Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Sarin (isopropylmethylphosphonofluoridate) is an organophosphorous nerve agent that irreversibly inhibits the essential enzyme acetylcholinesterase (AChE), resulting in excessive amounts of acetylcholine (ACh) in cholinergic synapses. This cholinergic hyperactivity rapidly evolves into a more generalized neuroexcitability, which triggers seizure activity in susceptible brain areas (Kadar et al., 1995; Shih and McDonough, 1997; Van Helden and Bueters, 1999). Novel treatment strategies are currently under development, because the current therapy (atropine, oxime, and diazepam) seemed inadequate with respect to the suppression of the seizure activity and subsequent brain pathology in primates (Hayward et al., 1990; Van Helden et al., 1996; Shih and McDonough, 1997; Lallement et al., 1998). In this respect, a possible role for the adenosine A₁ receptor-mediated inhibition of ACh release in the brain was explored recently (Van Helden et al., 1998; Van Helden and Bueters, 1999; Bueters et al., 2002). This approach is aimed at the prevention or attenuation of ACh accumulation, thereby stifling the neurotoxic cascade in its birth.

Intracerebral application of a series of adenosine A₁ receptor agonists could effectively inhibit the ACh release (Bueters et al., 2000; Materi et al., 2000). Moreover, promising results were obtained with the adenosine A₁ receptor agonist N⁶-cyclopentyladenosine (CPA) in treating organophosphate-intoxicated rats: cholinergic symptoms were suppressed and mortality was prevented (Van Helden et al., 1998; Bueters et al., 2002). Microdialysis studies confirmed that accumulation of central ACh was attenuated upon CPA administration, which presumably explains the observed protection (Bueters et al., 2002).

However, some other observations do not easily fit with the conclusion that the protective effect of CPA is mediated via presynaptic cholinergic modulation. In general, adenosine A₁ receptor agonists exhibit very poor blood-brain barrier transport properties, which makes it questionable whether sufficient concentrations of CPA reach the brain to achieve adequate ACh inhibition (Pardridge et al., 1994; Kurokawa et al., 1996). Moreover, in sarin-intoxicated and CPA-treated rats, AChE activity in the brain was partially spared from being inhibited relative to untreated animals, indicating that CPA interferes with the inactivation of AChE by sarin (Bueters et al., 2002). CPA may achieve this by either directly blocking the interaction of sarin with the catalytic site of AChE, or indirectly by diminishing the amount of sarin reaching the central nervous system (CNS).

The aim of the present study was to characterize the interaction between CPA, sarin, and AChE. Hereto, a kinetics of pharmacodynamics (K-PD) model has been developed to properly describe the in vivo three-way interaction between CPA, AChE in blood and brain, and the AChE inhibitor sarin. In addition, clinical symptoms associated with sarin poisoning were monitored examine a possible direct relationship between the severity of symptoms and the degree of AChE inhibition in blood and brain.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male Wistar rats (~300 g; n = 67; Harlan B.V., Horst, The Netherlands) were housed with two or three animals per cage. Temperature was kept at 20 ± 1°C and relative humidity at 60 ± 5%, and a 12-h light/dark cycle was maintained (lights on at 7:00 AM). Acidified water and standard rodent chow (Teklad Global Diet; Harlan B.V.) were freely accessible. The Ethical Committee on Animal Experimentation of TNO approved all experiments described.

Determination of the Direct Effect of CPA on AChE Inactivation in Vitro. Rats were sacrificed (n = 3), blood was collected in heparinized glass tubes, and the brain was dissected out on ice. The brain tissue was homogenized [900 rpm, 10% (w/v) homogenate] in ice-cold TENT buffer, which consisted of 50 mM Tris, 5 mM EDTA, 1 M NaCl, and 1% v/v Triton X-100, pH 7.4 (Sigma-Aldrich B.V., Zwijndrecht, The Netherlands). CPA (2 µM; RBI, Inc., Zwijndrecht, The Netherlands) was incubated in 250-µl aliquots of blood or brain homogenate for 1 min at 37°C. Subsequently, sarin was added to the blood samples resulting in final concentrations of 0, 1.2, or 2.4 µM, or to brain homogenate samples, yielding final concentrations of 0, 14, and 28 nM. The mixtures were incubated for another minute. During the incubations steps the samples were gently shaken in an oscillating water bath. Finally, a 25-µl aliquot was drawn in duplicate and transferred into 250 µl of saponine solution (1%) or 250 µl of 50 mM phosphate buffer (PB) in case of blood or brain homogenate, respectively. Samples were immediately frozen and stored at 70°C until analysis.

Surgical Procedures. Rats were anesthetized with a single i.p. injection containing 0.675 ml/kg hypnorm (Janssen Pharmaceutica, Beerse, Belgium) and 0.675 ml/kg dormicum (Roche Nederland B.V., Mijdrecht, The Netherlands). An indwelling cannula was implanted in the right femoral artery for the serial collection of arterial blood samples. The cannula was filled with buffered saline (NPBI B.V., Emmer-Compascuum, The Netherlands) containing 20 U/ml of heparine (LEO Pharmaceutical Products B.V., Weesp, The Netherlands) and tunneled subcutaneously to the back of the neck. The incisions were sprayed with Nobecutan (Astra Meditec, Rijswijk, The Netherlands) to aid recovery. Rats were individually placed and allowed to recover 24 to 36 h from the surgical procedure.

Animal Experiments. Three groups of rats were injected according to the following treatment regimen: group A, 144 µg/kg sarin s.c. followed by vehicle i.m. after 1 min (n = 20); group B, 144 µg/kg sarin s.c. followed by 2 mg/kg CPA i.m. after 1 min (n = 10); and group C, 2 mg/kg CPA i.m. followed by 144 µg/kg sarin s.c. after 2 min (n = 26).

Radiometric Determination of AChE Activity. After appropriate dilution, samples were assayed for AChE activity using a radiometric method described previously by Johnson et al. (1975). Briefly, acidified [³H]ACh iodide (37.0 MBq/5 ml ethanol; PerkinElmer Life Sciences, Boston, MA) was purified from [³H]choline by extraction with a mixture of toluene/isoamylalcohol (9:1) followed by extraction with diethyl ether. Subsequently, 10 ml of a freshly prepared ACh perchlorate solution was added, which resulted in a final ACh concentration of 3 mM (60-80 µCi). To 50 µl of the samples, 500 µl of 50 mM PB containing 50 mM NaCl, 2 mM MgSO₄, and 1% Triton X-100 and 25 µl of the labeled ACh were added, and the samples were incubated 20 min at ambient temperature. After the reaction was terminated with 100 µl of 10 M acetic acid, 4.5 ml of scintillation fluid [18 mM 2,5-diphenyloxazol, 0.25 mM 2,2'-p-phenylenbis(4-methyl-5-phenyloxazol and 11.1% (v/v) tert-mylalcohol in toluene] was added and the samples were firmly shaken. After both layers had separated, the AChE activity was counted. The activities were calculated with a calibration curve (15-7500 µU) on the basis of AChE from electric eel (Sigma-Aldrich B.V.) (r > 0.98; n = 7). The residual assay coefficient of variation for 75, 750, and 7500 µU were 1, 8, and 8%, respectively (n = 5).

Dose-Response-Time Model. Because current analytical methods are insufficiently robust and sensitive to adequately determine free sarin concentrations in blood and brain, the residual AChE activity in blood and brain was determined. This is considered a sensitive biomarker for sarin exposure (Jokanovic and Maksimovic, 1997; Lotti, 1995; Nigg and Knaak, 2000). Subsequently, the profile of AChE inhibition by sarin versus time was characterized with a newly developed K-PD model, which extracts information about kinetic parameters from the pharmacodynamic profiles (Fig. 1; Fisher and Wright, 1997; Gabrielsson et al., 2000). In this model it is assumed that sarin is distributed upon dosing to a biophase (i.e., site of action) compartment with rate constant k', from which it irreversibly inhibits AChE in both blood and brain. An irreversible response model described the AChE inhibition (Jusko, 1971). Because sarin stoichiometrically reacts with AChE, its concentration in the biophase compartment is directly linked to the irreversible response model of AChE. The rate of inactivation by sarin is then determined by the concentrations of sarin and AChE and a rate constant k_irr, which are different in blood and brain. To implement the influence of CPA on the distribution of sarin to the biophase compartment, the shift parameter was introduced. Under control conditions, i.e., in the absence of CPA, the value of this shift is 1. If CPA affects the distribution of sarin to the biophase, the shift parameter will deviate from 1: higher values indicate that a larger amount of sarin reaches the biophase compartments to inhibit AChE, and lower values mean that less sarin is transported to the biophase compartment. Given these interpretations, the biophase kinetics was modeled on the basis of two compartments, represented by the following equations

(1)

(2)

in which C_dose and C_bio reflect the concentration of sarin at the dose site and in the biophase, respectively. D is the dose of sarin administered, k' represents the rate constant for sarin transport properties, and shift represents the steady-state partition coefficient.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Proposed K-PD model to describe the observed changes in inhibition of the AChE activity by sarin (144 µg/kg s.c.) in the blood and brain in the presence of CPA (2 mg/kg i.m.). In this model it is assumed that sarin is distributed from a dose compartment to a biophase compartment with rate constant k', from which it inactivates AChE in blood and brain. The elimination constants k1irr and k2irr, and the concentrations of the enzyme and sarin describe this inhibition. The impact of CPA on the distribution of sarin to the biophase compartment is given by the shift parameter. Under control conditions, this factor is 1. When distribution of sarin is altered, which is reflected by a change in the rate of AChE inactivation, the factor will deviate from 1.

(3)

(4)

(5)

(6)

Logistic Regression Analysis. The relationship between the residual AChE activity in blood and hippocampus and the onset of occurrence of the three cholinergic symptoms studied (chewing, convulsions, and respiratory distress) was investigated. The appearance of a symptom was marked as a positive response; the absence of a symptom during the sarin exposure was marked as a negative response. Both negative and positive responses were related to the residual AChE activity. The data obtained from the three treatment groups were pooled and analyzed with linear logistic regression. The probability of a response was investigated based upon a linear logistic model governing the probability p_j of the jth response being 1. The logit of p_j is given by the following equation:

(7)

where ₁ and ₂ represent the fixed effect parameters for slope and intercept and AChE_j is the value of the residual AChE activity. The intercept is given by the log of the EC₅₀ value of the response times the slope parameter. The probability is then as follows:

(8)

The appearances of the different cholinergic signs were fitted simultaneously while taking only residual variability into account, i.e., interindividual random effects were not included in the model. The reason to exclude interindividual variation was because only one observation per clinical sign per individual rat was made. Yano et al. (2001) showed that in such a case estimates are not improved by including an interindividual error model. Similar to the previously described model, the residual variability was characterized by an additive error model (eq. 6). The values of the population  and ² were estimated with NONMEM and individual parameter estimates were obtained in a Bayesian post hoc step.

Statistical Analysis. Goodness-of-fit was analyzed using visual inspection, objective function, and assessment of parameter correlation. Statistical analysis was performed using one-way analysis of variance followed by the Student Newman-Keuls test, whenever appropriate. All data are represented as mean ± S.E.M. and differences were considered significant for P values <0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Direct Effect of CPA on the AChE Inactivation by Sarin in Vitro. Increasing concentrations of sarin dose dependently diminished the AChE activity in both blood and brain preparations, regardless of the presence of 2 µM CPA (Fig. 2). A 100-fold higher concentration of sarin was required to inactivate a similar amount of AChE in blood compared with the brain tissue.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Inactivation of the AChE levels by sarin in heparinized blood (A; closed circles) and brain homogenate (B; closed triangles). Preincubation with CPA (2 µM) does not affect the sarin-mediated inhibition of the enzyme in blood (open circles) or in brain (open triangles). Effect is expressed as the percentage of the AChE levels with no drugs present (mean ± S.E.M.; n = 3).

Effect of CPA on the Distribution of Sarin in Vivo. Fig. 3 shows the observed, predicted population, and individual AChE activity-time profiles in blood for the three treatment groups, based upon the K-PD model (Fig. 1). Population parameter estimates are summarized in Table 1. From each rat a blood sample was drawn before drug administration, of which the AChE activity was set at 100%. Control groups that received only vehicle treatment (n = 3) or vehicle followed by 2 mg/kg CPA (n = 5) showed no loss in AChE activity in blood and brain tissue (data not shown). Directly upon subcutaneous administration of sarin, a steep decrease in AChE activity was apparent in rats that did not receive CPA treatment (Fig. 3A; n = 20; group A). Within 4 min, the AChE activity was abolished to 0.6 ± 0.3% of control value. CPA treatment shortly after sarin was given, resulted in a small delay in inactivation of the AChE activity, but the AChE was still maximally inhibited (Fig. 3B; n = 10, group B). Treatment with CPA 2 min before sarin administration led to a more pronounced rightward shift of the inactivation curve of AChE in time (Fig. 3C; n = 26, group C). Moreover, in several individual animals some AChE activity remained after 15 min, resulting in an average residual AChE activity of 10.3 ± 4.7%. In this treatment group, one rat died after 10 min, which was excluded from further analyses. In Fig. 3D, the population profiles are summarized, in which the rightward shift as a consequence of the CPA treatment is clearly demonstrated.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   A to C, individual time-response profiles for the inhibition of AChE in blood by sarin for the treatment groups A to C (A: t = 0, 144 µg/kg sarin, t = 1, vehicle; B: t = 0, 144 µg/kg sarin, t = 1, 2 mg/kg CPA; and C: t = 2, 2 mg/kg CPA, t = 0, 144 µg/kg sarin). The observed AChE activities (symbols), the individual predictions (thin lines) and the population predictions (thick lines) are depicted. D, summarizes A to C to clearly demonstrate the influence of CPA on the sarin-mediated inactivation of AChE.

1

View larger version (18K): [in this window] [in a new window]   Fig. 4.   A and B, individual time-response profiles for the inhibition of AChE in the brain by sarin for the treatment groups A and C (A: t = 0, 144 µg/kg sarin, t = 1, vehicle; C: t = 2, 2 mg/kg CPA, t = 0, 144 µg/kg sarin). The observed AChE activities (symbols), the individual predictions (dotted lines), and the population predictions (thick lines) are depicted. C, summarizes A and B to clearly demonstrate the influence of pretreatment with CPA (2 mg/kg) on the sarin-mediated inactivation of AChE in brain.

Relationship between Symptomatology and AChE Activity. During the experiments described above, the occurrence of chewing, convulsions, and respiratory distress were monitored to obtain a relationship between the amount of AChE inhibited and the visible condition of the rats. The observed clinical signs from the three treatment groups were pooled and related to the residual AChE activity in blood and brain tissue with a linear logistic regression model (Fig. 5). The parameter estimates are listed in Table 2. Figure 5 shows that the AChE activity in blood does not correlate with the emergence of cholinergic symptoms. With complete AChE inhibition the probability of response is approximately 60%. In contrast, the residual AChE activity in the brain was very well correlated with the severity of symptoms. The critical AChE activity is approximately 7%; lower AChE activities are likely to result in convulsions and respiratory distress.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Continuous relationships of the occurrence of the sarin-induced symptoms chewing (closed circles), convulsions (open triangles), and respiratory distress (open circles) and the residual AChE activity in blood (A; n = 56) and hippocampus tissue (B; n = 48) obtained with logistic regression analysis (eqs. 7 and 8). The upper part depicts the responses of the pooled symptoms versus the residual AChE activity in blood (A) and brain (B). An appearance of a symptom is indicated by a tic above the horizontal line; the absence of a symptom by a tic below the horizontal line.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The aim of the present study was to characterize the interactions between CPA, sarin, and AChE and to examine whether these interactions could explain the observed protection by CPA against such poisoning. Hereto, a K-PD model has been developed to properly describe the in vivo three-way interaction between CPA, AChE in blood and brain, and sarin. From the in vitro results, it seemed that CPA did not block the catalytic site of AChE for sarin. However, CPA altered the transport of sarin from the injection site to the biophase, which was more pronounced when CPA was present at the moment of sarin administration. The effective dose of sarin reaching the biophase was markedly reduced upon CPA presence, resulting in a smaller inactivated fraction of AChE, specifically in the brain, where sarin exerts its main toxicological effects (Gupta et al., 1991; Kadar et al., 1995; Bueters et al., 2002). This spared fraction of AChE was sufficient to explain the absence of symptoms associated with severe AChE inhibition.

Direct Effect of CPA on the AChE Inactivation by Sarin in Vitro. Sarin dose dependently diminished the AChE activity in both blood and brain preparations, regardless of the presence of 2 µM CPA (Fig. 2). This points out that 2 µM CPA does not interfere in the interaction between sarin and AChE, which ultimately leads to the irreversible inactivation of the enzyme.

Relationship between AChE and the Symptomatology. Recently, analytical methods have been developed to determine sarin directly in plasma (D'Agostino et al., 1999, 2001; Spruit et al., 2000, 2001). However, these assays are neither sensitive nor robust enough to detect concentrations of free sarin in brain tissues. Therefore, the AChE activity is still used as a sensitive biomarker for sarin exposure (Lotti, 1995; Jokanovic and Maksimovic, 1997; Nigg and Knaak, 2000). The isoform of AChE on red blood cells is similar to the enzyme present in the synapses, and some therefore consider erythrocyte AChE inhibition a biomarker for organophosphate toxicity (Holstege et al., 1997; Jokanovic and Maksimovic, 1997). The results of the present study clearly demonstrate that this can be misleading and great care should be taken in extrapolating the amount of inhibited AChE in blood to that in the CNS. In all treatment groups, AChE in the blood is maximally or nearly maximally inhibited, whereas large differences exist between the remaining enzyme activity in the brain. Moreover, the AChE activity in blood seems to be a very poor predictor for the severity of the intoxication (Fig. 5). Therefore, inhibition of AChE activity in blood should only be interpreted as an indicator for exposure and not for toxicity. In contrast, the enzyme inhibition in the brain correlates very well with the severity of the intoxication. Neuronal AChE levels greater than 10 to 15% will probably not lead to visible effects in case of sarin poisoning, and are certainly sufficient to provide protection. This correlation only accounts for the acute toxicity and the development of the various symptoms and not for the situation many hours or even days after the intoxication, when adaptations may have altered this relationship (Lotti, 1995).

Modeling the Effect of CPA on Sarin Distribution. Because plasma concentrations of sarin were not available to judge and quantify possible changes herein upon CPA presence, a K-PD model was applied. Such a model extracts kinetic parameters from the response versus time profiles, which inherently contain information about turnover characteristics and biophase kinetics (Jusko, 1971; Fisher and Wright, 1997; Gabrielsson et al., 2000). This way we could still investigate the impact of CPA on the distribution pattern of sarin. The model was adopted from Jusko (1971), who applied it to quantitatively predict the efficacy of chemotherapeutic agents. To reduce the number of parameters, the biophase kinetics of sarin was described with a single rate constant, i.e., it was assumed that the distribution to the effect site was equal to the elimination from this site (Gabrielsson et al., 2000).

1

5

3

1

5

4

1