Structure-activity relationships of pyridoxal phosphate derivatives as potent and selective antagonists of P2X1 receptors

Article abstract of DOI:10.1021/jm9904203

Novel analogues of the P2 receptor antagonist pyridoxal-5′-phosphate 6-azophenyl-2′,5′ disulfonate (2) were synthesized and studied as antagonists in functional assays at recombinant rat P2X₁, P2X₂, and P2X₃ receptors expressed in Xenopus oocytes (ion flux stimulation) and at turkey erythrocyte P2Y₁ receptors (phospholipase C activation). Selected compounds were also evaluated as antagonists of ion flux and the opening of a large pore at the recombinant human P2X₇ receptor. Modifications were made in the 4-aldehyde and 5′-phosphate groups of the pyridoxal moiety: i.e. a CH₂OH group at the 4-position in pyridoxine was either condensed as a cyclic phosphate or phosphorylated separately to form a bisphosphate, which reduced potency at P2 receptors. 5-Methylphosphonate substitution, anticipated to increase stability to hydrolysis, preserved P2 receptor potency. At the 6-position, halo, carboxylate, sulfonate, and phosphonate variations made on the phenylazo ring modulated potency at P2 receptors. The p-carboxyphenylazo analogue, 4, of phosphate 2 displayed an IC₅₀ value of 9 nM at recombinant P2X₁ receptors and was 1300-, 16-, and > 10000-fold selective for P2X₁ versus P2X₂, P2X₃, and P2Y₁ subtypes, respectively. The corresponding 5-methylphosphonate was equipotent at P2X₁ receptors. The 5-methylphosphonate analogue containing a 6-[3,5-bis(methylphosphonate)]phenylazo moiety, 9, had IC₅₀ values of 11 and 25 nM at recombinant P2X₁ and P2X₃ receptors, respectively. The analogue containing a phenylazo 4-phosphonate group, 11, was also very potent at both P2X₁ and P2X₃ receptors. However, the corresponding 2,5-disulfonate analogue, 10, was 28-fold selective for P2X₁ versus P2X₃ receptors. None of the analogues were more potent at P2X₇ and P2Y₁ receptors than 2, which acted in the micromolar range at these two subtypes.

Full text of DOI:10.1021/jm9904203

340
J . Med. Chem. 2001, 44, 340-349
Str u ctu r e-Activity Rela tion sh ip s of P yr id oxa l P h osp h a te Der iva tives a s P oten t
a n d Selective An ta gon ists of P 2X1 Recep tor s
Yong-Chul Kim,† Sean G. Brown,‡ T. Kendall Harden,§ J ose´ L. Boyer,§ George Dubyak,| Brian F. King,‡
Geoffrey Burnstock,‡ and Kenneth A. J acobson*,†
Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney
Diseases, NIH, Bethesda, Maryland 20892-0810, Autonomic Neuroscience Institute, Royal Free and University College Medical
School, Rowland Hill Street, London NW3 2PF, U.K., Department of Pharmacology, University of North Carolina School of
Medicine, Chapel Hill, North Carolina 27599-7365, and Department of Physiology and Biophysics, Case Western Reserve
University School of Medicine, Cleveland, Ohio 44106
Received August 19, 1999
Novel analogues of the P2 receptor antagonist pyridoxal-5′-phosphate 6-azophenyl-2′,5′-
disulfonate (2) were synthesized and studied as antagonists in functional assays at recombinant
rat P2X1, P2X2, and P2X3 receptors expressed in Xenopus oocytes (ion flux stimulation) and at
turkey erythrocyte P2Y1 receptors (phospholipase C activation). Selected compounds were also
evaluated as antagonists of ion flux and the opening of a large pore at the recombinant human
P2X7 receptor. Modifications were made in the 4-aldehyde and 5′-phosphate groups of the
pyridoxal moiety: i.e. a CH2OH group at the 4-position in pyridoxine was either condensed as
a cyclic phosphate or phosphorylated separately to form a bisphosphate, which reduced potency
at P2 receptors. 5-Methylphosphonate substitution, anticipated to increase stability to
hydrolysis, preserved P2 receptor potency. At the 6-position, halo, carboxylate, sulfonate, and
phosphonate variations made on the phenylazo ring modulated potency at P2 receptors. The
p-carboxyphenylazo analogue, 4, of phosphate 2 displayed an IC50 value of 9 nM at recombinant
P2X1 receptors and was 1300-, 16-, and >10000-fold selective for P2X1 versus P2X2, P2X3, and
P2Y1 subtypes, respectively. The corresponding 5-methylphosphonate was equipotent at P2X1
receptors. The 5-methylphosphonate analogue containing a 6-[3,5-bis(methylphosphonate)]-
phenylazo moiety, 9, had IC50 values of 11 and 25 nM at recombinant P2X1 and P2X3 receptors,
respectively. The analogue containing a phenylazo 4-phosphonate group, 11, was also very
potent at both P2X1 and P2X3 receptors. However, the corresponding 2,5-disulfonate analogue,
10, was 28-fold selective for P2X1 versus P2X3 receptors. None of the analogues were more
potent at P2X7 and P2Y1 receptors than 2, which acted in the micromolar range at these two
subtypes.
In tr od u ction
may be antihypertensive. The P2Y1 subtype participates
in blood platelet aggregation, thus selective antagonists
may be useful in regulating hemostasis. It appears that
activation of the P2X3 receptor subtype mediates noci-
ception via the dorsal root ganglia, thus a selective
antagonist may be antinociceptive.10,11 Activation of the
P2X7 receptor subtype induces apoptosis in the immune
and inflammatory system, thus selective antagonists
may also be useful therapeutically.12
Adenine and uracil 5′-nucleotides act in cellular
signaling in the nervous, muscular, cardiovascular,
renal, and immune systems, through the activation of
P2 receptors.1-3 These nucleotide receptors consist of
two families of distinct structure and function: P2X
ligand-gated cation channels and G protein-coupled P2Y
receptors.4 P2X1-7 and P2Y1,2,4,6,11 designations have
been unambiguously assigned to mammalian nucleotide
receptors,5-7 although there is still uncertainty about
the correspondence of these sequences to the pharma-
cological phenotypes described prior to P2 receptor
cloning.8
The therapeutic potential of potent and selective
agonists and antagonists of P2 receptors has been
discussed, although in mainly hypothetical terms due
to the current lack of such agents.9 For example,
activation of the P2X1 subtype mediates vasoconstriction
at vascular smooth muscle, thus selective antagonists
Derivatives of pyridoxal-5′-phosphate in which an
azoaryl group is present at the 6-position were shown
to be nonselective P2 receptor antagonists.13 Compounds
1 (pyridoxal-R5-phosphate-6-azophenyl-2′,4′-disulfonic
acid) and 2 (the 2,5-disulfonate isomer) (Figure 1) have
been studied as antagonists of adenine nucleotides at
the turkey erythrocyte P2Y1 receptor14,15 and at P2X1-
like receptors in the rabbit vas deferens,16 urinary
bladder,17 isolated blood vessels,18 guinea-pig isolated
vas deferens,19 and perfused rat mesenteric arterial
bed.20 1 was also found to be a weak antagonist at the
P2X7 receptor in mouse microglial cells.21 Both 1 and 2
have been shown to be antagonists of ATP (adenosine-
5′-triphosphate) at P2X3-like receptors in sensory
neurons.22-24
* Correspondence to: Dr. K. A. J acobson, Chief, MRS, Bldg. 8A, Rm.
B1A-19, NIH, NIDDK, LBC, Bethesda, MD 20892-0810. Tel: (301) 496-
9024. Fax: (301) 480-8422. E-mail: kajacobs@helix.nih.gov.
† NIH.
‡ Royal Free Hospital School of Medicine.
§ University of North Carolina School of Medicine.
| Case Western Reserve University School of Medicine.
The present study demonstrates that both subtype
10.1021/jm9904203 CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/19/2000

SARs of Pyridoxal Phosphate Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 341
Sch em e 1. Synthesis of Anilinephosphonic Acid
Derivativesa
F igu r e 1. Structures of azo derivatives of pyridoxal-5′-
phosphate (1 and 2) and a cyclic pyridoxine-R4,5-monophos-
phate derivative (3), all of which act as P2 receptor antago-
nists.
Ta ble 1. MS and HPLCa Characterization of the
Pyridoxal-Phosphate Analogues Synthesized
FAB (M - H+)
HPLC (tR, min)
no.
formula
% yield calcd
found
A
B
5 C15H14N3O7P
7 C16H14N3O9P
9 C16H20N3O11P3
11 C14H14N3NaO8P2
13 C14H12ClN3NaO8PS
14 C15H14N3O7P
15 C16H14N3Na2O12PS2
16 C14H14N3O5P
71 378.0491 378.0494
59 422.0389 422.0383
21 522.0232 522.0221
68 436.0076 436.0071
4.6
5.2
5.2
5.5
10.50
12.9
12.4
10.0
12.8
10.2
13.4
10.1
8.7
59 469.9591 469.9608 10.5
43 378.0491 378.0481
100 579.9474 579.9475
28 334.0593 334.0607
23 327.9987 328.0004
80 635.9137 635.9103
73 545.9540 545.9545
81 432.0362 432.0339
8.2
7.2
7.2
4.5
5.5
7.8
6.7
a
Reagents: (a) (Ph)3P, CBr4 in ether, 25 ˚C, 12 h, 52%; (b)
17 C8H13NO9P2
P(OCH3)3, 80 ˚C, 6 h, 94%; (c) Pd/C, H2 in MeOH, 1 atm, 25 ˚C, 1
h, 100%; (d) TMS-Br in CH3CN, 25 ˚C, 12 h, 99%; (e) (EtO)2P(O)H,
triethylamine, tetrakis((Ph)3P)Pd(0) in toluene, 90 ˚C, 16 h, 18%.
18 C14H15N3Na2O15P2S2
19 C14H16ClN3O12P2S
20 C14H17N3O9P2
17.1
14.3
12.8
a
logues of the P2 receptor antagonists pyridoxal-5′-
phosphate and 6-azophenyl-2′,5′-disulfonate derivative
(2), in which the phosphate group was cyclized by
esterification to a CH2OH group at the 4-position in
pyridoxine, were synthesized (compounds 13-16). Fi-
nally, bisphosphate derivatives of pyridoxine (com-
pounds 18-20) were prepared (Scheme 2).
The new derivatives were characterized using NMR
and high-resolution mass spectroscopy, and purity of
95-98% was demonstrated using high-pressure liquid
chromatography (HPLC) in two different solvent sys-
tems. The compounds were tested in ion channel as-
says26 of ATP-induced current at recombinant rat P2X1,
P2X2, and P2X3 receptors, expressed in Xenopus oocytes,
using the twin-electrode voltage-clamping technique.
Several compounds (4, 8, 10) were previously reported
to antagonize agonist-induced cation flux at recombi-
nant P2X2 receptors, with IC50 values of 1-10 µM. Most
of the compounds were weak or inactive as antagonists
at the turkey erythrocyte P2Y1 receptor (Table 2), in
activation of phospholipase C activity27,28 induced by 10
nM 2-methylthioadenosine-5′-diphosphate (2-MeSADP).
Compound 11 displayed an IC50 value of 27 µM.
In inhibition of the inward current elicited by ATP
at recombinant P2X1 receptors, compounds 4-11, with
IC50 values of 9-42 nM, were found to be more potent
than 1 and 2 (IC50 values of 99 and 43 nM, respectively).
The p-carboxylate analogue, 4, having a phosphate
linkage similar to compound 1, selectively inhibited
ATP-evoked responses at P2X1 receptors with an IC50
value of 9 nM and displayed 1300-, 16-, and >10000-
fold selectivity versus P2X2, P2X3, and P2Y1 receptors,
respectively. The corresponding 5-methylphosphonate
analogue, 5, was equipotent to 4 at P2X1 receptors
(Figure 2). Thus, compound 5 was 16-fold selective for
P2X1 versus P2X3 receptors and 19-fold selective versus
P2X2 receptors. The monocarboxylic acid 5-ethylphos-
phonate analogue, 6, and the dicarboxylic acid deriva-
System A was 0.1 M triethylammonium acetate buffer:CH3CN
) 95:5 to 40:60 for 20 min with flow rate 1 mL/min. System B
was 5 mM tetrabutylammonium phosphate buffer:CH3CN 80:20
to 40:60 in 20 min with flow rate 1 mL/min.
selectivity and high affinity can be achieved through
structural modification of the pyridoxal-5′-phosphate
series of antagonists. Recently we reported14 that the
cyclic pyridoxine-R4,5-monophosphate corresponding to
2, i.e. compound 3, was a weak but selective antagonist
of ATP-evoked responses at rat P2X1 receptors. Fur-
thermore, this cyclic phosphate was inactive at phos-
pholipase C-coupled P2Y receptors, at the adenylate
cyclase-coupled P2Y receptors in rat C6 glioma cells, and
at adenosine receptors. In the present study, additional
analogues of both pyridoxal and pyridoxine are explored,
resulting in enhanced potency and/or selectivity at P2X1
receptors. In certain analogues, the polyanionic nature
of the derivatives is diminished, and in other analogues
the stability is potentially enhanced through the intro-
duction of a phosphonate linkage. As antagonists, many
of the compounds were found to be more freely revers-
ible, in comparison to 1 and 2. Some of the present 5′-
phosphate and 5′-phosphonate derivatives were reported
in a preliminary study25 and are now characterized more
fully at a broader range of recombinant P2X receptor
subtypes.
Resu lts
We have introduced 5′-phosphonate linkages,25 cyclic
phosphate diesters,14 bisphosphates, and modified func-
tional groups on the phenylazo moiety (Table 1), with
the aim of developing more potent and selective antago-
nists for P2 receptor subtypes. Halo, carboxylate, sul-
fonate, and phosphonate (Scheme 1) variations were
made on the phenyl ring (compounds 4-11). 5-Meth-
ylphosphonate and 5-ethylphosphonate substitutions of
the methylphosphate moiety were introduced. Ana-

342 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Kim et al.
Sch em e 2. Azo Coupling Reaction of Pyridoxal- and
The cyclic phosphates and bisphosphates derived from
pyridoxine (3, 13-20) were much less potent than the
pyridoxal derivatives in antagonizing the activation of
recombinant P2X receptors expressed in oocytes. The
cyclic phosphate analogue unsubstituted on the arylazo
ring, 16, was inactive.
Pyridoxine-Phosphate Derivativesa
Compound 15, a cyclic phosphate in which the 3-OH
group was acetylated, was completely inactive at P2X1
receptors, which in comparison to 3 demonstrates the
importance of the hydroxyl group for receptor recogni-
tion.
Selected derivatives of each major class of derivatized
pyridoxal-phosphate were also tested for relative an-
tagonistic actions on human recombinant P2X7 recep-
tors expressed in HEK (human embryonic kidney) 293
cells (Table 3). For comparison, three additional ana-
logues of 2, e.g. 21-23, and a nucleotide antagonist of
P2Y1 receptors,28 24, were examined in the P2X7 recep-
tor assay. Activation of this particular P2X family
receptor induces two distinct changes in plasma mem-
brane permeability: (1) a rapidly gated cation channel
permeable to Na+, K+, and Ca2+ and (2) a delayed pore
that is nonselectively permeable to both inorganic ions
and various organic molecules up to 800 Da in mass.
We first assayed the rapidly gated cation channel
activity by assaying the ATP-induced loss of total cell
K+. Given the strong inhibitory effects of extracellular
divalent cations on P2X7 receptor function,29 these
assays were performed using cells suspended in media
containing physiological concentrations of Mg2+ and
Ca2+ or in media containing no added Mg2+ or Ca2+. In
physiological saline containing 1.5 mM CaCl2 and 1 mM
MgCl2, the EC50 for ATP was 1200 µM (as assayed by
K+ efflux). Thus, we used 3 mM ATP as a maximally
active concentration of agonist. Under these conditions,
30 µM 2 produced an 81 ( 10% inhibition (n ) 3
experiments) of the K+ release triggered by 3 mM ATP.
None of the other derivatized pyridoxal-phosphate
compounds matched the efficacy of 2. Compound 10 was
the next most efficacious inhibitor (51 ( 6% inhibition)
followed by the related compound 24. A similar rank
order of efficacy was observed using HEK-hP2X7 cells
a
Reagents: TMS-polyphosphate, benzene, 2 days, either (a) 5%
solution, 40 ˚C or (b) 20% solution, 85 ˚C; (c) aromatic amine,
NaNO2, 6 N HCl, pH 10-11.
tive, 7, were slightly less potent than 5 at P2X1
receptors. Thus, the phosphonate substitution of the
5-methylphosphate moiety, anticipated to increase sta-
bility of the analogues to hydrolysis, was found to be
well-tolerated at P2X receptor subtypes.
Phosphonate substitution of the arylazo ring proved
to have distinctive effects on potency and selectivity. The
IC50 values for the bis(methylphosphonate), 9, at P2X1
and P2X3 receptors were 11 and 25 nM, respectively,
in the presence of 1 and 3 µM ATP, respectively, thus
the P2X3 receptor potency was enhanced. Compound 9
was approximately 25- and 1300-fold selective for the
P2X1 subtype versus P2X2 and P2Y1 receptors, respec-
tively. Compound 10, a 5-methylphosphonate having the
same arylazo sulfonate substitution as compound 2,
antagonized responses at P2X1 receptors with an IC50
value of 12 nM. Compound 10 was 28-fold selective for
P2X1 versus P2X3 receptors and 92-fold selective versus
P2X2 receptors. A p-phenylphosphonate derivative, 11
(Figure 3), was nearly as potent as 9 at both P2X1 and
P2X3 receptors. Thus, 11 was 79-fold selective for P2X1
versus P2X2 receptors.
The time course of washout at P2X1 and P2X3 recep-
tors of various potent antagonists was examined (Figure
4). Ion channel activity was more rapidly restored,
following a 20-min washout, for several active com-
pounds than for 1. The percent of control activity
recovered for 1 following washout was only 30-40%.
The action of 2 was fully reversible at both receptors,
and most other analogues (4, 8-11), except for com-
pounds 5 and 7 with 20-30%, displayed g75% recovery.
tested in medium lacking extracellular Mg2+ and Ca2+
.
The relative efficacy of all inhibitory compounds, includ-
ing 2, was reduced in the absence of extracellular Mg2+
and Ca2+
.
We next assayed the effects of these compounds on
ATP-induced pore formation by the P2X7 receptor.
Because pore formation is very strongly repressed by
extracellular Mg2+ and Ca2+, these experiments were
performed using divalent cation-free saline. Under these
assay conditions the EC50 for ATP-induced activation
of the pore was 120 µM. Thus, we used 300 µM ATP as
a maximally active concentration of the agonist. 30 µM
2 completely repressed ATP-induced pore formation.
Compound 10, the phosphonate derivative of 2, was
equally efficacious (100% inhibition at 30 µM). Introduc-
tion of bis(methylphosphonate) moieties into the pyri-
doxal-phosphate core structure (compound 9) further
reduced the antagonistic action at the human P2X7
receptor. Cyclic phosphate (compound 3) and bisphos-
phate derivatives (compounds 18 and 19) were inactive.
The same rank order of efficacy (2 > 10 > 24 > 9, 22,
21 . 3, 18, 19) characterized the inhibitory effects of

SARs of Pyridoxal Phosphate Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 343
Ta ble 2. Structures and Pharmacological Activities of Pyridoxal and Pyridoxine Derivatives (phosphates and phosphonates) at P2
Receptors
PLC assay
IC50 (µM)
positions
recombinant IC50 (µM)
a
a
a
c
compd
2′
3′
4′
5′
X
P2X1 or % inhib
P2X2
P2X3
P2Y1
1
SO3H
H
H
SO3H
H
CH2O
0.099 ( 0.006
1.6 ( 0.1
0.240 ( 0.038 16.6 ( 2.5
(PPADS)
2
SO3H
H
SO3H
CH2O
0.043 ( 0.018
0.398 ( 0.125 0.084 ( 0.004 21.4 ( 9.0
(IsoPPADS)
4d,f
5
H
H
H
H
H
H
H
H
COOH
COOH
COOH
H
H
H
H
H
H
H
CH2O
CH2
0.0094 ( 0.0017 11.9 ( 1.4
0.0081 ( 0.0021 0.150 ( 0.020 0.128 ( 0.019
2.4 ( 0.3
0.140 ( 0.011 ∼100
6d
CH2CH2 0.020 ( 0.003
CH2
CH2O
0.145 ( 0.057 ∼100
7
COOH
H
H
SO3H
H
COOH
H
CH2P(O)(OH)2 CH2
SO3H
H
0.037 ( 0.008
0.042 ( 0.006
0.011 ( 0.005
0.012 ( 0.003
0.012 ( 0.008
5.9 ( 1.8 (EC50
10.2 ( 2.6
43 ( 4%b
0.486 ( 0.023 0.330 ( 0.014
8d
1.2 ( 0.2
0.480 ( 0.090 54 ( 3
9f
CH2P(O)(OH)2
H
H
0.280 ( 0.030 0.025 ( 0.007 14.5 ( 2.1
10d,f
11f
12e
3e
CH2
CH2
1.1 ( 0.2
0.340 ( 0.040 46 ( 21
P(O)(OH)2
0.948 ( 0.019 0.036 ( 0.010 27 ( 3
g
)
inactive
inactive
inactive
58.3 ( 0.1
inactive
inactive
SO3H
Cl
H
SO3H
H
H
H
H
H
H
H
H
SO3H
SO3H
H
SO3H
H
H
H
H
13
14
15
16
17
18
19
20
COOH
H
H
39 ( 10%b
inactive
inactive
CH3CO
H
inactive
inactive
∼100
inactive
SO3H
Cl
H
H
H
H
H
H
H
SO3H
SO3H
H
27 ( 7%b
78 ( 7%b
30 ( 10%b
∼100
50 ( 4%b
>100
a
Inhibition of ion current (mean ( SEM, n ) 4), unless noted, induced by ATP (at the EC70 values in µM for respective subtypes: P2X1
1, P2X2 10, and P2X3 3) at recombinant P2X receptors expressed in Xenopus oocytes (b% inhibition at 10 µM). c Inhibition of 10 nM
d
2-MeSADP-stimulated phospholipase C in turkey erythrocyte membranes (mean ( SEM, n ) 3), labeled using [3H]inositol. Compounds
reported by Kim et al.25 e Compounds and values reported by J acobson et al.14 f Compound 4, MRS 2159; compound 5, MRS 2284;
compound 9, MRS 2257; compound 10, MRS 2191; compound 11, MRS 2273. g Compound 12 potentiates the effect of ATP at P2X receptors.14
all tested compounds on both indices of human P2X7
receptor activity. The apparent efficacies of the three
most effective compounds (2, 9, 10) as human P2X7
antagonists were significantly reduced in the presence
of physiological divalent cation concentrations. In ad-
dition, these three compounds were significantly less
potent and efficacious antagonists of the murine P2X7
receptor even when assayed in the absence of divalent
cations (data not shown). Previous studies have noted
significant differences in the pharmacological antago-
nism of human versus rodent P2X7 receptors.29
decreased at the P2X7 receptor.
Functional group variation on the 6-(phenylazo) sub-
stituent was found to greatly modulate the potency as
P2 receptor antagonists. For example, two analogues
bearing phosphonate groups on the arylazo ring, i.e.
compounds 9 and 11, were selective for “group 1” (P2X1
and P2X3 receptors) versus “groups 2-4” (P2X2,4,7
receptors.
)
The view that 1 is an irreversible antagonist stemmed
originally from pharmacokinetic data on blockade of
P2X receptors in whole tissues. Compound 1 was noted
as having a slow on-rate (slow equilibration) and slow
off-rate (slow reversibility) at native P2X receptors in
vascular and visceral smooth muscle (for a review, see
ref 13). The fact that antagonism does eventually
reverse resulted in the term “pseudo-irreversible” to
describe the slow kinetics of these blocking agents.32
Nonetheless, problems associated with diffusion, surface
adhesion, uptake, compartmentalization, and re-release
of 1 have never been clearly resolved where whole tissue
assays are employed to test their activity. Most of these
problems are avoided if recombinant P2X receptors are
used instead, since the antagonist has direct access to
the cell surface and receptor and drug concentration can
be controlled by fast application and washout.
Discu ssion
Compound 1 and its analogues have been studied at
a variety of P2 receptor subtypes. At P2Y1 receptors, a
range of potencies of 1 has been reported, from ap-
proximately 1 µM30 to 16.6 µM in the present study.
Boarder and colleagues31 also showed that 1 had a
curious potentiating effect at turkey P2Y1 receptors,
which may be based on the inhibition of locally released
ATP by a mechanosensory mechanism (stimulated by
washing cells). Nevertheless, the potency of analogues
of 2 prepared in the present study generally tended to
be greater at P2X1-3 subtypes than at P2Y1 receptors.
Furthermore, among the present set of analogues in
which the aldehyde group remained, potencies often
increased over 2 at the P2X1 and P2X3 subtypes and
The slow onset and slow reversal of blockade by 1 at
the recombinant P2X1, P2X2, and P2X3 receptors (rat)

344 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Kim et al.
F igu r e 2. Potent effects of compounds 5 (b) and 7 (9) on inward current induced by ATP at recombinant rat P2X1 (A) receptors
and intermediate potency at rat P2X2 (B) and rat P2X3 (C) receptors. Compounds 5 and 7 (10 µM) were inactive at rat P2X4 (D)
receptors. Receptors were expressed in Xenopus oocytes and current was measured using the twin-electrode voltage-clamping
technique (pH 7.5, Ba2+ Ringer’s solution). All data points were mean ( SEM of 4 observations. ATP concentrations were as in
Materials and Methods. IC50 values are given in Table 2. Hill coefficients (nH) were: (A) -1.20 ( 0.07 (5) and -1.19 ( 0.08 (7);
(B) -1.11 ( 0.009 (5) and -0.95 ( 0.15 (7); (C) -0.87 ( 0.04 (5) and -0.89 ( 0.05 (7).
subunits (K249 on rP2X1, K246 on rP2X2, K251 on
rP2X3).33 However, evidence for this assumption is far
from clear. Point mutation of the lysine in the rP2X2
subunit (K246E) did increase the rate of recovery from
blockade by 1, but it did not significantly affect the
potency of the antagonist.33 On the other hand, substi-
tution of glutamate for lysine at the rat P2X4 subunit
(E249K) increased the potency of 1 (IC50 ) >500 µM
(wt) and 2.6 µM (mutant)) yet did not noticeably
accelerate recovery from blockade.33 A similar introduc-
tion of lysine in the P2X6 subunit (L251K) increased the
potency of 1 (IC50 ) >500 µM (wt) and 2.0 µM (mutant))
without affecting the rate of recovery.34
Human and mouse forms of P2X4 do not possess a
lysine residue at position 249 yet are significantly more
F igu r e 3. Selectivity of compound 11 on inward current
induced by ATP at recombinant rat P2X1 (9) and rat P2X3 (b)
sensitive to 1 (27.5 and 21 µM, respectively) than
rP2X4.35,36 Although sensitive to the antagonist, human
and mouse P2X4 receptors also recover very slowly from
blockade by 1. Mutation of rP2X4 (N127K) to provide a
lysine unique to hP2X4 did not enhance the sensitivity
of the rat orthologue to 1.35 Furthermore, mP2X4 only
possesses extracellular lysyl residues that are common
to rP2X4,36 and consequently, the heightened sensitivity
to 1 of the former cannot be ascribed to a strategic
lysine. The P2X3 subunit lacks a lysine residue at the
receptors (“group 1”) versus rat P2X2 ([) receptors (“group 2”),
expressed in Xenopus oocytes, using the twin-electrode voltage-
clamping technique (pH 7.5, Ba2+ Ringer’s solution). All data
points were mean ( SEM of 4 observations. ATP concentra-
tions were as in Materials and Methods. IC50 values are given
in Table 2. Hill coefficients (nH) were: (P2X1) -1.05 ( 0.05,
(P2X2) -1.11 ( 0.06, and (P2X3) -1.03 ( 0.04.
led to the suggestion that the aldehyde group on the
pyridoxal moiety formed a Schiff’s base with a strategic
lysyl residue at equivalent extracellular positions in P2X

SARs of Pyridoxal Phosphate Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 345
Ta ble 3. Antagonistic Effects of Pyridoxal-Phosphate
Derivatives on the Function of Human P2X7 Receptors
Expressed in HEK Cells: Cation Channel Activity versus
Nonselective Pore
% inhibition of ATP-induced
K+ efflux
K+ efflux
ethidium
influxc
compd
(with cations)a
(cation-free)b
2
9
10
3
18
19
21d
22d
23d
24d
81 ( 10
0 ( 3
67 ( 14
-5 ( 3
28 ( 7
-5 ( 2
-7 ( 4
-5 ( 3
2 ( 2
100 ( 0
84 ( 10
100 ( 0
17 ( 15
15 ( 15
12 ( 15
92 ( 15
96 ( 7
51 ( 6
-1 ( 3
-1 ( 6
-1 ( 5
14 ( 2
4 ( 4
1 ( 2
-5 ( 6
37 ( 1
-8 ( 1
24 ( 10
40 ( 35
100 ( 0
a
Adherent HEK293 cells stably expressing the recombinant
human P2X7 receptor; antagonist, 30 µM; agonist, 3 mM ATP in
b
divalent cation-containing medium; n ) 3. Adherent HEK293
cells stably expressing the recombinant human P2X7 receptor;
antagonist, 30 µM; agonist, 3 mM ATP in divalent cation-free
medium; n ) 3. c Suspended HEK293 cells stably expressing the
recombinant human P2X7 receptor; antagonist, 30 µM; agonist,
d
300 µM ATP in divalent cation-containing medium; n ) 3. 21,
MRS 2157 ) pyridoxal-R5-phosphate-6-azophenyl-2′-chloro-5′-sul-
fonic acid; 22, MRS 2160 ) pyridoxal-R5-phosphate-6-azophenyl-
3′-chloro-4′-carboxylic acid; 23, MRS 2192 ) [4-formyl-3-hydroxy-
2-methyl-6-(2-chloro-5-sulfonylphenylazo)pyrid-5-yl]methylphos-
phonic acid;25 24, MRS 2179 ) 2′-deoxy-N6-methyladenosine-3′,5′-
bisphosphate.28
reversible antagonism at these P2X receptors. Recently,
a naphthyl derivative of 1 that also retains the aldehyde
group, pyridoxal-R5-phosphate-6-(2′-naphthylazo-6′-ni-
tro-4′,8′-disulfonic acid), was shown to be a potent
selective and reversible antagonist of P2X1 receptors.39
Some P2 receptor antagonists also may inhibit ecto-
nucleotidases, making these studies even more complex.
We are currently studying ecto-nucleotidase inhibition
by selected analogues of 2.41 Preliminary results with
the recombinant enzymes expressed in CHO cells
indicate that most of the analogues of 1 were better
inhibitors of ecto-apyrase than of ecto-ATPase. The
aldehyde group appears to be required for ecto-apyrase
inhibition. However, the efficacy of derivatives of 1 as
inhibitors of ecto-enzymes had no bearing on their
efficacy as antagonists at P2X1-3 receptors expressed
in Xenopus oocytes, since defolliculated oocytes are
largely devoid of surface enzymes that degrade ATP.42
In conclusion, we have identified highly potent and
selective antagonists of P2X1 receptors and antagonists
of combined P2X1,3 receptor selectivity. Damer et al.43
have also reported an antagonist with high affinity and
apparent P2X1 receptor selectivity: the suramin ana-
logue NF279 (8,8′-(carbonylbis(imino-4,1-phenylenecar-
bonylimino-4,1-phenylenecarbonylimino))bis(1,3,5-naph-
thalenetrisulfonic acid). A selective P2X1 receptor
antagonist may have potential utility in controlling
receptor-mediated contraction of visceral and vascular
smooth muscle (e.g. vascular hypertension and instabil-
ity of the urinary bladder detrusor muscle). A selective
P2X3 receptor antagonist may be useful in pain control.
The properties of selective P2 receptor antagonists in
the central nervous system remain to be examined.
F igu r e 4. Histogram showing the degree of recovery rat P2X1
(A) and rat P2X3 (B) receptors from full blockade by various
antagonists, 20 min after drug washout. Compounds are
indicated by their appropriate number, as shown in Table 2.
equivalent position, at which a threonine residue is
found (T235), and yet 1 is potent (IC50 ∼ 1 µM) and
recovery from blockade is rapid for both rat and human
orthologues.35,37 Thus, the critical role of a strategic lysyl
residue in condensation with the aldehyde of 1 is
difficult to reconcile given the available data.
In the present study, the IC50 value for 1 at the rP2X1
receptor was lower than at the P2X3 receptor (0.099
versus 0.240 µM; see Table 2), and the rate of recovery
was faster at the rP2X3 receptor. Compound 2 proved
to be more potent at both rP2X1 and rP2X3 receptors,
yet full recovery occurred at these receptors within 20
min of drug washout. Blockade by 2 at the P2X1-like
receptor in rat vas deferens was considered surmount-
able and a slow recovery reported.38 Similarly, blockade
by 2 was reversible at P2X receptors of rat vagus nerve
bundle,22 in which P2X3 mRNA is present. Furthermore,
it was shown that reversal of blockade by 1 at human
P2X3 receptors is accelerated by Cibacron blue which,
of itself, potentiates ATP responses at this P2X recep-
tor.40 Some of the derivatives of 1 tested in the present
study (e.g. compounds 4, 8-11) showed a level of
recovery of about 75% and greater at rP2X1 and rP2X3
receptors within 20 min of washout. Thus, it now seems
unnecessary to alter the aldehyde on the pyridoxal
moiety, as with a pyridoxine cyclic phosphate, 3,14 to
make allowances for Schiff’s base formation and obtain
Ma ter ia ls a n d Meth od s
Syn th esis. Ma ter ia ls a n d An a lytica l Meth od s. Com-
pounds 4, 6, 8, 10 and (4-formyl-3-hydroxy-2-methyl-5-pyridyl)-

346 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Kim et al.
methylphosphonic acid, as the starting compound for the
synthesis of compounds 5, 7, and 9-11, were prepared as
described in a previous report.25 Compounds 3 and 12 were
synthesized according to the literature.14 Pyridoxine and the
reagents for azo coupling reactions were purchased from
Aldrich (St. Louis, MO). Aniline-2,5-disulfonic acid was ob-
tained from K & K Laboratories, Inc. (Hollywood, CA).
1H NMR spectroscopy was performed on a Varian GEMINI-
300 spectrometer and spectra were taken in D2O or CDCl3.
The chemical shifts are expressed as relative ppm from HOD
peaks (4.78 ppm) or as ppm downfield from tetramethylsilane.
31P NMR spectra were recorded without proton decoupling
mode, at room temperature using a Varian XL-300 spectrom-
eter (121.419 MHz); orthophosphoric acid (85%) was used as
an external standard. High-resolution mass spectroscopy
(HRMS) using FAB (fast atom bombardment) or EI (electron
impact) was determined with a J EOL SX102 spectrometer, and
electron spray mass spectra were obtained using a Hewlett-
Packard 1100 LC-ESPRAY system.
(1H, s, -CHO); 31P NMR (D2O) δ 17.89 (t, J ) 20.9 Hz), 21.69
(m), 22.29 (m).
[4-For m yl-3-h ydr oxy-2-m eth yl-6-(4′-ph osph on ylph en yl-
a zo)p yr id -5-yl]m et h ylp h osp h on ic Acid Disod iu m Sa lt
(11). Following the general procedure A with 23 mg (0.115
mmol) of (4-formyl-3-hydroxy-2-methyl-5-pyridyl)methylphos-
phonic acid and 20 mg (0.115 mmol) of 29, 36 mg of 11 was
obtained (yield 68%): 1H NMR (D2O) δ 2.48 (3H, s, -CH3),
3.99 (2H, d, J ) 21.5 Hz, -CH2P-), 7.82-7.89 (4H, m, phenyl),
10.27 (1H, s, -CHO); 31P NMR (D2O) δ 12.83 (m), 17.89 (t, J
) 21.4 Hz).
P yr id oxin e-r4,5-cyclom on op h osp h a t e-6-a zop h en yl-2′-
ch lor o-5′-su lfon ic Acid Mon osod iu m Sa lt (13). Following
the general procedure A with 0.1 g (0.432 mmol) of 12 and
0.11 g (0.444 mmol) of aniline-2-chloro-5-sulfonic acid, 0.12 g
of 5 was obtained (yield 59%): 1H NMR (D2O) δ 2.47 (3H, s,
-CH3), 5.18 (2H, d, J ) 15.6 Hz, -CH2O-), 5.61 (2H, d, J )
15.6 Hz, -CH2O-), 7.56 (1H, d, J ) 8.8 Hz, phenyl), 7.67 (1H,
d, J ) 8.8 Hz, phenyl), 7.95 (1H, s, phenyl); 31P NMR (D2O) δ
5.48 (pen, J ) 15.9 Hz).
The determinations of purity were performed with a Hewlett-
Packard 1090 HPLC system using an OD-5-60 C18 analytical
column (250 mm × 4.6 mm; Separation Methods Technologies,
Inc., Newark, DE) in two different linear gradient solvent
systems. One solvent system (A) was 0.1 M triethylammonium
acetate buffer:CH3CN ) 95:5 to 40:60 for 20 min with flow
rate ) 1 mL/min. The other (B) was 5 mM tetrabutylammo-
nium phosphate buffer:CH3CN ) 80:20 to 40:60 in 20 min with
flow rate ) 1 mL/min. Peaks were detected by UV absorption
using a diode array detector.
P yr id oxin e-r4,5-cyclom on op h osp h a t e-6-a zop h en yl-4′-
ca r boxylic Acid (14). Following the general procedure A with
50 mg (0.216 mmol) of 12 and 0.03 g (0.216 mmol) of
4-aminobenzoic acid, 35 mg of 14 was obtained (yield 43%):
1H NMR (D2O) δ 2.32 (3H, s, -CH3), 5.00 (2H, d, J ) 15.6 Hz,
-CH2O-), 5.47 (2H, d, J ) 15.6 Hz, -CH2O-), 7.58 (2H, d, J
) 7.8 Hz, phenyl), 7.81 (2H, d, J ) 8.8 Hz, phenyl); 31P NMR
(D2O) δ 5.50 (pen, J ) 15.9 Hz).
O 3-Ac e t y lp y r id o x in e -r4,5-c y c lo m o n o p h o s p h a t e -6-
a zop h en yl-2′,5′-d isu lfon ic Acid Mon otr ieth yla m in e Di-
sod iu m Sa lt (15). A mixture of 23 mg (0.043 mmol) of 3, 100
µL of acetic anhydride and 100 µL of triethylamine in 1 mL of
DMF was stirred for 1 h. The reaction mixture was purified
with the same ion-exchange column chromatography in gen-
eral procedure A to give 27 mg of 15 (100% yield): 1H NMR
(D2O) δ 1.23 (9H, t, J ) 6.8 Hz, 3× -CH3), 2.42 (3H, s, -CH3),
2.49 (3H, s, -CH3), 3.15 (6H, q, J ) 6.8 Hz, 3× -CH2), 5.18
(2H, d, J ) 15.6 Hz, -CH2O-), 5.71 (2H, d, J ) 15.6 Hz,
-CH2O-), 7.91 (1H, s, phenyl), 8.05 (1H, d, J ) 7.8 Hz,
phenyl), 8.18 (1H, d, J ) 7.8 Hz, phenyl); 31P NMR (D2O) δ
5.08 (pen, J ) 15.9 Hz).
Gen er a l P r oced u r e A for Azo Cou p lin g Rea ction s. To
a stirred solution of an aniline analogue (0.11 mmol) and 13
mg (0.12 mmol) of Na2CO3 in 1 mL of H2O was added 8 mg
(0.12 mmol) of NaNO2 at 0 °C. 60 µL (0.36 mmol) of 6 N HCl
was added dropwise, and the mixture was stirred for 5-10
min at 0 °C. A solution of a pyridoxal- or pyridoxine-phosphate
analogue (0.1 mmol) in 1 mL of H2O was added at once, and
the pH of the mixture was adjusted to 10-11 by addition of 1
N NaOH (∼300 µL). The yellow color changed to red (some-
times gradually) as an indicator of the reaction occurring. After
stirring for 30 min at 0 °C and 30 min at 25 °C, the mixture
was purified by ion-exchange column chromatography using
Amberlite CG-50 resin (H+ form, weakly acidic) with the
elution of water (flow rate 0.5 mL/min). The red fraction
showing a single peak in HPLC was collected and lyophilized
to give the desired compound with more than 95% purity.
P yr idoxin e-6-azoph en yl-r4,5-cyclom on oph osph or ic Acid
Mon osod iu m Sa lt (16). Following the general procedure A
with 25 mg (0.108 mmol) of 12 and 24 mg (0.189 mmol) of
aniline hydrochloride, 10 mg of 16 was obtained (yield 28%):
1H NMR (D2O) δ 2.34 (3H, s, -CH3), 5.05 (2H, d, J ) 15.6 Hz,
-CH2O-), 5.49 (2H, d, J ) 15.6 Hz, -CH2O-), 7.37-7.39 (3H,
m, phenyl), 7.59-7.62 (2H, m, phenyl); 31P NMR (D2O) δ 5.30
(pen, J ) 15.9 Hz); HRMS (FAB-) calcd 334.0593, found
334.0607. HPLC retention time 7.2 min (purity >98%).
P yr id oxin e-r4,5-b isp h osp h a t e Tet r a a m m on iu m Sa lt
(17). To a suspension of 0.5 g (2.96 mmol) of pyridoxine in 2.5
mL of anhydrous benzene was added 5 mL of trimethylsilyl
polyphosphate under nitrogen atmosphere. The mixture was
stirred at 85 °C for 2 days and poured into 30 mL of anhydrous
ether. The white precipitate was collected by filtration, washed
with anhydrous ether, and dissolved in 30 mL of water. The
water solution was stirred at 70 °C for 12 h to cleave
multiphosphate groups and remove trimethylsilyl group, and
passed through Amberlite CG-50 resin (H+ form, weakly
acidic) with the elution of water (flow rate 0.5 mL/min). The
acidic fractions with UV absorption were combined, lyo-
phillized, redissolved 10 mL of water, and neutralized by
triethylamine (pH ∼7). The desired product was purified using
[4-For m yl-3-h ydr oxy-2-m eth yl-6-(4′-car boxyph en ylazo)-
p yr id -5-yl]m eth ylp h osp h on ic Acid Disod iu m Sa lt (5).
Following the general procedure A with 23 mg (0.1 mmol) of
(4-formyl-3-hydroxy-2-methyl-5-pyridyl)methylphosphonic acid
and 16 mg (0.12 mmol) of 4-aminobenzoic acid, 27 mg of 5 was
obtained (yield 71%): 1H NMR (D2O) δ 2.41 (3H, s, -CH3),
3.85 (2H, d, J ) 21.5 Hz, -CH2P-), 7.70 (2H, d, J ) 7.8 Hz,
phenyl), 7.88 (2H, d, J ) 7.8 Hz, phenyl), 10.24 (1H, s, -CHO;
31P NMR (D2O) δ -4.61 (t, J ) 21.4 Hz).
[4-F or m yl-3-h yd r oxy-2-m eth yl-6-(2′,5′-d ica r boxyp h en -
yla zo)p yr id -5-yl]m eth ylp h osp h on ic Acid Disod iu m Sa lt
(7). Following the general procedure A with 23 mg (0.115
mmol) of (4-formyl-3-hydroxy-2-methyl-5-pyridyl)methylphos-
phonic acid and 22 mg (0.12 mmol) of 2-aminoterephthalic acid,
25 mg of 7 was obtained (yield 59%): 1H NMR (D2O) δ 2.01
(3H, s, -CH3), 3.77 (2H, d, J ) 21.5 Hz, -CH2P-), 7.48 (1H,
d, J ) 7.8 Hz, phenyl), 7.74 (1H, d, J ) 7.8 Hz, phenyl), 8.01
(1H, s, phenyl), 10.09 (1H, s, -CHO); 31P NMR (D2O) δ -4.31
(t, J ) 21.5 Hz).
-
Sephadex-DEAE A-25 (HCO3 form) ion-exchange column
chromatography with a linear gradient elution of water and
0.5 M ammonium bicarbonate solution. The pure fractions
determined by HPLC were collected and repeated lyophiliza-
tion with water gave 0.215 g of 17 (yield 18%): 1H NMR (D2O)
δ 2.43 (3H, s, -CH3), 4.92 (2H, s, -CH2O-), 4.94 (2H, d, J )
2.9 Hz, -CH2O-), 7.86 (1H, s, H-6); 31P NMR (D2O) δ 3.51 (t,
J ) 5.5 Hz), 4.40 (pt).
[4-F or m yl-3-h yd r oxy-2-m e t h yl-6-a zo(3′,5′-b ism e t h -
ylp h osp h on ylp h en yl)-5-p yr id yl]m eth ylp h osp h on ic Acid
Disod iu m Sa lt (9). Following the general procedure A with
46 mg (0.2 mmol) of (4-formyl-3-hydroxy-2-methyl-5-pyridyl)-
methylphosphonic acid and 56 mg (0.2 mmol) of 1-amino-
benzene-3,5-bis(methylphosphonic acid), 24 mg of 9 was
obtained (yield 21%): 1H NMR (D2O) δ 2.54 (3H, s, -CH3),
3.19 (4H, d, J ) 21.5 Hz, 2× -CH2P-), 4.05 (2H, d, J ) 21.5
Hz, -CH2P-), 7.39 (1H, s, phenyl), 7.76 (2H, s, phenyl), 10.36
P yr id oxin e-r4,5-b isp h osp h a t e-6-a zop h en yl-2′,5′-d isu l-
fon ic Acid Disod iu m Sa lt (18). Following the general

SARs of Pyridoxal Phosphate Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 347
procedure A with 25 mg (0.063 mmol) of 17 and 16 mg (0.063
mmol) of aniline-2,5-disulfonic acid, 32 mg of 18 was obtained
(yield 80%): 1H NMR (D2O) δ 2.72 (3H, s, -CH3), 5.32 (2H, d,
J ) 8.8 Hz, -CH2O-), 5.68 (2H, d, J ) 6.8 Hz, -CH2O-),
8.04 (1H, d, J ) 7.8 Hz, phenyl), 8.13 (1H, s, phenyl), 8.18
(1H, d, J ) 7.8 Hz, phenyl); 31P NMR (D2O) δ -0.06 (t, J )
6.8 Hz), 1.67 (pt).
P yr id oxin e-r4,5-bisp h osp h a te-6-a zop h en yl-2′-ch lor o-5′-
su lfon ic Acid Mon osod iu m Sa lt (19). Following the general
procedure A with 25 mg (0.063 mmol) of 17 and 16 mg (0.063
mmol) of aniline-2-chloro-5-sulfonic acid, 26 mg of 19 was
obtained (yield 73%): 1H NMR (D2O) δ 2.68 (3H, s, -CH3),
5.31 (2H, d, J ) 8.8 Hz, -CH2O-), 5.67 (2H, d, J ) 6.8 Hz,
-CH2O-), 7.79 (1H, d, J ) 7.8 Hz, phenyl), 7.92 (1H, dd, J )
2.0, 7.8 Hz, phenyl), 8.18 (1H, d, J ) 2.0 Hz, phenyl); 31P NMR
(D2O) δ -0.06 (t, J ) 6.2 Hz), 1.69 (pt).
P yr idoxin e-6-azoph en yl-r4,5-bisph osph or ic Acid Mon o-
sod iu m Sa lt (20). Following the general procedure A with
25 mg (0.063 mmol) of 17 and 24 mg (0.189 mmol) of aniline
hydrochloride, 22 mg of 20 was obtained (yield 81%): 1H NMR
(D2O) δ 2.54 (3H, s, -CH3), 5.20 (2H, d, J ) 7.8 Hz, -CH2O-
), 5.55 (2H, d, J ) 4.9 Hz, -CH2O-), 7.57-7.59 (3H, m,
phenyl), 7.93-7.96 (2H, m, phenyl); 31P NMR (D2O) δ 0.08 (t,
J ) 5.2 Hz), 1.54 (t, J ) 7.9 Hz).
3,5-Bis(br om om eth yl)-1-n itr oben zen e (25). A solution
of 1.89 g of 5-nitro-m-xylene-(R,R′)-diol (10 mmol), 5.25 g of
triphenylphosphine (20 mmol), and 6.62 g of carbon tetrabro-
mide (20 mmol) in 50 mL of anhydrous ether was stirred at
25 °C for 12 h under N2 atmosphere. After evaporation, the
residue was purified with flash silica gel column chromatog-
raphy (Hex/EtOAc ) 5/1) to give 1.6 g of 25 as a bright yellow
solid (52%): 1H NMR (CDCl3) δ 4.53 (4H, s, 2× -CH2Br), 6.83
(1H, s, Ph), 7.76 (1H, s, Ph), 8.20 (1H, s, Ph); MS (EI) (M+)
309; HRMS (EI) calcd 306.8843, found 306.8859.
280.0127; HPLC retention time 4.1 min (purity >98%) using
solvent system A, 7.8 min (>98% purity) using solvent system
B.
Tetr a eth yl 4-Am in op h en ylp h osp h on a te (28). A mixture
of 0.69 g (4.0 mmol) of 4-bromoaniline, 0.57 mL (4.4 mmol) of
diethyl phosphite, 0.61 mL (4.4 mmol) of triethylamine and
0.23 g (0.2 mmol) of tetrakis(triphenylphosphine)palladium-
(0) in 1 mL of anhydrous toluene was heated at 90 °C for 16
h under N2 atmosphere. After cooling, 10 mL of diethyl ether
was added, and the precipitate was filtered off. The filtrate
was concentrated and purified with preparative thin-layer
chromatography (CHCl3/MeOH ) 20/1) and crystallized in
diethyl ether to give 0.18 g of 28 as a white solid (18%): 1H
NMR (CDCl3) δ 1.31 (6H, t, J ) 6.8 Hz, 2× -CH3), 3.98-4.14
(4H, m, -OCH2-), 6.69, 6.71 (2H, 2d, J ) 8.8 Hz, Ar), 7.56,
7.61 (2H, 2d, J ) 8.8 Hz, Ar); 31P NMR (CDCl3) δ 21.24 (m);
MS (EI) (M+) 229; HRMS (EI) calcd 229.0868, found 229.0867.
4-Am in op h en ylp h osp h on ic Acid (29). To a solution of
40 mg of 28 (0.18 mmol) in 2 mL of anhydrous CH3CN was
added 0.12 mL of trimethylsilyl bromide (0.88 mmol) at 25 °C
under N2 atmosphere. The mixture was stirred for 12 h, and
the solvent was removed by N2 stream. The residue was
partitioned between diethyl ether and 0.5 M ammonium
bicarbonate solution, and the aqueous fraction was purified
with the same ion exchange column chromatography in general
procedure to give 30 mg of 29 (99% yield): 1H NMR (D2O) δ
7.38, 7.39 (2H, 2d, J ) 8.8 Hz, Ar), 7.80, 7.84 (2H, 2d, J ) 8.8
Hz, Ar); 31P NMR (D2O) δ 12.00 (m); MS (FAB+) (M + H) 174;
HRMS (FAB+) calcd 174.0320, found 174.0327; HPLC reten-
tion time 2.1 min (purity >98%) using solvent system A, 3.3
min (>98% purity) using solvent system B.
P h a r m a cology. Precautions must be taken to shield these
P2X antagonists from light. In aqueous solution and upon
exposure to light, azo-containing derivatives, such as 1 and 2,
had half-lives of several hours (G. Semple, unpublished
observations). Both compounds were stable for days in solution
when kept in the dark and for over a month as solids in clear
bottles on the shelf.
An ta gon ist Activity a t P 2X Recep tor s. At r a t P 2X1,
P 2X2, or P 2X3 r ecep tor s: Xenopus oocytes were harvested
and prepared as previously described.26 Defolliculated oocytes
were injected cytosolically with rat P2X1, P2X2 or P2X3 receptor
cRNA (40 nL, 1 µg/mL), incubated for 24 h at 18 °C in Barth’s
solution and kept for up to 12 days at 4 °C until used in
electrophysiological experiments.
ATP-activated membrane currents (Vh ) -90 mV) were
recorded from cRNA-injected oocytes using the twin-electrode
voltage-clamp technique (Axoclamp 2B amplifier). Voltage
recording (1-2 MΩ tip resistance) and current-recording
microelectrodes (5 MΩ tip resistance) were filled with 3.0 M
KCl. Oocytes were held in an electrophysiological chamber and
superfused with Ringer’s solution (5 mL/min, at 18 °C)
containing (mM) NaCl, 110; KCl, 2.5; HEPES (N-[2-hydroxy-
ethyl]piperazine-N′-[3-propanesulfonic acid]), 5; BaCl2, 1.8,
adjusted to pH 7.5.
ATP (at the EC70 values in µM for respective subtypes:
P2X1, 1; P2X2, 10; P2X3, 3) was superfused over the oocytes
for 60-120 s then washed out for a period of 20 min. For
inhibition curves, data were normalized to the current evoked
by ATP, at pH 7.5. Test substances were added for 20 min
prior to ATP exposure; all compounds were tested for revers-
ibility of their effects. The concentration required to inhibit
the ATP response by 50% (IC50) was taken from Hill plots
constructed using the formula: log(I/Imax - I), where I is the
current evoked by ATP in the presence of an antagonist. Data
are presented as mean ( SEM (n ) 4) for data from different
batches of oocytes.
Tetr a m eth yl 1-Am in oben zen e-3,5-bis(m eth ylp h osp h o-
n a te) (26). A solution of 1.4 g of 25 (4.53 mmol) in 15 mL of
trimethyl phosphite was heated at 80 °C for 6h. After evapora-
tion, the residue was purified with flash silica gel column
chromatography (CHCl3/MeOH ) 30/1) to give 1.57 g of
tetramethyl 1-nitrobenzene-3,5-bis(methylphosphonate) as a
white solid (94%): 1H NMR (CDCl3) δ 3.25 (4H, d, J ) 21.5
Hz, 2× -CH2P-), 3.72 (3H, s, -CH3), 3.73 (3H, s, -CH3), 3.75
(3H, s, -CH3), 3.76 (3H, s, -CH3), 7.28 (1H, s, Ph), 7.60 (1H,
s, Ph), 8.06 (1H, s, Ph); 31P NMR (CDCl3) δ 27.09 (m), 33.48
(m); MS (EI) (M+) 367 HRMS (EI) calcd 367.0586, found
367.0592.
A solution of 40 mg of tetramethyl 1-nitrobenzene-3,5-bis-
(methylphosphonate) (4.53 mmol) and 10 mg of 10% Pd/C in
3 mL of methanol was stirred at 25 °C for 1 h under H2
atmosphere. The mixture was filtered through a Celite bed
and purified with preparative thin-layer chromatography
(CHCl3/MeOH ) 20/1) to give 37 mg of 26 as a white solid
(100%): 1H NMR (CDCl3) δ 3.06 (4H, d, J ) 21.5 Hz, 2×
-CH2P-), 3.66 (6H, s, 2× -CH3), 3.70 (6H, s, 2× -CH3), 6.55
(2H, s, Ph), 6.59 (1H, s, Ph); 31P NMR (CDCl3) δ 27.09 (m),
33.48 (m): with proton decoupling off mode; MS (EI) (M+) 337;
HRMS (EI) calcd 337.0844, found 337.0845.
1-Am in oben zen e-3,5-bis(m eth ylp h osp h on ic a cid ) Tet-
r a a m m on iu m Sa lt (27). To a solution of 35 mg of 26 (0.10
mmol) in 2 mL of anhydrous CH3CN was added 0.14 mL of
trimethylsilyl bromide (1.04 mmol) at 25 °C under N2 atmos-
phere. The mixture was stirred for 12 h, and the solvent was
removed by N2 stream. The residue was partitioned between
diethyl ether and 0.5 M ammonium bicarbonate solution. The
aqueous fraction was applied to an ion-exchange chromatog-
raphy column (Sephadex-DEAE A-25 resin) and eluted with
a linear gradient (0.01 to 0.5 M) of 0.5M ammonium bicarbon-
ate. UV and HPLC were used to monitor the elution to give
35 mg of 27 (100%): 1H NMR (D2O) δ 2.84 (2H, d, J ) 20.5
Hz, -CH2P-), 2.93 (2H, d, J ) 20.5 Hz, -CH2P-), 6.60 (2H,
s, Ar), 6.66 (H, s, Ar); 31P NMR (D2O) δ 18.85 (m), 20.29 (m);
MS (FAB-) (M - H) 280; HRMS (FAB-) calcd 280.0140, found
P 2X7 Recep tor Ch a n n el Activa tion . Cation channel
activity of P2X7 receptors was assayed as the ATP-induced loss
of intracellular K+ in adherent HEK293 cells stably expressing
recombinant human P2X7 receptors. HEK293 cells were plated
at 5 × 105 cells/well on polylysine-coated wells of 12-well tissue
culture chambers and grown to 90% confluence in DMEM
(Dulbecco’s modified eagle medium) plus 10% calf serum. The

348 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Kim et al.
growth medium was removed and replaced with 1 mL of
divalent cation-containing saline consisting of 130 mM NaCl,
5 mM KCl, 20 mM NaHEPES (pH 7.5), 1 mM MgCl2, 1.5 mM
CaCl2, 10 mM glucose, and 0.1% BSA. An alternative, divalent
cation-free medium with the same basic composition but
lacking the MgCl2 and CaCl2 was also tested. The cell mono-
layers were treated with the indicated concentrations of
pyridoxal-phosphate derivative prior to the addition of 3 mM
ATP (a maximal concentration of agonist under these ionic
conditions). After a 10-min incubation at 37 °C, the assay
medium was rapidly and completely aspirated and replaced
with 1 mL 10% HNO3. After a 2-h extraction, the total K+
content of the nitric acid extract was assayed by atomic
absorbance spectrophotometry. The extent of ATP-induced K+
efflux (plus or minus the various antagonists) was measured
relative to the K+ content of untreated cells. The K+ content
of control HEK cells bathed in divalent cation-containing saline
was 531 ( 130 nmol K+/well (n ) 3; with each experiment
performed in duplicate for all tested conditions). Following a
10-min stimulation with ATP, the K+ content was reduced to
141 ( 68 nmol/well. For cells bathed in divalent cation-free
medium, the K+ content was 512 ( 60 nmol K+/well, while
the K+ content of the ATP-stimulated cells was 135 ( 36 nmol/
well. Relative efficacies of the different antagonists were
calculated as the percentage inhibition of the ATP-induced K+
loss in the absence of any pyridoxal-phosphate derivatives.
Refer en ces
(1) North, R. A.; Barnard, E. A. Nucleotide receptors. Curr. Opin.
Neurobiol. 1997, 7, 346-357.
(2) Ralevic, V.; Burnstock, G. Receptors for purines and pyrimidines.
Pharmacol. Rev. 1998, 50, 413-492.
(3) Abbracchio, M. P.; Burnstock, G. Purinergic signaling: patho-
physiological roles. J pn. J . Pharmacol. 1998, 78, 113-145.
(4) Communi, D.; Boeynaems, J . M. Receptors responsive to extra-
cellular pyrimidine nucleotides. Trends Pharmacol. Sci. 1997,
18, 83-86.
(5) Burnstock, G.; King, B. F. Numbering of cloned P2 purinoceptors.
Drug Dev. Res. 1996, 38, 67-71.
(6) Fredholm, B. B.; Abbracchio, M. P.; Burnstock, G.; Dubyak, G.
R.; Harden, T. K.; J acobson, K. A.; Schwabe, U.; Williams, M.
Towards a revised nomenclature for P1 and P2 receptors. Trends
Pharmacol. Sci. 1997, 18, 79-82.
(7) Communi, D.; Govaerts, C.; Parmentier, M.; Boeynaems, J . M.
Cloning of a human purinergic P2Y receptor coupled to phos-
pholipase C and adenylyl cyclase. J . Biol. Chem. 1997, 272,
31969-31973.
(8) King, B. F.; Townsend-Nicholson, A.; Burnstock, G. Metabotropic
Receptors for ATP and UTP: Exploring the Correspondence
Between Native and Recombinant Nucleotide Receptors. Trends
Pharmacol. Sci. 1998, 19, 506-514.
(9) J acobson, K. A.; Knutsen, L. J . S. P1 and P2 purine and
pyrimidine receptors. Handbk. Exp. Pharmacol., in press.
(10) Burnstock, G.; Wood, J . N. Purinergic receptors - their role in
nociception and primary afferent neurotransmission. Curr. Opin.
Neurobiol. 1996, 6, 526-532.
(11) Cook, S. P.; Vulchanova, L.; Hargreaves, K. M.; Elde, R.;
McCleskey, E. W. Distinct ATP receptors on pain-sensing and
stretch-sensing neurons. Nature 1997, 387, 505-508.
(12) Di Virgilio, F.; Falzoni, S.; Mutini, C.; Sanz, J . M.; Chiozzi, P.
Purinergic P2X7 receptor: A pivotal role in inflammation and
immunomodulation. Drug Dev. Res. 1998, 45, 207-213.
(13) Lambrecht, G.; Ardanuy, U.; Ba¨umert, H. G.; Bo, X.; Hoyle, C.
H. V.; Nickel, P.; Pfaff, O.; Ralevic, V.; Windschief, U.; Ziganshin,
A. U.; Ziyal, R.; Mutscheler, E.; Burnstock, G. Design and
pharmacological characterization of selective P2-purinoceptor
antagonists. In Perspectives in Receptor Research; Giardina`, D.,
Piergentili, S., Pigini, M., Eds.; Elsevier Science Publishers:
Amsterdam, 1996; pp 337-350.
(14) J acobson, K. A.; Kim, Y.-C.; Wildman, S. S.; Mohanram, A.;
Harden, T. K.; Boyer, J . L.; King, B. F.; Burnstock, G. A
pyridoxine cyclic-phosphate and its 6-arylazo-derivative selec-
tively potentiate and antagonize activation of P2X1 receptors.
J . Med. Chem. 1998, 41, 2201-2206.
(15) Boyer, J . L.; Zohn, I. E.; J acobson, K. A.; Harden, T. K.
Differential effects of P2-purinergic receptor antagonists on
phospholipase C- and adenylyl cyclase-coupled P2Y purinergic
receptors. Br. J . Pharmacol. 1994, 113, 614-620.
P 2X7 Recep tor -Dep en d en t P or e F or m a tion . The pore-
forming function of P2X7 receptors was assayed as described
previously in suspended HEK293 cells stably expressing
recombinant human P2X7 receptors.29 Following trypsiniza-
tion, the cells were suspended in the divalent cation-free saline
described above. After transfer of a 1.5-mL aliquot of cell
suspension to a thermostated (37 °C), stirred fluorimeter
cuvette, 20 µM ethidium bromide was added. The cells were
then treated with the indicated concentrations of pyridoxal-
phosphate derivative prior to the addition of 300 µM ATP (a
supramaximal concentration of agonist under these ionic
conditions). The rates of ethidium influx were fluorimetrically
monitored and calculated as described previously.29 Relative
efficacies of the different antagonists were calculated as the
percentage inhibition of the ATP-induced rate of ethidium
influx in the absence of any pyridoxal-phosphate derivatives.
P h osp h olip a se C Assa y a t P 2Y Recep tor s. P2Y1 recep-
tor-promoted stimulation of inositol phosphate formation by
2-MeSADP (10 nM) was measured in turkey erythrocyte
membranes as previously described.27,28 The values were
averaged from 3-8 independent determinations. Briefly, 1 mL
of washed turkey erythrocytes was incubated in inositol-free
medium (DMEM; Gibco) with 0.5 mCi of 2-[3H]myo-inositol (20
Ci/mmol; American Radiolabeled Chemicals Inc., St. Louis,
MO) for 18-24 h in a humidified atmosphere of 95% air 5%
CO2 at 37 °C. Erythrocyte ghosts were prepared by rapid lysis
in hypotonic buffer (5 mM sodium phosphate, pH 7.4, 5 mM
MgCl2, 1 mM EGTA) as described.26 Phospholipase C activity
was measured in 25 µL of [3H]inositol-labeled ghosts (ca. 175
µg of protein, 200-500000 cpm/assay) in a medium containing
424 µM CaCl2, 0.91 mM MgSO4, 2 mM EGTA, 115 mM KCl, 5
mM KH2PO4, and 10 mM Hepes, pH 7.0. Assays (200 µL final
volume) contained 1 µM GTPγS and the indicated concentra-
tions of nucleotide analogues. Ghosts were incubated at 30 °C
for 5 min, and total [3H]inositol phosphates were quantitated
by anion-exchange chromatography as previously described.27,28
(16) Lambrecht, G.; Friebe, T.; Grimm, U.; Windscheif, U.; Bungardt,
E.; Hildebrandt, C.; Baumert, H. G.; Spatzkumbel, G.; Mut-
schler, E. PPADS, a Novel Functionally Selective Antagonist of
P2 Purinoceptor-Mediated Responses. Eur. J . Pharmacol. 1992,
217, 217-219.
(17) Ziganshin, A. U.; Hoyle, C.; Bo, X. N.; Lambrecht, G.; Mutschler,
E.; Ba¨umert, H. G.; Burnstock, G. PPADS selectively antago-
nizes P2X purinoceptor-mediated responses in the rabbit urinary-
bladder. Br. J . Pharmacol. 1993, 110, 1491-1495.
(18) Ziganshin, A. U.; Hoyle, C. H. V.; Lambrecht, G.; Mutschler, E.;
Baumert, H. G.; Burnstock, G. Selective antagonism by PPADS
at P2X-purinoceptors in rabbit isolated blood vessels. Br. J .
Pharmacol. 1994, 111, 923-929.
(19) McLaren, G. J .; Lambrecht, G.; Mutschler, E.; Baumert, H. G.;
Sneddon, P.; Kennedy, C. Investigation of the actions of PPADS,
a
novel of P2X-purinoceptor antagonist, in the guinea-pig
isolated vas deferens. Br. J . Pharmacol. 1994, 111, 913-917.
(20) Windscheif, U.; Ralevic, V.; Ba¨umert, H. G.; Mutschler, E.;
Lambrecht, G.; Burnstock, G. Vasoconstrictor and vasodilator
responses to various agonists in the rat perfused mesenteric
arterial bed - selective-inhibition by PPADS of contractions
mediated via P2X-purinoceptors. Br. J . Pharmacol. 1994, 113,
1015-1021.
(21) Chessell, I. P.; Simon, J .; Hibell, A. D.; Michel, A. D.; Barnard,
E. A.; Humphrey, P. P. A. Cloning and functional characteriza-
tion of the mouse P2X7 receptor. FEBS Lett. 1998, 439, 26-30.
(22) Trezise DJ , Kennedy I & Humphrey PPA. The use of antagonists
to characterize the receptors mediating depolarization of the rat
isolated vagus nerve by alpha, beta-methylene adenosine 5′-
triphosphate. Br. J . Pharmacol. 1994, 112, 282-288.
(23) Rae, M. G.; Rowan, E. G.; Kennedy, C. Pharmacological proper-
ties of P2X3-receptors present in neurones of the rat dorsal root
ganglia. Br. J . Pharmacol. 1998, 124, 176-80.
Ack n ow led gm en t. S.G.B. thanks Gilead Sciences
(Foster City, CA) for funding. We thank Dr. Lewis
Pannell and Wesley White for determination of high-
resolution mass spectroscopy and NMR. We thank
Graeme Semple Ph.D., AstraZeneca R&D, Mo¨lndal,
Sweden, for alerting us to the instability of azo-contain-
ing P2 antagonists. This work was supported by USPHS
Grants GM38213, HL34322, and HL54889.
(24) Grubb, B. D.; Evans, R. J . Characterization of cultured dorsal
root ganglion neuron P2X receptors. Eur. J . Neurosci. 1999, 11,
149-154.

SARs of Pyridoxal Phosphate Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 349
(25) Kim, Y.-C.; Camaioni, E.; Ziganshin, A. U.; J i, X.-J .; 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-
azoaryl-5′-phosphate and phosphonate derivatives as P2 receptor
antagonists. Drug Dev. Res. 1998, 45, 52-66.
(26) King, B. F.; Wildman, S. S.; Ziganshina, L. E.; Pintor, J .;
Burnstock, G. Effects of extracellular pH on agonism and
antagonism at recombinant P2X2 receptors. Br. J . Pharmacol.
1997, 121, 1445-1453.
(27) Harden, T. K.; Hawkins, P. T.; Stephens, L.; Boyer, J . L.;
Downes, P. Phosphoinositide hydrolysis by guanosine 5′-(gamma-
thio)triphosphate-activated phospholipase C of turkey erythro-
cyte membranes. Biochem. J . 1988, 252, 583-593.
(28) Camaioni, E.; Boyer, J . L.; Mohanram, A.; Harden, T. K.;
J acobson, K. A. Deoxyadenosine-bisphosphate derivatives as
potent antagonists at P2Y1 receptors. J . Med. Chem. 1998, 41,
183-190.
(29) Humphreys, B. D.; Virginio, C.; Surprenant, A.; Rice, J .; Dubyak,
G. R. Isoquinolines as antagonists of the P2X7 nucleotide
receptor: high selectivity for the human versus rat receptor
homologues. Mol. Pharmacol. 1998, 54, 22-32.
(30) Charlton, S. J .; Brown, C. A.; Weismann, G. A., Turner, J . T.;
Erb, L.; Boarder, M. R. PPADS and Suramin as Antagonists at
Cloned P2y- and P2u- Receptors. Br. J . Pharmacol. 1996, 118,
704-710.
(31) Brown, C. A.; Charlton, S. J .; Boarder, M. R. Enhancement of
the Response to Purinergic Agonists in P2Y1 Transfected 1321N1
Cells by Antagonists Suramin and PPADS. Br. J . Pharmacol.
1997, 120, 1049-1052.
(35) Garcia-Guzman, M.; Stuhmer, W.; Soto, F. Molecular charac-
terization and pharmacological properties of the human P2X3
purinoceptor. Brain Res. (Mol. Brain Res.) 1997, 47, 59-66.
(36) Townsend-Nicholson, A.; King, B. F.; Wildman, S. S.; Burnstock,
G. Molecular cloning, functional characterization and possible
cooperativity between the murine P2X4 and P2X4a receptors.
Brain Res. (Mol. Brain Res.) 1999, 64, 246-254.
(37) Lewis, C.; Neidhart, S.; Holy, C.; North, R. A.; Buell, G.;
Surprenant, A. Coexpression of P2X2 and P2X3 receptor subunits
can account for ATP-gated currents in sensory neurons. Nature
1995, 377, 432-435.
(38) Khakh, B. S.; Michel, A.; Humphrey, P. P. A. Estimates of
antagonist affinities at P2X purinoceptors in rat vas deferens.
Eur. J . Pharmacol. 1994, 263, 301-309.
(39) Lambrecht, G.; Rettinger, J .; Baumert, H.; Czeche, S.; Damer,
S.; Ganso, M.; Hildebrandt, C.; Niebel, B.; Schmalzing, G.;
Mutschler, E. The pyridoxal-6-naphthylazo-5′-phosphate deriva-
tive PPNDS potently antagonizes activation of P2X1 receptors.
In press.
(40) Alexander, K.; Niforatos, W.; Bianchi, B.; Burgard, E. C.; Lynch,
K. J .; Kowaluk, E. A.; J arvis, M. F.; van Biesen, T. Allosteric
modulation and accelerated resensitization of human P2X3
Receptors by Cibacron blue. J . Pharmacol. Exp. Ther. 1999, 291,
1135-1142.
(41) Hoffmann, C.; Pradel, G.; Heine, P.; J acobson, K. A.; Zimmer-
mann, H. Inhibition of ecto-ATPase and ecto-Apyrase by pyri-
doxalphosphate-related compounds. Drug Dev. Res. 2000, in
press.
(42) Ziganshin, A. U.; Ziganshina, L. E.; King, B. F.; Pintor, J .;
Burnstock, G. Effects of P2 purinoceptor antagonists on degra-
dation of adenine nucleotides by ecto-nucleotidases in folliculated
oocytes of Xenopus laevis. Biochem. Pharmacol. 1996, 51, 897-
901.
(43) Damer S.; Niebel B.; Czeche S.; Nickel P.; Ardanuy U.; Schmalz-
ing G.; Rettinger J .; Mutschler E.; Lambrecht G. NF279: a novel
potent and selective antagonist of P2X receptor-mediated re-
sponses. Eur. J . Pharmacol. 1998, 350, R5-6.
(32) Soto F, Garcia-Guzman M & Stuhmer W. Cloned ligand-gated
channels activated by extracellular ATP (P2X receptors). J .
Membr. Biol. 1997, 160, 91-100.
(33) Buell G, Lewis C, Collo G, North RA & Surprenant A. An
antagonist-insensitive P2X receptor expressed in epithelia and
brain. EMBO J . 1996, 15, 55-62.
(34) Collo, G.; North, R. A.; Kawashima, E.; Merlo-Pich, E.; Neidhart,
S.; Surprenant, A.; Buell, G. Cloning of P2X5 and P2X6 receptors
and the distribution and properties of an extended family of ATP-
gated ion channels. J . Neurosci. 1996, 16, 2495-2507.
J M9904203