- Email: [email protected]
Behavioral Effects Following Subacute Inhalation Exposure to m-Xylene or Trimethylbenzene in the Rat
NeuroToxicology 22 (2001) 79±89
Behavioral Effects Following Subacute Inhalation Exposure to m-Xylene or Trimethylbenzene in the Rat A Comparative Study Søawomir Gralewicz*, Dorota Wiaderna Department of Toxicology and Carcinogenesis, Nofer Institute of Occupational Medicine, Laboratory of Neurotoxicology, 8 Teresy St., P.O. Box 199, 90-950 èoÂdzÂ, Poland Received 13 April 1999; accepted 22 May 2000
Abstract Trimethylbenzene (TMB), like xylene (dimethylbenzene), is a signi®cant constituent of some industrial solvent mixtures. In earlier studies, we found that in the rat a subacute low-level inhalation exposure to some of the TMB isomers may result in behavioral alterations detectable weeks after the exposure [Neurotoxicol Teratol 19;1997:327; Int J Occup Med Environ Health 11;1998:319]. The purpose of the present study was to compare m-xylene (XYL) and each of the TMB isomers: 1,2,3-TMB (hemimellitene Ð HM), 1,2,4-TMB (pseudocumene Ð PS), and 1,3,5-TMB (mesitylene Ð MES) with respect to the ability for inducing behavioral effects in the rat. The rats (10±11 animals per group) were exposed repeatedly for 4 weeks (6 h per day, 5 days per week) to XYL (XYL group), HM (HM group), PS (PS group) or MES (MES group) at 100 ppm, or sham exposed (C group) in 1.3 cu/m dynamic inhalation chambers. Starting 2 weeks after exposure the following forms of rat's behavior were assessed: radial maze performance, spontaneous activity in an open ®eld, learning and retention of passive and active (two-way) avoidance response, and heat-induced paw licking before and after a 2 min footshock (a test for assessment of the stress response). None of the solvent-exposed groups differed considerably from the control one with respect to the radial maze performance. Compared to control rats, the rats of the XYL, PS and MES groups, but not those of HM group, showed a signi®cantly higher spontaneous locomotor activity in the open ®eld, an impaired passive avoidance learning and signi®cantly longer paw-lick latencies 24 h after footshock. Acquisition, but not retention, of the two-way active avoidance response was signi®cantly impaired in all solvent-exposed groups. The XYL group did not differ signi®cantly from PS, MES or HM group in any of the behavioral parameters. The above results show that a short-term exposure to any of the TMB isomers or m-xylene at concentration as low as 100 ppm may induce persistent behavioral alterations in the rat. # 2001 Elsevier Science Inc. All rights reserved.
Keywords: Trimethylbenzene; Xylene; Inhalation exposure; Behavior; Rat
INTRODUCTION Trimethylbenzene (TMB) is a component of solvent mixtures used in the paint and lacquer industry. In some petroleum products like Solvesso 100 (Exon Chemical, Belgium), Shellsol A (Shell Netherland Chemie B.V., Netherlands), Jolasol (J.L.C. Chemie, Austria), and *
Corresponding author. Tel.: 48-42-6314-664. E-mail address: [email protected] (S. Gralewicz).
Farbasol (Polifarb Cieszyn S.A., Poland), the joint contribution of the three TMB isomers: 1,2,4-TMB (pseudocumene Ð PS), 1,2,3-TMB (hemimellitene Ð HM), and 1,3,5-TMB (mesitylene Ð MES), may exceed 40%. TMB is also a countable constituent of motor fuels (Wesoøowski and Gromiec, 1996, 1997). It is recognized that occupational exposure to solvent vapors may result in persisting neurobehavioral disturbances in the exposed workers (Ikeda, 1992; Stollery, 1992; Baker, 1994; Mikkelsen, 1997). Little is
0161-813X/01/$ ± see front matter # 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 6 1 - 8 1 3 X ( 0 0 ) 0 0 0 0 3 - 6
80
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
known, however, about the TMB ability to produce such health effects. The toxicological database for this hydrocarbon includes just a few reports of epidemiological and experimental studies. In the epidemiological studies, transient disturbances of the respiratory, erythropoietic and nervous systems were found in workers exposed for years to 10±60 ppm of solvent vapor containing 80% of TMB, mainly 1,2,4-TMB (Battig et al., 1956, 1958). In an early experimental study no overt adverse effects were noted in rats exposed by inhalation to PS, 8 h per day for 14 days, at concentrations up to 2000 ppm (Cameron et al., 1938). Based mainly on the epidemiologic data, in US the TLV-TWA concentration for TMB has been set at 125 mg/m3 or 25 ppm (ACGIH, 1996). In Poland, the MAC value adopted for TMB is 100 mg/m3 or 20 ppm. A few years ago studies were undertaken at our institute to obtain more information on TMB toxicity. The experiments performed by Korsak and RydzynÄski (1996) revealed that a 3 month intermittent (6 h per day, 5 days per week) inhalation exposure to PS, or HM at concentration of 100 or 250 ppm led to an impairment of rotarod performance which increased gradually in the course of exposure. This impairment was markedly reduced 2 weeks after the exposure, which might suggest no long-lasting effect of exposure on sensory-motor functions. In subsequent studies (Gralewicz et al., 1997) more complex forms of the rat behavior were assessed starting 2 weeks after a 4-week inhalation exposure to PS at concentrations of 25, 100 or 250 ppm. In the case of 25 ppm no effect of exposure was found. However, in rats exposed to 100 or 250 ppm the behavioral testing revealed an impairment of passive avoidance and prolonged emotional response to pain. Similar results were also obtained in analogous studies on HM but in that case exposure to 25 ppm disrupted passive avoidance as effectively as exposure to 100 ppm (Wiaderna et al., 1998). These data suggested that in the rat persistent changes of the CNS functional state might result from exposure to TMB at a level approaching the hygienic standard adopted for this compound. The purpose of the experiment was to obtain more information on the TMB neurotoxic properties. Firstly, having established that even at a low exposure level TMB may cause persistent behavioral effects, we wanted to know whether TMB differ from xylene (dimethylbenzene) with respect to its potential for producing such effects. Xylene (XYL) is the most common of the industrial solvents and the primary suspect in inducing ``chronic toxic encephalopathy'' or
``organic solvent syndrome'' in the exposed workers (Arlien-Soborg, 1985; Toxicological Pro®le for Xylenes, 1995). Therefore, we considered it important to ®nd out whether exposure to TMB would be more or less effective than exposure to XYL in production of persisting behavioral changes. The major constituent of most technical XYL preparations is the meta-isomer (m-xylene), and, therefore, m-xylene was selected for the present study. Secondly, we wanted to check whether exposure to MES, could produce behavioral effects similar to those found in the studies on PS and HM. In these studies the concentration of 100 ppm appeared effective in inducing persisting behavioral effects both for PS and HM. Therefore, in the present studies each of the solvents tested was administered at this concentration. MATERIAL AND METHODS Animals Male 5-month-old Wistar rats (outbreed stock from own breeding colony) were used. Until the experiment onset the rats were housed in groups (four±six rats/ cage), and were not subjected to any experimental procedure or exposed to any chemicals. The rats were divided into ®ve groups: the XYL (XYL) group (n 11), the PS (PS) group (n 11), the MES (MES) group (n 11), the HM (HM) group (n 11) and the control (C) group (n 10). The grouping was done in such a way that the groups did not differ with respect to the mean body weight and the mean level of spontaneous locomotor activity in the open ®eld. During the experiment the rats were housed in single rat cages at 228C, with a light/dark cycle of 12/12 h (light on at 6.00 a.m.). In the home cages standard rat food pellets (Murigran) were accessible ad libitum. Access to water was limited only during the radial maze test (see below). Chemicals m-Xylene (1,3-dimethylbenzene, CAS No. 108-383) was obtained from Reachim. HM (1,2,3-trimethylbenzene, CAS No. 526-73-8), PS (1,2,4-trimethylbenzene, CAS No. 95-63-6), and MES (1,3,5trimethylbenzene, CAS No. 108-67-8) were obtained from Fluka. Conversion factor for xylenes: 1 ppm 4:35 mg/ 3 m , 1 mg=m3 0:231 ppm; conversion factor for TMBs: 1 ppm 4:92 mg/m3, 1 mg=m3 0:20 ppm.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
81
Exposure The rats were exposed to solvents in 1.3 m3 inhalation chambers for 4 weeks (6 h per day, from 8.00 a.m. to 2:00 p.m., 5 days per week). Within the chambers the rats were located in small (25 cm 20 cm 25 cm) wire-mesh cages, one rat per cage, with no sawdust. They had no access to food and water during the exposure. The chambers were designed to sustain dynamic and adjustable air¯ow (15±20 m3/h). Input air was ®ltered and conditioned to a mean temperature of 22±248C and relative humidity 30±70%. For the solvent groups the air ¯owing through the chamber contained solvent vapor (m-xylene or one of the TMB isomers) at concentration of 100 ppm (10 ppm) generated by bubbling an additional air¯ow through a ¯ask containing the test compound at 708C. Samples of chamber air were taken 20 min after the exposure onset to verify the chamber concentration and then at 30 min intervals, to monitor the exposure concentration. The samples (500 ml) were sucked in through a glass tube ®lled with charcoal (100 mg MERCK, 20±35 mesh) at the rate of 20 l/h. The solvents were desorbed from coal with 1.0 ml of CS2 for 30 min and analyzed using a Hewlett-Packard gas chromatograph with a ¯ame-ionization detector (FID) and HP-1 capillary column (30 m 0:54 mm 2:65 mm). For group C, the air ¯owing through the chamber contained no solvent vapor. The methodology was described in detail in previous studies from this laboratory (SÂwiercz et al., 1995). Body weight was determined weekly during exposure and then fortnightly until the end of the experiment.
(Woolfe and MacDonald, 1944) was performed in order to compare the groups for the decrease in responsiveness to a thermal stimulus (54:5 0:28C ) after a brief (2 min) intermittent footshock. According to Bolles and Fanselow (1980) exposure to an aversive stimulus conditions fear to the experimental context, and fear inhibits pain. Therefore, the footshockinduced increment in the latency of the unconditioned motor response on the hot-plate (i.e. paw licking), commonly regarded as an index of the stress related analgesia, may also be regarded as a rough measure of the emotional fear response. Like in studies on PS the behavioral testing was started on day 14 after exposure cessation. The sequence of the behavioral testing was such that the less stressful tests were carried out ®rst and were followed by more stressful ones.
Behavioral Testing
Radial Maze The radial maze consisted of a circular 30 cm diameter arena, from which eight arms, each 60 cm long, 12 cm wide and 4 cm high, radiated at equal angles. Each arm was equipped with a 0.5 ml plastic container fastened to the ¯oor near the back wall. The maze was located in the center of the room, 80 cm above the ¯oor level. The testing consisted of two stages: the preexposure stage and the postexposure stage. The preexposure stage (adaptation) consisted of ®ve trials (one trial per day). The rats were tested individually. During each trial of the adaptation stage the rats were allowed to explore the maze for 3 min daily. They were not water deprived before the trials and the arms contained no water. During the training stage the rats were trained (one trial per day) in seeking water in the arms of the maze. Two days before the ®rst training trial and during the 5 day training stage the rats had access to water in their home cages for 30 min a day (morning hours)
The battery of the behavioral tests used for evaluation of the effects of exposure was the same as that employed in earlier studies on PS (Gralewicz et al., 1997). Epidemiologic data suggest that exposure to solvents affects cognitive (memory) and emotional functions (Linz et al., 1986; Stollery, 1992; Mikkelsen, 1997). Therefore, the battery included tests to assess short-term memory, long-term memory, the magnitude of the stress response and a test to assess spontaneous locomotor activity. To evaluate the effect of exposure on short-term (spatial) memory, choice accuracy in a radial arm maze was tested (Olton and Samuelson, 1976; Innis and Macgillivray, 1987). The effect on spontaneous activity was evaluated by an open-®eld test. The in¯uence on long-term memory and learning ability was assessed by the conditioned passive and active avoidance tests (Olton, 1973). The hot-plate test
1. Radial maze Ð a week before exposure and days 14±18 after exposure. 2. Open-field activity Ð day 8 before the exposure and day 25 after exposure. 3. Passive avoidance Ð days 39±48 after exposure. 4. Hot-plate test Ð days 50 and 51 after exposure. 5. Active avoidance Ð days 54 and 60 after exposure. The time intervals between successive tests were set arbitrarily. In each case the testing was performed between 8.00 a.m. and 3.00 p.m. by experienced experimenter not aware of the group assignment of the rats. Apparatus and Testing Procedures
82
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
only. Before each training trial the arms were wiped with wet cloth and the containers were ®lled with water. Then, the rat was placed in the arena and allowed to freely enter the arms and empty the containers. The trial ended after eight choices had been made or 5 min had elapsed. The number of arms omitted (omission errors), the number of reentries (perseveration errors), and the trial duration were recorded. The remaining details of the procedure were as described earlier (Gralewicz et al., 1997). Open Field The ``open ®eld'' was a square (100 cm 100 cm) arena surrounded by 20 cm high walls and separated into 49 equal squares. Each rat was placed in the middle of the arena and observed for 5 min. The number of square borders crossed (locomotor activity), the number of rearings (exploratory activity), and the number of grooming episodes were recorded manually by the experimenter. Passive Avoidance The passive avoidance situation was a cage (80 cm 30 cm 30 cm) with transparent roof and front, and a metal grid ¯oor. A cuboidal 22 cm long, 12 cm high and 7 cm wide platform, made of hard cardboard, was fastened to the ¯oor in the central position. The test consisted of two preshock trials (trials 1 and 2), one shock trial (trial 3) and three postshock trials (trials 4, 5 and 6). Trials 1, 2 and 3 were performed at 24 h intervals. Trials 4, 5 and 6 were performed 24 h, 3 and 7 days, respectively, following trial 3. In preshock trials the rat was placed on the top of the platform and the latency to step down onto the ¯oor was measured. After stepping down, the rat was allowed to explore the inside of the box for 1 min. In trial 3, immediately after stepping down, the rat received a series of electric footshocks (100 ms 2.0 mA 1 Hz rectangular pulses) for 10 s. In trials 4, 5 and 6, before returning it to the home cage, the rat was allowed to remain on the platform for 3 min or, in case it stepped down before 3 min had elapsed, to stay on the ¯oor for 1 min. Hot Plate The apparatus consisted of a ¯at metal box ®lled with water (hot plate), a thermostat, and a shock chamber (a 30 cm 40 cm 30 cm compartment with electri®ed grid ¯oor). It was described in detail in an earlier report (Gralewicz and SocÂko, 1990). The temperature of the hot plate was kept at 54.58C (0.28). The testing consisted of three trials, a preshock trial
(trial 1) and two postshock trials (trials 2 and 3). Each trial involved placing the rat on the hot plate within a plastic enclosure where it remained until it responded as expected, i.e. licked the hind foot, or 60 s had elapsed. Then, it was removed from the enclosure, which ended the trial. Right after trial 1 the rat was transferred to the shock chamber where it received electric footshocks (100 ms, 2.0 mA), one every 2 s for 2 min. Trial 2 was performed several seconds after the shocking. Trial 3 was performed 24 h after trial 2. The latencies of the paw-lick response in trials 1, 2 and 3 were measured and denoted as L1, L2 and L3, respectively. Active Avoidance The test enclosure consisted of a box with a metal grid ¯oor. The box was divided into two identical compartments (40:0 cm 30:0 cm 30:0 cm each) by a partition. A rectangular 8:0 cm 8:0 cm opening was provided in the lower middle part of the partition. The conditioned stimulus (CS) was a 500 Hz pulsing tone from a miniature loud speaker located at the back wall of each compartment. The unconditioned stimulus (UCS) was a series of electric footshocks (100 ms 2.0 mA 1.0 Hz rectangular pulses) applied through the grid ¯oor. The rats were trained in avoiding the UCS by moving from one compartment to another after the presentation of the CS in the compartment occupied by the rat. The CS±UCS interval was 5 s. The CS and UCS terminated immediately after the rat's entry into the second compartment. The inter-trial interval varied from 20±40 s (30 s average). The testing comprised two sessions: an acquisition session and a retention session separated by a 7-day interval. During each of the sessions the rats were trained until they reached an avoidance criterion which consisted of four avoidance responses during ®ve successive trials. Statistics In the case of the open-®eld data and the activeavoidance data differences between groups were estimated with the Kruskall±Wallis non-parametric oneway ANOVA. Multiple comparisons were made with the Sheffe's test (Siegel, 1956). Data from the remaining tests were analyzed using repeated measures twofactor ANOVA (groups trials or measurements). In case of non-homogeneity of covariances, an approximation procedure, which enables to avoid assumption about equal covariances, was applied. In this procedure the degrees of freedom used in ®nding the critical values are reduced (see Winer, 1962, p. 305±6). When-
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
83
ever a signi®cant interaction between the main factors was found, Tukey posthoc comparisons were applied (Winer, 1962). RESULTS Body Weight At the onset of the experiment, the mean body weight was the lowest in the HM group (395.9 g, S:D: 25:4), and the highest in the MES group (422.7 g, S:D: 25:2). In all groups, body weight increased gradually during the experiment (effect of measurements: F(1, 49)103.73, P < 0:0001), but the effect of the group factor as well as the groups measurements interaction was not signi®cant (results not shown). Behavioral Effects of Exposure Radial Maze Performance The analysis (two-way ANOVA, groups days) of the data obtained during the adaptation period (before the exposure) revealed no signi®cant differences between groups in the number of arm entries during successive trials. After the exposure the groups did not differ with respect to the number of perseveration or omission errors in successive days of testing (Fig. 1), as well as in the trial duration (data not shown). Open Field Behavior As regards the open ®eld behavior, the differences between groups attained statistical signi®cance (oneway non-parametric Kruskall±Wallis ANOVA: X 2 15:09, df 4, P < 0:005) only with respect to locomotor (ambulatory) activity. Detailed comparisons have shown that rats of the PS group and the MES group made signi®cantly more border crossings during the 5 min observation period than rats of the C group and the HM group, but they did not differ in this respect from the XYL group (Fig. 2). Passive Avoidance There were large individual differences within each group in the step-down latency. Statistical comparisons of the data from all animals revealed differences between groups in trial 6, i.e. on day 7 after the footshock (F 4; 282 2:86, P < 0:05); in the MES group the step-down latencies were signi®cantly shorter than in the C group. In order to reduce the within-group variability, data on two rats, with the
Fig. 1. Radial maze performance of rats exposed for 4 weeks to m-xylene or a TMB isomer at concentration of 100 ppm. The test (one trial a day) was performed on days 14±18 after exposure. The diagrams illustrate the number of perseveration (upper diagram) and omission (lower diagram) errors in successive daily trials. Denotation of groups: control Ð sham-exposed group (n 10), XYL Ð m-xylene group (n 11), PS Ð PS group (n 11), MES Ð mesitylene group (n 11), HM Ð hemimellitene group (n 11). The bars represent group means and S.E.
lowest and the highest mean step-down latency in the ®rst postshock trial, were excluded from data sets for each group of rats. The following analysis revealed a signi®cant effect of trials (F 1; 39 41:97, P < 0:0001) and a signi®cant groups trials interaction (F 4; 39 22:85, P < 0:0001). In all groups, in the preshock trials the step-down latency decreased gradually and in the solvent-exposed groups this decrease was evidently steeper than in the control group (see Fig. 3). However, as the results of statistical analysis have shown, in all preshock trials as well as in
84
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
Fig. 2. A comparison of open-field locomotor activity in shamexposed and solvent exposed rats. The test was performed on day 25 after a 4-week exposure to m-xylene or a TMB isomer at concentration of 100 ppm. Denotation of groups as in Fig. 1. The bars represent group means and S.E.
the ®rst postshock trial (trial 4) the differences between groups were not signi®cant. They appeared to be signi®cant in trial 5 (F 4; 234 3:77, P < 0:01) and in trial 6 (F 4; 234 4:49, P < 0:002). In trial 5, rats of the MES group, and in trial 6 rats of the MES, PS and XYL groups stepped down onto the ¯oor after a signi®cantly shorter time of staying on the platform than did the rats of the C group.
The Hot-plate Behavior The analysis of the paw-lick latencies (two-way ANOVA, groups trials) revealed some general differences between groups (effect of groups: F 4; 49 3:88, P < 0:01). The effect of trials as well as the groups trials interaction were also signi®cant (effect of trials: F 1; 49 61:84, P < 0:0001; interaction: F 4; 49 33:40, P < 0:0001). Consequent comparisons within groups showed signi®cant differences between trials in group C (F 2:98 21:16, P < 0:0001); in group XYL (F 2; 98 16:74, P < 0:0001); in group PS (F 2; 98 10:86, P < 0:0001); in group MES (F 2; 98 9:90, P < 0:0002), and in group HM (F 2; 98 11:53, P < 0:0001). In all groups, L2 was signi®cantly longer than L1. In XYL and MES groups, L3 was also signi®cantly longer than L1. Comparisons within trials revealed no signi®cant differences between groups in trial 1 and trial 2. They appeared in trial 3 (F 4; 147 5; 24, P < 0:001); in this trial, the paw-lick latencies measured in the XYL, PS and MES group were signi®cantly longer than in the C group (Fig. 4). Active Avoidance The analysis (non-parametric one-way Kruskall± Wallis ANOVA) of the data obtained during the training session (number of trials to the avoidance criterion)
Fig. 3. Diagrams illustrating the effect of a four-week inhalation exposure to m-xylene or a TMB isomer at concentration of 100 ppm on the step-down response latency in the passive avoidance test. The test was performed on days 39±48 after exposure. Trials 1, 2 and 3 were performed at 24 h intervals. The step-down response was punished by a 10 s footshock in trial 3 only. Trials 4, 5 and 6 were performed 24 h, 3 days and 7 days after trial 3, respectively. The maximum time of staying of the platform was 180 s. Denotations of groups as in Fig. 1. The bars represent group means and S.E.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
85
Fig. 4. A comparison of sham exposed and solvent exposed rats with respect to the latency of the paw-lick response to heat (54.58C) before (L1), several seconds after (L2) and 24 h after a 2 min intermittent footshock. The test was performed on day 50 and 51 after a 4-week inhalation exposure to m-xylene or a TMB isomer at concentration of 100 ppm. Denotation of groups as in the figure. The bars represent group means and S.E.
showed signi®cant differences between groups (X 2 13:19, df 4, P < 0:02). In all solvent exposed groups the mean number of trials to criterion was signi®cantly higher than in the control (C) group. Among the
solvent-exposed groups, the number of trials to criterion during training was highest in the XYL group and lowest in the HM group, but the differences were not signi®cant (Fig. 5). Analysis of the data obtained
Fig. 5. Active avoidance learning in rats after a 4-week inhalation exposure to to m-xylene or a TMB isomer at concentration of 100 ppm. In one massed-trial session (inter-trial interval Ð 20±40 s; maximum number of trials Ð 60) the rats learned to shuttle between two neighboring compartments in order to avoid a footshock. The test was performed on day 54±60 after exposure. Denotation of groups as in Fig. 1. The bars represent group means and S.E. of the number of trials.
86
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
during retraining revealed no differences between groups. DISCUSSION The results of the present experiments show that in the rat a 4-week inhalation exposure to 100 ppm of mxylene or any one of the TMB isomers, while having no in¯uence on the general health status, resulted in behavioral alterations detectable weeks after exposure. Generally, in PS, MES and XYL groups, the observed changes were qualitatively similar to those found in previous experiments on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998). The rats from these groups did not differ from controls with respect to the radial maze performance. However, they showed an increased locomotor activity in the open ®eld, increased latency of the paw-lick response to heat (the hot-plate test) and impaired acquisition of passive avoidance response. All exposed groups were de®cient in learning, but not in retention of the active avoidance response in shuttle box. Taken together, these changes do not indicate cognitive impairments but rather a decreased ability to inhibit locomotor response, especially in fear-inducing environment. Such an interpretation is suggested directly by the results of the open®eld and the passive avoidance test and may be applied for explaining results of the hot-plate test and the active avoidance test. In the hot-plate test, an augmented locomotion (attempts to escape from the enclosure), competitive with the recuperative response (paw licking), might be responsible for the increase in paw-lick latency. In the active avoidance situation, ®nding the only successful escape route, (i.e. the small, halfcovered opening in the partition separating the two compartments) might be more dif®cult for rats showing a higher locomotor response to US and CS. The main question asked in the present studies concerned the relative potential of TMBs and m-xylene in inducing long-lasting neurobehavioral alterations. The literature data on XYL suggest relatively low behavioral toxicity of this solvent; usually the effects were noted at exposure concentrations much higher than 100 ppm (see Toxicological Pro®le for Xylenes, 1995). As regards TMB, some effects were noted in rats during and shortly after a subchronic inhalation exposure at concentration of 100±250 ppm (see Section 1). Moreover, in rats exposed to HM these effects were more pronounced than in rats exposed to PS (Korsak and RydzynÄski, 1996). Thus, we might have expected that: (i) at the 100 ppm exposure level, XYL
would have no behavioral consequences or the effects would be evidently weaker than those produced by exposure to TMBs; (ii) the effects of exposure to HM would be more pronounced than those of PS and presumably MES. However, the results of the present experiment did not con®rm the above. The behavioral effects of exposure to 100 ppm of XYL were no less pronounced than those seen after exposure to TMBs. MES appeared as effective as PS and XYL but the effects of HM were apparently weaker than those of PS or MES. However, taking into account the results of our previous studies on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998), the results of the present work may not necessarily denote that HM is less toxic than PS or MES. In these experiments rats were exposed to graded concentrations (0, 25, 100 or 250 ppm) of TMBs. The behavioral effects of exposure to 100 ppm of HM were similar to those produced by exposure to PS or MES in the present experiment. Impaired passive avoidance was also observed in rats exposed to 25 ppm but in rats exposed to 250 ppm none effect was detected. A similar concentration±effect relationship was noted in the experiment on PS. However, in case of PS exposure 250 ppm was effective, although less than 100, and 25 ppm appeared totally ineffective. The above observations suggest that, at least with respect to postexposure behavioral alterations, the concentration±effect curve for both, HM and PS, will have the shape of inverted ``U'' and that the HM concentration most effective in producing behavioral alterations seems to be lower that the most effective PS concentration. It is likely then, that if the concentration selected for the present experiment were less than 100 ppm, the behavioral effect of exposure to HM would be more pronounced than those of PS and possibly MES. We also presume that the discrepancies in the results of exposure to 100 ppm of HM in the present and previous experiment may have resulted from different sensitivity of examined rat populations to the solvents or from undetected differences in other experimental variables. Irrespective of the cause of the discrepancies, the results of previous and the present experiment indicate that HM differs, albeit quantitatively rather than qualitatively, from the remaining TMBs and XYL in its capacity for affecting rat behavior. A question then arises about the distinctive properties of HM that could be regarded as the possible cause of this difference. Of the four solvents used in the present study HM is distinguished by the most asymmetric and therefore the most polar structure. Moreover, it is the only compound with the methyl substituent in the ortho-con®guration.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
Both these features are likely to contribute to the HM af®nity to some biological constituents (e.g. lipids and proteins) as well as to its toxokinetics. The data obtained by Jarnberg and Johanson (1995) suggest that, compared to PS and MES, HM has a higher af®nity for water and lower af®nity for fats. The oil/ blood partition coef®cient (calculated value) of HM (164) was slightly lower than that of PS (173) and signi®cantly lower than that of MES (230) (Jarnberg and Johanson, 1995). On the other hand, in human volunteers no distinct differences between TMBs were noted in respiratory uptake and blood concentration in the course of a 2 h (Jarnberg et al., 1996) or 8 h (Kostrzewski et al., 1997) inhalation exposure. Taken together, the above observations suggest that compared to MES but not to PS, HM may be less readily redistributed from blood to body fats, and hence to the neural tissue. However, HM is metabolized an excreted at much lower rate than all the remaining solvents used in the present study. In the rat, elimination of an oral dose (1.2 g/kg) of HM was reported to take even more than ten days. The elimination of MES and PS proceeded three to four times faster (Mikulski and Wiglusz, 1975). In the experiments by Huo et al. (1989), as much as 99% of a single oral dose of PS was excreted within 24 h after exposure. In an inhalation experiment on rats exposed to solvents for 3 days (12 h per day) at concentration of 100 ppm, no detectable amounts of solvent was found 12 h after the last exposure in the brains of rats exposed to PS or XYL (Zahlsen et al., 1992) which precluded accumulation. (It is worth emphasizing, that at the end of exposure day 3, the brain concentrations of XYL and PS were similar (XYL 28:6 7:3 mmol/kg; PS 36:5 2:2 mmol/kg)). Based on the above data, it may be presumed that during the present experiment the brains of rats from the HM group might be subjected to the action of the solvent and its metabolites for a longer time and at higher concentration (due to accumulation) than the brains of rats of the XYL, PS and possibly MES group. Consequently, in the HM group the exposure induced neuroadaptative changes related to the postexposure behavioral performance might differ somehow from those in the remaining groups. Aromatic hydrocarbons are known to interact with the catecholaminergic systems and this action is regarded as speci®c for this group of organic solvents (see Kyrklund (1992) for review). The dopaminergic system appears to be the most sensitive; persistent changes suggestive of an increased sensitivity, were observed after repeated inhalation exposures to some aromatic hydrocarbons at concentrations as low as
87
80 ppm (Von Euler et al., 1993, 1994; Hillefors-Berglund et al., 1995). From the point of view of the behavioral ®ndings of our present as well as previous studies the latter observations deserve particular attention. First, it is known that pharmacological activation of the dopaminergic system may result in a disinhibition of motoric responses suppressed by punishment (see Jackson and Westlind-Danielsson (1994) for review), which conforms with the general de®cit in our solvent exposed rats. Second, the changes in the dopaminergic system resulting from a low-level solvent exposure were reported to persist weeks after exposure cessation and show a non-linear concentration±effect relationship just like the behavioral effects of exposure to PS or HM. It has been shown, for instance, that 4 weeks after a 4-week (6 h per day, 5 days per week) inhalation exposure to toluene, the af®nity of dopamine D2 agonist binding in the rat striatum is signi®cantly increased but only in rats which were exposed at concentration of 80 ppm or less. Concentrations higher than 80 ppm produced little or no effect (Hillefors-Berglund et al., 1995). If exposure to TMB isomers and XYL could affect the dopaminergic system in a similar manner as toluene does, it might account for the presence and the nature of the behavioral effects observed in our studies, as well as the non-linearity of the concentration±effect relationship noted in case of PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998). Accordingly, the lower ef®cacy of HM compared to that of PS, MES and XYL in the present experiment might indicate that in case of HM the concentration of 100 ppm was to high to induce persistent changes in dopaminergic system comparable with those produced by exposure to the other solvents. One more possible explanation of the behavioral effects of the low-level exposures to TMBs or XYL is worth considering. Namely, it may be presumed that they may have been consequences of stress responseinduced repeatedly by the daily exposures. Bushnell and Peele (1988) have found out that xylene is perceived by the rat as an aversive stimulus at concentrations much below 100 ppm. According to recent ®ndings, in the rat the odor of many solvents, including toluene and mesitylene, induces a speci®c EEG response, identical with that induced by the odor of natural rat predators (Heale et al., 1994; Zibrowski and Vanderwolf, 1997; Zibrowski et al., 1998). This would suggest that the odor of at least some solvents may be a threat signal for this species. It is known that a repeated, or even single exposure to an aversive stimulus or any other stressor may result in a persistent
88
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
sensitization of the subject to the same and other stressors (see Antelman et al., 1997), which may be manifested as an augmented locomotor response to a test stressor (Herman et al., 1984; Kalivas et al., 1986; Kalivas and Stewart, 1991). However, depending on whether the intensity of the sensitizing stimulus is low or high the ensuing change may consist in an increase or decrease in the behavioral response to a subsequent administration of a stressor (Antelman et al., 1991). In the light of the above, the behavioral effect of exposure to solvents in our experiments, may be regarded as a consequence of rat sensitization to aversive events by repeated exposure to the aversive stimulus: solvent odor. The smaller range of behavioral effects in the HM group, compared to the remaining groups, might indicate that at 100 ppm the HM odor was more aversive for the rat than the odor of the remaining solvents, and therefore in the HM group the activating effect on locomotion was less pronounced . The above hypothesis is not inconsistent with that linking the behavioral effects of low-level solvent exposure with changes in the dopaminergic system. Involvement of the dopaminergic system in stress response is well documented. (Finlay and Zigmond, 1997; Pani et al., 2000). Therefore, it cannot be excluded that the changes within this system observed after exposure to solvents (see above) were Ð at least in some instances Ð related more to the exposure induced stress response than to a direct action of the solvent on the dopaminergic neurons. The above assumption, relating the effects of lowlevel solvent exposure to stress may also appear as relevant for explaining some effects of solvent exposure in humans (Otto et al., 1992) and, therefore, deserves further testing. Summing up, results of the present study demonstrate that long-term behavioral alterations may follow a 4-week inhalation exposure to m-xylene, or any of the TMB isomers at concentration as low as 100 ppm. The character of these alterations suggests a decreased ability to inhibit the locomotor response especially in a stress-inducing experimental context. m-Xylene does not differ reliably from TMB isomers with respect to the type and magnitude of the behavioral effects. Among the TMB isomers, at the concentration used in this study, HM appears to be less potent in inducing behavioral alterations than PS or MES. However, considering the apparent non-linearity of the concentration±effect function found in earlier studies on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998), the latter conclusion must be limited to the 100 ppm concentration only.
ACKNOWLEDGEMENTS This work was supported by Grant PB 1093/PO5/97/ 13 from the Polish Committee for Scienti®c Research. The skilful work of Wanda Majcherek (exposure level setting and control) is highly appreciated.
REFERENCES American Conference of Governmental Industrial Hygienists. Threshold limit values for chemical substances and physical agents. Biological Exposure Indices, 1996. Antelman SM, Caggiula AR, Kocan D, Knopf S, Meyer D, Edwards DJ, Barry H. One experience with ``lower'' or ``higher'' intensity stressors, respectively enhances or diminishes responsiveness to haloperidol weeks later: implications for understanding drug variability. Brain Res 1991;566:276±83. Antelman SM, Soares JC, Gershon S. Time-dependent sensitization Ð possible implications for clinical pharmacology. Behav Pharmacol 1997;8:505±14. Arlien-Soborg P. Chronic effects of organic solvents on the central nervous system and diagnostic criteria. Organic solvents and the central nervous system. Copenhagen: World Health Organization, 1985. p. 197±218. Baker E. A review of recent research on health effects of human occupational exposure to organic solvents. JOM 1994;36:1079± 91. Battig K, Grandjean E, Turian V. Gesundheitschaden nach langdauernder trimethylbenzol exposition in einer malerwerkstatt. Z Praeventiv Med 1956;1:389±404. Battig K, Grandjean E, Rossi L, Rickenbacher J. Toxicologische Untersuchungen uber trimethylbenzol. Arch Gewerbepathol 1958;16:555±66. Bolles RC, Fanselow MS. A perceptual-defensive-recuperative model of fear and pain. Behav Brain Sci 1980;3:291±301. Bushnell PJ, Peele D. Conditioned flavor aversion induced by inhaled p-xylene in rats. Neurotoxicol Teratol 1988;10:273±7. Cameron GR, Paterson JLH, DeSarin GSW, Thomas JC. The toxicity of some methyl derivatives of benzene with special reference to PS and heavy coal naphtha. J Pathol Bacteriol 1938;46:95±107. Finlay JM, Zigmond MJ. The effects of stress on central dopaminergic neurons: possible clinical implications. Neurochem Res 1997;22:1387±94. Gralewicz S, SocÂko R. Effects of a single exposure to chlorphenvinphos, an organophosphate insecticide on hot-plate behaviour in rats. Pol J Occup Med 1990;3:215±20. Gralewicz S, Wiaderna D, Tomas T, RydzynÄski K. Behavioral changes following four-week inhalation exposure to (1,2,4trimethylbenzene) pseudocumene in the rat. Neurotoxicol Teratol 1997;19:327±33. Heale VR, Vanderwolf CH, Kavaliers M. Components of weasel and fox odors elicit fast wave burst in the dentate gyrus of rats. Behav Brain Res 1994;31:159±65. Herman J-P, Stinus L, le Moal M. Repeated stress increases locomotor response to amphetamine. Psychopharmacology 1984;84:431±5.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89 Hillefors-Berglund M, Liu Y, von Euler G. Persistent, specific and dose-dependent effects of toluene exposure on dopamine D2 agonist binding in the rat caudete-putamen. Toxicology 1995;100:185±94. Huo J, Aldous S, Campbell K, Davies N. Distribution and metabolism of 1,2,4-trimethylbenzene (pseudocumene) in the rat. Xenobiotica 1989;19:161±70. Ikeda M. Public health problems of organic solvents. Toxicol Lett 1992;64/65:191±201. Innis NK, Macgillivray M. Radial maze performance under food and water deprivation. Behav Proc 1987;15:167±79. Jackson DM, Westlind-Danielsson A. Dopamine receptors: molecular biology, biochemistry and behavioural aspects. Pharmacol Ther 1994;64:291±369. Jarnberg J, Johanson G. Liquid/air partition coefficients of the trimethylbenzenes. Toxicol Ind Health 1995;11:81±8. Jarnberg J, Johanson G, Lof A. Toxicokinetics of inhaled trimethylbenzenes in man. Toxicol Appl Pharmacol 1996;140:281±8. Kalivas PW, Richardson-Carlson R, Van Orden G. Crosssensitization between foot shock stress and enkephalin-induced motor activity. Biol Psychiat 1986;21:939±50. Kalivas PW, Stewart J. Dopamine transmission in the initiation and expression of drug- and stress-induced sensitization of motor activity. Brain Res Rev 1991;16:223±44. Korsak Z, RydzynÄski K. Neurotoxic effects of acute and subchronic inhalation exposure to trimethylbenzene isomers (pseudocumene, mesitylene, hemimellitene) in rats. Int J Occup Med Environ Health 1996;9:341±9. Kostrzewski P, Wiaderna-Brycht A, Czerski B. Biological monitoring of experimental human exposure to trimethylbenzene. Sci Total Environ 1997;199:73±81. Kyrklund T. The use of experimental studies to reveal suspected neurotoxic chemicals as occupational hazards: acute and chronic exposures to organic solvents. Am J Ind Med 1992;21:15±24. Linz DH, de Garmo PL, Morton WE, Wiens AN, Coull BM, Maricle RA. Organic solvent-induced encephalopathy in industrial painters. J Occup Med 1986;28:119±26. Mikkelsen S. Epidemiologic update on solvent neurotoxicity. Environ Res 1997;73:101±12. Mikulski PI, Wiglusz R. The comparative metabolism of mesitylene, pseudocumene, and hemimellitene in rats. Toxicol Appl Pharmacol 1975;31:21±31. Otto D, Molhave L, Rose G, Hudnell HK, House D. Neurobehavioral and sensory irritant effects of controlled exposure to a complex mixture of volatile organic solvents. Neurotoxicol Teratol 1992;12:649±52. Olton DS. Shock-motivated avoidance and the analysis of behavior. Psychol Bull 1973;79:243±51.
89
Olton DS, Samuelson RJ. Remembrance of places passed: spatial memory in rats. J Exp Psychol 1976;2:97±116. Pani L, Porcella A, Gessa GL. The role of stress in the pathophysiology of the dopaminergic system. Mol Psychatr 2000;5:14±21. Siegel S. Nonparametric statistics for the behavioral science. New York: McGraw-Hill, 1956. Stollery B., Organic solvents. In: Smith AP, Jones DM, editors. Handbook of human performance. The physical environment, vol. 1. London: Academic Press, 1992. p. 149±175. SÂwiercz R, Korsak Z, RydzynÄski K. Kinetics of n-butyl alcohol and m-xylene in blood during single and combined inhalation exposure in rats. Int J Occup Med Environ Health 1995;8: 361±5. Toxicological Profile for Xylenes. US Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, August 1995. Von Euler G, Ogren S-O, Li XM, Fuxe K, Gustafsson J-A. Persistent effects of subchronic toluene exposure on spatial learning and memory, dopamine-mediated locomotor activity and dopamine D2 agonist binding in the rat. Toxicology 1993;77:223±32. Von Euler G, Ogren S-O, Li XM, Fuxe K, Gustafsson J-A. Persistent effects of 80 ppm toluene on dopamine-regulated locomotor activity and prolactin secretion in the male rat. Neurotoxicology 1994;15:621±4. Wesoøowski W, Gromiec JP. Prediction of exposure to engine fuel vapours. Med Pracy 1996;67:20±9. Wesoøowski W, Gromiec JP. Occupational exposure in Polish paint and lacquer industry. Int J Occup Med Environ Health 1997;10:71±81. Wiaderna D, Gralewicz S, Tomas T. Behavioral changes following four-week inhalation exposure to hemimellitene (1,2,3-trimethylbenzene) in the rat. Int J Occup Med Environ Health 1998;11:319±34. Winer BJ. Statistical principles in experimental design. New York: McGraw Hill, 1962. Woolfe G, MacDonald A. The evaluation of the analgesic action of pethidine hydrochlorine demerol. J Pharmacol Exp Ther 1944;80:300±7. Zahlsen K, Eide I, Nilsen AM, Nilsen OG. Inhalation kinetics of C6 to C10 aliphatic, aromatic and naphtenic hydrocarbons in rat after repeated exposure. Pharmacol Toxicol 1992;71:144±9. Zibrowski EM, Hoh TE, Vanderwolf CH. Fast wave activity in the rat rhinencephalon: elicitation by the odors of phytochemicals, organic solvents, and a rodent predator. Brain Res 1998;800:207±15. Zibrowski EM, Vanderwolf CH. Oscillatory fast wave activity in the rat pyriform cortex: relations toolfaction and behavior. Brain Res 1997;766:39±49.
Behavioral Effects Following Subacute Inhalation Exposure to m-Xylene or Trimethylbenzene in the Rat A Comparative Study Søawomir Gralewicz*, Dorota Wiaderna Department of Toxicology and Carcinogenesis, Nofer Institute of Occupational Medicine, Laboratory of Neurotoxicology, 8 Teresy St., P.O. Box 199, 90-950 èoÂdzÂ, Poland Received 13 April 1999; accepted 22 May 2000
Abstract Trimethylbenzene (TMB), like xylene (dimethylbenzene), is a signi®cant constituent of some industrial solvent mixtures. In earlier studies, we found that in the rat a subacute low-level inhalation exposure to some of the TMB isomers may result in behavioral alterations detectable weeks after the exposure [Neurotoxicol Teratol 19;1997:327; Int J Occup Med Environ Health 11;1998:319]. The purpose of the present study was to compare m-xylene (XYL) and each of the TMB isomers: 1,2,3-TMB (hemimellitene Ð HM), 1,2,4-TMB (pseudocumene Ð PS), and 1,3,5-TMB (mesitylene Ð MES) with respect to the ability for inducing behavioral effects in the rat. The rats (10±11 animals per group) were exposed repeatedly for 4 weeks (6 h per day, 5 days per week) to XYL (XYL group), HM (HM group), PS (PS group) or MES (MES group) at 100 ppm, or sham exposed (C group) in 1.3 cu/m dynamic inhalation chambers. Starting 2 weeks after exposure the following forms of rat's behavior were assessed: radial maze performance, spontaneous activity in an open ®eld, learning and retention of passive and active (two-way) avoidance response, and heat-induced paw licking before and after a 2 min footshock (a test for assessment of the stress response). None of the solvent-exposed groups differed considerably from the control one with respect to the radial maze performance. Compared to control rats, the rats of the XYL, PS and MES groups, but not those of HM group, showed a signi®cantly higher spontaneous locomotor activity in the open ®eld, an impaired passive avoidance learning and signi®cantly longer paw-lick latencies 24 h after footshock. Acquisition, but not retention, of the two-way active avoidance response was signi®cantly impaired in all solvent-exposed groups. The XYL group did not differ signi®cantly from PS, MES or HM group in any of the behavioral parameters. The above results show that a short-term exposure to any of the TMB isomers or m-xylene at concentration as low as 100 ppm may induce persistent behavioral alterations in the rat. # 2001 Elsevier Science Inc. All rights reserved.
Keywords: Trimethylbenzene; Xylene; Inhalation exposure; Behavior; Rat
INTRODUCTION Trimethylbenzene (TMB) is a component of solvent mixtures used in the paint and lacquer industry. In some petroleum products like Solvesso 100 (Exon Chemical, Belgium), Shellsol A (Shell Netherland Chemie B.V., Netherlands), Jolasol (J.L.C. Chemie, Austria), and *
Corresponding author. Tel.: 48-42-6314-664. E-mail address: [email protected] (S. Gralewicz).
Farbasol (Polifarb Cieszyn S.A., Poland), the joint contribution of the three TMB isomers: 1,2,4-TMB (pseudocumene Ð PS), 1,2,3-TMB (hemimellitene Ð HM), and 1,3,5-TMB (mesitylene Ð MES), may exceed 40%. TMB is also a countable constituent of motor fuels (Wesoøowski and Gromiec, 1996, 1997). It is recognized that occupational exposure to solvent vapors may result in persisting neurobehavioral disturbances in the exposed workers (Ikeda, 1992; Stollery, 1992; Baker, 1994; Mikkelsen, 1997). Little is
0161-813X/01/$ ± see front matter # 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 6 1 - 8 1 3 X ( 0 0 ) 0 0 0 0 3 - 6
80
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
known, however, about the TMB ability to produce such health effects. The toxicological database for this hydrocarbon includes just a few reports of epidemiological and experimental studies. In the epidemiological studies, transient disturbances of the respiratory, erythropoietic and nervous systems were found in workers exposed for years to 10±60 ppm of solvent vapor containing 80% of TMB, mainly 1,2,4-TMB (Battig et al., 1956, 1958). In an early experimental study no overt adverse effects were noted in rats exposed by inhalation to PS, 8 h per day for 14 days, at concentrations up to 2000 ppm (Cameron et al., 1938). Based mainly on the epidemiologic data, in US the TLV-TWA concentration for TMB has been set at 125 mg/m3 or 25 ppm (ACGIH, 1996). In Poland, the MAC value adopted for TMB is 100 mg/m3 or 20 ppm. A few years ago studies were undertaken at our institute to obtain more information on TMB toxicity. The experiments performed by Korsak and RydzynÄski (1996) revealed that a 3 month intermittent (6 h per day, 5 days per week) inhalation exposure to PS, or HM at concentration of 100 or 250 ppm led to an impairment of rotarod performance which increased gradually in the course of exposure. This impairment was markedly reduced 2 weeks after the exposure, which might suggest no long-lasting effect of exposure on sensory-motor functions. In subsequent studies (Gralewicz et al., 1997) more complex forms of the rat behavior were assessed starting 2 weeks after a 4-week inhalation exposure to PS at concentrations of 25, 100 or 250 ppm. In the case of 25 ppm no effect of exposure was found. However, in rats exposed to 100 or 250 ppm the behavioral testing revealed an impairment of passive avoidance and prolonged emotional response to pain. Similar results were also obtained in analogous studies on HM but in that case exposure to 25 ppm disrupted passive avoidance as effectively as exposure to 100 ppm (Wiaderna et al., 1998). These data suggested that in the rat persistent changes of the CNS functional state might result from exposure to TMB at a level approaching the hygienic standard adopted for this compound. The purpose of the experiment was to obtain more information on the TMB neurotoxic properties. Firstly, having established that even at a low exposure level TMB may cause persistent behavioral effects, we wanted to know whether TMB differ from xylene (dimethylbenzene) with respect to its potential for producing such effects. Xylene (XYL) is the most common of the industrial solvents and the primary suspect in inducing ``chronic toxic encephalopathy'' or
``organic solvent syndrome'' in the exposed workers (Arlien-Soborg, 1985; Toxicological Pro®le for Xylenes, 1995). Therefore, we considered it important to ®nd out whether exposure to TMB would be more or less effective than exposure to XYL in production of persisting behavioral changes. The major constituent of most technical XYL preparations is the meta-isomer (m-xylene), and, therefore, m-xylene was selected for the present study. Secondly, we wanted to check whether exposure to MES, could produce behavioral effects similar to those found in the studies on PS and HM. In these studies the concentration of 100 ppm appeared effective in inducing persisting behavioral effects both for PS and HM. Therefore, in the present studies each of the solvents tested was administered at this concentration. MATERIAL AND METHODS Animals Male 5-month-old Wistar rats (outbreed stock from own breeding colony) were used. Until the experiment onset the rats were housed in groups (four±six rats/ cage), and were not subjected to any experimental procedure or exposed to any chemicals. The rats were divided into ®ve groups: the XYL (XYL) group (n 11), the PS (PS) group (n 11), the MES (MES) group (n 11), the HM (HM) group (n 11) and the control (C) group (n 10). The grouping was done in such a way that the groups did not differ with respect to the mean body weight and the mean level of spontaneous locomotor activity in the open ®eld. During the experiment the rats were housed in single rat cages at 228C, with a light/dark cycle of 12/12 h (light on at 6.00 a.m.). In the home cages standard rat food pellets (Murigran) were accessible ad libitum. Access to water was limited only during the radial maze test (see below). Chemicals m-Xylene (1,3-dimethylbenzene, CAS No. 108-383) was obtained from Reachim. HM (1,2,3-trimethylbenzene, CAS No. 526-73-8), PS (1,2,4-trimethylbenzene, CAS No. 95-63-6), and MES (1,3,5trimethylbenzene, CAS No. 108-67-8) were obtained from Fluka. Conversion factor for xylenes: 1 ppm 4:35 mg/ 3 m , 1 mg=m3 0:231 ppm; conversion factor for TMBs: 1 ppm 4:92 mg/m3, 1 mg=m3 0:20 ppm.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
81
Exposure The rats were exposed to solvents in 1.3 m3 inhalation chambers for 4 weeks (6 h per day, from 8.00 a.m. to 2:00 p.m., 5 days per week). Within the chambers the rats were located in small (25 cm 20 cm 25 cm) wire-mesh cages, one rat per cage, with no sawdust. They had no access to food and water during the exposure. The chambers were designed to sustain dynamic and adjustable air¯ow (15±20 m3/h). Input air was ®ltered and conditioned to a mean temperature of 22±248C and relative humidity 30±70%. For the solvent groups the air ¯owing through the chamber contained solvent vapor (m-xylene or one of the TMB isomers) at concentration of 100 ppm (10 ppm) generated by bubbling an additional air¯ow through a ¯ask containing the test compound at 708C. Samples of chamber air were taken 20 min after the exposure onset to verify the chamber concentration and then at 30 min intervals, to monitor the exposure concentration. The samples (500 ml) were sucked in through a glass tube ®lled with charcoal (100 mg MERCK, 20±35 mesh) at the rate of 20 l/h. The solvents were desorbed from coal with 1.0 ml of CS2 for 30 min and analyzed using a Hewlett-Packard gas chromatograph with a ¯ame-ionization detector (FID) and HP-1 capillary column (30 m 0:54 mm 2:65 mm). For group C, the air ¯owing through the chamber contained no solvent vapor. The methodology was described in detail in previous studies from this laboratory (SÂwiercz et al., 1995). Body weight was determined weekly during exposure and then fortnightly until the end of the experiment.
(Woolfe and MacDonald, 1944) was performed in order to compare the groups for the decrease in responsiveness to a thermal stimulus (54:5 0:28C ) after a brief (2 min) intermittent footshock. According to Bolles and Fanselow (1980) exposure to an aversive stimulus conditions fear to the experimental context, and fear inhibits pain. Therefore, the footshockinduced increment in the latency of the unconditioned motor response on the hot-plate (i.e. paw licking), commonly regarded as an index of the stress related analgesia, may also be regarded as a rough measure of the emotional fear response. Like in studies on PS the behavioral testing was started on day 14 after exposure cessation. The sequence of the behavioral testing was such that the less stressful tests were carried out ®rst and were followed by more stressful ones.
Behavioral Testing
Radial Maze The radial maze consisted of a circular 30 cm diameter arena, from which eight arms, each 60 cm long, 12 cm wide and 4 cm high, radiated at equal angles. Each arm was equipped with a 0.5 ml plastic container fastened to the ¯oor near the back wall. The maze was located in the center of the room, 80 cm above the ¯oor level. The testing consisted of two stages: the preexposure stage and the postexposure stage. The preexposure stage (adaptation) consisted of ®ve trials (one trial per day). The rats were tested individually. During each trial of the adaptation stage the rats were allowed to explore the maze for 3 min daily. They were not water deprived before the trials and the arms contained no water. During the training stage the rats were trained (one trial per day) in seeking water in the arms of the maze. Two days before the ®rst training trial and during the 5 day training stage the rats had access to water in their home cages for 30 min a day (morning hours)
The battery of the behavioral tests used for evaluation of the effects of exposure was the same as that employed in earlier studies on PS (Gralewicz et al., 1997). Epidemiologic data suggest that exposure to solvents affects cognitive (memory) and emotional functions (Linz et al., 1986; Stollery, 1992; Mikkelsen, 1997). Therefore, the battery included tests to assess short-term memory, long-term memory, the magnitude of the stress response and a test to assess spontaneous locomotor activity. To evaluate the effect of exposure on short-term (spatial) memory, choice accuracy in a radial arm maze was tested (Olton and Samuelson, 1976; Innis and Macgillivray, 1987). The effect on spontaneous activity was evaluated by an open-®eld test. The in¯uence on long-term memory and learning ability was assessed by the conditioned passive and active avoidance tests (Olton, 1973). The hot-plate test
1. Radial maze Ð a week before exposure and days 14±18 after exposure. 2. Open-field activity Ð day 8 before the exposure and day 25 after exposure. 3. Passive avoidance Ð days 39±48 after exposure. 4. Hot-plate test Ð days 50 and 51 after exposure. 5. Active avoidance Ð days 54 and 60 after exposure. The time intervals between successive tests were set arbitrarily. In each case the testing was performed between 8.00 a.m. and 3.00 p.m. by experienced experimenter not aware of the group assignment of the rats. Apparatus and Testing Procedures
82
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
only. Before each training trial the arms were wiped with wet cloth and the containers were ®lled with water. Then, the rat was placed in the arena and allowed to freely enter the arms and empty the containers. The trial ended after eight choices had been made or 5 min had elapsed. The number of arms omitted (omission errors), the number of reentries (perseveration errors), and the trial duration were recorded. The remaining details of the procedure were as described earlier (Gralewicz et al., 1997). Open Field The ``open ®eld'' was a square (100 cm 100 cm) arena surrounded by 20 cm high walls and separated into 49 equal squares. Each rat was placed in the middle of the arena and observed for 5 min. The number of square borders crossed (locomotor activity), the number of rearings (exploratory activity), and the number of grooming episodes were recorded manually by the experimenter. Passive Avoidance The passive avoidance situation was a cage (80 cm 30 cm 30 cm) with transparent roof and front, and a metal grid ¯oor. A cuboidal 22 cm long, 12 cm high and 7 cm wide platform, made of hard cardboard, was fastened to the ¯oor in the central position. The test consisted of two preshock trials (trials 1 and 2), one shock trial (trial 3) and three postshock trials (trials 4, 5 and 6). Trials 1, 2 and 3 were performed at 24 h intervals. Trials 4, 5 and 6 were performed 24 h, 3 and 7 days, respectively, following trial 3. In preshock trials the rat was placed on the top of the platform and the latency to step down onto the ¯oor was measured. After stepping down, the rat was allowed to explore the inside of the box for 1 min. In trial 3, immediately after stepping down, the rat received a series of electric footshocks (100 ms 2.0 mA 1 Hz rectangular pulses) for 10 s. In trials 4, 5 and 6, before returning it to the home cage, the rat was allowed to remain on the platform for 3 min or, in case it stepped down before 3 min had elapsed, to stay on the ¯oor for 1 min. Hot Plate The apparatus consisted of a ¯at metal box ®lled with water (hot plate), a thermostat, and a shock chamber (a 30 cm 40 cm 30 cm compartment with electri®ed grid ¯oor). It was described in detail in an earlier report (Gralewicz and SocÂko, 1990). The temperature of the hot plate was kept at 54.58C (0.28). The testing consisted of three trials, a preshock trial
(trial 1) and two postshock trials (trials 2 and 3). Each trial involved placing the rat on the hot plate within a plastic enclosure where it remained until it responded as expected, i.e. licked the hind foot, or 60 s had elapsed. Then, it was removed from the enclosure, which ended the trial. Right after trial 1 the rat was transferred to the shock chamber where it received electric footshocks (100 ms, 2.0 mA), one every 2 s for 2 min. Trial 2 was performed several seconds after the shocking. Trial 3 was performed 24 h after trial 2. The latencies of the paw-lick response in trials 1, 2 and 3 were measured and denoted as L1, L2 and L3, respectively. Active Avoidance The test enclosure consisted of a box with a metal grid ¯oor. The box was divided into two identical compartments (40:0 cm 30:0 cm 30:0 cm each) by a partition. A rectangular 8:0 cm 8:0 cm opening was provided in the lower middle part of the partition. The conditioned stimulus (CS) was a 500 Hz pulsing tone from a miniature loud speaker located at the back wall of each compartment. The unconditioned stimulus (UCS) was a series of electric footshocks (100 ms 2.0 mA 1.0 Hz rectangular pulses) applied through the grid ¯oor. The rats were trained in avoiding the UCS by moving from one compartment to another after the presentation of the CS in the compartment occupied by the rat. The CS±UCS interval was 5 s. The CS and UCS terminated immediately after the rat's entry into the second compartment. The inter-trial interval varied from 20±40 s (30 s average). The testing comprised two sessions: an acquisition session and a retention session separated by a 7-day interval. During each of the sessions the rats were trained until they reached an avoidance criterion which consisted of four avoidance responses during ®ve successive trials. Statistics In the case of the open-®eld data and the activeavoidance data differences between groups were estimated with the Kruskall±Wallis non-parametric oneway ANOVA. Multiple comparisons were made with the Sheffe's test (Siegel, 1956). Data from the remaining tests were analyzed using repeated measures twofactor ANOVA (groups trials or measurements). In case of non-homogeneity of covariances, an approximation procedure, which enables to avoid assumption about equal covariances, was applied. In this procedure the degrees of freedom used in ®nding the critical values are reduced (see Winer, 1962, p. 305±6). When-
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
83
ever a signi®cant interaction between the main factors was found, Tukey posthoc comparisons were applied (Winer, 1962). RESULTS Body Weight At the onset of the experiment, the mean body weight was the lowest in the HM group (395.9 g, S:D: 25:4), and the highest in the MES group (422.7 g, S:D: 25:2). In all groups, body weight increased gradually during the experiment (effect of measurements: F(1, 49)103.73, P < 0:0001), but the effect of the group factor as well as the groups measurements interaction was not signi®cant (results not shown). Behavioral Effects of Exposure Radial Maze Performance The analysis (two-way ANOVA, groups days) of the data obtained during the adaptation period (before the exposure) revealed no signi®cant differences between groups in the number of arm entries during successive trials. After the exposure the groups did not differ with respect to the number of perseveration or omission errors in successive days of testing (Fig. 1), as well as in the trial duration (data not shown). Open Field Behavior As regards the open ®eld behavior, the differences between groups attained statistical signi®cance (oneway non-parametric Kruskall±Wallis ANOVA: X 2 15:09, df 4, P < 0:005) only with respect to locomotor (ambulatory) activity. Detailed comparisons have shown that rats of the PS group and the MES group made signi®cantly more border crossings during the 5 min observation period than rats of the C group and the HM group, but they did not differ in this respect from the XYL group (Fig. 2). Passive Avoidance There were large individual differences within each group in the step-down latency. Statistical comparisons of the data from all animals revealed differences between groups in trial 6, i.e. on day 7 after the footshock (F 4; 282 2:86, P < 0:05); in the MES group the step-down latencies were signi®cantly shorter than in the C group. In order to reduce the within-group variability, data on two rats, with the
Fig. 1. Radial maze performance of rats exposed for 4 weeks to m-xylene or a TMB isomer at concentration of 100 ppm. The test (one trial a day) was performed on days 14±18 after exposure. The diagrams illustrate the number of perseveration (upper diagram) and omission (lower diagram) errors in successive daily trials. Denotation of groups: control Ð sham-exposed group (n 10), XYL Ð m-xylene group (n 11), PS Ð PS group (n 11), MES Ð mesitylene group (n 11), HM Ð hemimellitene group (n 11). The bars represent group means and S.E.
lowest and the highest mean step-down latency in the ®rst postshock trial, were excluded from data sets for each group of rats. The following analysis revealed a signi®cant effect of trials (F 1; 39 41:97, P < 0:0001) and a signi®cant groups trials interaction (F 4; 39 22:85, P < 0:0001). In all groups, in the preshock trials the step-down latency decreased gradually and in the solvent-exposed groups this decrease was evidently steeper than in the control group (see Fig. 3). However, as the results of statistical analysis have shown, in all preshock trials as well as in
84
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
Fig. 2. A comparison of open-field locomotor activity in shamexposed and solvent exposed rats. The test was performed on day 25 after a 4-week exposure to m-xylene or a TMB isomer at concentration of 100 ppm. Denotation of groups as in Fig. 1. The bars represent group means and S.E.
the ®rst postshock trial (trial 4) the differences between groups were not signi®cant. They appeared to be signi®cant in trial 5 (F 4; 234 3:77, P < 0:01) and in trial 6 (F 4; 234 4:49, P < 0:002). In trial 5, rats of the MES group, and in trial 6 rats of the MES, PS and XYL groups stepped down onto the ¯oor after a signi®cantly shorter time of staying on the platform than did the rats of the C group.
The Hot-plate Behavior The analysis of the paw-lick latencies (two-way ANOVA, groups trials) revealed some general differences between groups (effect of groups: F 4; 49 3:88, P < 0:01). The effect of trials as well as the groups trials interaction were also signi®cant (effect of trials: F 1; 49 61:84, P < 0:0001; interaction: F 4; 49 33:40, P < 0:0001). Consequent comparisons within groups showed signi®cant differences between trials in group C (F 2:98 21:16, P < 0:0001); in group XYL (F 2; 98 16:74, P < 0:0001); in group PS (F 2; 98 10:86, P < 0:0001); in group MES (F 2; 98 9:90, P < 0:0002), and in group HM (F 2; 98 11:53, P < 0:0001). In all groups, L2 was signi®cantly longer than L1. In XYL and MES groups, L3 was also signi®cantly longer than L1. Comparisons within trials revealed no signi®cant differences between groups in trial 1 and trial 2. They appeared in trial 3 (F 4; 147 5; 24, P < 0:001); in this trial, the paw-lick latencies measured in the XYL, PS and MES group were signi®cantly longer than in the C group (Fig. 4). Active Avoidance The analysis (non-parametric one-way Kruskall± Wallis ANOVA) of the data obtained during the training session (number of trials to the avoidance criterion)
Fig. 3. Diagrams illustrating the effect of a four-week inhalation exposure to m-xylene or a TMB isomer at concentration of 100 ppm on the step-down response latency in the passive avoidance test. The test was performed on days 39±48 after exposure. Trials 1, 2 and 3 were performed at 24 h intervals. The step-down response was punished by a 10 s footshock in trial 3 only. Trials 4, 5 and 6 were performed 24 h, 3 days and 7 days after trial 3, respectively. The maximum time of staying of the platform was 180 s. Denotations of groups as in Fig. 1. The bars represent group means and S.E.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
85
Fig. 4. A comparison of sham exposed and solvent exposed rats with respect to the latency of the paw-lick response to heat (54.58C) before (L1), several seconds after (L2) and 24 h after a 2 min intermittent footshock. The test was performed on day 50 and 51 after a 4-week inhalation exposure to m-xylene or a TMB isomer at concentration of 100 ppm. Denotation of groups as in the figure. The bars represent group means and S.E.
showed signi®cant differences between groups (X 2 13:19, df 4, P < 0:02). In all solvent exposed groups the mean number of trials to criterion was signi®cantly higher than in the control (C) group. Among the
solvent-exposed groups, the number of trials to criterion during training was highest in the XYL group and lowest in the HM group, but the differences were not signi®cant (Fig. 5). Analysis of the data obtained
Fig. 5. Active avoidance learning in rats after a 4-week inhalation exposure to to m-xylene or a TMB isomer at concentration of 100 ppm. In one massed-trial session (inter-trial interval Ð 20±40 s; maximum number of trials Ð 60) the rats learned to shuttle between two neighboring compartments in order to avoid a footshock. The test was performed on day 54±60 after exposure. Denotation of groups as in Fig. 1. The bars represent group means and S.E. of the number of trials.
86
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
during retraining revealed no differences between groups. DISCUSSION The results of the present experiments show that in the rat a 4-week inhalation exposure to 100 ppm of mxylene or any one of the TMB isomers, while having no in¯uence on the general health status, resulted in behavioral alterations detectable weeks after exposure. Generally, in PS, MES and XYL groups, the observed changes were qualitatively similar to those found in previous experiments on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998). The rats from these groups did not differ from controls with respect to the radial maze performance. However, they showed an increased locomotor activity in the open ®eld, increased latency of the paw-lick response to heat (the hot-plate test) and impaired acquisition of passive avoidance response. All exposed groups were de®cient in learning, but not in retention of the active avoidance response in shuttle box. Taken together, these changes do not indicate cognitive impairments but rather a decreased ability to inhibit locomotor response, especially in fear-inducing environment. Such an interpretation is suggested directly by the results of the open®eld and the passive avoidance test and may be applied for explaining results of the hot-plate test and the active avoidance test. In the hot-plate test, an augmented locomotion (attempts to escape from the enclosure), competitive with the recuperative response (paw licking), might be responsible for the increase in paw-lick latency. In the active avoidance situation, ®nding the only successful escape route, (i.e. the small, halfcovered opening in the partition separating the two compartments) might be more dif®cult for rats showing a higher locomotor response to US and CS. The main question asked in the present studies concerned the relative potential of TMBs and m-xylene in inducing long-lasting neurobehavioral alterations. The literature data on XYL suggest relatively low behavioral toxicity of this solvent; usually the effects were noted at exposure concentrations much higher than 100 ppm (see Toxicological Pro®le for Xylenes, 1995). As regards TMB, some effects were noted in rats during and shortly after a subchronic inhalation exposure at concentration of 100±250 ppm (see Section 1). Moreover, in rats exposed to HM these effects were more pronounced than in rats exposed to PS (Korsak and RydzynÄski, 1996). Thus, we might have expected that: (i) at the 100 ppm exposure level, XYL
would have no behavioral consequences or the effects would be evidently weaker than those produced by exposure to TMBs; (ii) the effects of exposure to HM would be more pronounced than those of PS and presumably MES. However, the results of the present experiment did not con®rm the above. The behavioral effects of exposure to 100 ppm of XYL were no less pronounced than those seen after exposure to TMBs. MES appeared as effective as PS and XYL but the effects of HM were apparently weaker than those of PS or MES. However, taking into account the results of our previous studies on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998), the results of the present work may not necessarily denote that HM is less toxic than PS or MES. In these experiments rats were exposed to graded concentrations (0, 25, 100 or 250 ppm) of TMBs. The behavioral effects of exposure to 100 ppm of HM were similar to those produced by exposure to PS or MES in the present experiment. Impaired passive avoidance was also observed in rats exposed to 25 ppm but in rats exposed to 250 ppm none effect was detected. A similar concentration±effect relationship was noted in the experiment on PS. However, in case of PS exposure 250 ppm was effective, although less than 100, and 25 ppm appeared totally ineffective. The above observations suggest that, at least with respect to postexposure behavioral alterations, the concentration±effect curve for both, HM and PS, will have the shape of inverted ``U'' and that the HM concentration most effective in producing behavioral alterations seems to be lower that the most effective PS concentration. It is likely then, that if the concentration selected for the present experiment were less than 100 ppm, the behavioral effect of exposure to HM would be more pronounced than those of PS and possibly MES. We also presume that the discrepancies in the results of exposure to 100 ppm of HM in the present and previous experiment may have resulted from different sensitivity of examined rat populations to the solvents or from undetected differences in other experimental variables. Irrespective of the cause of the discrepancies, the results of previous and the present experiment indicate that HM differs, albeit quantitatively rather than qualitatively, from the remaining TMBs and XYL in its capacity for affecting rat behavior. A question then arises about the distinctive properties of HM that could be regarded as the possible cause of this difference. Of the four solvents used in the present study HM is distinguished by the most asymmetric and therefore the most polar structure. Moreover, it is the only compound with the methyl substituent in the ortho-con®guration.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
Both these features are likely to contribute to the HM af®nity to some biological constituents (e.g. lipids and proteins) as well as to its toxokinetics. The data obtained by Jarnberg and Johanson (1995) suggest that, compared to PS and MES, HM has a higher af®nity for water and lower af®nity for fats. The oil/ blood partition coef®cient (calculated value) of HM (164) was slightly lower than that of PS (173) and signi®cantly lower than that of MES (230) (Jarnberg and Johanson, 1995). On the other hand, in human volunteers no distinct differences between TMBs were noted in respiratory uptake and blood concentration in the course of a 2 h (Jarnberg et al., 1996) or 8 h (Kostrzewski et al., 1997) inhalation exposure. Taken together, the above observations suggest that compared to MES but not to PS, HM may be less readily redistributed from blood to body fats, and hence to the neural tissue. However, HM is metabolized an excreted at much lower rate than all the remaining solvents used in the present study. In the rat, elimination of an oral dose (1.2 g/kg) of HM was reported to take even more than ten days. The elimination of MES and PS proceeded three to four times faster (Mikulski and Wiglusz, 1975). In the experiments by Huo et al. (1989), as much as 99% of a single oral dose of PS was excreted within 24 h after exposure. In an inhalation experiment on rats exposed to solvents for 3 days (12 h per day) at concentration of 100 ppm, no detectable amounts of solvent was found 12 h after the last exposure in the brains of rats exposed to PS or XYL (Zahlsen et al., 1992) which precluded accumulation. (It is worth emphasizing, that at the end of exposure day 3, the brain concentrations of XYL and PS were similar (XYL 28:6 7:3 mmol/kg; PS 36:5 2:2 mmol/kg)). Based on the above data, it may be presumed that during the present experiment the brains of rats from the HM group might be subjected to the action of the solvent and its metabolites for a longer time and at higher concentration (due to accumulation) than the brains of rats of the XYL, PS and possibly MES group. Consequently, in the HM group the exposure induced neuroadaptative changes related to the postexposure behavioral performance might differ somehow from those in the remaining groups. Aromatic hydrocarbons are known to interact with the catecholaminergic systems and this action is regarded as speci®c for this group of organic solvents (see Kyrklund (1992) for review). The dopaminergic system appears to be the most sensitive; persistent changes suggestive of an increased sensitivity, were observed after repeated inhalation exposures to some aromatic hydrocarbons at concentrations as low as
87
80 ppm (Von Euler et al., 1993, 1994; Hillefors-Berglund et al., 1995). From the point of view of the behavioral ®ndings of our present as well as previous studies the latter observations deserve particular attention. First, it is known that pharmacological activation of the dopaminergic system may result in a disinhibition of motoric responses suppressed by punishment (see Jackson and Westlind-Danielsson (1994) for review), which conforms with the general de®cit in our solvent exposed rats. Second, the changes in the dopaminergic system resulting from a low-level solvent exposure were reported to persist weeks after exposure cessation and show a non-linear concentration±effect relationship just like the behavioral effects of exposure to PS or HM. It has been shown, for instance, that 4 weeks after a 4-week (6 h per day, 5 days per week) inhalation exposure to toluene, the af®nity of dopamine D2 agonist binding in the rat striatum is signi®cantly increased but only in rats which were exposed at concentration of 80 ppm or less. Concentrations higher than 80 ppm produced little or no effect (Hillefors-Berglund et al., 1995). If exposure to TMB isomers and XYL could affect the dopaminergic system in a similar manner as toluene does, it might account for the presence and the nature of the behavioral effects observed in our studies, as well as the non-linearity of the concentration±effect relationship noted in case of PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998). Accordingly, the lower ef®cacy of HM compared to that of PS, MES and XYL in the present experiment might indicate that in case of HM the concentration of 100 ppm was to high to induce persistent changes in dopaminergic system comparable with those produced by exposure to the other solvents. One more possible explanation of the behavioral effects of the low-level exposures to TMBs or XYL is worth considering. Namely, it may be presumed that they may have been consequences of stress responseinduced repeatedly by the daily exposures. Bushnell and Peele (1988) have found out that xylene is perceived by the rat as an aversive stimulus at concentrations much below 100 ppm. According to recent ®ndings, in the rat the odor of many solvents, including toluene and mesitylene, induces a speci®c EEG response, identical with that induced by the odor of natural rat predators (Heale et al., 1994; Zibrowski and Vanderwolf, 1997; Zibrowski et al., 1998). This would suggest that the odor of at least some solvents may be a threat signal for this species. It is known that a repeated, or even single exposure to an aversive stimulus or any other stressor may result in a persistent
88
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89
sensitization of the subject to the same and other stressors (see Antelman et al., 1997), which may be manifested as an augmented locomotor response to a test stressor (Herman et al., 1984; Kalivas et al., 1986; Kalivas and Stewart, 1991). However, depending on whether the intensity of the sensitizing stimulus is low or high the ensuing change may consist in an increase or decrease in the behavioral response to a subsequent administration of a stressor (Antelman et al., 1991). In the light of the above, the behavioral effect of exposure to solvents in our experiments, may be regarded as a consequence of rat sensitization to aversive events by repeated exposure to the aversive stimulus: solvent odor. The smaller range of behavioral effects in the HM group, compared to the remaining groups, might indicate that at 100 ppm the HM odor was more aversive for the rat than the odor of the remaining solvents, and therefore in the HM group the activating effect on locomotion was less pronounced . The above hypothesis is not inconsistent with that linking the behavioral effects of low-level solvent exposure with changes in the dopaminergic system. Involvement of the dopaminergic system in stress response is well documented. (Finlay and Zigmond, 1997; Pani et al., 2000). Therefore, it cannot be excluded that the changes within this system observed after exposure to solvents (see above) were Ð at least in some instances Ð related more to the exposure induced stress response than to a direct action of the solvent on the dopaminergic neurons. The above assumption, relating the effects of lowlevel solvent exposure to stress may also appear as relevant for explaining some effects of solvent exposure in humans (Otto et al., 1992) and, therefore, deserves further testing. Summing up, results of the present study demonstrate that long-term behavioral alterations may follow a 4-week inhalation exposure to m-xylene, or any of the TMB isomers at concentration as low as 100 ppm. The character of these alterations suggests a decreased ability to inhibit the locomotor response especially in a stress-inducing experimental context. m-Xylene does not differ reliably from TMB isomers with respect to the type and magnitude of the behavioral effects. Among the TMB isomers, at the concentration used in this study, HM appears to be less potent in inducing behavioral alterations than PS or MES. However, considering the apparent non-linearity of the concentration±effect function found in earlier studies on PS (Gralewicz et al., 1997) and HM (Wiaderna et al., 1998), the latter conclusion must be limited to the 100 ppm concentration only.
ACKNOWLEDGEMENTS This work was supported by Grant PB 1093/PO5/97/ 13 from the Polish Committee for Scienti®c Research. The skilful work of Wanda Majcherek (exposure level setting and control) is highly appreciated.
REFERENCES American Conference of Governmental Industrial Hygienists. Threshold limit values for chemical substances and physical agents. Biological Exposure Indices, 1996. Antelman SM, Caggiula AR, Kocan D, Knopf S, Meyer D, Edwards DJ, Barry H. One experience with ``lower'' or ``higher'' intensity stressors, respectively enhances or diminishes responsiveness to haloperidol weeks later: implications for understanding drug variability. Brain Res 1991;566:276±83. Antelman SM, Soares JC, Gershon S. Time-dependent sensitization Ð possible implications for clinical pharmacology. Behav Pharmacol 1997;8:505±14. Arlien-Soborg P. Chronic effects of organic solvents on the central nervous system and diagnostic criteria. Organic solvents and the central nervous system. Copenhagen: World Health Organization, 1985. p. 197±218. Baker E. A review of recent research on health effects of human occupational exposure to organic solvents. JOM 1994;36:1079± 91. Battig K, Grandjean E, Turian V. Gesundheitschaden nach langdauernder trimethylbenzol exposition in einer malerwerkstatt. Z Praeventiv Med 1956;1:389±404. Battig K, Grandjean E, Rossi L, Rickenbacher J. Toxicologische Untersuchungen uber trimethylbenzol. Arch Gewerbepathol 1958;16:555±66. Bolles RC, Fanselow MS. A perceptual-defensive-recuperative model of fear and pain. Behav Brain Sci 1980;3:291±301. Bushnell PJ, Peele D. Conditioned flavor aversion induced by inhaled p-xylene in rats. Neurotoxicol Teratol 1988;10:273±7. Cameron GR, Paterson JLH, DeSarin GSW, Thomas JC. The toxicity of some methyl derivatives of benzene with special reference to PS and heavy coal naphtha. J Pathol Bacteriol 1938;46:95±107. Finlay JM, Zigmond MJ. The effects of stress on central dopaminergic neurons: possible clinical implications. Neurochem Res 1997;22:1387±94. Gralewicz S, SocÂko R. Effects of a single exposure to chlorphenvinphos, an organophosphate insecticide on hot-plate behaviour in rats. Pol J Occup Med 1990;3:215±20. Gralewicz S, Wiaderna D, Tomas T, RydzynÄski K. Behavioral changes following four-week inhalation exposure to (1,2,4trimethylbenzene) pseudocumene in the rat. Neurotoxicol Teratol 1997;19:327±33. Heale VR, Vanderwolf CH, Kavaliers M. Components of weasel and fox odors elicit fast wave burst in the dentate gyrus of rats. Behav Brain Res 1994;31:159±65. Herman J-P, Stinus L, le Moal M. Repeated stress increases locomotor response to amphetamine. Psychopharmacology 1984;84:431±5.
S. Gralewicz, D. Wiaderna / NeuroToxicology 22 (2001) 79±89 Hillefors-Berglund M, Liu Y, von Euler G. Persistent, specific and dose-dependent effects of toluene exposure on dopamine D2 agonist binding in the rat caudete-putamen. Toxicology 1995;100:185±94. Huo J, Aldous S, Campbell K, Davies N. Distribution and metabolism of 1,2,4-trimethylbenzene (pseudocumene) in the rat. Xenobiotica 1989;19:161±70. Ikeda M. Public health problems of organic solvents. Toxicol Lett 1992;64/65:191±201. Innis NK, Macgillivray M. Radial maze performance under food and water deprivation. Behav Proc 1987;15:167±79. Jackson DM, Westlind-Danielsson A. Dopamine receptors: molecular biology, biochemistry and behavioural aspects. Pharmacol Ther 1994;64:291±369. Jarnberg J, Johanson G. Liquid/air partition coefficients of the trimethylbenzenes. Toxicol Ind Health 1995;11:81±8. Jarnberg J, Johanson G, Lof A. Toxicokinetics of inhaled trimethylbenzenes in man. Toxicol Appl Pharmacol 1996;140:281±8. Kalivas PW, Richardson-Carlson R, Van Orden G. Crosssensitization between foot shock stress and enkephalin-induced motor activity. Biol Psychiat 1986;21:939±50. Kalivas PW, Stewart J. Dopamine transmission in the initiation and expression of drug- and stress-induced sensitization of motor activity. Brain Res Rev 1991;16:223±44. Korsak Z, RydzynÄski K. Neurotoxic effects of acute and subchronic inhalation exposure to trimethylbenzene isomers (pseudocumene, mesitylene, hemimellitene) in rats. Int J Occup Med Environ Health 1996;9:341±9. Kostrzewski P, Wiaderna-Brycht A, Czerski B. Biological monitoring of experimental human exposure to trimethylbenzene. Sci Total Environ 1997;199:73±81. Kyrklund T. The use of experimental studies to reveal suspected neurotoxic chemicals as occupational hazards: acute and chronic exposures to organic solvents. Am J Ind Med 1992;21:15±24. Linz DH, de Garmo PL, Morton WE, Wiens AN, Coull BM, Maricle RA. Organic solvent-induced encephalopathy in industrial painters. J Occup Med 1986;28:119±26. Mikkelsen S. Epidemiologic update on solvent neurotoxicity. Environ Res 1997;73:101±12. Mikulski PI, Wiglusz R. The comparative metabolism of mesitylene, pseudocumene, and hemimellitene in rats. Toxicol Appl Pharmacol 1975;31:21±31. Otto D, Molhave L, Rose G, Hudnell HK, House D. Neurobehavioral and sensory irritant effects of controlled exposure to a complex mixture of volatile organic solvents. Neurotoxicol Teratol 1992;12:649±52. Olton DS. Shock-motivated avoidance and the analysis of behavior. Psychol Bull 1973;79:243±51.
89
Olton DS, Samuelson RJ. Remembrance of places passed: spatial memory in rats. J Exp Psychol 1976;2:97±116. Pani L, Porcella A, Gessa GL. The role of stress in the pathophysiology of the dopaminergic system. Mol Psychatr 2000;5:14±21. Siegel S. Nonparametric statistics for the behavioral science. New York: McGraw-Hill, 1956. Stollery B., Organic solvents. In: Smith AP, Jones DM, editors. Handbook of human performance. The physical environment, vol. 1. London: Academic Press, 1992. p. 149±175. SÂwiercz R, Korsak Z, RydzynÄski K. Kinetics of n-butyl alcohol and m-xylene in blood during single and combined inhalation exposure in rats. Int J Occup Med Environ Health 1995;8: 361±5. Toxicological Profile for Xylenes. US Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry, August 1995. Von Euler G, Ogren S-O, Li XM, Fuxe K, Gustafsson J-A. Persistent effects of subchronic toluene exposure on spatial learning and memory, dopamine-mediated locomotor activity and dopamine D2 agonist binding in the rat. Toxicology 1993;77:223±32. Von Euler G, Ogren S-O, Li XM, Fuxe K, Gustafsson J-A. Persistent effects of 80 ppm toluene on dopamine-regulated locomotor activity and prolactin secretion in the male rat. Neurotoxicology 1994;15:621±4. Wesoøowski W, Gromiec JP. Prediction of exposure to engine fuel vapours. Med Pracy 1996;67:20±9. Wesoøowski W, Gromiec JP. Occupational exposure in Polish paint and lacquer industry. Int J Occup Med Environ Health 1997;10:71±81. Wiaderna D, Gralewicz S, Tomas T. Behavioral changes following four-week inhalation exposure to hemimellitene (1,2,3-trimethylbenzene) in the rat. Int J Occup Med Environ Health 1998;11:319±34. Winer BJ. Statistical principles in experimental design. New York: McGraw Hill, 1962. Woolfe G, MacDonald A. The evaluation of the analgesic action of pethidine hydrochlorine demerol. J Pharmacol Exp Ther 1944;80:300±7. Zahlsen K, Eide I, Nilsen AM, Nilsen OG. Inhalation kinetics of C6 to C10 aliphatic, aromatic and naphtenic hydrocarbons in rat after repeated exposure. Pharmacol Toxicol 1992;71:144±9. Zibrowski EM, Hoh TE, Vanderwolf CH. Fast wave activity in the rat rhinencephalon: elicitation by the odors of phytochemicals, organic solvents, and a rodent predator. Brain Res 1998;800:207±15. Zibrowski EM, Vanderwolf CH. Oscillatory fast wave activity in the rat pyriform cortex: relations toolfaction and behavior. Brain Res 1997;766:39±49.