Design and synthesis of novel pyridoxine 5′-phosphonates as potential antiischemic agents

Article abstract of DOI:10.1021/jm0300678

On the basis of previous reports that the natural cofactor pyridoxal 5′-phosphate 1 appears to display cardioprotective properties, a series of novel mimetics of this cofactor were envisioned. As pyridoxal 5′-phosphate is a natural compound and is subject to biological degradation and elimination pathways, the objective was to generate active phosphonates that are potentially less light sensitive and more stable in vivo than the parent vitamer. Several phosphonates were designed and synthesized, and in particular, compounds 10 and 14 displayed similar biological traits to natural phosphate 1 in the rat model of regional myocardial ischemia and reperfusion. A reduction in infarct size was observed in animals treated with these compounds. In an effort to identify other relevant cardioprotective models in order to potentially define structure - activity relationships, these three compounds were tested in the rat working heart model. Compounds 1, 10, and 14 were compared to dichloroacetic acid (DCA) as positive control in this model. As with DCA, compounds 1, 10, and 14 were found to induce a shift from fatty acid oxidation toward glucose oxidation.

Full text of DOI:10.1021/jm0300678

3680
J . Med. Chem. 2003, 46, 3680-3687
Design a n d Syn th esis of Novel P yr id oxin e 5′-P h osp h on a tes a s P oten tia l
An tiisch em ic Agen ts
Vinh Pham, Wenlian Zhang, Vince Chen, Tara Whitney, J ohn Yao, Doug Froese, Albert D. Friesen,†
J ames M. Diakur, and Wasim Haque*
Medicure Inc. and CanamBioresearch Inc., 6-1200 Waverley Street, Winnipeg, Manitoba, Canada, R3T 0P4
Received February 11, 2003
On the basis of previous reports that the natural cofactor pyridoxal 5′-phosphate 1 appears to
display cardioprotective properties, a series of novel mimetics of this cofactor were envisioned.
As pyridoxal 5′-phosphate is a natural compound and is subject to biological degradation and
elimination pathways, the objective was to generate active phosphonates that are potentially
less light sensitive and more stable in vivo than the parent vitamer. Several phosphonates
were designed and synthesized, and in particular, compounds 10 and 14 displayed similar
biological traits to natural phosphate 1 in the rat model of regional myocardial ischemia and
reperfusion. A reduction in infarct size was observed in animals treated with these compounds.
In an effort to identify other relevant cardioprotective models in order to potentially define
structure-activity relationships, these three compounds were tested in the rat working heart
model. Compounds 1, 10, and 14 were compared to dichloroacetic acid (DCA) as positive control
in this model. As with DCA, compounds 1, 10, and 14 were found to induce a shift from fatty
acid oxidation toward glucose oxidation.
In tr od u ction
5′-phosphates. In particular, 1 (Figure 1), the metabol-
ically active coenzyme form of vitamin B6, plays a vital
role as a cofactor in several crucial enzymatic pathways
and is formed by the in vivo oxidation of the correspond-
ing pyridoxine 5′-phosphate 2. The major degradation
pathway for PLP in vivo is its conversion to pyridoxal
by the action of alkaline phosphatase. There is sufficient
evidence to warrant further exploration into the pos-
sibility that the B6 vitamins may serve as potential
leads for the design of agents for the treatment of
certain cardiovascular diseases. Such agents should be
designed to selectively mimic one or more of the known
modes of actions of the vitamin B6 cogeners with similar
potency as these natural compounds. Along this line,
mimetics functionally similar to PLP have been re-
ported.9
In 1949, Rinehart and Greenberg observed that
monkeys given a vitamin B6 deficient diet developed
atherosclerosis.1 More recent studies using isolated
perfused rat hearts have suggested that pyridoxal 5′-
phosphate (PLP) is an effective ATP receptor antagonist
with activity in the 10-50 µM range.2 In humans, the
situation is more complex, and work on optimizing the
therapeutic benefits of vitamin B6 in treating heart
disease is ongoing. Reports have confirmed the fact that
patients who have suffered a myocardial infarction
display lower levels of PLP.3 Suggestive evidence has
been provided to support the notion that a high daily
dose of supplemental pyridoxine (PN) may delay or even
prevent cardiovascular events.4 The authors of this
study postulated that the protective effect of vitamin
B6 was attributed to increased levels of PLP, which is
the active coenzyme form required for the catabolism
of homocysteine. Elevated levels of this atherogenic
amino acid are now considered to be an independent
risk factor for atherosclerotic disease,5 and the clinical
relevance of the inverse relationship between homocys-
teine and B vitamin levels is still under investigation.6,7
It has been further postulated that pyridoxine supple-
mentation can significantly improve endothelial func-
tion, and potential beneficial effects in terms of the
reduction of transplant coronary artery disease have
been noted.8
The phosphate moiety is a common feature in several
important biomolecules. In particular, this epitope
serves to target well-defined phosphate binding pockets
of a variety of enzymes. It is therefore of interest to
develop novel phosphate mimetics of natural compounds
that can potentially target phosphate binding pockets,
but are themselves relatively resistant to degradation
by phosphatase enzymes. The replacement of a carbon
atom for the oxygen in the P-O-C phosphate bond
renders the resulting phosphonate linkage relatively
resistant toward chemical and enzymatic hydrolysis.
Additionally, the phosphonate substitution pattern
PCR1R2C allows for a wide range of structural modifica-
tions including variations in both substituents R1 and
R2 at the tetravalent carbon center. For example,
Blackburn has pioneered the use of R-fluorophosponates
(R1 ) F and R2 ) H) and difluorophosphonates (R1 and
R2 ) F) as phosphate surrogates.10 The biological
activity of the former class of mimetics, however,
remains relatively unexplored.11 The second deproto-
The vitamin B6 group primarily consists of three
interrelated vitamers including pyridoxine, pyridoxal
(PL) and pyridoxamine (PM). All three pyrimidine
derivatives are found in tissue as their corresponding
* To whom correspondence should be addressed: CanamBioresearch
Inc., 6-1200 Waverely St., Winnipeg, MB, CA R3T 0P4. Tel: (204) 487-
2328. Fax (204) 488-9823. E-mail: whaque@canambioresearch.com.
† Medicure Inc, Winnipeg, MB.
10.1021/jm0300678 CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/11/2003

Novel Pyridoxine 5′-Phosphonates
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3681
Sch em e 1. Synthesis of Pyridoxine 5′-Phosphonates 8
and 10a
F igu r e 1. B-Vitamers and phosphonate analogues.
a
Reagents and conditions: (a) MnO2, C7H8, 40 °C; (b) (t-
BuO)2PO-Na+, THF, 0 °C f rt; (c) HOAc:H2O 4:1, 75 °C; (d) DAST,
CH2Cl2, -78 °C f rt.
F igu r e 2. Acidity (pKa of the second deprotonation) of
phosphate/phosphonates.
Resu lts a n d Discu ssion
nation constants for these phosphatase stable mimetics
are shown in Figure 2.12 In protein binding events at
physiological pH, it is generally assumed that the
natural phosphate moiety is in the completely ionized
state. It can therefore be anticipated that the increased
acidity for both the CHF- and CF2-phosphonates would
not impair their ability to function as mimetics. In fact,
the pKa of the former is nearly identical to that of the
natural phosphonate.
Ch em ist r y. The synthesis of pyridoxine 5′-phos-
phonates 8 and 10 is shown in Scheme 1. The key
starting material for the preparation of both of these
phosphonates was the aldehyde 6, which was in turn
obtained from its precursor 5′-alcohol following the
procedure of Korytnyk and Ikawa,18 with slight modi-
fication. Thus, mild oxidation of a toluene solution of
519 with manganese(IV) oxide at 40 °C gave pyridoxal
derivative 6 in 83% yield. Formation of blocked phos-
phonate 7 was accomplished by the addition of 6 to a
solution of di-tert-butyl phosphite anion in THF at 0
°C,20 followed by reaction at room temperature for 1 h.
Aqueous workup and subsequent recrystallization gave
the racemic R-hydroxyphosphonate 7 in 82% yield.
Concomitant removal of all protecting groups was
achieved by heating a solution of blocked 7 in acetic acid:
water 4:1 at 75 °C overnight. The deprotected R-hy-
droxyphosphonate 8 precipitated from the solution upon
cooling to room temperature and, combined with a
second crop, was obtained in 89% yield.
Pyridoxal 5′-phosphate-dependent enzymes are in-
volved in the catalysis of many important metabolic
processes. The phosphate moiety typically binds near
the N terminus of the anchoring R-helix, and in concert
with Schiff base formation with an active-site lysine,
determines the placement and conformation of 1.13
Recently, there has been renewed interest in the ap-
plication of phosphonate analogues of 1. The ethylphos-
phonate 314 has been utilized as a structural probe in
the detailed investigation into the ionization state of 1
in glycogen phosphorylase.15 Meanwhile, the 4-formyl
derivative of phosphonate derivative 4 served as a
precursor for the development of pyridoxal-6-arylazo-
5′-phosphonate-based P2 receptor antagonists.16 Ad-
ditionally, beneficial effects in animal models have
indicated the possibility of treating cardiovascular
pathologies such as ischemia reperfusion injury with
natural cofactor 1.17 Ischemia typically occurs in an
organ that fails to receive a sufficient supply of blood,
and this situation leads to structural and functional
abnormalities. Subsequent resumption of blood flow,
known as ischemic reperfusion, additionally contributes
to injury of the organ. When the organ affected by these
insults is the heart, the condition is referred to as
myocardial ischemia reperfusion injury. One of the
primary goals in the treatment of acute myocardial
infarction is the early and total restoration of infarct-
related artery perfusion. In the present study, we
describe the preparation of a series of pyridoxine 5′-
phosphonates as agents that potentially mimic the
efficacy of the natural cofactor in reducing scar size in
rats that have been subjected to ischemia reperfusion
injury.
We next focused our attention on the substitution of
a single hydrogen for fluorine in phosphonate template
4. The preparation of 10 was initiated from 7. Thus,
direct substitution of the hydroxyl group by fluorine
upon treatment with DAST (1.3 equiv) proceeded
smoothly to provide the 9 in 82% yield. Routine depro-
tection as before, gave the target R-fluorophosphonate
10 in good yield.
The synthesis of R,R-difluoroketophosphonate 14 is
outlined in Scheme 2. Thus, treatment of blocked
aldehyde 6 with the lithium salt of diethyl difluoro-
methylphosphonate (1.0 equiv) in THF provided difluo-
rohydroxyphosphonate 11 in excellent (80%) yield.
Subsequent oxidation with manganese(IV) oxide at 60
°C in toluene overnight furnished 12 (48% yield). The
pyridoxyl ring H-6 proton of this diketophosphonate was
significantly downfield shifted to 8.87 ppm as antici-
pated. Two-step deprotection involving removal of the
isopropylidine protection under acidic conditions to
provide 13, followed by treatment with triethylsilyl
bromide in acetonitrile, provided phosphonic acid 14.
The absence of a 13C carbonyl carbon resonance (190-

3682 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Sch em e 2. Synthesis of Difluoroketophosphonate 14a
Pham et al.
a
Reagents and conditions: (a) LDA/n-BuLi/CHF2PO(OCH2CH3)2, THF, -78 °C f rt; (b) MnO2, C7H8, 60 °C; 18 h; (c) HOAc:H2O 4:1,
80 °C, 5 h; (d) (CH3CH2)3SiBr/CH3CN.
Sch em e 3. Synthesis of Difluorophosphonate 16a
a
Reagents and conditions: (a) LDA/THF, (C6H5SO2)2NF, -78 °C; (b) HOAc:H2O 4:1, 75 °C, 18 h.
Sch em e 4. Synthesis of Phosphonate 17a
a
Reagents and conditions: (a) [(CH3)2CH]2NCH2CH3 /CH2Cl2, SO3-Py/DMSO, -8 °C, 1.5 h; (b) HOAc:H2O 4:1, 75 °C, 18 h.
200 ppm) suggested that this compound, and possibly
precursor 13, were in their respective cyclic hemiketal
forms.
death are ongoing. A detailed understanding of these
processes would hopefully pave the way for the rational
development of treatment regimens to delay myocardial
cell death. Nevertheless, myocardial infarct size is well
established as an important predictor of cardiac function
and thus prognosis.22
The synthesis of difluorophosphonate 16 is shown in
Scheme 3. Treatment of a solution of the lithium anion
of 9 in THF at -78 °C with excess NFSI (1.33 equiv)
gave crude 15 after workup. The crude blocked difluo-
rophosphonate was unstable and was immediately
deprotected without prior purification. Purification of
the final deprotected material by silica gel chromatog-
raphy gave 16, albeit in modest yield (38%). Scale-up
preparation of this compound was difficult, and suf-
ficient amounts of 16 of consistent purity for subsequent
biological evaluation proved to be elusive. The synthesis
of phosphonate 17 (Scheme 4), however, was straight-
forward; thus, oxidation of the R-hydroxyl moiety,
followed by immediate deprotection after workup, gave
cyclic hemiketal 17 in good yield.
We investigated the ability of our compounds to
reduce infarct size in the male Wistar rat model of left
anterior descending coronary artery occlusion (25 min)
and reperfusion (2 h).23 A comparison of compounds 1,
8, 10, and 14 (bolus injection of 0.02 mmol/kg) for their
ability to reduce infarct size relative to saline control is
shown in Figure 3. The reduction in infarct size under
these conditions was found to be 33% (SEM ( 3, p <
0.05), 19% (SEM ( 7, p > 0.05), 35% (SEM ( 3, p <
0.05), and 46% (SEM ( 2, p < 0.05), respectively. As
the reduction of infarct size with 0.02 mmol/kg of
compound 8 was not found to be statistically significant
(p > 0.05), we reevaluated this compound along with
pyridoxine phosphate 2 and literature compounds 3 and
4 in the rat model of ischemia reperfusion injury at a
higher dose of 0.1 mmol/kg (n ) 10 for each group). None
of these compounds were found to significantly reduce
infarct size; however, a trend toward a reduction in
infarct size (20%, SEM ( 5, p ) 0.07) was noted with
compound 4. The reduction in infarct size with com-
pound 8 (0.1 mmol) as the treatment again did not reach
statistical significance.
Isch em ia Rep er fu sion Stu d ies. Ischemia-reperfu-
sion injury is a root cause of a majority of the important
cardiovascular diseases including myocardial infarction
and thrombotic stroke. Myocardial ischemia is a condi-
tion that exists when the uptake of oxygen in the heart
is below the level needed to maintain the rate of cellular
oxidation.21 Prolonged myocardial ischemia leads to
significant tissue injury and ultimately myocyte necro-
sis. Efforts to establish the precise sequence of biochemi-
cal events involved in recovery upon reperfusion or cell

Novel Pyridoxine 5′-Phosphonates
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3683
F igu r e 3. Effects of compounds 1, 8, 10, and 14 on infarct
size after ischemia (25 min) followed by reperfusion (120 min).
The compounds (0.02 mmol/kg) were administered via bolus
injection 5 min prior to occlusion. Infarct size was determined
by staining the heart with p-nitro-blue tetrazolium (NBT, 0.5
mg/mL) for 40 min at 37 °C. The infarct size (IS) is reported
as a percentage of the area at risk (AR) (standard error is
indicated by the error bars).
F igu r e 4. Effects of dichloroacetic acid (positive control) and
compounds 1, 10, and 14 on metabolic flux through fatty acid
oxidation. The hearts were perfused with either TCA (3 mM)
or test compound (1 mM), while the control group was
untreated (n ) 6 hearts in each group).
TCA Cycle Stu d ies. As glycogen phosphorylase
requires PLP as a cofactor, we decided to screen select
phosphonates for beneficial metabolic modulation prop-
erties. Under normal physiological conditions, the heart
primarily utilizes carbohydrates and fatty acids as
convenient energy sources in order to maintain cardiac
homeostasis. Under certain pathological conditions
where plasma fatty acid levels become elevated, such
as diabetes mellitus, or during myocardial infarction,
alterations in energy metabolism occur. In these pa-
thologies, there is a dramatic reduction in glucose
oxidation with fatty acid oxidation predominating.24
This increased fatty acid oxidation has recently been
linked to increased ischemic injury that is due, in part,
to a significant uncoupling between glycolysis and
glucose oxidation during reperfusion. The resulting
imbalance subsequently leads to increased lactate re-
lease along with increased proton generation from
glycolytically derived ATP, which in turn leads to
greater Na+ and thus Ca2+ influx.
Recently, a novel therapeutic approach for the treat-
ment of ischemic heart disease has been proposed based
on the concept of modulating myocardial metabolism.25
The approach essentially targets increasing glucose
oxidation at the expense of fatty acid oxidation. The net
effect of this shift in metabolism is the improvement of
cardiac efficiency due to better oxygen utilization along
with decreased acidosis in the myocardium resulting
from improved coupling of glycolysis to glucose oxida-
tion.
One such metabolic modulator, namely dichloroacetic
acid (DCA), has recently been shown to behave as a
direct pyruvate dehydrogenase complex activator (PDH
activator).26 The PDH complex irreversibly catalyzes
oxidation of pyruvate to acetyl CoA, and this carbohy-
drate-derived acetyl CoA can then directly participate
in the tricarboxylic acid cycle in mitochondria. The
resultant elevation in the level of glucose oxidation at
the expense of fatty acid oxidation can potentially
improve the mechanical function of the reperfused
human heart. We therefore investigated the effect of
compound 1 and two phosphonates, namely 10 and 14,
for their ability to alter cardiac metabolism.
F igu r e 5. Effects of dichloroacetic acid (positive control) and
compounds 1, 10, and 14 on metabolic flux through glycolysis.
The hearts were perfused with either TCA (3 mM) or test
compound (1 mM), while the control group was untreated (n
) 6 hearts in each group); p < 0.05 vs control for DCA, 1 and
10, p ) 0.052 for compound 14.
solution containing calcium (2.5 mmol/L), glucose (5.5
mmol/L), palmitate (0.4 mmol/L), and 3% bovine serum
albumin. Fatty acid oxidation and glucose oxidation
were measured simultaneously by perfusing hearts with
[3H]-palmitate and [14C]-glucose, respectively. The effect
of compounds 1, 10, and 14 on the fatty acid utilization
rate is shown in Figure 4. DCA (positive control) did
not alter palmitate oxidation rates as compared to the
untreated control group (397 ( 29 vs 418 ( 24,
respectively). Phosphate 1 (398 ( 15) and phosphonates
10 (409 ( 17) and 14 (414 ( 12) likewise failed to alter
palmitate oxidation compared to control (418 ( 24).
Glucose oxidation rates were determined by the
qualitative measurement of 14CO2 production.28 As
shown in Figure 5, DCA effected an increase in glucose
oxidation rates as compared to control {2422 ( 140 (p
) 0.001) vs 1580 ( 183}, as did each of compounds 1,
10, and 14 {2253 ( 230 (p ) 0.045), 2244 ( 137 (p )
0.016), 2339 ( 293 (p ) 0.052), vs 1580 ( 183}.
As shown in Figure 6, the contribution from palmitate
oxidation and glucose oxidation in the control under the
conditions employed in this study were nearly identical
(%52 ( 2 vs %48 ( 2). In the DCA-treated hearts, the
contribution from these substrates was altered signifi-
cantly. Treatment of rat hearts with this agent resulted
in a decrease in contribution from palmitate oxidation
with a concomitant increase in glucose oxidation (%37
( 2 vs %63 ( 2, respectively). Compounds 1, 10, and
14 were found to decrease the contribution from palmi-
tate oxidation and simultaneously increase the glucose
oxidation contribution (%42 ( 1 vs %58 ( 2; %43 ( 1
vs %57 ( 2; and %42 ( 1 vs %58 ( 2, respectively).
The compounds were tested in the Langendorff per-
fusion model following published protocol.27 Working rat
hearts were perfused with modified Krebs-Henseleit

3684 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Pham et al.
Exp er im en ta l Section
Pyridoxine was purchased from Fluka. All other reagents
were obtained from standard commercial sources and were
used without further purification unless otherwise indicated.
All reactions were carried out under a dry nitrogen atmosphere
using dry solvents. Reactions were monitored by thin-layer
chromatography carried out on Sigma-Aldrich 0.25 mm silica
gel plates (60 Å, fluorescent indicator). Spots were visualized
under UV light (254 nm). Flash column chromatography was
performed on silica gel (60, particle size 0.032-0.063 mm) from
Scientific Adsorbents Inc. Reverse phase chromatography was
performed on C-18 silica gel (60, particle size 0.032-0.063 mm)
from Scientific Adsorbents Inc. HPLC analysis was performed
using a Waters 996 PDA High-Performance Liquid Chromato-
graph equipped with a Waters 600 Controller using a gradient
elution; 0-100% (v/v) methanol or acetonitrile vs 0.1% aqueous
trifluoroacetic acid over 20 min; 1 mL/min flow rate @ 25 °C;
Zorbax SB-C8 (4.6 × 150 mm; 3.5 µm particle size) column
equipped with a Zorbax SB-Aq (4.6 × 12.5 mm; 5 µm particle
size) guard column. Signals were detected with a photodiode
array detector (set at maxplot 254-400 nm). NMR spectra
were recorded on a Bruker AM-300 instrument (13C, 19F, and
31P at 75.5, 282, and 121 MHz, respectively) and were
calibrated using residual nondeuterated solvent as the internal
reference. Spectra were processed using Spinworks version 1.3
(developed by Dr. Kirk Marat, Department of Chemistry,
University of Manitoba). All 19F spectra are reported using
hexafluorobenzene (δ -162.9 ppm) as the external standard,
while 31P spectra were collected using 85% H3PO4 (δ 0.0 ppm)
as the external reference. Compound 3 was prepared according
to the method of Hullar,14 and compound 4 was prepared
following the procedure published by Kim et al.16
2,2,8-Tr im et h yl-4H -[1,3]d ioxin o[4,5-c]p yr id in e-5-ca r -
boxa ld eh yd e (6). The reaction conditions employed were
similar to those previously reported18 with minor modification.
To a solution of 517 (25 g, 0.12 mol) in toluene (900 mL) was
added manganese(IV) oxide (85%), 33 g, 0.32 mol), and the
reaction was stirred under inert atmosphere at 40 °C for 4 h.
After this time, a second portion of MnO2 (16.5 g, 0.16 mol)
was added, and stirring was continued for an additional 4 h.
The solvents were evaporated, and the resulting yellow solid
was dissolved in hexane:diethyl ether 1:1 and filtered and the
filtrate kept at rt. The resulting solid was collected by
filtration, and the mother liquors were recrystallized from
hexane:diethyl ether 7:3 to give 20.6 g (83% combined yield)
of 6. This material was of sufficient purity for use in the
subsequent reaction.
F igu r e 6. Effects of DCA and compounds 1, 10, and 14 on
the TCA cycle contribution; % palmitate, light gray shading;
% glucose, black shading.
As the phosphonates described herein are based on a
natural vitamer scaffold that behaves as a cofactor in
numerous enzymatic processes, an acute toxicity test
was performed on the monosodium salt of compound 10
(prepared by titrating this compound with 1 equiv of
standard sodium hydroxide solution). The LD50 from
intravenous administration of this salt (12 mg/mL) to
rats was found to be greater than 100 mg/kg.
Con clu sion
In contrast to aldehydes 1 and 3 that are known to
be light sensitive, phosphonates 8, 10, 14, and 17 were
found to be stable for 24 h in aqueous solutions kept at
room temperature unshielded from light as assayed by
HPLC (see Experimental Section). Ischemia reperfusion
injury represents an extremely complex pathology.
Active B-vitamer 1 and phosphonate derivates 10 and
14 have been shown to reduce infarct size in preliminary
experiments in the rat model of ischemia reperfusion
injury. At least one general pathway for their potential
action in the nonischemic heart has been identified,
namely that of involvement in energy metabolism.
Neither DCA, nor compounds 1, 10, and 14, displayed
any measurable influence on palmitate oxidation rates
in the isolated working rat heart. These compounds did,
however, alter glucose metabolism. The experimental
conditions employed for the metabolic studies were
designed to provide equal contributions from fatty acids
and glucose to the TCA cycle. DCA and compounds 1,
10, and 14 were able to induce a desirable shift away
from fatty acid metabolism toward glucose metabolism.
Hemodynamic parameters including heart rate, mean
arterial pressure, and pressure rate index were mea-
sured for all animal groups in the metabolic study. No
significant differences in these three parameters were
observed between the test animals and control group,
demonstrating that the observed effects on glucose
metabolism are not occurring secondarily to changes in
contractile function. Although these studies were per-
formed on nonischemic hearts, an increase in glucose
oxidation both during and postischemia would be ex-
pected to provide therapeutic benefit by improving the
coupling of glycolysis and glucose oxidation which leads
to a decrease in intracellular proton production.28 The
net effect of improved coupling is therefore an improve-
ment of cardiac efficiency. Compounds 1, 10, and 14
were able to induce a shift toward glucose oxidation,
and further work to uncover more details on their
specific mode of action is currently underway.
[Hydr oxy-(2,2,8-tr im eth yl-4H-[1,3]dioxin o[4,5-c]pyr idin -
5-yl)m eth yl]p h osp h on ic Acid Di-ter t-bu tyl Ester (7). Di-
tert-butyl phosphite (44.1 g, 227 mmol) was added to a
suspended solution of sodium hydride (9.1 g, 60% in mineral
oil, 227 mmol) in THF (100 mL) under nitrogen atmosphere
at 0 °C. The mixture was warmed to rt and stirred for 15 min.
The reaction mixture was again cooled to 0 °C, and a solution
of aldehyde 6 (34.2 g, 165 mmol) in THF (300 mL) was added.
The resulting mixture was slowly warmed to rt and then
stirred for 1 h further. After this time, the reaction was
quenched with saturated aqueous NaHCO3 (100 mL). The
organic compounds were extracted with diethyl ether, and the
organic solution was successively washed with saturated
aqueous NaHCO3 and brine, dried (MgSO4), filtered, and
evaporated. The resulting crude solid was dissolved in hexane:
diethyl ether 52:48 (500 mL) and kept at rt overnight. The
colorless solid was collected by filtration and washed with
hexane then diethyl ether (quickly). The mother liquor was
evaporated, washed with hexane, and again recrystallized to
provide 54.5 g (82% combined yield) of 7 as a colorless solid.
1H NMR (CDCl3) δ 8.13 (d, J ) 2.6 Hz, 1H), 5.12 (dd, J ) 16.4,
1.9 Hz, 1H), 4.92 (dd, J ) 16.4, 1.4 Hz, 1H), 4.73 (dd, J ) 10.6,
3.6 Hz, 1H), 3.00 (dd, J ) 10.6, 3.9 Hz, 1H), 2.40 (d, J ) 2.0
Hz, 3H), 1.56 (s, 3H), 1.52 (s, 3H), 1.47 (s, 9H), 1.43 (s, 9H).
31P NMR (CDCl3, 1H-decoupled) δ 13.3. HRMS (ES+) calcd for
C
19H33NO6P (M + H+) m/z, 402.204; found 402.2039. Anal.
(C19H32NO6P): C, H, N.

Novel Pyridoxine 5′-Phosphonates
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3685
[Hyd r oxy-(5-h yd r oxy-4-h yd r oxym eth yl-6-m eth ylp yr i-
d in -3-yl)m eth yl]p h osp h on ic Acid (8). The fully protected
compound 7 (48.9 g, 120.5 mmol) was stirred in acetic acid-
water (4:1, 800 mL) at 75 °C for 17 h. After this time, the
reaction mixture was allowed to cool to rt, and the precipitate
was collected by filtration. The colorless solid was washed with
water, methanol, and then ethyl acetate. The mother liquor
was evaporated and washed with water, methanol, and ethyl
acetate. The HPLC profile for both solids was found to be
identical, thus furnishing 26.8 g (89% combined yield) of 8 as
a colorless solid. 1H NMR (D2O) δ 8.24 (d, J ) 2.2 Hz, 1H),
5.25 (d, J ) 13.5 Hz, 1H), 5.05 (d, J ) 14.0 Hz, 1H), 4.97 (d,
temperature for 10 min. The solution was then cooled further
to -78 °C, and a solution of aldehyde 6 (5.18 g, 25 mmol) in
THF (10 mL) was added dropwise to the solution containing
the diethyl phosphite anion. The reaction mixture was then
allowed to warm to room temperature. The solvent was
evaporated, and the residue was dissolved in dichloromethane,
washed with saturated aqueous NaHCO3 and brine, dried (Na2-
SO4), and evaporated. Purification by column chromatography
over silica gel using dichloromethane:hexane:methanol 5:5:1
1
gave 7.72 g (80%) of 11 as a colorless syrup. H NMR (CDCl3)
δ 8.10 (s, 1H), 5.08 (ddd, J ) 23.0, 4.5, 1.2 Hz, 1H), 5.03 (d, J
) 16.5 Hz, 1H), 4.9 (d, J ) 16.5 Hz, 1H), 4.37-4.21 (m, 4H),
2.40 (s, 3H), 1.55 (s, 3H), 1.52 (s, 3H), 1.38 (t, 3H), 1.36 (t,
3H). 19F NMR (CDCl3) δ -126.08 (dd, J ) 304.5, 105.4 Hz),
-114.59 (dd, J ) 304.5, 97.2 Hz). 31P NMR (CDCl3, 1H-
decoupled) δ 7.24 (dd, J ) 105.4, 96.7 Hz).
1
J ) 14.0 Hz, 1H), 2.63 (d, 1.2 Hz, 3H). H NMR (0.1 M NaOH
in D2O): δ 7.73 (d, J ) 2.0 Hz, 1H), 5.09 (d, J ) 12.3 Hz, 1H),
4.87 (d, J ) 12.3 Hz, 1H), 4.65 (d, J ) 12.3 Hz, 1H), 2.41 (d,
J ) 0.8 Hz, 3H). 13C NMR (0.1 M NaOH in D2O) δ 161.0, 141.9
(J ) 1.4 Hz), 139.5 (d, J ) 4.3 Hz), 137.9 (d, J ) 1.3 Hz), 124.4
(d, J ) 2.9 Hz), 68.2 (d, J ) 141.7 Hz), 55.6, 15.6. 31P NMR
[1,1-Diflu or o-2-oxo-2-(2,2,8-tr im eth yl-4H-[1,3]d ioxin o-
[4,5-c]p yr id in -5-yl)eth yl]p h osp h on ic Acid Dieth yl Ester
(12). A mixture of alcohol 11 (3.30 g, 8.35 mmol) and MnO2
(6.30 g, 72 mmol) in toluene (80 mL) was stirred at 60 °C
overnight. After this time, the reaction mixture was filtered
and the filtrate evaporated. The oily residue was purified by
flash column chromatography using hexane:ethyl acetate 2:1
to 1:2 as eluent. Ketone 12, 1.59 g (48% yield), was obtained
1
(D2O, H-decoupled) δ 14.8; (0.1 M NaOH in D2O) δ 14.3 (d, J
) 12.3 Hz); 1H decoupled δ 14.3 (s). HRMS (ES+) calcd for
C8H13NO6P (M + H+) m/z, 250.0475; found 250.0477. HPLC
99% purity. Anal. (C8H12NO6P): C, H, N.
[F lu or o-(2,2,8-tr im eth yl-4H-[1,3]d ioxin o[4,5-c]p yr id in -
5-yl)m eth yl]p h osp h on ic Acid Di-ter t-bu tyl Ester (9). Di-
ethylaminosulfur trifluoride (1.0 g, 6.2 mmol) was added to a
solution of 7 (1.9 g, 4.73 mmol) in dichloromethane (200 mL)
at -78 °C under nitrogen atmosphere. The resulting solution
was allowed to slowly warm to rt and then stirred for an
additional 1.5 h. Saturated aqueous NaHCO3 (100 mL) was
added slowly to quench the reaction, After separation of the
phases, the aqueous layer was further extracted with dichlo-
romethane (2×). The combined organic solution was dried
(MgSO4), filtered, and evaporated. Flash chromatography
(silica, hexane:ethyl acetate 9:1 to 1:1) gave 1.56 g (82%) of 9
as yellow oil. 1H NMR (CDCl3) δ 8.01 (s, 1H), 5.39 (dd, J )
44.5, 9.1 Hz, 1H), 5.06 (dt, J ) 16.9 Hz, J ) 1.4 Hz, 1H), 4.99
(d, J ) 16.9 Hz, 1H), 2.39 (dd, J ) 1.4, 1.4 Hz, 3H), 1.53 (s,
3H), 1.50 (s, 3H), 1.43 (s, 9H), 1.42 (s, 9H). 19F NMR (CDCl3,
1H decoupled) δ -202.87 (d, J ) 84.9 Hz). 31P NMR (CDCl3,
1H decoupled) δ 6.86 (d, J ) 84.9 Hz). HRMS (ES+) calcd for
C19H32NO5FP (M + H+) m/z, 404.1997; found 404.1995.
1
as light yellow syrup. H NMR (CDCl3) δ 8.87 (s, 1H), 5.08 (s,
2H), 4.38-4.31 (m, 4H), 2.50 (s, 3H), 1.56 (s, 6H), 1.41 (t, J )
7.2 Hz). 19F NMR (CDCl3) δ -109.9 (dd, J ) 94.9, 1.5 Hz). 31P
NMR (CDCl3, 1H-decoupled) δ 3.94 (t, J ) 95.0 Hz). HRMS
(ES+) calcd for C16H23NO6F2P (M + H+) m/z, 394.1226; found
394.1226. Anal. (C16H22F2NO6P): C, H, N.
[1,1-Diflu or o-2-(5-h yd r oxy-4-h yd r oxym et h yl-6-m et h -
ylp yr id in -3-yl)-2-oxoeth yl]p h osp h on ic Acid Dieth yl Es-
ter (13). Compound 12 (1.59 g, 4.05 mmol) in 80% HOAc (40
mL) was stirred at 80 °C for 5 h. The solution was allowed to
cool to room temperature and then evaporated to dryness. The
resulting crude mixture was purified by flash column chro-
matography using dichloromethane:hexane:methanol 5:5:1 as
1
eluant to give 760 mg (53%) of 13 as a white solid. H NMR
(DMSO-d6) δ 9.73 (s, 1H), 8.03 (s, 1H), 7.81 (s, 1H), 5.07 (d, J
) 14.0 Hz, 1H), 4.94 (d, J ) 14.0 Hz, 1H), 4.18-4.02 (m, 4H),
3.31 (br s, 1H), 2.49 (s, 3H), 1.24 (t, J ) 7.0 Hz, 3H), 1.22 (t,
J ) 7.0 Hz, 3H). 19F NMR (DMSO-d6) δ -118.74 (ddd, J )
303.8, 98.9, 12.7 Hz), -116.61 (ddd, J ) 303.8, 98.5, 18.6 Hz).
31P NMR (DMSO-d6) δ 6.26 (td, J ) 98.2, 2.7 Hz).
[Flu or o-(5-h ydr oxy-4-h ydr oxym eth yl-6-m eth ylpyr idin -
3-yl)m eth yl]p h osp h on ic Acid (10). Fully protected com-
pound 9 (1.56 g, 3.8 mmol) was stirred in acetic acid:water
4:1 (30 mL) at 75 °C for 15 h. The solvents were removed, and
the residue was coevaporated with toluene to give a yellow
solid. This crude product was dissolved in hot water (20 mL)
and filtered while still hot. The filtrate was slowly cooled to
fridge temperature (5 °C), and the precipitated product was
collected by filtration. The solid was washed with cold water
(10 mL), methanol (10 mL), and dichloromethane (10 mL) to
give a colorless solid. The mother liquor was subsequently
evaporated, dissolved in water (10 mL), and precipitated
repeatedly to provide additional colorless solid. The HPLC
profile for both solids was found to be identical, and 484 mg
(50% combined yield) of 10 was obtained as a colorless solid.
1H NMR (0.1 M NaOH in D2O) δ 7.71 (d, J ) 1.6 Hz, 1H),
5.87 (dd, J ) 44.4, 8.2 Hz, 1H), 4.83 (d, J ) 12.6 Hz, 1H), 4.64
(d, J ) 12.6 Hz, 1H), 2.43 (s, 3H). 13C NMR (0.1 M NaOH in
D2O) δ 160.4, 143.1, 138.7 (d, J ) 3.6 Hz), 134.5 (d, J ) 17.8
Hz), 124.4, 89.2 (dd, J ) 177.5, 145.5 Hz), 55.68, 15.7. 19F NMR
[(3,7-Dih ydr oxy-6-m eth yl-1,3-dih ydr ofu r o[3,4-c]pyr idin -
3-yl)d iflu or om eth yl]p h osp h on ic Acid (14). Phosphonic
acid diethyl ester 13 (1.05 g, 2.97 mmol) in dry acetonitrile
(20 mL) was treated with excess trimethylsilyl bromide (3.13
mL, 17.8 mmol), and the reaction was stirred at rt overnight.
The solvent was then evaporated, ammonium hydroxide added,
and the solvent again evaporated. Chromatography over
reverse phase (C-18) silica using CH3CN:H2O 9:1 as elution
solvent gave 590 mg (60%) of 14 as a white solid. 1H NMR
(D2O) δ 7.91 (s, 1H), 5.31 (d, J ) 15 Hz, 1H), 5.21 (d, J ) 15
Hz, 1H), 2.53 (s, 3H). 13C NMR (D2O) δ 157.19, 145.0, 143.53,
133.93, 121.8, 71.0, 14.9. 19F NMR (D2O) δ -121.80 (dd, J )
295.8, 81.2 Hz), -119.52 (dd, J ) 295.8, 80.1 Hz). 31P NMR
(D2O) δ 4.40 (br t, J ) 80.5 Hz). HRMS (ES+) calcd for C9H11
-
NO6F2P (M + H+) m/z, 298.0287; found 298.0286; m/z (ES-):
296.2 (M-H+). HPLC 100% purity.
[Diflu or o-(5-h yd r oxy-4-h yd r oxym eth yl-6-m eth ylp yr i-
d in -3-yl)m eth yl]p h osp h on ic Acid (16). Blocked monofluo-
rophosphonate 9 (86 mg, 0.21 mmol) was dissolved in 5 mL of
THF and cooled to -78 °C under dry nitrogen. Lithium
diisopropylamide (2.0 M in THF/n-heptane, 0.14 mL, 0.28
mmol) was then introduced, and the reaction mixture was
stirred at -78 °C for 1 h. After this time, a solution of
N-fluorobenzenesulfonimide (95 mg, 0.30 mmol) in THF (5 mL)
was added, and stirring was continued at this temperature
for an additional 1 h. The reaction was quenched with
saturated aqueous NaHCO3 and then allowed to warm to rt.
The mixture was extracted with ethyl acetate, and the organic
solution was dried over MgSO4, filtered, and concentrated.
Crude 15 obtained in this way was immediately dissolved in
1
(0.1 M NaOH in D2O) δ -196.8 (dd, J ) 69.0, 44.4 Hz); H-
decoupled δ -196.8 (d, J ) 69.0 Hz). 31P NMR (0.1 M NaOH
1
in D2O) δ 9.38 (dd, J ) 69.0, 8.2 Hz); H-decoupled δ 9.38 (d,
J ) 69.0 Hz). HRMS (ES+) calcd for C8H12NO5FP (M + H+)
m/z, 252.0432; found 252.0433. HPLC 98% purity. Anal. (C8H11
-
FNO5P): C, H, N.
[1,1-Diflu or o-2-h yd r oxy-2-(2,2,8-t r im et h yl-4H -[1,3]d i-
oxin o[4,5-c]p yr id in -5-yl)eth yl]p h osp h on ic Acid Dieth yl
Ester (11). To a THF solution (100 mL) at -40 °C under dry
nitrogen were added lithium diisopropylamide (2.0 M, 12.5 mL,
25 mmol) and n-BuLi (0.5 M, 1.25 mL, 2.5 mmol). Diethyl
difluoromethylphosphonate (4.08 mL, 25 mmol) was then
introduced to this mixture, and the reaction was stirred at this

3686 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Pham et al.
HOAc:H2O 4:1 (10 mL) and then stirred at 75 °C under
nitrogen for 18 h. The volatiles were evaporated under reduced
pressure, and the crude product was purified by column
chromatography over silica gel using CHCl3:CH3OH:NH4OH
Wor k in g Hea r t P er fu sion s. Male Sprague-Dawley rats
(0.3-0.35 kg) were used in this study, and the rat hearts were
cannulated and prepared for working heart perfusions as per
literature protocol.27 The working rat hearts were perfused
with a modified Krebs-Henseleit solution containing calcium
(2.5 mmol/L), glucose (5.5 mmol/L), palmitate (0.4 mmol/L),
and 3% bovine serum albumin. Spontaneously beating hearts
were employed in all perfusions and were subjected to aerobic
perfusion (60 min). The test compounds were perfused at a
concentration of 1 mM while the positive control DCA was
perfused at a concentration of 3 mM. Fatty acid oxidation and
glucose oxidation were measured simultaneously by perfusing
the hearts with [3H]-palmitate and [14C]-glucose, respectively.
The control group received no treatment. The total myocardial
3H2O and 14CO2 production was measured at 10-min intervals
during the aerobic period, and these data were used to
calculate fatty acid and glucose oxidation, respectively.
1
65:50:25 as eluant to furnish 20 mg (35%) of 16 as a solid. H
NMR (0.1 M NaOH in D2O) δ 7.68 (s, 1H), 4.81 (s, 2H), 2.30
(s, 3 H). 19F NMR (0.1 M NaOH in D2O) δ 100.65 (d, J ) 90.1
Hz). 31P NMR (0.1 M NaOH in D2O) δ 6.00 (t, J ) 90.0 Hz).
HRMS (ES+) calcd for C8H11NO5F2P (M + H+) m/z, 270.0337;
found 270.034. HPLC 91% purity.
(3,7-Dih ydr oxy-6-m eth yl-1,3-dih ydr ofu r o[3,4-c]pyr idin -
3-yl)p h osp h on ic Acid (17). To a solution of R-hydroxyphos-
phonate 7 (3.5 g, 8.7 mmol) in dichloromethane (100 mL) at
-8 °C under nitrogen atmosphere was added diisopropylethyl-
amine (3.39 g, 26.2 mmol) followed by a solution of SO3-
pyridine complex (4.17 g, 26.2 mmol) in DMSO (20 mL). The
yellow solution was stirred for 1.5 h at this temperature and
then diluted with diethyl ether (300 mL). The organic solution
was successively washed with water, saturated aqueous
NaHCO3, and brine and then dried (MgSO4), filtered, and
concentrated to give the crude blocked R-ketophosphonate as
a yellow oil. This material was deprotected by stirring in
HOAc:H2O 4:1 (60 mL) at 75 °C for 18 h. After this time, the
solvent was evaporated, and the resulting solid was washed
with water, methanol, and finally dichloromethane to furnish
1.85 g (86%) of 17 as an off-white solid. 1H NMR (0.1 M NaOH
in D2O) δ 7. 73 (s, 1H), 5.25 (dd, J ) 15.0, 6.2 Hz, 1H), 5.08 (d,
J ) 15.0 Hz, 1H), 2.42 (s, 3H). 13C NMR (0.1 M NaOH in D2O)
δ 158.1, 144.2, 143.1 (d, J ) 4.0 Hz), 137.8 (d, J ) 8.5 Hz),
120.5, 107.6 (d, J ) 193.4 Hz), 71.4 (d, J ) 5.2 Hz), 49.3, 14.9.
31P NMR (0.1 M NaOH in D2O) δ 11.1 (d, J ) 6.2 Hz); 1H
decoupled δ 11.1 (s). HRMS (ES+) calcd for C8H11NO6P (M +
H+) m/z, 248.0319; found 248.032. HPLC 99.8% purity. Anal.
(C8H10NO6P‚1.5H2O): C, H, N.
Ack n ow led gm en t. We thank Dr. Michelle Mc-
Donald (William Harvey Research Ltd. London, En-
gland) and Nicole S. Wayman for performing the
ischemia reperfusion work. We also thank Rick Barr
(Metabolic Modulators Research, Ltd., University of
Alberta, Edmonton, AB, Canada) for carrying out the
metabolic studies and Dr. J ason Dyck (Departments of
Pediatrics and Pharmacology, University of Alberta,
Edmonton, AB, Canada) for helpful discussions regard-
ing the metabolic modulation studies. Toxicity tests
were carried out by Nucro-Technics Inc. (Ontario,
Canada). We are indebted to Dr. Angelina Morales-
Izquierdo (Mass Spectrometry Laboratory, Department
of Chemistry, University of Alberta) for providing the
high resolution ES MS data, and Mrs. Darlene Mahlow
(Microanalytical Laboratory, Department of Chemistry,
University of Alberta) for performing the combustion
analysis.
HP LC Assessm en t of Ligh t Sen sitivity. Aqueous solu-
tions of compounds 1, 3, 8, 10, 14, and 17 were tested for
stability toward light. HPLC analysis of solutions exposed to
natural light at room temperature under the conditions
outlined in the general methods revealed that compounds 8,
10, 14, and 17 remained unchanged after 24 h, while 42% and
70% of compounds 1 and 3, respectively, remained after this
period.
Su p p or tin g In for m a tion Ava ila ble: HPLC purity data
for compounds 8, 10, and 14. This material is available free
of charge via the Internet at http://pubs.acs.org.
Isch em ia Rep er fu sion Stu d ies. This study was carried
out at the William Harvey Research Ltd. (London, England)
as per literature protocol.23 Briefly, male Wistar rats (200-
300 g) were anesthetized (thiopentone sodium), tracheoto-
mized, intubated, and ventilated while maintaining the body
temperature at 38 ( 1 °C. The right carotid artery was
cannulated to monitor mean arterial blood pressure and heart
rate. The right jugular vein was cannulated for the adminis-
tration of the test compounds prior to lateral thoractomy. A
snare occluder was positioned around the left anterior de-
scending artery (LAD), and the animals were allowed to
stabilize for 30 min prior to LAD occlusion. The rats were then
given either saline (infusion) or test compound (bolus injec-
tion: 0.02 mmol/kg or 0.1 mmol/kg), and after 5 min, the
occluder was tightened. Following 25 min of acute ischemia,
the occluder was reopened to allow reperfusion for 2 h. Upon
completion of the reperfusion period, the coronary artery was
reoccluded, and Evans blue dye (2% w/v, 1 mL) was admin-
istered into the left ventrical (LV) via the right carotid artery
cannula to identify perfused and nonperfused (area at risk of
infarction, AR) regions. The Evans blue dye stains perfused
myocardium while leaving the nonperfused (AR) unstained.
AR was obtained as a percentage of the LV. The animals were
sacrificed, and after sectioning the heart, the AR was separated
form the stained nonischemic area. The AR was cut into small
pieces and incubated with p-nitro-blue tetrazolium solution
(NBT, 0.5 mg/mL) for 40 min at 37 °C. In viable myocardium,
the presence of intact dehydrogenase activity leads to the
formation of a dark blue formazan in the presence of NBT
while necrotic regions fail to stain. The pieces were then
separated based on staining and weighed to determine the
infarct size (expressed as a percentage of AR).
Refer en ces
(1) Rineheart, J . F.; Greenberg, L. D. Arteriosclerotic lesions in
pyridoxine-deficient monkeys. Am. J . Pathol. 1949, 25, 481-
491.
(2) Wang, X.; Dakshinamurti, K.; Musat, S.; Dhalla, N. S. Pyridoxal
5-phosphate is an ATP-receptor antagonist in freshly isolated
rat cardiomyocytes. J . Mol. Cell Cardiol. 1999, 31, 1063-1072.
(3) Serfontein, W. J .; Ubbink, J . B.; De Villiers, L. S.; Rapley, C.
H.; Becker, P. J . Plasma pyridoxal-5-phosphate level as risk
index for coronary artery disease. Atherosclerosis 1985, 55, 357-
361.
(4) Ellis, J . M.; McCully, K. S. Prevention of myocardial infarction
by vitamin B6. Res. Commun. Pathol. Pharmacol. 1995, 89, 208-
220.
(5) Abdelmouttaleb, I.; Danchin, N.; Aimone-Gastin, I.; Namour, F.;
Angioi, M.; Gelot, M.-A.; Bennani, N.; Lambert, D.; J eandel, C.;
Gue´any. Homocysteine, vitamins B6, B12, folate, and risk of
coronary artery disease in patients undergoing diagnostic coro-
nary angiography. Amino Acids 2000, 18, 139-146.
(6) Robinson, K.; Mayer, E. L.; Miller, D. P.; Green, R.; van Lente,
F.; Gupta, A.; Kottke-Marchant, K.; Savon, S. R.; Selub, J .;
Nissen, S. E.; Kutner, M.; Topol, E. J .; J acobsen, D. W.
Hyperhomocysteinemia and low pyridoxal phosphate. Circula-
tion 1995, 92, 2825-2830.
(7) Robinson, K.; Arheart, K.; Refsum, H.; Brattstro¨m, L.; Boers,
G.; Ueland, P.; Rubba, P.; Palma-Reis, R.; Meleady, R.; Daly,
L.; Witteman, J .; Graham, I. Low circulating folate and vitamin
B6 concentrations risk factors for stroke, peripheral vascular
disease, and coronary disease. Circulation 1998, 97, 437-443.
(8) Miner, S. E. S.; Cole, D. E. C.; Evrovski, J .; Forrest, Q.;
Hutchison, S.; Holmes, K.; Ross, H. J . Pyridoxine improves
endothelial function in cardiac transplant recipients. J . Heart
Lung Transplant. 2001, 20, 964-969.
(9) Haque, W. Cardioprotective Phosphonates and Malonates. WO
Patent 0164692.

Novel Pyridoxine 5′-Phosphonates
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3687
(10) Blackburn, G. M.; Kent, D. E.; Kolkman, F. The synthesis and
metal binding characteristics of novel, isopolar phosphonate
analogues of nucleotides. J . Chem. Soc., Perkin Trans. 1 1984,
1119-1125.
(11) Berkowitz, D. B.; Bose, M.; Pfannenstiel, T. J .; Doukov, T.
R-Fluorinated phosphonates as substrate mimics for glucose
6-phosphate dehydrogenase: the CHF stereochemistry matters.
J . Org. Chem. 2000, 65, 4498-4508.
(12) O’Hagen, D.; Rzepa, H. S. Some influences of fluorine in
bioorganic chemistry. J . Chem. Soc., Chem Commun. 1997, 645-
652.
(13) Denesyuk, A. I.; Denessiouk, K. A.; Korpela, T.; J ohnson, M. S.
Functional Attributes of the phosphate group binding cup of
pyridoxal phosphate-dependent enzymes. J . Mol. Biol. 2002, 316,
155-172.
(14) Hullar, T. L. Pyridoxal phosphate: I phosphonic acid analogues
of pyridoxal phosphate. Synthesis via Wittig reactions and
enzyme evaluation. J . Med. Chem. 1969, 12, 58-63.
(15) Stirtan, W. G.; Withers, S. G. Phosphonate and R-fluorophos-
phonate analogue probes of the ionization state of pyridoxal 5′-
phosphate (PLP) in glycogen. Biochemistry 1996, 35, 15057-
15064.
(20) Pudovik, A. N.; Konovalova, I. V. Addition reactions of esters of
phosphorous III acids with unsaturated systems. Synthesis 1979,
81-96.
(21) Ferrari, R.; Pepi, P.; Ferrari, F.; Nesta, F.; Benigno, M.; Visioli,
O. Metabolic Derangement in Ischemic Heart Disease and its
Therapeutic Control. Am. J . Cardiol. 1998, 82 (5A), 2K-13K.
(22) Dellborg, M.; Herlitz, J .; Risenfors, M.; Swedberg, K. Electro-
cardiographic assessment of infarct size: comparison between
QRS scoring of 12-lead electrocardiography and dynamic vector
cardiography. Int. J . Cardiol. 1993, 40, 167-172.
(23) Zacharowski, K.; Blackburn, B.; Thiemermann, C. Ranolazine,
a
partial fatty acid oxidation inhibitor, reduces myocardial
infarct size and cardiac troponin C release in the rat. Eur. J .
Pharmacol. 2001, 418, 105-110.
(24) Kantor, P. F.; Dyck, J . R. B.; Lopaschuk, G. D. Fatty acid
oxidation in the reperfused ischemic heart. Am. J . Med. Sci.
1999, 318, 3-14.
(25) Lopaschuk, G. D. Optimizing cardiac energy metabolism: a new
approach to treating ischemic heart disease. Eur. Heart J . Suppl.
1999, 1 (Suppl. O), O32-O39.
(16) Kim, Y.-C.; Camaioni, E.; Ziganshin, A. U.; J i, X.-D.; King, B.
F.; Wildman, S. S.; Rychkov, A.; Yoburn, J .; Kim, H.; Mohanram,
A.; Harden, T. K.; Boyer, J . L.; Burnstock, G.; J acobson, K. A.
Synthesis and structure-activity relationships of pyridoxal-6-
arylazo-5′-phosphate and phosphonate derivatives as P2 receptor
antagonists. Drug Dev. Res. 1998, 45, 52-66.
(17) Dhalla, N. S.; Sethi, R.; Dakshinamurti, K. Treatment of
Cardiovascular and Related Pathologies. US Patent 6,043,259.
(18) Korytnyk, W.; Ikawa, M. Synthesis of vitamin B6 analogues.
Methods Enzymol. 1970, 18A, 524-566.
(26) Bersin, R. M.; Stacpoole, P. W. Dichloroacetate as metabolic
therapy for myocardial ischemia and failure. Am. Heart J . 1997,
134, 841-855.
(27) Lopaschuk, G. D.; Barr, R. L. Measurement of fatty acid and
carbohydrate metabolism in isolated working rat hearts. Mol.
Cell Biochem. 1997, 172, 137-147.
(28) Lopaschuk, G. D.; Wambolt, R. B.; Barr, R. L. An imbalance
between glycolysis and glucose oxidation is a possible explana-
tion for the detrimental effects of high levels of fatty acid during
aerobic reperfusion of ischemic hearts. J . Pharmacol. Exp. Ther.
1993, 264, 135-144.
(19) Brooks, H. G., J r.; Laasko, J . W.; Metzler, D. E. Synthesis of
vitamin B6 derivatives 1. Preparation of some isopyridoxamine
derivatives from isopyridoxal. J . Heterocycl. Chem. 1966, 126-
128.
J M0300678