Ethylene bis-imidazoles are highly potent and selective activators for isozymes VA and VII of carbonic anhydrase, with a potential nootropic effect

Article abstract of DOI:10.1039/c4cc02346c

A series of ethylene bis-imidazoles was synthesized via a novel microwave-mediated synthesis. Biological testing on eight isozymes of carbonic anhydrase (CA) present in the human brain revealed compounds with nanomolar potency against CA VA and CA VII, also displaying excellent selectivity against other CA isozymes present in this organ. the Partner Organisations 2014.

Full text of DOI:10.1039/c4cc02346c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Ethylene bis-imidazoles are highly potent and
selective activators for isozymes VA and VII of
carbonic anhydrase, with a potential nootropic
effect†
Cite this: Chem. Commun., 2014,
50, 5980
Received 29th March 2014,
Accepted 14th April 2014
Bogdan Draghici,a Daniela Vullo,b Suleyman Akocak,a Ellen A. Walker,a
DOI: 10.1039/c4cc02346c
Claudiu T. Supuran*b and Marc A. Ilies*a
www.rsc.org/chemcomm
A series of ethylene bis-imidazoles was synthesized via a novel
microwave-mediated synthesis. Biological testing on eight iso-
zymes of carbonic anhydrase (CA) present in the human brain
revealed compounds with nanomolar potency against CA VA and
CA VII, also displaying excellent selectivity against other CA iso-
zymes present in this organ.
The dynamic state of all organisms relates directly to their metabolic
activity. Carbon dioxide is the end product of catabolic processes of
higher vertebrates and its rapid transport and elimination are
critical for achieving fast metabolic rates. Carbonic anhydrases
(E.C. 4.2.1.1) are zinc enzymes that catalyze the reversible hydration
of CO2 under physiological conditions: CO2 + H2O 2 HCO3ꢀ + H+.
Fig. 1 Schematics of CO2 flux, from mitochondria to the cytoplasm and
to the exterior of the cell, revealing that the CA isozymes involved in the
ꢀ
process and the three different CO2/HCO3 pools (extracellular (e),
cytosolic (c) and mitochondrial (m)) equilibrated by them.
Through the activity of mitochondrial isozymes, especially CA VA, such as the NMDA receptor.4 CAs are directly involved in
cytosolic isozymes CA I, CA II, CA VII and membrane-bound cerebrospinal fluid (CSF) secretion at the level of the choroid
isozymes CA IV, CA IX, CA XII, CA XIV, carbon dioxide can be plexus, generating bicarbonate from CO2 and H2O, and regulat-
rapidly hydrated and transported from mitochondria through the ing through ion hydration the volume, buffering properties,
cytoplasm towards the exterior of the cell (Fig. 1).1 About 85% of and turnover of CSF.5
ꢀ
total CO2 produced in the body is transported as HCO3 to lungs.
The key role played by CAs in brain homeostasis and normal
The above-mentioned CA isozymes are particularly important for cerebral function is emphasized by CA deficiency syndrome – a
the brain2 which is a major CO2 producer and where the CO2/ distinct genetic disorder characterized by severe pathological modi-
HCO3 buffer constitutes the main buffer in the interstitial/extra- fications of the brain, lungs, kidneys, and bone metabolism.6
ꢀ
cellular space.3 These CA isozymes actively neutralize the steep pH Patients with this disorder lack CA II in these tissues/organs and
gradients (as high as 0.1 pH units) generated by neuronal discharge blood and display a range of symptoms including cerebral calcifica-
that are superimposed on the normal flux of protons caused by tion, respiratory and renal tubular acidosis, inherited osteopetrosis,
aerobic metabolism.3 Strict pH control maintains membrane polari- and mental retardation.6 Moreover, in the neurodegenerative dis-
zation and the proper ionization (hence the action) of various order of memory and cognitive function, Alzheimer’s disease, the
key neurotransmitters, such as glutamate, GABA, and dopamine. level of CA is significantly diminished.7 A similar decrease in the
Fluctuation of local pH has been shown to affect receptor responses expression level of CA was observed in the brain of older rats
relative to young animals,8 being associated with a reduced resis-
tance to dehydration, alteration in choroidal control of brain
homeostasis, and reduced cerebrospinal fluid production.
a Department of Pharmaceutical Sciences and Moulder Center for Drug Discovery
Research, Temple University School of Pharmacy, 3307 N Broad Street,
Philadelphia, PA-19140, USA. E-mail: mailies@temple.edu; Fax: +1 215-707-5620;
Tel: +1 215-707-1749
b NEUROFARBA Department, Pharmaceutical Sciences Section, Universita degli
Studi di Firenze, Polo Scientifico, Via Ugo Schiff no. 6, 50019 Sesto Fiorentino,
Florence, Italy. E-mail: claudiu.supuran@unifi.it; Fax: +39-055-4573385;
Tel: +39-055-4573005
Therefore, activation of the CAs present in the brain represents a
promising and rather unexplored approach to increase brain activity,
especially memory and learning, with significant potential benefits
in older individuals or in patients affected by Alzheimer’s disease
and other forms of dementia. Sun and Alkon first suggested
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4cc02346c that carbonic anhydrase activators (CAAs) can enhance learning
5980 | Chem. Commun., 2014, 50, 5980--5983
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
and memory, thus eliciting a nootropic effect.9 For example, in demonstrated the superiority of the imidazole structural unit
the hidden-platform water maze, rats injected with the CAAs (pKa = 6–7) against the primary amino moiety (pKa = 9–10) as a
phenylalanine or imidazole have significantly higher latencies proton-shuttling group.12 Activation of CA with bis-azoles such as
to escape. In further support of the role of CA in this learning bis-imidazole 2 was discovered almost simultaneously. In these
model, the CA inhibitor acetazolamide impaired memory and N-alkylated bis-azoles 2, one imidazole ring acts as a hydrophobic
prevented the improvements in learning observed after injection anchor and the other one as a proton shuttle (Chart 1).13 Subsequent
of both phenylalanine and imidazole.9
studies performed by Silverman’s group on activation of CA V using
Memory improvements in rodents have also been observed histidine analogues 4-bromoethylimidazole (4-BEI, 3), 2-chloromethyl-
through CA activation with the dipeptide carnosine.10 Finding imidazole (2-CMI), 4-chloromethyl-imidazole (4-CMI, 4) as covalent CA
potent and selective CAAs with a pharmacokinetic profile that activators confirmed the efficiency of imidazoles as CA activators.14 An
could ensure good brain penetrability represents the critical extensive SAR study on pyridinium azoles by our team15 confirmed
next step towards the generation of a potential nootropic effect. the imidazole moiety as the best proton shuttle (e.g., compound 5),
CA activators act by speeding up the deprotonation of zinc due to its pKa in the maximum activity range for most CA isozymes
bound water (the rate-determining step, eqn (2) in the catalytic (6.5–8.5 pKa units). The study demonstrated the efficiency of the
mechanism), with the generation of the active form of the enzyme:11 pyridinium moiety to anchor the activator at the rim of the active
site.15 Subsequent CA activation studies with various imidazole
ꢀ
EZn2þ ꢀ OH þ CO2 Ð EZn2þ ꢀ HCO3
ꢀ
derivatives (4-methylimidazole 6, L- and D-histidine (7, 8), carnosine
9 and congeners), and with the 4-(pyridinium-ethyl)imidazoles 10–12
recently synthesized by our groups confirmed this conclusion.16–18
The CAA design approach was supported by site-directed muta-
genesis studies and X-ray crystal structures of the complexes of
various CA isozymes with imidazole-based activators.17–19
(1)
þH2O
Ð
EZn2þ ꢀ OH2 þ HCO3
ꢀ
EZn2þ ꢀ OH2 Ð EZn2þ ꢀ OHꢀ þ Hþ -rate determining step
(2)
The positively-charged molecules 5 and 10–12 were shown to
generate membrane-impermeant CA activators useful for studying
the physiology of membrane-bound CA isozymes. Moreover, the
same studies15,18 revealed that small differences in the structure of
activators 10–12 could trigger dramatic changes in the activation
profile of these compounds through the exploitation of the differ-
ences in the amino acids lining the active site of various isozymes.
Very recently, McKenna’s team revealed the existence of several
In the presence of an activator A, eqn (2) becomes (3):11
Â
Ã
EZn2þ ꢀ OH2 þ A Ð EZn2þ ꢀ OH2 ꢀ A
enzyme-activator complexes
Â
þÃ
Â
Ã
EZn2þ ꢀ OHꢀ ꢀ AH Ð EZn2þ ꢀ OHꢀ ꢀ AHþ
Ð EZn2þ ꢀ OHꢀ þ AHþ
(3)
Comprehensive structure–activity relationship (SAR) studies done by binding pockets for imidazole CAAs such as 4-methylimidazole 6
us or by others11 revealed that efficient CA activators possess a and its isomers.21 Encouraged by the very promising results obtained
proton-shuttling group attached to a hydrophobic/amphiphilic with CAAs of type 10–12, we decided to investigate compact bis-
aromatic/heterocyclic anchor via a short, flexible linker (Chart 1, imidazoles 13 in which two imidazole moieties were C-linked via an
inset). We identified imidazole as a very potent proton shuttling ethyl linker, attached to the 4(5)-position of the imidazole ring. In
moiety, in combination with various anchors (Chart 1). In the study these new CAAs, one imidazole moiety acts as the proton-shuttling
of CA activation with histamine 1, X-ray crystallography revealed that moiety, while the other imidazole anchors the activator at the rim of
the activator is anchored at the entrance of the active site cavity the active site, being structurally related to their congeners 2 but
through the amino group. The imidazole moiety participates in having the full proton shuttling capabilities conferred by a free
shuttling protons between the active site and the bulk solvent, thus imidazole NH. Hydrophobic substituents of increased steric
acting as a second proton shuttle of the enzyme. This study bulk were placed in the 2-position of the imidazole ring in order
Chart 1 Evolution of CA activator design: the activation potency of representative lead compounds 1–13 at a concentration of 10 mM against different
CA isozymes,11,13,15,16,18,20 and the proposed design for the novel series of CA activators 13. The general structure of carbonic anhydrase activators is
shown in the inset.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5980--5983 | 5981

View Article Online
ChemComm
Communication
to facilitate the anchoring of the activator on the rim of the activator within this class of compounds, thus validating our
different isozymes of CA. These substituents also increase new design.
lipophilicity of the bis-imidazoles 13, thus facilitating the
crossing of the blood–brain barrier which is critical for achieving were the mitochondrial isozyme CA VA and the brain cytosolic
the nootropic effect. isozyme CA VII, followed by membrane-bound hCA IV, hCA XII,
The most susceptible isozymes activated with bis-imidazoles 13
In terms of synthesis, bis[1H-imidazol-4(5)yl]alkanes can be pre- hCA IX and hCA XIV. The cytosolic isozymes hCA I and hCA II were
pared from a-dihaloalkyldione and formamide at elevated tempera- the least susceptible to be activated with this class of compounds.
tures (180 1C) via Bredereck cyclization.22 However, this method is
Importantly, in the case of hCA VA and hCA VII, we were able
limited to systems in which the two a-keto units in the molecule are to identify activators with nanomolar potency. For these isozymes
separated by four to eight C atoms, with longer spacers generating the increase in steric bulk of the substituents placed in position 2
higher yields. A better option for the synthesis of bis-imidazoles of of the imidazole rings generally caused a decrease in the potency
type 13 having short alkyl spacers is the strategy of Schunack et al.,23 of the activators, probably due to the limited space available in the
who prepared them through the condensation of amidines with active site and the decreased mobility of the heterocyclic rings
a-dihaloalkyldione in the presence of ammonia. Since this method substituted with bulky substituents such as iPr (13d) and Ph (13e).
requires an autoclave and generates only moderate yields even after The most potent activators were 13a (R = H), with a KA = 21 nM for
20 h reaction time, we tested a microwave-mediated procedure hCA VA, KA = 15 nM for hCA VII and 13b (R = CH3) with a KA =
based on this synthetic strategy.
9 nM for hCA VA and KA = 20 nM for hCA VII. Importantly, these
The novel MW procedure allows the achievement of high activators were three orders of magnitude more selective for hCA VA
temperatures and moderate pressures, decreasing the reaction and hCA VII as compared with the other cytosolic and membrane-
time significantly. Using this procedure, we were able to generate bound CA isozymes tested (Table 1).
bis[1H-imidazol-4(5)yl]ethane derivatives 13a–e, substituted at
Thus, bis-imidazoles 13 displayed only micromolar potency
position 2 of the imidazole ring, from 1,6-dibromohexane-2,5- against membrane-bound hCA IV, hCA XII, hCA IX and hCA XIV
dione 1524 and corresponding alkylamidines 1625 in the presence and cytosolic isozymes hCA I and hCA II. Moreover, for these
of ammonia solution in methanol (Scheme 1, ESI†).
isozymes the increase in steric bulk of substituents placed in
Bis-imidazoles 13a–e were subsequently tested for the acti- position 2 of the imidazole rings in compounds 13 was not
vation of eight CA isozymes involved in brain homeostasis, dramatically decreasing the activation potency. This particular
namely the mitochondrial hCA VA, the cytosolic hCA I, hCA II, activation profile may suggest a different binding location of
hCA VII and the membrane-bound hCA IV, hCA IX, hCA XII and bis-imidazoles 13 in the active sites of hCA VA and hCA VII as
hCA XIV, using the CO2 hydration assay method (Table 1, see compared with the other CA isozymes tested.
ESI† for details).
The recent identification of multiple binding sites for dif-
Data from Table 1 revealed an excellent activation profile for ferently substituted imidazoles within the active site of hCA II21
the new compounds. For all isozymes we identified activators supports this hypothesis. Work is in progress to shed light on
that were more potent than histamine 1, the prototype CA the structural origin of the potency and selectivity of these
novel CA activators.
In conclusion, we are proposing a novel set of nanomolar CA
activators that act selectively on CA VA and CA VII, key CA isozymes
involved in brain metabolism. Their potency and selectivity against
other CA isozymes present at the level of the brain, together with
their optimum lipophilicity recommend them as novel probes for
elucidation of the role of CA VA and CA VII in brain physiology and
potential modulators of learning, memory and cognition.
This work was supported by TUSP Dean’s Office and an EU
Grant (Dynano).
Scheme 1 Synthesis of bis[1H-imidazol-4(5)yl]ethane derivatives 13a–e.
Reactions and conditions: (i) Br2, H2SO4, MeOH, 0 1C to r.t., 12 h; (ii) HCl,
EtOH 0 1C to r.t., 2 h then NH3 7 N, MeOH 4 h r.t.; (iii) NH3–MeOH 7 N, MW
200 W, 120 1C, 40 min.
Notes and references
1 C. T. Supuran, Nat. Rev. Drug Discovery, 2008, 7, 168; V. Alterio,
A. Di Fiore, K. D’Ambrosio, C. T. Supuran and G. De Simone, Chem.
Rev., 2012, 112, 4421.
Table
imidazoles 13a–e
1
Activation profile of human CA isozymes (hCAs) with bis-
2 S. Parkkila, EXS, 2000, 79; C. K. Tong, L. P. Brion, C. Suarez and
M. Chesler, J. Neurosci., 2000, 20, 8247; G. N. Shah, B. Ulmasov,
A. Waheed, T. Becker, S. Makani, N. Svichar, M. Chesler and W. S. Sly,
Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 16771.
KA (mM)
hCA hCA hCA hCA hCA
hCA hCA hCA
IX XII XIV
No.
I
II
IV
VA
VII
3 M. Chesler, Physiol. Rev., 2003, 83, 1183.
Histamine 2.1
125
37.5
4 S. Makani and M. Chesler, J. Neurosci., 2007, 27, 7438.
5 T. H. Maren, C. W. Conroy, G. C. Wynns and D. R. Godman,
J. Pharmacol. Exp. Ther., 1997, 280, 98.
6 W. S. Sly and P. Y. Hu, Annu. Rev. Biochem., 1995, 64, 375; W. S. Sly,
D. Hewett-Emmett, M. P. Whyte, Y. S. Yu and R. E. Tashian, Proc.
Natl. Acad. Sci. U. S. A., 1983, 80, 2752.
13a
13b
13c
13d
13e
16.4
23.5
1.13
4.81
5.96
68.5 1.25 0.021
69.3 4.51 0.009
76.8 2.37 0.037
80.4 9.50 0.131
78.7 12.4 0.052
0.015 9.51
0.020 13.1
0.071 24.7
0.054 20.1
0.089 18.5
8.63 13.9
5.40 10.2
6.13 18.5
9.85 24.6
12.6 25.9
5982 | Chem. Commun., 2014, 50, 5980--5983
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
7 W. Meier-Ruge, P. Iwangoff and K. Reichlmeier, Arch. Gerontol. 16 A. Scozzafava and C. T. Supuran, J. Med. Chem., 2002, 45, 284.
Geriatr., 1984, 3, 161.
8 C. Masseguin, S. LePanse, B. Corman, J. M. Verbavatz and
J. Gabrion, Neurobiol. Aging, 2005, 26, 917.
9 M. K. Sun and D. L. Alkon, J. Pharmacol. Exp. Ther., 2001, 297, 961;
M. K. Sun and D. L. Alkon, Trends Pharmacol. Sci., 2002, 23, 83.
10 A. R. Hipkiss, J. Alzheimer’s Dis., 2007, 11, 229.
11 C. T. Supuran and A. Scozzafava, EXS, 2000, 197; M. Ilies, A. Scozzafava
and C. T. Supuran, in Carbonic Anhydrase - its Inhibitors and Activators,
ed. C. T. Supuran, A. Scozzafava and J. Conway, CRC Press, Boca Raton,
17 D. Duda, C. Tu, M. Qian, P. Laipis, M. Agbandje-McKenna, D. N.
Silverman and R. McKenna, Biochemistry, 2001, 40, 1741; I. Elder,
C. Tu, L. J. Ming, R. McKenna and D. N. Silverman, Arch. Biochem.
Biophys., 2005, 437, 106; C. Temperini, A. Scozzafava, D. Vullo and
C. T. Supuran, Chem. – Eur. J., 2006, 12, 7057.
18 K. Dave, A. Scozzafava, D. Vullo, C. T. Supuran and M. A. Ilies, Org.
Biomol. Chem., 2011, 9, 2790; K. Dave, M. A. Ilies, A. Scozzafava,
C. Temperini, D. Vullo and C. T. Supuran, Bioorg. Med. Chem. Lett.,
2011, 21, 2764.
2004, p. 317; C. Temperini, A. Scozzafava and C. T. Supuran, in Drug 19 K. M. Jude, S. K. Wright, C. Tu, D. N. Silverman, R. E. Viola and
Design of Zinc-Enzyme Inhibitors – Functional, Structural, and Disease D. W. Christianson, Biochemistry, 2002, 41, 2485.
Applications, ed. C. T. Supuran and J. Y. Winum, Wiley, Hoboken, 2009, 20 C. T. Supuran, A. Dinculescu and A. T. Balaban, Rev. Roum. Chim., 1993,
p. 473.
38, 343; C. T. Supuran, M. Barboiu, C. Luca, E. Pop, M. E. Brewster and
A. Dinculescu, Eur. J. Med. Chem., 1996, 31, 597; M. A. Ilies, M. D. Banciu,
M. Ilies, F. Chiraleu, F. Briganti, A. Scozzafava and C. T. Supuran,
Eur. J. Med. Chem., 1997, 32, 911.
12 F. Briganti, S. Mangani, P. Orioli, A. Scozzafava, G. Vernaglione and
C. T. Supuran, Biochemistry, 1997, 36, 10384.
13 C. T. Supuran, A. T. Balaban, P. Cabildo, R. M. Claramunt, J. L.
Lavandera and J. Elguero, Biol. Pharm. Bull., 1993, 16, 1236; C. T. 21 M. Aggarwal, B. Kondeti, C. Tu, M. Maupin, D. N. Silverman and
Supuran, R. M. Claramunt, J. L. Lavandera and J. Elguero, Biol.
Pharm. Bull., 1996, 19, 1417.
R. McKenna, IUCrJ, 2014, 1, 129.
22 B. Miller, J. Altman and W. Beck, Synthesis, 1997, 347.
14 J. N. Earnhardt, S. K. Wright, M. Qian, C. Tu, P. J. Laipis, R. E. Viola 23 C. Leschke, J. Altman and W. Schunack, Synthesis, 1993, 197.
and D. N. Silverman, Arch. Biochem. Biophys., 1999, 361, 264.
15 M. Ilies, M. D. Banciu, M. A. Ilies, A. Scozzafava, M. T. Caproiu and
C. T. Supuran, J. Med. Chem., 2002, 45, 504.
24 C. Meziere, M. Fourmigue, E. Canadell, R. Clerac, K. Bechgaard and
P. Auban-Senzier, Chem. Mater., 2000, 12, 2250.
25 P. L. Barker, P. L. Gendler and H. Rapoport, J. Org. Chem., 1981, 46, 2455.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5980--5983 | 5983