An efficient route for the synthesis of glycosyl phosphinic acids

Article abstract of DOI:10.1021/jo034475v

An efficient method for the synthesis of glycosyl phosphinic acids (21) from the corresponding C-phosphonates is described. The route developed involves three steps: reduction of the glycosyl C-phosphonate to a primary phosphine, reaction of this product

Full text of DOI:10.1021/jo034475v

An Efficien t Rou te for th e Syn th esis of Glycosyl P h osp h in ic Acid s
Charla A. Centrone and Todd L. Lowary*
Department of Chemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210
lowary.2@osu.edu
Received April 14, 2003
An efficient method for the synthesis of glycosyl phosphinic acids (21) from the corresponding
C-phosphonates is described. The route developed involves three steps: reduction of the glycosyl
C-phosphonate to a primary phosphine, reaction of this product with an alkylating agent to afford
a secondary phosphine, and finally oxidation to the phosphinic acid. Deprotection provides
compounds suitable for testing as glycosyl phosphate analogues. Although the focus of this report
is the synthesis of analogues of arabinofuranosyl-containing phosphate esters, the method should
be readily applicable to other systems, carbohydrate or otherwise.
In tr od u ction
common, although there are routes available for their
preparation.5c,6
Glycosyl phosphates and the corresponding phosphate
diesters have important biological roles, serving for
example as key intermediates in the biosynthesis of
oligosaccharides.1 Representative examples of these com-
pounds are provided in Chart 1 and include glucose-1-
phosphate (1) and dolichol phosphate mannose (2). The
biological importance of these and other such compounds
has, for many years, stimulated interest both in the
synthesis of nonhydrolyzable glycosyl phosphate ana-
logues and in their subsequent evaluation as inhibitors
of the enzymes involved in glycosyl phosphate metab-
olism.2-5
Significant work has been done on the synthesis of
glycosyl C-phosphonates (e.g., 3). Routes commonly used
for the preparation of such compounds include (1) a
Michaelis-Arbuzov reaction between an alkyl halide and
a trialkyl phosphite4 and (2) a Horner-Emmons olefi-
nation of a protected reducing sugar with subsequent
cyclization of the resulting vinyl phosphonate.5 These
methods are well developed, and glycosyl C-phosphonates
are, in general, straightforwardly prepared. Syntheses
of glycosyl C-phosphonate diesters (e.g., 4) are less
In contrast, there have been few reported syntheses
of glycosyl phosphinic acids, analogues of glycosyl phos-
phates in which two of the P-O bonds of the phosphate
moiety have been replaced with P-C bonds (e.g., 5). To
date, the only reported phosphinic acid analogues of
naturally occurring carbohydrate phosphates are 67 and
7,8 as well as nucleotides derived from 7. In the assembly
of 6, the carbon-phosphorus bonds were formed by the
reaction of alkene 8 with hypophosphorus acid derivative
9, followed by the coupling of this product with 10.7 The
key step in the synthesis of 7 was the coupling of the
phosphorus anion derived from 11 with 12; subsequent
functional group transformations afforded the target.8
Related to these investigations are a series of papers that
describe the synthesis of monosaccharide analogues in
which the ring oxygen has been replaced with phospho-
rus.9 Included among these “phosphosugars” are phos-
phinic acid derivatives such as 13.
We previously described the synthesis of C-phospho-
nate analogues (14-19, Chart 2) of decaprenolphospho-
arabinose (DPA, 20).10 DPA is the donor substrate used
by the arabinosyltransferases that synthesize the ara-
binan portions of two polysaccharides that are the major
structural components of the cell wall complex in myco-
bacteria, including the human pathogen Mycobacterium
tuberculosis.11 Nonhydrolyzable analogues of 20 are of
interest not only as biochemical tools but also as lead
compounds for new antituberculosis agents. Such species
(1) (a) Burda, P.; Aebi, M. Biochim. Biophys. Acta 1999, 1426, 239.
(b) Freeze, H. Metabolism of Sugars and Sugar Nucleotides. In
Carbohydrates in Chemistry and Biology; Ernst, B., Hart, G. W., Sinay¨,
P., Eds.; Wiley-VCH: Weinheim, 2000.
(2) (a) Street, I. P.; Withers, S. G. Biochem. J . 1995, 308, 1017. (b)
Martin, J . L.; J ohnson, L. N.; Withers, S. G. Biochemistry 1990, 29,
10745. (c) Maryanoff, B. E.; Nortey, S. O.; Inners, R. R.; Campbell, S.
A.; Reitz, A. B.; Liotta, D. Carbohydr. Res. 1987, 171, 259. (d) Witte,
J . F.; McClard, R. W. Bioorg. Chem. 1996, 24, 29. (e) McClard, R. W.;
Witte, J . F. Bioorg. Med. Chem. Lett. 1994, 4, 1537.
(3) (a) Nicotra, F. Synthesis of Glycosyl Phosphate Mimics. In
Carbohydrate Mimics; Chapleur, Y., Ed.; Wiley-VCH: Weinheim, 1998.
(b) Cipolla, L.; La Ferla, B.; Nicotra, F. Carbohydr. Polym. 1998, 37,
291.
(4) For examples see: (a) Casero, F.; Cipolla, L.; Lay, L.; Nicotra,
F.; Panza, L.; Russo, G. J . Org. Chem. 1996, 61, 3428. (b) Nicotra, F.;
Ronchetti, F.; Russo, G. J . Org. Chem. 1982, 47, 4459. (c) Nicotra, F.;
Panza, L.; Russo, G.; Senaldi, A.; Burlini, N.; Tortora, P. J . Chem. Soc.,
Chem. Commun. 1990, 1396.
(6) (a) Brooks, G.; Edwards, P. D.; Hatto, J . D. I.; Smale, T. C.
Southgate, R. Tetrahedron 1995, 29, 7999. (b) Qiao, L.; Vederas, J . C.
J . Org. Chem. 1993, 58, 3480. (c) Borodkin, V. S.; Milne, F. C.;
Ferguson, M. A. J .; Nikolaev, A. V. Tetrahedron Lett. 2002, 3, 7821.
(7) Dubert, O.; Gautier, A.; Condamine, E.; Piettre, S. R. Org. Lett.
2002, 4, 359.
(8) Collingwood, S. P.; Baxter, A. D. Synlett 1995, 703.
(9) (a) Hanaya, T.; Yamamoto, H. Helv. Chim Acta 2002, 85, 2608.
(b) Yamamoto, H.; Hanaya, T. Sugar Analogues Containing Carbon-
Phosphorus Bonds; Atta-ur-Rahman, Ed.; Studies in Natural Products
Chemistry, Vol. 6; Elsevier: Amsterdam 1990; p 351 and references
therein.
(5) For examples, see: (a) McClard, R. W.; Witte, J . F. Bioorg. Chem.
1990, 18, 165. (b) McClard, R. W.; Tsimikas, S.; Schriver, K. E. Arch.
Biochem. Biophys. 1986, 245, 282. (c) Borodkin, V. S.; Ferguson, M.
A. J .; Nikolaev, A. V. Tetrahedron Lett. 2001, 42, 5305.
(10) Centrone, C. A.; Lowary, T. L. J . Org. Chem. 2002, 67, 8862.
(11) Lowary, T. L. J . Carbohydr. Chem. 2002, 21, 691.
10.1021/jo034475v CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/25/2003
J . Org. Chem. 2003, 68, 8115-8119
8115

Centrone and Lowary
CHART 1
CHART 2
designing routes for the synthesis of these compounds,
we found previously reported methods for the synthesis
of carbohydrate-containing phosphinic acids unattractive,
given the number of steps involved. We also considered
the routes that have been used to synthesize phosphinic
acid containing peptides (e.g., 22), which have been
studied as protease inhibitors.13 The approach most
commonly used for the synthesis of the phosphinic acid
moiety of these peptides is similar to the one used for
the preparation of 6 and involves the radical addition of
hypophosphorus acid derivatives to an alkene. We envi-
sioned a more efficient route to the glycosyl phosphinic
acids of interest to us. The retrosynthetic analysis of this
approach is shown in Figure 1. We postulated that the
targets could be obtained by oxidation of secondary
phosphines of the general type 27, which were to be
prepared by monoalkylation of primary phosphine 26. We
hoped to access 26 via reduction of the easily prepared
C-phosphonate 25.
are expected to arrest cell wall arabinan biosynthesis by
competing with 20 for the active site of mycobacterial
arabinosyltransferases. These enzymes have been vali-
dated as suitable targets for drug action in that one of
the drugs used to treat tuberculosis, ethambutol, is an
arabinosyltransferase inhibitor.12 In our earlier investi-
gation, we demonstrated that one of these C-phosphonate
compounds, 19, was active in vitro against M. tubercu-
losis, with an MIC value of 3.13 µg/mL.10 This compound
is currently being tested to determine its potency in vivo.
On the basis of these results, we endeavored to
synthesize additional DPA analogues and turned our
attention to phosphinic acids of the general type 21. In
Resu lts a n d Discu ssion
With this route in mind, C-phosphonate 25 was syn-
thesized in three steps from the commercially available
2,3,5-tri-O-benzyl arabinofuranose, 23, as shown in
Scheme 1. Wittig olefination of 23 followed by cyclization
of the resulting alkene with iodine14 yielded 24, which
was then heated in refluxing triethyl phosphite to afford
(12) (a) Mikusova´, K.; Slayden, R. A.; Besra, G. S.; Brennan, P. J .
Antimicrob. Agents Chemother. 1995, 39, 2484. (b) Deng, L.; Mikusova´,
K.; Robuck, K. G.; Scherman, M.; Brennan, P. J .; McNeil, M. R.
Antimicrob. Agents Chemother. 1995, 39, 694. (c) Khoo, K.-H.; Douglas,
E.; Azadi, P.; Inamine, J . M.; Besra, G. S.; Mikusova´, K.; Brennan, P.
J .; Chatterjee, D. J . Biol. Chem. 1996, 271, 28682.
(13) Representative examples: (a) Malachowski, W. P.; Coward, J .
K. J . Org. Chem. 1994, 59, 7625. (b) Buchardt, J .; Ferreras, M.; Krog-
J ensen, C.; Delaisse, J .-M.; Foged, N. T.; Meldal, M. Chem. Eur. J .
1999, 5, 2877. (c) Ellsworth, B. A.; Tom, N. J .; Bartlett, P. A. Chem.
Biol. 1996, 3, 37.
8116 J . Org. Chem., Vol. 68, No. 21, 2003

Synthesis of Glycosyl Phosphinic Acids
F IGURE 1. Retrosynthetic analysis of glycosyl phosphinic acid derivatives 21.
SCHEME 1. Syn th esis of 21a -21ea
CHART 3
The reduction of the phosphonate moiety in 25 was
achieved upon treatment with lithium aluminum hydride
in ether at room temperature.16 The product phosphine,
26, was prone to air oxidation, but it was nevertheless
possible to purify the product by quickly passing the
crude reaction mixture though a column of silica gel.
Following this purification step, 26 was obtained in 69%
yield. In addition to a signal for the desired compound,
the mass spectrum indicated the formation of the corre-
sponding oxidation byproduct, 29 (Chart 3), which is
1
presumably produced in the ion source.17 In the H NMR
spectrum of 26 (recorded in CDCl3) the phosphine
appeared as a broad singlet between 2.60 and 3.15 ppm
and the 1H-decoupled 31P NMR spectrum showed a single
peak at -150.6 ppm (relative to external phosphoric acid
at 0.0 ppm).
We initially explored the alkylation of phosphine 26
using n-butyllithium and an alkyl halide in THF.18
However, under these conditions no alkylation of 26 was
observed; similar results were obtained when t-BuLi was
used as the base. We then turned our attention to a
previous report that described the monoalkylation of
primary phosphines using a phosphazene (Schweshinger)
base.19 These alkylation reactions, which were carried out
by treating a solution of phosphine 26 in diethyl ether
with an alkylating agent and the P4-Schweshinger base,
a
(a) Ph3PCH3Br, n-BuLi, THF, rt, 77%. (b) I2, Et2O, NaHCO3,
rt, 95%. (c) (EtO)3P, reflux, 85%. (d) LiAlH4, Et2O, rt, 69%. (e) 30,
R-I or 32, diethyl ether, rt. (f) I2, pyridine, H2O, rt, 26-64% from
26. (g) H2, Pd/C, CH3OH, rt 70-88%.
25. As was previously reported15 for the synthesis of the
C-2 epimer of 25, when this Michaelis-Arbuzov reaction
was attempted in refluxing trimethyl phosphite, iodide
24 was recovered unreacted.
(16) Prabhu, K. R.; Pillarsetty, N.; Gali, H.; Katti, K. V. J . Am.
Chem. Soc. 2000, 122, 1554.
(17) A signal at m/z ) 489.1716 was observed, which corresponds
to the Na+ adduct of 29 (M + Na+ ) 489.1806). A resonance arising
from 29 was not present in the 31P NMR spectrum of 26.
(18) Yan, Y.-Y.; RajanBabu, T. V. Org. Lett. 2000, 2, 4137.
(19) Uhlig, F.; Puschner, B.; Herrmann, E.; Zobel, B.; Bernhardt,
H.; Uhlig, W. Phosphorus Sulfur Silicon Relat. Elem. 1993, 81, 155.
(14) Freeman, F.; Robarge, K. D. Carbohydr. Res. 1987, 171, 1.
(15) McGurk, P.; Chang, G. X.; Lowary, T. L.; McNeil, M.; Field, R.
A. Tetrahedron Lett. 2001, 42, 2231.
J . Org. Chem, Vol. 68, No. 21, 2003 8117

Centrone and Lowary
30 (Chart 3), were successful. For the synthesis of
secondary phosphines 27a -27d , commercially available
alkyl iodides were used as the electrophile. The alkyl
iodide required for the preparation of 27e is not available,
and we therefore used tosylate 32 (Chart 3), which was
synthesized from 1-eicosanol. Like the primary phosphine
26, the secondary phosphines produced in these reactions
were also air-sensitive; however, their purification could
be achieved by rapid chromatography on silica gel.20
Given their air sensitivity, these compounds were not
further purified or characterized but were instead im-
mediately dissolved in a solution of pyridine and water
and then oxidized upon treatment with iodine.21 After
stirring for 3 days at room temperature, the reaction
mixtures were purified by chromatography. Following
chromatography, the products were typically contami-
nated with traces of pyridine, which were removed upon
stirring in methanol with Amberlite IR-120 (H+) resin.
The yields of the oxidized products 28a -28e from
secondary phosphine 26 ranged from 26% to 64%.
In an effort to improve the overall yields and efficiency
of the process, we investigated oxidation of the crude
secondary phosphines 27 immediately following the
alkylation reaction. Unfortunately, although the oxida-
tion proceeded without difficulty, we were unable to
separate the phosphinic acid products 28 from other
byproducts produced during these reactions. It appears,
therefore, that despite the air-sensitivity of phosphines
27a -27e, a rapid purification of these products prior to
treatment with iodine provides the best overall results.
A disadvantage of the iodine/pyridine/water oxidation
system employed here is that the method requires
extended reaction times (3 days). We therefore explored
the use of H2O2 or m-CPBA to carry out this reaction.
With both of these reagents, however, the product
obtained was only partially oxidized and was resistant
to further oxidation. For example, reaction of 27a with
either H2O2 or m-CPBA yielded 31. The structure of this
product was confirmed by mass spectrometry and also
hypothesized that the free phosphinic acids may be
aggregating in both CD3OD and CDCl3, thus leading to
the poor quality spectra. Given the relatively well-
resolved spectra obtained for the crude pyridinium salts
(above) we investigated the use of deuterated pyridine
as the NMR solvent. Dissolution of 28a -28e in pyridine-
d5 should lead to the immediate formation of the corre-
sponding pyridinium salts. We were pleased to see that
when these NMR spectra were recorded in pyridine-d5,
the resonances in the 1H and 13C NMR spectra sharpened
substantially. In addition, the 1H-decoupled 31P NMR
spectrum of each product showed a singlet at approxi-
mately 50 ppm (relative to external phosphoric acid at
0.0 ppm), which is consistent with the compounds being
phosphinic acids.22 Similarly, in the 13C NMR spectra,
the presence of the two C-P bonds was established by
two resonances between 30 and 32 ppm, which appeared
1
as doublets with J C,P magnitudes of approximately
90 Hz.
Having developed a successful method for the synthesis
of the protected phosphinic acids 28a -28e, we depro-
tected them without difficulty. Hydrogenation (H2, Pd/
C, CH3OH) provided the targets 21a -21e in yields of 70-
88%.23 As was true with the tribenzylated phosphinic
acids, the best NMR data for 21a -21e was obtained
when pyridine-d5 was used as the NMR solvent. In all
cases, the 1H-decoupled 31P NMR spectra showed the
characteristic phosphinic acid singlet around 50 ppm, and
in the 13C NMR spectra two doublets (1J C,P ≈ 90 Hz)
arising from the carbons directly bonded to the phospho-
rus atoms were present between 30 and 32 ppm.
In summary, we have developed an efficient method
for the preparation of glycosyl phosphinic acid derivatives
via a route that involves monalkylation of a primary
glycosyl phosphine and subsequent oxidation of the
resulting product. Although the focus of this paper has
been on the synthesis of analogues of arabinofuranosyl-
containing glycosyl phosphate esters, this method should
be straightforwardly applied to other carbohydrate sys-
tems. In many cases, the required C-phosphonate start-
ing materials (e.g., 33, Chart 3) have already been
reported.3,4 The extension of this method to the synthesis
of phosphinic acid analogues of non-carbohydrate phos-
phate esters should also be readily achieved. Testing of
21a -21e as inhibitors of mycobacterial growth is in
progress.
1
by H-coupled 31P NMR spectroscopy, which showed the
presence of two diastereomeric oxidation products (δP )
1
34.33 and 33.92 ppm, CDCl3) with J P,H magnitudes of
462 and 476 Hz, as would be expected for compounds of
this type.22
Our initial attempts to characterize 28a -28e by NMR
spectroscopy were complicated by the fact that in either
1
CDCl3 or CD3OD the resonances in the H and 13C NMR
Exp er im en ta l Section
spectra were significantly broadened. Further complicat-
ing the issue was that the spectra of the crude reaction
mixtures following evaporation of the reaction solvent
(the pyridinium salts of 28a -28e) were well-resolved and
showed the products to be of reasonable purity prior to
purification. One explanation for these observations is
that the compounds degrade upon silica gel chromatog-
raphy, but we viewed this as unlikely. Instead, we
Gen er a l. General experimental procedures and analytical
data for new compounds (1H NMR, 13C NMR, 13P NMR, HRMS,
[R]D) are provided in Supporting Information.
1-(Octyl)-2,5-a n h yd r oglu cityl P h osp h in ic Acid (21a ).
Phosphinic acid 28a (90 mg, 0.15 mmol) was dissolved in CH3-
OH (5 mL) and 10% Pd/C (30 mg) was added. The reaction
mixture was stirred under H2 overnight at atmospheric
pressure before being filtered and concentrated. The resulting
oil was purified by chromatography on Iatrobeads (1% pyridine
in CH2Cl2 f 3:1 CH2Cl2/CH3OH) providing a compound that
was then redissolved in water and lyophilized to afford 21a
(40 mg, 81%) as an off-white solid.
(20) We attempted to obtain mass spectrometric data for these
phosphines; however, the only signal detected was for the correspond-
ing oxidation product, the phosphinous acid derivative, e.g., 31 (Chart
3), which we propose is produced in the ion source. For example, the
mass spectrum recorded with 27a (M + Na+ ) 585.3104) showed only
a signal at m/z ) 601.3024, which arises from the sodium adduct of
31 (M + Na+ ) 601.3059).
(21) Lindh, I.; Stawinski, J . J . Org. Chem. 1989, 54, 1338.
(22) Gorenstein, D. G. Prog. Nucl. Magn. Reson. Spectrosc. 1983,
16, 1.
(23) Prior to the hydrogenation step, we converted the pyridinium
salts of 28a -28e, which were generated upon dissolving the protected
phosphinic acids in pyridine-d5, back to the free phosphinic acids. This
was achieved by stirring these pyridinium salts with Amberlite IR-
120 (H+) resin in methanol overnight at room temperature.
8118 J . Org. Chem., Vol. 68, No. 21, 2003

Synthesis of Glycosyl Phosphinic Acids
1-(Decyl)-2,5-a n h yd r oglu cityl P h osp h in ic Acid (21b).
Phosphinic acid 28b (140 mg, 0.23 mmol) was hydrogenated
in CH3OH (5 mL) using 10% Pd/C (50 mg) as described for
the preparation of 21a . Purification of the product was done
as described for 21a to provide 21b (65 mg, 82%) as an off-
white solid.
1-(Dodecyl)-2,5-an h ydr oglu cityl P h osph in ic Acid (21c)-
. Phosphinic acid 28c (98 mg, 0.15 mmol) was hydrogenated
in CH3OH (5 mL) using 10% Pd/C (40 mg) as described for
the preparation of 21a . Purification of the product was done
as described for 21a to provide 21c (50 mg, 88%) as an off-
white solid.
1-(H exa d ecyl)-2,5-a n h yd r oglu cit yl P h osp h in ic Acid
(21d ). Phosphinic acid 28d (189 mg, 0.27 mmol) was hydro-
genated in CH3OH (5 mL) using 10% Pd/C (70 mg) as described
for the preparation of 21a . The compound was purified by
crystallization from methanol to provide 21d (81 mg, 70%) as
an off-white solid.
1-(E icosa n yl)-2,5-a n h yd r oglu cit yl P h osp h in ic Acid
(21e). Phosphinic acid 28e (29 mg, 0.038 mmol) was hydro-
genated in CH3OH (5 mL) using 10% Pd/C (15 mg) as described
for the preparation of 21a . The compound was purified by
crystallization from methanol to provide 21e (15 mg, 80%) as
an off-white solid.
Dieth yl 3,4,6-tr i-O-ben zyl-2,5-a n h yd r oglu cityl P h os-
p h on a te (25). Iodide 2414 (3.85 g, 7.08 mmol) was dissolved
in triethyl phosphite (20 mL), and the reaction mixture was
heated at reflux (156 °C) for 12 h. The excess triethyl phosphite
was evaporated by heating under high vacuum, and the
resulting oil was purified by chromatography (hexane/EtOAc
4:1 f EtOAc/hexane 2:1) to afford 25 (3.34 g, 85%) as a
colorless oil.
3,4,6-Tr i-O-ben zyl-2,5-a n h yd r oglu cityl P h osp h in e (26).
A solution of phosphonate 25 (235 mg, 0.42 mmol) in anhy-
drous Et2O (5 mL) was added dropwise to a mixture of LiAlH4
(40 mg, 1.10 mmol) in anhydrous Et2O (5 mL) stirring at room
temperature. After 20 min, EtOAc (1 mL) was added to the
mixture, followed a few minutes later by H2O (0.2 mL). After
all gas evolution had subsided, the mixture was filtered though
Celite and concentrated to give a clear oil. Purification of the
product by rapid passage of the crude reaction mixture through
a column of silica gel (6:1 hexane/EtOAc) afforded 26 (132 mg,
69%) as a colorless oil.
1-(Octyl)-3,4,6-tr i-O-ben zyl-2,5-a n h yd r oglu cityl P h os-
p h in ic Acid (28a ). Phosphine 26 (245 mg, 0.54 mmol) was
dissolved in Et2O (10 mL) and stirred at room temperature
before 30 (600 µL of a 1 M solution in hexane, 0.60 mmol) was
added, followed by 1-iodooctane (100 µL, 0.60 mmol). After 30
min, the mixture was neutralized with AcOH, and the salts
that precipitated were removed by filtration through a cotton
plug, which was rinsed with Et2O. The solvent was evaporated,
and the residue was purified by rapid elution though a short
column of silica gel (6:1 hexane/EtOAc) to afford 27a as a
colorless oil: Rf 0.36 (6:1 hexane/EtOAc). This oil was im-
mediately dissolved in pyridine/H2O (98:2, 5 mL), I2 (189 mg,
0.751 mmol) added, and the reaction mixture stirred for 3 days
at room temperature. The mixture was then diluted with CH2-
Cl2, washed with an aqueous 5% NaHSO3, solution, and dried
(Na2SO4). The solvent was then evaporated, and the residual
oil was purified by chromatography (12:1 CH2Cl2/CH3OH). The
product following chromatography was contaminated with
traces of pyridine, which were removed by redissolving the
material in CH3OH (5 mL) and then stirring the solution with
Amberlite 120 (H+) resin (150 mg) overnight. Filtration of the
resin and evaporation of the solvent afforded 28a (111 mg, 34%
from 26) as a clear oil.
EtOAc as the eluant. This product was immediately oxidized
as described for the preparation of 28a using I2 (187 mg, 0.74
mmol) in pyridine/H2O (98:2, 6 mL). Purification of the oxidized
product was achieved via chromatography (12:1 CH2Cl2/
CH3OH) and subsequent treatment of the resulting residue
with ion-exchange resin as described for 28a . Phosphinic acid
28b was isolated (158 mg, 57% from 26) as a clear oil.
1-(Dodecyl)-3,4,6-tr i-O-ben zyl-2,5-an h ydr oglu cityl P h os-
p h in ic Acid (28c). Alkylation of phosphine 26 (208 mg, 0.46
mmol) with 1-iodododecane (125 µL, 0.51 mmol) was achieved
as described for the preparation of 27a using 30 (490 µL of a
1 M solution in hexane, 0.49 mmol) in Et2O (10 mL). Phosphine
27c (Rf 0.30, 9:1 hexane/EtOAc) was obtained following the
purification process outlined above for 27a using 9:1 hexane/
EtOAc as the eluant. This product was immediately oxidized
as described for the preparation of 28a using I2 (143 mg, 0.57
mmol) in pyridine/H2O (98:2, 6 mL). Purification of the oxidized
product was achieved via chromatography (12:1 CH2Cl2/CH3-
OH) and subsequent treatment of the resulting residue with
ion-exchange resin as described for 28a . Phosphinic acid 28c
was isolated (118 mg, 40% from 26) as a clear oil.
1-(H exa d ecyl)-3,4,6-t r i-O-b en zyl-2,5-a n h yd r oglu cit yl
P h osp h in ic Acid (28d ). Alkylation of phosphine 26 (189 mg,
0.42 mmol) with 1-iodohexadecane (155 mg, 0.44 mmol) was
achieved as described for the preparation of 27a using 30 (440
µL of a 1 M solution in hexane, 0.44 mmol) in Et2O (10 mL).
Phosphine 27d (Rf 0.33, 8:1 hexane/EtOAc was obtained fol-
lowing the purification process outlined above for 27a using
9:1 hexane/EtOAc as the eluant. This product was immediately
oxidized as described for the preparation of 28a using I2 (216
mg, 0.86 mmol) in pyridine/H2O (98:2, 6 mL). Purification of
the oxidized product was achieved via chromatography (12:1
CH2Cl2/CH3OH) and subsequent treatment of the resulting
residue with ion-exchange resin as described for 28a . Phos-
phinic acid 28d was isolated (189 mg, 64% from 26) as a clear
oil.
1-(E icosa n yl)-3,4,6-t r i-O-b en zyl-2,5-a n h yd r oglu cit yl
P h osp h in ic Acid (28e). Alkylation of phosphine 26 (79 mg,
0.18 mmol) with 1-p-toluenesulfonyloxy-eicosane (32, 130 mg,
0.29 mmol) was achieved as described for the preparation of
27a using 30 (180 µL of a 1 M solution in hexane, 0.18 mmol)
in Et2O (5 mL). Phosphine 27e (Rf 0.26, 12:1 hexane/EtOAc)
was obtained following the purification process outlined above
for 27a using 12:1 hexane/EtOAc as the eluant. This product
was immediately oxidized as described for the preparation of
28a using I2 (66 mg, 0.26 mmol) in pyridine/H2O (98:2, 4 mL).
Purification of the oxidized product was achieved via chroma-
tography (12:1, CH2Cl2/CH3OH,) and subsequent treatment of
the resulting resin with ion-exchange residue as described for
28a . Phosphinic acid 28e was isolated (35 mg, 26% from 26)
as a clear oil.
1-p-Tolu en esu lfon yloxy-eicosa n e (32). 1-Eicosanol (500
mg, 1.67 mmol) was dissolved in THF (15 mL) and stirred at
room temperature. n-Butyllithium (1.1 mL of a 1.6 M solution
in hexanes, 1.84 mmol) was added, followed by p-toluenesulfo-
nyl chloride (639 mg, 3.35 mmol) in THF (15 mL). After 3 h,
the mixture was diluted with EtOAc (20 mL), washed with
H2O (25 mL) and brine (25 mL), and then dried over Na2SO4.
The solvent was evaporated and the solid was purified by
chromatography (hexane/EtOAc 10:1) to afford 32 (501 mg,
66%) as a white solid.
Ackn owledgm en t. The National Institutes of Health
(AI44045-01) supported this work. C.A.C. is a recipient
of a GAANN fellowship from the U.S. Department of
Education. We thank T. V. RajanBabu for helpful
discussions.
1-(Decyl)-3,4,6-tr i-O-ben zyl-2,5-a n h yd r oglu cityl P h os-
p h in ic Acid (28b). Alkylation of phosphine 26 (203 mg, 0.45
mmol) with 1-iododecane (100 µL, 0.50 mmol) was achieved
as described for the preparation of 27a using 30 (500 µL of a
1 M solution in hexane, 0.50 mmol) in Et2O (10 mL). Phosphine
27b (Rf 0.26, 9:1 hexane/EtOAc) was obtained following the
purification process outlined above for 27a using 9:1 hexane/
Su p p or tin g In for m a tion Ava ila ble: Analytical data and
1H, 13C, and 31P NMR spectra for new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O034475V
J . Org. Chem, Vol. 68, No. 21, 2003 8119