The first synthesis of a cyclic dihydroxyacetone phosphate, a new molecule of biological importance

Article abstract of DOI:10.1016/S0040-4039(01)02178-5

A six-membered cyclic dihydroxyacetone phosphate (CDHAP) 3 (2-oxo-2-phenoxy-235-[1,2,3]-dioxaphosphinane-5-one) which is a new and interesting molecule of biological interest has been synthesised for the first time. Though dihydroxyace-tone phosphate (DHAP) 2 is very well known and is the precursor for enzymatic synthesis of sugars, the six-membered cyclic dihydroxyacetone phosphate and its synthesis have not been reported to our knowledge.

Full text of DOI:10.1016/S0040-4039(01)02178-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 503–505
The first synthesis of a cyclic dihydroxyacetone phosphate,
a new molecule of biological importance
Shyamaprosad Goswami* and Avijit Kumar Adak
Department of Chemistry, Bengal Engineering College (Deemed University), Howrah 711 103, West Bengal, India
Received 21 August 2001; revised 23 October 2001; accepted 14 November 2001
Abstract—A six-membered cyclic dihydroxyacetone phosphate (CDHAP) 3 (2-oxo-2-phenoxy-2l5-[1,2,3]-dioxaphosphinane-5-
one) which is a new and interesting molecule of biological interest has been synthesised for the first time. Though dihydroxyace-
tone phosphate (DHAP) 2 is very well known and is the precursor for enzymatic synthesis of sugars, the six-membered cyclic
dihydroxyacetone phosphate and its synthesis have not been reported to our knowledge. © 2002 Elsevier Science Ltd. All rights
reserved.
Dihydroxyacetone (DHA) exists as a dimer 1. Nor-
mally, dihydroxyacetone phosphate (DHAP) 2 is syn-
thesised enzymatically as well as chemically. DHAP has
been used extensively in organic synthesis for enzymatic
aldol reactions in the synthesis of sugars.1 Six-mem-
bered cyclic phosphates are important constituents
The DHA dimer 1 has been chemically phosphorylated
to 2 via POCl3,8 (PhO)2P(O)Cl9 and also by Wong’s
improved method.1 We reacted
1 with (p-NO2–
PhO)P(O)Cl2 in various solvents containing Et3N such
as CH2Cl2, dioxan, CH3CN and also in pyridine, but
the yield was very poor (5–10%) in all the cases due to
the notorious insolubility of 1 in the solvent used. We
reasoned that 5 could be used as a more suitable
starting material for making CDHAP 3 via conven-
tional methods. The exo-methylene function could eas-
ily be converted into a keto group by ozonolysis at low
temperature and under neutral conditions in which the
cyclic phosphate moiety should survive.
present in
a
number of biologically important
molecules, e.g. cyclic adenosine phosphate2 and Form Z
(a precursor) of molybdenum cofactor.3,4 Rajagopalan
and co-workers5,6 have shown that the substrate of
Form Z is an a-phosphorylated ketone where the phos-
phate is a part of the six-membered ring. In our syn-
thetic studies7 on the molybdenum cofactor, we were
interested to have an efficient synthesis of CDHAP 3.
We report here the synthesis of the six-membered cyclic
phosphate 3 and also the exo-methylene analogue 4.
We utilised a low temperature ozonolysis reaction for
the synthesis of 3 from a stable phosphate triester
5-methylene-2-oxo-2-phenoxy-[1,2,3]-dioxaphosphori-
nane 4. The b-hydroxyl group of Form Z, may be
generated by NaBH4 reduction of the keto group of
CDHAP.
We thus report here our successful approach to prepare
3 by this new and efficient method (Scheme 1) in good
yield. High dilution coupling of 5 and (PhO)P(O)Cl2 in
dry CH2Cl2 and Et3N afforded a thick, light yellow
liquid 4, which on ozonolysis gave the pale brown
semi-solid 3 in 70% overall yield. Compound 4 was also
made via the formation of intermediate 6 by the reac-
tion of 5 with POCl3 or PCl5. Intermediate 6 was not
Keywords: dihydroxyacetone dimer; cyclic phosphate; molybdenum cofactor; cyclic keto phosphate; ozonolysis.
* Corresponding author. E-mail: spgoswamical@yahoo.com
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02178-5

504
S. Goswami, A. K. Adak / Tetrahedron Letters 43 (2002) 503–505
Scheme 1.
isolated, but directly converted to compound 4 by
reaction with phenol in dry CH2Cl2 at room tempera-
ture. However, the (PhO)P(O)Cl2 route was found to be
more efficient for the synthesis of 4. Compounds 3 and
4 were well characterised10 by spectroscopic means
including 31P NMR.
A solid state X-ray study11 of 2-oxo-2-phenoxy-1,2,3-
dioxaphosphorinane 7 (R=H) revealed that the phen-
oxy substituent favours an axial orientation in the chair
conformation. In 3 and 4, the phosphorus is more
strongly coupled to the equatorial protons (3JPꢀH (eq.)
=
18.73 Hz for 3 and 18.91 Hz for 4) compared to the
axial protons (3JPꢀH (ax.)<4.5 Hz both for 3 and 4) (Fig.
1 and Fig. 2). These findings are consistent with the
3JPꢀH coupling constants for axial and equatorial pro-
tons in 7 (R=CH3)12 and those found in other
dioxaphosphorinane ring systems13 and thus, it is likely
that 3 and 4 exist as a single conformer at room
temperature.
1
Figure 2. 1H decoupled 31P NMR (top) and H undecoupled
31P NMR (bottom) of 4 at 22°C.
Acknowledgements
In conclusion, we have developed a new route for the
synthesis of a cyclic dihydroxyacetone phosphate. The
new compounds, e.g. 5-methylene-2-oxo-2-phenoxy-
[1,2,3]-dioxaphosphorinane and CDHAP are promising
synthons for the synthesis of keto sugars or analogues
attached to different groups having biological
importance.
We thank C.S.I.R., Govt. of India for financial support
and Bose Institute, Kolkata for the paid facility of 500
MHz NMR. We also acknowledge Professor S. Ghosh
for the facility of Welbash Ozonapor instrument at the
Department of Organic Chemistry at IACS, Kolkata.
References
1. Pederson, R. L.; Esker, J.; Wong, C. H. Tetrahedron
1991, 47, 2643–2648 and references cited therein.
2. Lipkin, D.; Cook, W. H.; Markham, R. J. Am. Chem.
Soc. 1959, 81, 6198.
3. Hille, R. Chem. Rev. 1996, 96, 2757–2816.
4. Pateman, J. A.; Coves, D. J.; Rever, B. M.; Roberts, D.
B. Nature 1964, 201, 58.
5. Johnson, J. L.; Wuebbens, M. M.; Rajagopalan, K. V. J.
Biol. Chem. 1989, 264, 13440–13447.
6. Wuebbens, M. M.; Rajagopalan, K. V. J. Biol. Chem.
1993, 268, 13493–13498.
7. (a) Pilato, R. S.; Erickson, K. A.; Greaney, M. A.; Stiefel,
E. I.; Goswami, S. P.; Kilpatrick, L.; Spiro, T. G.;
Taylor, E. C.; Rhiengold, A. L. J. Am. Chem. Soc. 1991,
113, 9372–9374; (b) Goswami, S. P. Heterocycles 1993,
1
Figure 1. 1H decoupled 31P NMR (top) and H undecoupled
31P NMR (bottom) of 3 at 22°C.

S. Goswami, A. K. Adak / Tetrahedron Letters 43 (2002) 503–505
505
35, 1552–1572; (c) Goswami, S. P.; Adak, A. K. On
asymmetric synthesis of Form Z, a precursor to molybde-
num co-factor. International Symposium on Current
Trends in Drug Discovery Research, CTDDR-2001 11–
15th Feb, CDRI, Lucknow, India, 2001; Abs c OP-
XVIV-5.
(CDCl3, 500 MHz): l 7.28 (d, 2H, J=8.0 Hz), 7.19 (d,
2H, J=8.0 Hz), 7.12 (t, 1H, J=7.3 Hz), 5.23 (d, 2H,
J=9.4 Hz), 4.89 (dd, 2H, JHꢀH=12.5 Hz, JPꢀOꢀCꢀH=4.5
Hz (ax.)), 4.72 (dd, 2H, JHꢀH=12.8 Hz, JPꢀOꢀCꢀH=19.0
Hz (eq.)). 13C NMR (CDCl3, 500 MHz): 150.79 (d,
J
PꢀC=6.7 Hz, ꢁCH2), 135.12 (d, 3JPꢀC=9.6 Hz, C),
130.31 (CH), 125.61 (d, J=1 Hz, CH), 120.02 (d, J=5.1
Hz, CH), 117.89 (CH), 72.2 (d, 2JPꢀC=7.5 Hz, 2×CH2).
1H decoupled 31P NMR (CDCl3, 500 MHz): 4.78. Mass
(FAB): 227.1 (MH+).
8. Effenberger, F.; Strub, A. Tetrahedron Lett. 1987, 28,
1641.
9. Colbran, R. L.; Jones, J. K. N.; Matheson, N. K.;
Rozema, I. Carbohydr. Res. 1967, 4, 352.
10. Spectral analysis: For 3: 1H NMR (CDCl3, 300 MHz):
7.34–7.29 (m, 3H), 7.18–7.15 (m, 2H), 4.84–4.60 (m, 4H).
13C NMR (CDCl3, 500 MHz): 202.09 (CꢁO), 133.43
(CH), 130.38 (CH), 125.84 (CH), 120.02 (d, J=5.1 Hz,
11. Geise, H. J. Rec. Trav. Chim. 1967, 86, 362.
12. Hall, L. D.; Malcolm, R. B. Can. J. Chem. 1972, 2092–
2101.
13. Hulst, R.; Visser, J. M.; Koen de Vries, N.; Zijlstra, R.
W. J.; Kooijman, H.; Smeets, W.; Speck, A. L.; Feringa,
B. L. J. Am. Chem. Soc. 2000, 122, 3135–3150 and
references cited therein.
2
CH), 118.26 (CH), 116.42 (CH), 72.39 (d, JPꢀC=7.3 Hz,
2×CH2). 1H decoupled 31P NMR (CDCl3, 500 MHz):
5.20. Mass (FD): 228 (M+), 36%. For 4: 1H NMR