Effect of Different Oximes on Rat and Human Cholinesterases Inhibited by Methamidophos: A Comparative In Vitro and In Silico Study

Authors


Author for correspondence: Félix Alexandre Antunes Soares, Department of Chemistry, Federal University of Santa Maria, Santa Maria, RS CEP 97105-900, Brazil (fax + 55 55 3220 8978, e-mail felix@ufsm.br)

Abstract

Methamidophos is one of the most toxic organophosphorus (OP) compounds. It acts via phosphorylation of a serine residue in the active site of acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), leading to enzyme inactivation. Different oximes have been developed to reverse this inhibition. Thus, our work aimed to test the protective or reactivation capability of pralidoxime and obidoxime, as well as two new oximes synthesised in our laboratory, on human and rat cholinesterases inhibited by methamidophos. In addition, we performed molecular docking studies in non-aged methamidophos-inhibited AChE to understand the mechanisms involved. Our results suggested that pralidoxime protected and reactivated methamidophos-inhibited rat brain AChE. Regarding human erythrocyte AChE, all oximes tested protected and reactivated the enzyme, with the best reactivation index observed at the concentration of 50 μM. Concerning BChE, butane-2,3-dionethiosemicarbazone oxime (oxime 1) was able to protect and reactivate the methamidophos-inhibited BChE by 45% at 50 μM, whereas 2(3-(phenylhydrazono)butan-2-one oxime (oxime 2) reactivated 28% of BChE activity at 100 μM. The two classical oximes failed to reactivate BChE. The molecular docking study demonstrated that pralidoxime appears to be better positioned in the active site to attack the O-P moiety of the inhibited enzyme, being near the oxyanion hole, whereas our new oximes were stably positioned in the active site in a manner similar to that of obidoxime. In conclusion, our work demonstrated that the newly synthesised oximes were able to reactivate not only human erythrocyte AChE but also human plasma BChE, which could represent an advantage in the treatment of OP compounds poisoning.

Acetylcholinesterase (EC 3.1.1.7, AChE) is an enzyme that catalyses the hydrolysis of acetylcholine to choline and acetate. The biological function of this enzyme is to terminate acetylcholine activity in the terminal nervous junction with its effector organs or post-synaptic sites [1]. The mechanism of action of organophosphorus (OP) compounds with anticholinesterase activity involves the phosphorylation of the serine hydroxyl group in the active site of AChE, leading to an inactive enzyme (AChE-OP) [2]. The inactivation of AChE results in the accumulation of acetylcholine at cholinergic receptor sites, causing a cholinergic crisis that can lead to death [3].

Methamidophos is a potent AChE inhibitor used to control plague of insects on a variety of crops [4] and may present anticholinesterase activity against human cholinesterases [5, 6]. In Brazil in particular, methamidophos is commonly used on many crops, including cotton, soybeans and wheat, resulting in food [7] and occupational exposure [8]. A recent study demonstrated that over 90% of small farmers use methamidophos and nearly 60% of them exhibit typical OP intoxication symptoms [9]. Mechanistically, methamidophos may inhibit AChE by covalently binding a serine residue (Ser203) in the active site, and this moiety could form two different enantiomers, Sp and Rp. The Sp enantiomer is more likely to occur and is more easily reactivated by oximes [10].

Currently, treatment for OP poisoning consists of patient stabilisation, reduction of OP absorption, the use of an anti-muscarinic agent, an oxime reactivator (e.g. pralidoxime and obidoxime) and an anti-convulsant. The mechanism of oxime action is based on displacing the phosphoryl group of the AChE-OP complex, owing to the fact that oxime having a higher affinity for the enzyme than OP does, and the high nucleophilic power of the oxime [2]. Furthermore, no oxime can restore AChE activity in structurally different OP moieties. Thus, the structure of the oxime determines its capacity to restore the function of AChE [11].

Despite various positive experimental evidence, the clinical trials with oximes have been controversial [12-14]. As pralidoxime and obidoxime are the most common reactivators available for clinical use, and considering the scarcity of literature data concerning the use of oximes in cases of methamidophos poisoning in physiological conditions, further studies are required to clarify the efficacy of oximes against methamidophos for different cholinesterases and develop new oximes capable of acting as cholinesterase reactivators against different OPs with better results. Furthermore, several in silico studies exemplified the capacity of simulation models to predict the molecular interactions with relative reliability. Therefore, the aim of our work was to study the effectiveness of the newly synthesised oximes in reactivating methamidophos-inhibited AChE in comparison with the effectiveness of the classical oximes (pralidoxime and obidoxime) and perform a molecular docking study to clarify the possible interaction between the oximes and the active site of AChE.

Materials and Methods

Chemicals

2, 2-dinitro-5, 5-ditiobenzoic acid (DTNB) and acetylthiocholine iodide were purchased from Merck; pralidoxime chloride (2-hydroxyiminomethyl-1-methyl-pyridinium), obidoxime dichloride (1,1′-(Oxydimethylene)bis(pyridinium-4-carbaldoxime) dichloride) and methamidophos (O,S-Dimethyl phosphoramidothioate) were purchased from Sigma-Aldrich. The newly synthesised oximes were designated as oxime 1 (butane-2,3-dionethiosemicarbazone oxime) and oxime 2 (3-(phenylhydrazono)butan-2-one oxime) (fig. 1). All other chemicals were of the highest grade available commercially.

Figure 1.

Chemical structures of obidoxime, pralidoxime, butane-2,3-dionethiosemicarbazone oxime (oxime 1) and 3-(phenylhydrazono)butan-2-one oxime (oxime 2).

Animals

Adult male Wistar rats (200–250 g) were obtained from University of Santa Maria and maintained in an air-conditioned room (20–25°C) under natural lighting conditions (cycle 12:12 hr) with water and food (Guabi-RS, Brazil) ad libitum. All experimental procedures were performed according to the guidelines of the Committee on Care and Use of Experimental Animal Resources of the Federal University of Santa Maria, Brazil.

Preparation of erythrocyte ghosts

Haemoglobin-free erythrocyte ghosts were prepared according to previously described method [15] with minor modifications. Briefly, blood of non-fasted healthy voluntary donors was collected. Heparinised human blood was centrifuged (3000 × g, 10 min.) and the plasma removed and kept to test butyrylcholinesterase (BChE) activity. Erythrocytes were washed three times with two volumes of sodium/potassium phosphate buffer (0.1 M, pH 7.4). Then, the packed erythrocytes were diluted in 20 volumes (w/v) of hypotonic sodium/potassium phosphate buffer (6.7 mM, pH 7.4) to facilitate the haemolysis, followed by centrifugation at 30,000 × g (30 min., 4°C). The supernatant was removed and the pellet resuspended in hypotonic phosphate buffer. After two additional washing cycles, the pellet was resuspended in sodium/potassium phosphate buffer (0.1 M, pH 7.4), passed through one more centrifugation at 30,000 × g (30 min., 4°C) and were gently removed. Next, the AChE activity was adjusted to the original activity by appropriate dilution with phosphate buffer (0.1 M, pH 7.4). Aliquots of the erythrocyte ghosts were stored at −20°C until use.

Haemoglobin content present in ghost membranes was measured at 540 nm as the cyano-met-Hb form, but no haemoglobin was detected.

Cerebral tissue preparation

Animals were anaesthetised and killed by decapitation. Brain was quickly removed, placed on ice and homogenised in 10 volumes of 0.1 M sodium/potassium phosphate buffer (pH 7.4) and Triton X100. The homogenate was centrifuged at 400 × g at 4°C for 10 min. to yield a low-speed supernatant fraction (S1) that was used in the experiments. Aliquots were stored at −20°C until use.

Cholinesterases activity

Brain and ghost erythrocyte AChE and human plasma butyrylcholinesterase activities were estimated by the method of Ellman [16], using acetylthiocholine and butyrylthiocoline iodide as substrate, respectively. The rate of hydrolysis of substrate was measured at 412 nm through the release of the thiol compound that, when reacted with DTNB, produces the colour-forming compound TNB [17].

Reactivation/protection of OP-inhibited cholinesterase

Brain and ghost erythrocyte AChE and human plasma butyrylcholinesterase were exposed to oximes in two different assay conditions to indentify a possible protective or a reactivation capacity of oximes:

  1. Protection: the enzyme was exposed to methamidophos and oximes at the same time making up a total incubation period of 60 min.
  2. Reactivation: the enzyme was firstly exposed to methamidophos for 10 min. After this, the oxime tested was added into the medium and the incubation followed for 50 min. (to perform an equal total incubation time period of 60 min.). Enzyme activities were referred to control activity.

The values of the reactivation index (R) were calculated according to the following equation:

display math

The symbol ∆A0 is absorbance provided by mixture with intact AChE (in the final mixture, there was no organophosphate as well as no reactivator); ∆Ai is the absorbance of mixture with inhibited AChE (inhibition by organophosphate, no reactivator). Absorbance provided by mixture where AChE activity was influenced by organophosphate and consequently by reactivator was presented by the last symbol ∆Ar [18].

For the determination of reactivation rate constants, the method described by Worek et al. [19] was used, with some modifications. Briefly, an aliquot of ghosts inhibited by methamidophos (25 μM) for 10 min. was incubated with different oxime concentrations. For the AChE activity measurement, mixture aliquots (40 μL) were transferred to cuvettes containing sodium/phosphate buffer (0.1 M) at pH 7.4, DTNB (0.35 mM) and acetylthiocholine (0.9 mM), at various time intervals within 2-hr incubation.

According to Aldridge and Reiner [20], the reactivation consists of two consecutive steps, represented by the following scheme:

display math

where [EP] is the phosphonylated enzyme, [Ox] the reactivator, [EPOx] the phosphonylenzyme-oxime complex, [E] the reactivated enzyme and [POx] the phosphonylated oxime; KD ([EP] × [Ox]/[EPOx]) is the dissociation constant which describes the affinity of the oxime to [EP] and kr the rate constant for the displacement of the phosphonyl residue from [EPOx] by the oxime. In case of complete reactivation and with [Ox] » [EP]0, a pseudo first-order rate equation can be derived for the reactivation process [21]:

display math

kobs is determined at various oxime concentrations. The amount of kobs is not proportional to the oxime concentration but underlies saturation kinetics [22]. Under pseudo first-order conditions, the observed rate constant for the reactivation of the phosphonylated enzyme can be expressed by the following equation [23]:

display math

representing the activities of the control enzyme (vo), the inhibited enzyme (vi) and the reactivated enzyme at time t (vt). If [Ox] << KD, the second-order reactivation rate constant kr2 (mM/min.), describing the specific reactivity, can be derived from the equation:

display math

Molecular docking

Docking simulations of the oximes with Mus musculus AChE were carried out using AutoDock Vina 1.1.1 [24], followed by redocking with Autodock 4.0.1. The non-aged methamidophos-inhibited Mus musculus AChE obtained from the RCSB Protein Data Bank (http://www.rcsb.org/pdb/) was used as macromolecule (PDB code 2jge). The oximes 1 (butane-2,3-dionethiosemicarbazone oxime) and 2 (3-(phenylhydrazono)butan-2-one oxime), as well as the two classical oximes, were constructed using the programme Avogadro 0.9 and their geometry were optimised with the MMFF 94 force field. Both ligands and macromolecule are previously prepared using AutoDock Tools [25] and Chimera 1.5 [26]. All rotatable bonds within the ligands were allowed to rotate freely, and the receptor was considered rigid. The grid was centred on the active site of AChE and the dimensions of the grid box consisted of 30 × 22 × 30 Å points, with spacing of 1 Å. The exhaustiveness was set to 50. All other parameters were used as defaults. For each ligand docked, the conformation from the lowest binding free energy with inferred inhibitory reactivity was accepted as the best affinity model. The redocking calculation was carried out using Autodock 4.0.1, following the method of Musilek et al. [27]. Briefly, a Lamarckian genetic algorithm (Amber force field) was used, and a population of 150 individuals and 2,500,000 function evaluations were applied. The structure optimisation was performed for 27,000 generations. Docking calculations were set to 100 runs. At the end of calculation, Autodock performed cluster analysis. The 3D affinity grid box was designed to include the full active and peripheral site of AChE. The number of grid points in the x-, y- and z-axes was 60, 60 and 60 with grid points separated by 0.253 Å. The conformations and interactions were analysed using the programmes Accelrys Discovery Studio Visualizer 2.5 and PyMOL [28].

Statistical analysis

Differences between groups were evaluated by one-way anova, followed by Duncan's multiple range tests when appropriate. All values were presented as mean ± S.E.M. and the differences were considered significant when < 0.05. Data analysis and calculation of kinetic constants by linear and nonlinear regression analysis were performed with GraphPad Prism 5.0 (GraphPad, San Diego, CA, USA).

Results

Erythrocyte ghosts were exposed to methamidophos at various concentrations (ranging from 1 to 250 μM) and an aliquot was removed at different time intervals to test the AChE activity (fig. 2). The concentration of 25 μM of methamidophos was chosen to the next experiments, to have an inhibition of approximately 50% of AChE activity for approximately all the incubation period and thus, avoiding a possible re-inhibition process by the excess of methamidophos in the oxime-induced reactivation protocols. A spontaneous reactivation of AChE was not observed in the time interval used at this concentration. The same protocol was used to rat brain AChE and human plasma BChE, with similar results (data not shown). None of the tested oximes had significant effects per se on the activities of AChE and BChE at the tested concentrations. In addition, no direct hydrolytic effect on acetylthiocholine or butyrylthiocholine was observed (data not shown) in the protocols used.

Figure 2.

Kinetics of methamidophos inhibition on AChE activity. The activities are expressed as percentage of control. Enzyme preparations were treated as described in the Materials and Methods section.

Effects of oximes on methamidophos-inhibited rat brain AChE

Fig. 3 shows the effects of exposure to 25 μM methamidophos and the protective effect of oximes on AChE from the rat brain. Obidoxime (fig. 3A) had a slight, but significant, protective effect on methamidophos-induced AChE inactivation. Pralidoxime (fig. 3B) significantly protected AChE from methamidophos-induced inhibition at all tested concentrations (p < 0.05). Similarly, oxime 1 (fig. 3C) at concentrations of 50 μM and 100 μM significantly (p < 0.05) protected against methamidophos-induced AChE inhibition, whereas oxime 2 was able to significantly protect the enzyme activity only at higher concentrations (fig. 3D).

Figure 3.

Protective effect of obidoxime (A), pralidoxime (B), oxime 1 (C) and oxime 2 (D) on methamidophos (MAP)-inhibited AChE activity from rat brain homogenate. Oximes were added to the reaction medium at the same time of methamidophos. The data are expressed as percentage from control group (100%). *Statistically different from control group (p < 0.05); #statistically different from methamidophos 25 μM group (p < 0.05). Results are presented as the mean ± S.E. of at least three independent experiments performed in duplicate.

Effect of oximes on methamidophos-inhibited human erythrocyte ghost AChE

Fig. 4 shows the protective effects of oximes on methamidophos-inhibited AChE from human erythrocyte ghosts. Obidoxime at 50 μM (fig. 4A) plus methamidophos significantly increased cholinesterase activity in comparison with that observed for methamidophos alone (p < 0.05); however, it did not restore the enzyme activity to control levels. Pralidoxime (fig. 4B) exhibited significant protection of AChE activity at both concentrations tested compared with the findings for the methamidophos group (p < 0.05); however, this protective effect did not allow the enzyme to reach control activity levels. Oxime 1 (fig. 4C) exhibited protective activity for AChE (p < 0.05), as AChE activity was maintained at control levels at both concentrations of oxime 1 in the presence of methamidophos. Oxime 2 (fig. 4D) exerted a significant (p < 0.05) protective effect on methamidophos-induced AChE inactivation at 100 μM, with control levels of enzyme activity being observed at this concentration.

Figure 4.

Protective effect of obidoxime (A), pralidoxime (B), oxime 1 (C) and oxime 2 (D) on methamidophos (MAP)-inhibited AChE activity from human erythrocyte ghost. Oximes were added to the reaction medium at the same time that methamidophos. The data are expressed as % from control group (100%). *Statistically different from control group (p < 0.05); #statistically different from methamidophos 25 μM group (p < 0.05). Results are presented as the mean ± S.E. of at least three independent experiments performed in duplicate.

Effect of oximes on methamidophos-inhibited plasma BChE

Fig. 5 shows the protective effects of oximes on methamidophos-inhibited BChE from human plasma. Obidoxime (fig. 5A), pralidoxime (fig. 5B) and oxime 2 (fig. 5D) exerted no protective effects on methamidophos-induced BChE inhibition. Oxime 1 (fig. 5C) was the only oxime that significantly (p < 0.05) protected against methamidophos-induced BChE inhibition relative to the activity of BChE observed in the methamidophos group.

Figure 5.

Protective effect of obidoxime (A), pralidoxime (B), oxime 1 (C) and oxime 2 (D) on methamidophos (MAP)-inhibited BChE activity from human plasma. Oximes were added to the reaction medium at the same time as methamidophos. The data are expressed as percentage from control group (100%). *Statistically different from control group (p < 0.05); #statistically different from methamidophos 25 μM group (p < 0.05). The results are presented as the mean ± S.E. of at least three independent experiments performed in duplicate.

Reactivation of methamidophos-inhibited cholinesterases by oximes

The results from the reactivation protocol were calculated as the percentage of reactivation, as summarised in table 1. For AChE from erythrocyte ghosts, obidoxime and pralidoxime had reactivation rates of 67.1% and 57.77%, respectively, at 50 μM, but this tendency was not statistically significant compared with the protective effects of the newly synthesised oximes 1 and 2 at the same concentration. For all tested oximes, better results were achieved at 50 μM than at 100 μM. For rat brain AChE, pralidoxime provided the best results regarding reactivation capacity, whereas obidoxime and the two new oximes exhibited low reactivation capacities compared with that of pralidoxime.

Table 1. Per cent reactivation of methamidophos-inhibited AChE by oximes in human erythrocyte ghost and rat brain AChE, and BChE from human plasma
 ObidoximePralidoximeOxime 1Oxime 2
50 μM100 μM50 μM100 μM50 μM100 μM50 μM100 μM
  1. Superscript letters indicate statistical difference between oximes at the same concentrations.

  2. Superscript numbers indicate statistical difference between the different enzymes.

  3. AChE, acetylcholinesterase; BChE, butyrylcholinesterase.

AChEery67.1 ± 6.53a,122.23 ± 9.69a,158.77 ± 30.77a,130.57 ± 11.65a,141.50 ± 18.13a,13.97 ± 3.3b,147.98 ± 8.59a,17.96 ± 3.03a,1
AChEbra3.55 ± 1.89a,20.95 ± 1.2a,252.74 ± 5.89b,166.05 ± 4.70b,215.71 ± 6.3a,25.31 ± 5.1a,16.36 ± 2.67a,25.89 ± 4.37a,1
BChE<0<0<0<044.98 ± 8.18a,1<05.71 ± 1.96b,228.27 ± 16.11a,1

Concerning BChE, the two currently available oximes, pralidoxime and obidoxime, had no reactivation effect on methamidophos-inhibited BChE, whereas oximes 1 (44.98%) and 2 (28.27%) exhibited significant reactivation activity at 50 and 100 μM, respectively.

The reactivation potencies, characterised by 1/KD (affinity) and kr (reactivity), were calculated for reactivation of methamidophos-inhibited AChE and BChE. In human erythrocyte AChE, the order of reactivity (kr) and affinity (KD) was pralidoxime > obidoxime > oxime 2 > oxime 1. The comparison of the second-order rate constants, kr2, demonstrated a superior potency of pralidoxime.

In human plasma BChE, no rate constants could be calculated for the classical oxime (obidoxime and pralidoxime), owing to the absence of reactivation of the inhibited enzyme. For the new oximes 1 and 2, the rate constant analysis shows a higher reactivation and affinity constants for oxime 1, which is reflected in the second-order rate constant (kr2).

Molecular docking results

Molecular docking studies were performed on all tested oximes to rationalise their possible interactions with the oximes. Pralidoxime which had the best reactivation capacity (table 1) presented binding energies of −5.83 and −5.3 kcal/mol for the Sp and Rp conformations of methamidophos, respectively. In the Rp conformation, the oxime group of pralidoxime is stabilised by hydrogen bonds with residues Arg296 and Phe295 in the acyl-binding pocket region (Trp236, Phe295, Phe297 and Phe338), and the aromatic ring is sandwiched among the peripheral anionic site (PAS) residues. For the Sp conformation, the oxime group of pralidoxime is stabilised by hydrogen bonds with residues Tyr133, Ala127, and Gly120, stabilising in the region between the oxyanion hole and the anionic subsite. The pyridine ring is sandwiched among the residues Trp86 (anionic subsite), Gly121 (oxyanion hole) and Ser203 (catalytic triad).

In the Rp conformation of methamidophos, obidoxime (fig. 6B) has a binding energy of −9.61 kcal/mol, and it is stabilised by hydrogen bonds with residues Tyr72, Asn87, and Arg296, being positioned between the acyl-binding pocket and the PAS. The linker between the pyridine rings is sandwiched among the residues Tyr124, Tyr337 and Tyr341 (PAS), forming a hydrogen bond with Tyr124. In the Sp conformation, the binding energy is −9.59 kcal/mol, and obidoxime exhibits a similar conformation to that of the Rp form of methamidophos, with the molecule being positioned in the PAS region among residues Asp74 and Ser125 and with hydrogen bonds formed with Tyr72, Asp74 and Ans87. The linker between the pyridine rings is stabilised among residues Tyr124, Tyr337 and Tyr341 (PAS).

Figure 6.

Docking of the conformation corresponding to the oximes: oxime 1 (A), oxime 2 (B), pralidoxime (C) and obidoxime (D) inside MmAChE active site. The left panel refers to the SGX conformation of the modified Ser203 by methamidophos, whereas the right panel refers to the SGR conformation.

The oxime moiety of the newly synthesised oxime 1 (fig. 6C) is stabilised by hydrogen bonds with residues Phe295, Arg296 and Phe338, being positioned between the PAS residues (Trp286 and Tyr341) and the acyl-binding pocket residues (Phe295, Phe297 and Phe338). The binding energy was −6.01 kcal/mol for both Rp and Sp enantiomers, and the conformation within the active site was similar. Oxime 2 (fig. 6D) exhibited a binding energy of −7.68 kcal/mol for both Rp and Sp enantiomers as well, and the oxime conformation was practically the same for both enantiomers. Overall, the molecule was stabilised by hydrogen bonds with residues Tyr72, Asp74, Asn87 and Tyr124, with the aromatic ring sandwiched among the PAS residues (Tyr337 and Tyr341), acyl-binding pocket (Phe295, Phe297 and Phe338), and the catalytic residue Ser203, and the oxime group is close to the PAS residue Asp74.

Discussion

The main objective of our study was to investigate methamidophos-induced toxicity and oxime-mediated protection/reactivation of methamidophos-inhibited cholinesterases. We compared the effects of two classical oximes (pralidoxime and obidoxime) and two newly synthesised oximes in protecting or reactivating methamidophos-inhibited cholinesterases. The two newly synthesised oximes had previously been found to be related with a protective effect on LDL oxidation [29] and have shown no signs of toxicity during in vivo or ex vivo use [30, 31]. In addition, these new oximes have been found to have LD50 values similar to other oximes, including pralidoxime and obidoxime [32]. It has also been previously demonstrated that the oximes 1 and 2 reactivate AChE inhibited by chlorpyrifos, diazinon and malathion with a potency similar to that of pralidoxime, but to a lesser extent than obidoxime [33]. The results found in this study showed that both new oximes could reactivate AChE, although less than the classical oximes, and, unlike the classical oximes, reactivate BChE.

Methamidophos has become an important public health issue, particularly in Brazil, largely because of its frequent use on crops, resulting in high rates of occupational [8] and food [7] exposure. In regard to oxime reactivation of methamidophos-inhibited AChE, Pohanka et al. [34] showed an in vitro reactivation of human AChE by obidoxime and pralidoxime; however, the experimental protocol used focused on the qualitative analysis of contaminated water, with no physiological conditions included in the analysis. Furthermore, Wan et al. [35] showed a positive effect of pralidoxime in individuals poisoned with methamidophos, and Satar et al. [36] demonstrated a hepatoprotective effect of pralidoxime against methamidophos-induced (30 mg/kg) toxicity in rats. Necrosis of the diaphragm induced by methamidophos was also found to be protected against by pralidoxime under in vivo conditions [37]. In our work, we also observed a protection/reactivation effect of oximes on methamidophos-inhibited AChE, using obidoxime and pralidoxime as positive controls for comparison with the new oximes.

There are numerous studies in the literature relating oximes and the reactivation of organophosphate-inhibited AChE, with a published review by Jokanovic and Stojiljkovic [2]. However, the understanding of the structure–activity relationships for oxime reactivation of cholinesterases remains limited [11]. The mechanism of action of oxime reactivation is based on the displacement of the phosphonyl group of the AChE-OP complex [2]. The reactivation potency of the oximes depends on the nucleophilicity and orientation of the oxime [38], making molecular studies vital for the comprehension of this phenomenon. Here, the two new oximes tested have only one aldoxime group, similar to pralidoxime, while obidoxime has two aldoxime groups. Kassa et al. [39] previously demonstrated that the number of aldoxime groups is not that critical to enzyme reactivation. However, Cabal et al. [40] showed that other characteristics of the oxime could interfere with its reactivation power, such as the number of pyridinium rings and the position of the oxime group within the pyridinium ring. In our case, none of the new oximes tested contained pyridinium rings and, to some extent, could reactivate both AChE and BChE inhibited by methamidophos (tables 2 and 3). Additionally, the compounds showed reactivation of AChE inhibited by chlorpyrifos, diazinon and malathion to a similar level as pralidoxime [33], indicating that oximes not based on pyridinium rings could be promising for further studies with modifications to their molecular structure.

Table 2. Reactivation constants
OximeReactivation constant
kr (/min)KD (μM)Kr2 (mM/min.)
  1. Human erythrocyte ghosts were incubated with methamidophos at 25 μM for 10 min. at 37°C. Inhibited acetylcholinesterase was reactivated by oximes (in six different concentrations, ranging from 0.01 to 1 mM) by adding oximes to inhibited ghosts. Small aliquots were transferred after various intervals to a cuvette for measurement of AChE activity.

Obidoxime0.013712.90.018
Pralidoxime0.014475.30.029
Oxime 10.0041008.20.004
Oxime 20.008873.30.009
Table 3. Reactivation constants
OximeReactivation constant
kr (min−1)KD (μM)Kr2 (mM/min.)
  1. Human plasma was incubated with methamidophos at 25 μM for 10 min. at 37°C. Inhibited acetylcholinesterase was reactivated by oximes (in six different concentrations, ranging from 0.01 to 1 mM) by adding oximes to inhibited ghosts. Small aliquots were transferred after various intervals to a cuvette for measurement of butyrylcholinesterase activity.

Obidoxime
Pralidoxime
Oxime 10.007763.90.009
Oxime 20.004968.30.004

Wong et al. [10] previously demonstrated that the oxyanion hole could be the subsite through which pralidoxime attacks the OP bound on the tabun-modified Ser203, while Musilek et al. [27] demonstrated that mono-bisquaternary AChE reactivators appear to be positioned between the PAS and the acyl-binding pocket. Our docking results strongly agree with these previous works, with pralidoxime being positioned in the oxyanion hole and obidoxime between the PAS and the acyl-binding pocket. In this way, pralidoxime seems to be better positioned into the active site for the reactivation process in the case of methamidophos-inhibited AChE. Surprisingly, the new oximes were not positioned as deeply into the active site as pralidoxime, but near the mouth of the gorge similar to obidoxime. This could explain the lesser reactivation potency of these oximes compared with pralidoxime and the fact that the new oximes protected AChE from inhibition to the same degree as the classical oximes, but with less potency for the reactivation of AChE. Furthermore, Wong et al. [10] also demonstrated that Sp enantiomers of cycloheptyl methylphosphonate-conjugated AChE are more likely to occur and are easily reactivated by oximes. Our molecular docking suggests the same behaviour for methamidophos-inhibited AChE, while pralidoxime had the best ‘pose’ in the Sp enantiomer (SGX in fig. 6).

Butyrylcholinesterase is able to catalyse the hydrolysis of acetylcholine to a lesser extent than AChE, thus regulating cholinergic neurotransmission. It has been suggested that BChE could act as a co-regulator for the action of acetylcholine (see Darvesh et al. [41]), and its inhibition can lead to a dose-dependent increase in the levels of acetylcholine in the brain [42]. Neuroanatomical data demonstrated that BChE is expressed by specific populations of central [43, 44] and peripheral [45] neurons, consistent with a co-regulatory role of BChE in cholinergic neurotransmission. In AChE-null mice, BChE was suggested to have a prominent role in the hydrolysis of acetylcholine, once inhibition of BChE shown to be lethal [46]. This situation is analogous to AChE inhibition by OP compounds and, as BChE is present at 10 times higher levels in the human body than AChE (about 680 nmol of BChE and 62 nmol of AChE) [47], the hydrolysis of acetylcholine by the BChE may represent a way to deal with the high levels of acetylcholine caused by AChE inhibition. In this way, our results show that the new oximes reactivated BChE, while the classical oximes failed. This fact could represent an advantage in the treatment of a cholinergic crisis, with a minor reactivation effect on AChE. Modifications to the oximes' molecular structure may increase AChE reactivation, without loss of the effect on BChE.

Conclusions

Our work demonstrates that the newly synthesised oximes are able to protect and reactivate human erythrocyte AChE, however, less efficiently than pralidoxime and obidoxime, and reactivate human plasma BChE, where the classical oximes failed. In addition, our work demonstrates that the attack through the oxyanion hole of the active site of AChE appears to be the best subsite for the nucleophilic attack of oximes on methamidophos-inhibited AChE, which could guide the development of new oximes.

Acknowledgements

This work was supported by the FINEP research grant ‘Rede Instituto Brasileiro de Neurociências (IBN-Net)’ # 01.06.0842-00. P.G. receives a fellowship from PIBIC/CNPq/UFSM. T.H L. receives a fellowship from CAPES. J.B.T.R, F.A.A.S., M.E.P., R.A.S. and N.B.V.B receive a fellowship by CNPq. Additional support was given by CAPES, CNPq, FAPERGS and INCT – National Institute of Science and Technology for Excitotoxicity and Neuroprotection/CNPq; and to Scripps Research Institute and Accelrys Inc. for providing free academic licence of programmes used in this study.

Conflict of Interest

The authors declare that there are no conflicts of interest.

Ancillary