(α-L-rhamnopyranosyl)methylphosphonic acids: Experimental evidence of the analogy with α-L-rhamnopyranosyl phosphate

Article abstract of DOI:10.1002/ejoc.200500440

(α-L-Rhamnopyranosyl)methylphosphonic acid and methyl (2-O-methyl-α-L-rhamnopyranosyl)methylphosphonate have been synthesized starting from allyl 3,4-di-O-benzyl-2-O-p-methoxybenzyl-α-L- rhamnopyranoside. The complete NMR spectroscopic characterization of

Full text of DOI:10.1002/ejoc.200500440

FULL PAPER
(α-L
-Rhamnopyranosyl)methylphosphonic Acids: Experimental Evidence of the
Analogy with α- -Rhamnopyranosyl Phosphate
L
Silvia Ronchi,[a] Federica Compostella,*[a] Luigi Lay,[b] Fiamma Ronchetti,[a] and
Lucio Toma[c]
Keywords: C-Glycosides / Conformational analysis / Phosphonate analogues / Capsular polysaccharides
(α-
(2-O-methyl-α-
been synthesized starting from allyl 3,4-di-O-benzyl-2-O-p-
methoxybenzyl-α- -rhamnopyranoside. The complete NMR
L
-Rhamnopyranosyl)methylphosphonic acid and methyl
coupling constants indicates a marked preference for the 1C4
conformation for both compounds, in analogy with the natu-
ral phosphate.
L-rhamnopyranosyl)methylphosphonate have
L
spectroscopic characterization of these two compounds and
the comparison of the theoretical and experimental vicinal
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
formational behavior. Natural α-l-rhamnosyl phosphate 1
in the 1C4 conformation (Figure 1) has substituents at posi-
tions 1 and 2 in an axial orientation and those at positions
3, 4, and 5 in equatorial orientations. Even in the presence
of a bulky aglycon group the ring does not revert to this
normally preferred conformation;[4] the anomeric effect is
one of the factors that contribute to this preference. Its
phosphono analogue 2 does not utilise the anomeric effect
such that the 4C1 conformation can be more easily attained
and therefore produce a significant contribution to the
overall conformational population.
Introduction
The replacement of the anomeric oxygen atom with a
methylene group has been proposed as a strategy for ob-
taining stabilized analogues of glycosides[1] and, in particu-
lar, glycosyl phosphates.[2] This approach may be used in
various areas, e.g. in the synthesis of analogues which inter-
fere in the glycosylation process as competitive inhibitors of
glycosyl transferases or for the preparation of carbohydrate
analogues stable to hydrolysis. In this case, the search for
analogues is stimulated by the poor stability in water of
many types of polysaccharides, e.g. capsular polysaccha-
rides.[3] For example, the capsular polysaccharide of Strep-
tococcus pneumoniae type 19F consists of a trisaccharide
repeating
unit,
(Ǟ4)-β-d-ManpNAc-(1Ǟ4)-α-d-Glcp-
–
(1Ǟ2)-α-l-Rhap-(1-PO4 Ǟ), with a phosphodiester bridge
which connects a residue of rhamnose at the reducing end
of one unit to the nonreducing end of the following unit. A
phosphate group at the anomeric position of rhamnose is a
reason for its limited stability in water, due to the high pro-
pensity of hydrolysis of this group at an acetalic position.
In such cases, the replacement of the anomeric oxygen atom
with a methylene group transforms an acetyl phosphate
into a more stable phosphonate. However, though a methyl-
ene group is isosteric with oxygen, its stereoelectronic prop-
erties are quite different, so that replacement of the oxygen
atom could, in principle, produce major changes in the con-
[a] Dipartimento di Chimica, Biochimica e Biotecnologie per la
Medicina, Università di Milano,
Via Saldini 50, 20133 Milano, Italy
Figure 1. Structures of the natural α-l-rhamnosyl-1-phosphate (1)
and its phosphono analogues 2, 3, and 4.
Fax: +39-02-50316036
E-mail: federica.compostella@unimi.it
[b] Dipartimento di Chimica Organica e Industriale, Università di
Milano,
Via Venezian 21, 20133 Milano, Italy
[c] Dipartimento di Chimica Organica, Università di Pavia,
Via Taramelli 10, 27100 Pavia, Italy
This hypothesis is confirmed by the spectroscopic data
observed for the perbenzylated phosphono analogue of the
α-l-rhamnosyl phosphate 3.[5] This protected derivative of
Eur. J. Org. Chem. 2005, 4459–4463
DOI: 10.1002/ejoc.200500440
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4459

S. Ronchi, F. Compostella, L. Lay, F. Ronchetti, L. Toma
FULL PAPER
2 shows that the 1H NMR vicinal coupling constants of the here the synthesis of compound 2 and its dimethyl deriva-
pyranose ring are incompatible with a ring geometry of the tive 4. Their spectroscopic characterization has been used
1C4 type as the sole populated conformation and are in to shed light on the conformational behavior of these α-l-
1
good agreement with an equilibrium between the C4 and rhamnosyl-1-phosphate mimics.
4C1 chair conformations. As a consequence, the similarity
between the analogue and the natural reference compound
ceases to exist and this becomes a problem for the phos-
phono analogues of α-l-rhamnose, if the preservation of
Results and Discussion
the conformational features is a requisite for their biological
The syntheses of the target compounds 2 and 4 required
activity. On the other hand, the J4,5 and J5,6 coupling con- the preparation of phosphonate 9 as a key intermediate and
stants observed for the ammonium salt of (α-l-rhamnopyr- this is outlined in Scheme 1. The 2-hydroxy group of the
anosyl)methylphosphonic acid (2)[5] are higher than those known allyl rhamnopyranoside 5[7] was protected as a p-
of the protected derivative, thus indicating that the equilib- methoxybenzyl ether, and the allyl group was then removed
rium is shifted towards the 1C4 conformation. However, the under standard conditions to afford rhamnopyranose 7 in
J2,3 coupling constant, which is diagnostic of the relative 96% yield. Reaction of 7 with the tetramethyl methylenedi-
orientation of the substituents at positions 2 and 3, was phosphonate anion[5] gave an α,β-unsaturated phosphonate,
not determined[5] such that an ultimate assessment on the which spontaneously underwent an intramolecular Michael
conformational preference of this compound cannot be de- reaction to yield compound 8 in 77% yield. Two isomers
termined.
were formed and the preference of the desired α-isomer was
1
A different scenario may emerge when the rhamnosidic detected by H NMR spectroscopy; the 2-H signal of the
residue is inserted in a polysaccharide chain, as in the Strep- major product appeared at δ = 4.42 ppm, analogous to the
tococcus pneumoniae 19F polysaccharide. In this case, its 2-H resonance of the corresponding perbenzylated com-
phosphono analogue can be mimicked by compound 4, pound 3.[5]
which presents two methyl groups at the points of chain
Since the diastereoisomers of 8 could not be separated
elongation; this simple compound 4 can be considered the by flash chromatography, the mixture was treated with
minimum substructure representing a rhamnosylphosphon- DDQ in order to selectively remove the p-methoxybenzyl
ate inserted in the polysaccharide chain. Theoretical calcu- group. The product was obtained in 81% yield, and the
lations on compound 4 predicted a conformational equilib- anomeric mixture was separated by flash chromatography.
1
4
rium between the C4 and C1 conformations with a 14% The desired dibenzyl α-phosphonate 9 was isolated and
1
contribution of the latter conformation to the overall pop- fully characterized. The H NMR spectrum shows vicinal
ulation.[6] However, due to a degree of uncertainty in the coupling constants (J2,3 = 5.0 Hz; J4,5 = J5,6 = 6.3 Hz) that
theoretical calculations of the modeling of highly polar are very similar to those observed for the corresponding
molecules, such as carbohydrates bearing ionizable groups, perbenzylated compound 3,[5] revealing the presence of a
an experimental validation of the behavior of the dimethyl conformational equilibrium between the 1C4 and 4C1 geom-
derivative 4 appears necessary. With this aim, we describe etries and also confirming the expected trend for the pro-
Scheme 1. Reagents and conditions: i) PMBCl, NaH, THF; ii) tBuOK, DMSO, then I2, py, THF/H2O; iii) [(MeO)2PO]2CH2, NaH, DME,
0 °C Ǟ room temp.; iv) DDQ, DCM/H2O; v) Me3SiI, CCl4, 0 °C; vi) MeI, Ag2O, DMF; vii) PhSH, TEA, THF; viii) H2, Pd(OH)2/C,
MeOH.
4460
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4459–4463

(α-l-Rhamnopyranosyl)methylphosphonic Acids
FULL PAPER
tected phosphono analogues of the natural phosphate 1. group is isosteric to the anomeric oxygen atom in stabilized
The (α-l-rhamnopyranosyl)methylphosphonic acid (2) was analogues of α-rhamnopyranosyl phosphates. This can be
obtained from 9 by removal of the protecting groups with seen as a new stimulus towards the syntheses of phosphono
Me3SiI.[5]
analogues of natural phosphates with high guarantees of
On the other hand, the dimethyl derivative 4 was ob- good mimic properties.
tained in three steps from 9. The 3-hydroxy group of 9 was
first methylated by treatment with MeI in the presence of
Ag2O[8] to give 10. Next, one of the ester methyl groups was
selectively removed,[9] which led to compound 11, which
was debenzylated by catalytic hydrogenation to yield com-
pound 4 in quantitative yield.
Experimental Section
General: 1H and 13C NMR spectra were recorded with a Bruker
1
AVANCE-500 spectrometer at a sample temperature of 298 K. H
and 13C NMR spectra for compound 2 were recorded with a sam-
ple concentration of 25 mm in D2O (99.9%D) at pD of 1.55. The
pD value corresponded to the reading on a pH meter equipped
with a glass electrode calibrated with pH 4.01 and 7.01 standard
buffers in H2O and is not corrected for the deuterium isotope ef-
Compounds 2 and 4 were fully characterized. In particu-
1
lar, all the resonances in their H NMR spectra were as-
signed, and the vicinal coupling constants of the ring hydro-
gen atoms, diagnostic of the conformational preference of
the compounds, were determined (Table 1). Our previous
theoretical conformational study of compound 4[6] indi-
cated a greater preference for the 1C4 conformation, though
1
fect. Acetonitrile was used as internal standard for H NMR (δ =
2.0 ppm) and 13C NMR (δ = 0.2 ppm) in D2O solutions, whereas
NMR spectra recorded in CDCl3 or CD3OD were calibrated by
using the signal of TMS as an internal reference. Spectra for com-
pound 4 were recorded with a sample concentration of 25 mm in
CD3OD (99.8%D). One bond 1H NMR–13C NMR correlation
maps were obtained from HMQC experiments. Optical rotations
were measured with a 241 Perkin–Elmer polarimeter at 20°C. Mass
spectra were recorded in the negative or positive mode on a
Thermo Quest Finningan LCQ™deca spectrometer using electro-
spray ionization as indicated. All reactions were monitored by TLC
on Silica Gel 60 F-254 plates (Merck), spots being developed with
5% sulfuric acid in methanol/water (1:1), or with a phosphomo-
lybdenate-based reagent. Flash column chromatography was per-
formed on Silica Gel 60 (230–400 mesh, Merck). Organic solutions
were dried with sodium sulfate. All evaporations were carried out
under reduced pressure at 40°C. Dry solvents were distilled prior
to use: THF and 1,2-dimethoxyethane (DME) were distilled from
sodium; dichloromethane (DCM) was distilled from calcium hy-
dride. DMF, DMSO, and methanol were dried using 4 Å molecular
sieves; triethylamine (TEA) was distilled from calcium hydride.
NaH was washed three times with hexane prior to use. DDQ: 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone.
4
a 14% contribution of the C1 conformation to the overall
population was predicted. The same computational ap-
proach was applied to compound 2, which gave similar re-
sults. For the most stable 1C4 and 4C1 conformations of
both compounds, the vicinal coupling constants have been
calculated by using the Haasnoot et al. equation[10] and are
reported in Table 1 and are compared with the experimental
data. The experimental values are close to those for the 1C4
conformation but very different from those for the 4C1 con-
formation, indicating a greater preference for the former ge-
ometry or even its exclusive presence.
Table 1. Comparison of the experimental 1H NMR vicinal coupling
constants (Hz) of compounds 2 and 4 with those calculated for
1
4
their C4 and C1 conformations.
J2,3
J3,4
J4,5
J5,6
2 (exp.)
2.0
1.3
9.1
1.7
1.1
7.9
3.2
3.1
3.0
3.5
3.7
4.8
9.5
9.0
2.8
9.1
9.5
2.8
9.5
9.2
1.3
9.1
9.2
1.5
1C4 (calcd.)
4C1 (calcd.)
4 (exp.)
Allyl 3,4-Di-O-benzyl-2-O-p-methoxybenzyl-α-L-rhamnopyranoside
(6): Allyl 3,4-di-O-benzyl-α-l-rhamnopyranoside (5)[7] (1.50 g,
3.90 mmol) was dissolved in dry THF (15 mL), and p-methoxyben-
zyl chloride (0.64 mL, 4.68 mmol) and NaH (60% dispersion in
mineral oil, 0.31 g, 7.80 mmol) were added, and the mixture was
stirred at room temperature for 4 d. It was then quenched with
MeOH, concentrated, and the residue dissolved in EtOAc (40 mL),
and washed with water (50 mL). The aqueous layer was further
extracted with EtOAc (2×40 mL). The combined organic layers
were dried and concentrated, and the product was isolated by flash
chromatography (hexane/EtOAc, 8:2) to afford 6 as a white solid
1C4 (calcd.)
4C1 (calcd.)
The data for phosphonate 4, closely related to those of
compound 2, suggests that this compound still maintains
the same conformational preference as α-l-rhamnopyrano-
syl phosphate 1.
1
(1.87 g, 95%). [α]2D0 = –34.2 (c = 1.0, CHCl3). H NMR (CDCl3):
Conclusions
δ = 1.33 (d, J5,6 = 6.0 Hz, 3 H, 6-H), 3.61 (t, J3,4 = J4,5 = 9.5 Hz,
1 H, 4-H), 3.70 (dq, J4,5 = 9.5, J5,6 = 6.0 Hz, 1 H, 5-H), 3.77–3.81
(m, 4 H, OCH3, 2-H), 3.87 (dd, J2,3 = 2.7, J3,4 = 9.5 Hz, 1 H, 3-
H), 3.90 (br. dd, Ja,aЈ = 13.0, Ja,b = 6.9 Hz, 1 H, OCHaHaЈ-
CHb=CH2), 4.11 (br. dd, Ja,aЈ = 13.0, JaЈ,b = 5.2 Hz, 1 H, OCHaHaЈ-
CHb=CH2), 4.56–4.70 (m, 5 H, CH2Ph), 4.76 (br. s, 1 H, 1-H), 4.95
(d, Jgem = 10.7 Hz, 1 H, CH2Ph), 5.14 (br. d, Jb,c = 10.5 Hz, 1 H,
CHb=CHcHd), 5.20 (br. d, Jb,d = 17.0 Hz, 1 H, CHb=CHcHd),
5.79–5.89 (m, 1 H, CHb=CHcHd), 6.81–6.85 (m, 2 H, arom.), 7.25–
7.36 (m, 12 H, arom.) ppm. 13C NMR (CDCl3): δ = 18.0, 55.3,
67.7, 68.1, 72.1, 72.4, 74.3, 75.4, 80.2, 80.6, 97.2, 113.8, 114.0,
α-l-Rhamnopyranosyl)methylphosphonic acid 2 and its
dimethyl derivative 4 have been synthesized from allyl 3,4-
di-O-benzyl-α-l-rhamnopyranoside 5, through the selec-
tively protected common phosphono intermediate 9. The
complete NMR spectroscopic characterization of the com-
pounds and the comparison of the experimental vicinal
coupling constants with the computed values indicated a
marked preference, or even the exclusive presence, of the
1C4 conformation, in analogy with the natural phosphate.
Our study provides experimental proof that the methylene 117.1, 127.5–130.4 (13C), 133.9, 138.6, 138.7, 159.3 ppm. ESI-MS
Eur. J. Org. Chem. 2005, 4459–4463
www.eurjoc.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4461

S. Ronchi, F. Compostella, L. Lay, F. Ronchetti, L. Toma
FULL PAPER
(positive-ion mode): m/z = 522.2 [M + NH4]+. C31H36O6 (504.61):
calcd. C 73.79, H 7.19; found C 74.01, H 7.35.
perature for 2 h, it was then diluted with CH2Cl2 (50 mL), washed
with saturated NaHCO3 (2×80 mL) and brine (1×80 mL), and
dried and concentrated. The crude was purified by flash
chromatography (toluene/acetone, 6:4 to 2:8), which first gave
0.30 g of the minor isomer and then 0.74 g of the major isomer 9
as a colorless oil (overall yield 82%). [α]2D0 = –7.5 (c = 1.0, CHCl3).
1H NMR (CDCl3): δ = 1.32 (d, J6,7 = 6.3 Hz, 3 H, 7-H), 2.06 (ddd,
J1a,2 = 8.5, J1a,1b = 15.5, J1a,P = 18.4 Hz, 1 H, 1a-H), 2.15 (ddd,
J1b,2 = 5.7, J1a,1b = 15.5, J1b,P = 19.5 Hz, 1 H, 1b-H), 2.62 (d, JOH,3
= 5.7 Hz, 1 H, OH), 3.46 (t, J5,6 = J5,4 = 6.3 Hz, 1 H, 5-H), 3.69–
3.82 (m, 8 H, 2 OCH3, 4-H, 6-H), 3.86 (ddd, J2,3 = 4.7, J3,4 = 3.5,
JOH,3 = 5.7 Hz, 1 H, 3-H), 4.23 (dddd, J1a,2 = 8.5, J1b,2 = 5.7, J2,3
= 4.7, J2,P = 10.5 Hz, 1 H, 2-H), 4.56–4.72 (m, 4 H, 2CH2Ph),
7.24–7.36 (m, 10 H, arom.) ppm. 13C NMR (CDCl3): δ = 17.3 (7-
C), 26.7 (d, JC,P = 142.7 Hz, 1-C), 52.4 (d, JC,P = 6.2 Hz, POCH3),
52.6 (d, JC,P = 6.2 Hz, POCH3), 69.1 (d, JC,P = 10.8 Hz, 3-C), 69.2
(2-C), 70.1 (6-C), 72.3 (OCH2Ph), 73.9 (OCH2Ph), 78.3 (4-C and
5-C), 127.8–128.6 (10 C arom.), 138.0 (2 C arom.) ppm. ESI-MS
(positive-ion mode): m/z (%) = 917.9 (100) [2 M + NH4]+, 467.9
(35) [M + NH4]+, 451.1 (65) [M + H]+. C23H31O7P (450.56): calcd.
C 61.33, H 6.94; found C 61.21, H 6.69.
3,4-Di-O-benzyl-2-O-p-methoxybenzyl-L-rhamnopyranose
(7):
tBuOK (4.0 g, 36.0 mmol) was slowly added to a solution of com-
pound 6 (1.80 g, 3.60 mmol) in dry DMSO (75 mL). After 3 h, the
solvent was evaporated under reduced pressure (high vacuum
pump), and the residue dissolved in CHCl3 (100 mL), washed with
H2O (1×80 mL), dried, and concentrated. The crude was dissolved
in THF/H2O (4:1, 90 mL), and pyridine (1.16 mL, 14.40 mmol) and
iodine (1.83 g, 7.20 mmol) were added, and the mixture was stirred
at room temperature for 3 h. The mixture was then diluted with
EtOAc (150 mL), washed with saturated Na2S2O3 (1×80 mL), 5%
HCl (1×80 mL), water (1×80 mL), dried, and the solvents evapo-
rated. The crude was purified by flash chromatography (hexane/
EtOAc, 75:25) to give compound 7 as a colorless oil (1.60 g, 96%).
[α]2D0 = –18.6 (c = 1.0, CHCl3). 1H NMR (CDCl3): δ = 1.32 (d, J5,6
= 6.2 Hz, 2.4 H, 6-H major), 1.34 (d, J5,6 = 6.2 Hz, 0.6 H, 6-H
minor), 2.79 (br. s, 1 H, OH), 3.32–3.39 (m, 0.2 H, 5-H min.), 3.54–
3.58 (m, 0.4 H, 3-H and 4-H min.), 3.62 (t, J3,4 = J4,5 = 9.4 Hz,
0.8 H, 4-H maj.), 3.76–3.85 (m, 4 H, OCH3, 2-H), 3.88–3.96 (m,
1.6 H, 3-H and 5-H maj.), 4.57–4.77 (m, 5 H, 1-H min. and
CH2Ph), 4.92–5.05 (m, 1.2 H, CH2Ph), 5.12 (br. s, 0.8 H, 1-H maj.),
6.81–6.91 (m, 2 H, arom.), 7.24–7.42 (m, 12 H, arom.) ppm. 13C
NMR (CDCl3, selected signals): δ = 17.9 (6-C min.), 18.1 (6-C
maj.), 55.3 (OCH3), 93.0 (1-C maj.), 93.3 (1-C min.) ppm. ESI-MS
(positive-ion mode): m/z = 482.1 [M + NH4]+. C28H32O6 (464.55):
calcd. C 72.39, H 6.94; found C 72.45, H 7.00.
Dimethyl 2,6-Anhydro-1,7-dideoxy-4,5-di-O-benzyl-3-O-methyl-
L-
glycero- -talo-heptit-1-yl Phosphonate (10): Compound 9 (0.17 g,
L
0.38 mmol) was dissolved under argon in dry DMF (10.5 mL).
CH3I (0.28 mL, 4.53 mmol) and Ag2O (0.70 mg, 3.02 mmol) were
added, and the mixture was stirred in the dark at room temperature
for 72 h, during which time a further aliquot of CH3I (0.52 mL,
8.31 mmol) and Ag2O (0.35 g, 1.51 mmol) were added. The mixture
was then diluted with Et2O, filtered through Celite, and washed
with H2O (1×80 mL). The aqueous layer was extracted with Et2O
(2×50 mL), and the combined organic layers were dried and con-
centrated. Product 10 was recovered after flash chromatography
(toluene/acetone, 1:1) as a colorless oil (0.12 g, 70%). [α]2D0 = –5.3
Dimethyl
oxybenzyl-
2,6-Anhydro-1,7-dideoxy-4,5-di-O-benzyl-3-O-p-meth-
-glycero- -talo-heptit-1-yl Phosphonate and Dimethyl
-gly-
solution of tet-
L
L
2,6-Anhydro-1,7-dideoxy-4,5-di-O-benzyl-3-O-p-methoxybenzyl-
L
cero- -galacto-heptit-1-yl Phosphonate (8):
L
A
ramethyl methylenediphosphonate (4.50 g, 19.38 mmol) in dry 1,2-
dimethoxyethane (6 mL) was slowly added to a suspension of NaH
(0.46 g, 19.38 mmol) in dry DME (15 mL) at 0°C. After 30 min, a
solution of compound 7 (1.50 g, 3.23 mmol) in dry DME (5 mL)
was added, and the resulting mixture was allowed to warm to room
temperature. After 2 days, it was quenched with MeOH (8 mL) and
concentrated. The residue was dissolved in EtOAc (200 mL), and
washed with saturated NH4Cl (1×100 mL) and brine (1×100 mL).
The aqueous layers were further extracted with Et2O, and the com-
bined organic layers were then dried, and the solvents evaporated.
The crude was purified by flash chromatography (toluene/acetone,
1
(c = 1.0, CHCl3). H NMR (CDCl3): δ = 1.30 (d, J6,7 = 6.5 Hz, 3
H, 7-H), 1.98 (ddd, J1a,2 = 6.5, J1a,1b = 15.5, J1a,P = 19.7 Hz, 1 H,
1a-H), 2.09 (ddd, J1b,2 = 7.7, J1a,1b = 15.5, J1b,P = 19.0 Hz, 1 H,
1b-H), 3.43 (s, 3 H, OCH3), 3.49 (t, J2,3 = J3,4 = 3.5 Hz, 1 H, 3-
H), 3.50 (t, J4,5 = J5,6 = 7.5 Hz, 1 H, 5-H), 3.57–3.64 (m, 1 H, 6-
H), 3.68–3.74 [m, 7 H, PO(OCH3)2, 4-H], 4.33–4.40 (m, 1 H, 2-H),
4.56–4.84 (m, 4 H, 2CH2Ph), 7.23–7.37 (m, 10 H, arom.) ppm. 13
C
NMR (CDCl3): δ = 17.9 (7-C), 26.1 (d, JC,P = 142 Hz, 1-C), 52.4
(d, JC,P = 6.4 Hz, POCH3), 52.7 (d, JC,P = 6.5 Hz, POCH3), 58.1
(OCH3), 68.3 (2-C), 69.8 (6-C), 72.8 (OCH2Ph), 74.8(OCH2Ph),
77.5 (4-C), 79.0 (d, JC,P = 9.8 Hz, 3-C), 79.9 (5-C), 127.7–128.4
(10 C arom.), 138.1 (arom.), 138.4 (arom.) ppm. ESI-MS (positive-
ion mode): m/z (%) = 951.0 (100) [2 M + Na]+), 946.1 (55) [2 M +
NH4]+, 482.0 (30) [M + NH4]+, 465.2 (25) [M + H]+. C24H33O7P
(464.49): calcd. C 62.06, H 7.16; found C 61.86, H 6.81.
1
7:3) to give compound 8 as a colorless oil (1.60 g, 87%). H NMR
(CDCl3): δ = 1.30 (d, J6,7 = 6.0 Hz, 0.9 H, 7-H min.), 1.35 (d, J6,7
= 6.0 Hz, 2.1 H, 7-H maj.), 1.74–1.85 (m, 0.3 H, 1a-H min), 2.06
(dd, J1,2 = 7.3, J1,P = 19.3 Hz, 1.4 H, 1-H maj.), 2.09–2.19 (m, 0.3
H, 1b-H min.), 3.34–3.41 (m, 0.3 H, 6-H min.), 3.54–3.84 (m, 13
H, ArOCH3, 2POCH3, 2-H min, 3-H, 4-H, 5-H, 6-H maj), 4.35–
4.44 (m, 0.7 H, 2-H maj.), 4.48–4.99 (m, 6 H, 3 CH2Ph), 6.82–6.89
(m, 2 H, arom.), 7.22–7.42 (m, 12 H, arom.) ppm. 13C NMR
(CDCl3, selected signals): δ = 17.8 (7-C maj.), 18.1 (7-C min.), 26.4
[d, J(C,P) = 140.7 Hz, 1-C maj.], 27.7 (d, JC,P = 141.5 Hz, 1-C
min.), 51.9 (d, JC,P = 6.7 Hz, POCH3 min.), 52.3 (d, JC,P = 6.7 Hz,
POCH3 maj.), 52.6 (d, JC,P = 6.7 Hz, POCH3 maj.), 52.8 (d,
JC,P = 6.7 Hz, POCH3 min.), 55.3 (OCH3) ppm. ESI-MS (positive-
ion mode): m/z (%) = 1162.9 (100) [2 M + Na]+, 588.1 (40)
[M + NH4]+. C31H39O8P (570.61): calcd. C 65.25, H 6.89; found C
65.52, H 7.04.
Methyl 2,6-Anhydro-1,7-dideoxy-4,5-di-O-benzyl-3-O-methyl-
L-gly-
cero- -talo-heptit-1-yl Phosphonate (11): Compound 10 (0.11 g,
L
0.24 mmol) was dissolved in dry THF (2.5 mL). Freshly distilled
thiophenol (0.098 mL, 0.96 mmol) and TEA (0.16 mL, 1.20 mmol)
were added, and the solution was stirred at room temperature. Af-
ter 48 h, the mixture was diluted with TEA, and the solvents evapo-
rated to dryness. Flash chromatography (CH2Cl2/MeOH, from 9:1
to 7:3) gave compound 11 as a white solid (0.050 g, 46%). [α]2D0
+11.1 (c = 1.0, CH3OH). H NMR (CD3OD): δ = 1.25 (d, J6,7
=
=
1
6.4 Hz, 3 H, 7-H), 1.90–2.00 (m, 2 H, 2×1-H), 3.45–3.51 (m, 4 H,
5-H and OCH3), 3.56–3.65 (m, 4 H, 6-H and POCH3), 3.82–3.87
(m, 2 H, 3-H and 4-H), 4.35–4.42 (m, 1 H, 2-H), 4.61–4.87 (m, 4
Dimethyl
2,6-Anhydro-1,7-dideoxy-4,5-di-O-benzyl-L-glycero-L-
talo-heptit-1-yl Phosphonate (9): DDQ (0.76 g, 3.36 mmol) was H, 2×CH2Ph), 7.23–7.36 (m, 10 H, arom.) ppm. 13C NMR
added to a mixture of compound 8 (1.60 g, 2.80 mmol) in CH2Cl2/
H2O (18:1, 60 mL). The reaction mixture was stirred at room tem-
(CD3OD): δ = 18.6 (7-C), 28.1 (d, JC,P = 135.5 Hz, 1-C), 51.8 (d,
JC,P = 5.8 Hz, POCH3), 58.4 (OCH3), 70.5 (6-C), 72.1 (2-C), 72.7
4462
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4459–4463

(α-l-Rhamnopyranosyl)methylphosphonic Acids
(OCH2Ph), 76.0 (OCH2Ph), 80.1 (m, 2 C, 3-C and 4-C), 81.5 (5- 1]–. C7H15O7P (242.16): calcd. C 34.72, H 6.24; found C 34.85, H
FULL PAPER
C), 128.7–129.4 (10 C arom.), 139.8 (arom.), 139.9 (arom.) ppm.
ESI-MS (negative-ion mode): m/z = 449.5 [M – 1]–. C23H31O7P
(450.46): calcd. C 61.33, H 6.94; found C 61.40, H 7.13.
6.47.
Computational Methods: The conformational behavior of com-
pound 2 was investigated through a DFT approach at the B3LYP/
6-31G(d) level using the Gaussian03 package according to the same
procedure already described for compound 4.[6]
Methyl 2,6-Anhydro-1,7-dideoxy-3-O-methyl-L-glycero-L-talo-hept-
it-1-yl Phosphonate (4): Compound 11 (0.04 g, 0.089 mmol) was
dissolved in MeOH (2 mL), and a catalytic amount of palladium
hydroxide on charcoal was added under argon. The mixture was
hydrogenated at atmospheric pressure for 40 h, then diluted with
MeOH, filtered through a MILLIPORE filter, and the solvents
evaporated to dryness giving compound 4 as a white solid (0.024 g,
Acknowledgments
This work was supported by MIUR-Italy (COFIN 2004: “Chemi-
cal approach to new formulation vaccines through the synthesis of
complex saccharidic antigens and new adjuvant apt to potentiate
the immune response”).
quant.). [α]2D0 = +0.5 (c = 1.0, CH3OH). H NMR (CD3OD): δ =
1
1.26 (d, J6,7 = 6.3 Hz, 3 H, 7-H), 1.92–2.05 (m, 2 H, 2×1-H), 3.34
(t, J4,5 = J5,6 = 9.1 Hz, 1 H, 5-H), 3.47 (s, 3 H, OCH3), 3.50 (dq,
J5,6 = 9.1, J6,7 = 6.3 Hz, 1 H, 6-H), 3.62 (d, JC,P = 10.7 Hz, 3 H,
POCH3), 3.67 (dd, J2,3 = 1.75, J3,4 = 3.5 Hz, 1 H, 3-H), 3.71 (dd,
J3,4 = 3.5, J4,5 = 9.1 Hz, 1 H, 4-H), 4.35–4.42 (m, 1 H, 2-H) ppm.
13C NMR (CD3OD): δ = 18.2 (7-C), 27.2 (d, JC,P = 139.7 Hz, 1-
C), 52.4 (d, JC,P = 6.3 Hz, POCH3), 58.2 (OCH3), 70.2 (6-C), 71.5
(2 C, 2-C and 4-C), 74.5 (5-C), 83.1 (d, JC,P = 11.2 Hz, 3-C) ppm.
ESI-MS (negative-ion mode): m/z = 269.4 [M – 1]–. C9H19O7P
(270.22): calcd. C 40.00, H 7.09; found C 39.83, H 6.88.
[1] a) P. Compain, O. R. Martin, Bioorg. Med. Chem. 2001, 9,
3077–3092; b) A. Dondoni, A. Marra, Chem. Rev. 2000, 100,
4395–4421.
[2] R. Engel, Chem. Rev. 1977, 77, 349–367.
[3] P. Costantino, F. Berti, F. Norelli, A. Bartoloni, Int. Appl. WO
03/080678 A1, 2003 [Chem. Abstr. 2003, 139, 275732].
[4] N. Ohno, T. Yadomae, T. Miyazaki, Carbohydr. Res. 1980, 80,
297–304.
[5] a) L. Cipolla, B. La Ferla, F. Nicotra, L. Panza, Tetrahedron
Lett. 1997, 38, 5567–5568; b) L. Cipolla, B. La Ferla, L. Panza,
F. Nicotra, J. Carbohydr. Chem. 1998, 17, 1003–1013.
[6] F. Compostella, F. Marinone Albini, F. Ronchetti, L. Toma,
Carbohydr. Res. 2004, 339, 1323–1330.
[7] E. Bousquet, M. Khitri, L. Lay, F. Nicotra, L. Panza, G.
Russo, Carbohydr. Res. 1998, 311, 171–181.
[8] T. Hanaya, R. Okamoto, Y. V. Prikhodko, M. A. Armour,
A. M. Hogg, H. Yamamoto, J. Chem. Soc., Perkin Trans. 1
1993, 14, 1663–1671.
[9] a) J. F. Dellaria, Jr., R. G. Maki, H. H. Stein, J. Cohen, D.
Whittern, K. Marsh, D. J. Hoffman, J. J. Plattner, T. J. Perun,
J. Med. Chem. 1990, 33, 534–542; b) G. W. Daub, E. E. van Ta-
melen, J. Am. Chem. Soc. 1977, 99, 3526–3528.
[10] C. A. G. Haasnoot, F. A. A. M. de Leeuw, C. Altona, Tetrahe-
dron 1980, 36, 2783–2792.
2,6-Anhydro-1,7-dideoxy-L-glycero-L-talo-heptit-1-yl
Phosphonic
Acid (2): Compound 2 was synthesized from 9 (0.08 g, 0.18 mmol)
by reaction with Me3SiI according to the procedure of Nicotra et
al.,[5b] but recovered in its acid form after loading on a cation ex-
change resin column (Dowex 50×8, H+ form), elution with meth-
anol, and evaporation of the solvent under reduced pressure
1
(0.036 g, 85%). H NMR (D2O): δ = 1.20 (d, J6,7 = 6.2 Hz, 3 H,
7-H), 2.02 (ddd, J1a,2 = 6.5, J1a,1b = 15.5, J1a,P = 20.0 Hz, 1 H, 1a-
H), 2.17 (ddd, J1b,2 = 8.5, J1a,1b = 15.5, J1b,P = 18.8 Hz, 1 H, 1b-
H), 3.38 (t, J4,5 = J5,6 = 9.5 Hz, 1 H, 5-H), 3.55 (dq, J5,6 = 9.5, J6,7
= 6.2 Hz, 1 H, 6-H), 3.74 (dd, J3,4 = 3.2, J4,5 = 9.5 Hz, 1 H, 4-H),
3.94 (dd, J2,3 = 2.0, J3,4 = 3.2 Hz, 1 H, 3-H), 3.96–4.03 (m, 1 H, 2-
H) ppm. 13C NMR (D2O): δ = 16.4 (7-C), 26.7 (d, JC,P = 133.5 Hz,
1-C), 69.0 (6-C), 69.4 (4-C), 71.1 (d, JC,P = 11.3 Hz, 3-C), 71.8 (5-
C), 73.7 (2-C) ppm. ESI-MS (negative-ion mode): m/z = 241.3 [M –
Received: June 14, 2005
Published Online: September 1, 2005
Eur. J. Org. Chem. 2005, 4459–4463
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4463