Carbonic anhydrase activators: Design of high affinity isozymes I, II, and IV activators, incorporating tri-/tetrasubstituted-pyridinium-azole moieties

Article abstract of DOI:10.1021/jm011031n

A series of tight binding carbonic anhydrase (CA) activators was obtained by reaction of amino-azoles (3-amino-pyrazole, 2-amino-imidazole, and 5-amino-tetrazole) with tri- or tetrasubstituted pyrylium salts. Many of the new pyridinium salts incorporating

Full text of DOI:10.1021/jm011031n

504
J . Med. Chem. 2002, 45, 504-510
Ca r bon ic An h yd r a se Activa tor s: Design of High Affin ity Isozym es I, II, a n d IV
Activa tor s, In cor p or a tin g Tr i-/Tetr a su bstitu ted -p yr id in iu m -a zole Moieties
Monica Ilies,† Mircea D. Banciu,‡ Marc A. Ilies,† Andrea Scozzafava,§ Miron T. Caproiu,| and
Claudiu T. Supuran*,§
Department of Chemistry, Faculty of Biotechnologies, University of Agricultural Sciences and Veterinary Medicine,
B-dul Marasti 59, 71331-Bucharest, Romania, Department of Organic Chemistry, Polytechnic University, Splaiul
Independentei 313, Bucharest, Romania, Laboratorio di Chimica Inorganica e Bioinorganica, Universita` degli Studi,
Via Gino Capponi 7, I-50121, Florence, Italy, and “C.D. Nenitzescu” Institute of Organic Chemistry,
Splaiul Independentei 202B, Bucharest, Romania
Received August 27, 2001
A series of tight binding carbonic anhydrase (CA) activators was obtained by reaction of amino-
azoles (3-amino-pyrazole, 2-amino-imidazole, and 5-amino-tetrazole) with tri- or tetrasubstituted
pyrylium salts. Many of the new pyridinium salts incorporating azole moieties reported here
proved to be efficient in vitro activators of three CA isozymes, CA I, II, and IV. Very good
activity was detected against hCA I and bCA IV (h ) human; b ) bovine isozymes), for which
some of the new compounds showed affinities in the low nanomolar range, whereas against
hCA II, their affinities were in the range of 95-150 nM. Substitution patterns of the pyridinium
ring leading to best activity included 4-phenyl-2,6-dialkyl moieties or 2,4,6-tri- and 2,3,4,6-
tetraalkyl groups. Ex vivo experiments showed some of the new activators to strongly enhance
CA activity after incubation with human erythrocytes. Furthermore, due to their cationic nature,
some of these compounds (the imidazole and pyrazole derivatives) are membrane-impermeant,
discriminating thus between cytosolic and membrane-bound CA isozymes. The present paper
is the first report of membrane-impermeant CA activators. The pyridinium tetrazole derivatives
on the other hand do penetrate through biological membranes. Such CA activators might lead
to the development of drugs/diagnostic tools for the management of CA deficiency syndromes
as well as for the pharmacological enhancement of synaptic efficacy, spatial learning, and
memory. This may constitute a new approach for the treatment of Alzheimer disease and other
conditions in need of achieving memory therapy.
In tr od u ction
portant pharmacological enhancement of synaptic effi-
cacy, spatial learning, and memory, proving that this
class of unexplored enzyme modulators may be used for
the management of conditions in which learning and
memory are impaired, such as Alzheimer’s disease or
aging. It should be also mentioned that it was previously
reported that the levels of several CA isozymes are
significantly diminished in the brain of patients affected
by Alzheimer’s disease,7 a fact strongly supporting the
involvement of CAs in cognitive functions.4,6,7
In previous contributions from this laboratory9-13 it
was shown that effective CA activators can be designed
by considering histamine 1 as a lead molecule.2 Indeed,
the X-ray crystallographic structure of the adduct of
human CA II (hCA II) with histamine, a moderate-weak
activator (activation constant, KA ) 125 µM), showed
the activator molecule to be bound at the entrance of
the active site cavity, where it is anchored by hydrogen
bonds to three amino acid side chains and to a water
molecule. These hydrogen bonds involve only the nitro-
Activation of the zinc enzyme carbonic anhydrase (CA,
EC 4.2.1.1) by a multitude of physiologically relevant
compounds such as biogenic amines (histamine, sero-
tonin, catecholamines), amino acids, or oligopeptides/
small proteins among others, has only recently been
explained at the molecular level.1-3 By means of elec-
tronic spectroscopy, X-ray crystallography, and kinetic
measurements, it has been proved that the activator
molecule binds within the enzyme active cavity at a site
distinct of the inhibitor or substrate binding sites,
participating thereafter in the rate-determining step of
the catalytic cycle, i.e., the proton-transfer processes
between the active site and the environment.1-3
Except for clarifying basic aspects of the catalytic
mechanism of this class of widely spread enzymes over
the phylogenetic tree,4,5 CA activators might also pos-
sess pharmacological applications, although this field
is largely unexplored for the moment. Thus, very
recently it has been reported6 that phenylalanine, a CA
activator first investigated by this group,1,3 when ad-
ministered to experimental animals produces an im-
* To whom correspondence should be addressed. Telephone: +39-
055-2757551. Fax: +39-055-2757555. E-mail: claudiu.supuran@unifi.it
† University of Agricultural Sciences and Veterinary Medicine.
‡ Polytechnic University.
§ Universita` degli Studi.
| “C.D. Nenitzescu” Institute of Organic Chemistry.
10.1021/jm011031n CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/19/2001

Carbonic Anhydrase Activators
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2 505
Sch em e 1a
Ta ble 1. CA Activatory Properties against Isozymes CA I, II,
and IV of Compounds 4-45 Prepared in the Present Study,
Compared with Those of Histamine as Standard
a
A, B, D ) N or C.
KA(µM)a
gen atoms of the imidazole moiety of histamine (the Nδ1
and Nꢀ2 are engaged in hydrogen bonds with the side
chains of Asn 62, His 64, Gln 92, and with Wat 152,
respectively), whereas the aliphatic amino group is not
experiencing any contact with the enzyme but is ex-
tending away from the cavity into the solvent.2 Posi-
tioned in such a favorable way, histamine facilitates the
rate-limiting step of CA catalysis but also allows its easy
derivatization (at the aliphatic amino group) in order
to obtain tighter-binding activators, as it has been
reported by us for compounds incorporating acyl or
arylsulfonyl moieties, of type 1a , which behaved as
much stronger CA activators against isozymes CA I, II,
and IV, as compared to the lead 1.8-10
Again considering histamine as lead, this group
reported12 the good CA activatory properties of a series
of aminoethyl-1,3,4-thiadiazoles of type 2, also incorpo-
rating substituted pyridinium moieties in their mol-
ecule. A further increase of the CA activatory properties
was then found for a series of 1,2,4-triazoles incorporat-
ing substituted pyridinium moieties, of type 3.13 It
became thus clear that effective CA activators should
incorporate two main structural elements: (i) a moiety
able to participate in proton-transfer reactions between
the active site and the bulk solvent, preferably with a
pKa in the range of 6-8;1,11,14 (ii) additional moieties that
ensure a tight binding to the enzyme, of the arysulfon-
amido,8 acylamido,9 or substituted pyridinium12,13 type.
Considering the excellent CA activatory properties
reported in a preliminary communication13 for the
pyridinium-triazoles 3, we decided to investigate in
more detail this type of derivative. Here we report the
synthesis and CA activation of three physiological
relevant isozymes (CA I, II, and IV) with three series
of novel 1-azolyl-pyridinium derivatives incorporating
pyrazole, imidazole, and tetrazole, respectively, as
proton shuttle moieties between the CA active site and
the environment. A comparative CA in vitro and ex vivo
activation study was performed, and some elements of
structure-activity relationship in this class of CA
activators are also discussed.
hCA hCA
bCA
IVd
compd
R1
R2
R3
R4
R5
Me
Me
Me
Ph
Me
Et
Ib
IIc
1 (histamine)
4
5
6
7
8
9
2.10
125
41
Me
Me
H
H
Me
Me
H
Me
0.052 0.480 0.045
0.049 0.457 0.038
0.037 0.412 0.033
0.875 83
0.035 0.359 0.017
0.027 0.306 0.015
Me; R2,R4 ) (CH2)9, R3 ) H
Me
Me
Et
n-Pr
i-Pr
n-Bu
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
1.68
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
n-Pr 1.35
i-Pr 0.021 0.155 0.008
n-Bu 1.76
75
1.20
140
96
118
135
130
2.59
1.85
3.41
8.23
5.29
Ph
Ph
Ph
Ph
Ph
Ph
i-Pr
Me
Me
Ph
Me
Et
0.64
2.34
2.78
2.50
5.64
32
Et
n-Pr
i-Pr
n-Bu
Ph
i-Pr
Me
>150 >50
>150 >50
0.038 0.264 0.016
0.036 0.347 0.021
0.022 0.305 0.020
0.360 59
0.021 0.153 0.013
0.018 0.144 0.009
Me; R2,R4 ) (CH2)9, R3 ) H
Me
Me
Et
n-Pr
i-Pr
n-Bu
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.975
n-Pr 0.247 48
i-Pr 0.012 0.095 0.006
n-Bu 0.651 86
0.620
1.89
1.24
2.56
5.76
3.14
9.55
Ph
Ph
Ph
Ph
Ph
Ph
i-Pr
Me
Et
0.303 62
Et
1.05
1.90
1.54
2.40
15
84
n-Pr
i-Pr
n-Bu
Ph
i-Pr
Me
130
107
130
>150 >50
0.024 0.130 0.008
0.076 1.39
0.064 1.23
0.063
0.055
1.55
0.012
6.80
Et
n-Pr
i-Pr
n-Bu
Me
n-Pr 2.12
i-Pr 0.050 1.15
n-Bu 1.80
136
145
108
140
Ph
Ph
Ph
Ph
Ph
Ph
1.10
3.05
4.21
2.87
5.24
12.52
Et
n-Pr
i-Pr
n-Bu
Ph
>150 15.50
134 10.13
10.50 >150 >50
41 >150 >50
a
Activation constant, mean from at least three determinations
by the esterase method. Standard error was in the range of 5-10%.
[hCA I] ) 25 nM. c [hCA II] ) 7 nM. [bCA IV] ) 33 nM.
b
d
Ex Vivo Activa tion . Data with some of the best in
Resu lts
vitro activators, against human red cell isozymes (hCA
I + hCA II), after incubation of red cells with the
activator solution for different periods of time, are
presented in Table 2.
Syn th esis. Preparation of the new compounds 4-45
is shown in Scheme 1 and involved the reaction of tri-/
tetrasubstituted pyrylium salts15 46 with the amino
azoles 47-49.15-18 All new compounds were character-
ized by standard analytical and spectroscopic methods
that confirmed the proposed structures.
Discu ssion
Ca r bon ic An h yd r a se Activa tion . In vitro activa-
tion data of isozymes hCA I, hCA II, and bCA IV with
the new derivatives 4-45, as well as standard activa-
tors, is presented in Table 1.
Ch em istr y. A series of 2,4,6-tri-/2,3,4,6-tetrasubsti-
tuted pyrylium salts were condensed with 3-amino-
pyrazole 47, 2-amino-imidazole 48, and 5-amino-tetra-
zole 49 in Baeyer-Piccard reaction conditions, affording

506 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2
Ilies et al.
Ta ble 2. Ex Vivo CA Activation Data after 30 and 60 min of
Incubation of Lysed Human Erythrocytes with Solutions
Containing 5 µM Activators: Histamine (as standard), 11, 23,
26, and 38
the case of some of the amines 47-49 investigated here,
and consequently, despite laborious synthetic and pu-
rification procedures (see Experimental Section for
details), yields in pyridinium salts 4-45 varied in a wide
range (10-98%), being linked with the above-mentioned
parameters.
% CA activitya
activator
30 min
60 min
histamine
121 ( 3
189 ( 5
192 ( 8
256 ( 7
173 ( 6
130 ( 5
234 ( 9
233 ( 11
298 ( 10
212 ( 8
Thus, especially for pyrylium salts possessing branched
alkyl chains (e.g., i-propyl, tert-butyl, nonamethylene)
in the 2- and/or 6-positions of the heterocyclic ring, the
desired product could difficultly be obtained in pure
form, due to the following problems: (i) preponderance
of side reactions (mainly the above-mentioned C-
cyclization, generating the corresponding substituted
anilino-azoles of type 54) and (ii) particularly high water
solubility of the obtained products 53 and 54. We were
confronted with these problems during the prep-
aration of 2,4,6-trimethyl- and 2,3,4,6-tetramethyl-
pyridinium-tetrazole as well as 2,3,4,6-tetramethyl-
pyridinium-imidazole (this situation has also been
encountered at the preparation of pyridinium-1,2,4-
triazoles13 and pyridinium-benzenesulfonamides22 re-
ported previously by our group). Consequently, although
TLC indicated the formation of these products, they
were not included in the list of derivatives 4-45 shown
in Table 1, due to their unsatisfactory purity.
11
23
26
38
a
Mean ( standard error (n ) 3); erythrocyte CA activity (hCA
I + hCA II) in the absence of activator is taken as 100%.
Sch em e 2
For insight into the structure-activity relationship
for this class of CA activators, we pursued some
structural variations in the choice of the reactants, as
follows: (i) the number of azole nitrogen atoms (2 or 4)
and their position (pyrazole/imidazole), also for compar-
ing these new activators with the previously mentioned
analogues incorporating pyridinium-triazoles moieties,
of type 3;13 (ii) the substitution pattern of the pyrylium
ring (alkyl/aryl groups as well as different combina-
tions of them), as this was one of the parameters that
strongly influenced the efficacy of the pyridinium de-
rivatives, both as CA activators11,13 as well as CA
inhibitors (for some pyridinium-sulfonamides reported
previously).22,23
three series of new pyridinium salts: 4-19, 20-34, 35-
45, respectively (Scheme 1). Although apparently very
simple, the reaction is in reality a multistep process
(Scheme 2), the key factors influencing it being the azole
amino group nucleophilicity, the pyrylium salt reactiv-
ity, the steric hindrance of the heterocyclic ring, and
possible side reactions, as shown previously by these
and other researchers.13,15,16,19-23 Thus, the nucleophilic
attack of a primary amine RNH2 on pyrylium cations
generally occurs in the R position, with the formation
of intermediates of type 50, which by deprotonation in
the presence of bases lead to the 2-amino-tetradehydro-
pyran derivatives 51. In many cases the deprotonation
reaction is promoted by the amine itself, when this is
basic enough (this being the reason in many cases one
works at the molar ratio pyrylium salt:amine of 1:2), or
by external catalysts added to the reaction mixture, such
as triethylamine.15,16,19-22 The derivatives 51 are gener-
ally unstable, being tautomers with the ketodiene-
amines 52, which are the key intermediates for the
conversion of pyryliums into pyridiniums.15,16,19-22 In
acidic media, in the rate-determining step of the whole
process, ketodieneamines 52 may be converted to the
corresponding pyridinium salts 53, although other
products such as vinylogous amides with diverse struc-
It must also be mentioned that the pyridinium tet-
razoles 35-45 reported here were isolated as zwitter-
ionic species (and not as perchlorates, as for the corre-
sponding imidazoles and pyrazoles). This may be due
to the very acidic character of this ring system (a typical
isosteric replacement of the COOH moiety) and has also
been observed for the pyridinium amino acid derivatives
reported by Balaban’s group.15,16
In Vit r o CA Act iva t ion . The pyridinium-azoles
4-45 were assayed for activatory properties against
isozymes hCA I, hCA II, and bCA IV, using histamine
1 as standard. Indeed, the new derivatives 4-45 gener-
ally showed good CA activatory properties. Activation
data given in Table 1 reveal significant differences in
isozyme specificity related to both “classical” activators
such as histamine and the newly synthesized com-
pounds. Thus, the most susceptible isozyme to activation
by histamine was hCA I, followed at a rather large
distance by bCA IV, whereas isozyme hCA II was shown
to be less activated by this compound (more precisely,
the difference in the three isozymes sensitivity ex-
pressed as activation constants is about 1 order of
magnitude).2 In the case of the pyridinio-azoles 4-45,
all three isozymes showed a 3-fold increased susceptibil-
ity to be activated by some of the new derivatives, as
tures have also been isolated in such reactions.15
A
supplementary complication appears when the moiety
substituting the 2- and/or 6-position(s) of the pyrylium
ring is methyl (or CH2R′ and CHR′R′′), in which case a
concurrent cyclization with formation of the anilines 54
in addition to the pyridinium salts 53 may take place
too.15,16,19-22 These concurrent reactions mentioned
above are generally important when the amine to be
converted into the pyridinium salt possesses weak
nucleophilicity or basicity. This probably happens to be

Carbonic Anhydrase Activators
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2 507
compared to histamine. The isozymes affinity for the
new activators was also different: although the cytosolic
rapid isozyme hCA II remained the most reluctant to
be activated (as for histamine), there were no significant
differences between isozymes hCA I and bCA IV, which
were both strongly activated by some of the compounds
4-45. Regarding the intensity of the activation effect,
the histamine pattern is maintained, the activation
constants for hCA II being an order of magnitude higher
than those for hCA I and bCA IV isozymes.
Regarding the previous study13 on CA activation, the
investigated isozyme affinity for the analogous series
of pyridinium-1,2,4-triazoles varied in a different way,
i.e., bCA IV . hCA I > hCA II. It should be noted that,
similarly to histamine, CA activation by the pyridinium-
triazoles reported earlier13 only led to micromolar
activators, whereas many of the derivatives obtained
in the present work showed nanomolar affinity to
isozymes I and IV.
chain increase to n-propyl and n-butyl reduced the
activatory properties, probably due to steric hindrance
effects, which destabilize the activator interaction with
the CA active site. The optimal fit has been achieved
by 2,6-di-i-propyl-4-phenyl groups, with compounds 11,
26, and 38, respectively, being the best activators in the
whole series and against all three CA isozymes.
The nature of the azole ring is probably the second
other important variable determining the activation
effect in these series of compounds. Presuming an
identical substitution pattern, the following order re-
lated to the activatory properties has been observed:
imidazoles > pyrazoles . tetrazoles. The explanation
of this activity pattern may reside in the pKa values of
the azolyl moieties in aqueous system: the imidazole
ring system, with a pKa of 7.1 for proton acceptance (and
17.5 for H+ donation), is the best option for shuttling
protons between the active site and the external me-
dium.11,14 Pyrazole is much less basic than imidazole
(pKa ) 2.5 for proton acceptance), and this may explain
the decreased CA activatory properties of its derivatives,
as compared to the corresponding imidazoles.11,14 Tet-
razole, on the other hand, is a quite acidic compound
(pKa ) 4.9 for proton donation) and hence less appropri-
ate to act as a CA activator (in neutral state). It should
also be stressed that the pKa of the imidazole/pyrazole/
tetrazole derivatives reported by us may be rather
different from the pKa value of the parent ring systems,
but this parameter should vary in the same manner for
the three classes of derivatives investigated here and
may be thus enlightening for better understanding SAR
of such CA activators.
The influence of the pyridinium ring substitution
pattern on the biological activity of compounds 4-45
confirmed and extended the conclusions of our prelimi-
nary study:13 (i) derivatives possessing 2,6-dialkyl-4-
phenyl moieties, such as 8-11, 23-26, and 35-38,
respectively, were the most active against all the
investigated isozymes (with the 2,6-alkyl groups: i-Pr
> Et > Me), confirming that elongation of the activator
molecule in the direction of the axis passing through
the Zn(II) ion of the enzyme, the activator molecule
itself, and the exit of the active site is the most beneficial
substitution pattern leading to strong CA activators; (ii)
the 2,4,6-trimethyl- and 2,3,4,6-tetramethyl-pyridinium-
azoles (4, 20, and 5) were slightly less active than the
previously mentioned derivatives; (iii) the increased
volume of the pyridinium ring moieties generated an
opposite effect on the CA activation properties, with 2,4-
diphenyl-6-alkyl-substituted derivatives (13-17, 28-
32, 40-44, respectively) acting as moderate-weak ac-
tivators [mention should be made that, as part of this
pattern substitution, the most active derivative was the
first term of the series, i.e., the compound bearing 2,4-
diphenyl-6-methyl-pyridinium groups (13, 28, 40, re-
spectively), with the higher homologues much less active
(i.e., the 6-ethyl-, 6-propyl-, or 6-butyl derivatives)]; (iv)
the 2,4,6-triphenyl-pyridinium azoles were practically
devoid of activity, confirming the limited amount of
space available in the activators’ binding site;1,13 and
(v) the 2,6-dimethyl-3,5-nonamethylene-pyridinium azoles
6 and 21 (a substitution pattern newly introduced in
this study) proved good in vitro CA activation proper-
ties. These lipophilic, flexible lariate side chain sub-
stituted derivatives showed indeed remarkable CA
activatory properties, being only slightly less effective
than the 2,6-dialkyl-4-phenyl-substituted azoles men-
tioned above.
We also observed that the pyridinium-tetrazole de-
rivatives reported here possess a zwitterionic nature
(see structures 35-45). Thus, it is rather probable that
the pyridinium-imidazole/pyrazole derivatives reported
in this paper bind to the enzyme as cations (and by
means of their azole moieties shuttle protons between
the Zn-bound water and the environment), whereas the
pyridinium-tetrazoles prepared by us may bind as
zwitterions, activating CAs by the same mechanism of
action as the first type of compounds mentioned above,
i.e., favoring the rate-determining proton-transfer reac-
tions between the active site and the reaction medium.
It may be mentioned again that the pKa of a com-
pound bound within an enzyme active site may be
quite different from the pKa of the same compound in
solution.
Ex Vivo CA Activa tion . After incubation of normal
blood red cells (containing approximately 150 µM of hCA
I and 20 µM of hCA II)24-26 with micromolar concentra-
tions of histamine, the total CA activity (measured by
the esterase method, with 4-nitrophenyl acetate as
substrate)25 in homogenates of treated cells is enhanced
as compared to that of cells treated in a blank experi-
ment only with buffer (Table 2). Performing the same
type of experiment with some of the new activators
synthesized in the present work (such as the pyrazole
11, or the imidazoles 23 and 26) does not lead to any
CA activatory effects (data not shown). This is obviously
due to the cationic nature of these derivatives, and in
consequence, to their membrane impermeability, as
already proved by us for positively charged sulfonamide
CA inhibitors of type 55-58.22,23 Indeed, such positively
Also in agreement with the structure-activity rela-
tionships data pointed out in our preliminary study,13
we wish to stress here again that the alkyl chain length
of the 2,6-substituents of the pyridinium ring is the most
important factor determining the intensity of CA activa-
tion. Thus, considering the series of the 2,6-dialkyl-4-
phenyl-substituted derivatives reported here, it was
observed that passing from methyl to ethyl led to a
slight increase in activity. On the contrary, the further

508 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2
Ilies et al.
Exp er im en ta l Section
Gen er a l. Melting points were determined on a Boetius
heating plate microscope and are uncorrected. The IR spectra
were recorded in KBr pellets, on a Nicolet Avatar 360 FTIR
spectrometer, in the range 400-4000 cm-1. The NMR spectra
were recorded in DMSO-d6, at ≈295 K with a Varian Gemini
300BB spectrometer operating at 300 MHz for 1H and at 75
MHz for 13C. Chemical shifts are reported as δ values, using
TMS as internal standard for proton spectra and the solvent
resonance (39.50 ppm in DMSO-d6) for carbon spectra. At-
tributions were done by means of chemical shifts, peak
integration, attached proton test (APT), COSY (1H-13C), hy-
drogen multiple bond coherence (HMBC) experiments, dy-
namic NMR spectra, selective deuteration, and model spectra.
Elemental analyses were done by combustion, for C, H, N, with
an automated Carlo Erba analyzer, and the results were found
to be (0.4% within the theoretical values.
All reactions were monitored by thin-layer chromatography
(TLC), using 0.25 mm thick precoated silica gel plates (E.
Merck) eluted with MeOH:CHCl3 1:4 v/v. 3-Aminopyrazole,
2-amino-imidazole sulfate, and 5-amino-tetrazole hydrate were
from Sigma-Aldrich (Milan, Italy). Pyrylium salts were pre-
pared by literature procedures, optimized in our laboratories.15
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
4-34. Amino-azole (5 mmol) and the appropriate pyrylium salt
(6 mmol) were suspended in 10 mL of anhydrous methanol.
In the case of 2-amino-imidazole, the free amine was obtained
as methanolic solution by refluxing the corresponding sulfate
(Sigma) with the stoechiometric amount of MeONa in MeOH
for 1 h, followed by filtration of the precipitated Na2SO4.
Triethylamine (6 mmol) was added, and the obtained
solution was refluxed under stirring for 10 min, then AcOH
(12 mmol) was added, and the mixture was refluxed for
another 3-4 h. After that time, reflux was discontinued, 0.5
mL of concentrated NH3 solution was added, and the solution
was heated again for 5 min in order to convert the unreacted
pyrylium salt into the corresponding pyridine. After cooling,
the final reaction mixture was poured in 80 mL of diethyl
ether, when crystallization usually occurred. The crude product
was filtered, washed with ether, and recrystallized from
alcohols (MeOH or EtOH/iPrOH mixtures). Yields were in the
range of 15-98%.
charged derivatives cannot cross the plasma membrane
and are unable to arrive within the cytosol for activating
CA I and II isozymes abundant in this cellular localiza-
tion. To observe the ex vivo CA activatory properties of
these pyridinium-azole derivatives, the experimental
protocols described above had to be changed. Thus, red
blood cells were first lysed in order to liberate the
present CA isozymes, and then the lysate was incubated
either with histamine or with the newly synthesized
compounds 11, 23, or 26 (Table 2). After 30 or 60 min
incubation, a clear-cut activatory effect was observed
with the two types of compounds (histamine as stand-
ard, and the new positively charged activators reported
here). Thus, histamine produced only a weak activation
of around 120% after half an hour incubation, and of
around 130% of the basal CA activity after 1 h incuba-
tion with red cells, either in the first as well as in the
second type of experiment. Some of the new derivatives
tested ex vivo (which showed strong in vitro CA activity
enhancements) produced activations of around 190-
255% after 30 min and 233-300% after 1 h incubation,
respectively. The stronger in vitro CA activator 26 was
also the most effective ex vivo derivative, as compared
to the weaker activators 11 and 23 investigated in some
detail. In contrast to the pyrazole and imidazole cationic
derivatives mentioned above, the zwitterionic tetrazoles
(such as 38) do penetrate through the plasma mem-
branes, since the ex vivo activatory effect of this
compound is similar when incubating the blood red cells
with the activator followed by lysis or when lysing first
blood red cells and incubating them with the activator
(Table 2). Thus, similarly to histamine and in contrast
to the pyridinium-pyrazole/imidazole derivatives, the
zwitterionic pyridinium tetrazoles do penetrate mem-
branes activating thus cytosolic CA isozymes.
Gen er a l P r oced u r e for th e P r ep a r a tion of Com p ou n d s
35-45. 5-Aminotetrazole hydrate (5 mmol) and the appropri-
ate pyrylium salt (6 mmol) were suspended in 3 mL of
anhydrous methanol and poured into a stirred mixture of 25
mmol triethylamine and 15 mmol acetic anhydride previously
cooled at 0 °C (ice bath). After 5 min of stirring, 10 mL of
methanol were added, and the mixture was heated to reflux
for 10 min. Then 15 mmol of acetic acid was added, and the
mixture was refluxed for 4 h. The workup was the same as in
case of compounds 4-34. In several cases, the purified
compounds still contained variable amounts of ammonium
salts which co-crystallized with the desired material. For
eliminating these salts, the crude products were suspended
in 5 mL of distilled water and treated dropwise with 10%
aqueous NaOH under stirring, until pH 13 was reached,
filtered, washed with water and 5% aqueous HClO4 solution,
and recrystallized again. Yields were in the range of 10-84%.
The new derivatives 4-45 were characterized by elemental
Con clu sion s
Three series of 1-azolyl-tri-/tetrasubstituted pyri-
dinium salts were synthesized by reaction of the corre-
sponding pyrylium salts with 3-amino-pyrazole, 2-amino-
imidazole, and 5-amino-tetrazole, respectively. Many of
these new compounds proved to be powerful in vitro CA
activators against isozymes I, II, and IV, with activities
in the low nanomolar range for isozymes I and IV, and
low micromolar range for isozyme II, showing thus a
3-fold increased activity as compared to histamine, the
lead molecule used for their preparation. The different
isozymes showed diverse affinities for these activators,
with CA I ≈ CA IV > CA II. Furthermore, the best in
vitro activators strongly enhanced ex vivo red cell lysate
CA activity. SAR in these series was also discussed.
These results confirmed and completed a previous CA
activation study of our group on a series of 1-[1,2,4-
triazole-(1H)-3-yl]-2,4,6-trisubstituted-pyridinium salts.
Furthermore, this is the first report of membrane-
impermeant CA activators.
1
analysis and detailed spectroscopic data (IR and H/13C NMR
spectra). In some cases, advanced NMR techniques were used
for complete structural attributions, i.e., selective decoupling,
bidimensional correlations spectra COSY (1H-13C), and dy-
namic NMR spectroscopy. These experiments also permitted
the evaluation of the barrier to rotation around the sp2-sp2
Npyridinium-Cazole bond, data that will be reported elsewhere.
1-(P yr a zol-3-yl)-2,6-d im eth yl-4-ph en yl-pyr idin iu m p er -
ch lor a te, 8: cream crystals, mp 246-8 °C (EtOH-MeOH 1:1
v/v). FTIR (KBr), cm-1: 624 (ClO4-), 776, 974, 1082 (ClO4-),
1221, 1299, 1334, 1380, 1455, 1473, 1529, 1558, 1595, 1634,
2881, 2947, 2967, 3021, 3076, 3145, 3256, 3373; 1H NMR
(DMSO-d6), δ, ppm, J , Hz: 13.83 (bs, 1H, NH, deuterable),
8.47 (s, 2H, Hâ of Py+), 8.20 (d, 3J HH ) 2.4, 1H, H-3 of pyrazole),

Carbonic Anhydrase Activators
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2 509
3
4
8.11 (dd, 2H, Hortho of Phe, J HH ) 7.3, J HH ) 2.3), 7.68 (m,
Ex Vivo CA Activa tion . An amount of 2 mL of freshly
isolated human blood was thoroughly washed several times
with 5 mL of Tris buffer (pH 7.40, 5 mM) and centrifuged for
10 min. The obtained erythrocytes were then treated with 2
mL of a 5 µM solution of CA activator. Incubation has been
done at 37 °C with gentle stirring, for periods of 30-60 min.
After that time, the red cells were centrifuged again for 10
min, the supernatant discarded, and the cells washed three
times with 5 mL of the above-mentioned buffer, to eliminate
all unbound compound. The cells were then lysed in 5 mL of
distilled water and centrifuged for eliminating membranes and
other insoluble materials, and the CA activity has been
assayed as described above. Blank experiments were done in
which no activator has been added to the blood red cells treated
as described above, and CA activity determined in such
conditions has been taken as 100%.23,26 The same type of
experiment was then repeated, lysing the red blood cells first,
and incubating the obtained supernatant with the activators
as described above.
3
3H, Hmeta of Phe + Hpara of Phe), 6.71 (d, J HH ) 2.4, 1H, H-4
of pyrazole), 2.48 (s, 6H, CH3); 13C NMR (DMSO-d6), δ, ppm:
156.99 (Cq, CR of Py+), 155.88 (Cq, Cγ of Py+), 145.26 (Cq, C-3
of pyrazole), 133.55 (Cq, Cipso of Phe), 132.66 (CH, C-5 of
pyrazole), 132.66 (CH, Cpara of Phe), 129.77 (CH, Cmeta of Phe),
128.30 (CH, Cortho of Phe), 123.22 (CH, Câ of Py+), 101.96 (CH,
C-4 of pyrazole), 21.08 (CH3). Anal. (C16H16N3+ClO4-) C, H,
N.
1-(Im idazol-2-yl)-2,6-dim eth yl-4-ph en yl-pyr idin iu m per -
ch lor a te, 23: cream crystals, mp 246-8 °C (EtOH-MeOH 1:1
v/v). FTIR (KBr), cm-1
:
625 (ClO4-), 772, 803, 884, 1094
(ClO4-), 1216, 1294, 1333, 1376, 1449, 1473, 1539, 1559, 1590,
1
1633, 2845, 2888, 2920, 3049, 3109, 3180; H NMR (DMSO-
d6), δ, ppm, J , Hz: 8.53 (s, 2H, Hâ of Py+), 8.16 (dd, 2H, Hortho
3
4
of Phe, J HH ) 7.1, J HH ) 2.1), 7.71 (m, 3H, Hmeta of Phe +
Hpara of Phe), 7.43 (s, 2H, CH of imidazole), 2.47 (s, 6H, CH3);
13C NMR (DMSO-d6), δ, ppm: 157.49 (Cq, CR of Py+), 157.10
(Cq, Cγ of Py+), 135.70 (Cq, C-2 of imidazole), 133.33 (Cq, Cipso
of Phe), 132.92 (CH, Cpara of Phe), 129.88 (CH, Cmeta of Phe),
Ack n ow led gm en t. This research was financed by
the EU grant ERB CIPDCT 940051, by an Italian CNR
grantsTarget Project Biotechnology, and by the Ro-
manian grant CNFIS 141.
128.54 (CH, Cortho of Phe), 124.37 (CH, C-4,5 of imidazole),
123.25 (CH, Câ of Py+), 20.19 (CH3). Anal. (C16H16N3+ClO4
)
-
C, H, N.
1-(Tet r a zol-5-yla t e)-2,6-d im et h yl-4-p h en yl-p yr id in i-
u m , 35: cream crystals, mp 279-81 °C (desc.) (MeOH). FTIR
(KBr), cm-1: 606, 694, 770, 878, 997, 1009, 1022, 1063, 1081,
1128, 1146, 1190, 1336, 1380, 1404, 1457, 1559, 1635, 2846,
Refer en ces
1
(1) Supuran, C. T.; Scozzafava, A. Activation of carbonic anhydrase
isozymes. In The Carbonic Anhydrases - New Horizons; Cheg-
widden, W. R., Carter, N., Edwards, Y., Eds.; Birkhauser
Verlag: Basel, Switzerland, 2000; pp 197-219.
(2) Briganti, F.; Mangani, S.; Orioli, P.; Scozzafava, A.; Vernaglione,
G.; Supuran, C. T. Carbonic anhydrase activators: X-ray
crystallographic and spectroscopic investigations for the interac-
tion of isozymes I and II with histamine. Biochemistry 1997, 36,
10384-10392.
(3) Briganti, F.; Iaconi, V.; Mangani, S.; Orioli, P.; Scozzafava, A.;
Vernaglione, G.; Supuran, C. T. A ternary complex of carbonic
anhydrase: X-ray crystallographic structure of the adduct of
human carbonic anhydrase II with the activator phenylalanine
and the inhibitor azide. Inorg. Chim. Acta 1998, 275-276, 295-
300.
(4) (a) Supuran, C. T.; Scozzafava, A. Carbonic anhydrase inhibitors.
Curr. Med. Chem. - Imm., Endoc., Metab. Agents 2001, 1, 61-
97. (b) Supuran, C. T.; Scozzafava, A. Carbonic anhydrase
inhibitors and their therapeutic potential. Expert Opin. Ther.
Pat. 2000, 10, 575-600.
(5) The Carbonic Anhydrases - New Horizons; Chegwidden, W. R.,
Carter, N., Edwards, Y., Eds.; Birkhauser Verlag: Basel,
Switzerland, 2000.
(6) Sun, M. K.; Alkon, D. L. Pharmacological enhancement of
synaptic efficacy, spatial learning and memory through carbonic
anhydrase activation in rats. J . Pharmacol. Exp. Ther. 2001,
297, 961-967.
(7) Meier-Ruge, W.; Iwangoff, P.; Reichlmeier, K. Neurochemical
enzyme changes in Alzheimer’s and Pick’s disease. Arch. Ger-
ontol. Geriatr. 1984, 3, 161-165.
(8) (a) Briganti, F.; Scozzafava, A.; Supuran, C. T. Novel carbonic
anhydrase isozymes I, II and IV activators incorporating sul-
fonyl-histamino moieties. Bioorg. Med. Chem. Lett. 1999, 9,
2043-2048. (b) Supuran, C. T.; Scozzafava, A. Carbonic anhy-
drase activators. Amino acyl/dipeptidyl histamine derivatives
bind with high affinity to isozymes I, II and IV and act as
efficient activators. Bioorg. Med. Chem. 1999, 7, 2915-2924.
(9) (a) Scozzafava, A.; Supuran, C. T. Carbonic anhydrase activators.
Part 24. High affinity isozymes I, II and IV activators, deriva-
tives of 4-(4-chlorophenylsulfonylureido-amino acyl)ethyl-1H-
imidazole. Eur. J . Pharm. Sci. 2000, 10, 29-41. (b) Scozzafava,
A.; Iorga, B.; Supuran, C. T. Carbonic anhydrase activators.
Synthesis of high affinity isozymes I, II and IV activators,
derivatives of 4-(4-tosylureido-amino acyl)ethyl-1H-imidazole
(histamine derivatives). J . Enzyme Inhib. 2000, 15, 139-161.
(10) (a) Scozzafava, A.; Supuran, C. T. Carbonic anhydrase activators.
Part 21. Novel activators of isozymes I, II and IV incorporating
carboxamido- and ureido histamine moieties. Eur. J . Med. Chem.
2000, 35, 31-39. (b) Supuran, C. T.; Scozzafava, A. Carbonic
anhydrase activators. Synthesis of high affinity isozymes I, II
and IV activators, derivatives of 4-(arylsulfonylureido-amino
acyl)ethyl-1H-imidazole. J . Enzyme Inhib. 2000, 15, 471-486.
(c) Scozzafava, A.; Supuran, C. T. Carbonic anhydrase activa-
tors: High affinity isozymes I, II and IV activators, incorporating
a beta-alanyl-histidine scaffold. J . Med. Chem. 2001, 45, 284-
291.
2928, 3083, 3227, 3351; H NMR (DMSO-d6), δ, ppm, J , Hz:
8.49 (s, 2H, Hâ of Py+), 8.12 (dd, 2H, Hortho of Phe, J HH ) 7.2,
3
4J HH ) 2.2), 7.68 (m, 3H, Hmeta of Phe + Hpara of Phe), 2.36 (s,
6H, CH3); 13C NMR (DMSO-d6), δ, ppm: 157.26 (Cq, C-5 of
tetrazole), 156.37 (Cq, CR of Py+), 155.85 (Cq, Cγ of Py+), 133.56
(Cq, Cipso of Phe), 132.41 (CH, Cpara of Phe), 129.66 (CH, Cmeta
of Phe), 128.31 (CH, Cortho of Phe), 123.27 (CH, Câ of Py+), 20.51
(CH3). Anal. (C14H14N4) C, H, N.
En zym e P r ep a r a tion s. Human CA I and CA II cDNAs
were expressed in Escherichia coli strain BL21 (DE3) from the
plasmids pACA/hCA I and pACA/hCA II described by Lindskog
et al.27 (the two plasmids were a gift from Prof. Sven Lindskog,
Umea University, Sweden). Cell growth conditions were those
described by this group,28 and enzymes were purified by
affinity chromatography according to the method of Khalifah
et al.29 Enzyme concentrations were determined spectrophoto-
metrically at 280 nm, utilizing a molar absorptivity of 49 mM-1
cm-1 for CA I and 54 mM-1 cm-1 for CA II, respectively, based
on Mr ) 28.85 kDa for CA I and 29.30 kDa for CA II,
respectively.30 CA IV was isolated from bovine lung mi-
crosomes as described by Maren et al., and its concentration
has been determined by titration with ethoxzolamide.31
Initial rates of 4-nitrophenyl acetate hydrolysis catalyzed
by different CA isozymes were monitored spectrophotometri-
cally, at 400 nm, with a Cary 3 instrument interfaced with an
IBM-compatible PC.25 Solutions of substrate were prepared
in anhydrous acetonitrile; the substrate concentrations varied
between 2 × 10-2 and 1 × 10-6 M, working at 25 °C. A molar
absorption coefficient ꢀ of 18 400 M-1 cm-1 was used for the
4-nitrophenolate formed by hydrolysis, in the conditions of the
experiments (pH 7.40), as reported in the literature.25 Non-
enzymatic hydrolysis rates were always subtracted from the
observed rates. Duplicate experiments were done for each
activator concentration, and the values reported throughout
the paper are the mean of such results. Stock solutions of
activator (1 mM) were prepared in distilled-deionized water
with 10-15% (v/v) DMSO (which is not inhibitory/activatory
at these concentrations) and dilutions up to 0.1 nM were done
thereafter with distilled-deionized water. Activator and en-
zyme solutions were preincubated together for 15 min at room
temperature prior to assay, in order to allow for the formation
of the E-A complex. The activation constant KA was deter-
mined as described by Briganti et al.2 Enzyme concentrations
were 7 nM for hCA II, 25 nM for hCA I, and 33 nM for bCA
IV (this isozyme has a decreased esterase activity,32 and higher
concentrations had to be used for the measurements).

510 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 2
Ilies et al.
(11) (a) Supuran, C. T.; Balaban, A. T.; Cabildo, P.; Claramunt, R.
M.; Lavandera, J . L.; Elguero, J . Carbonic anhydrase activators.
VII. Isozyme II activation by bis-azolylmethanes, -ethanes and
related azoles. Biol. Pharm. Bull. 1993, 16, 1236-1239. (b)
Supuran, C. T.; Claramunt, R. M.; Lavandera, J . L.; Elguero, J .
Carbonic anhydrase activators. XV. A kinetic study of the
interaction of bovine isozyme with pyrazoles, bis- and tris-azolyl
methanes. Biol. Pharm. Bull. 1996, 19, 1417-1422. (c) Clare,
B. W.; Supuran, C. T. Carbonic anhydrase activators. Part 3.
Structure-activity correlations for a series of isozyme II activa-
tors. J . Pharm. Sci. 1994, 83, 768-779.
(12) Supuran, C. T.; Barboiu, M.; Luca, C.; Pop, E.; Brewster, M. E.;
Dinculescu, A. Carbonic anhydrase activators. Part 14. Synthesis
of mono and bis pyridinium salt derivatives of 2-amino-5-(2-
aminoethyl)- and 2-amino-5-(3-aminopropyl)-1,3,4-thiadiazole
and their interaction with isozyme II. Eur. J . Med. Chem. 1996,
31, 597-606.
(22) Supuran, C. T.; Scozzafava, A.; Ilies, M. A.; Iorga, B.; Cristea,
T.; Briganti, F.; Chiraleu, F.; Banciu, M. D. Carbonic anhydrase
inhibitors. Part 53. Synthesis of substituted-pyridinium deriva-
tives of aromatic sulfonamides: The first non polymeric mem-
brane-impermeable inhibitors with selectivity for isozyme IV.
Eur. J . Med. Chem. 1998, 33, 577-594.
(23) Scozzafava, A.; Briganti, F.; Ilies, M. A.; Supuran, C. T. Carbonic
anhydrase inhibitors. Synthesis of membrane-impermeant low
molecular weight sulfonamides possessing in vivo selectivity for
the membrane-bound versus the cytosolic isozymes J . Med.
Chem. 2000, 43, 292-300.
(24) Wistrand, P. J .; Lindqvist, A. Design of carbonic anhydrase
inhibitors and the relationship between the pharmacodynamics
and pharmacokinetics of acetazolamide. In Carbonic Anhydrase
- From Biochemistry and Genetics to Physiology and Clinical
Medicine; Botre`, F., Gros, G., Storey, B. T., Eds.; VCH: New
York-Weinheim, 1991; pp 352-378.
(13) Ilies, M. A.; Banciu, M. D.; Ilies, M.; Chiraleu, F.; Briganti, F.;
Scozzafava, A.; Supuran, C. T. Carbonic anhydrase activators.
Part 17. Synthesis and activation studies of a series of 1-(1,2,4-
triazole-1H-3-yl)-2,4,6-trisubstituted-pyridinium salts against
isozymes I, II and IV. Eur. J . Med. Chem. 1997, 32, 911-918.
(14) Supuran, C. T.; Balaban, A. T. Carbonic anhydrase activators.
Part 8. pKa - activation relationship for amino acid derivatives,
activators of isozyme II. Rev. Roum. Chim. 1994, 39, 107-113.
(15) (a) Balaban, A. T.; Dinculescu, A.; Dorofeenko, G. N.; Fischer,
G. W.; Koblik, A. V.; Mezheritskii, V. V.; Schroth, W. Pyrylium
Salts: Syntheses, Reactions and Physical Properties. In Ad-
vances in Heterocyclic Chemistry; Katritzky, A. R., Ed.; Academic
Press: New York, 1982; pp 8-360. (b) Schroth, W.; Balaban, A.
T. Methoden der Organischen Chemie (Houben-Weil), Vol. E7b,
Hetarene (Teil 2); Kreher, R. P., Ed.; G. Thieme Verlag:
Stuttgart, 1992; pp 755-963.
(25) Pocker, Y.; Stone, J . T. The catalytic versatility of erythrocyte
carbonic anhydrase. III. Kinetic studies of the enzyme-catalyzed
hydrolysis of p-nitrophenyl acetate. Biochemistry 1967, 6, 668-
678.
(26) (a) Sly, W. S. Carbonic anhydrase II deficiency syndrome:
Clinical delineation, interpretation and implications. In The
Carbonic Anhydrases; Dodgson, S. J .; Tashian, R. E.; Gros, G.;
Carter, N. D., Eds.; Plenum Press: New York and London, 1991;
pp. 183-196. (b) Sly, W. S.; Hu, P. Y. Human carbonic anhy-
drases and carbonic anhydrase deficiencies. Annu. Rev. Biochem.
1995, 64, 375-401.
(27) Behravan, G.; J onsson, B. H.; Lindskog, S. Fine-tuning of the
catalytic properties of carbonic anhydrase. Studies of a Thr200-
His variant of human isoenzyme II. Eur. J . Biochem. 1990, 190,
351-357.
(28) Lindskog, S.; Behravan, G.; Engstrand, C.; Forsman, C.; J onsson,
B. H.; Liang, Z.; Ren, X.; Xue, Y. Structure-function relations
in human carbonic anhydrase II as studied by site-directed
mutagenesis. In Carbonic anhydrase - From biochemistry and
genetics to physiology and clinical medicine; Botre`, F., Gros, G.,
Storey, B. T., Eds.; VCH: Weinheim, 1991; pp 1-13.
(29) Khalifah, R. G.; Strader, D. J .; Bryant, S. H.; Gibson, S. M.
Carbon-13 nuclear magnetic resonance probe of active site
ionization of human carbonic anhydrase B. Biochemistry 1977,
16, 2241-2247.
(30) Steiner, H.; J onsson, B. H.; Lindskog, S. The catalytic mecha-
nism of carbonic anhydrase. Hydrogen-isotope effects on the
kinetic parameters of the human C isoenzyme. Eur. J . Biochem.
1975, 59, 253-259.
(31) Maren, T. H.; Wynns, G. C.; Wistrand, P. J . Chemical properties
of carbonic anhydrase IV, the membrane-bound enzyme. Mol.
Pharmacol. 1993, 44, 901-906.
(32) Baird, T. T.; Waheed, A.; Okuyama, T.; Sly, W. S.; Fierke, C. A.
Catalysis and inhibition of human carbonic anhydrase IV.
Biochemistry 1997, 36, 2669-2678.
(16) Dinculescu, A.; Balaban, A. T. Reaction of pyrylium salts with
nucleophiles. XIV. New pyridinium salts with potential biological
activity. Rev. Roum. Chim. 1980, 25, 1505-1528.
(17) Supuran, C. T.; Pop, E.; Dinculescu, A. Reaction of pyrylium salts
with amino acid derivatives. Part 3. Ribonuclease A inactivation
by trisubstituted pyrylium salts. Heterocycles 1994, 37, 667-
671.
(18) Supuran, C. T.; Manole, G.; Dinculescu, A.; Schiketanz, A.;
Gheorghiu, M. D.; Puscas, I.; Balaban, A. T. Carbonic anhydrase
inhibitors. Part 5. Pyrylium salts in the synthesis of isozyme-
specific inhibitors. J . Pharm. Sci. 1992, 81, 716-719.
(19) Balaban, A. T.; Toma, C. Reaction of pyrylium salts with
nucleophiles. IV. Isolation of an intermediate in the conversion
of 2,4,6-triphenylpyrylium perchlorate into 2,4,6-triphenylpyri-
dine by amonia. Tetrahedron 1966, Suppl 7, 1-8.
(20) Katritzky, A. R.; Manzo, R. H. Kinetics and mechanism of the
reactions of primary amines with pyrylium cations. J . Chem.
Soc., Perkin Trans. 2 1981, 2, 571-575.
(21) Katritzky, A. R.; Lloyd, J . M.; Patel, R. C. The preparation of
pyridinium from pyryliums. J . Chem. Soc., Perkin Trans. 1 1982,
117-123.
J M011031N