Synthesis of dihydroxylated polyamines from an erythronolactone

Article abstract of DOI:10.1016/S0040-4039(03)00693-2

The opening of protected erythronolactone by an amine or a diamine furnished hydroxy-amides. Their multistep functional conversion led to either selectively protected or free dihydroxy-polyamines.

Full text of DOI:10.1016/S0040-4039(03)00693-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3451–3453
Synthesis of dihydroxylated polyamines from an erythronolactone
Myriam Le Roch, Jacques Renault, Kristell Penlae¨ and Philippe Uriac*
UPRES 2234: Synthe`se et Extraction de Mole´cules a` Vise´e The´rapeutique, Universite´ de Rennes I,
Faculte´ des Sciences Biologiques et Me´dicales, 2, Avenue du Pr. Le´on Bernard, 35043 Rennes, France
Received 14 February 2003; revised 12 March 2003; accepted 13 March 2003
Abstract—The opening of protected erythronolactone by an amine or a diamine furnished hydroxy-amides. Their multistep
functional conversion led to either selectively protected or free dihydroxy-polyamines. © 2003 Elsevier Science Ltd. All rights
reserved.
Synthetic analogues and derivatives of polyamines (PA)
are potential therapeutic agents in various biological
disorders and attract considerable attention in organic
chemistry.1 Despite promising biological properties, the
neurotoxicity of PA derivatives, due probably to their
high affinity for receptors involved in neurotransmis-
sion, is a major limitation to their development as
therapeutics. However, recent studies2 have demon-
strated that the hydroxylation of PA structures may be
a means of decreasing their toxicity. Few routes for the
preparation of hydroxylated PA have been described.
They mostly require the preparation and the opening of
chiral epoxides.2
In accordance with this idea, we report here the synthe-
sis of di-hydroxylated putrescine, spermidine and sper-
mine from an erythronolactone (X represents two
protected hydroxyl groups, n=1, Scheme 1). Our pur-
pose was not only to prepare di-hydroxylated natural
PA but also to access unsymmetrical protected interme-
diates likely to undergo classical functionalisation.
The synthesis of our hydroxylated putrescine analogues
(Scheme 2) started with the easy opening of the com-
mercially available lactone 1 with benzylamine. Ini-
tially, we planned to modify the hydroxyl group of 2
prior to reducing its amide function. The amide would
hence play the role of a protected amine until the end
of the synthesis. Unfortunately, when submitted to
mesylation, the amido-alcohol 2 furnished incomplete
and complicated reactions. The undesired cyclisation to
lactam is a possible reason for this failure. Therefore,
the amide function of the synthon 2 was first reduced3
leading to a secondary amine. This chemical step was
carefully optimised. Borane (BH3/THF) was used ini-
tially but always led to unexploitable mixtures. Lithium
aluminium hydride was then chosen. The classical
acidic work up was replaced by the use of concentrated
sodium hydroxide to avoid hydroxyl groups deprotec-
To our knowledge, lactones have not been used as
building blocks for the synthesis of PA compounds.
However, their simple aminolysis may lead to amido-
alcohols whose functional groups can be independently
converted to amines. This approach could take advan-
tage of their natural and synthetic diversity. Scheme 1
shows the general route to this type of ‘unsymmetrical
branched PA precursor’. In this general example, the
diversity may come from R (amines or diamines), X
substituents including chiral groups and the number of
methylenes n.
Scheme 1.
Keywords: polyamines; hydroxylated polyamines; erythronolactone.
* Corresponding author. E-mail: philippe.uriac@univ-rennes1.fr
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00693-2

3452
M. Le Roch et al. / Tetrahedron Letters 44 (2003) 3451–3453
Scheme 2. Reagents and conditions: (a) C6H5CH2NH2, THF, D, 22 h; (b) H4LiAl, THF, D, 22 h; (c) (Boc)2O, CH2Cl2, 20 h; (d)
CH3SO2Cl, TEA, CH2Cl2, 0°C, 2 h; (e) C6H5CH2NH2, D, 72 h; (f) H2 3 bars, Pd(OH)2, CH3OH, 24 h; (g) HCl/EtOH 3N.
tion. The choice of the solvent (THF or diethyl ether)
took into account amide solubility and heating necessity.
In the best experimental conditions, the yield of this sole
reduction step reached 65%.
Concurrently to this first route we carried out a similar
preparation of di-hydroxylated spermidine and spermine
(Scheme 3).
Firstly the propane-1,3-diamine reacted with 1 to lead
very easily to the hydroxy amide 6. This intermediate was
poorly soluble either in diethyl ether or in THF. More-
over its reduction would furnish a diamino alcohol whose
hydrophilicity could cause problematic work-up. For
these reasons, we first protected its primary amine, then
reduced the amido group to a secondary amine that was
Boc-protected. The hydroxyl group was next converted
to mesyloxy group to give 7. Despite our efforts to
optimise the experimental procedure, the reduction step
never exceeded 45% with the result of a moderate overall
yield from 6 to 7.
After Boc protection of the secondary amine, the
hydroxyl group underwent a simple reaction with
methanesulfonyl chloride leading to compound 3.4
The mesyl group of 3 may undergo nucleophilic
displacement5 by various amines in order to obtain
diverse N-substituted putrescines. For this work, due to
the incomplete reaction with ammonia, we used benzyl-
amine as a nucleophile, then proceeded to debenzyla-
tion.6 Contrary to what could be expected,7 our experi-
mental conditions only led to mono-debenzylation,
providing the potentially useful synthon 4. In spite of our
efforts, the second catalytic debenzylation remained
unsuccessful as long as the Boc group was present. This
finding led us to treat the compound 4 as follows:
deprotection of amino and hydroxyl groups in acidic
medium, alkaline treatment then catalytic debenzylation,
finally acidic treatment to isolate 5 as the hydrochloride
salt.8 The overall yield of this sequence reached 14%.
The synthon 7 was divided in two parts: the first
underwent reaction with benzylamine then catalytic
debenzylation. For the second, an excess of propane-1,3-
diamine was used. These nucleophilic displacements
furnished the protected hydroxy PA 8 and 10. Their final
treatment in acidic medium led respectively to 9 and 11
in good yield.8
Scheme 3. Reagents and conditions: (a) propane-1,3-diamine, THF, D, 22 h; (b) (Boc)2O, CH2Cl2:CH3OH (75:25), 20 h; (c)
H4LiAl, Et2O, D, 22 h; (d) CH3SO2Cl, TEA, CH2Cl2, 0°C, 2 h; (e) C6H5CH2NH2, D, 72 h; (f) H2 3 bars, Pd(OH)2, CH3OH, 24
h; (g) propane-1,3-diamine, D, 48 h; (h) HCl/EtOH 3N.

M. Le Roch et al. / Tetrahedron Letters 44 (2003) 3451–3453
3453
In conclusion, we have presented an efficient method
for the synthesis of the dihydroxylated PA 5, 9 and 11
whose biological investigations are in progress The
protected intermediates 3, 4, 7, 8 and 10 (bearing amino
or mesyloxy groups) were easily produced in a 0.5–1 g
scale. They represent potentially useful synthons for the
preparation of diverse di-hydroxylated PA conjugates
whose biological activity deserves interest. The exten-
sion of this method to other types of lactones should be
of interest for the synthesis of diverse branched PA.
Abboud, K. A.; Phanstiel, O. J. Org. Chem. 2002, 67,
7865–7868; (b) Nagamani, D.; Ganesh, K. N. Org. Lett.
2001, 3, 103–106; (c) Rajeev, K. G.; Sanjavan, G. J.;
Ganesh, K. N. J. Org. Chem. 1997, 62, 5169–5173; (d)
Renault, J.; Lebranchu, M.; Lecat, A.; Uriac, P. Tetra-
hedron Lett. 2001, 42, 6655–6658.
6. Boyle, P. H.; Davis, A. P.; Dempsey, K. J.; Hosken, G. D.
Tetrahedron: Asymmetry 1995, 6, 2819–2828.
7. Badorrey, R.; Cativiela, C.; Diaz-de-Villegas, M. D.;
Galvez, J. A. Tetrahedron 1997, 53, 1411–1416.
8. Analytical data (NMR spectra were recorded on a Bruker
DMX 500 WB apparatus at 500 MHz. IR spectra were
recorded on a Perkin–Elmer 16PC FTIR apparatus by
diffuse reflectance. The sample were diluted in a KBr
matrix. Mass spectra were recorded on a Varian MAT 311
apparatus).
Acknowledgements
We gratefully acknowledge the Association de
Recherche sur le Cancer (ARC), The Ligue Nationale
Contre le Cancer (LNCC, Comite´ d’Ille et Vilaine) for
financial support, and to Margaret Uriac, Jean-Guy
Delcros, David Cuperly and Jean-Franc¸ois Cupif for
technical assistance to this work.
1
meso-2,3-Dihydroxyputrescine 5: H NMR (D2O) l (ppm):
3.05 (dd, J=9.22 Hz, J=13.03 Hz, 2H, H-1, H-4), 3.37
(m, 2H, H-1, H-4), 3.86 (m, 2H, H-2, H-3). 13C NMR
(D2O) l (ppm): 42.00 (C-1, C-4), 69.38 (C-2, C-3). MS
(Electrospray): (M+H)+ theor.: 121.0977, found: 121.0974.
Rf=0.43 (CH3OH:NH4OH, 50:50). FTIR w (cm−1): 3262
(OH), 3213, 3049, 2988, 2907 (NH3+). [h]D21.5=0° (c 2.1×
10−3, H2O).
References
(6S,7R)-6,7-Dihydroxyspermidine hydrochloride 9: 1H
NMR (D2O) l (ppm): 2.15 (m, 2H, H-2), 3.04* (dd,
J=9.60 Hz, J=13.20 Hz, 1H, H-8), 3.13 (m, 2H, H-1),
3.15* (m, 1H, H-5), 3.24 (t, J=8.05 Hz, 2H, H-3), 3.36*
(dd, J=2.97 Hz, J=13.20 Hz, 1H, H-8), 3.42* (dd, J=
2.81 Hz, J=12.94 Hz, 1H, H-5), 3.85* (m, 1H, H-7), 3.92*
(m, 1H, H-6). 13C NMR (D2O) l (ppm): 24.01 (C-2), 36.98
(C-1), 41.92 (C-8), 45.21 (C-3), 50.02 (C-5), 68.63* (C-6),
69.38* (C-7). MS (LSIMS): (M+H)+ theor.: 178.1556,
1. For recent reviews on polyamine synthesis and biological
activity see: (a) Karigiannis, G.; Papaioannou, D. Eur. J.
Org. Chem. 2000, 10, 1841–1863; (b) Kuksa, V.; Buchan,
R. Synthesis 2000, 9, 1189–1207 and references cited
therein.
2. (a) Bergeron, R. J.; Mu¨ller, R.; Huang, G.; McManis, J.
S.; Algee, S. E.; Yao, H.; Weimar, W. R.; Wiegand, J. J.
Med. Chem. 2001, 44, 2451–2459; (b) Bergeron, R. J.;
Mu¨ller, R.; Bussenius, J.; McManis, J. S.; Merriman, R.
L.; Smith, R. E.; Yao, H.; Weimar, W. R. J. J. Med.
Chem. 2000, 43, 224–235; (c) Bergeron, R. J.; Bussenius, J.;
Mu¨ller, R.; McCosar, B. H.; McManis, J. S. Tetrahedron:
Asymmetry 1999, 10, 4285–4294.
3. Vassis, S.; Karigiannis, G.; Balayannis, G.; Militsopoulou,
M.; Mamos, P.; Francis, G. W.; Papaioannou, D. Tetra-
hedron Lett. 2001, 42, 1579–1582.
4. Initially, the opening of the lactone 1 was undertaken by
ammonia leading to a primary amide that was subjected to
a similar sequence (b, c, d, Scheme 2) but resulted in the
intramolecular cyclisation to give the corresponding sub-
stituted N-Boc-protected pyrrolidine.
found: 178.1551. Rf=0.25 (CH3OH:NH4OH, 50:50).
FTIR w (cm−1): 3373 (OH), 3020, 2892, 2818, 2410 (NH3
NH2+). [h]2D1.5=−5.1° (c 3.34×10−3, H2O).
,
+
meso-6,7-Dihydroxyspermine hydrochloride 11: 1H NMR
(D2O) l (ppm): 2.15 (m, 4H, H-2, H-11), 3.14 (t, J=7.80
Hz, 4H, H-1, H-12), 3.17 (m, 2H, H-5, H-8), 3.25 (t,
J=7.97 Hz, 4H, H-3, H-10), 3.42 (d, J=12.92 Hz, 2H,
H-5, H-8), 3.93 (m, 2H, H-6, H-7). 13C NMR (D2O) l
(ppm): 23.57 (C-2, C-11), 36.54 (C-1, C-12), 44.78 (C-3,
C-10), 49.46 (C-5, C-8), 68.16 (C-6, C-7). MS (Electro-
spray): (M+H)+ theor.: 235.2134, found: 235.2134. Rf=
0.13 (CH3OH:NH4OH, 50:50). FTIR w (cm−1): 3387 (OH),
2969, 2901, 2807, 2516, 2422 (NH3+, NH2+). [h]D21.5=0° (c
4.42×10−3, H2O).
5. For applications to the PA synthesis, see: (a) Wang, C.;