Discovery of Novel Non-Oxime Reactivators Showing In Vivo Antidotal Efficiency for Sarin Poisoned Mice

A family of novel efficient non-oxime compounds exhibited promising reactivation efficacy for VX and sarin inhibited human acetylcholinesterase was discovered. It was found that aromatic groups coupled to Mannich phenols and the introduction of imidazole to the ortho position of phenols would dramatically enhance reactivation efficiency. Moreover, the in vivo experiment was conducted, and the results demonstrated that Mannich phenol L10R1 (30 mg/kg, ip) could afford 100% 48 h survival for mice of 2*LD50 sarin exposure, which is promising for the development of non-oxime reactivators with central efficiency.


Introduction
Organophosphates (OPs) including pesticides (e.g., paraoxon, parathion, phorate, dichlorvos and chlorophos, Figure 1) and nerve agents (e.g., sarin, VX, tabun and soman, Figure 1) are highly toxic compounds [1]. OPs potently inhibit the cholinergic acetylcholinesterase (AChE) through phosphorylation of the enzyme's catalytic serine residue, and render it incapable of hydrolyzing the neurotransmitter acetylcholine. This inhibition causes accumulation of acetylcholine (ACh), and leads to cholinergic crisis, respiratory distress, convulsive seizures and ultimately death [2]. Nerve agents have been used for war and terrorist attacks (e.g., subway attack in Tokyo in 1995) [3]. Organophosphorus pesticides poisoning is also a serious public health issue with about 3,000,000 acute intoxications and over 200,000 fatalities annually worldwide [4].

Introduction
Organophosphates (OPs) including pesticides (e.g., paraoxon, parathion, phorate, dichlorvos and chlorophos, Figure 1) and nerve agents (e.g., sarin, VX, tabun and soman, Figure 1) are highly toxic compounds [1]. OPs potently inhibit the cholinergic acetylcholinesterase (AChE) through phosphorylation of the enzyme's catalytic serine residue, and render it incapable of hydrolyzing the neurotransmitter acetylcholine. This inhibition causes accumulation of acetylcholine (ACh), and leads to cholinergic crisis, respiratory distress, convulsive seizures and ultimately death [2]. Nerve agents have been used for war and terrorist attacks (e.g., subway attack in Tokyo in 1995) [3]. Organophosphorus pesticides poisoning is also a serious public health issue with about 3,000,000 acute intoxications and over 200,000 fatalities annually worldwide [4].   A combination of intramuscular injections of an AChE reactivator of the pyridinium aldoxime family (e.g., pralidoxime (2-PAM), trimedoxime (TMB-4), obidoxime, HI-6, Figure 1) [5,6], a muscarinic receptor antagonist (e.g., atropine), and an anticonvulsant (e.g., diazepam) is approved antidotal therapy for the treatment of OP poisoning in humans currently [7,8]. Pyridinium aldoximes are very potent nucleophiles that can break the strong phosphorus oxygen bond of OP-AChE and restore the enzyme's activity [9]. However, due to their permanent positive charge, these quaternary reactivators were poorly distributed in the central nervous system (CNS) [10], while the brain was a major target of nerve agents [11]. Consequently, various nonquaternary AChE reactivators were designed and synthesized, such as monoisonitrosoacetone [12,13] and amidine-oximes [14,15], but they were even less efficient than 2-PAM in vitro. In recent years, a series of pyridyl aldoxime conjugates were reported as superior reactivators for OP poisoning in comparison to HI-6 and obidoxime in vitro [16][17][18][19][20][21], but none of these pyridyl aldoxime conjugates has ever been reported showing in vivo antidotal efficiency. Previously, we had also reported a number of salicylic aldoxime conjugates as efficient nonquaternary reactivators [22][23][24], but they were proven as poor reactivators for sarin poisoned mice in vivo.
It seems that research of nonquaternary oxime reactivators encountered a bottleneck presently. Nevertheless, it was gratifying that two non-oxime intermediates (L6R1 and L10R1, Figure 2) in our previous study were found, showing reactivating ability to sarin and VX inhibited hAChE (Figure 3), which represented a totally different reactivator scaffold to the traditional oximes. Moreover, Francine et al. had discovered some similar Mannich phenols (such as ADQ and ADCQ, Figure 2) exhibiting reactivation efficacy for paraoxon or DFP inhibited AChE recently [25,26]; they further found that a series of nonoxime compounds containing imidazole moiety (SP134 and SP138, Figure 2) displayed reactivating ability [27]. In addition, de Koning et al. found that an imidazole derivative (3q, Figure 2) of ADOC exhibited reactivating efficacy for OP poisoned AChE at high concentration (1 mM) [28]. Intriguingly, a series of efficient imidazolium aldoxime reactivators was studied in our previous research [24]. These findings inspired us that maybe imidazole moiety play an important role in the reactivating process of OP inhibited AChE. As a preliminary structure modification of L6R1 and L10R1, imidazole was introduced to replace the diethylamine moiety in the Mannich phenols. It is gratifying that the resulting compounds (L6R4 and L10R4, Figure 2) exhibited higher in vitro reactivating ability for both sarin and VX inhibited hAChE ( Figure 3). Furthermore, a preliminary in vivo experiment disclosed that L10R1 could afford complete protection for sarin poisoned mice. To the best of our knowledge, L10R1 was the first reported non-oxime reactivator showing in vivo antidotal efficiency for sarin poisoning.

Synthesis
The synthetic routes to prepare these novel non-oxime compounds were outlined in Scheme 1. Firstly, R1 was obtained through a Mannich reaction by using 4-hydroxy-benzaldehyde 1 and paraformaldehyde in isopropanol. It then underwent a reductive amination reaction with L6 or L10 by using Hantzsch ester diludine and iodine to give L6R1 or L10R1 in the mixed solvents of dichloromethane and methanol. Synthesis of L6R4 was commenced with chloromethylation of 4-hydroxybenzaldehyde 1 to give the intermediate 2; then, condensation of 2 and imidazole in acetonitrile provided R4. Finally, reductive amination between R4 and L6 afforded compound L6R4. L10R4 was obtained by using a similar reductive amination reaction by using L10 and R4.

Synthesis
The synthetic routes to prepare these novel non-oxime compounds were outlined in Scheme 1. Firstly, R1 was obtained through a Mannich reaction by using 4-hydroxybenzaldehyde 1 and paraformaldehyde in isopropanol. It then underwent a reductive amination reaction with L6 or L10 by using Hantzsch ester diludine and iodine to give L6R1 or L10R1 in the mixed solvents of dichloromethane and methanol. Synthesis of L6R4 was commenced with chloromethylation of 4-hydroxybenzaldehyde 1 to give the intermediate 2; then, condensation of 2 and imidazole in acetonitrile provided R4. Finally, reductive amination between R4 and L6 afforded compound L6R4. L10R4 was obtained by using a similar reductive amination reaction by using L10 and R4.

Synthesis
The synthetic routes to prepare these novel non-oxime compounds were outlined in Scheme 1. Firstly, R1 was obtained through a Mannich reaction by using 4-hydroxy-benzaldehyde 1 and paraformaldehyde in isopropanol. It then underwent a reductive amination reaction with L6 or L10 by using Hantzsch ester diludine and iodine to give L6R1 or L10R1 in the mixed solvents of dichloromethane and methanol. Synthesis of L6R4 was commenced with chloromethylation of 4-hydroxybenzaldehyde 1 to give the intermediate 2; then, condensation of 2 and imidazole in acetonitrile provided R4. Finally, reductive amination between R4 and L6 afforded compound L6R4. L10R4 was obtained by using a similar reductive amination reaction by using L10 and R4.

In Vitro Inhibition and Reactivation Experiments
The in vitro experiments were conducted with human acetylcholinesterase (hAChE) serving as an enzyme source. Two most common nerve agents (VX and sarin) were used for the in vitro reactivation experiment. The enzyme activity was measured using a similar method of Ellman et al. [29]. Firstly, the inhibition experiment was necessary for these novel compounds because strong inhibition of hAChE was unfavorable to the reactivating process of OP poisoned enzyme. The in vitro inhibition experiment demonstrated that L6R1, L6R4, and L10R4 were weak inhibitors of hAChE with IC 50 greater than 400 µM, while L10R1 was a moderate inhibitor (with IC 50 lower than 40 µM, Table 1). The in vitro reactivation experiment showed that both L6R4 and L10R4 exhibited superior reactivation efficacy to L6R1 and L10R1 for sarin and VX inhibited hAChE, and they even exceeded HI-6 for VX inhibited hAChE at high concentrations ( Figure 3). Due to its inhibition potency towards hAChE, L10R1 did not exhibit reactivating ability at high concentrations, but it was confirmed that L10R1 was able to reactivate poisoned hAChE at low concentrations ( Figure 3). Table 1. Reactivation rate constant (k r ), dissociation constant (K D ), second order reactivation rate constant (k r2 ) of HI-6 and novel non-oxime reactivators for nerve agent inhibited hAChE. IC 50 and predicted pKa of these compounds.

Determination of Reactivation Kinetics
Determination of maximal reactivation rate constant k r , dissociation constant K D and second order reactivation rate constant k r2 (k r2 = k r /K D ) would help obtain a deeper comprehension of the reactivating ability. Results of the reactivation kinetics constants were reported in Table 1. For L6R1, introduction of imidazole not only increased reactivation rate constant k r but also enhanced binding affinity towards inhibited hAChE (indicated by lower dissociation constant K D ), which resulted in dramatically improved reactivation efficacy of L6R4 in contrast to L6R1, especially in the case of VX poisoning. For L10R4, introduction of imidazole greatly decreased the inhibition ability of hAChE; it was interesting that its inhibition potency to hAChE decreased at the same time. Although its reactivation rate constant k r was increased, the decreased binding affinity (higher dissociation constant K D ) made L10R4 a less efficient reactivator. Due to its greatly enhanced binding affinity, L10R1 was even 2-fold more efficient than L6R4 and near 3-fold more efficient than L10R4 for sarin poisoned hAChE. Hence, L10R1 was confirmed as an efficient reactivator for inhibited hAChE, but we cannot come to the conclusion that stronger inhibitor would bind tightly to poisoned hAChE because, although L6R1, L6R4, and L10R4 were almost equal inhibitors of hAChE, their binding affinity for poisoned hAChE varied greatly.
It could be concluded that the introduction of imidazole normally increased the reactivation rate constant, which was indicated by higher reactivating ability at relatively high concentrations, but it had the opposite influence on binding affinity for L10R4. The pKa of these compounds were predicted (Table 1), and it seems that L6R1 and L10R1 were a stronger base than L6R4 and L10R4, so they might be easier to protonate under physiological pH, which might increase their binding affinity towards poisoned hAChE, but L10R4 actually did not show binding efficiency. However, it was noteworthy that both L10R1 and L10R4 exhibited higher binding affinity to the inhibited hAChE than L6R1 and L6R4 (see values of K D in Table 1), especially in the case of L10R1 vs. L6R1. Hence, L10 seemed to be a better ligand than L6 for the construction of more efficient AChE reactivators.

Molecular Docking Simulation
Additionally, we tried to explain the reactivation mechanism of these non-oximes through a molecular docking simulation study (Figure 4). The potential binding pocket was explored according to the ligand from the crystal structure of VX inhibited hAChE in complex with HI-6 (PDB code: 6CQW, resolution 2.28 Å) [30]. In the case of L6R1, the benzamide ring (L6) was located at the peripheral anionic site of hAChE, which was fixed via strong hydrophobic and π-π stacking interactions with Trp286 and Tyr72, and it was further stabilized by a H-bond interaction with Val282 and a water molecular, while the Mannich phenol moiety was submerged in the active gorge. In the case of L6R4, its ligand L6 interacted with Trp286 and Tyr 72 in a similar way as L6R1, but an H-bond was formed with Glu285 at the peripheral anionic site of hAChE; the molecular was further stabilized by an H-bond between the imidazole moiety and Trp286, which may account for high binding affinity of L6R4 to VX inhibited hAChE. Ligand L10 of L10R1 interact with Trp286 and Tyr 72 through a π-π sandwiching way at the peripheral anionic site, while only weak hydrophobic interactions between L10 and Tyr 72 existed for L10R4, but an additional Hbond between the secondary amine of L10 and Asp74 was observed during the simulation. The phenolic hydroxyl group of both L10R1 and L10R4 formed an H-bond with Phe338 at the active site of hAChE, while L10R4 was further stabilized by an H-bond interaction between its imidazole moiety and Val282. However, we noticed that the imidazole moiety of L6R4 and L10R4 was far from His447, hence we thought that the imidazole moiety might not act as an internal base as histidine mimic in the OP-inhibited triad to promote the reactivation process [25].  The key amino acid residues in the peripheral anionic site include Tyr72, Tyr124, Trp286, and Glu285; and in the active site include Ser203, Glu334, Tyr337, and His447. The docked conformations of the reactivators are depicted as a bond stick model and the key amino acid residues as a slim stick model.

In Vivo Biological Experiments
Furthermore, a preliminary in vivo experiment was conducted to test the protection of these new reactivators to mice of sarin exposure. The better reactivators L10R1, L6R4, and L10R4 were selected. In the animal paradigm used herein, mice were observed for neurological toxicity symptoms such as muscles twitching, seizures, and convulsions af- For all four of the compounds, we noticed that their nucleophilic phenolic hydroxyl group did not orient to the inhibited Ser203 at the active site of hAChE, so it seemed that these non-oximes did not reactivate inhibited AChE through a nucleophilic process as traditional oximes [9]. Given the above analysis, we speculated that interaction between these non-oximes and OP-inhibited AChE would induce conformation changes of the enzyme, which would help to restart the spontaneous reactivation process in the active gorge. However, the proposed reactivating mechanism was sketchy, and it should be supported by the experimental data.

In Vivo Biological Experiments
Furthermore, a preliminary in vivo experiment was conducted to test the protection of these new reactivators to mice of sarin exposure. The better reactivators L10R1, L6R4, and L10R4 were selected. In the animal paradigm used herein, mice were observed for neurological toxicity symptoms such as muscles twitching, seizures, and convulsions after sarin or antidotes administration, and 48 h survival was finally recorded. Firstly, animals were pretreated with L10R1, L6R4, or L10R4 at a high dose of 60 mg/kg (ip) to evaluate possible acute toxicity ( Table 2, experiment 1). In parallel, two sets of mice pretreated with isotonic saline alone were challenged with 2*LD 50 dose of sarin (85 µg/Kg), and one set of the mice was treated with atropine sulphate 1 min later (0.5 mg/Kg, control 2 of experiment 1 in Table 2). The results demonstrated that no apparent toxicity was observed for these new reactivators administered alone at a relatively high dose of 60 mg/kg (ip, >180 µM/Kg). For sarin (2*LD 50 dose) poisoned animals, significant CNS poisoning symptoms were observed (such as muscles twitching, strong seizures, and convulsions) and no saline treated mice (control 1 in experiment 1) survived to the 36 h time point, while only 1/8 of the atropine treated mice (control 2 in experiment 1) survived to the 48 h time point.  Next, in order to maximize these non-oximes' antidotal ability in a preliminary in vivo experiment, mice were pretreated with different antidotes (including 2-PAM, HI-6, L10R1, L6R4, and L10R4) 15 min before the administration of 2*LD 50 dose of sarin, and treated with atropine sulfate 1 min later (0.5 mg/Kg). The results demonstrated that the best in vitro non-oxime reactivator L10R1 exhibited the highest antidotal efficacy in vivo, along with HI-6 providing complete protection for sarin poisoned mice. More importantly, only slight CNS poisoning symptoms (such as muscles twitching, lack of spontaneous activity, and decreased interest in food consumption) were observed for L10R1 treated animals, while some mice pretreated with HI-6 were observed with slight seizures and convulsions. Although the in vitro reactivation efficiency of L10R1 was inferior to that of HI-6, the predicted LogBB and LogP of L10R1 were much higher than that of HI-6 ( Table 2), which might help L10R1 to provide higher CNS protection than quaternary HI-6, while HI-6 provided higher peripheral protection and resulted in high survival. However, real data (e.g., BBB penetration ability, blood/brain cholinesterase activities) were needed to support these estimations, and our research group would conduct these experiments in the near future. In contrast, 2-PAM protected only 2/10 mice, the imidazole bearing reactivators L6R4 and L10R4 protected only 3/10 and 2/10 mice separately, along with heavy poisoning symptoms such as strong seizures and convulsions, which might be due to their lower reactivation efficiency (k r2 , Table 1) for sarin inhibited hAChE and their relatively lower LogBB and LogP (predicted values, Table 2).
Nonetheless, it was noteworthy that L10R1 exhibited moderate irreversible inhibition ability to hAChE and mice were pretreated with L10R1, which meant that L10R1 might serve as a protecting agent for OP poisoning in a similar way to pyridostigmine at the same time [31,32]. Given the fact that the weak inhibitors (L6R4 and L10R4, Table 1) of hAChE could not provide protection efficiency for sarin poisoned mice, further experiments need to be conducted to understand the complete antidotal mechanism of L10R1.

Chemicals
All reagents and solvents were used as received from commercial sources. 1 H NMR and 13 C NMR spectra were recorded at 400 MHz and 100 MHz on a Bruker-400 instrument in CDCl 3 or DMSO-d 6 , respectively. Proton and carbon chemical shifts are expressed in parts per million (ppm) relative to internal tetramethylsilane (TMS), and coupling constants (J) are expressed in Hertz (Hz).

Computational Methods
Molecular docking simulations were conducted by using the "SYBYL-X 2.0" software. The potential binding pocket was explored according to the ligand from the crystal structure of VX inhibited hAChE in complex with HI-6 (PDB code: 6CQW, resolution 2.28 Å) and from the crystal structure of hAChE in complex with HI-6 (PDB code: 6CQU, resolution 2.308 Å) [7]. The main protocols and the parameters set for the docking were as follows:

General In Vitro AChE Screening Information
Human acetylcholinesterase (hAChE, 20 U/mL, dissolved in 20 mM HEPES, pH 8.0), bovine serum albumin (BSA), acetylthiocholine (ATCh), and 5, 5-dithiobis-2-nitrobenzoic acid (DTNB) were purchased from Sigma-Aldrich. HI-6 and obidoxime were synthesized according to the literature protocols [33,34]. Sarin and VX were from the Anti Chemical Command and Engineering Institute of the Chinese People's Liberation Army. (Caution! OPs used in our research are highly toxic and must be handled with extreme care by well-trained personnel. Use of these materials has been approved by the Anti Chemical Command and Engineering Institute of the Chinese People's Liberation Army. After reactivation studies, biochemical samples were neutralized by stirring with 2 M NaOH for 12 h. and the remaining solutions were brought back to pH~7 and disposed in chemical waste.) In addition, 10 mM concentration solutions of the final compounds were prepared in water containing 20% methanol. They were further diluted by PBS (0.1 M, pH 7.4) to 3 mM concentrations. It was found that there was no effect of methanol on hAChE by a control experiment. All the biological evaluation experiments were conducted in a 96-well plate; the enzyme activity was measured by the time-dependent hydrolysis of ATCh in which the product (thiocholine) was detected by reaction with the Ellman's reagent, 5, 5'-dithiodis-2-nitrobenzoic acid (DTNB), and absorbance at 412 nm [29].

Procedures of hAChE Inhibition Experiments
The procedures of inhibition experiments were as follows:

Procedures of Reactivation Experiments
The procedures of reactivation experiments were as follows: (1) A stock solution of hAChE was diluted 2000-fold with PBS (0.1 M, pH = 7.4, 0.1% BSA); the concentrations of different nerve agents were determined by a pre-experiment similar to the inhibition experiment to attain an inhibition plateau from 90% to 95%.
We tried carefully to control the dosage of OP used to avoid 100% inhibition of hAChE, which meant that all OP used had bound to the enzyme, and there was no OP presented in the reaction mixture. The final concentrations of OPs were as follows: VX, 3*10 7 fold diluted; sarin, 1.6*10 6 fold diluted.

Determination of Reactivation Kinetics
To further investigate the reactivating mechanism, the %reactivation at different time intervals and at different concentrations were measured by using the same method we described in Section 3.6. The observed first-order rate constant k obs for each reactivator concentration, the dissociation constant K D of inhibited enzyme-reactivator complex (EP-R), and the reactivation rate constant k r were calculated by nonlinear fitting using the standard reactivator concentration dependent reactivation equation derived from the following scheme [35,36]:

Determination of Reactivation Kinetics
To further investigate the reactivating mechanism, the %reactivation at different time intervals and at different concentrations were measured by using the same method we described in Section 3.6. The observed first-order rate constant kobs for each reactivator concentration, the dissociation constant KD of inhibited enzyme-reactivator complex (EP-R), and the reactivation rate constant kr were calculated by nonlinear fitting using the standard reactivator concentration dependent reactivation equation derived from the following scheme [35,36]: In this scheme, EP is the phosphylated enzyme, [EP-R] is the reversible Michaelis-type complex between EP and the reactivators [R], E is the active enzyme and P-R the phosphylated reactivator or hydrolysised OPs. KD is equal to the ratio (k − 1 + kr)/k1, and it typically approximates the dissociation constant of the [EP-R] complex, where it follows that: kr2 = kr/KD.
For more details, we conducted the reactivation experiment at different reactivator concentrations with discontinuous determination of enzyme activity at different reactivation times. Basically, the concentration of the reactivated AChE is proportional to the enzyme activity, kobs was calculated from the continuous recording of d[S]/dt, and the velocity of substrate hydrolysis (v) may be expressed as a pseudo-first-order process of reactivation Equation (1): in which vt represents velocity at time t, v0 represents maximum velocity (normal control), and vi represents minimum velocity (poisoned control). Alternatively, for each reactivator concentration, the kobs value was determined by linear regression analysis applying Equation (2): Integration of (2) results in Equation (3): which was used for nonlinear regression analysis of the data points from individual reactivator concentrations.
Concentrations of the reactivators used to determine the concentration dependence of the apparent reactivation rate kobs for the reactivation of OPs inhibited hAChE were In this scheme, EP is the phosphylated enzyme, [EP-R] is the reversible Michaelistype complex between EP and the reactivators [R], E is the active enzyme and P-R the phosphylated reactivator or hydrolysised OPs. K D is equal to the ratio (k −1 + k r )/k 1 , and it typically approximates the dissociation constant of the [EP-R] complex, where it follows that: k r2 = k r /K D .
For more details, we conducted the reactivation experiment at different reactivator concentrations with discontinuous determination of enzyme activity at different reactivation times. Basically, the concentration of the reactivated AChE is proportional to the enzyme activity, k obs was calculated from the continuous recording of d[S]/dt, and the velocity of substrate hydrolysis (v) may be expressed as a pseudo-first-order process of reactivation Equation (1): in which v t represents velocity at time t, v 0 represents maximum velocity (normal control), and v i represents minimum velocity (poisoned control). Alternatively, for each reactivator concentration, the k obs value was determined by linear regression analysis applying Equation (2): Integration of (2) results in Equation (3): which was used for nonlinear regression analysis of the data points from individual reactivator concentrations.
Concentrations of the reactivators used to determine the concentration dependence of the apparent reactivation rate k obs for the reactivation of OPs inhibited hAChE were shown in Table 3, plots of k obs vs. concentrations of HI-6 and the new synthesized compounds were shown in supporting information [37]. Due to different effect-acting concentrations for different reactivators, different concentration scales were used for HI-6 and L10R4 or other phenols. Table 3. Reactivator concentrations (µM) used for determination of the observed first-order rate constant k obs .

Details of the In Vivo Reactivation Experiments
In order to maximize the antidotal ability of these new reactivators, an atropine dose as low as possible should be used. For the in vivo experiment, if only atropine was used for sarin (2*LD 50 dose) poisoned animals, we found that a dose of 0.5 mg/Kg would result in 1/8 of mice (control 2 in experiment 1) surviving to the 48 h time point. The procedure of in vivo protection experiments for sarin exposure was as follows: For experiment 1: 1. Animals were pretreated with L10R1, L6R4, and L10R4 at a dose of 60 mg/kg (ip); in parallel, two sets of mice pretreated with isotonic saline alone were challenged with 2*LD 50 dose of sarin (85 µg/Kg), and one set of the mice was treated 1 min later with atropine sulfate (0.5 mg/Kg, control 2 in Table 2).

2.
Mice were observed for neurological toxicity symptoms such as muscles twitching, seizures, and convulsions after sarin or antidote administration, and the 48 h survival was finally recorded.

3.
Mice were observed for neurological toxicity symptoms such as muscles twitching, seizures, and convulsions after sarin or antidote administration, and the 48 h survival was finally recorded.

Conclusions
In conclusion, a family of novel non-oxime compounds displayed promising reactivation efficacy for VX and sarin inhibited hAChE were discovered in this paper. L6R4, L10R1, and L10R4 were proven as efficient reactivators for sarin and VX inhibited hAChE in vitro. Aromatic groups coupled to Mannich phenol seemed to be key structures for construction of efficient reactivators, while the introduction of imidazole to the ortho position of phenols would promote the reactivating ability at high concentrations, but it decreased binding affinity towards the poisoned hAChE and the resulting L6R4 did not exhibit superior reactivating ability to L10R1 at low concentrations. Moreover, due to its improved in vitro reactivating efficiency and lipophilicity, L10R1 emerged as a potential and efficient antidote which afforded complete 48 h protection in an animal survival experiment of 2*LD 50 dose sarin exposure. Different from the traditional quaternary oxime reactivators, a totally novel nonquaternary non-oxime structural scaffold was explored, and exciting reactivation results were obtained in this study. These findings provided a completely new starting point for the development of improved reactivators with centrally antidotal efficiency.
Supplementary Materials: The following are available online: The NMR and HR-MS Spectra for R4, L6R1, L10R1, L6R4, and L10R4; Table S1. %Reactivation of non-oxime reactivators for VX and sarin inhibited hAChE; Table S2. Selected observed first-order rate constant Kobs of HI-6 and novel non-oxime reactivators; Plot of k obs vs. concentrations of HI-6 and new synthesized compounds.

Conflicts of Interest:
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.