Evaluation of the acute behavioral effects and abuse potential of a C8–C9 isoparaffin solvent

ELSEVIER Drug and Alcohol Dependence 46 (1997) 125-135

Evaluation of the acute behavioral effects and abuse potential of a C8-C9 isoparaffin solvent1

Robert L. Balder *, Scott E. Bowen, Eric B. Evans 2, Mary E. Tokarz

Department qf Phurmucology and Toxicology. Medical College of Virginia, Virginia Commonwealth University, Richmond VA 23298.0613, USA

Received 22 February 1995; accepted 1 April 1997

Abstract

We hypothesized that the abuse potential of certain types of inhalants could be evaluated in animals by determining the overlap in their profile of behavioral effects with that of CNS depressant drugs and other depressant-like abused inhalants. For our first attempt in evaluating a solvent with an unknown abuse potential we tested ISOPAR-ETM. ISOPAR-ETM is a mixture of predominately C&C9 isoparaffinic hydrocarbons that is being used more and more frequently as a solvent in industrial and consumer products, including, but not limited to, typewriter correction fluids. Presently, nothing is known about the potential for abuse of products containing this solvent. In the present studies, we compared the volatility of ISOPAR-ETM and the abused solvent 1,l .I-trichloroethane (TCE) in our exposure systems. Additionally, five behavioral procedures were conducted in mice to compare the effects of the two compounds. The results demonstrate that: (1) ISOPAR-ETM was less volatile than TCE; (2) ISOPAR-ETM produced a somewhat different profile of effects than did TCE as assessed with a functional observational battery; (3) unlike TCE, ISOPAR-ETM did not affect performance on tests of motor coordination; (4) TCE and ISOPAR-ETM produced concentration-related decreases in schedule-controlled operant performance with recovery from TCE being somewhat more rapid; (5) ISOPAR-ETM produced cross dependence in TCE-dependent mice; and (6) both TCE and ISOPAR-ETM produced substantial levels of ethanol-lever responding in a drug discrimination procedure, although the ethanol-like effects of ISOPAR-ETM only occurred at response rate decreasing concentrations. Overall, there was a poorer separation of behavioral and lethal concentrations for ISOPAR-ETM than for TCE. Although a somewhat different profile of behavioral effects was obtained with ISOPAR-ETM and TCE, we cannot say with certainty if enough similarities exist with abused inhalants to predict that ISOPAR-ETM would be subject to depressant-like abuse. Nonetheless, the feasibility of preclinical assessment of abuse potential of inhalants was demonstrated. 0 1997 Elsevier Science Ireland Ltd.

Keywords: Solvents; Inhalant abuse; Mice; 1,1,1 -Trichloroethane; Isoparaffins; Operant behavior; Functional observational battery; Drug discrimination

1. Introduction

Many household and industrial products are subject to voluntary inhalation because they contain volatile

* Corresponding author. Tel.: + I 804 82828402; fax: + 1 804 8282117.

’ A preliminary report of this work was presented at the 1994 Annual Meeting of the College on the Problems of Drug Depen- dence.

’ Present address: National Starch and Chemical Co., 10 Finderne Avenue. Bridgewater, NJ 08807, USA

organic compounds with abuse potential (Sharp, 1992). One strategy to reduce abuse by inhalation of these products is to reformulate them with new, less abusable solvents. However, this process requires knowing some- thing about the abuse potential of individual solvents (Balster, 1987). The present study was designed to evaluate the reformulation of a typewriter correction fluid, by comparing the profile of behavioral effects of 1 ,l ,l-trichloroethane (TCE), used in the early formulation, to that of ISOPAR-ETM, a branched C8-C9 isoparaffin solvent used in the new formulation. TCE is a

0376-8716/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights reserved PII SO376-8716(97)00055-O

widely abused solvent that is still used in a number of commercial and industrial products. Previous research in our laboratory and in others has shown that TCE has a profile of acute effects in mice which resembles that of abused depressant drugs such as ethanol and the barbiturates (Evans and Balster, 1991; Bowen et al., 1996).

The central concept to this research is that solvents are abused because they are psychoactive and thus produce an intoxication (Balster, 1987). Solvents without effects on behavior, or those whose behavioral effects occur at impractically high concentrations, should have little or no abuse potential. We also have evidence from a number of research studies in animals that the type of psychoactive effects produced by com- monly abused solvents such as TCE are very similar to those produced by ethyl alcohol and other abused depressant drugs (Balster, 1987; Evans and Balster, 1991; Bowen et al., 1996). Therefore, evidence concern- ing the ability of ISOPAR-ETM to produce behavioral and pharmacological effects similar to those of ethanol and TCE should help predict its potential for abuse.

Ideally, studies of the behavioral and psychological effects of a new solvent such as ISOPAR-ETM would be carried out in human subjects; however, this is often not possible because of safety concerns for the test subjects. We are not aware of any direct human studies of the abuse potential of solvent vapors, although studies of this type with anesthetics are beginning to be performed (Dohrn et al., 1992). For this reason, we have developed animal test procedures using mice for conducting research on the behavioral effects of inhalants (Balster, 1987). For the studies reported here, we have utilized five of these procedures to make direct comparisons between ISOPAR-ETM and TCE. These test procedures include basic studies to determine if ISOPAR-ETM has psychoactivity as evidenced by behavioral effects in mice as well as more direct studies of effects thought to be specifically predictive of abuse potential.

2. Methods

2. I. Arzimul,s

Adult male CFW (ChasRiver Swiss) albino mice (Charles River Breeding Laboratories, Wilmington, MA) were used in all experiments and weighed 27-40 g when testing began. Mice were housed individually in 18 x 29 x 13 cm plastic cages containing wood chip bedding and fitted with steel wire tops. The animal housing facility had a controlled temperature of 22- 24°C and was maintained on a 12 h light/dark cycle. Separate groups of mice were used in each of the behavioral studies conducted. Mice used in the func-

tional observational battery and the cross-dependence studies were allowed free access to food and water. Animals used in the operant and discrimination studies were allowed to gain weight to a maximum of 35 -t 5 g by post-session feeding of 3-4 g/day of rodent chow (Rodent Laboratory Chow, Ralston-Purina, St. Louis, MO). Mice were transported to the laboratory for all testing which was carried out during the light cycle.

2.2. Static exposure chambers

Static exposure chambers were used for mice tested in the functional observational battery and ethanol discrimination protocols. Vapor exposures were conducted in 29 1 cylindrical jars (47cm in height x 35 cm in diameter) which have been described previously (Bal- ster and Moser, 1987). Vapor generation began when liquid solvent was injected through a port onto filter paper suspended below the sealed lid. A fan, mounted on the underside of the lid, was then turned on which volatilized and dispersed the agent within the chamber. Nominal chamber concentrations did not vary by more than 10% from measured concentrations as determined by single wavelength monitoring infrared spectrometry (Miran lA, Foxboro Analytical, North Haven, CT). All vapor exposures were 30 min in duration.

2.3. Dynumic exposure chambers

For dynamic exposures, two of six mouse operant conditioning chambers were modified to allow for solvent vapor exposure (Balster et al., 1982). Vapor generation occurred by initially directing air flow through a bubbler that was immersed in a 500 ml solvent bath contained in a l-l round bottom flask. Air saturated with vapor exited the bath and was mixed with filtered laboratory (fresh) air that was then delivered to the exposure chamber. Control of the vapor concentrations was accomplished with a multiple Dyna-blender (Model 8219, Matheson, Montgomeryville, PA) which monitored and controlled the air flow rate through two valves, one for pure air and one for vapor-laden air. An IBM-compatible microcomputer (AT 486, WIN Labo- ratories, Fairfield, VA) was interfaced with the Dyna- Blender and dictated what flow rates were needed for each test concentration. The total flow entering the chambers was held constant at 10 l/min. Animals were regularly cycled from one of the four air-only chambers on non-test days (Monday, Wednesday and Friday) to the vapor exposure chambers on test days (Tuesday and Thursday).

For the dependence studies, dynamic vapor exposures were also conducted in a 20.8 1 glass rectangular tank (41 x 21 x 27 cm) fitted with a Teflon-lined lid into which vapors were continuously delivered in the air inflow. The methodology for this solvent exposure

R.L. Bdstrr rr al. Drug and Alcohol Dependmcr 46 (1997) 125-1.15 127

system has also been described previously (Evans and Balster, 1993). Briefly, a flow regulator allowed filtered air to pass through a gas dispersion tube immersed in a 2 1 flask containing TCE. This vapor laden air was then combined with filtered air from an additional flow regulator to achieve the final test concentration.

Vapor concentrations in both dynamic systems were monitored on line using a single wavelength monitoring infrared spectrometer (Miran IA, Foxboro Analytical, North Haven, CT). Both static and dynamic exposure systems were housed in a fume hood, which also served to provide white background noise and isolation from the laboratory environment.

2.4. Functional obsrrvational battery

The functional observational battery protocol used in this study was based on a previous version (Tegeris and Balster. 1994; Bowen et al., 1996) and consisted of observations of mice in the exposure chamber, their response to handling when removed from the chamber, post-exposure observations in an open-field, and ma- nipulative behavioral measures.

During the last 2 min of solvent exposure, mice were scored on the following measures: arousal, rearing, clonic movements, tonic movements, palpebral closure, and gait. Immediately after exposure the exposure chamber was opened and the mice were removed within 5 s of vapor termination and evaluated for ease of removal and handling reactivity. Piloerection, righting reflex, forelimb grip strength, the inverted screen task, landing foot splay. approach response, click response, touch response, tail pinch response, arousal, rearing and mobility were evaluated (in the order listed) over the next 4 min (see Tegeris and Balster, 1994 for definitions). Scoring of the functional observational battery was done by a single trained technician who was blind to both the chemical and concentration being tested.

On the day of testing, prior to inhalant exposure and functional observational battery testing, mice were trained on the inverted screen test (Tegeris and Balster, 1994). Each mouse was required to climb to the top of the inverted screen within 10 s during three consecutive training tests. This 10 s cutoff was used during testing in the functional observational battery. The concentrations tested were ISOPAR-ETM (0, 2000, 4000 and 6000 ppm); and TCE (0, 4000, 8000, 10 000, 13 300 and 18 000 ppm). Separate groups of eight mice were used for each test concentration.

The subjects were trained in six two-lever mouse operant-conditioning chambers. The response levers were located on the front wall 8 cm apart and 2.5 cm

above a stainless steel floor and extended 0.8 cm into the chamber. Located midway between the levers was a 3 cm diameter opening containing a trough into which 0.02 ml of sweetened-condensed milk (1 part sugar, 1 part condensed milk, and 2 parts water by volume) could be delivered via a calibrated peristaltic infusion pump (Masterflex, Cole-Parmer Instr.). Illumination of two houselights, located above each of the levers, sig- nalled that the session was in progress. Mice were trained to lever press during daily (5 days per week), 30 min sessions on a fixed-ratio 20 (FR-20) schedule. Animals were trained daily for about 2 months before entering into the testing phase of the experiment. Sol- vent tests were conducted on Tuesdays and Fridays, with training sessions occurring between test days.

To define the concentration-effect function for both ISOPAR-ETM and TCE, two exposure regimens were utilized with eight animals in each group. In the initial exposure regimen, a single concentration of ISOPAR- ETM and TCE was tested in each session consisting of a 20 min exposure preceded and followed by 5 min air-only exposure. An initial 5 min air only exposure served as a control period with the post-exposure 5 min air-only exposure included as an opportunity to observe recovery. The second regimen consisted of testing five concentrations within a single behavioral session. For the first 5 min of the session, the mice were exposed to air only (control period); subsequently, during each successive 5 min period, the vapor concentration was progressively increased. This allowed for the determina- tion of a complete concentration-effect curve within a single 30 min session. The within-session method for each solvent was conducted in duplicate.

2.6. Ethanol discrimination training

Nine mice were initially trained under a fixed-ratio 1 (FRl) schedule to press either of two response levers for milk presentation during daily (Monday-- Friday) 15 min sessions. After several sessions of stable lever pressing, responding on only one lever was reinforced dependent upon whether the mouse had received an i.p. injection 20 min earlier of 1.25 g/kg of ethanol or saline. Subjects were returned to their home cages between injections and placement into the test chamber. The assignment of drug and vehicle levers was random for each subject. During this first phase of drug discrimination training, the mice were gradually shaped through progressive increases in the FR requirement to respond under a FR20 schedule on alternating levers. This required the subjects to produce 20 consecutive responses on the correct lever for milk presentation.

Following reliable FR20 responding, each of the subjects was tested for stimulus control by ethanol and saline injections with test sessions conducted on each Tuesday and Friday. Completion of the response re-

quirement on either lever during the 2 min test sessions produced a liquid reinforcer. These test sessions were followed immediately by a 13 min training session in which only correct responses were reinforced. Acquisi- tion of the discrimination was defined as the successful completion of four consecutive test sessions (two ethanol and two saline). Success was defined as both completing the first FR and responding at least 85% on the correct lever. After successful acquisition training, a short 2 week period of continued training occurred during which subjects were placed in the static exposure chamber 20 min prior to the training session and sub- jected to ‘air-only’ exposures in an effort to adapt subjects to the inhalation procedure.

Following the acquisition of discrimination between 1.25 g/kg ethanol and saline and the adaptation period, animals were tested for ethanol generalization after static inhalation exposures to air only or various concentrations of TCE and ISOPAR-ETM. Discrimination tests were performed on Tuesdays and Fridays contin- gent on the subject completing the first FR on the correct response lever and having over 85% correct- lever responding over the entire training session preceding the test day. Mice continued to be trained on the double alternation sequence of ethanol and saline training sessions between test days in order to preserve the discrimination. On test days, mice were placed into the operant chambers for a 2 min session in which responding on either lever was reinforced. Following the 2 min test, the subjects were returned to their home cages.

All vapors were administered for 20 min beginning 20 min prior to a test session with all animals receiving all of the concentrations. Animals were rapidly removed from the exposure chambers and placed in the test chambers with test sessions beginning within 30 s of termination of the exposure.

2.7. Cross-dependence

Groups of ten mice were placed in the chamber and continuously exposed to 2000 ppm TCE vapor for 4 days. Each day, the chamber was maintained (food and water replenished), the solvent flasks were filled and the mice were removed from the chamber for approximately 15 min for weighing. After 4 days, TCE exposure was abruptly discontinued and the mice were removed from the exposure chamber and placed in standard mouse cages. To measure and quantify the withdrawal reaction, mice were scored for convulsions elicited by handling on a scale developed for assessment of ethanol dependence (Goldstein, 1972). Convulsions were elicited by lifting the mouse by the tail, and scoring was as follows: 0, no effect; 1, tonic convulsion observed when mouse is lifted and given a 180” turn; 2, tonic convulsion when mouse is only lifted and turned; 3, production of a tonic-clonic convulsion elicited by

only a lift; and 4, a severe tonic-clonic convulsion frequently continuing after release of the mouse. Post- exposure convulsions were assessed hourly for 12 h, with additional assessments at approximately 12-24 h intervals thereafter until recovery occurred.

To assess the ability of TCE and ISOPAR-ETM to modify the withdrawal reaction, each compound was administered before either the 3 or 4 h post-exposure assessment, corresponding to the time of peak withdrawal. Post-exposure administration of TCE and ISO- PAR-ETM vapor was accomplished by returning the mice to the exposure chamber during the post-exposure phase of the experiment, adjusting the flow regulators to generate the desired test concentration and exposing the mice for a predetermined period of time (30 or 60 min).

2.8. Data analysis

Concentration-effect curves for operant response rates were analyzed separately for each of the compounds tested using analysis of variance (ANOVA) and Tukey post-hoc comparisons (P < 0.05). The procedure for FOB data analysis was similar to a method used by the US Environmental Protection Agency (Creason, 1989; Tilson and Moser, 1992), with modification for between-subject data versus repeated measures. Contin- uous and count measures were analyzed by means of separate general linear model (GLM) procedures (SAS Institute, Cary, NC). Tukey post hoc tests were used to specify differences (from control) revealed by the overall analysis. Categorical data were analyzed with CAT- MOD (SAS Institute, Cary, NC), a procedure designed to provide a model of analysis of variance (ANOVA) for categorical data. When appropriate, frequencies of behaviors for individual concentrations were compared to control frequencies. In addition, analyses were conducted on each domain of solvent effects by performing CATMOD procedures on overall severity scores derived from tallying severity scores on each individual measure within the domain.

2.9. Chemicals

The solvents used were ISOPAR-Er-M (Exxon Chemi- cal Corporation) and 1 ,l, 1 -trichloroethane (T-391, Fisher Scientific, Fairlawn, NJ).

3. Results

3.1. Performance of’ TCE und ISOPAR-E TM in the exposure systems

To validate nominal concentrations of test atmo- spheres in the static chamber an IR spectrometer was

R.L. Balster et al. i Drug and Alcohol Dependence 46 (1997) 125- 135 129

connected in conjunction with a circulating pump to the ports on the lid of the tank. For both TCE and ISOPAR-ETM, 20 min recordings of infrared absorbance confirm that the test concentrations were rapidly achieved and maintained throughout exposure (Fig. 1). This figure also shows that TCE was com- pletely volatilized more quickly than ISOPAR-ETM. TCE reached maximal concentration within 0.5 min of injection and turning on the fan. ISOPAR-ETM required 2 min to reach an asymptotic level. Further validation was performed by directly comparing the absorbance reading while monitoring in-line with the exposure tank and those obtained from closed loop calibration (i.e. actual versus expected). For three different concentrations of ISOPAR-ETM, the absorbance recorded from the exposure tank never deviated more than 9% from the expected closed loop value.

In the dynamic exposure system, for each solvent a plot of the calibrated flow rate through the solvent bath versus concentration illustrates a relatively linear rela- tionship (Fig. 2). This figure also graphically illustrates the difference in volatility between these compounds. For ISOPAR-ETM, flow rates approximately ten-fold higher than those of TCE were required to generate the same concentrations.

3.2. Functional observational battery

Table 1 shows a summary of the results for the effects of TCE and ISOPAR-ETM on each measure included within the five domains of the functional observational battery. For both compounds, results are shown only for tests conducted immediately after exposure. The table demonstrates those measures for which significant effects were obtained and whether these effects were obtained at only one or at two or more

1 5 10 15 20

60OO~m l,l,l-Trlchlow,thane

--l---T

Time (min)

1’1g I. Time course of static exposure chamber concentrations as determined by IR hpectrometry after injections of TCE and ISOPAR- FT”’ The fan in the chamber was turned on at the arrows. Note stability of exposure for 20 min and the longer time required for ISOPAR-F”“’ to reach asymptotic concentration.

Dynamic Exposure System

0 ISOPAR ET” . TCE

Concentration (ppm)

Fig. 2. Flow rates through the bubbler needed in the dynamic exposure system to produce various concentrations of ISOPAR-ETM and TCE. Target concentrations were measured by IR spectrometry. Note that about IO-fold higher flow rates through the liquid solvent were required to volatilize ISOPAR-ETM than for TCE.

concentrations. It can be seen that ISOPAR-ETM produced few effects in the concentration range tested, and most effects were seen only at one concentration, the highest concentration tested (6000 ppm). Initially, a higher concentration of 8000 ppm ISOPAR-ETM was examined, however deaths occurred due to convulsions. Thus, in the mouse, there is a small separation between concentrations of ISOPAR-ETM which exert observable behavioral effects and those which are lethal. These results contrast with those for TCE which affected a wide range of measures in the functional observational battery and whose effects were often observed over a wide range of concentrations. In addition, concentrations as high as 18 000 ppm TCE have been tested without lethality (Woolverton and Balster, 1981). Thus, for TCE, there was a greater separation of behaviorally active and lethal concentrations than for ISOPAR-E’rM.

ISOPAR-ETM produced some effects like those of TCE, such as decreased arousal during the last 2 min of exposure (Fig. 3); however, on other measures the effects of ISOPAR-ETM differed from those produced by TCE. For example, none of the test concentrations of ISOPAR-ETM induced loss of righting reflex, while TCE produced a concentration-dependent loss of righting reflex (Fig. 4). Likewise, TCE decreased forelimb grip strength (data not shown) and impaired the ability of mice on the inverted screen task while ISOPAR-ETM

130 R.L. Bukter e, rd. Drug trnd ,4lcoi~~l Dtpnclencr 46 (1997) 125- 135

Table I Summary of effects of acute exposure to I,l,l-trichloroethane and ISOPAR-ErM on individual measures of the functional observational battery

TCE ISOPAR-ErM

CNS activity Arousal I “1 *1 Rearing 2 ** 1

CNS excitability Ease of removal I ** t Handling reactivity 7 ** 1

Autonomic effects Urination I Defecation 7 Lacrimation 3 Piloerection 4

Muscle tone~equilibrium Gait I *1 Mobility 2 **

Righting reflex 3 ** i

:i

Forelimb grip strength 3 ** 1 Inverted screen 5 **

Landing foot splay 6 ** t **1 Sensorimotor reactivity

Approach response I *1 “1 Click response 2 ** 1 Touch response 3 **

Tail pinch response 4 *11

* Affected only at the highest concentration (PC .05); ** two or more doses are different from vehicle on domain or test (PC .05); -, no effect compared to vehicle; TL. denotes direction of change in measure.

failed to disrupt grip strength and the inverted screen task at any concentration tested (Fig. 5).

Under a few conditions ISOPAR-ETM produced effects opposite to those of TCE. Exposure to 2000-6000

.

. . . r

Fig. 3. Effects of ISOPAR-E-r”’ and TCE on the arousal measure of the functional observational battery, displayed as the percentages of animals receiving a normal score. Arousal was scored during the last 2 mm of a 20 min exposure to the concentrations shown. Control values were obtained from air only test sessions (n = 8 mice/concentration).

RIGHTING REFLEX

i

Fig. 4. Effects of ISOPAR-ETM and TCE on the righting reflex measure of the functional observational battery. Shown are the percentages of mice showing a normal righting reflex. The righting reflex was measured within 2-3 min of removal from 20 min exposures to the concentrations shown. Control values were obtained from air only test sessions (n = 8 mice/concentration).

ppm ISOPAR-ETM decreased hindlimb foot splay, whereas TCE increased foot splay (Fig. 6). During exposure in the static tank, ISOPAR-ETM and TCE had a different profile of effects on the overall activity level of mice as measured by the number of rears recorded (data not shown). As the concentration of ISOPAR- ETM was increased (4000-6000 ppm) a concentration- related increase in rearing behavior was observed. Exposure to increasing TCE concentrations resulted in a depressant-like biphasic affect on the number of rears with 8000 ppm producing an increase and higher levels reducing rearing behavior.

INVERTED SCREEN TASK

Fig. 5. Effects of ISOPAR-ETM and TCE on motor coordination as measured by the inverted screen test. Shown are the percentage of mice that successfully climbed to the top of the screen within 60 s when tested within 2 min of removal from 20 min exposures to the concentrations shown. Control values were obtained from air only tests sessions (n = 8 mice/concentration).

HINDLIMB FOOT SPLAY

30 I

CONTRUL 2000 4000 5000 CONTROL 4.300 8000 10000 ,IfW ,800O

ISWAR E’~ @pm, TCE Iwml

Fig. 6. Effects of ISOPAR-ETM and TCE on the hind limb foot splay measure of the functional observational battery. Hind limb foot splay was measured within 2 min of removal from 20 min exposures to the concentrations shown. Control values were obtained from air only tests sessions (II = 8 mice:concentration).

Exposure to either TCE or ISOPAR-ETM resulted in concentration-dependent decreases in response rate (Fig. 7, left panel: between-session exposure; right panel: within-session exposure). Using the between-session exposure protocol, both solvents produced decreases in rates of responding in the 2000-6000 ppm range. with slight but variable effects noted as low as 1000 ppm (ANOVA for TCE concentration, Fc5,35) = 38.38, P < 0.001; ANOVA for ISOPAR-ETM concentration, Fc5,3s) = 9.72, P < 0.001). It is noteworthy that concentrations above 6000 ppm of ISOPAR-ETM were not tested since one death occurred at this concentration. Response rates increased in the 5 min period of air exposure following the 20 min exposure to higher concentrations of both TCE and ISOPAR-ETM indicating

Fig. 7. Effects of ISOP.L\R-E”’ and TCE on lever pressing behavior under a fixed-ratio 20 schedule of milk reinforcement. Shown are concentration effect relationships obtained in t\*o ways. For the data shoun in the Icft panel. each concentration was evaluated in a separate tcbt session. For the data shown in the right panel, all concentrations wel-e evaluated during successive 5 min exposures within a single test session. Rates of responding data are expressed as the percentage of the 5 min air only segments that occurred at the beginning of each Icst session (11 = 8 mice/concentration).

Fig. 8. Concentration-effect curve for TCE and ISOPAR-ETM in mice trained to discriminate 1.25 g/kg ethanol from saline. Percentage of ethanol-lever responding (mean i S.E.M.) are shown as filled circles: response rates (mean + S.E.M.) are shown as empty circles. Control data for both ethanol and saline represent results from the ethanol and saline-test sessions which occurred immediately before and after solvent testing (n = 9 mice;concentrdtion).

recovery (data not shown). There was some evidence that responding recovered more rapidly after TCE than after ISOPAR-ETM.

Under the within-session exposure protocol, effects of TCE and ISOPAR-ETM were comparable to those obtained under the 20 min exposure protocol (Fig. 7, right panel). Both solvents produced concentration-dependent decreases in response rates between 4000 and 6000 ppm (ANOVA for TCE concentration, Fc5,3s) = 52.07, P < 0.001; ANOVA for ISOPAR-ETM concentration, Fc5,35) = 53.60, P < O.OOl]. As compared to the 20 min exposure, TCE had slightly greater response rate decreasing effects as illustrated by the small shift in the concentration-effect curve to the left. Overall, re- gardless of the exposure protocol, both compounds produced concentration-dependent behavioral effects over a similar range of concentrations and therefore, appear to share similar behavioral potency.

3.4. Ethanol-like discriminative stimulus c@cts

Exposure to increasing concentrations of TCE produced ethanol-like effects as demonstrated by the increase in the percentage of ethanol-lever responding (Fig. 8, left panel). Nearly complete substitution was observed at the 4000 and 8000 ppm (75’%1 ethanol-lever responding) without significant effects on rates of responding. TCE also was substituted for ethanol at 14000 ppm, but a greater than 7.5”/0 decrease in response rate was observed at this concentration. Follow- ing ISOPAR-ETM exposure, the percentage of ethanol-lever responses increased in a concentration-dependent manner, reaching a maximum of 65% at 6000 ppm (Fig. 8, right panel). The data of individual subjects reveals that, in five out of nine mice, ISOPAR- ETM produced 80% or greater ethanol-lever responding in the 2000&6000 ppm concentration range (data not shown). However, in all subjects, substitution of ISO-

PAR-E.rM for ethanol occurred at concentrations which decreased rates of responding. Thus, ISOPAR-ETM produced some ethanol-like effects in this procedure, but not as clearly as the results obtained with TCE, since ethanol-like effects of ISOPAR-ETM only occurred at concentrations that impaired performance.

ln mice exposed to 2000 ppm TCE for 4 days, termination of TCE exposure resulted in handling induced convulsions (data not shown). The withdrawal syndrome was rapid in onset; a few mice convulsed immediately upon removal from the exposure tank (0 h assessment). consistent with the rapid elimination kinet- ics of TCE. Maximum withdrawal severity (SO”/;, convulsions) occurred at 3-5 hour post-exposure. The withdrawal reaction was over within 24-48 hours.

Re-exposure to TCE during the withdrawal period reduced the frequency and severity of convulsions (Fig. 9, top left panel). The extent of the reduction of withdrawal convulsions was exposure dependent. Exposure to 2000 ppm TCE for 30 min immediately preceding the 3 h post-exposure assessment decreased the percentage of mice convulsing to 50%. with 200/o demonstrating tonic-clonic convulsions, as compared to 80 and 60X, respectively. recorded during the preceding 2 h post-exposure assessment. Immediately after the effects of this 30 min re-exposure were recorded, mice were once again placed back in the exposure chamber and exposed to the same concentration (2000 ppm), but for 60 additional min. Exposure to TCE for 60 min resulted in greater reduction in withdrawal as compared to the 30 min exposure. This is evident in the 4 h post-exposure assessment shown in Fig. 9 (top left panel), where only 20% of the mice convulsed and 10% showed tonic- clonic convulsions. In another group of mice, which were exposed twice during withdrawal to a higher concentrations of 4000 ppm TCE for 60 min each, only 10% of the mice convulsed, with none showing tonic- clonic convulsions (Fig. 9, top right panel). The withdrawal reaction reemerged in hour 5 as the acute suppression of withdrawal by TCE subsided.

Like with TCE, short exposures to ISOPAR-ETM temporarily suppressed withdrawal convulsions that were well under way after 2000 ppm x 4 days TCE exposure. The return of the withdrawn mice to the inhalation chamber for a 30 min exposure to 2000 ppm ISOPAR-ETM prior to the 3 h assessment decreased the percentage of mice convulsing to 500/o, with no mice receiving a score of 2 (Fig. 9, bottom left panel). In the same mice, after the effects of 30 min re-exposure were recorded, an additional exposure to the same concentration (2000 ppm) but for a longer duration (60 min) produced a slightly greater reduction in withdrawal convulsions (40% convulsing at the 4 h assessment).

Exposure to a higher concentration of 4000 ppm ISO- PAR- ETM for 30 min reduced convulsions to 40% but, not a substantially greater suppression than was obtained with 2000 ppm ISOPAR-ETM for 30 min, whereas mice exposed to 4000 ppm for 60 min exhibited no withdrawal convulsions (Fig. 9, lower right panel). As with acute TCE re- exposures, the withdrawal reaction returned by hours 5 and 6 as the suppression by ISOPAR-ETM subsided.

4. Discussion

The purpose of the present studies was to use a variety of animal models to compare and contrast the profile of behavioral effects of ISOPAR-ETM with the abused solvent TCE in an attempt to arrive at a prediction of the abuse potential of this new solvent. To our knowledge, this is the first attempt to present a model for preclinical evaluation of the abuse potential of novel solvents. TCE, tested here as a positive control. clearly produced a depressant-like profile of effects which is consistent with earlier reports (Evans and Balster, 1991). These effects included pentobarbital and ethanol-like observable effects on the functional observational battery (Tegeris and Balster, 1994; Bowen et al., 1996), reversible response rate decreasing effects on FR performance (Balster et al., 1982), ethanol-like discriminative stimulus effects (Rees et al., 1987a) and physical dependence (Evans and Balster, 1993).

ISOPAR-ETM produced some effects which were similar to those of TCE, and other effects which differed from those of TCE. The effective concentrations of ISOPAR-ETM reported here (2000-6000 ppm with 20 min exposure) were comparable to those of TCE used in the present set of experiments and to those reported in the literature (Evans and Balster, 1993; Bowen et al., 1996). This is the first report of an isoparaffin’s effects on responding under a FR schedule. The uniform concentration-effect curve observed for ISOPAR-ETM and TCE during 20 min exposures and within-session exposures is consistent with previous reports of TCE and other vapors with reversible depressant drug-like effects which decreased operant responding in a concentration- dependent manner (Balster et al., 1982; Moser and Balster, 1981, 1985; Moser et al., 1985a,b). However, it is worth noting that unlike the other abused vapors, ISOPAR-ETM appears to have less separation between the behaviorally effective and toxic concentrations which is evident by the fatalities that were observed at the highest concentration of ISOPAR-ETM.

Evidence for concentration-dependent generalization from ethanol was obtained for both inhaled TCE and ISOPAR-ETM. However, ISOPAR-ETM was less effective than TCE in producing ethanol-like discriminative stimulus effects. While TCE produced greater than 75O/o

R.L. Baker et al. : Drug and Alcohol Dependence 46 (1997) 125-135 133

80

60

8 .-

f

‘t? 100 s

80

60

Period of Period of Re-exposure to Exposure to 2000 ppm TCE 4000 ppm TCE

-+- % Convulsing m % with Score of Two

Period of Exposure to 2000 ppm ISOPAR ETM

0 2 4 6 8 24 48 0 2 4 6 8 24 48

Period of Exposure to 4000 ppm ISOPAR ETM

I

Hours Post-Exposure FGg. 9. Effects of 2000 and 4000 ppm TCE and ISOPAR-ET” exposure in TCE-dependent mice. Withdrawal effects are shown in mice made dependent on TCE by 4 days of exposure to 2000 ppm. Solid circles show the percentage of mice which had any handling-induced convulsive zffect. Bar graphs show percentage of mice with a score of two. No mice had a score higher than two at any time (tonic-clonic convulsion). Mice were exposed to TCE and ISOPAR-E TM for 20 or 60 min during the period of peak withdrawal effects (N = IO mice/concentration).

ethanol-lever responding at one or more concentrations in nearly all of the animals tested, ISOPAR-ETM produced a maximum mean of only 65% ethanol-lever responding at any single concentration. In addition,

effects on response rates generally required concentrations of TCE that were two to three times higher than those required for ethanol-lever responding. In general, these stimulus generalization test results are consistent

with previous research showing ethanol- (Rees et al., 1987b; Bowen and Balster, 1997) and pentobarbital-like (Rees et al.. 1985, 1987a) discriminative stimulus effects in mice with TCE and other anesthetics and abused inhalants. Conversely, the effects of ISOPAR-ETM on response rates were observed at the same concentrations required for drug-lever responding. In our experience, test compounds which produce results as clearly different as were obtained with ISOPAR-ETM and TCE often prove to produce considerably different types of intoxication in humans as well. While the present results provide some evidence for overlap in the discriminative stimulus effects of ethanol and ISOPAR-ETM, we would have to conclude that the results of the drug discrimination tests were inconclusive.

Reduction of the withdrawal effects in TCE-dependent mice was observed for both TCE and ISOPAR- Er”. Like with TCE, when mice who had been made dependent on TCE were exposed to ISOPAR-ETM during withdrawal, a reduction in the frequency and severity of handling-induced convulsions was seen. This is almost certainly due to a central nervous system effect of ISOPAR-El‘“. Previous investigations using CNS depressant drugs (barbiturates, benzodiazepines, ethanol) to establish physical dependence have found that administration of other CNS depressant drugs will maintain the dependent state and suppress the symp- toms of withdrawal (Yanagita, 1981; Woods et al., 1987). Evidence for this cross-dependence also exists for some abused solvents and several CNS depressant drugs (Evans and Balster, 1993). While the present results do not predict with certainty that ISOPAR-ETM would produce dependence itself, it does provide evidence that ISOPAR-ETM shares this pharmacological property in common with the abused depressant solvents such as TCE and toluene (Evans and Balster, 1993).

On the other hand, ISOPAR-ETM produced a number of notable differences with TCE in several of the tests conducted. In the present studies, we were unable to achieve anesthetic-like depressant effects with ISO- PAR-ETM in the static exposure chambers. With concentrations of up to 6000 ppm, ISOPAR-ETM did not produce central nervous depressant effects on the functional observational battery. ISOPAR-ETM was also inactive in the inverted screen test of motor impairment at the highest concentrations that could be tested. In fact, ISOPAR-ETM produced what appeared to be more like excitatory effects on some of the measures of the functional observational battery. At 8000 ppm, several mice died from convulsions without the production of the anesthesia-like state produced by TCE. The results of the present study are similar to those of an earlier report in which flurothyl, a fluorinated ether and pur- ported ‘excitatory’ vapor, was investigated using the functional observational battery (Bowen et al., 1996).

One possible explanation for the lack of reports con- cerning flurothyl abuse may be in part due to its excitatory and convulsant properties. The excitatory component of the effects of ISOPAR-Er”, if it also occurs in humans, could also possibly make this solvent undesirable to solvent abusers.

TCE and ISOPAR-ETM also differed in their ease and rate of volatilization. This was especially evident in the dynamic exposure conditions, where the flow rate through the bubblers containing the liquid solvents were usually about IO-fold greater for ISOPAR-ETM than for TCE (see Fig. 2). This is consistent with the 5fold lower vapor pressure for ISOPAR-ETM than for TCE. With regards to abuse potential, a solvent abuser would have a more difficult time arranging for the vaporization and delivery of ISOPAR-ETM than TCE. Whether this difference in volatility is enough to deter the abuse of ISOPAR-ETM is not known. We note that other abused solvents (e.g. toluene) are also less volatile than TCE.

Two additional aspects of our results deserve com- ment. We obtained a much poorer separation of behavioral and lethal effects for ISOPAR-ErM than for TCE. One implication for this may be that, if ISOPAR-ETM is voluntarily inhaled to produce concentrations with effects on the brain and behavior, abusers are poten- tially exposing themselves to toxic concentrations. Fur- ther research would be necessary to substantiate this. Even if ISOPAR-ETM is not abused, this observation would be of importance in situations involving spills with ISOPAR-ErM. We also found that recovery from ISOPAR-ETM exposure was less rapid than from TCE exposures. This could also have implications for acute high concentration exposures.

To our knowledge, this is the first attempt to use animal behavioral testing to evaluate the abuse potential of a new solvent. If we had obtained a pattern of acute behavioral effects with ISOPAR-ErM that was identical to that of TCE, we would have confidently predicted that ISOPAR-ETM had TCE-like abuse potential. Other abused vapors, such as toluene, ether and volatile anesthetics, do produce a pattern of effects very similar to what was shown here with TCE (Evans and Balster, 1991; Bowen and Balster, 1997). Unfortunately, we cannot say with certainty that ISOPAR-ETM produced enough similarities in effects with abused inhalants to predict whether or not it would be subject to abuse. Its ethanol-like discriminative stimulus effects and cross-dependence with TCE support a prediction of abuse potential. Its lack of acute anesthetic-like effects and effects on motor performance typical of abused inhalants would not. Its lower volatility may also miti- gate against abuse potential, Clinical experience with the actual abuse, or lack thereof, of ISOPAR-ETM-containing commercial products will be very helpful in determining the usefulness of these different tests for

abuse potential prediction. We also think it is impor- tant to point out that our results showed a poorer separation of acute behavioral effects and lethal effects for ISOPAR-ETM than for TCE, suggesting that if it were to be abused, it might prove to be very toxic to users. Despite the inconclusive results obtained with ISOPAR-ETM, we feel that this study exemplifies how one might begin to use animal tests to predict abuse potential of novel solvents, ultimately providing the scientific information needed to make informed deci- sions about reformulation of abused products.

Acknowledgements

This research project was supported by Wite-Out Office Products, Beltsville, MD. The help and inspira- tion of Mr George Korper in its initiation is gratefully acknowledged. Preparation of the report was supported by NIDA grants DA-031 12 and DA-05670 and by NIEHS training grant ES-07087. The expert technical assistance of Josiah Hamilton is also acknowledged.

References

Balster, R.L., 1987. Abuse potential evaluation of inhalants. Drug Alcohol Depend. 49, 7715.

Balster, R.L., Moser. V.C., 1987. Pentobarbital discrimination in the mouse. Drug Alcohol Depend. 7, 233-242.

Balster, R.L., Moser, V.C., Woolverton, W.L., 1982. Concurrent measurement of solvent vapor concentrations and effects on operant behavior using a dynamic exposure system. J. Pharmacol. Methods 8, 2999309.

Bowen. S.E.. Balster, R.L.. 1997. Desflurane, enflurane, isoflurane, and ether produce ethanol-like discriminative stimulus effects in mice. Pharmacol Biochem. Behav., 57, 191-198.

Bowen, SE., Wiley. J.L., Evans. E.B., Tokarz, M.E.. Balster, R.L., 1996. Functional observational battery comparing effects ethanol, I, I, 1 -trichloroethane, ether. and flurothyl. Neurotoxicol. Teratol. IX. I 9.

Creason, J.P., 1989. Data evaluation and statistical analysis of functional observational battery data using a linear models approach. J. Am. Coil. Toxicol. 8. 157-169.

Dohrn, C.S., Lichtor. J.L., Finn, R.S.. Uitvlugt, A., Coalson, D.W., Rupani, G.. de Wit, H., Zacny, J.P., 1992. Subjective and psycho- motor effects of nitrous oxide in healthy volunteers. Behev. Pharmacol. 3, 19.-30.

Evans. E.B., Balster, R.L., 1991. CNS depressant effects of volatile organic solvents. Neurosci. Biobehav. Rev. 15, 233-241.

Evans, E.B., Balster, R.L., 1993. Inhaled 1,1, I-trichloroethane produced physical dependence in mice: effects of drugs and vapors on withdrawal. J. Pharmacol. Exp. Ther. 264, 726- 733.

Goldstein, D.B.. 1972. An animal model for testing effects of drugs on alcohol reactions. J. Pharmacol. Exp. Ther. 183. 14- 22.

Moser, V.C., Balster, R.L., 1981. The effects of acute and repeated toluene exposure on operant behavior in mice. Neurobehav. Tox- icol. Teratol. 3, 471-475.

Moser, V.C., Balster, R.L., 1985. Effects of toluene, halothane and ethanol vapor on fixed-ratio performance in mice. Pharmacol. Biochem. Behav. 22, 7977802.

Moser, V.C., Coggeshall, E.M., Balster. R.L.. 1985a. Effects of xylene isomers on operant responding and motor performance in mice. Toxicol. Appl. Pharmacol. 80, 293-298.

Moser, V.C., Scimeca, J.A., Balster, R.L., 1985b. Minimal tolerance to the effects of l,l, I- trichloroethane on fixed-ratio responding in mice. Neurotoxicology 6, 35-42.

Rees, D.C., Coggeshall, E., Balster. R.L., 1985. Inhaled toluene produces pentobarbital-like discriminative stimulus effects in mice. Life Sci. 37, 1319- 1325.

Rees, D.C., Knisely, J.S., Breen, T.J., Balster, R.L.. I987a. Pentobar- bital-like discriminative stimulus properties of halothane, I ,I. I - trichloroethane, isoamyl nitrite, flurothyl and oxazepam in mice. J. Pharmacol. Exp. Ther. 241, 507-515.

Rees, D.C., Knisely, J.S., Breen, T.J., Balster, R.L.. 1987b. Toluene, halothane, 1, I, 1 -trichloroethane and oxazepam produce ethanol- like discriminative stimulus effects in mice. J. Pharmacol. Exp. Ther. 243, 931-937.

Sharp, C. 1992. Introduction to inhalant abuse. In: Sharp, C., Beau- vais, F., Spence, R. (Eds.), Inhalant Abuse: A Volatile Research Agenda. National Institute on Drug Abuse Research Monograph, 129. DHHS, Rockville, MD, pp. 13-42.

Tegeris, J.S., Balster, R.L., 1994. A comparison of the acute behavioral effects of alkylbenzenes using a functional observational battery in mice. Fundam. Appl. Toxicol. 22, 240 250.

Tilson, H.A., Moser, V.C., 1992. Comparison of screening ap- proaches. Neurotoxicology 13, I - 14.

Woods, J.H., Katz, J.L., Winger, G., 1987. Abuse liability of benzodiazepines. Pharmacol. Rev. 39, 251.-419.

Woolverton, W.L., Balster, R.L., 1981. Behavioral and lethal effects of combinations of oral ethanol and inhaled I, 1 ,l -trichloroethane in mice. Toxicol. Appl. Pharmacol. 59, I-7.

Yanagita, T., 1981, Dependence-producing effects of anxiolytics. In: Hoffmeister, F., Stille, G. (Eds.), Handbook of Experimental Pharmacology, 55. Springer, New York. pp. 396- 406.

Documents

Evaluation of the acute behavioral effects and abuse potential of a C8–C9 isoparaffin solvent