| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
There are acetylcholine receptors throughout the central nervous system, and they may mediate some forms and aspects of convulsive activity. Most high-affinity binding sites on nicotinic acetylcholine receptors for nicotine, cytisine, and epibatidine in the brain contain the ß2 subunit of the receptor. Transitional inhaled compounds (compounds less potent than predicted from their lipophilicity and the Meyer-Overton hypothesis) and nonimmobilizers (compounds that do not produce immobility despite a lipophilicity that suggests anesthetic qualities as predicted from the Meyer-Overton hypothesis) can produce convulsions. The nonimmobilizer flurothyl [di-(2,2,2,-trifluoroethyl)ether] blocks the action of -aminobutyric acid on -aminobutyric acidA receptors, whereas the nonimmobilizer 1,2-dichlorohexafluorocyclobutane (2N, also called F6) does not. 2N can block the action of acetylcholine on nicotinic acetylcholine receptors. We examined the relative capacities of these compounds to cause convulsions in mice having and lacking the ß2 subunit of the acetylcholine receptor. The partial pressure causing convulsions in half the mice (the 50% effective concentration [EC50]) was the same as in control mice. For the knockout mice, the EC50 for flurothyl was 0.00170 ± 0.00030 atm (mean ± SD), and for 2N, it was 0.0345 ± 0.0041 atm. For the control mice, the respective values were 0.00172 ± 0.00057 atm and 0.0341 ± 0.0048 atm. The ratio of the 2N to flurothyl EC50 values was 20.8 ± 3.5 for the knockout mice and 21.7 ± 7.0 for the control mice. These results do not support the notion that acetylcholine receptors are important mediators of the capacity of 2N or flurothyl to cause convulsions. However, we also found that both nonimmobilizers inhibit rat 4ß2 neuronal nicotinic acetylcholine receptors at EC50 partial pressures (0.00091 atm and 0.062 atm for flurothyl and 2N, respectively) that approximate those that produce convulsions (0.0015 atm and 0.04 atm). IMPLICATIONS: The results from the present study provide conflicting data concerning the notion that acetylcholine receptors mediate the capacity of nonimmobilizers to produce convulsions.
Several observations suggest that acetylcholine receptors may be important to the initiation and propagation of convulsions. Such receptors provide excitatory neurotransmission throughout the central nervous system (1,2) and can play a role in the genesis of seizures (3). Direct application of acetylcholine to the cerebral cortex can precipitate seizures (4), an effect that is exaggerated in the isolated cortex (5). Strains of rats with larger concentrations of brain acetylcholine have an increased susceptibility to seizures (6), and mutations of 4 and ß2 nicotinic acetylcholine receptors subunits have been associated with epilepsy (7,8). Blockade of cerebral acetylcholinesterase predisposes to convulsions (9). Nicotine application to the brain can produce convulsions, and this action is blocked by nicotine-blocking drugs (10). But, clearly other receptors can mediate convulsive activity (11,12). For example, localized seizures produced by the topical application of strychnine (a blocker of glycine receptors) do not increase acetylcholine release (9).
Volatile and gaseous compounds, particularly transitional compounds and nonimmobilizers (13), can produce convulsions. Such compounds tend to be additive in their capacities to produce convulsions or modestly deviate from additivity (14). One pair of nonimmobil-izers that does deviate is flurothyl [di-(2,2,2,-tri-fluoroethyl)ether] and 1,2-dichlorohexafluorocyclobu-tane (2N, also called F6), which show a 17% antagonism (i.e., the combined effect is 17% less than the predicted additive effect) (14). Studies of these and other immobilizers and transitional compounds suggested the possibility that at least two mechanisms underlay their convulsive activity (15). One mechanism was suggested by a correlation of the 50% effective concentration (EC50) for the convulsiveness with the lipophilicity (nonpolarity) of the compound (2N lies in this group). The other mechanism (particularly for flurothyl) was a capacity to block the action of
We obtained mice genetically engineered to lack the nicotinic ß2 gene product on the background of the c57 Bl/6J strain (ß2 knockout). Most of the high-affinity binding sites on nicotinic acetylcholine receptors for nicotine, cytisine, and epibatidine in the brain contain the ß2 subunit (21). Nearly all nicotinic acetylcholine receptors in the locus ceruleus contain the ß2 subunit; such neurons provide most of the noradrenergic innervation in the brain, and nicotine potentiates noradrenaline release from their terminals (22). Mice lacking the nicotinic ß2 gene display a reduced antinociceptive effect of nicotine on the hot-plate test and diminished sensitivity to nicotine in the tail-flick test (23). They have mild deficits in learning, particularly in older animals (24). They show decreased place preference to 5 mg/kg of cocaine, and dopamine turnover is decreased in normal mice after treatment with 5 mg/kg of cocaine but not in knockout mice (25). The mice otherwise seem normal. The ß2-containing receptors probably mediate most of the reinforcing properties of nicotine (21). We hoped that the study of the convulsive concentrations of flurothyl and 2N in knockout versus control mice would provide further insights into the mechanistic basis for the convulsant effects of the nonimmobilizers. The present study examined this possibility. In addition, we studied the capacity of flurothyl to block the action of acetylcholine on rat
The Committee on Animal Research of the University of California, San Francisco, approved our study of male mice genetically engineered to lack the nicotinic ß2 gene product on the background of the c57 Bl/6J strain (ß2 knockout) and closely related wild type cousins. Mice were housed five to a cage in the animal care facility for several weeks before experimentation. The mice were exposed to 12-h cycles of light and dark and had food and water ad libitum. Mice were approximately 6 mo old at the time of experimentation. We purchased 2N and flurothyl from PCR Incorporated (Gainesville, FL). The concentration of flurothyl and 2N that caused convulsions in 50% of mice (the EC50) was determined in 12 ß2-knockout mice and 12 control (wild type) mice. Each of six mice (two to four mice from either the control or knockout strains) was placed in a clear plastic tube that did not compress the mouse but was of a diameter that prevented the mouse from turning back on itself. Rubber stoppers at each end of each tube sealed the tube except at openings that permitted the ingress and egress of gases and the sampling of gases exiting the tube. Oxygen was delivered via a manifold to all six tubes at a flow rate that exceeded the average minute ventilation of each mouse by a factor of at least four. This was determined by measuring the exiting partial pressure of carbon dioxide and finding that the average was <10 mm Hg (i.e., the gases respired by the mouse were diluted at least four-fold, assuming an alveolar carbon dioxide partial pressure of 40 mm Hg). Temperatures were not measured because we previously found that 2N does not affect temperature regulation (26). Flurothyl or 2N was added to the inflowing gas stream from temperature compensated vaporizers. The initial convulsant concentration was less than that which produced convulsions, and this concentration was held constant for at least 20 min. The concentration then was increased by 15%30% of the previous value and again held constant for 20 min. All mice were continuously monitored by an observer blinded to the genetic identity of each mouse. This process continued until all mice had convulsed (i.e., had a tonic-clonic seizure in which the head was extended, the lips retracted, the eyes closed, and the fore and hind limbs demonstrated repetitive flexion and extension). As soon as a mouse convulsed, it was removed and returned to its home cage. A total of 24 mice were studied. One mouse in each group (knockout and control) died, and their data are not included (i.e., n = 11 for both groups). Flurothyl and 2N concentrations were separately analyzed using a Gow Mac model 580E gas chromatograph (Gow-Mac Instrument Corp, Bridgewater, NJ) equipped with a 40-ft-long SF-96 column at 100°C, through which flowed a nitrogen carrier gas stream (10 mL/min). The flame ionization detector (at 192°C) was maintained with hydrogen (40 mL/min) and air (300 mL/min). Gas samples were injected into a 0.2-mL sample loop. The gas chromatograph was calibrated with primary standards. For each mouse, the convulsive EC50 partial pressure was calculated as the mean of the highest partial pressure of flurothyl or 2N that did not produce convulsions and the lowest partial pressure that did. The mean ± SD for each group was calculated and an unpaired t-test applied by which we compared the EC50 values for the knockout versus the control group. For each mouse, we also obtained a value for the ratio of the EC50 value for 2N to the value for flurothyl. The mean and SD for these ratios for the knockout versus the control groups were obtained and an unpaired t-test applied. We accepted a P value of <0.05 as significant.
Studies of the inhibitory effects of 2N and flurothyl on neuronal nicotinic acetylcholine receptors were conducted as described in our previous work (19). Indeed, the data for 2N are taken from that work, whereas the data for flurothyl are new. Complimentary DNAs for the rat
For the knockout mice, the EC50 for convulsions from flurothyl was 0.00170 ± 0.00030 atm (mean ± SD) and from 2N was 0.0345 ± 0.0041 atm. These values did not differ from those for the control mice, with the respective values being 0.00172 ± 0.00057 atm and 0.0341 ± 0.0048 atm (n = 11 for all values). The ratio of the 2N to flurothyl EC50 values was 20.8 ± 3.5 for the knockout mice and 21.7 ± 7.0 for the control mice. These ratios did not differ. The average carbon dioxide partial pressure issuing from the cylinders that housed the knockout mice was 7.3 ± 2.8 mm Hg, and that from the control mice cylinders was 6.7 ± 2.4 mm Hg.
Both nonimmobilizers reversibly inhibited rat
Our data indicate that the EC50 for convulsions from flurothyl or 2N in mice lacking the nicotinic ß2 gene product does not differ from the EC50 in control mice. Further, the ratio of the EC50 for convulsions from flurothyl to the EC50 for convulsions from 2N does not differ between the two types of mice. Such data do not support a role of acetylcholine receptors as mediators of convulsive activity from nonimmobilizers. Absence of a role is also consistent with the finding that enhanced choline acetyltransferase activity does not explain the action of inhaled convulsants, particularly 2N and flurothyl (27). It is also consistent with the finding that 2N and inhaled anesthetics both can block human neuronal acetylcholine receptors expressed in oocytes (28), including nicotinic acetylcholine receptors (19,20). Inhaled anesthetics usually suppress convulsions, including those produced by 2N (29). The data from the present report may be subject to limitations that compromise interpretation of the results. In particular, in knockout animals there may be compensatory changes in other systems. There might be an upregulation of other acetylcholine receptor subunits that would compensate for the absence of the ß2 subunit, although this has not been found (30). Even if it occurred, such an effect would have to exactly match the effect of the loss of the ß2 subunit because we found not even a hint of a trend of a difference between the control and knockout mice. Also, as noted in the Introduction, acetylcholine receptors containing the ß2 subunit are the predominate receptors mediating the neuronal nicotinic effects of acetylcholine. Thus, the studies of knockout mice do not support a role for acetylcholine receptors as mediators of the convulsions produced by flurothyl and 2N. However, the studies of receptors might be considered to be supportive of the view that acetylcholine is important. Although, as noted in the Introduction, the application of acetylcholine or an enhancement of its effects may cause convulsions, blockade may also have this effect. Thus, the application of d-tubocurarine to the brain can cause convulsive activity (3133). Atropine applied topically to the brain can cause spiking activity but also blocks the convulsive activity of topically applied acetylcholine (34). Thus, it seems that disturbances in acetylcholine-mediated activity, either increases or decreases, can mediate convulsions. The 50% effective dose values for inhibition of the current induced by application of acetylcholine to receptors in Xenopus oocytes (Fig. 1) are 25 ± 3 µM for flurothyl and 29 ± 2 µM for 2N. A partition coefficient for flurothyl of 0.7 at 37°C (35) and 0.0119 for 2N (13) allows a calculation of the partial pressures these compounds would produce at the EC50 concentrations (0.00091 atm [0.091%] of flurothyl and 0.062 atm [6.2%] of 2N). Such values are close to those that produce convulsions with these drugs (0.0015 atm of flurothyl and 0.04 atm of 2N (14)). The findings for blockade of neuronal nicotinic receptors at concentrations close to those that produce convulsions suggests the possibility of a causal connection. However, inhaled anesthetics, compounds that normally suppress convulsions, also block neuronal acetylcholine receptors, doing so at EC50 values less than MAC and sometimes at a small fraction of MAC (19,36,37). Thus, if acetylcholine-receptor inhibition by inhaled compounds can mediate the development of seizures, counterbalancing effects of inhaled anesthetics, effects not available in nonimmobilizers must prevent such seizures. Flurothyl blocks the action of GABA in GABAA receptors in oocytes (17), but 2N does not (16). At the crab neuromuscular junction, flurothyl blocks inhibitory transmission (GABA) more than excitatory transmission (38). This may explain the greater potency of flurothyl as a convulsant, which supports the notion that blockade of GABAA receptors may mediate the convulsant effects of some inhaled compounds. However, the concentration of flurothyl that blocks the action of GABA in GABAA receptors in oocytes (17) is much larger than the concentration that produces convulsions. In contrast, the concentration that blocks acetylcholine receptors is close to that which produces convulsions (see Results), suggesting the greater importance of acetylcholine blockade. Further mechanistic investigations of the underlying basis of convulsions from inhaled compounds might be of clinical importance. Some previously and presently used volatile anesthetics have convulsant properties. Enflurane can cause convulsions (39), but enflurane is no longer widely used. Sevoflurane also can cause convulsions (4044).
Supported, in part, by NIH grant 1PO1GM4781808. The mice studied for this report were a gift from J.P. Changeux at the Institute Pasteur, Paris, France.
Dr Eger is a paid consultant to Baxter Healthcare Corp.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|