A concise synthesis of α-D-ribofuranosyl alkylphosphonates - Putative substrate intermediates for the carbon-phosphorous lyase system

Article abstract of DOI:10.1139/V06-038

Carbon-phosphorous lyase is a multienzyme system found in many species of bacteria that is distinguished by its ability to hydrolyze a broad array of unactivated alkylphosphonates. α-D-Ribofuranosyl alkylphosphonates are potential metabolic intermediates

Full text of DOI:10.1139/V06-038

743
A concise synthesis of ␣-D-ribofuranosyl
alkylphosphonates — Putative substrate
intermediates for the carbon–phosphorous lyase
system1
Yan Luo and David L. Zechel
Abstract: Carbon–phosphorous lyase is a multienzyme system found in many species of bacteria that is distinguished
by its ability to hydrolyze a broad array of unactivated alkylphosphonates. α-D-Ribofuranosyl alkylphosphonates are po-
tential metabolic intermediates generated by the carbon–phosphorous lyase pathway. Here we describe a facile synthesis
of α-D-ribofuranosyl alkylphosphonates using β-D-ribofuranosyl trichloracetimidate as a glycosyl donor.
Key words: carbon–phosphorous lyase, phn operon, phnN, phosphonates, glycosyl trichloroacetimidate donor, α-D-
ribofuranosyl ethylphosphonate.
Résumé : La lyase carbone–phosphore est un système multienzyme qu’on retrouve dans plusieurs espèces et qui se
distingue par sa facilité à hydrolyser une grande variété d’alkylphosphonates qui ne sont pas activés. Les alkylphospho-
nates de α-D-ribofuranosyle sont des intermédiaires métaboliques potentiels générés par la voie réactionnelle de la lyase
carbone–phosphore. Dans ce travail, on décrit une méthode facile de synthèse d’alkylphosphonates de α-D-ribofurano-
syle utilizant le trichloroacétimidate de β-D-ribofuranosyle comme donneur de glycosyle.
Mots clés : lyase carbone–phosphore, phn opéron, phnN, phosphonates, trichloroacétimidate de glycosyle comme don-
neur, éthylphosphonate de α-D-ribofuranosyle. Luo and Zechel 747
Introduction
herbicide bialaphos (4). 2-Amino-3-phosphonopropionic
acid and 2-aminoethylphosphonic acid are the most widely
distributed biogenic organophosphonates, occurring as con-
stituents of lipids, proteins, and polysaccharides in virtually
all organisms, including man (5). Due to the global abun-
dance of organophosphonates, bacteria have also evolved a
number of strategies to cleave the CP bond to produce inor-
ganic phosphate when the latter is scarce in the environment
(1, 6). At present, four distinct CP bond-cleaving activities
are known. Three of these activities, comprised of
phosphonoacetaldehyde hydrolase (7), phosphonopyruvate
hydrolase, (8) and phosphonoacetate hydrolase (9), share a
β-carbonyl group as a feature of their respective substrates
(1 in Scheme 1). A mechanism for the cleavage of the CP
bond in these substrates can be envisioned as direct
nucleophilic attack by water or an enzyme residue on phos-
phorous, displacing a carbanion to form an enzyme stabi-
lized enolate. In the case of phosphonoacetaldehyde
hydrolase, the substrate β-carbonyl is activated as a Schiff
base (2) followed by attack of a carboxylate nucleophile to
form an acyl phosphate intermediate (3); hydrolysis forms
acetaldehyde (4) and inorganic phosphate (7).
Unlike phosphonoacetaldehyde hydrolase, CP-lyase is a
multienzyme system that is distinguished by its ability to
cleave a broad array of “unactivated” alkylphosphonates (5)
yielding alkanes (7) and inorganic phosphate as products
(Scheme 2) (10). In Escherichia coli, the phn operon encod-
ing CP-lyase consists of 14 genes (phnCDEFGHIJKLMNOP)
(11, 12); mutational and sequence analysis of the phn operon
in whole cell metabolic studies has shown that phnCDE en-
codes a phosphonate transport system and phnG–phnP en-
Phosphate is a fundamental functional group in biology as
exemplified by its many structural, signalling, and energy
storage roles. Organophosphonates are characterized by a
highly stable carbon–phosphorous (CP) bond that is compa-
rable in energy to a carbon–carbon bond (1). In addition to
other reduced forms of phosphate such as phosphite and
hypophosphite, organophosphonates are significant contribu-
tors to the global phosphate pool. They may well have
served as the original source of water soluble phosphorous
for the assembly of biological precursors on prebiotic earth,
as inorganic phosphate would have been largely inaccessible
as the mineral apatite (2). Although anthropogenic genera-
tion of organophosphonates is considerable (1), natural pro-
duction is also substantial. The biosynthesis of the majority
of organophosphonate natural products begins with the rear-
rangement of phosphenolpyruvate to phosphonopyruvate by
phosphoenolpyruvate mutase (3). From this precursor,
Streptomyces species biosynthesize a number of bioactive
phosphonates, including the antibiotic fosfomycin and the
Received 1 September 2005. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 17 May 2006.
Y. Luo and D.L. Zechel.2 Department of Chemistry,
Chernoff Hall, Queen’s University, Kingston, ON K7L 3N6,
Canada.
1This article is part of a Special Issue dedicated to Professor
Walter A. Szarek.
2Corresponding author (e-mail: dlzechel@chem.queensu.ca).
Can. J. Chem. 84: 743–747 (2006)
doi:10.1139/V06-038
© 2006 NRC Canada

744
Can. J. Chem. Vol. 84, 2006
Scheme 1. Cleavage of an “activated” carbon–phosphorous bond by phosphonoacetaldehyde hydrolase (7).
Asp
Asp
O
HO P O
O
Mg2+
O
O
O
O
Mg2+
2 x H2O
O
O
O P O
O
P O
O
H
P O
O
H
H
H
3
O
H2O
NH
NH
Lys
1
O
2
Lys
4
Scheme 2. A hypothetical reaction pathway depicting the cleavage of “unactivated” alkylphosphonates by CP-lyase utilizing 6 as an in-
termediate. The subsequent phosphorylation reaction by phnN produces a precursor (9) for NADH biosynthesis (14).
O
R
O P OH
O
7
O
phnC-M
???
HO
O P O
O
phnC-M
O
O P
O
O
O
R
O
O P
O
O
O P O
O
R
OH OH
OH OH
5
6
8
ATP
phnN
ADP
O
O P O
O
O
O
O
P O P O
O
OH OH
O
O
9
codes enzymes that are required for CP bond cleavage and
transcription regulation (13). The sequences of several of
these enzymes are unique, yielding no insight via homology
into their possible function. An in vitro reconstruction of
CP-lyase activity has not yet been achieved, further hamper-
ing functional assignments. This is primarily due to the diffi-
culty in expressing several of these enzymes, some of which
appear to be associated with the membrane of bacterial cells
(13). The only in vitro reaction assigned to a phn protein is
that of phnN, which was recently shown to be an ATP-
dependent kinase, phosphorylating 5-phospho-α-D-ribofuran-
osyl phosphate (8) to yield 5-phospho-α-D-ribofuranosyl
diphosphate (PRPP, 9), a precursor in the biosynthesis of the
essential cofactor NAD (14). Intriguingly, α-D-ribofuranosyl
ethylphosphonate (6) was observed to reach significant con-
centrations in the culture medium when E. coli with a mu-
tated, yet partly functional phn operon subsisted on
ethylphosphonate. Free phosphate was not produced. By
contrast, a mutant strain of E. coli with an inoperative phn
operon did not produce 6 when fed ethylphosphonate, indi-
cating that the some of the CP-lyase enzymes were likely in-
volved in its biosynthesis (15). The substrate specificity of
phnN and the biosynthesis of 6 by CP-lyase enzymes
strongly suggest that a phn enzyme can cleave the CP bond
of 6 (or a derivative) and in doing so support NAD
biosynthesis. The chemical mechanism of CP-lyase is of
great interest, not only for its ability to cleave the remark-
ably stable carbon–phosphorous bond, but also for the devel-
opment of environmental remediation catalysts that will de-
grade phosphonates, such as nerve agents and herbicides. As
part of our current program in characterizing the functions
and reactions of individual CP-lyase enzymes, we have syn-
thesized 6 and other phosphonate analogues as putative CP-
lyase substrate intermediates.
Results and discussion
A major challenge facing the stereoselective synthesis of
α-D-ribofuranosyl phosphates is the reactivity and 1,2-cis ar-
rangement of the glycosidic linkage (the product itself is a
glycosyl donor (16)). The diphosphate 8 has been synthe-
sized enzymatically (17) by direct condensation of the per-
O-acetylated sugar with phosphoric acid (18, 19) or by reac-
tion of the ribofuranosyl bromide with dibenzyl phosphate
(20, 21). A synthetically useful enzymatic route to 6 is not
yet possible, and the current synthetic methods suffer from
low yields or the use of toxic and difficult to handle reagents
(e.g., HBr gas). We report that β-D-ribofuranosyl trichloro-
acetimidate serves as a high yielding and stereoselective do-
nor for the synthesis of 6.
The required glycosyl acceptors (11a and 11b) were con-
veniently prepared in two steps (Scheme 3). The correspond-
ing phosphonochloridates were reacted with benzyl alcohol
to yield dibenzyl methylphosphonate (10a) and dibenzyl
ethylphosphonate (10b) (22). Refluxing with DABCO in an-
hydrous toluene afforded mono-protected 11a and 11b quan-
© 2006 NRC Canada

Luo and Zechel
745
1
Scheme 3. Reagents and conditions: (a) BnOH, toluene–
pyridine, 0 °C to RT, 5 h; (b) DABCO, toluene, reflux, 3 h.
performed under a dry nitrogen atmosphere. H NMR spectra
was recorded on a 400 MHz Bruker spectrometer using
CDCl3 (δ 7.26 ppm), MeOD (δ 3.35 ppm), and D2O (δ 4.81 ppm)
as internal references. 13C NMR and 31P NMR spectra were
recorded at 100.6 and 162.0 MHz, respectively. 31P NMR
spectra are reported relative to phosphate. Mass spectra were
determined by electrospray ionization (ESI) or fast atom
bombardment (FAB) mass spectroscopy. Analytical TLC
was perfomed on Silicycle F254 silica gel plates. Visualiza-
tion was performed under UV light or by staining with ceric
ammonium molybdate. Flash chromatography was per-
formed using Silicycle (40–63 µm) silica gel.
O
Cl P Cl
R
O
BnO P OBn
R
O
HO P OBn
R
a
b
10a R = Me
10b R = Et
11a R = Me
11b R = Et
titatively (23). The trichloroacetimidate glycosyl donor 12
was prepared from methyl 2,3,5-tri-O-benzyl-α-D-ribofur-
anoside after solvolysis in 5% aq. HCl – dioxane and treat-
ment of the free anomeric hydroxyl with trichloroacetonitrile
and potassium carbonate in dry dichloromethane (Scheme 4).
Phosphonic acids 11a and 11b were reacted with the donor
12 in dry dichloromethane in the presence of 4 Å molecular
sieves to produce a 2:1 α/β mixture of the ribofuranosyl
alkylphosphonates 14 and 15 in 60% yield. Reaction with
dibenzyl phosphoric acid produced a similar mixture of the
corresponding ribofuranosyl phosphate (13). An attempt to
increase the α/β ratio by reducing the reaction temperature
to –20 °C was unsuccessful. However, the β anomers readily
isomerized to the thermodynamically stable α anomer in the
presence of BF3·Et2O and 2-chloropyridine. The β anomer
was notably unstable, as we observed that 13β decomposed
overnight at room temperature. The stereochemistry of the α
anomer was assigned based on NOESY experiments: the
signal corresponding to H-1 (δ 6.09) exhibits an NOE with
both H-2 and H-3, indicating that H-1, H-2, and H-3 are on
the same side of the ribofuranose ring. Compounds 13α,
14α, and 15α were deprotected by catalytic hydrogenolysis
to afford the desired α-D-ribofuranosyl phosphate (16) and
alkylphosphonates (17 and 18) in quantitative yield. These
were subsequently converted to the triethylammonium salts.
The putative substrate intermediates (17 and 18) are very
likely to be crucial for assigning the functions of CP-lyase
proteins as they are all, or in part, produced in a soluble
form suitable for in vitro studies. The stereoselective and
high yielding synthetic route described here is considerably
more facile than previously reported methods. We can now
readily synthesize a wide array of putative CP-lyase interme-
diates that can also include radiolabels, chromaphores, or
fluorophores. Of particular interest is the substrate specific-
ity of the CP-lyase enzyme that ultimately cleaves the CP
bond: Does this enzyme cleave simple alkylphosphonates or
only glycosylated derivatives like 6 or both? Likewise,
which CP-lyase enzyme is responsible for forming 6? As the
ultimate CP bond cleaving mechanism is suspected to be
radical based (10), we can also readily incorporate radical
stabilizing functionalities (thio, allyl, etc.) into 17 and 18 to
serve as radical probes. Our current success in producing
soluble CP-lyase proteins will benefit from access to these
substrates and will be the topic of future reports.
Synthesis
Dibenzyl ethylphosphonate (10b)
A
suspension of ethylphosphonylchloridate (1 mL,
9.4 mmol) was stirred rapidly in dry toluene (5 mL) at 0 °C.
A mixture of dry benzyl alcohol (2 mL, 18.8 mmol) and dry
pyridine (1.5 mL, 18.8 mmol) was added over 90 min while
the temperature was maintained. After the addition was
complete, the reaction was allowed to reach room tempera-
ture and stirred for a further 3 h. The solid product, dibenzyl
ethylphosphonate (10b), was removed by filtration and
washed with toluene, 2 mol/L NaOH, water, then dried over
Na2SO4 and concentrated in vacuo. Crude 10b was used in
the next reaction without further purification. Data for 10b:
1H NMR (400 MHz, CDCl3) δ: 7.4–7.3 (m, 10H, aromatic),
5.08 (m, 2H, OCH2Ph), 4.98 (m, 2H, OCH2Ph), 1.76 (m,
2H, CH2), 1.22 (m, 3H, CH3). 31P NMR (CDCl3) δ: 35, 37.
Benzyl ethylphosphonic acid (11b)
To a stirred solution of crude 10b (3.0 g, 10 mmol) in dry
toluene (20 mL) was added DABCO (1.12 g, 10 mmol) un-
der a nitrogen atmosphere. The reaction mixture was refluxed
for 3 h before the solvent was removed in vacuo and the res-
idue was redissolved in 5% aq. HCl. The aqueous layer was
extracted with ethyl acetate, dried over Na2SO4, then evapo-
rated under reduced pressure to yield benzyl ethylphosphonic
acid (11b) (1.87 g, 99% calcd. from ethylphosphonylchloridate).
1
Data for 11b: H NMR (400 MHz, CDCl3) δ: 5.07 (m, 2H,
OCH2Ph), 1.8 (m, 2H, CH2), 1.2 (m, 3H, CH3). 31P NMR
(CDCl3) δ: 36.
Dibenzyl methylphosphonate (10a)
The procedure for the preparation of 10b was followed.
Data for 10a: H NMR (400 MHz, CDCl3) δ: 7.4–7.3 (m,
10H, aromatic), 5.08 (m, 2H, OCH2Ph), 4.98 (m, 2H,
OCH2Ph), 1.5 (d, J = 17.6 Hz, 3H, CH3). 13C NMR (CDCl3)
δ: 67.16 (OCH2Ph), 67.09 (OCH2Ph), 12.42, 10.99 (CH3).
31P NMR (CDCl3) δ: 31.
1
Experimental
Benzyl methylphosphonic acid (11a)
The procedure for the preparation of 11b was followed.
General
1
Data for 11a: H NMR (400 MHz, CDCl3) δ: 7.3–7.4 (m,
Solvents were distilled under dry nitrogen by standard
methods (THF from Na/benzophenone and CH2Cl2 from
CaH2) prior to use. All commercially available reagents
were used without further purification. All reactions were
5H, aromatic), 5.07 (m, 2H, OCH2Ph), 1.55 (d, J = 17.9 Hz,
3H, CH3). 13C NMR (CDCl3) δ: 66.59 (OCH2Ph), 12.39,
11.14 (CH3).
© 2006 NRC Canada

746
Can. J. Chem. Vol. 84, 2006
Scheme 4. Reagents and conditions: (a) phosphonic acid, dichloromethane, 0 °C to RT, 3 h; (b) BF3·Et2O, 2-chloropyridine, RT, 2 h;
(c) Pd/C (10%), H2, 12 h.
O
NH
BnO
HO
BnO
BnO
O
O P OBn
R
O
O
O
O
O
a
O
c
CCl3
+
O P OBn
O P OH
R
OBnOBn
OH OH
OBn OBn
OBn OBn
R
12
13α R = OBn
14α R = Me
15α R = Et
13β R =OBn
14β R = Me
15β R = Et
16 R = OH
17 R = Me
18 R = Et
b
(C3), 73.16, 72.54, 72.27 (OCH2Ph), 72.19 (C5), 66.8
2,3,5-Tri-O-benzyl-D-ribofuranosyl benzyl ethylphosphonate
(15)
(POCH2Ph). 31P NMR (CDCl3) δ: 34.
A solution of 2,3,5-tri-O-benzyl-D-ribofuranose (1 g,
2.5 mmol), potassium carbonate (396 mg, 2.87 mmol), and
trichloroacetonitrile (4 mL, 40 mmol) in anhydrous dichloro-
methane (20 mL) was stirred overnight at room temperature.
The mixture was filtered through Celite and concentrated in
vacuo. Flash chromatography using ethyl acetate – hexanes
(1:2) yielded pure 2,3,5-tri-O-benzyl-β-D-ribofuranosyl
2,3,5-Tri-O-benzyl-D-ribofuranosyl dibenzylphosphate (13)
The procedure for 15 was followed. Data for 13α: 1H
NMR (400 MHz, CDCl3) δ: 7.34–7.2 (m, 25H, aromatic),
6.1 (t, J1,2 = 1.5 Hz, J1,P = 2.1 Hz, 1H, H1), 5.08 (m, 4H,
P(OCH2Ph)2), 4.7 (m, 2H, OCH2Ph), 4.5–4.6 (m, 4H,
OCH2Ph), 4.4 (m, 1H, H4), 4.02 (d, 2H, H2,H3), 3.5 (d, 2H,
H5). 13C NMR (CDCl3) δ: 99.1, 99.06 (C1), 8.4.64 (C4),
79.05 (C2), 75.44 (C3), 73.55, 72.68, 72.63 (OCH2Ph),
69.79 (C5), 69.31, 69.25 (P(OCH2Ph)2). 31P NMR (CDCl3)
1
trichloracetimidate (12). Data for 12: H NMR (400 MHz,
CDCl3) δ: 8.5 (s, 1H, NH), 7.4–7.3 (m, 15H, aromatic), 6.35
(s, 1H, H1), 4.8–4.4 (m, 7H, OCH2Ph, H2), 4.15 (m, 2H,
H3, H4), 3.7 (dd, 2H, H5a, H5b). 13C NMR (CDCl3) δ:
160.77 (C=NH), 103.43 (C1), 82.15 (C4), 78.48 (C2), 77.22
(C3), 73.25, 72.54, 72.18 (CH2Ph), 70.19 (C5). To an ice-
cold solution of 12 (698 mg, 1.24 mmol) in anhydrous di-
chloromethane was added 11b (527 mg, 2.6 mmol) in the
presence of 4 Å molecular sieves. The mixture was stirred
for 3 h whereupon TLC indicated complete conversion of
12. The mixture was filtered through Celite and concentrated
in vacuo to yield an anomeric mixture of 15 (450 mg, 60%,
α/β = 2:1). The anomers were resolved by flash chromatog-
raphy (ethyl acetate – hexanes, 1:3 to 1:1). The β anomer 15
β (100 mg, 0.17 mmol) was subsequently dissolved in anhy-
drous dichloromethane (5 mL) followed by BF3·Et2O
(10 µL, 0.087 mmol) and 2-chloropyridine (16 µL,
0.174 mmol). The mixture was stirred at room temperature
for 1.5 h then neutralized with triethylamine and evaporated
in vacuo. The crude product was purified by flash chroma-
tography (ethyl acetate – hexane, 1:1) to yield the α anomer
15α (75 mg, 75%). Data for 15α: 1H NMR (400 MHz,
δ: –2.51. Data for 13β: 1H NMR (400 MHz, CDCl3) δ:
7.34–7.2 (m, 25H, aromatic), 5.86 (d, J = 4.9 Hz, 1H, H1),
5.0–5.1 (m, 4H, P(OCH2Ph)2), 4.7–4.5 (m, 6H, OCH2Ph),
4.4 (m, 1H, H4), 4.1 (m, 1H, H3), 3.9 (m, 1H, H2), 3.6–3.7
(dd, 2H, H5). 13C NMR (CDCl3) δ: 102.65 (C1), 81.08 (C4),
79.13 (C2), 77.71 (C3), 73.18, 72.58, 72.32 (OCH2Ph), 70.4
(C5), 69.33, 69.29 (P(OCH2Ph)2). 31P NMR (CDCl3) δ: –3.1.
2,3,5-Tri-O-benzyl-D-ribofuranosyl benzyl
methylphosphonate (14)
The procedure for 15 was followed. Data for 14α: 1H
NMR (400 MHz, CDCl3) δ: 7.4–7.3 (m, 20H, aromatic),
6.09 (dd, 1H, J1,2 = 1.6 Hz, J1,P = 3.8 Hz, H1), 4.9–5.1 (m,
2H, POCH2Ph), 4.7 (d, 2H, OCH2Ph), 4.6 (m, 2H,
OCH2Ph), 4.5 (d, 2H, OCH2Ph), 4.3 (m, 1H, H4), 3.9 (m,
2H, H2, H3), 3.4 (m, 2H, H5). 13C NMR (CDCl3) δ: 97.37
(C1), 84.13 (C4), 78.63 (C2), 75.27 (C3), 73.52 (OCH2Ph),
72.73 (OCH2Ph), 72.49 (OCH2Ph), 69.81 (C5), 67.15
1
(POCH2Ph). 31P NMR (CDCl3) δ: 32, 31. Data for 14β: H
NMR (400 MHz, CDCl3) δ: 7.4–7.3 (m, 20H, aromatic),
5.87 (d, 1H, J = 5.8 Hz, H1), 5.0 (m, 2H, POCH2Ph), 4.7 (d,
2H, OCH2Ph), 4.6 (d, 2H, OCH2Ph), 4.5 (m, 2H, OCH2Ph),
CDCl3) δ: 7.1–7.4 (m, 20H, aromatic), 6.09 (dd, J1,2
=
4.4 Hz, J1,P = 4.6 Hz, 1H, H1), 5.0–5.1 (m, 2H, POCH2Ph),
4.8–4.4 (m, 6H, OCH2Ph), 4.38 (m, 1H, H4), 3.9 (d, 2H, H2,
H3), 3.5 (dd, 2H, H5), 1.9 (m, 2H, CH2), 1.2 (m, 3H, CH3).
13C NMR (CDCl3) δ: 97 (C1), 84.3 (C4), 78.97 (C2), 75.54
(C3), 73.53, 72.56, 72.63 (OCH2Ph), 69.92 (C5), 66.20
(POCH2Ph), 18.89 (CH2), 6.44 (CH3). 31P NMR (CDCl3) δ:
34. ESI-MS m/z: 603.22 (50%, [M + H]+), 295.12 (100%,
[M – C16H20O4P – H]+). HR-MS m/z calcd. for C35H40O7P
4.1 (m, 1H, H4), 3.9 (m, 2H, H2, H3), 3.59 (m, 2H, H5). 13
C
NMR (CDCl3) δ: 100.75 (C1), 80.07 (C4), 79.99 (C2), 76.65
(C3), 73.17 (OCH2Ph), 72.58 (OCH2Ph), 72.26 (OCH2Ph),
69.84 (C5), 66.95 (POCH2Ph). 31P NMR (CDCl3) δ: 30, 31.
␣-D-Ribofuranosyl ethylphosphonate (18)
1
[M + H]+: 603.25116; found: 603.21992. Data for 15β: H
To a solution of compound 15α (200 mg, 0.33 mmol) in
MeOH (10 mL) was added 10% Pd/C. The reaction mixture
was stirred under H2 overnight, filtered through Celite, and
concentrated to afford 18 (82 mg, 100%). The crude product
was purified through an Amberlite IR-120 ion exchange col-
umn, titrated with triethylamine, then dried in vacuo to ob-
tain the triethylammonium salt. Data for 18: 1H NMR
NMR (400 MHz, CDCl3) δ: 7.1–7.4 (m, 20H, aromatic),
5.88 (dd, J1,2 = 5.9 Hz, J1,P = 6.1 Hz, 1H, H1), 4.9–5.1 (m,
2H, POCH2Ph), 4.5–4.7 (m, 6H, OCH2Ph), 4.4 (m, 1H, H4),
4.15 (m, 1H, H3), 3.9 (m, 1H, H2), 3.7 (dd, 1H, H5a), 3.55
(dd, 1H, H5b), 1.9 (m, 2H, CH2), 1.2 (m, 3H, CH3). 13C
NMR (CDCl3) δ: 100.63 (C1), 81.54 (C4), 80.01 (C2), 77.03
© 2006 NRC Canada

Luo and Zechel
747
2. R.M. de Graaf, J. Visscher, and A.W. Schwartz. Nature (Lon-
don), 378, 474 (1995).
3. J. Kim and D. Dunaway-Mariano. Biochemistry, 35, 4628
(1996).
4. H. Seto and T. Kuzuyama. Nat. Prod. Rep. 16, 589 (1999).
5. S.A. Tan and L.G. Tan. Clin. Physiol. Biochem. 7, 303 (1989).
6. S.V. Kononova and M.A. Nesmeyanova. Biochemistry (Mos-
cow), 67, 184 (2002).
7. M.C. Morais, G. Zhang, W. Zhang, D.B. Olsen, D. Dunaway-
Mariano, and K.N. Allen. J. Biol. Chem. 279, 9353 (2004).
8. A.N. Kulakova, G.B. Wisdom, L.A. Kulakov, and J.P. Quinn.
J. Biol. Chem. 278, 23426 (2003).
9. M.Y. Galperin and M.J. Jedrzejas. Proteins, 45, 318 (2001).
10. J.W. Frost. J. Am. Chem. Soc. 109, 2166 (1987).
11. A.K. White and W.W. Metcalf. J. Bacteriol. 186, 4730 (2004).
12. C.M. Chen, Q.Z. Ye, Z.M. Zhu, B.L. Wanner, and C.T. Walsh.
J. Biol. Chem. 265, 4461 (1990).
(400 MHz, D2O) δ: 4.78 (s, 1H, H1), 4.03 (m, 1H, H3), 3.9–
3.8 (m, 2H, H2, H4), 3.7 (dd, 1H, H5a), 3.5 (dd, 1H, H5b),
1.3 (m, 2H, CH2), 0.9 (m, 3H, CH3). 13C NMR (D2O) δ:
107.82 (C1), 82.69 (C2), 74.07 (C4), 70.64 (C3), 62.62
(C5), 21 (CH2), 7 (CH3). 31P NMR (D2O) δ: 25.89.
␣-D-Ribofuranosyl phosphate (16)
The procedure for 18 was followed. Data for 16: 1H NMR
(400 MHz, D2O) δ: 4.77 (s, 1H, H1), 4.02 (m, 1H, H2), 3.8–
3.9 (m, 2H, H3, H4), 3.67 (dd, 1H, H5a), 3.5 (m, 1H, H5b).
13C NMR (CDCl3) δ: 107.82 (C1), 82.7 (C2), 74.08 (C4),
69.57 (C3), 61.40 (C5). 31P NMR (CDCl3) δ: 0.11.
␣-D-Ribofuranosyl methylphosphonate (17)
The procedure for 18 was followed. Data for 17: 1H NMR
(400 MHz, MeOD) δ: 4.76 (s, 1H, H1), 4.04 (m, 1H, H3),
3.96 (m, 1H, H2), 3.89 (d, 1H, H4), 3.74 (dd, 1H, H5a), 3.55
(dd, 1H, H5b), 1.3 (s, 3H, CH3). 13C NMR (MeOD) δ:
108.44 (C1), 83.44 (C2), 74.77 (C4), 71.27 (C3), 63.61
(C5), 15 (CH3). 31P NMR (MeOD) δ: 30.24.
13. G.M. Yakovleva, S.K. Kim, and B.L. Wanner. Appl.
Microbiol. Biotechnol. 49, 573 (1998).
14. B. Hove-Jensen, T.J. Rosenkrantz, A. Haldimann, and B.L.
Wanner. J. Bacteriol. 185, 2793 (2003).
15. L.Z. Avila, K.M. Draths, and J.W. Frost. Biorg. Med. Chem.
Lett. 1, 51 (1991).
16. O.J. Plante, E.R. Palmacci, R.B. Andrade, and P.H. Seeberger.
J. Am. Chem. Soc. 123, 9545 (2001).
17. H. Klenow and R. Emberland. Arch. Biochem. Biophys. 58,
276 (1955).
18. R. Hanna and J. Mendicino. J. Biol. Chem. 245, 4031 (1970).
19. D.L. MacDonald. J. Org. Chem. 27, 1107 (1961).
20. B.E. Maryanoff, A.B. Reitz, and S.O. Nortey. Tetrahedron, 44,
3093 (1988).
Acknowledgements
This paper is dedicated to Professor Walter Szarek, not
only for his inspirational contributions to the fields of carbo-
hydrate chemistry and Alzheimer’s research, but also for his
infallible warmth and good humor as a colleague. We also
thank the Natural Sciences and Engineering Research Coun-
cil of Canada (NSERC) and the Canadian Foundation for In-
novation (CFI) for financial support.
21. G.M. Tener and H.G. Khorana. J. Am. Chem. Soc. 80, 1999
(1958).
22. P.C. Bulman Page, M.J. McKenzie, and J.A. Gallagher. Synth.
References
Commun. 32, 211 (2002).
1. N.G. Ternan, J.W. McGrath, G. McMullan, and J.P. Quinn.
23. M. Saady, L. Lebeau, and C. Mioskowski. J. Org. Chem. 60,
2946 (1995).
World J. Microbiol. Biotechnol. 14, 635 (1998).
© 2006 NRC Canada