Introduction

The fatal effect of poisoning by organophosphorus compounds (OPCs) like pesticides or chemical warfare agents is due to inhibition of the active site of the enzyme acetyl-cholinestrase (EC 3.1.1.7). Today it is generally accepted (Volans 1996; Lee 2003) that the treatment of OPC-poisoning is based on the immediate and simultaneous administration of an antimuscarinic (e.g. atropine), a reactivator (pyridinium oximes like 2-PAM or obidoxime), and an anticonvulsant (e.g. diazepam). In a biological environment (aqueous solution, pH 7.4) pyridinium oximes form a zwitterion (betaine) with a high electron density at the O-atome of the oximate group. This enables the oximate to react as a nucleophilic agent and to displace the phosphyl (phosphonyl or posphoryl) residue from the enzyme. Thereby a phosphyloxime (POX) intermediate is generated. Furthermore, oximes are able to react spontaneously with a free OPC also forming a POX. This more or less stable intermediate is decomposed spontaneously into less toxic reaction products.

After the first description of the direct reaction of different PAM isomers with sarin by Green and Saville 1956, various attempts have been made to characterize and to quantify the reaction. However, the studies are hardly to compare, since the authors did not only use different methods, OPCs, and oximes, but also different buffers, pH values, temperatures, and reactant concentrations (Hackley et al. 1959; Lamb et al. 1965; Okonek 1972; Nenner 1974; Harvey et al. 1986; Waser et al. 1992). Therefore, experiments were carried out to study the interaction of OPCs and oximes under identical conditions. In a previous communication in this journal we reported upon the direct reaction of seven pyridinium oximes (2-PAM, obidoxime, HS 6, HGA 1, HGG 12, HI 6, and HLö 7) with the OPCs sarin, soman, and tabun (Becker et al. 1997). This work was continued and is presented here giving the results of the interaction studies of the same seven pyridinium oximes with the OPCs cyclosarin, crotylsarin, and VX. Taken together, experiments with 42 different combinations of oximes and OPCs were carried out and described. Since these studies were performed under identical conditions (spectrophotometry, 0.1 M MOPS buffer, pH 7.4 at 25°C, oxime 15 μM, nerve agent 1.5 mM), the results are comparable.

The aim of the work was to elucidate whether the direct reaction could contribute to the therapeutic effect of oxime treatment after OPC-poisoning. Furthermore, there was the question whether the oximes could serve as non-corrosive decontaminating compounds, especially for sensitive biological surfaces like skin, mucous membranes, or wounds.

Materials and methods

Chemicals

Crotylmethylphosphonofluoridate (crotylsarin, 97% pure), cyclohexylmethylphosphonofluoridate (cyclosarin, > 98% pure), and O-Ethyl S-2-(diisopropylamino)ethyl methylphosphonothiolate (VX, 98% pure) were made available by the German Ministry of Defence (purity checked by gas chromatography). Morpholinopropanesulfonic acid (MOPS) and pyridinium-2-aldoxime methiodide (2-PAM iodide or pralidoxime iodide) were obtained from Sigma (Deisenhofen, Germany).

HGG 12 (pyridinium, 1-[[3-(benzoyl-pyridinio)methoxy]methyl]-2-[(hydroxyimino)methyl]dichloride or dinitrate) and obidoxime (1,1′-[oxy-bis(methylene)]bis[4-(hydroxyimino)methyl]pyridinium dichloride) were obtained from Merck (Darmstadt, Germany). HI 6 (pyridinium, 1-[[[4-(aminocarbonyl)pyridinio]methoxy]methyl]-2-[(hydroxyimino)methyl]dichloride) was a gift of Dr. J. G. Clement (DRES, Ralston, Alberta, Canada). HLö 7 (pyridinium, 1-[[[4-(aminocarbonyl)pyridinio]methoxy]methyl]-2,4-bis[(hydroxyimino)methyl]dimethanesulfonate) was obtained from J. Braxmeier (Chemisches Laboratorium, Döpshofen, Germany). HGA 1 (3-(N,N-dimethylamino-carbonyloxy)-2-(hydroxyiminomethyl)-1-methylpyridinium iodide) was synthesized by Dr. H. Eckert (Institut für Organische Chemie I, TU München, Germany), and HS 6 (pyridinium, 1-[[[3-(aminocarbonyl)pyridinio]methoxy]methyl]-2-[(hydroxyimino)methyl]dichloride) was obtained from Dr. H. Oldiges (Fraunhofer-Institut für Umweltchemie und Ökotoxikologie, Schmallenberg, Germany).

Solutions

Stock solutions of the oximes (30 μM) and OPCs (3 mM crotylsarin, cyclosarin, or VX) were freshly prepared in 0.1 M MOPS buffer, pH 7.4 at 25°C immediately before use. To facilitate the dispersion and solution of VX in the buffer, the preparation was shaken for 2 min by means of a Vortex mixer (Vortex Genie 2, step 6).

UV spectra

The oxime spectra (250–430 nm) were recorded using a 1-cm quartz cuvette and a UV-VIS Beckman type DU 70 instrument with temperature control. The reference spectrum of 15 μM oxime (in the absence of OPC) was determined in 0.1 M MOPS buffer, pH 7.4 at 25°C. Immediately after mixing of 1.5 ml of oxime solution with 1.5 ml of the OPC stock solution in the quartz cuvette, the reaction spectra, i.e. the time dependent changes of the oxime concentrations were followed at 25°C after various reaction time intervals (0.25–540 min). At the start of the reaction the final concentrations of each of the oximes and OPCs were 15 μM and 1.5 mM, respectively, i.e. the oxime to OPC ratio was 1:100. In the spectral region of interest neither MOPS buffer nor the OPCs showed any absorbance. The peak absorbance in the region of interest (250 nm–400 nM), representing the zwitterionic (betaine) form of the oxime (Schoene 1973; Waser et al. 1992), was used for quantification and calculation of the decomposition half-life times t 1/2, the first-order rate constants k 1, the second-order rate constants k 2, and the maximum reaction velocities v max at the reactant concentrations indicated.

Calculations

The semilogarithmic plot of oxime concentrations versus time showed a straight line indicating a monomolecular reaction (pseudo-first-order kinetics). After the graphic determination of the half-life time t 1/2 the first-order rate constant k 1 is available, since at that point of time the initial oxime concentration is reduced to ½:

  • k 1 × t 1/2 = −ln ½ = −2.303 log ½

  • t 1/2 = −1/k 1 × 2.303 × (−0.301) = 0.693/k 1 (min)

  • for practical use: k 1 = 0.693/t 1/2 (min−1)

k 1 leads to the second-order rate constant k 2:

  • k 2 = k 1/(initial OPC concentration) = k 1/1.5 × 10−3 (M−1 min−1)

By means of k 2 the maximum reaction velocity can be calculated:

  • vmax = k 2 × (oxime concentration) × (OPC concentration) (M min−1)

  • v max = k 2 × (15 × 10−6) × (1.5 × 10−3) (M min−1)

  • for practical use: v max = k 2 × 22.5 × 10−9 (M min−1)

All kinetic parameters summarized in Table 1 were calculated this way.

Table 1 Decomposition half-lifes (t 1/2), rate constants (k 1, k 2), and maximum reaction velocities (v max) of different oximes (15 μM) in the presence of 1.5 mM of crotylsarin, cyclosarin, or VX in 0.1 M MOPS buffer, pH 7.4 at 25°C
Full size table

Results

The direct reaction of the oxime HI 6 and crotylsarin is shown in Fig. 1. The spectrum of HI 6 (reference in the absence of crotylsarin) is characterized by three maxima at 273, 300, and 354 nm, respectively. After addition of an 100-fold excess of crotylsarin, a marked but transient increase of the peak at 273 nm over a period of 10 min could be observed, whereas the two other maxima decreased. The half-time of HI 6 concentration decrease was 3.7 min according the diminution of the zwitterion peak at 354 nm.

Fig. 1
figure 1

Time related spectral changes of a mixture of HI 6 (15 μM) and crotylsarin (1.5 mM) in 0.1 M MOPS buffer, pH 7.4 at 25°C. Time of recording relative to the reaction start by mixing of the components (HI 6, 30 μM + crotylsarin, 3 mM, in 0.1 M MOPS buffer, 1.5 ml each) given in minutes. For the full chemical name of the oxime see Materials and methods

Full size image

Figure 2 shows the direct reaction of HI 6 and VX. It is obvious that the change of the HI 6 spectrum was different, compared to the experiments with crotylsarin. The maxima at 300 and 354 nm decreased similar to the experiments with crotylsarin, whereas the peak at 273 nm was transformed into a shoulder. On the basis of the decreasing absorbances at 354 nm the half-life time of HI 6 decrease was determined to be 16.4 min.

Fig. 2
figure 2

Time related spectral changes of a mixture of HI 6 (15 μM) and VX (1.5 mM) in 0.1 M MOPS buffer, pH 7.4 at 25°C. Time of recording relative to the reaction start by mixing of the components (HI 6, 30 μM + VX, 3 mM, in 0.1 M MOPS buffer, 1.5 ml each) given in minutes. For the full chemical name of the oxime see Materials and methods

Full size image

The spectrum of 2-PAM (reference in the absence of cyclosarin) displays a peak at 294 nm and a shoulder (zwitterion) at 330 nm (Fig. 3). After addition of an 100-fold excess of cyclosarin the maximum and the shoulder decreased with increasing reaction time. The peak at 294 nm showed a hypsochromic shift during the incubation time, while the shoulder was stationary. The half-life time of 2-PAM was 7.5 min (294 nm) or 6.0 min (326 nm), respectively.

Fig. 3
figure 3

Time related spectral changes of a mixture of 2-PAM (15 μM) and cyclosarin (1.5 mM) in 0.1 M MOPS buffer, pH 7.4 at 25°C. Time of recording relative to the reaction start by mixing of the components (2-PAM, 30 μM + cyclosarin, 3 mM, in 0.1 M MOPS buffer, 1.5 ml each) given in minutes. For the full chemical name of the oxime see Materials and methods

Full size image

The experimental results and the calculated kinetic parameters are summarized in Table 1. As the dimension (min−1) of k 1 in Table 1 shows, this value characterizes the decay of a substance (OPC) per minute. The first-order rate constant k 1 is independent of the concentration. The multiplication of k 1 with 100 gives the percentage of decay per minute; for example, the first line in Table 1 shows a k 1 value of 0.1307 min−1, that means a disappearance per minute of 13.07% of crotylsarin. In contrast to k 1 the second-order rate constant k 2 is dependent on the concentration. The knowledge of k 2 is necessary for the calculation of the maximum reaction velocity v max. This parameter cannot be exceeded and describes the maximum quantity of a substance (moles per litre) that react per minute.

It can be taken from Table 1 that the reaction velocities increased in the order VX < cyclosarin < crotylsarin. The reaction of VX and HGA 1 was too slow (t 1/2 > 540 min) to calculate a half-time. Even the doubling of the VX concentration from 1.5 to 3.0 mM did not increase the reaction velocity. The same phenomenon was to be observed (t 1/2 > 540 min) with the reaction of VX and HLö 7 (354 nm; Table 1).

Discussion

The results indicate that the oximes investigated react spontaneously with the three OPCs. The reaction velocities, however, were markedly different. The sarin analogues showed high reaction velocities with half-times of the oximes between 3.5 and 9.1 min (crotylsarin) and between 5.8 and 12.2 min (cyclosarin). In contrast, the decrease of the oxime concentration was much slower in the presence of VX (half-times between 16.4 and 151 min). In two cases, HGA 1 at 340 nm and HLö 7 at 354 nm, the direct reaction with VX was too slow to calculate half-lifes of the oximes (t 1/2 > 540 min). The explanation might be that the leaving group of the sarin analogues is the small fluoride ion (MW ca. 19 Da) whereas the leaving group of VX (the anion of diisopropylaminothioethanol) shows more than the eightfold mass (MW ca. 160 Da). Other reasons for the different behaviour of the sarin analogues and VX might be different water solubility, different electron density at the P-atome, and steric factors.

The direct reaction is accomplished at least in three steps, (a) generation of the oxime zwitterion (betaine), (b) formation of a phosphyloxime (POX) intermediate, and (c) irreversible decomposition of the POX into less toxic compounds (Steinberg and Solomon 1966). POXs were identified as highly toxic anticholinesterases, even 10- to 100-fold more toxic than their parent compounds (Hackley et al. 1959; Lamb et al. 1964, 1965; Schoene 1973; Nenner 1974; de Jong and Ceulen 1978; Harvey et al. 1984, 1986; Worek et al. 2000). The stability of the POX intermediate seems to be an important factor determining the velocity of the direct reaction. During the reaction of HI 6 and crotylsarin, an immediate increase followed by a diminution of the peak at 273 nm was observed (Fig. 1). This could be interpreted as formation and degradation of a metastable transition state complex in which the POX and other reaction products are involved. In the same way the alteration of the HI 6 spectrum at 273 nm during the reaction with VX (Fig. 2) and the marked hypsochromic shift during the reaction of 2-PAM and cyclosarin (Fig. 3) could be explained.

Recently, Ashani et al. (2003) reported upon the distinctive properties of the POXs formed from 2-, 3-, and 4-PAM with different OPCs (e.g. sarin, cyclosarin, soman, tabun, VX). The t 1/2 values of the non-enzymatic decomposition of the 2-PAM POXs ranged from 0.7 to 1.5 min. We observed similar half-lifes from 3.9 to 7.5 min for the complete direct reaction (formation and decomposition of the POXs) of cyclo- or crotylsarin with 2-PAM. However, our results obtained with 2-PAM and VX at 294 and 326 nm show much higher values (t 1/2 > 60 min). The reason for this difference might be the leaving group effect during POX formation mentioned above. Furthermore, it must be taken in account that Ashani et al. (2003) worked with highly diluted HEPES buffer (0.01 M), pH 7.8 at 29°C. The concentration of the POXs was in the order of magnitude of 0.05 mM (=50 μM) as described in an earlier paper of this group (Leader et al. 1999). Compared to the conditions we used (0.1 M MOPS buffer, pH 7.4 at 25°C; 15 μM oxime and 1.5 mM nerve agent), the different buffer type and buffer concentration, pH value, and reaction temperature may have an accelerating effect on the decomposition velocity of the POXs and may explain the short half-times given by Ashani et al. (2003).

The question whether the direct reaction could contribute to the detoxication of nerve agents in vivo during OPC poisoning can be evaluated by means of the maximum reaction velocities (v max) in Table 1. These values reach from 0.070 × 10−6 to 2.970 × 10−6 M min−1, hence, the pair HLö 7/VX shows the lowest and the combination obidoxime/crotylsarin the highest reaction velocity (Table 1). It can be assumed that during poisoning the OPC concentration in vivo is in the nanomolar range and the reasonable therapeutic concentration of an oxime is approximately 15 μM. Under these conditions the maximum reaction velocities are v max = 0.465 × 10−12 M min−1 (at 10 nM VX and 15 μM HI 6) and v max = 19.80 × 10−12 M min−1 (at 10 nM crotylsarin and 15 μM obidoxime).

The v max values of all the other OPC oxime combinations in Table 1 are to be found within this range. These results are in agreement with a previous study (Becker et al. 1997) using soman/HI 6. In this case the maximum reaction velocity was calculated to be v max = 2.55 × 10−12 M min−1 (at 10 nM soman and 15 μM HI 6). This order of magnitude (picomoles per litre) demonstrates that the direct reaction cannot markedly contribute to the therapeutic effect of oximes in vivo.

Conclusion

The results presented here with crotyl- and cyclosarin demonstrate that the direct reaction of OPCs and pyridinium oximes could be used for mild decontamination/detoxication purposes (skin, mucous membranes, wounds). Pyridinium oximes could substitute the efficient but corrosive preparations of hypochlorites (NaOCl solution, calcium hypochlorite/magnesium oxide powder). This holds true for previously published data obtained with sarin, soman, and tabun (Becker et al. 1997). For practical administrations it might be useful to accelerate the third step of the direct reaction, i.e., the degradation of the POXs, by addition of bacterial (Ashani et al. 1998; Leader et al. 1999; Herkenhoff et al. 2004) or human plasma enzymes (Kiderlen et al. 2000; Worek et al. 2000).

The importance of the direct reaction in the detoxication during human OPC poisoning may be assessed by considering the generated data (Table 1) and by regarding the concentrations of OPCs and oximes, which can be expected in human OPC poisoning. Hence, the maximum reaction velocity was calculated to be in the range of 10−12 M min−1 (picomoles per litre). This demonstrates that in vivo the direct reaction is only of small significance. However, it can be considered as a beneficial side effect to the oxime action in vivo.