure
a- and b-monoalkylphosphonates, which was our origi-
nal aim, but also for accessing carbohydrate-derived
phostones, a class of compounds with potentially inter-
esting biological activities.
nosyl-C-phosphonate 3.3 To this effect (Scheme 1) the
phosphonate 4 was prepared as described by Reitz et al.
and obtained as a mixture of anomers not separable by
chromatography.4 Selective cleavage of the phosphonyl
esters (TMSBr), activation of the resulting phosphonic
acid (CCl3CN) according to Vasella et al.5 and treat-
ment by eicosanol provided a mixture of protected
monophosphonates 6 which was converted to the mix-
ture of free sugar phosphonates 7. At this point, separa-
tion of anomers became necessary and acetylation was
performed in the hope that they could be separated at
the triacetate level. As shown in Scheme 1, the triacetyl-
ated a-phosphonate 8 was indeed formed but, instead
The method could also be used for the preparation of
free phostonic acids and free a- and b-arabinofura-
nosyl-C-phosphonic acids. Thus, alkaline treatment of
phosphonate 4 provided monoethylphosphonate 10.
Hydrogenolysis of the benzyl protecting groups and
acetylation as above again gave a triacetylated a-phos-
phonate 12 and a mixture of phostones 13. Final con-
* Corresponding author: Tel.: +33-3-8933-6858; fax: +33-3-8933-6860;
e-mail: j.eustache@uha.fr
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00240-5

2348
M. Bosco et al. / Tetrahedron Letters 44 (2003) 2347–2349
Scheme 1. Reagents and conditions: (a) TMSBr (4 equiv.), CH2Cl2, 0°C to rt, 16 h, then NEt3, quantitative; (b) (i) Cl3CCN (2
equiv.), pyridine, 15 min (ii) C20H41OH, 65°C, 15 h, 83%; (c) H2 (1 atm), 20% Pd(OH)2/C, CHCl3/MeOH (1/1), 24 h, quantitative;
(d) NEt3, then Ac2O (20 equiv.), pyridine, 15 h, 51% (8) and 31% (9) from 7, 51% (12) and 33% (13) from 11; (e) NaOH (0.5 M,
5 equiv.), 2 h, THF, then DOWEX 50W (H+ form), 90% (7a), 85% (7b); (f) NaOH (2 M, 9 equiv.), EtOH, 40°C, 3 days, then
DOWEX 50W (H+ form), 97%.
the crude mixture by 31P NMR showed the complete
disappearance of signals in the 20 ppm region (charac-
teristic of acyclic phosphonates) while four new signals
in the proportion 100/13/10/1 had appeared in the 50
ppm region (characteristic of five-membered phos-
tones). Chromatography allowed the isolation of the
two isomeric phostones 18a (22%) and 18b (26%) (abso-
lute configuration at the phosphorus atom not deter-
mined) as major products thereby confirming
preferential formation of a mixed anhydride during the
acetylation reaction and showing that the method was
Scheme 2. Reagents and conditions: (a) TMSBr (2 equiv.),
CH2Cl2, 0°C to rt, 15 h, then NEt3, quantitative; (b) NaOH (1
M, 10 equiv.), EtOH, 50°C, 15 h, then DOWEX 50W (H+
form), 96%; (c) (i) DOWEX 50W (H+ form), MeOH, 30 min
(ii) CH2N2, Et2O, 79%; (d) (i) TMSBr (4 equiv.), CH2Cl2, 0°C
to rt, 15 h, (ii) NaOH (0.5 M, 6 equiv.), water, 2 h, then
DOWEX 50W (H+ form), 83%.
indeed applicable for the preparation of free phostones
(Scheme 3).10
In conclusion, we have developed a simple and efficient
method for the formation of carbohydrate phostones
which complements the existing ones. Particularly note-
worthy is the fact that the method works well with
unprotected precursors. In our laboratories, the method
was successfully applied to other, more elaborated sys-
tems (see companion paper).
Acknowledgements
Scheme 3. Reagents and conditions: (a) Ac2O (1.2 equiv.),
pyridine, 15 h, 22% (18a), 26% (18b).
We thank Drs. D. Le Nouen and S. Bourg for help
with NMR recording and spectra interpretation. This
work was done within the frame of a COST program
(action D13 New Molecules for Human Health).
version is shown in Scheme 2: selective cleavage of the
ethyl group in 13 (TMSBr) and alkaline hydrolysis
provided in excellent yield the b-phosphonic acid 15
(Scheme 2). Surprisingly, TMSBr treatment of the a-
monoethylphosphonate 12 failed to provide the corre-
sponding phosphonic acid. The problem could be
solved by prior conversion to the ethyl methyl phos-
phonate 16 (CH2N2) then TMSBr-induced cleavage fol-
lowed by saponification of the acetates to afford
phosphonic acid 17 in high yield.
References
1. Wolucka, B. A.; McNeil, M. R.; de Hoffmann, E.; Choj-
nacki, T.; Brennan, P. J. J. Biol. Chem. 1994, 269,
23328–23335.
2. Scherman, M. S.; Kalbe-Bournonville, L.; Bush, D.; Xin,
Y.; Deng, L.; McNeil, M. J. Biol. Chem. 1996, 271,
29652–29658.
3. While this manuscript was in preparation, the synthesis
of analogs of decaprenolphosphoarabinose as inhibitors
of arabinosyltransferases was disclosed. See: Centrone, C.
A.; Lowary, T. L. J. Org. Chem. 2002, 67, 8862–8870.
The absence of b-triacetate formation during the acetyl-
ation of 7, implied that the mixed phosphonic-acetic
anhydride had been formed prior to the acetates and
suggested that direct access to non-protected phostonic
esters was possible. Thus, the free phosphonate 7b was
treated with acetic anhydride (1.2 equiv.). Analysis of

M. Bosco et al. / Tetrahedron Letters 44 (2003) 2347–2349
2349
4. Maryanoff, B. E.; Nortey, S. O.; Inners, R. R.; Campbell,
S. A.; Reitz, A. B. Carbohydr. Res. 1987, 171, 259–278.
5. Billault, I.; Vasella, A. Helv. Chim. Acta 1999, 82, 1137–
1149.
(1H, dd, J=14.8, 11.6 Hz, H-1%), 2.09 (1H, td, J=15.4,
5.5 Hz, H-1), 1.66 (2H, qn, J=7 Hz, PO(OCH2CH2)),
1.25 (34H, m, aliphatic chain), 0.87 (3H, t, J=6.5 Hz,
CH2CH3). 13C NMR (100.69 MHz, CDCl3, 300 K) l
88.41 (d, J=8.5 Hz), 87.69, 78.40, 67.20 (d, J=7.1 Hz),
61.88, 32.07, 30.63 (d, J=5.6 Hz), 29.86 (11C), 29.80,
29.74, 29.67, 29.50, 29.30, 26.89 (d, J=122.3 Hz), 25.62,
22.83, 14.25. 31P NMR (161.9 MHz, CDCl3, 300 K) l
49.86 ppm, s. HRMS calcd for C26H51O6P (M+H+)
491.3501, found 491.3490. 18b: 1H NMR (400 MHz,
CDCl3, 300 K) l 4.80 (1H, dt, J=4.5, 31.7 Hz, H-2), 4.68
(1H, t, J=4.5 Hz, H-3), 4.45 (1H, d, J=4.3 Hz, H-4),
4.12 (2H, q, J=6.8 Hz, PO(OCH2CH2)), 3.89–3.86 (2H,
m, H-5+H-6%), 3.71 (1H, m, H-6), 2.24 (1H, dd, J=16.1,
10.6 Hz, H-1%), 2.16 (1H, td, J=15.6, 5.3 Hz, H-1), 1.68
(2H, qn, J=6.8 Hz, PO(OCH2CH2)), 1.24 (34H, m,
aliphatic chain), 0.87 (3H, t, J=6.6 Hz, CH2CH3). 13C
NMR (100.69 MHz, CDCl3, 300 K) l 89.01 (d, J=9.2
Hz), 86.93, 78.77, 77.01 (d, J=6.6 Hz), 67.81 (d, J=6.3
Hz), 61.92, 32.07, 30.62 (d, J=5.7 Hz), 29.84 (11C),
29.79, 29.71, 29.64, 29.49, 29.28, 26.26 (d, J=120.9 Hz),
25.53, 22.83, 14.24. 31P NMR (161.9 MHz, CDCl3, 300
K) l 49.82 ppm, s. HRMS calcd for C26H51O6P (M+H+)
491.3501, found 491.3499.
6. (a) Yamamoto, H.; Yamamoto, K.; Inokawa, S.;
Yamashita, M.; Armour, M.-A.; Nakashima, T. T. J.
Org. Chem. 1983, 48, 435–440; (b) Racha, S.; Vargeese,
C.; Vemishetti, P.; El-Subbagh, H. I.; Abushanab, E.;
Panzica, R. P. J. Med. Chem. 1996, 39, 1130–1135.
7. Casero, F.; Cipolla, L.; Lay, L.; Nicotra, F.; Panza, L.;
Russo, G. J. Org. Chem. 1996, 61, 3428–3432.
8. For another example of unexpected phostonic acid for-
mation, see: Morr, M.; Kakoschke, C.; Ernst, L. Tetra-
hedron Lett. 2001, 42, 8841–8843.
9. Jones, G. H.; Albrecht, H. P.; Damodaran, N. P.; Mof-
fatt, J. G. J. Am. Chem. Soc. 1970, 92, 5510–5511.
10. Selected analytical data:
1
18a: H NMR (400 MHz, CDCl3, 300 K) l 4.79 (1H, dt,
J=19, 4 Hz, H-2), 4.56 (1H, t, J=4 Hz, H-3), 4.53 (1H,
broad s, H-4), 4.14–4.03 (2H, m, PO(OCH2CH2)), 3.92–
3.87 (2H, m, H-6%+H-5), 3.74 (1H, broad dd, J=11.6,
2Hz, H-6), 3.50 (1H, broad d, J=2 Hz, C(H-4)OH), 2.29