Regio- and chemoselective one-step 3-O-alkylenation of unprotected ascorbic acid using ω-iodoalkanols

Article abstract of DOI:10.1055/s-0028-1087662

A regio- and chemoselective alkylenation employing unprotected ascorbic acid and a series of unprotected iodoalkanols in the presence of sodium hydrogen carbonate in dimethyl sulfoxide is described. This atom economic high yielding procedure delivers the

Full text of DOI:10.1055/s-0028-1087662

LETTER
217
Regio- and Chemoselective One-Step 3-O-Alkylenation of Unprotected
Ascorbic Acid Using w-Iodoalkanols
O
ne-Ste
p
3-O-A
h
lkylenation
i
of
U
np
e
rotecte
d
A
s
r
corbic
A
r
c
id y Muller,a Paul Heuschling,b Bang Luu*c
a
Institut für Organische Chemie, Universität Karlsruhe (TH), Fritz-Haber-Weg 6, 76131 Karlsruhe, Germany
Laboratoire de NeuroBiologie, Unité de Recherche Sciences de la Vie, Faculté des Sciences, de la Technologie et de la Communication,
Université du Luxembourg, 1511 Luxembourg, Luxembourg
Laboratoire de Chimie Organique des Substances Naturelles, UMR 7177 CNRS, Université Louis Pasteur,
67084 Strasbourg Cedex, France
b
c
Fax +33(388)411672; E-mail: bang.luu@orange.fr
Received 22 September 2008
Various protocols use 5,6-isopropylidene-L-ascorbic acid
as starting material and deliver a mixture of 2- and 3-alky-
lation products.8 In some cases however, regio- and
chemoselective 3-O-alkylation is observed.9 Selective 2-
Abstract: A regio- and chemoselective alkylenation employing un-
protected ascorbic acid and a series of unprotected iodoalkanols in
the presence of sodium hydrogen carbonate in dimethyl sulfoxide is
described. This atom economic high yielding procedure delivers the
corresponding 3-O-alkylene ethers in a single step without prior O-alkylation can be achieved after protection of the more
protection. Specific 3-O-etherification was also observed with
acidic position 3.
unprotected 16-iodohexadecanoic acid and with 2-(10-iodode-
cyl)isoindole-1,3-dione. Furthermore, an 2-O-alkylene derivative
In addition to these procedures, several groups have de-
veloped methods to address 3-alkyl ethers without the
need for prior protection. Unprotected L-ascorbic acid
can, for example, be regio- and chemoselectively alkylat-
ed at position 3 under Mitsunobu conditions in good
yields.10
was obtained when 3-O-benzyl ascorbic acid was reacted with un-
protected 10-iododecanol under slightly modified conditions. Final-
ly, the antioxidative activity of all compounds was determined and
compared to vitamin C and E.
Key words: ascorbic acid, regioselective, chemoselective, 3-O-
alkylenation, w-iodoalkanols
HO
OBn
HO
OH
HO
I
a
b
(CH2)n
(CH2)n
(CH2)10
2
1
3
For some time now, our respective groups have been in-
volved in the synthesis and biological evaluation of hybrid
compounds combining an antioxidant moiety with a w-al-
kanol side chain for the treatment of neurodegenerative
and neuroinflammatory diseases.1 This has resulted in dif-
ferent lead structures, namely cyclohexenon-,2 toco-
pherol-,3 quinol-4 and lately resveratrol-5 fatty alcohols
having interesting pharmacological properties. Here we
report on preliminary results concerning the antioxidative
activity of a new series of hybrid compounds merging the
nucleus of vitamin C at position 3 with different iodoalky-
lene side chains.
c
d
e
BnO
OBn
MsO
I
NPhth
I
(CH2)10
(CH2)10
(CH2)10
4
5
6
O
O
O
f
I
HO
(CH2)15
(CH2)14
7
8
Scheme 1 Syntheses of the different chains. Reagents and conditi-
ons: (a) n = 10, 12, 14, 16: aq HI (57%), toluene, 90 °C, 6 h (61–78%);
(b) n = 10: NaH, BnBr, NBu4Br, THF, reflux, 12 h (49%); (c) n = 10:
NaH, BnBr, THF, reflux, 12 h (95%); (d) n = 10: phthalimide, DEAD,
Ph3P, CH2Cl2, 0 °C to r.t., 24 h (93%); (e) MsCl, Et3N, CH2Cl2, 0 °C
to r.t., 18 h (quant); (f) aq HI (57%), AcOH, 60 °C, 6 h (96%).
Ascorbic acid or vitamin C is one of Nature’s most potent
water-soluble antioxidants which protects cells and tis-
sues against oxidative damage. Besides, vitamin C is
known to play an important role in the prevention of sev-
eral chronic diseases.6 Although its high hydrosolubility
facilitates oral uptake, it seriously limits its application in
certain domains, e.g. as lipid peroxidation inhibiting
agent.7 As a consequence, numerous efforts have been
made to enhance its lipophilicity, mostly by adding alkyl
side chains. The majority of the literature references on
this subject focus on the preparation of 2- or 3-O-alkyl
ascorbic acids.8–13
Regioselective one-pot 3-O-alkylation of unprotected
ascorbic acid has also been achieved using alkyl mesy-
lates in the presence of sodium hydrogen carbonate in
DMSO11 or by refluxing alkyl mesylates or sulfonates
with triethylamine in alcohols.12 O-Alkylenation reactions
of vitamin C are much rarer though13 and to the best of our
knowledge, there is no example to date of a one-pot syn-
thesis of such a compound.
Here we present an efficient one-pot synthesis for 3-O-
alkylene ethers of vitamin C without need for protecting
groups which is based on a slight modification of a known
SYNLETT 2009, No. 2, pp 0217–0220
Advanced online publication: 15.01.2009
DOI: 10.1055/s-0028-1087662; Art ID: G28608ST
x
x.
x
x.
2
0
0
9
© Georg Thieme Verlag Stuttgart · New York

218
T. Muller et al.
LETTER
chemo- and regioselective approch.11 These compounds Using crude mesylate 6, the corresponding 3-O-ether 11a
are very interesting as potent amphiphilic antioxidants. In was obtained in only 42% yield after 24 hours at 60 °C
addition to the corresponding alkyl ethers of vitamin C, al- (Table 1, entry 1). We consequently looked for other leav-
ready used in this context, our compounds with their ter- ing groups and turned towards the corresponding iodide 4.
minal functionalization offer the possibility for further We were pleased to note that the corresponding ether 11a
derivatization.
was obtained in 68% yield in 19 hours (entry 2). Next, we
tried reacting directly unprotected 10-iododecan-1-ol (2a)
which resulted in decomposition of the starting materials
during the acidic workup.11b Performing the workup in
neutral conditions led to the desired 3-O-alkylene ascor-
bic acid 11b in 80% yield in only three hours (entry 3).16
All attempts to use another solvent (acetone, toluene,
methanol) in order to avoid the distillation under reduced
pressure of DMSO, only led to recovery of the starting
materials. Using the thus adapted method allowed to ob-
tain 3-O-ethers 11c–e with even longer alkanol chains
(entries 4–6) as well as the corresponding 3-alkyl ethers
11f–i (entries 7–10) in high yields. It was only when very
long alkyl chains were reacted that the yield started to
drop (entry 10). Finally, we tried to regioselectively 3-O-
alkylenate vitamin C using 2-(10-iododecyl)isoindole-
1,3-dione (5; entry 11) and unprotected 16-iodohexade-
canoic acid (8; entry 12). As before, the resulting com-
pounds 11j,k were obtained as single product in good
yields.
In order to access the different 3-O-alkylene ethers, we
first had to prepare the different alkylene chains
(Scheme 1). Diols 1 were monoiodinated using a 57%
aqueous solution of hydroiodic acid in toluene.3a Com-
pound 5 was obtained as single product via a Mitsunobu
reaction of iodoalcohol 2a and phthalimide.14 Benzylation
of 2a in the presence of sodium hydride gave [(10-iodo-
decyloxy)methyl]benzene (4). Monobenzylation of de-
can-1,10-diol (1a) afforded 10-(benzyloxy)decan-1-ol
(3)3b which was further converted into mesylate 6.11 16-
Iodohexadecanoic acid (8) was obtained by reacting lac-
tone 7 in glacial acetic acid in the presence of hydroiodic
acid.15
With all the different alkylene chains in hand, we started
investigating the regioselective etherification of vitamin
C. We began to explore the procedure developed by
Beifuss et al. reacting unprotected ascorbic acid with me-
sylates and sodium hydrogen carbonate in DMSO
(Scheme 2, Table 1).11
We next adapted our method based on a slightly modified
procedure11 to the synthesis of 2-O-alkylene ascorbic
acids. Starting from 3-O-benzyl ascorbic acid (12)10a and
unprotected 10-iododecan-1-ol (2a), 2,3-O,O-dialkyl
ether 13 was obtained in good yield using only a slightly
stronger base (K2CO3) as before (Scheme 3). 2-O-Alky-
lene ascorbic acid 14 was finally obtained by hydrogena-
tion over Pd/C in excellent yield.
HO
HO
HO
HO
O
O
O
NaHCO3
O
R2
R1
+
(CH2)n
DMSO
60 °C
HO
OH
O
R1
OH
11
(CH2)n
9
Scheme 2 3-O-Alkylenation attempts of unprotected ascorbic acid
Table 1 Specific 3-O-Alkylenation of Ascorbic Acid
Finally, we evaluated the antioxidative capacity of all
ascorbic acid derivatives. Their free radical scavenging
activity was determined by the corresponding IC50 using
the DPPH [2,2¢-di(4-tert-octylphenyl)-1-picrylhydrazyl]
test (Table 2).13
Entry Reactant R1
R2
n
Time Product
(h)
24
19
3
(yield, %)
11a (42)
11a (68)
11ba (80)
11ca (71)
11da (74)
11ea (78)
11fa (84)
11ga (79)
11ha (80)
11ia (42)
11ja (79)
11ka (65)
Table 2 Results of the DPPH Antioxidative Test
1
2
6
4
OBn
OBn
OH
OMs
10
10
10
12
14
16
9
Entry
Compound
a-tocopherol
Trolox®
vitamin C
11b
IC50 (mM)
900
Error
60
30
20
60
40
50
–
I
I
I
I
I
I
I
I
I
I
I
1
2
3
4
5
6
7
8
9
3
2a
2b
2c
80
4
OH
3
80
5
OH
3
730
6
2d
10a
10b
10c
10d
5
OH
3
11f
860
7
Me
3
11j
790
8
Me
11
15
17
16
10
3
a
11k
–
9
Me
3
a
13
–
–
10
11
12
Me
3
14
130
20
NPhth
CO2H
3
a No effect was detected.
8
3
a Reaction workup was performed under non-acidic conditions.
Synlett 2009, No. 2, 217–220 © Thieme Stuttgart · New York

LETTER
One-Step 3-O-Alkylenation of Unprotected Ascorbic Acid
219
HO
HO
HO
HO
HO
O
HO
O
O
O
a
b
O
O
BnO
OH
BnO
O
HO
O
OH
(CH2)10
OH
(CH2)10
12
13
14
Scheme 3 2-O-Alkylenation attempts of 3-O-benzyl ascorbic acid. Reagents and conditions: (a) 10-iododecanol, K2CO3, DMSO, 60 °C, 3 h
(77%); (b) H2, 5% Pd/C, EtOH, r.t., 2 h (96%).
Vitamin E (a-tocopherol), Trolox®, a water-soluble deriv- Acknowledgment
ative of the latter, and vitamin C were used as control
We thank the Luxembourg Ministry of Culture, Higher Education
and Research (grant to T.M.).
compounds (entries 1–3). Trolox® has generally a better
antioxidative activity than vitamin E due to its increased
solubility in ethanol. The IC50s of a given series, e.g. 11b–
References and Notes
e and 11f–i were not affected by the different chain
lengths (data not shown). 3-O-Alkanol ascorbic acid 11b
(entry 4) has a slightly better antioxidative activity than
the corresponding 3-O-alkyl ether 11f (entry 5) which
might again be explained by a better solubility of 11b in
ethanol owing to an additional alcohol function. Both
compounds do however not compete with vitamin C
which has a significantly lower IC50. These findings are in
contradiction with the results of Kato et al. who found
comparable reducing activities for all three types of com-
pounds.13 Derivative 11j showed reducing activity in the
range of compounds 11b and 11f (entry 6). Carboxylic
acid 11k as well as compound 13 completely lost their an-
tioxidative activity (entries 7 and 8). These results are in
accordance with those of Kato et al. who also observed
complete loss of reducing activity for ascorbic acid deriv-
atives bearing specific functional groups as well as for
2,3-O,O-disubstitued compounds.13 Finally, 2-O-alkylene
ether 14 exhibited an almost identical reductive ability as
vitamin C (entries 3 and 9) which is again in agreement
with the findings of Kato et al.13
(1) Jover, E.; Gonzalez de Aguilar, J.-L.; Luu, B.; Lutz-Becher,
B. Eur. J. Pharmacol. 2005, 516, 197.
(2) Girlanda-Junges, C.; Keyling-Bilger, F.; Schmitt, G.; Luu,
B. Tetrahedron 1998, 54, 7735.
(3) (a) Muller, T.; Grandbarbe, L.; Morga, E.; Heuschling, P.;
Bang, L. Bioorg. Med. Chem. Lett. 2004, 14, 6023.
(b) Muller, T.; Coowar, D.; Hanbali, M.; Heuscling, P.; Luu,
B. Tetrahedron 2006, 62, 12025.
(4) (a) Hanbali, M.; Vela-Ruiz, M.; Bagnard, D.; Luu, B.
Bioorg. Med. Chem. Lett. 2006, 16, 2637. (b) Hanbali, M.;
Bagnard, M.; Luu, B. Bioorg. Med. Chem. Lett. 2006, 16,
3917.
(5) Hauss, F.; Liu, J.; Michelucci, A.; Coowar, D.; Morga, E.;
Heuschling, P.; Luu, B. Bioorg. Med. Chem. Lett. 2007, 17,
4218.
(6) (a) Jacob, R. A.; Sotoudeh, G. Nutr. Clin. Care 2002, 5, 66.
(b) Mayne, S. T. J. Nutr. 2003, 133, 933.
(7) Nihro, Y.; Miyataka, H.; Sudo, T.; Matsumoto, H.; Satoh, T.
J. Med. Chem. 1991, 34, 2152.
(8) Kukarni, M. G.; Thopate, S. R. Tetrahedron 1996, 52, 1293.
(9) (a) Wimalasena, K.; Mahindaratne, M. P. D. J. Org. Chem.
1994, 59, 3427. (b) Olabisi, A. O.; Wimalasena, K. J. Org.
Chem. 2004, 69, 7026. (c) Kulkarni, M. G.; Kate, S. D.
Synth. Commun. 2004, 34, 2365.
(10) (a) Tahir, H.; Hindsgaul, O. J. Org. Chem. 2000, 65, 911.
(b) Vilaça, G.; Rubio, C.; Susperregui, J.; Latxague, L.;
Déléris, G. Tetrahedron 2002, 58, 9249.
(11) (a) Beifuss, U.; Kunz, O.; Aguado, G. P. Synlett 1999, 147.
(b) Beifuss, U.; Kunz, O.; Voss, G. Tetrahedron 2000, 56,
357.
(12) Kikugawa, Y.; Sakamoto, T.; Sato, S.; Asami, K.
Heterocycles 2005, 66, 579.
In conclusion we have developed a regioselective 3-O-
alkylation reaction of vitamin C with a series of different
unprotected iodoalkylene derivatives. These amphiphilic
products offer the possibility of further derivatization.
Moreover, this optimized procedure allows for the
straightforward synthesis of 2-O-alkylene ether deriva-
tives using the same readily available iodides. The deter-
mination of the antioxidative capacity of the different
compounds showed that there appears to be a difference
between 2-O- and 3-O-monoalkylene derivatives al-
though further 2-O-alkylene ethers will have to be pre-
pared and tested to confirm this result. Furthermore, the
functionalization of the attached chain can strongly influ-
ence on the reducing activity of given compounds.
(13) Kato, K.; Terao, S.; Shimamoto, N.; Hirata, M. J. Med.
Chem. 1988, 31, 793.
(14) Mulzer, J.; Angermann, A.; Schubert, B.; Seilz, A. J. Org.
Chem. 1986, 51, 5294.
(15) Kruizinga, W. H.; Kellogg, R. M. J. Am. Chem. Soc. 1981,
103, 5183.
(16) Typical Experimental Procedure; Preparation of (R)-5-
[(S)-1,2-Dihydroxyethyl)]-3-hydroxy-4-(10-hydroxy-
decyloxy)furan-2 (5H)-one (11b): Ascorbic acid (3.20 g,
18.20 mmol, 2.5 equiv) and NaHCO3 (1.83 g, 21.85 mmol,
3 equiv) were added to a solution of 10-iododecanol (2.07 g,
7.28 mmol, 1 equiv) in anhyd DMSO (18 mL) and the
resulting mixture was heated at 60 °C. After 3 h, DMSO was
evaporated under reduced pressure and the residue was
suspended in ethanol (12 mL) and brine (12 mL). EtOAc
(200 mL) was added to the suspension which was then dried
over MgSO4, filtered and evaporated under reduced
There are ongoing studies in our respective groups on the
synthesis of additional 2-O-alkylene ascorbic acids and on
biological screening reactions to determine potent candi-
dates for the treatment of neurological diseases.
Synlett 2009, No. 2, 217–220 © Thieme Stuttgart · New York

220
T. Muller et al.
LETTER
pressure. The crude was purified via silica gel column
chromatography (CH2Cl2–MeOH, 9:1) and was
3.64 (m, 2 H, CH2OH), 3.82 (m, 1 H, CHOH), 4.48 (m, 2 H,
H-1¢), 4.75 (d, J = 1.8 Hz, 1 H, CH). 13C NMR (75 MHz,
CD3OD): d = 25.2, 25.5 (C-2¢, C-3¢), 28.2–29.2 (C-4¢ to C-
8¢), 32.2 (C-9¢), 61.6 (CH2OH), 70.0 (CHOH), 70.5 (C-10¢),
72.4 (C-1¢), 75.2 (CH), 119.0 (=COH), 150.8 (=CO-alkyl),
171.8 (C=O).
recrystallized (EtOAc–hexane) to give a white solid (80%).
Rf = 0.30 (EtOAc); mp 71–73 °C. IR (CHCl3): 3396 (OH),
1748 (CyO), 1693 (CyO) cm–1. 1H NMR (300 MHz,
CD3OD): d = 1.31 (br s, 12 H, H-3¢ to H-8¢), 1.51 (m, 2 H,
H-9¢), 1.72 (m, 2 H, H-2¢), 3.52 (t, J = 6.6 Hz, 2 H, H-10¢),
Synlett 2009, No. 2, 217–220 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.