Competitive inhibitors of type B ribose 5-phosphate isomerases: design, synthesis and kinetic evaluation of new d-allose and d-allulose 6-phosphate derivatives

Article abstract of DOI:10.1016/j.carres.2009.02.024

This study reports syntheses of d-allose 6-phosphate (All6P), d-allulose (or d-psicose) 6-phosphate (Allu6P), and seven d-ribose 5-phosphate isomerase (Rpi) inhibitors. The inhibitors were designed as analogues of the 6-carbon high-energy intermediate pos

Full text of DOI:10.1016/j.carres.2009.02.024

Carbohydrate Research 344 (2009) 869–880
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Competitive inhibitors of type B ribose 5-phosphate isomerases: design,
synthesis and kinetic evaluation of new D-allose and D-allulose
6-phosphate derivatives
Sandrine Mariano a, Annette K. Roos b, , Sherry L. Mowbray c, Laurent Salmon a,
*
a Laboratoire de Chimie Bioorganique et Bioinorganique, CNRS-UMR 8182, Institut de Chimie Moléculaire et des Matériaux d’Orsay, Bâtiment 420, Université
Paris-Sud XI, F-91405 Orsay, France
b Department of Cell and Molecular Biology, Uppsala University, Biomedical Center, Box 596, SE-751 24 Uppsala, Sweden
c Department of Molecular Biology, Swedish University of Agricultural Sciences, Biomedical Center, Box 590, SE-751 24 Uppsala, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
This study reports syntheses of
D-allose 6-phosphate (All6P), D-allulose (or D-psicose) 6-phosphate
Received 26 November 2008
Received in revised form 20 February 2009
Accepted 23 February 2009
Available online 3 March 2009
(Allu6P), and seven
logues of the 6-carbon high-energy intermediate postulated for the All6P to Allu6P isomerization reaction
(Allpi activity) catalyzed by type B Rpi from Escherichia coli (EcRpiB). 5-Phospho- -ribonate, easily
obtained through oxidative cleavage of either All6P or Allu6P, led to the original synthon 5-dihydrogeno-
phospho- -ribono-1,4-lactone from which the other inhibitors could be synthesized through nucleophilic
addition in one step. Kinetic evaluation on Allpi activity of EcRpiB shows that two of these compounds, 5-
phospho- -ribonohydroxamic acid and N-(5-phospho- -ribonoyl)-methylamine, indeed behave as new
efficient inhibitors of EcRpiB; further, 5-phospho- -ribonohydroxamic acid was demonstrated to have
D-ribose 5-phosphate isomerase (Rpi) inhibitors. The inhibitors were designed as ana-
D
D
Keywords:
Sugar phosphates
Enzyme inhibitor
Hydroxamic acids
D
D
D
competitive inhibition. Kinetic evaluation on Rpi activity of both EcRpiB and RpiB from Mycobacterium
tuberculosis (MtRpiB) shows that several of the designed 6-carbon high-energy intermediate analogues
are new competitive inhibitors of both RpiBs. One of them, 5-phospho-D-ribonate, not only appears as
Ribose 5-phosphate isomerase
Allose 6-phosphate isomerase
Mycobacterium tuberculosis
the strongest competitive inhibitor of a Rpi ever reported in the literature, with a Ki value of 9 lM for
MtRpiB, but also displays specific inhibition of MtRpiB versus EcRpiB.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
in almost all organisms including humans, Spinacia oleracea, and
Escherichia coli (Ec). The second type, RpiB, is found mainly in bac-
teria, including Mycobacterium tuberculosis (Mt) and E. coli. Due to
important differences in the two active sites including a change in
the sequence position and nature of the catalytic base, M. tubercu-
losis and E. coli RpiBs were proposed to be representative of two
distinct sub-families.5 Indeed, we recently demonstrated, through
kinetic analyses and structural studies, that EcRpiB, but not MtRpiB
is also a functional allose 6-phosphate isomerase (Allpi), catalyzing
Ribose 5-phosphate isomerase (Rpi, E.C. 5.3.1.6), a key enzyme
of the pentose phosphate pathway and of the Calvin cycle of plants,
is the metal-independent aldose–ketose isomerase that catalyzes
the reversible conversion of D-ribose 5-phosphate (R5P) into D-
ribulose 5-phosphate (Ru5P). As observed in the case of triosephos-
phate isomerase (TIM),1 phosphoglucose isomerase (PGI),2,3 and
phosphomannose isomerase (PMI),3,4 the isomerization reaction
catalyzed is thought to proceed through a proton transfer mecha-
nism between the two carbon atoms C-1 and C-2 of the substrates,
concomitant with a proton transfer between the two oxygen atoms
O-1 and O-2, and thus involves a 1,2-cis-enediol(ate) high-energy
intermediate (HEI) as depicted in Chart 1.
the reversible isomerization of the 6-carbon sugars
D
-allose 6-
phosphate (All6P) and D-allulose (also called D-psicose) 6-phos-
phate (Allu6P).6,7 A proton transfer mechanism similar to that pro-
posed for the Rpi activity would involve the 6-carbon HEI depicted
in Chart 1.
Two unrelated forms of the enzyme, with no amino acid se-
quence similarity and different structures, are known to catalyze
the isomerization. RpiA is the most common form and is found
While a number of strong competitive inhibitors designed as
mimics of the HEI proposed to be involved in the isomerization
of the 5-carbon sugars R5P and Ru5P have been reported in the
literature, such as 4-phospho-
D
-erythronate,8,9 4-deoxy-4-
and 4-phospho- -ery-
phosphonomethyl-
D
-erythronate,10
D
* Corresponding author. Tel.: +33 1 69 15 63 11; fax: +33 1 69 15 72 81.
E-mail address: laurent.salmon@u-psud.fr (L. Salmon).
Present address: Structural Genomics Consortium, Oxford University, Old Road
Campus Research Building, Roosevelt Drive, Oxford, OX3 7DQ, UK.
thronohydroxamic acid,9 to date no corresponding inhibitors based
on the All6P to Allu6P isomerization reaction have been kinetically
evaluated. Competitive inhibitors of Allpi activity would be
0008-6215/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2009.02.024

870
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
CHO
OH
O-
OH
CH2OH
O
1
2
2-O3PO
O
OH
OH
OH
OH
OH
OH
OH
OH
HO
2-
2-
2-
CH2OPO3
CH2OPO3
CH2OPO3
R5P
Ru5P
1,2-cis-enediolate
high-energy intermediates
2-
CHO
OH
O-
CH2OH
O
OPO3
O
1
2
2-O3PO
OH
OH
O
OH
OH
OH
OH
OH
OH
HO
OH
OH
OH
OH
OH
HO
OH
HO
OH
All6P
2-
2-
2-
CH2OPO3
CH2OPO3
CH2OPO3
Allu6P
Chart 1. Isomerization reactions of
D-ribose 5-phosphate (R5P) to D-ribulose 5-phosphate (Ru5P) and D-allose 6-phosphate (All6P) to D-allulose 6-phosphate (Allu6P).
important tools for future structural and mechanistic studies, as
well as promising compounds for the design of species-specific
therapeutic agents. Indeed, Allpi activity was suggested to be a
characteristic of certain type B Rpis,11,12 which are mainly found
in bacteria, but not of type A Rpis (like human Rpi), which display
only Rpi activity. Thus, specific inhibition of enzymes possessing
Allpi activity while leaving unaffected those that have only Rpi
activity could lead to a new type of highly selective antibiotics.
For these reasons, we have actively sought inhibitors that are able
to distinguish between enzymes catalyzing the two types of
reactions.
A knockout study in M. tuberculosis has not yet been per-
formed, but some evaluation of rpiB essentiallity is still possible.6
Using transposon site hybridization experiments, Sassetti et al.
identified rpiB as being essential for optimal growth of M. tubercu-
losis under some, but not all, circumstances.13 Further, an Rpi of
either type A or B (or both) is present in every complete genome,
indicating that the activity is likely to play an important role in an
organism’s ability to maintain the correct balance of different
nutrients in the cell. This is supported by a recent medical study
that revealed that a human deficiency in RpiA causes destruction
of myelin sheaths (leukoencephalopathy), which in turn leads to
extensive brain abnormalities.14 In the pathogens Trypanosoma
cruzi and M. tuberculosis, the fact that an RpiB is the only Rpi pres-
ent, and that such enzymes have no homologues in upper eukary-
otic organisms, opens up the possibility of considering them as
potential targets for the treatment of Chagas disease and
tuberculosis.15
In this study, we report the synthesis of five new potential Rpi
inhibitors, and an improved preparation of two previously de-
scribed (one potential and one known) Rpi inhibitors, designed
as analogues of the HEI postulated for the mechanism of the All6P
to Allu6P isomerization (Chart 2). We also report a kinetic evalua-
tion of these compounds on the Allpi activity of EcRpiB, as well as
on the Rpi activities of both EcRpiB and MtRpiB. One of them ap-
pears as the strongest Rpi inhibitor reported to date.
OH
NH2
O
NH
O
O-
O
NH2
O
NH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
2-
2-
2-
2-
CH2OPO3
CH2OPO3
CH2OPO3
CH2OPO3
1
2
3
4
5PRH
5PRA
5PRAm
5PRHz
CH3
NH
H
N
H
N
O
O
COO-
O
COO-
OH
OH
OH
OH
OH
OH
OH
OH
OH
2-
2-
2-
CH2OPO3
CH2OPO3
CH2OPO3
5
6
7
5PRMA
5PRGly
5PRGABA
Chart 2. Structure of the inhibitors evaluated in this study: 5-phospho-
(5PRAm, 3), N-(5-phospho- -ribonoyl)-hydrazine (5PRHz, 4), N-(5-phospho-
phospho- -ribonoyl)- -aminobutanoate (5PRGABA, 7).
D
-ribonohydroxamic acid (5PRH, 1), 5-phospho-
D-ribonoyl)-methylamine (5PRMA, 5), N-(5-phospho-D-ribonoyl)-glycine (5PRGly, 6), N-(5-
D
-ribonate (5PRA, 2), 5-phospho-D-ribonamide
D
D
c

S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
871
HO
HO
TrO
HO
TrO
BzO
HO
a
c
O
O
b
O
O
O
O
O
O
BzO
97%
96%
94%
HO
O
HO
BzO
O
O
BzO
O
8
9
10
OPO3(NH4)2
OPO(OPh)2
OPO(OH)2
OPO3Na2
O
d
e
f
g
O
O
O
O
O
O
O
HO
HO
BzO
O
95%
90%
66%
100%
OH
OH
O
OH
BzO
O
O
HO
O
11
12
13
14
All6P
O
Scheme 1. Synthesis of D-allose 6-phosphate, disodium salt (All6P, 14) using diphenylphosphochloridate as the phosphorylating agent. Reagents and conditions: (a) TrCl,
DMAP, pyridine, 60 °C, 12 h; (b) BzCl, pyridine, rt, 12 h; (c) H2 (30 bar)/Pd–C (10%), CH2Cl2/MeOH (4/1), 5 days; (d) (PhO)2POCl, pyridine, rt, 5 h; (e) H2 (4 bar)/PtO2, MeOH, rt,
4 days; (f) NH3/H2O, MeOH, rt, 12 h; (g) 1-TFA, H2O, rt, 2 h, 2-DowexÒ-50WX8 (H+), 3-DowexÒ-50WX8 (Na+).
OPO(OBn)2
OPO(OH)2
OPO3(NH4)2
HO
a
b
c
O
O
O
O
O
O
O
O
BzO
BzO
BzO
HO
61%
95%
75%
O
O
O
O
BzO
BzO
BzO
HO
10
15
16
13
Scheme 2. Synthesis of 1,2-O-isopropylidene-a-D-allofuranose 6-phosphate, diammonium salt (13) using dibenzyl N,N-diisopropylphosphoramidite as the phosphorylating
agent. Reagents and conditions: (a) 1-(BnO)2P–N[CH(CH3)2]2, 1,2,4-triazole, CH3CN, rt, 12 h; 2-tBuOOH, CH2Cl2, rt, 2 h; (b) H2 (10 bar)/Pd–C (10%), MeOH, rt, 4 h; (c) NH3/H2O,
MeOH, rt, 12 h.
calibration of the thiobarbituric acid (TBA) solution (see Section
4), as well as for the kinetic studies. Synthesis of All6P (or Allu6P)
provided the starting compounds needed for a straightforward
2. Results and discussion
As depicted in Chart 2, 5-phospho-D-ribonohydroxamic acid
(5PRH, 1),16 which can be considered as a weak acid with a gener-
ally accepted pKa in the range of 9–9.5 for hydroxamic acids, and 5-
synthesis of the new potential Allpi inhibitor, 5-phospho-
ate (5PRA, 2), which, upon cyclization in acidic medium, led to
the original synthon 5-dihydrogenophospho- -ribono-1,4-lactone.
D-ribon-
phospho-
anionic 1,2-cis-enediolate HEI. 5-Phospho-
3) and N-(5-phospho- -ribonoyl)-hydrazine (5PRHz, 4), which
can be considered as the respective neutral analogues of 2 and 1,
as well as N-(5-phospho- -ribonoyl)-methylamine (5PRMA, 5),
were designed as mimics of the neutral 1,2-cis-enediol HEI. Finally,
N-(5-phospho- -ribonoyl)-glycine (5PRGly, 6) and N-(5-phospho-
-ribonoyl)-4-aminobutanoate (5PRGABA, 7) were designed as
D
-ribonate (5PRA, 2)8,17 were designed as mimics of the
D
Indeed, we had previously reported the successful synthesis of sev-
eral new and strong PGI inhibitors from the synthon 5-dihydro-
D
-ribonamide (5PRAm,
D
genophospho-
corresponding aldonate 5-phospho-
D
-arabinono-1,4-lactone,18 easily obtained from its
D
-arabinonate.19 Similarly, the
D
reaction of defined nucleophilic amines with the synthon here al-
lowed us to achieve the one-step and almost quantitative synthe-
ses of new potential Allpi inhibitors 1 and 3–7, as depicted in
Scheme 4.
D
D
new types of amino acid derivatives of phosphorylated monosac-
charides valuable for studying the influence of the size of the upper
moiety of Allpi inhibitors. With the exception of 2, the inhibitors
depicted in Chart 2 were all synthesized from a common synthon,
itself obtained from All6P or Allu6P (2 was directly obtained from
All6P or Allu6P).
2.1. Synthesis of enzyme substrates
D
-Allose 6-phosphate (All6P) was obtained in seven steps from
1,2-O-isopropylidene-
a-D
-allofuranose20,21 with an overall yield
of 50% (Scheme 1). Alternative chemical and enzymatic syntheses
of All6P have been reported earlier starting from much less afford-
The synthesis of All6P is depicted in Schemes 1 and 2. Allu6P,
the synthesis of which is depicted in Scheme 3, was used for the
(HO)2OPO
(BnO)2OPO
Na2O3PO
c
HO
O
O
O
OH
O
O
O
O
a
b
OH
O
O
O
76%
95%
98%
HO
OH
O
O
O
O
O
O
17
18
19
Allu6P
Scheme 3. Synthesis of D-allulose (or D-psicose) 6-phosphate, disodium salt (Allu6P, 19). Reagents and conditions: (a) 1-(BnO)2P–N[CH(CH3)2]2, 1,2,4-triazole, CH3CN, rt,
12 h; 2-tBuOOH, CH2Cl2, rt, 1 h; (b) H2 (10 bar)/Pd–C (10%), MeOH, rt, 4 h; (c) 1-TFA, H2O, rt, 2 h, 2-DowexÒ-50WX8 (H+), 3-DowexÒ-50WX8 (Na+).

872
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
CONHOH
OH
CONH2
OH
OH
OH
OH
OH
2-
2-
CH2OPO3
CH2OPO3
CONHNH2
OH
1
3
2-
OPO3
5PRAm
5PRH
OH
OH
O
OH
e
d
OH
a
f
94%
100%
2-
CH2OPO3
91%
43%
HO
OH
COO-
OH
OPO3H2
O
4
19
c
5PRHz
O
Allu6P
OH
86%
OH
CONHCH3
OH
HO
OH
2-
2-
OPO3
CH2OPO3
20
5PRL
b
g
O
OH
2
HO
100%
28%
5PRA
OH
OH
98%
97%
OH
2-
i
h
CH2OPO3
OH
14
All6P
CONH(CH2)3COO-
CONHCH2COO-
5
5PRMA
OH
OH
OH
OH
OH
OH
2-
2-
CH2OPO3
CH2OPO3
7
6
5PRGly
5PRGABA
Scheme 4. Synthesis of 5-phospho-
hydrazine (5PRHz, 4), N-(5-phospho-
(5PRGABA, 7), and 5-dihydrogenophospho-
NaOH 0.5 M, rt, 48 h; (b) O2, NaOH 0.5 M, rt, 72 h; (c) 1-DowexÒ-50WX8 (H+), 2-lyophilization; (d) NH2OH/H2O, rt, 15 min; (e) NH3/H2O, rt, 2 h; (f) H2NNH2/H2O, rt, 3 h; (g)
D
-ribonohydroxamic acid (5PRH, 1), 5-phospho-
-ribonoyl)-methylamine (5PRMA, 5), N-(5-phospho-
-ribono-1,4-lactone (5PRL, 27) from -allulose (or
D
-ribonate (5PRA, 2), 5-phospho-
D
-ribonamide (5PRAm, 3), N-(5-phospho-D-ribonoyl)-
D
D-ribonoyl)-glycine (5PRGly, 6), N-(5-phospho-
D-ribonoyl)-4-aminobutanoate
D
D
D-psicose) 6-phosphate (Allu6P, 19) or -allose 6-phosphate (All6P, 14): (a) O2,
D
CH3NH2, rt, 12 h; (h) glycine, MeONa/MeOH, reflux, 10 min, then rt, 2 h; (i)
c
-aminobutyric acid, reflux, 10 min, then rt, 2 h.
able
D
-ribose 5-phosphate22 and
D
-allose,23 respectively. The par-
D
-Allulose 6-phosphate (Allu6P), also called
phate, was prepared in three steps from 1,2:3,4-di-O-isopropyli-
dene-
-psicofuranose28–30 with an overall yield of 71% (Scheme
D-psicose 6-phos-
tially protected 1,2-O-isopropylidene-
a-
D-allofuranose, synthe-
sized in three steps from 1,2:5,6-di-O-isopropylidene-
a
-D
-
a-D
glucofuranose according to literature procedures,20,21,24–26 allowed
3). The synthesis was based on the strategy initially developed
by Herve du Penhoat and Perlin,31 with the exception of the phos-
phorylation step. The starting compound was obtained in four
steps from
phinylation of the primary hydroxyl group by dibenzyl N,N-dii-
sopropylphosphoramidite and oxidation by tert-
selective tritylation of the primary hydroxyl group to obtain 8 in
94% yield. The fully orthogonally protected D-allose intermediate
9 was obtained almost quantitatively through benzoylation of
the secondary hydroxyl groups. The primary alcohol 10 was then
easily formed by hydrogenolysis which allowed its phosphoryla-
tion by diphenylphosphochloridate in pyridine to furnish the fully
D
-fructose according to literature procedures.28–37 Phos-
butylhydroperoxide27 yielded the new phosphorylated psicofura-
nose 17 (76%). Although direct introduction of the phosphate
group using diphenylphosphochloridate gave us the expected
product as described,31 the subsequent hydrogenolysis step over
PtO2 that would have led to 18 repeatedly did not work. The dihy-
drogenophosphate 18 was instead obtained quantitatively through
hydrogenolysis of 17 over Pd/C 10%. Final hydrolysis of the isopro-
pylidene groups with trifluoroacetic acid and ion-exchange chro-
protected D-allose 6-phosphate derivative 11 in 91% yield (two
steps). Compound 12 was obtained in 90% yield by hydrogenation
over PtO2, which not only deprotected the phosphate group, but
also reduced benzoyl to cyclohexanoyl groups, as we previously re-
ported for 2,3-di-O-benzoyl-5-diphenyphospho-D-ribono-1,4-lac-
tone.16 Removal of the cyclohexanoyl groups by aqueous
ammonia gave the side product cyclohexylamide and the ammo-
nium salt of 1,2-O-isopropylidene-6-phospho-
(66% yield); the latter was quite hygroscopic. Acid hydrolysis of
the remaining isopropylidene group followed by ion-exchange
D
-allofuranose 13
matography gave
D
-allulofuranose (or
D-psicofuranose) 6-
phosphate 19 as the disodium salt31 in almost quantitative yield
(98%).
chromatography gave D-allose 6-phosphate 14 as the disodium salt
in quantitative yield. As an alternative to the use of the platinium
catalyst described above, we also achieved phosphorylation of
compound 10 by dibenzyl N,N-diisopropylphosphoramidite and
subsequent oxidation by tert-butylhydroperoxide27 to obtain the
2.2. Synthesis of substrate-derived enzyme inhibitors
All6P and Allu6P, which have the same absolute configuration
at carbon atoms C-3, C-4, and C-5 as compounds 1–7 have at car-
bon atoms C-2, C-3, and C-4, respectively, were considered as
two possible starting compounds (Scheme 4). The synthesis of 5-
phospho-D-ribonate 2 was achieved by oxidative cleavage of All6P
14 in 28% yield or from Allu6P 19 in 43% yield with molecular diox-
phosphorylated
D-allofuranose intermediate 15 in 61% yield
(Scheme 2). Deprotection of the phosphate group through hydro-
genation over Pd/C yielded the dihydrogenophosphate intermedi-
ate 16 (95%), which reacted with aqueous ammonia as above to
give 13 in 75% yield and the side product benzylamide.
ygen and NaOH (Spengler–Pfannenstiel conditions), following the

S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
873
Table 1
strategy reported for the synthesis of 5-phospho-
D
-arabinonate
from D D-fructose 6-phos-
-glucose 6-phosphate (G6P)38 or from
Inhibition parameters (IC50 and Ki values) of compounds 1–7 for the All6P to Allu6P
isomerization (Allpi activity) catalyzed by EcRpiB, and the R5P to Ru5P isomerization
(Rpi activity) catalyzed by EcRpiB and MtRpiB
phate (F6P).19 For reasons which remain unclear to us, the reaction
appears less efficient on All6P or Allu6P than on G6P or F6P (where
Inhibitor
Enzyme inhibition
EcRpiB Allpia
EcRpiB Rpib
MtRpiB Rpic,d
78% yields were obtained). 5-Dihydrogenophospho-D-ribono-1,4-
5PRH (1)
IC50
Ki
0.62 0.05
0.43 0.04
0.17 0.02
0.09 0.02
0.44 0.02
lactone (5PRL, 20) was then obtained in 86% yield by freeze-drying
of an aqueous solution of the acidic form of 2. Although formation
of 20 was mentioned in the kinetic study of the oxidation of R5P by
chromium(VI)39 or vanadium(V)40 in perchloric acid media, we re-
port here the first synthesis of this important phosphorylated su-
gar. Indeed, treatment of synthon 20, which can be considered as
an activated ester, with the nucleophilic amine derivatives hydrox-
ylamine, ammonia, hydrazine, methylamine, glycine, and 4-
0.26 0.01e
5PRA (2)
IC50
Ki
1.33 0.09
1.31 0.06
n.d.
0.031 0.002
0.009 0.001
n.d.f
5PRAm (3)
5PRHz (4)
5PRMA (5)
5PRGly (6)
5PRGABA (7)
IC50
Ki
1.8 0.3g
n.d.
0.20 0.02
0.07 0.05
0.22 0.02
0.04 0.01
IC50
Ki
1.9 0.3
n.d.
n.d.
n.d.
1.0 0.1
n.d.
IC50
Ki
0.35 0.01
n.d.
n.i.i
n.d.
0.36 0.02
0.18 0.03
0.19 0.01
0.11 0.02
aminobutyric acid easily and efficiently gave 5-phospho-
nohydroxamic acid (5PRH, 1), 5-phospho- -ribonamide (5PRAm,
3), N-(5-phospho- -ribonoyl)-hydrazine (5PRHz, 4), N-(5-phos-
pho- -ribonoyl)-methylamine (5PRMA, 5), N-(5-phospho- -ribo-
noyl)-glycine (5PRGly, 6), and N-(5-phospho- -ribonoyl)-4-
D-ribo-
D
IC50
Ki
9
1
0.57 0.04
0.34 0.01
D
n.d.
D
D
IC50
Ki
1.6 0.3
n.d.
2.0 0.2
n.d.
1.42 0.03
n.d.
D
aminobutanoate (5PRGABA, 7), respectively, with yields ranging
from 91% to quantitative (Scheme 4). With the exception of the
hydroxamic acid 1, for which we had previously reported a differ-
ent synthesis,16 all the products obtained are original compounds
and were fully characterized.
IC50 values (in italic) and Ki values are given in mM.
a
Km (All6P) = 2.9 0.8 mM (lit.7: Km (All6P) = 0.5 0.2 mM).
b
Km (R5P) = 1.5 0.4 mM (lit.7: Km (R5P) = 1.1 0.2 mM; lit.47
:
Km
(R5P) = 1.23 mM).
c
Km (R5P) = 3.1 0.7 mM (lit.5: Km (R5P) = 3.7 mM; lit.7: Km (R5P) = 1.0 0.4
mM).
d
e
f
2.3. Kinetic evaluation of substrate-derived enzyme inhibitors
IC50 (All6P) = 2.0 0.4 mM; IC50 (Allu6P) = 6.3 0.5 mM.
Lit.16: Ki = 0.40 0.04 mM.
n.d.: not determined.
Estimated value.
n.i.: no inhibition.
Substrate-derived compounds 1–7 were tested for inhibition of
the All6P to Allu6P isomerization (Allpi activity) catalyzed by
EcRpiB, and of the R5P to Ru5P isomerization (Rpi activity) cata-
lyzed by both EcRpiB and MtRpiB. Inhibition constants (Ki) and/
or IC50 values obtained are reported in Table 1.
g
i
had initially expected, these Ki values show that the design of
inhibitors as mimics of the 6-carbon HEI postulated for Allpi activ-
ity is appropriate for inhibition of EcRpiB.
Kinetic evaluation on Allpi activity of EcRpiB shows that two of
these compounds, 5-phospho-D-ribonohydroxamic acid (5PRH, 1)
and N-(5-phospho- -ribonoyl)-methylamine (5PRMA, 5), are new
D
In addition to EcRpiB Allpi and Rpi activities, inhibition proper-
ties of compounds 1–7 were evaluated on the Rpi activity of
MtRpiB (Table 1), a more interesting enzyme from a therapeutic
point of view, and for which we previously demonstrated that no
Allpi activity exists.7 Indeed, while most of compounds 1–7 were
originally designed as mimics of the 6-carbon HEI postulated for
Allpi activity, not for Rpi activity, several of them behave as new
good competitive inhibitors of MtRpiB, such as 5PRH (1), 5PRMA
(5), and even the rather long 5PRGly (6), with respective Ki values
of 0.26, 0.11, and 0.34 mM (to be compared to the Km value of
3.1 mM under the conditions used in this study). More interest-
ingly, the two shortest of the evaluated compounds, namely 5-
efficient inhibitors of the enzyme, with IC50 values of 0.62 and
0.35 mM, respectively. Because the TBA assay is not optimized
for Ki determination, only 5PRH (1) was tested further; this com-
pound displayed competitive inhibition, as shown by the Linewe-
aver–Burk plots depicted in Figure 1A, which gave a Ki value of
0.43 mM (Fig. 1B). In comparison to the Km value of 2.9 mM we
determined in this study for All6P isomerization, 5PRH (1) appears
as a relatively good HEI analogue inhibitor of EcRpiB Allpi activity
with a Km/Ki ratio of about 7. As expected, shorter (2 and 3) and
longer (6 and 7) HEI analogues gave higher IC50 values and thus be-
have as weak inhibitors (except 6 which does not inhibit EcRpiB
Allpi activity). Although closely related to 5PRH (1) in structure,
5PRHz (4) is also a weak inhibitor of EcRpiB Allpi activity. Indeed,
as discussed in the case of other aldose–ketose isomerases such as
PGI,18 this observation would be in accordance with the rather an-
ionic nature of the HEI thought to be involved in the rate-deter-
mining step of the isomerization reaction of All6P catalyzed by
EcRpiB.
Considering inhibition of EcRpiB Rpi activity by compounds 1–
7, Table 1 shows that, most often, close if not identical IC50 and/or
Ki values were obtained (slight differences observed in some cases
appear to be due to instability of EcRpiB when diluted). As Ki is the
simple dissociation constant for the enzyme–inhibitor complex,
and the same active site of EcRpiB acts on both 5- and 6-carbon lin-
ear substrates,7 we indeed expect the Ki values to be the same for
both of the activities. The Ki values we report for 5PRH (1), 5PRAm
(3), and 5PRMA (5), 0.09, 0.07, and 0.18 mM, respectively, are
among the best values ever reported against EcRpiB (Hamada
phospho-D-ribonate (5PRA, 2) and 5-phospho-D-ribonamide
(5PRAm, 3), behave as new strong competitive inhibitors of MtRpiB
with Ki values of 0.009 (Fig. 1C and D) and 0.04 mM, respectively,
which gives corresponding Km/Ki ratios of 344 and 77. To our
knowledge, 5PRA (2) is the strongest competitive inhibitor of an
Rpi ever reported in the literature. It should be noted that, from
a theoretical point of view, the two inhibitors 5PRA (2) and 5PRAm
(3) are slightly too long to act as mimics of the 5-carbon HEI pos-
tulated for the R5P isomerization reaction, having an additional
oxygen atom and NH2 group, respectively. Nevertheless, they
prove to have favorable interactions in the MtRpiB active site that
explain the observed strong stabilization of these ‘imperfect’ HEI
analogues. We recently reported the high-resolution 3D crystal
structures of a MtRpiB-5PRA complex [PDB (The Protein Data
Bank:42 www.rcsb.com) ID code: 2VVQ], which was obtained by
fortuitious hydrolysis of 5PRH (1) under the crystallization condi-
tions. This structure clearly shows the unexpected conformation
of the upper moiety of the inhibitor in the enzyme active site
(Fig. 2A) when compared to the structures of the enzyme com-
et al. reported a value of about 0.07 mM using D-glucose 6-phos-
phate41). In comparison to the Km value of 1.5 mM we measured
for R5P, the three new EcRpiB inhibitors have 10–20 times more
affinity for EcRpiB than the substrate. Although not as low as we
plexed to linear R5P/Ru5P (PDB ID code: 2VVP) or 4-phospho-
D-
erythronohydroxamic acid (PDB ID code: 2BES).7

874
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
Figure 1. (A) Inhibition of EcRpiB (50 mM TrisꢀHCl buffer, 37 °C, pH 7.5) by 5-phospho-
versus All6P concentration obtained at various concentrations of inhibitor: d, no inhibitor; h, [I] = 100
Km/Vmax values (slopes of the straight lines in the primary graph A) versus inhibitor concentration. (C) Inhibition of MtRpiB (50 mM TrisꢀHCl buffer, 37 °C, pH 7.5) by 5-
phospho- -ribonate (5PRA, 2); double reciprocal plot of initial reaction velocity versus R5P concentration obtained at various concentrations of inhibitor: d, no inhibitor; h,
[I] = 3 M; Ç, [I] = 5 M; s, [I] = 8 M; j, [I] = 10 M. (D) Secondary plot of apparent Km/Vmax values (slopes of the straight lines in the primary graph C) versus inhibitor
D-ribonohydroxamic acid (5PRH, 1); double reciprocal plot of initial reaction velocity
l
M; Ç, [I] = 200 M; s, [I] = 300 M. (B) Secondary plot of apparent
l
l
D
l
l
l
l
concentration. In graphs A and C, lines drawn were obtained from a linear regression fit of the observed data using Michaelis–Menten equation for competitive inhibition. In
graphs B and D, the x intercept, which is equal to ꢁKi, was determined by linear regression.
Interestingly, while 5PRAm (3) is also a rather strong inhibitor
of EcRpiB Rpi activity, with a Ki value of 0.07 mM, 5PRA (2) and
5PRGly (6) are quite weak inhibitors of the Rpi activity, with IC50
values of 1.31 and 9 mM, respectively. The observed specific inhi-
bition of MtRpiB versus EcRpiB by 5PRA (2) and 5PRGly (6) clearly
points out the important differences between the two RpiBs active
sites in their stabilization of the HEI, confirming the initially pro-
posed existence of two sub-families of type B Rpis, based on their
sequences alone.5
The most striking difference between the active sites of M.
tuberculosis and E. coli enzymes is the nature of the catalytic base,
Glu75 and Cys66, respectively. Consideration of the active sites
helps to explain some of the kinetic results we obtained. Figure 2
shows the active site of MtRpiB complexed to 5PRA (Fig. 2A, actual
structure from 2VVQ) and to 5PRH (Fig. 2B, manually docked, using
the position of the phosphate group and carbon atoms 2–5 of the
inhibitor). The analysis clearly predicts steric hindrance for 5PRH
with an improbable H-bond of 1.5 Å between the N-bonded hydro-
xyl group and the backbone NH group of Ser71, to be compared to
5PRA which gives a much more satisfactory H-bond of 2.7 Å. This
structural comparison is in accordance with the respective Ki val-
ues of 0.26 and 0.009 mM we obtained. Figure 2 also shows the ac-
tive site of EcRpiB complexed to 5PRA (Fig. 2C) and 5PRH (Fig. 2D)
based on the partial ligand electron density seen in the EcRpiB-
5PRH structure (Roos et al., unpublished). The figure clearly shows
loss of two H-bonds when 5PRH is replaced by 5PRA. Again, this
structural comparison is in accordance with the IC50 values we
determined for the two compounds, 0.17 and 1.31 mM, respec-
tively. Other results were more difficult to link to particular aspects
of the protein structures. More structural work will be required to
determine how the differences arise among the compounds, and
between the two enzymes, particularly in the case of the E. coli en-
zyme, for which no well-defined inhibitor complexes are at present
available.
3. Conclusion
This study reports syntheses of All6P, Allu6P and subsequently,
Rpi inhibitors 1–7, designed as analogues of the 6-carbon HEI pos-
tulated for EcRpiB Allpi activity. Several analogues, notably 5PRH
(1), 5PRAm (3), and 5PRMA (5), appear as new efficient competitive
inhibitors of Allpi/Rpi activity of EcRpiB. In addition, compounds 1–
7 not only inhibit Allpi/Rpi activity of EcRpiB, but also Rpi activity
of MtRpiB. Therefore, it appears that specific inhibition of Allpi/Rpi
activity of EcRpiB versus Rpi activity of MtRpiB has not been

S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
875
Figure 2. (A) Representation of the active site of MtRpiB complexed to 5PRA (2) from the parent X-ray 3D crystal structure (PDB ID code 2VVQ). (B) Model of the active site of
MtRpiB complexed to 5PRH (1) obtained by manual docking based on the same structure (steric hindrance of the hydroxamic acid function is depicted as large circles). (C)
Representation of the active site of EcRpiB complexed to 5PRA (2) obtained by manual docking using an EcRpiB–5PRH structure (undeposited, in which the electron density
allows the placement of the phosphate of the ligand with some confidence, and the hydroxamic acid moiety with somewhat less precision; the OH groups of carbons 2 and 3
are not visible in the density). (D) Representation of the active site of EcRpiB complexed to 5PRH (1) from the same parent X-ray structure. Selected H-bonds are depicted as
small circles. Some active site residues, water molecules, and H-bonds were omitted for clarity of the figure.
reached. However, species-specific inhibition of MtRpiB versus
EcRpiB, which could be demonstrated for 5PRA (2) and 5PRGly
(6), highlights the importance of testing the inhibition properties
of 6- and 5-carbon HEI-analogue inhibitors on both Allpi and/or
Rpi activities of type A Rpis (notably human RpiA) and type B Rpis
of therapeutical interest, a subject that is now under investigation.
MeOH, CH2Cl2, MeCN, and acetone were dried by refluxing with,
respectively, KOH, Mg/I2, CaH2, P2O5, and CaSO4, then distilled. Un-
less used immediately, dried MeOH and pyridine were stored over
molecular sieves (respectively 0.3 and 0.4 nm) under argon. Col-
umn chromatography was performed using silica gel (70–
200
lm, E. Merck). Flash chromatography was performed using sil-
ica gel (35–70
l
m, E. Merck) under N2 pressure. Ion-exchange
chromatography was performed on DowexÒ-50WX8 cation-ex-
4. Experimental
change resin (H+ or Na+ form, 50–100 or 100–200 mesh, Acros),
eluting with purified water (Millipore, 18.2 MX). Unless otherwise
4.1. General materials and methods
stated, all organic extracts were dried over MgSO4 and filtered.
Concentration of solutions was performed under diminished pres-
sure at temperature <30 °C using a rotary evaporator. All air- and
moisture-sensitive reactions were performed under an atmosphere
Unless otherwise stated, all chemical reagents were of analyti-
cal grade, obtained from Acros or Aldrich, and used without further
purification. Solvents were obtained from VWR-Prolabo. Pyridine,

876
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
of argon. Analytical TLC was performed using Silica Gel 60 F254 pre-
coated aluminum plates (E. Merck). Spots were visualized by treat-
ment with 5% ethanolic H2SO4 followed by heating and/or by
absorbance of UV light at 254 nm. Melting points were measured
in open capillaries on a Büchi apparatus and are uncorrected. Opti-
cal rotations were measured on a Jasco DIP-370 digital polarimeter
at 589 nm. NMR spectra were recorded at 297 K in CDCl3, CD3OD,
and D2O with a Bruker DPX 250 (1H at 250.13 MHz, 13C at
62.90 MHz, and 31P at 101.26 MHz), DRX 300 (1H at 300.13 MHz
and 13C at 75.47 MHz), or Avance 360 (1H at 360.13 MHz and 13C
at 90.56 MHz) spectrometer using NMR Notebook 2.0 software.
Chemical shifts are reported in ppm (d) and coupling constants
in hertz (Jij). 1H NMR spectra were referenced to internal residual
CHCl3 (d 7.26), CD2HOD (d 3.31), and HOD (d 4.80) for solns in
CDCl3, CD3OD, and D2O, respectively. For CH2 groups, H-10, H-50,
and H-60 resonances appear arbitrarily at higher field than H-1,
H-5, and H-6 resonances, respectively. 13C NMR spectra were refer-
enced to solvent for solns in CDCl3 (d 77.0) and CD3OD (d 49.1), and
to dioxane (d 67.4) for solns in D2O. 31P NMR spectra were refer-
enced externally to 85% aq H3PO4 (d 0.0). In most cases, COSY,
HSQC, J-mod, and/or DEPT135 NMR spectra were recorded for
assigning resonances. Infrared spectra were recorded with a FT-
IR Bruker IFS-66 spectrometer. Low-resolution mass spectrometry
(MS) and high-resolution mass spectrometry (HRMS) analyses
were performed by the ‘Service de Spectrométrie de Masse’
(ICMMO, Orsay) using electrospray with positive ionization mode
(ESI+) for non-ionic compounds or negative ionization mode (ESIꢁ)
for anionic compounds. Elemental analyses were performed by the
‘Service de Microanalyse’ (ICSN-CNRS, Gif-sur-Yvette).
was dissolved in a minimum vol of water with 1 g of DowexÒ-
50WX8 (H+ form) resin. The product was then successively eluted
with water on two DowexÒ-50WX8 ion-exchange columns (1-H+
form, 2-Na+ form) and lyophilized to afford 2 (220 mg, 43%) as a
white powder. (B) The corresponding procedure was applied start-
ing from D-allopyranose 6-phosphate 14 (1 g, 3.3 mmol) to afford 2
(270 mg, 26%): ½a D28
ꢂ
ꢁ 2:8 (c 1, water); 1H NMR (250 MHz, D2O): d
4.16 (d, 1H, J2,3 2.3 Hz, H-2), 4.10–3.80 (m, 4H, H-3, H-4, H-5, H-50);
13C NMR (62.9 MHz, D2O): d 178.1 (C-1), 73.2 (C-3), 72.5 (C-2), 70.0
(d, JC,P 7.1 Hz, C-4), 66.4 (d, JC,P 4.3 Hz, C-5), in relatively good
agreement with the data reported for the dilithium salt;17 31P
NMR (101.2 MHz, D2O): d 1.30; ESIꢁMS: m/z 288.9 (19) [MꢁNa]ꢁ,
266.9 (100) [Mꢁ2Na+H]ꢁ.
4.2.3. 5-Phospho-D-ribonamide, disodium salt (3)
Compound 20 (105 mg, 0.46 mmol) was dissolved in 30% aq
ammonia (3 mL, 2.3 mmol) and stirred at rt for 14 h. Following
evaporation under diminished pressure, the bis(ammonium) salt
of the title compound was eluted with water through a DowexÒ-
50WX8 (Na+ form) ion-exchange column. Lyophilization afforded
the disodium salt 3 (140 mg, 100%) as a white solid: 1H NMR
(250 MHz, D2O): d 4.39 (d, 1H, J2,3 3.3 Hz, H-2), 4.03–3.88 (m, 6H,
H-3, H-4, H-5, H-50, NH2); 13C NMR (62.9 MHz, D2O): d 177.2 (C-
1), 72.2 (C-2, C-3), 70.1 (d, JC,P 5.8 Hz, C-4), 65.1 (d, JC,P 4.5 Hz, C-
5); 31P NMR (101.2 MHz, D2O): d 4.3; ESIꢁMS: m/z 244.0 (100)
[Mꢁ2Na+H]ꢁ; HRESIꢁMS: calcd for C5H11NO8P: 244.0228
[Mꢁ2Na+H]ꢁ. Found m/z 244.0227.
4.2.4. N-(5-Phospho-D-ribonoyl)-hydrazine, bis(hydrazinium)
salt (4)
4.2. Syntheses
Compound 20 (96 mg, 0.42 mmol) was dissolved in water
(1.5 mL) and the soln was cooled to 4 °C and stirred. Aq hydrazine
4.2.1. Preparation of 5-phospho-
D
-ribonohydroxamic acid,
(64%, 205 lL, 4.2 mmol) was added and the soln stirred for 15 min.
monohydroxylammonium salt (1)16
Water and excess hydrazine were removed under high vacuum.
The residue was lyophilized to afford the bis(hydrazinium) salt 4
(125 mg, 91%) as a very hygroscopic white powder: 1H NMR
(250 MHz, D2O): d 4.24 (d, 1H, J23 3.2 Hz, H-2), 3.83–3.70 (m, 5H,
H-3, H-4, H-5, H-50, NH); 13C NMR (62.9 MHz, D2O): d 172.4 (C-
1), 72.1 (C-3), 71.7 (C-2), 69.9 (d, JC,P 5.7 Hz, C-4), 64.8 (d, JC,P
3.7 Hz, C-5); 31P NMR (101.2 MHz, D2O): d 4.5; ESIꢁMS: m/z
259.0 (100) [Mꢁ2NH3NH2+H]ꢁ; HRESIꢁMS calcd for C5H12N2O8P:
259.0337 [Mꢁ2NH3NH2+H]ꢁ, found m/z 259.0348.
A soln of compound 20 (98 mg, 0.43 mmol) in water (1.5 mL)
was cooled to 4 °C. Aq NH2OH (50% w/w, 260 lL, 4.4 mmol) was
then added and the reaction mixture was stirred for 15 min. Fol-
lowing removal of excess hydroxylamine and water under high
vacuum, lyophilization gave 1 (119 mg, 94%) as a very hygroscopic
white powder: ½a D28
ꢂ
þ 3:3 (c 1, water); 1H NMR (250 MHz, D2O): d
4.10 (d, 1H, J2,3 3.7 Hz, H-2), 3.71–3.60 (m, 4H, H-3, H-4, H-5, H-50);
13C NMR (62.9 MHz, D2O): d 170.2 (C-1), 71.7 (C-3), 71.2 (C-2), 69.7
(d, JC,P 6.5 Hz, C-4), 65.0 (d, JC,P 2.9 Hz, C-5); 31P NMR (101.2 MHz,
D2O): d 3.7. All NMR data are in agreement with our previously re-
ported values for the bis(hydroxylammonium) salt;16 ESIꢁMS: m/z
260.0 (100) [MꢁNH3OH]ꢁ, lit.16 260.1; HRESIꢁMS: calcd for
C5H11NO9P: 260.0177 [MꢁNH3OH]ꢁ, found m/z 260.0173.
4.2.5. N-(5-Phospho-D-ribonoyl)-methylamine, disodium salt
(5)
Compound 20 (94 mg, 0.41 mmol) was dissolved in 40% aq
methylamine (2 mL, 2.3 mmol) and the soln was stirred at rt for
14 h. Evaporation, under diminished pressure, of water and excess
methylamine gave the bis(methylammonium) salt of the title com-
pound, which was thereafter eluted with water through a DowexÒ-
50WX8 (Na+ form) ion-exchange column. Lyophilization afforded
the disodium salt 5 (160 mg, 100%) as a white solid: 1H NMR
(250 MHz, D2O): d 4.19 (d, 1H, J2,3 3.8 Hz, H-2), 3.75 (m, 5H, H-3,
H-4, H-5, H-50, NH), 2.59 (s, 3H, CH3); 13C NMR (62.9 MHz, D2O):
d 174.9 (C-1), 72.6 (C-3), 72.2 (C-2), 70.1 (d, JC,P 5.2 Hz, C-4), 64.9
(d, JC,P 3.7 Hz, C-5), 25.4 (C-6, CH3); 31P NMR (101.2 MHz, D2O): d
4.8; ESIꢁMS: m/z 258.0 (100) [Mꢁ2Na+H]ꢁ; HRESIꢁMS: calcd for
C6H13NO8P: 258.0384 [Mꢁ2Na+H]ꢁ. Found: m/z 258.0382.
4.2.2. Preparation of 5-phospho-
D-ribonate, trisodium salt
(2)8,17
(A)
D-Allulofuranose 6-phosphate, disodium salt 19 (0.5 g,
1.65 mmol) was dissolved in 0.5 M aq NaOH (7.5 mL) and vigor-
ously stirred under oxygen (1 bar) for 72 h, during which time oxy-
gen consumption was 25 mL (1.12 mmol, 68%). Stirring for an
additional 24 h period did not increase oxygen consumption. The
reaction mixture was adjusted to pH 1.5 with concentrated HCl
(0.5 mL) and bubbled with N2. Adjustment to pH 5 was then
achieved with a satd aq Ba(OH)2 soln (5.5 mL) and the barium salt
of 5-phospho-D-ribonic acid precipitated by the addition of abs
EtOH (2 vols, 27 mL). The filtered barium salt was suspended in
water (3 mL) and the soln was adjusted again to pH 1.5 with concd
HCl, bubbled with N2, adjusted to pH 5 with satd aq Ba(OH)2
(3.5 mL), and precipated as before (13 mL abs EtOH). The filtered
barium salt was suspended again in water (3.5 mL) and a third pre-
cipitation was performed in the same manner, but with only a sin-
gle volume (7 mL) of abs EtOH. The filtered and dried barium salt
4.2.6. N-(5-Phospho-D-ribonoyl)-glycine, trisodium salt (6)
Compound 20 (100 mg, 0.43 mmol) was dissolved under argon
in anhyd MeOH (10 mL) and the soln was stirred. Glycine (330 mg,
4.4 mmol) and MeONa (140 mg, 2.6 mmol) were added and the
soln was stirred at reflux for 10 min, then at rt for 2 h. The solvent
was removed under diminished pressure and the residue was
eluted with water on a DowexÒ-50WX8 (Na+ form) ion-exchange

S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
877
column to furnish 6 (156 mg, 97%) as a white solid after lyophiliza-
tion: 1H NMR (360 MHz, D2O): d 4.33 (d, 1H, J2,3 2.4 Hz, H-2), 3.96–
26.7 and 26.6 (2CH3); ESI+MS: m/z 693.4 (100) [M+Na]+; HRESI+MS:
calcd for C42H38O8Na: 693.2459 [M+Na]+. Found m/z 693.2469.
3.85 (m, 5H, H-3, H-4, H-5, H-50, NH), 3.80 (d, 1H, J6,6 17.5 Hz, H-6),
0
3.68 (d, 1H, H-60); 13C NMR (75.0 MHz, D2O): d 178.6 (COOꢁ), 173.8
(C-1), 72.4 (C-3), 72.3 (C-2), 69.9 (d, JC,P 5.8 Hz, C-4), 65.4 (d, JC,P
4.2 Hz, C-5), 42.9 (CH2COOꢁ); 31P NMR (101.2 MHz, D2O): d 2.41.
4.2.10. 3,5-Di-O-benzoyl-1,2-O-isopropylidene-a-D-
allofuranose (10)
Compound 9 (2.04 g, 3.0 mmol) was dissolved in a mixture of
anhyd CH2Cl2 (22 mL) and MeOH (6 mL). Following addition of
Pd/C 10% (350 mg), the reaction mixture was stirred under H2
(30 bar) for 4 d. The soln was then filtered and concentrated under
diminished pressure. The residue was purified by column
chromatography using initially CH2Cl2 as eluent to remove triphe-
nylmethane, then 97:3 CH2Cl2–MeOH to furnish 10 (1.25 g, 96%) as
4.2.7. N-(5-Phospho-D-ribonoyl)-c-aminobutanoate, trisodium
salt (7)
Compound 20 (98 mg, 0.43 mmol) was dissolved under argon in
anhyd MeOH (10 mL) and the soln was stirred. 4-Aminobutanoic
acid (450 mg, 4.3 mmol) and MeONa (145 mg, 2.6 mmol) were
added and the soln was stirred at reflux for 10 min, then at rt for
2 h. The solvent was removed under diminished pressure and the
residue was eluted with water on a DowexÒ-50WX8 (Na+ form)
ion-exchange column, then lyophilized to give 7 (170 mg, 98%) as
a white solid: 1H NMR (300 MHz, D2O): d 4.33 (d, 1H, J2,3 3.2 Hz,
H-2), 3.98–3.90 (m, 5H, H-3, H-4, H-5, H-50, NH), 3.19 (m ,2H,
NHCH2), 2.19 (m, 2H, CH2COOꢁ), 1.74 (m, 2H, NHCH2CH2). 13C
NMR (75.0 MHz, D2O): d 182.7 (COOꢁ), 173.9 (C-1), 72.5 (C-3),
72.1 (C-2), 69.9 (d, JC,P 6.1 Hz, C-4), 65.5 (d, JC,P 3.8 Hz, C-5), 38.8
(NHCH2), 34.7 (CH2COOꢁ), 25.3 (NHCH2CH2); ESIꢁMS: m/z 330.1
(100) [Mꢁ3Na+2H]; HRESIꢁMS: calcd for C9H17NO10P: 330.0596
[Mꢁ3Na+2H], found: m/z 330.0598.
a white crystalline powder (very hygroscopic): ½a D27
þ 105:3 (c 1,
ꢂ
CH2Cl2); 1H NMR (250 MHz, CDCl3): d 7.95–7.91 (m, 4H, CHm
Ph), 7.48–7.42 (m, 2H, CHp Ph), 7.34–7.22 (m, 4H, CHo Ph), 5.88
0
(d, 1H, J1,2 3.8 Hz, H-1), 5.43 (ddd, 1H, J5,6 4.0, J5,6 5.4, J5,4 6.0 Hz,
H-5), 5.17 (dd, 1H, J3,2 5.0, J3,4 8.9 Hz, H-3), 4.99 (dd, 1H, H-2),
0
4.65 (dd, 1H, H-4), 4.02 (dd, 1H, J6,6 12.2 Hz, H-6), 3.96 (dd, 1H,
H-60), 1.57, 1.32 (2s, 6H, 2CH3); 13C NMR (62.9 MHz, CDCl3): d
166.0 and 165.5 (2PhC@O), 133.1 and 133.0 (2Cp Ph), 129.6 (4Cm
Ph), 129.3 and 128.8 (2Cipso Ph), 128.2 and 128.1 (4Co Ph), 113.3
(CMe2), 104.2 (C-1), 77.6 (C-2), 76.0 (C-4), 74.4 (C-3), 74.2 (C-5),
61.8 (C-6), 26.6 and 26.5 (2CH3); ESI+MS: m/z 451.1 (100)
[M+Na]+; HRESI+MS: calcd for C23H24O8Na: 451.1363 [M+Na]+.
Found m/z 451.1373. Anal. Calcd for C23H24O8: C, 64.48; H, 5.65;
O, 29.88. Found: C, 63.84; H, 5.78; O, 28.95.
4.2.8. 1,2-O-Isopropylidene-6-O-trityl-
a-D-allofuranose (8)
1,2-O-Isopropylidene-
a-D
-allofuranose21 (3.03 g, 13.7 mmol),
4.2.11. 3,5-Di-O-benzoyl-1,2-O-isopropylidene-a-D-
trityl chloride (4.21 g, 15.1 mmol), and DMAP (0.42 g, 3.43 mmol)
were dissolved under argon in pyridine (6 mL) at rt. The reaction
mixture was stirred at 60 °C for 12 h and concentrated. The crude
product was purified by column chromatography (1:1 pentane–
EtOAc) to give 8 (5.97 g, 94%) as a white powder: mp 64 °C;
allofuranose 6-diphenylphosphate (11)
Compound 10 (1.15 g, 2.68 mmol) was dissolved in anhyd pyri-
dine under argon. Diphenylphosphate chloride was added and the
reaction mixture was stirred at rt for 5 h. The soln was concentrated,
water (50 mL) was added to the residue, and the product was ex-
tracted with CH2Cl2 (3 ꢃ 50 mL). The organic phases were dried on
Na2SO4, filtered, and concentrated to afford pure 11 (1.68 g, 95%) as
½
a 2D8
ꢂ
þ 12:9 (c 1, CH2Cl2); Rf 0.48 (1:1 pentane–EtOAc); 1H NMR
(250 MHz, CDCl3): d 7.51–7.27 (m, 15H, 3Ph), 5.81(d, 1H, J1,2
a white powder: mp 115 °C; ½a D27
þ 82:4 (c 1, CH2Cl2); Rf 0.51 (2:1
ꢂ
3.8 Hz, H-1), 4.65 (dd, 1H, J2,3 5.1 Hz, H-2), 4.15 (m, 1H, J5,6 3.8,
pentane–EtOAc); 1H NMR (250 MHz, CDCl3): d 7.86 (d, 4H, CHm
PhC@O), 7.49–7.40 (m, 2H, CHp PhC@O), 7.30–7.15 (m, 14H, CHo
PhC@O, CH PhO), 5.91 (d, 1H, J1,2 3.8 Hz, H-1), 5.55 (m, 1H, H-5),
5.14 (dd, 1H, J3,2 4.9, J3,4 8.8 Hz, H-3), 5.02 (dd, 1H, H-2), 4.66 (m,
3H, H-4, H-6, H-60), 1.57, 1.35 (2s, 6H, 2CH3); 13C NMR (62.9 MHz,
CDCl3): d 164.9 and 164.8 (2PhC@O), 150.0 (d, JC,P 6.7 Hz, Cipso
PhO), 149.9 (d, JC,P 6.7 Hz, Cipso PhO), 132.8 (2Cp PhC@O), 129.4
and 129.3 (4Cm PhC@O), 129.3 (2Cp PhO), 128.5 and 128.3 (2Cipso
PhC@O), 127.8 and 127.7 (4Co PhC@O), 125.1 and 125.0 (4Cm
PhO), 119.6 (d, JC,P 5.3 Hz, 2Co PhO), 119.5 (d, JC,P 5.3 Hz, 2Co PhO),
113.0 (CMe2), 104.0 (C-1), 77.5 (C-2), 74.5 and 74.4 (C-3, C-4), 71.7
(d, JC,P 7.1 Hz, C-5), 66.8 (d, JC,P 4.6 Hz, C-6), 26.4 and 26.3 (2CH3);
31P NMR (101.2 MHz, CDCl3): d ꢁ12.1; ESI+MS: m/z 683.1 (100)
[M+Na]+; HRESI+MS: calcd for C35H33O11PNa: 683.1676 [M+Na]+,
found m/z 683.1662. Anal. Calcd for C35H33O11P: C, 63.63; H, 5.04;
P, 4.69. Found: C, 63.93; H, 5.08; P, 4.86.
0
J5,6 5.6, J5,OH 7.0, J5,4 8.8 Hz, H-5), 4.06 (m, 1H, J3,OH 3.2, J3,4
0
4.4 Hz, H-3), 3.92 (dd, 1H, H-4), 3.45 (dd, 1H, J6,6 10.1 Hz, H-6),
3.32 (dd, 1H, H-60), 2.84 (d, 1H, OH-5), 2.48 (d, 1H, OH-3), 1.61,
1.40 (2s, 6H, 2CH3); 13C NMR (62.9 MHz, CDCl3): d 143.5 (3Cipso
Ph), 128.6 (6Cm Ph), 127.8 (6Co Ph), 127.1 (3Cp Ph), 112.8
(CMe2), 103.8 (C-1), 87.1 (CPh3), 79.8 (C-4), 79.3 (C-2), 71.9 (C-5),
71.0 (C-3), 64.3 (C-6), 26.6 and 26.4 (2CH3); ESI+MS: m/z 485.3
(100) [M+Na]+; HRESI+MS: calcd for C28H30O6Na: 485.1935
[M+Na]+. Found m/z 485.1950.
4.2.9. 3,5-Di-O-benzoyl-1,2-O-isopropylidene-6-O-trityl-a-D-
allofuranose (9)
Compound 8 (1.9 g, 4.1 mmol) was dissolved under argon in dry
pyridine (6.5 mL) before addition of benzoyl chloride (1 mL,
8.62 mmol). The reaction mixture was stirred at rt for 15 h, then
concentrated under diminished pressure. Following addition of
water and extraction with CH2Cl2, evaporation under diminished
pressure of the solvent yielded pure 9 (2.67 g, 97%) as a white pow-
4.2.12. 3,5-Di-O-cyclohexanoyl-1,2-O-isopropylidene-a-D-
allofuranose 6-dihydrogenophosphate (12)
der: mp 72 °C; ½a 2D6
þ 71:5 (c 1, CH2Cl2); Rf 0.84 (1:1 pentane–
ꢂ
EtOAc); 1H NMR (250 MHz, CDCl3): d 8.03 (dd, 2H, CH PhCO),
7.92 (dd, 2H, CH PhCO), 7.46–7.33 (m, 11H, CH arom), 7.25–7.21
(m, 10H, CH arom), 5.85 (d, 1H, J1,2 3.8 Hz, H-1), 5.74 (ddd, 1H,
Compound 11 (1.33 g, 2.01 mmol) dissolved in anhyd MeOH
(20 mL) and PtO2 (133 mg) were stirred under H2 (4 bar) for 3 d.
The soln was then filtered and concentrated under diminished
pressure to give pure 12 (1.1 g, 100%) as a white solid: mp 57 °C;
0
J5,6 4.1, J5,4 5.0, J5,6 6.4 Hz, H-5), 5.16 (dd, 1H, J3,2 5.0, J3,4 8.8 Hz,
½
a 2D9
ꢂ
þ 82:6 (c 1, MeOH); 1H NMR (250 MHz, CDCl3): d 9.16 (m,
0
H-3), 4.99 (dd, 1H, H-2), 4.72 (dd, 1H, H-4), 3.52 (dd, 1H, J6,6
10.1 Hz, H-6), 3.48 (dd, 1H, H-60), 1.60, 1.35 (2s, 6H, 2CH3); 13C
NMR (62.9 MHz, CDCl3): d 165.4 (2PhC@O), 143.6 (3Cipso Ph trityl),
133.1, 133.0 (2Cp PhC@O), 129.7 (4Cm PhC@O), 129.0 (2Cipso
PhC@O), 128.6 (6Cm Ph trityl), 128.2 (4Co PhC@O), 127.7 (6Co Ph
trityl), 127.0 (3Cp Ph trityl), 113.1 (CMe2), 104.3 (C-1), 86.7
(CPh3), 77.5 (C-2), 76.2 (C-4), 73.7 (C-3), 72.3 (C-5), 62.5 (C-6),
2H, PO(OH)2), 5.82 (d, 1H, J1,2 3.6 Hz, H-1), 5.31 (m, 1H, H-5),
5.85 (dd, J2,3 4.8 Hz, H-2), 4.78 (dd, 1H, J3,4 8.5 Hz, H-3), 4.31 (dd,
1H, J4,5 4.5 Hz, H-4), 4.23 (m, 1H, H-6), 4.10 (m, 1H, H-60), 2.37
(m, 2H, CHipso C6H11), 1.95–1.26 (m, 20H, CH2 C6H11), 1.53, 1.33
(2s, 6H, 2CH3); 13C NMR (62.9 MHz, CDCl3): d 176.3 and 176.2
(2C6H11CO), 114.1 (C-7), 105.9 (C-1), 78.7 (C-2), 77.4 (C-4), 74.4

878
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
(C-3), 72.4 (d, JC,P 8.4 Hz, C-5), 65.3 (d, JC,P 3.0 Hz, C-6), 44.2 and
43.9 (2CHipso C6H11), 30.2, 30.1, 30.0 and 29.8 (4CH2CHipso
C6H11), 27.0 (2CH3); 26.8 (2CH2CH2CH2CHipso C6H11), 26.4 and
26.3 (4CH2CH2CHipso C6H11); 31P NMR (101.2 MHz, CDCl3): d
0.95; ESIꢁMS: m/z 519.2 (100) [MꢁH]ꢁ; ESI+MS: m/z 543.2 (100)
[M+Na]+; ESIꢁHRMS: calcd for C23H36O11P: 519.1990 [MꢁH]ꢁ,
found m/z 519.2005. Anal. Calcd for C23H37O11P: C, 53.07; H,
7.16. Found: C, 52.64; H, 7.51.
uum (0.1 mbar) for 30 min. A soln of 1,2,4-triazole (7 g, 93 mmol)
in anhyd MeCN (100 mL) was then added, and the soln was stirred
at rt under argon for 15 h. Following dilution with CH2Cl2 (80 mL)
and addition of tert-butylhydroperoxide (11 mL, 93 mmol), the
soln was stirred vigorously at rt for 1 h. The mixture was neutral-
ized with 1 M aq Na2S2O3 (100 mL) and 1 M aq NaHCO3 (100 mL),
and then extracted with CH2Cl2 (2 ꢃ 200 mL). The combined or-
ganic phases were dried (MgSO4), filtered, and concentrated under
diminished pressure. The residue (22 g) was purified by two suc-
cessive column chromatographies (4:1 pentane–EtOAc, then 1:1
toluene–EtOAc) to give 15 (7.3 g, 60%) as a white powder: Rf 0.19
(4:1 pentane–EtOAc); 1H NMR (250 MHz, CDCl3): d 7.90–7.86 (m,
4H, CH arom), 7.50–7.41 (m, 2H, CH arom), 7.32–7.20 (m, 14H,
CH arom), 5.88 (d, 1H, J1,2 3.8 Hz, H-1), 5.52 (m, 1H, H-5), 5.12
(dd, J3,2 4.9, J3,4 8.8 Hz, H-3), 5.06 (d, 2H, JH,P 8.0 Hz, OCH2Ph),
4.99 (m, 3H, H-2, OCH2Ph), 4.62 (dd, 1H, J4,5 6.9 Hz, H-4), 4.41
(m, 2H, H-6, H-60), 1.56, 1.33 (2s, 6H, 2CH3); 13C NMR (62.9 MHz,
CDCl3): d 165.3 (2PhCO), 135.5 (d, JC,P 7.0 Hz, 2Cipso PhCH2),
133.0 (2Cp PhCO), 129.7 and 129.6 (2Cm PhCO), 129.0 and 128.7
(2Cipso PhCO), 128.4 (2Cp PhCH2), 128.3 (2Co PhCH2), 128.1 (2Co
PhCO), 127.8 and 127.7 (4Co PhCH2), 113.3 (CMe2), 104.2 (C-1),
77.6 (C-2), 74.9 (C-4), 74.5 (C-3), 72.0 (d, JC,P 7.7 Hz, C-5), 69.2 (d,
JC,P 5.4 Hz, 2PhCH2), 65.8 (d, JC,P 5.1 Hz, C-6), 26.7 and 26.6
(2CH3); 31P NMR (101.2 MHz, D2O): d ꢁ1.1.
4.2.13. 1,2-O-Isopropylidene-a-D-allofuranose 6-phosphate,
diammonium salt (13)
Aq ammonia (60 mL) was added to compound 16 (5.1 g,
99 mmol) dissolved in MeOH (15 mL). The soln was stirred at rt
for 8 h. Co-evaporation of solvents with toluene under diminished
pressure yielded a white powder that was dissolved in MeOH
(5 mL). Addition of diethylether allowed precipitation of the title
compound with the side-product benzamide staying in the liquid
phase. Upon filtration and drying, pure 13 was obtained as a white
powder (2.5 g, 75%). The same procedure was used starting from
compound 12 (1.33 g, 2.55 mmol) to obtain 13 (0.564 g, 66%) with
the side-product cyclohexylamide: ½a D26
ꢂ
þ 26:8 (c 1, water); 1H
NMR (250 MHz, CD3OD): d 5.75 (d, 1H, J1,2 3.5 Hz, H-1), 4.59 (dd,
1H, J2,3 4.6 Hz, H-2), 5.18 (dd, J3,4 8.8 Hz, H-3), 4.05–4.02 (m, 1H,
H-5), 3.99–3.92 (m, 3H, H-4, H-6, H-60), 1.56, 1.36 (2s, 6H, 2CH3);
13C NMR (62.9 MHz, CD3OD): d 113.8 (CMe2), 105.3 (C-1), 81.5
(C-2), 72.2 (C-3, C-4), 67.0 (C-5), 66.6 (d, JC,P 4.9 Hz, C-6), 27.2
and 26.8 (2s, 2CH3); 13C NMR (62.9 MHz, D2O): d 114.0 (CMe2),
104.2 (C-1), 80.1 (C-2), 79.7 (C-4), 70.6 (C-3), 70.3 (d, JC,P 7.5 Hz,
C-5), 65.9 (d, JC,P 2.5 Hz, C-6), 26.2 and 26.0 (2CH3); 31P NMR
4.2.16. 3,5-Di-O-benzoyl-1,2-O-isopropylidene-a-D-
allofuranose 6-dihydrogenophosphate (16)
Compound 15 (7.2 g, 104 mmol) dissolved in MeOH (60 mL)
and Pd/C 10% (2.4 g) were stirred under H2 (15 bar) for 2.5 h. The
mixture was then filtered and concentrated to give 16 (5.1 g,
95%) as a white solid: 1H NMR (360 MHz, CD3OD): d 7.90 (d, 2H,
J 7.3 Hz, Ph), 7.82 (d, 2H, J 7.3 Hz, Ph), 7.52–7.44 (m, 2H, Ph),
7.33–7.23 (m, 4H, Ph), 5.94 (d, 1H, J1,2 3.6 Hz, H-1), 5.51 (m, 1H,
H-5), 5.18 (dd, J3,2 4.9, J3,4 8.8 Hz, H-3), 5.02 (dd, 1H, H-2), 4.62
(dd, 1H, J4,5 7.7 Hz, H-4), 4.45–4.33 (m, 2H, H-6, H-60), 1.55, 1.33
(2s, 6H, 2CH3); 13C NMR (90.5 MHz, CD3OD): d 167.1 and 166.7
(2C@O), 134.4 (2Cp Ph), 130.8 and 130.7 (2Co Ph), 130.2 (2Cipso
Ph), 129.4 (2Cm Ph), 114.5 (CMe2), 106.0 (C-1), 79.2 (C-2), 76.3
(C-3, C-4), 74.2 (d, JC,P 8.0 Hz, C-5), 66.1 (d, JC,P 4.6 Hz, C-6), 27.1
and 27.0 (2CH3); ESI+MS: m/z 553.1 (100) [MꢁH+2Na]+. Anal. Calcd
for C23H25O11P: C, 54.33; H, 4.96; O, 34.62. Found: C, 52.66; H, 5.21;
O, 34.59.
(101.2 MHz, CD3OD):
d
3.35; ESIꢁMS: m/z 299.0 (100)
[Mꢁ2NH4+H]ꢁ; HRESIꢁMS: calcd for C9H16O9P: 299.0526
[Mꢁ2NH4+H]ꢁ, found m/z 299.0538.
4.2.14. Preparation of
salt (14)22,23
D-allopyranose 6-phosphate, disodium
Compound 13 (431 mg, 1.29 mmol) was dissolved in water
(3 mL) and trifluoroacetic acid (20 mL). The soln was stirred at rt
for 2 h and concentrated under diminished pressure. The residue
was eluted with water through an ion-exchange column (Dowex-
50WX8, H+ form) to remove ammonium salts. Evaporation of the
solvents afforded 6-dihydrogenophospho-D-allopyranose (350 mg,
100%): 1H NMR (250 MHz, D2O): d 4.60 (d, 1H, J1,2 8.2 Hz, H-1),
3.89–3.77 (m, 3H, H-3, H-6, H-60), 3.61 (m, 1H, H-5), 3.41 (dd,
1H, J4,3 2.3, J4,5 9.9 Hz, H-4), 3.12 (dd, 1H, J2,3 2.7 Hz, H-2); 13C
4.2.17. 1,2:3,4-Di-O-isopropylidene-b-
D
-psicofuranose 6-
NMR (62.9 MHz, D2O): d 93.3 (C-1b), 92.7 (C-1
a
), 71.9 (d, JC,P
dibenzylphosphate (17)
7.4 Hz, C-5), 70.9 (C-2), 70.8 (C-3), 66.1 (C-4), 65.3 (d, JC,P 4.3 Hz,
A mixture of 1,2:3,4-di-O-isopropylidene-b-D
-psicofuranose28
C-6); 31P NMR (101.2 MHz, D2O): d ꢁ0.087. 6-Dihydrogenophos-
(8.7 g, 33 mmol) and dibenzyl N,N-diisopropylphosphoramidite
(23 g, 67 mmol) was dried under vacuum (0.1 mbar) for 30 min.
A soln of 1,2,4-triazole (9 g, 134 mmol) in anhyd MeCN (200 mL)
was then added, and the soln was stirred at rt under argon for
2 h. Following dilution with CH2Cl2 (200 mL) and addition of tert-
butylhydroperoxide (18.3 mL, 134 mmol), the soln was stirred vig-
orously at rt for 1 h. The mixture was neutralized with 1 M aq
Na2S2O3 (300 mL) and 1 M aq NaHCO3 (300 mL), and then ex-
tracted with CH2Cl2 (2 ꢃ 400 mL). The combined organic phases
were dried (MgSO4), filtered, and concentrated under diminished
pressure. The residue (20 g) was purified by two successive column
chromatographies (4:1 pentane–EtOAc, then 9:1 toluene–EtOAc),
then concentrated under diminished pressure to give 17 (11.43 g,
67%) as a white powder: Rf 0.17 (4:1 pentane–EtOAc); 1H NMR
pho-a-D-allopyranose was then eluted with water on an ion-ex-
change column (DowexÒ-50WX8, Na+ form) and lyophilized to
afford 14 (392 mg, 100%) as a white solid: ½a D27
þ 11:5 (c 1, water);
ꢂ
FTIR (KBr): m
max 3415, 1037 cmꢁ1; 1H NMR (250 MHz, D2O): d 5.05
(d, 1H, J1 3.5 Hz, H-1), 4.79 (d, 1H, J1b,2 8.2 Hz, H-1b), 4.08 (dd,
a
,2
0
1H, J3,2 2.8, J3,4 9.7 Hz, H-3), 4.00 (m, 1H, J6,5 4.6, J6,6 11.4 Hz, H-
6), 3.93 (m, 1H, J6 ,5 5.5 Hz, H-60), 3.80 (ddd, 1H, J5,4 3.0 Hz, H-5),
0
3.61 (dd, 1H, H-4), 3.33 (dd, 1H, H-2), in agreement with the liter-
ature for the monosodium salt;23 13C NMR (62.9 MHz, D2O): d 93.5
(C-1b), 92.6 (C-1a), 72.6 (d, JC,P 7.4 Hz, C-5
a, C-5b), 71.3 (C-3
a
a
),
),
71.1 and 71.0 (C-2b, C-3b), 66.9 (C-2a), 66.3 (C-4b), 65.7 (C-4
64.3 (d, JC,P 3.5 Hz, C-6b), 64.0 (d, JC,P 3.0 Hz, C-6a), in agreement
with the literature22 (nature of the salt not reported); 31P NMR
(101.2 MHz, D2O): d ꢁ0.67, lit.23 3.55 for the monosodium salt.
(250 MHz, CDCl3):
d 7.34 (s, 10H, 2Ph), 5.07–5.00 (m, 4H,
2PhCH2O), 4.66 (d, 1H, J4,3 5.8 Hz, H-4), 4.51 (d, 1H, H-3), 4.25 (d,
0
0
4.2.15. 3,5-Di-O-benzoyl-1,2-O-isopropylidene-
allofuranose 6-dibenzylphosphate (15)
a
-
D
-
1H, J1,1 9.9 Hz, H-1), 4.20 (t, 1H, J5,6 7.3, J5,6 7.3 Hz, H-5), 4.01 (d,
1H, H-10), 3.99 (m, 2H, H-6, H-60), 1.41, 1.38, 1.34, 1.28 (4s, 12H,
4CH3); 13C NMR (62.9 MHz, CDCl3): d 135.6 (d, JC,P 6.6 Hz, 2Cipso
Ph), 127.9 and 128.5 (4Co Ph, 4Cm Ph, 2Cp Ph), 113.5 (C-2), 112.6
A
mixture of 10 (10 g, 23 mmol) and dibenzyl N,N-
diisopropylphosphoramidite (16 g, 46 mmol) was dried under vac-

S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
879
and 111.6 (2CMe2), 84.8 (C-3), 83.0 (d, JC,P 9.0 Hz, C-5), 81.8 (C-4),
69.5 (C-1), 69.3 (d, JC,P 5.4 Hz, 2PhCH2), 66.7 (d, JC,P 5.8 Hz, C-6),
26.3 (2CH3), 26.2 and 24.9 (2CH3); 31P NMR (101.2 MHz, D2O): d
ꢁ1.1; ESI+MS: m/z 543.1 (100) [M+Na]+; HRESI+MS: calcd for
C26H33O9PNa: 543.1754 [M+Na]+, found m/z 543.1764. Anal. Calcd
for C26H33O9P: C, 59.99; H, 6.39. Found: C, 59.97; H, 6.39.
assay described previously,43 in which the production of Ru5P is
monitored as a change in absorbance at 290 nm (
= 72 Mꢁ1 cmꢁ1
e
)
at 37 °C in 50 mM Tris–HCl buffer (pH 7.5). Allpi activity was eval-
uated by the thiobarbituric acid (TBA) assay described previously,7
in which the production of Allu6P is measured at 37 °C in 50 mM
Tris–HCl buffer (pH 7.5) and 5 mM MESNA after a 2-min reaction
time (a blank of identical composition but without enzyme was
used as the reference). Following dehydration, dephosphorylation,
and reaction with TBA, a characteristic yellow TBA adduct is
4.2.18. 1,2:3,4-Di-O-isopropylidene-b-D-psicofuranose 6-
dihydrogenophosphate (18)
Compound 17 (11.43 g, 22 mmol) dissolved in anhyd MeOH
(60 mL) and Pd/C 10% (0.11 g) were stirred under H2 (20 bar) for
3 h. The mixture was then filtered and concentrated to give 18
(7.3 g, 100%): 1H NMR (360 MHz, CD3OD): d 5.38 (br s, 2H, 2OH),
formed which strongly absorbs at 438 nm (e
= 27800 Mꢁ1 cmꢁ1).
Kinetic constants were determined from linear regression using
Lineweaver–Burk or Hanes-Woolf plots at various concentrations
of inhibitors, and replots of apparent Km/Vmax values vs inhibitor
concentration. The x intercept, which is equal to ꢁKi, was deter-
mined by linear regression. For substances showing weak inhibi-
tion, only IC50 values were acquired. These were measured using
a constant R5P concentration equal to the Km value for Rpi activity
measurements, or a constant All6P concentration equal to 0.8 mM
(except for compounds 4 and 7, for which it was 1.0 mM) for Allpi
activity measurements.
0
4.85 (d, 1H, J4,3 5.8 Hz, H-4), 4.66 (d, 1H, H-3), 4.26 (d, 1H, J1,1
0
9.7 Hz, H-1), 4.23 (t, 1H, J5,6 7.3, J5,6 7.3 Hz, H-5), 3.97–3.01 (m,
3H, H-10, H-6, H-60), 1.44, 1.42, 1.37, 1.31 (4CH3); 13C NMR
(90.6 MHz, CD3OD): d 114.9 (C-2), 113.7 and 112.9 (2CMe2), 86.2
(C-3), 84.8 (d, JC,P 9.3 Hz, C-5), 83.3 (C-4), 70.6 (C-1), 67.1 (d, JC,P
5.0 Hz, C-6), 26.7, 26.7, 26.6 and 25.1 (4CH3); 31P NMR
(101.2 MHz, CD3OD): d 0.0.
4.2.19. Preparation of
D
-allulofuranose (or
D-psicofuranose) 6-
4.4. Structural comparisons and analyses
phosphate, disodium salt (19)31
Compound 18 (650 mg, 1.9 mmol) was dissolved in water
(0.5 mL) and trifluoroacetic acid (4.5 mL). The soln was stirred at
rt for 4 h and concentrated under diminished pressure to afford
Manual docking of 5PRA (2) and 5PRH (1) was performed in the
graphics program O,44 which was used together with OPLOT45 and
Molray46 to prepare Figure 2.
6-dihydrogenophospho-D
-psicofuranose (520 mg, 100%): 31P NMR
(101.2 MHz, D2O): d ꢁ0.8. 6-Dihydrogenophospho-
D
-psicofuranose
Acknowledgments
was then successively eluted with water on two ion-exchange
columns DowexÒ-50WX8 (1-H+ form, 2-Na+ form) and lyophilized
The authors are grateful to Jean-Pierre Baltaze for COSY, HSQC,
and J-mod NMR spectra, and to Karine Leblanc, Delphine Arquier,
and Yannick Charvet for MS data. This work was supported by
grants from the Swedish Research Council (VR) to SM, and by Upp-
sala University.
to afford 19 (633 mg, 100%) as a white solid: ½a D28
ꢂ
ꢁ 29 (c 1, water);
), 81.8 (d,
13C NMR (62.9 MHz, CDCl3): d 105.8 (C-2b), 103.4 (C-2
a
JC,P 7.8 Hz, C-5
C-4b), 70.5 and 70.2 (C-3
(d, JC,P 4.7 Hz, C-6 ), 63.2 (C-1
the literature31 for the dihydrogenophosphate form; 31P NMR
(101.2 MHz, D2O): d 0.7 ( ), 0.5 (b).
a
), 81.5 (d, JC,P 7.8 Hz, C-5b), 74.7 and 71.0 (C-3b,
, C-4 ), 66.0 (d, JC,P 4.7 Hz, C-6b), 64.8
), 62.4 (C-1b), in agreement with
a
a
a
a
References
a
1. Rieder, S. V.; Rose, I. A. J. Biol. Chem. 1959, 234, 1007–1010.
2. Bruns, F.; Okada, S. Nature 1956, 177, 87–88.
3. Topper, Y. J. J. Biol. Chem. 1957, 225, 419–425.
4.2.20. 5-Dihydrogenophospho-D-ribono-1,4-lactone (20)
Compound 2 (617 mg, 2.0 mmol) dissolved in water (5 mL) was
deposited on a DowexÒ-50WX8 (H+ form) ion-exchange column
and eluted with water. Concentration under diminished pressure,
lyophilization, and drying under high vacuum for 3 d in a desicca-
tor containing P2O5 yielded compound 20 (386 mg, 86%) as a pale
yellow viscous oil: 1H NMR (250 MHz, D2O): d 4.49 (d, 1H, J3,2
4. Gracy, R. W.; Noltman, E. A. J. Biol. Chem. 1968, 243, 5410–5419.
5. Roos, A. K.; Andersson, C. E.; Bergfors, T.; Jacobsson, M.; Karlèn, A.; Unge, T.;
Alwyn Jones, T.; Mowbray, S. L. J. Mol. Biol. 2004, 335, 799–809.
6. Roos, A. K. PhD thesis, Uppsala University, 2007.
7. Roos, A. K.; Mariano, S.; Kowalinski, E.; Salmon, L.; Mowbray, S. L. J. Mol. Biol.
2008, 382, 667–679.
8. Woodruff, W. W.; Wolfenden, R. J. Biol. Chem. 1979, 254, 5866–5867.
9. Burgos, E.; Salmon, L. Tetrahedron Lett. 2004, 45, 753–756.
10. Burgos, E.; Salmon, L. Tetrahedron Lett. 2004, 45, 3465–3469.
11. Kim, C.; Song, S.; Park, C. J. Bacteriol. 1997, 179, 7631–7637.
12. Poulsen, T. S.; Chang, Y.-Y.; Hove-Jensen, B. J. Bacteriol. 1999, 181, 7126–7130.
13. Sassetti, C. M.; Boyd, D. H.; Rubin, E. J. Mol. Microbiol. 2003, 48, 77–84.
14. Huck, J. H.; Verhoeven, N. M.; Struys, E. A.; Salomons, G. S.; Jakobs, C.; van der
Knaap, M. S. Am. J. Hum. Genet. 2004, 74, 745–751.
0
0
5.6 Hz, H-3), 4.44 (dd, 1H, J4,5(5 ) 6.2, J4,5 (5) 2.5 Hz, H-4), 4.26 (d,
1H, H-2), 3.96 (m, 2H, H-5, H-50); 13C13C NMR (62.9 MHz, D2O): d
178.0 (C-1), 84.3 (d, JC,P 7.9 Hz, C-4), 69.1 (C-3), 68.6 (C-2), 64.6
(d, JC,P 4.6 Hz, C-5); 31P NMR (101.2 MHz, D2O): d ꢁ0.8; ESIꢁMS:
m/z 227.0 (100) [MꢁH]ꢁ; HRESIꢁMS: calcd for C5H8O8P:
226.9962 [MꢁH]ꢁ, found m/z 226.9975.
15. Stern, A. L.; Burgos, E.; Salmon, L.; Cazzulo, J. J. Biochem. J. 2007, 401, 279–285.
16. Burgos, E.; Roos, A. K.; Mowbray, S. L.; Salmon, L. Tetrahedron Lett. 2005, 46,
3691–3694.
17. Bigham, E. C.; Gragg, C. E.; Hall, W. R.; Kelsey, J. E.; Mallory, W. R.; Richardson,
D. C.; Benedict, C.; Ray, P. H. J. Med. Chem. 1984, 27, 717–726.
18. Hardré, R.; Salmon, L. Carbohydr. Res. 1999, 318, 110–115.
19. Chirgwin, J. M.; Noltmann, E. A. J. Biol. Chem. 1975, 250, 7272–7276.
20. Fischer, J.-C.; Horton, D. Carbohydr. Res. 1977, 59, 477–503.
21. Estevez, J. C.; Fairbanks, A. J.; Fleet, G. W. J. Tetrahedron 1998, 54, 13591–
13620.
22. Franke, F. P.; Kapuscinsky, M.; Lyndon, P. Carbohydr. Res. 1985, 143, 69–76.
23. Chenault, H. K.; Mandes, R. F.; Hornberger, K. R. J. Org. Chem. 1997, 62, 331–336.
24. Wood, W. W.; Watson, G. M. J. Chem. Soc., Perkin Trans. I 1987, 2681–2687.
25. Sowa, W.; Thomas, H. S. Can. J. Chem. 1966, 44, 836–838.
26. Andersson, F.; Samuelsson, B. Carbohydr. Res. 1984, 129, C1–C3.
27. Vieira de Almeida, M.; Dubreuil, D.; Cleophax, J.; Verre-Sebrie, C.; Pipelier, M.;
Prestat, G.; Vass, G.; Gero, S. D. Tetrahedron 1999, 55, 7251–7270.
28. Brady, R. F., Jr. Carbohydr. Res. 1971, 20, 170–175.
4.3. Kinetic evaluations
Both enzymes were cloned, expressed, and purified as previ-
ously described.5,7 In a minor modification, the last purification
step consisted of a size exclusion chromatography in a buffer of
20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 10% glycerol, 10 mM b-
mercaptoethanol, 0.1 mM EDTA. For storage, the salt concentration
in the buffer was reduced to 25 mM; for EcRpiB the reducing agent
was also replaced by 5 mM MESNA. Protein samples were stored at
ꢁ20 °C at a concentration of 1–10 mg/mL.
Prior to the kinetic experiments, aliquots were thawed and di-
luted with 50 mM Tris–HCl (pH 7.5); final concentrations in the as-
says were 93 nM for the Rpi activity measurements and 52 nM for
Allpi activity. Rpi activity was evaluated by the spectrophotometric
29. Prisbe, E. J.; Smejkal, J.; Verheyden, J. P. H.; Moffatt, J. G. J. Org. Chem. 1976, 41,
1836–1845.
30. Mio, S.; Kumagawa, Y.; Sugai, S. Tetrahedron 1991, 47, 2133–2144.

880
S. Mariano et al. / Carbohydrate Research 344 (2009) 869–880
31. Herve du Penhoat, P. C. M.; Perlin, A. S. Carbohydr. Res. 1979, 71, 149–167.
32. Brady, R. F., Jr. Carbohydr. Res. 1970, 15, 35–40.
33. Wang, Z. X.; Tu, Y.; Frohn, M.; Zhang, J. R.; Shi, Y. J. Am. Chem. Soc. 1997, 119,
11224–11235.
40. Sen Gupta, K. K.; Sen Gupta, S.; Mandal, S. K.; Basu, S. N. Carbohydr. Res. 1986,
145, 193–200.
41. Hamada, K.; Ago, H.; Sugahara, M.; Nodake, Y.; Kuramitsu, S.; Miyano, M. J. Biol.
Chem. 2003, 278, 49183–49190.
34. Guo, J.; Frost, J. W. J. Am. Chem. Soc. 2002, 124, 528–529.
35. Nieto, N.; Molas, P.; Benet-Buchholz, J.; Vidal-Ferran, A. J. Org. Chem. 2005, 70,
10143–10146.
42. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.; Weissig, H.;
Shindyalov, I. N.; Bourne, P. E. Nucleic Acids Res. 2000, 28, 235–242.
43. Wood, T. Anal. Biochem. 1970, 33, 297–306.
36. McDonald, E. J. Carbohydr. Res. 1967, 5, 106–108.
37. Trifonova, A.; Földesi, A.; Dinya, Z.; Chattopadhyaya, J. Tetrahedron 1999, 55,
4747–4762.
38. Hardré, R.; Bonnette, C.; Salmon, L.; Gaudemer, A. Bioorg. Med. Chem. Lett. 1998,
8, 3435–3438.
44. Jones, T. A.; Zou, J.-Y.; Cowan, S. W.; Kjeldgaard, M. Acta Crystallogr., Sect. A.
Found. Crystallogr. 1991, 47, 110–119.
45. Jones, T. A.; Kjeldgaard, M. O. Methods Enzymol. 1997, 277, 173–288.
46. Harris, M.; Jones, T. A. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2001, 57, 1201–
1203.
39. Sen Gupta, K. K.; Sen Gupta, S.; Mandal, S. K.; Basu, S. N. Carbohydr. Res. 1985,
139, 185–190.
47. Zhang, R.-G.; Andersson, C. E.; Skarina, T.; Evdokimova, E.; Edwards, A. M.;
Joachimiak, A.; Savchenko, A.; Mowbray, S. L. J. Mol. Biol. 2003, 332, 1083–1094.