I

GI

a

AA

KPAMM

1

paocpsaievobou[mstisa

HT

0d

Chemico-Biological Interactions 187 (2010) 172–176

Contents lists available at ScienceDirect

Chemico-Biological Interactions

journa l homepage: www.e lsev ier .com/ locate /chembio int

nteractions of pyridinium oximes with acetylcholinesterase

oran Sinko ∗, Josipa Brglez, Zrinka Kovariknstitute for Medical Research and Occupational Health, P.O. Box 291, HR - 10001 Zagreb, Croatia

r t i c l e i n f o

rticle history:vailable online 24 April 2010

eywords:yridinium oxime

a b s t r a c t

Catalytic activity of acetylcholinesterase (AChE; EC 3.1.1.7) was studied in the presence of oximes HI-6,K114, K127 and K203, and inhibition constants were determined for the reversible enzyme–inhibitorcomplex (KI). Based on the mixed inhibition model, inhibition constants were 0.020 mM for HI-6,0.0021 mM for K114, 0.175 mM for K127, and 0.036 mM for K203. Molecular modelling of AChE–oxime

cetylcholinesteraseixed inhibitionolecular modelling

complexes was used to determine amino acid residues of the active site involved in the interactions. Bis-oxime K114 achieved the best stabilization in the active site due to �–� interaction between its threearomatic rings and Tyr124, Tyr341 and Trp86, and hydrogen bonds formed by its oxime groups withGly121 and Glu285. Mono-oximes HI-6 and K203, which inhibited the enzyme with similar potency,showed similar positions of their pyridinium rings in the active site. The weakest inhibitor, K127, alsoformed several hydrogen bonds with the active site residues, but due to its long linker it was more likely

al site

stabilized at the peripher

. Introduction

Pyridinium oximes are compounds of interest owing to theirotency to reactivate the organophosphorus compound inhibitedcetylcholinesterase (AChE, EC 3.1.1.7) [1]. Pesticides based onrganophosphorus compounds (OP compounds), along with OPhemical warfare agents, are recognized as a constant threat ofossible terroristic attacks [2,3]. The mechanism of nerve agent poi-oning involves phosphorylation of the serine hydroxyl group in thective site of AChE, leading to inactivation of this important enzymenvolved in neurotransmission [4]. The role of pyridinium oximes isssential for public threat reduction due to their efficacy of reacti-ation of organophosphorus inhibited AChE. Reversible inhibitionf AChE by oximes is also an important property because oximeinding protects AChE from inhibition by OP compounds. Thus far,nly four pyridinium oximes have been approved for emergencyse under special conditions: TMB-4, 2-PAM, HI-6 and obidoxime5,6]. On the other hand, there is no universal antidote against

ajority of OP compounds and, therefore, development of a broadpectrum oxime antidote is an ongoing task. Several features need

o be included in an oxime antidote to be adequate in the detoxify-ng treatment: (a) reactivation of inhibited AChE by oxime antidotehould be fast and complete, (b) oxime antidote needs to have rel-tively high affinity for both AChE and OP-inhibited AChE, and (c)

∗ Corresponding author at: Institute for Medical Research and Occupationalealth, Biochemistry Unit, Ksaverska cesta 2, HR-10 000 Zagreb, Croatia.el.: +385 1 4673 188; fax: +385 1 4673 303.

E-mail address: gsinko@imi.hr (G. Sinko).

009-2797/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.cbi.2010.04.017

(Tyr124), which could explain lower AChE affinity for this oxime.© 2010 Elsevier Ireland Ltd. All rights reserved.

low toxicity of the oxime itself is important also due to the toxiceffects of OP-compounds present in affected persons.

In this work, structure–activity relationships of four pyridiniumoximes were analyzed based on their affinity for AChE, and aminoacids of the AChE active site gorge involved in binding the testedpyridinium oxime were indentified.

2. Materials and methods

2.1. Chemicals

Oximes K114 [(E)-1,1′-(1,4-phenylenebis(methylene))bis(4-((E)-(hydroxyimino)-methyl)-pyridinium) bromide], K127 [(E)-4-carbamoyl-1-(2-(2-(4-((hydroxyimino)-methyl)pyridinium-1-yl)-ethoxy)ethyl)pyridinium bromide] and K203 [4-carbamoyl-1-((E)-4-(4-((E)-(hydroxyimino)-methyl)pyridinium-1-yl)but-2-enyl)pyridinium bromide] [7] were provided by Dr. KamilKuca (Department of Toxicology, Faculty of Military HealthSciences, Hradec Králové, Czech Republic). HI-6 [(E)-1-(((4-carbamoylpyridinium-1-yl)methoxy)methyl)-2-((hydroxyimino)methyl)pyridinium bromide] was synthesized in the Departmentfor Organic Chemistry, Faculty of Science, University of Zagreb,Croatia. Recombinant human acetylcholinesterase (AChE, EC3.1.1.7) was provided by Professor Palmer Taylor (Skaggs Schoolof Pharmacy and Pharmaceutical Sciences, UCSD, La Jolla, USA).

Enzyme substrate acetylthiocholine iodide (ATCh) and the assayreagent 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) were purchasedfrom Sigma Chemical Co., St. Louis, MO, USA. All experiments weredone in 0.1 M sodium phosphate buffer, pH 7.4, at 25 ◦C. Allchemicals used were of analytical grade.

ical Interactions 187 (2010) 172–176 173

2

i(ist

2

mA(hIc

v

V

wtfipv

2

eotATihJ2ttaaaactrt

3

3

Awtvaia

The inhibition constants determined are given in Table 1. Theutilized mixed inhibition model resulted in slightly lower con-stants than previously evaluated for native human erythrocyteAChE using the Hunter-Downs plot when Ki were 0.031, 0.011 and

G. Sinko et al. / Chemico-Biolog

.2. Enzyme assay

Enzyme activity was measured spectrophotometrically accord-ng to the Ellman procedure at 25 ◦C, with the thiol reagent DTNB0.3 mM final concentration) and ATCh as substrate [8–10]. Increasen absorbance was read at 412 nm up to 2 min on a CARY 300pectrophotometer (Varian Inc., Australia) with a temperature con-roller and stopped-flow attachment (SFA-20, HI-Tech Ltd, UK) [11].

.2.1. Reversible inhibition of AChEReversible inhibition of AChE by the oxime was measured in a

edium containing AChE suspended in buffer, DTNB, the oxime andTCh. Concentrations of the oxime (2.5 �M–0.8 mM) and substrate

25 �M–1.0 mM) were selected so that the oxime-induced ATChydrolysis did not interfere with the studied interactions [10,11].

nhibition constants were evaluated from the effect of substrateoncentration (S) on the degree of inhibition according to equation:

i = V ′mS

K ′m + S

′m = Vm

11 + (I/˛KI)

K ′m = Km

1 + (I/KI)1 + (I/˛KI)

here S is the substrate ATCh, I is the inhibitor (oxime), KI ishe enzyme–oxime inhibition (dissociation) constant of a complexormed at the catalytic site, ˛KI is the Michaelis complex–oximenhibition (dissociation) constant of a complex formed at theeripheral site, Km is the Michaelis complex, and Vm is maximalelocity.

.3. Molecular modelling

Molecular modelling was performed using the Accelrys Discov-ry Studio 2.1. Flexible docking protocol (CHARMm forcefield [12])n AChE (PDB code 1B41; [13]) was used and software was allowedo change conformation of residues of the following amino acids:sp74, Trp86, Tyr124, Trp286, Phe295, Phe297, Tyr337 and Phe338.he protocol resulted in clusters of oxime poses. Validation of dock-ng poses was performed using consensus scoring [14], after posesave been scored with PMF04 [15], LigScore2 [16], PLP2 [17] and

ain [18] scoring functions. Selected poses were ranked among top0% in consensus scoring. Although this methodology focused onhe top-ranked ligand pose, with the underlying assumption thathe orientation/conformation of the docked compound is the mostccurate, this may vary from crystallographic structure of the lig-nd. The ones were selected in which oxime binds to the catalyticctive site and the peripheral anionic site, PAS, simultaneously, ort least 6 Å away from Trp86 (the choline binding site) where theholine part of the substrate molecule would probably be posi-ioned in the Michaelis complex [19,20]. Crystal structure of humanecombinant AChE was prepared by removing the fasciculin pep-ide, alternate conformations of residues and water molecules.

. Results and discussion

.1. Inhibition potency

Tested bis-pyridinium oximes were reversible inhibitors ofChE, and all four oximes inhibited AChE in the same manner,hich the authors describe as mixed inhibition: the presence of

he oxime reduces the maximum velocity Vm, and increases the Km

alue (Fig. 1). This means that oximes can bind to the free enzyme,nd to the Michaelis complex of the enzyme and substrate. Similar-ty in Km and Vm constants determined from inhibition by oximesdditionally confirms selection of the model of mixed inhibition as

Fig. 1. Representative plot of AChE activity and the effect of substrate concentrationon AChE activity in the presence and absence of oxime K127. Experimental data oftwo series were normalized by the total enzyme concentration.

the proper one (Table 1). Parameter ˛ describes the decrease in theMichaelis complex affinity for an oxime compared to the affinityof the free enzyme for an oxime. Average ˛ value 3.9 and the highstandard deviation of ˛ values suggest that there is no significantdifference in ˛ values calculated from the inhibition by four oximes.The common ˛ value would mean that oximes interact with thesame enzyme–substrate specie, the Michaelis complex. The noticeddecrease in the Michaelis complex affinity for an oxime is probablycaused by repulsion between the positively charged choline part ofthe substrate and that of positively charged pyridinium rings of theoxime.

Fig. 2. Structural formulas of HI-6, K114, K127 and K203.

174 G. Sinko et al. / Chemico-Biological Interactions 187 (2010) 172–176

Fig. 3. Orientation of K114 and K127 inside the active site gorge of AChE. (A) K114 forms interactions simultaneously with Trp86 (the choline binding site) and Tyr124, aconstituent of the peripheral anionic site (PAS). The third aromatic ring creates a �–� interaction with Trp337, providing additional stabilization of K114 inside the AChEgorge. (B) Orientation of K114 preferably at PAS simulates binding to the Michaelis complex away from Trp86, where a substrate molecule should be present. Formation ofa est afe to theH 341. (Di

0Hpctt

TK

�–� sandwich with Tyr124 of PAS agrees with the kinetic data showing the highnable simultaneous binding at catalytic and peripheral binding sites of AChE, due-bonds with Glu285 and Glu202, and a cation–� interaction with Tyr124 and Tyr

nteraction with Tyr341. Hydrogen bonds are shown as dark green dashed lines.

.090 mmol dm−3 for HI-6, K114 and K203, respectively [21–23].

owever, regardless of the calculation applied, K114 was the mostotent inhibitor of AChE, with 10 times lower inhibition constantompared to HI-6 (Table 1). Furthermore, the affinity of K114 forhe Michaelis complex was higher compared to affinities of otherhree oximes for the free enzyme. This clearly suggests that the

able 1inetic parameters of oxime inhibition of AChE.

Oxime KI/mmol dm−3 ˛

HI-6 0.020 ± 0.004 3.08 ± 0.98K114 0.0021 ± 2 × 10−5 3.85 ± 0.06K127 0.175 ± 0.018 4.83 ± 0.93K203 0.036 ± 0.003 3.98 ± 0.65

finity of K114 for the Michaelis complex. (C) Structure of K127 is not adequate tolength of the linker and the presence of the oxygen atom. However, K127 creates) K127 positioned at PAS creates H-bonds with Glu285 and Tyr124, and cation–�

para-xylene group of the linker greatly contributes to the overallaffinity of AChE for K114. These findings suggest K114 to be a good

candidate for reactivation trials. The classical oxime HI-6 was usedas a reference. Surprisingly, HI-6 and K203 show similar inhibi-tion potency although their linkers and the oxime group positionon the pyridinium ring are quite different (Fig. 2). K127 was the

Vm/min Km/mmol dm−3

0.251 ± 0.006 0.081 ± 0.0080.253 ± 0.0002 0.088 ± 0.00050.268 ± 0.003 0.091 ± 0.0050.252 ± 0.003 0.088 ± 0.004

G. Sinko et al. / Chemico-Biological Interactions 187 (2010) 172–176 175

Table 2List of oxime–human acetylcholinesterse interactions.

Oxime Linker structure N of oxime groups, position on pyridine ring Molecular interactions of complex

EI ESIa

HI-6 CH2–O–CH2 One group, ortho- position (D) Ser203 (2.5 Å)b

(A) Tyr124 (2.1 Å)(�–�) Trp86, Tyr341

(D) Ser298 (2.3 Å), Phe297 (2.0 Å)(A) Phe295 (2.2 Å), Arg296 (1.8 Å)(�–�) Trp86, Tyr341, Tyr72, Tyr124

K114 CH2–C6H4–CH2 Two groups, para- position (D) Glu285 (2.0 Å)(A) Ser298 (2.2 Å), Phe299 (2.2 Å),Gly121 (2.4 Å)(�–�) Trp86, Tyr72, Tyr124, Tyr341

(D) Glu285 (2.3 Å)(A) none(�–�) Tyr72, Tyr124, Tyr341

K127 CH2CH2–O–CH2CH2 One group, para- position (D) Glu202 (2.0 Å), Glu285 (2.1 Å)(A) Ser298 (2.1 Å)(�–�) Tyr124, Tyr341

(D) Glu285 (2.0 Å), Tyr124 (2.1 Å)(A) Ser298 (2.1 Å)(�–�) Tyr124, Tyr341

K203 CH2–CH CH–CH2 One group, para- position (D) His447 (1.9 Å)(A) none(�

(D) Glu285 (2.5 Å), Asp74 (2.3 Å)(A) Ser298 (2.3 Å)

hydroi

otnbAr(l

3

csrodMosfT

agcGawiabTiirwatfKfwgat

Ser203 of the catalytic triad. Moreover, the oxime group of HI-6 andSer203 create an H-bond. Lower KI of HI-6 compared to KI of K203can be due to an additional H-bond between the oxygen atom inthe linker of HI-6 and Tyr124, and larger proximity between the

a ESI complex was simulated without substrate (see Materials and methods).b H-bond value is distance between hydrogen and H-bond acceptor, (D) donor of

nteraction.

xime with the lowest inhibition potency. This can be attributed tohe structure of the linker of K127. In stretched conformation, twoitrogen atoms of pyridinium rings are 7.5 Å apart, e.g. the distanceetween two nitrogen atoms of pyridinium rings of HI-6 is 4.9 Å.dditionally, the presence of an oxygen atom in the linker causesigidity in its centre due to higher energy of rotation of O–C bondscf. Ref [11,24]). It seems therefore that the linker of K127 is tooong to allow similar accommodation of K127 compared to HI-6.

.2. Modelling of binding interactions

Binding of oximes to the free enzyme and to the Michaelisomplex was simulated by molecular modelling. Flexibility of theelected AChE residues, in the authors’ opinion, approached aealistic situation in solution that enabled oximes to create vari-us interactions with gorge residues. Due to lack of experimentalata, it was not possible to simulate the substrate position in theichaelis complex. Therefore, the criterion was introduced that the

xime molecule should be at least 6 Å away from Trp86 involved inubstrate stabilization during the enzyme hydrolysis. Interactionsormed between AChE active site gorge and oximes are listed inable 2.

Position of K114 in AChE shows �–� interactions with Trp86nd the residues of PAS, Tyr124 and Trp286 (Fig. 3A). One oximeroup of K114 forms a hydrogen bond (H-bond) with the mainhain of Gly121 while another oxime group forms H-bonds withlu285, and with amino group of Ser298 and Phe299. Addition-lly, the para-xylene group of the linker forms �–� interactionith Tyr341. These interactions enable strong stabilization of K114

nside the unoccupied gorge of AChE. K114 also forms strong inter-ctions with the Michaelis complex of AChE (Fig. 3B): H-bondetween Glu285 and one oxime group, �–� sandwich betweenyr124 and two pyridinium rings, Tyr72 and Phe279 create �–�nteractions with one of the pyridinium rings. Closer inspection ofnteractions between K127, the weakest inhibitor, and AChE gorgeesidues reveals kinetic behaviour of K127. K127 forms interactionith Trp86, but instead of stacking of two aromatic rings there is

n interaction between the indole ring of Trp86 and � electrons ofhe C–N double bond of the oxime group (Fig. 3C). This oxime grouporms also an H-bond with Glu202. The second pyridinium ring of127 interacts with Tyr124 and Tyr72, but its position is not optimal

or �–� stacking with Tyr124. Tyr341 forms cation–� interactionsith both pyridinium rings of K127. Hydrogen atom from its amide

roup forms an H-bond with Glu285 while the oxygen atom formsn H-bond with the main chain of Ser298. Moreover, rigidity ofhe K127 linker may be responsible for insufficient �–� stacking

–�) Trp86, Tyr341 (�–�) Tyr72, Tyr124, Tyr341

gen bond, (A) acceptor of hydrogen bond, and (�–�) pi–pi interaction or cation–�

with Tyr124. The second orientation of K127 simulates binding tothe Michaelis complex (Fig. 3D). K127 is mostly positioned at PASin perpendicular orientation of the pyridinium ring compared toTyr124 forming H-bond with Glu285 and Ser298. Amide group ofK127 forms an H-bond with the oxygen of Tyr124 and the sec-ond pyridinium ring of K127 interacts with Tyr341 with a cation–�interaction.

For HI-6 and K203, structures were chosen in which the oximegroup was positioned toward the catalytic site of AChE and thesuperpositioned structures are given in Fig. 4. Pyridinium ringsare overlapped while oxime groups are positioned away from oneanother. From the aspect of reactivation, accommodation of HI-6 and its oxime group is favourable because it is directed toward

Fig. 4. Superposition of docked HI-6 and K203. Both oximes, HI-6 (grey) and K203(black), interact with Trp86 and other residues inside the active site gorge of AChE.Residues are shown with different conformations.

1 ical In

ppomic

4

dria(cioacrtfots

C

A

mrHrS

R

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

76 G. Sinko et al. / Chemico-Biolog

yridinium ring of HI-6 and indole ring of Trp86, rather than toroximity between the pyridinium ring of K203 and Trp86. Thexime group of K203 forms an H-bond with the oxygen from theain chain of His447, a constituent of the catalytic triad. These find-

ngs are in agreement with crystallographic structure of HI-6·AChEomplex [25].

. Conclusion

High affinity for AChE for bis-pyridinium oximes greatlyepends on the structure of the linker between two pyridiniumings. As shown earlier, it is important that the length and flexibil-ty of the linker are sufficient for the oxime to bind simultaneouslyt the choline binding site (Trp86) and the peripheral anionic siteTyr124, Trp286). If the linker is more than 5 carbon atoms long orontains an oxygen atom, the oxime affinity for AChE is lower, ast was shown for K127. Nevertheless, reactivation potency of thexime is not only contributed by its affinity for AChE but also by itsbility to form a transition state with an AChE-organophosphateonjugate, and resulting in successful reactivation. For example,eactivation of tabun inhibited AChE was six times faster by K203han by K114, although K114 had three times higher affinityor phosphorylated AChE [22,23]. A new approach to the designf a potent oxime reactivator requires simulation of the transi-ion state that occurs during reactivation as an insight into thetructure–activity relationship.

onflict of interest statement

None declared.

cknowledgements

We thank Professor Palmer Taylor, Skaggs School of Phar-acy and Pharmaceutical Sciences, UCSD, La Jolla, USA, for the

ecombinant enzyme, and Dr. Kamil Kuca, Faculty of Militaryealth Sciences, Hradec Kralove, Czech Republic, for K oximes. This

esearch was supported by the Ministry of Science, Education andports of the Republic of Croatia (Grant no. 022-0222148-2889).

eferences

[1] R.M. Dawson, Review of oximes available for the treatment of nerve agentpoisoning, J. Appl. Toxicol. 14 (1994) 317–331.

[2] H. Thiermann, L. Szinicz, P. Eyer, N. Felgenhauer, T. Zilker, F. Worek, Lessonsto be learnt from organophosphorus pesticide poisoning for the treatment ofnerve agent poisoning, Toxicology 233 (2007) 145–154.

[3] M. Nagao, T. Takatori, Y. Matsuda, M. Nakajima, H. Iwase, K. Iwadate, Definitiveevidence for the acute sarin poisoning diagnosis in the Tokyo subway, Toxicol.Appl. Pharmacol. 144 (1997) 198–203.

[

teractions 187 (2010) 172–176

[4] S.M. Somani, J.A. Romano Jr., Chemical Warfare Agents: Toxicity at Low Levels,CRS Press LLC, Boca Raton, Florida, 2001.

[5] B. Antonijevic, M.P. Stojiljkovic, Unequal efficacy of pyridinium oximes in acuteorganophosphate poisoning, Clin. Med. Res. 5 (2007) 71–82.

[6] M.G. Hamilton, P.M. Lundy, Medical countermeasures to WMDs: defenceresearch for civilian and military use, Toxicology 233 (2007) 8–12.

[7] K. Musilek, D. Jun, J. Cabal, J. Kassa, F. Gunn-Moore, K. Kuca, Design of a potentreactivator of tabun-inhibited acetylcholinesterases synthesis and evaluationof (E)-1-(4-carbamoylpyridinium)-4-(4-hydroxyiminomethylpyridinium)-but-2-ene dibromide (K203), J. Med. Chem. 50 (2007) 5514–5518.

[8] G.L. Ellman, K.D. Courtney V.Jr., R.M. Andres, Featherstone, New and rapid col-orimetric determination of acetylcholinesterase activity, Biochem. Pharmacol.7 (1961) 88–95.

[9] E. Reiner, G. Sinko, M. Skrinjaric-Spoljar, V. Simeon-Rudolf, Comparison of pro-tocols for measuring activities of human blood cholinesterases by the Ellmanmethod, Arh. Hig. Rada Toksikol. 51 (2000) 13–18.

10] G. Sinko, M. Calic, A. Bosak, Z. Kovarik, Limitation of the Ellman method:cholinesterase activity measurement in the presence of oximes, Anal. Biochem.370 (2007) 223–227.

11] G. Sinko, M. Calic, Z. Kovarik, para- and ortho-Pyridinium aldoximes in reactionwith acetylthiocholine, FEBS Lett. 580 (2006) 3167–3172.

12] B.R. Brooks, R.E. Bruccoleri, B.D. Olafson, D.J. States, S. Swaminathan, M. Karplus,CHARMM: a program for macromolecular energy, minimization, and dynamicscalculations, J. Comp. Chem. 4 (1983) 187–217.

13] G. Kryger, M. Harel, K. Giles, L. Toker, B. Velan, A. Lazar, C. Kronman, D. Barak,N. Ariel, A. Shafferman, I. Silman, J.L. Sussman, Structures of recombinantnative and E202Q mutant human acetylcholinesterase complexed with thesnake-venom toxin fasciculin-II, Acta Crystallogr. D Biol. Crystallogr. 56 (2000)1385–1394.

14] M. Kontoyianni, L.M. McClellan, G.S. Sokol, Evaluation of docking performance:comparative data on docking algorithms, J. Med. Chem. 47 (2004) 558–565.

15] I. Muegge, PMF scoring revisited, J. Med. Chem. 49 (2006) 5895–5902.16] S.L. Mayo, B.D. Olafson, W.A. Goddard, DREIDING: a generic force field for

molecular simulations, J. Phys. Chem. 94 (1990) 8897–8909.17] D.K. Gehlhaar, D. Bouzida, P.A. Rejto, in: L. Parrill, M.R. Reddy (Eds.), Rational

Drug Design: Novel Methodology and Practical Applications, American Chem-ical Society, Washington, DC, 1999, pp. 292–311.

18] A.N. Jain, Scoring noncovalent protein-ligand interactions: a continuous differ-entiable function tuned to compute binding affinities, J. Comput. Aided Mol.Des. 10 (1996) 427–440.

19] J.P. Colletier, D. Fournier, H.M. Greenblatt, J. Stojan, J.L. Sussman, G. Zaccai,I. Silman, M. Weik, Structural insights into substrate traffic and inhibition inacetylcholinesterase, EMBO J. 25 (2006) 2746–2756.

20] Y. Bourne, Z. Radic, G. Sulzenbacher, E. Kim, P. Taylor, P. Marchot, Substrateand product trafficking through the active center gorge of acetylcholinesteraseanalyzed by crystallography and equilibrium binding, J. Biol. Chem. 281 (2006)29256–29267.

21] M. Skrinjaric-Spoljar, V. Simeon, E. Reiner, B. Krauthacker, Bispyridinium com-pounds: inhibition of human erythrocyte acetylcholinesterase and protectionof the enzyme against phosphylation, Acta Pharm. Jugosl. 38 (1988) 101–109.

22] Z. Kovarik, A. Lucic Vrdoljak, S. Berend, M. Katalinic, K. Kuca, K. Musilek, B.Radic, Evaluation of oxime K203 as antidote in tabun poisoning, Arh. Hig. RadaToksikol. 60 (2009) 19–26.

23] M. Calic, A. Bosak, K. Kuca, Z. Kovarik, Interactions of butane, but-2-eneor xylene-like linked bispyridinium para-aldoximes with native and tabun-inhibited human cholinesterases, Chem. Biol. Interact. 175 (2008) 305–308.

24] Z. Kovarik, M. Calic, A. Bosak, G. Sinko, D. Jelic, In vitro evaluation of aldoxime

interactions with human acetylcholinesterase, Croat. Chem. Acta 81 (2008)47–57.

25] F. Ekström, A. Hörnberg, E. Artursson, L.-G. Hammarström, G. Schneider, Y.-P.Pang, Structure of HI-6·sarin-acetylcholinesterase determined by X-ray crys-tallography and molecular dynamics simulation: reactivator mechanism anddesign, PLoS One 4 (2009) e5957, doi:10.1371/journal.pone.0005957.