Abstract
This article reports neurobehavioral tests in rats with C5-C11 isoparaffinic and cycloparaffinic hydrocarbons. Testing, conducted shortly after exposure, evaluated the effects in several domains including clinical effects, motor activity, functional observations, and visual discrimination performance. Isopentane and cyclopentane did not produce any evidence of acute central nervous system (CNS) effects at levels up to 20 000 mg/m3. A C6/C7 mixed cycloparaffinic solvent produced minor, reversible changes in latency to response in visual discrimination testing at 14 000 mg/m3; the no-effect level was 4200 mg/m3. A C8 isoparaffin produced no effects at 14 000 mg/m3, the highest level tested. A C9/C11 isoparaffinic solvent produced minor acute CNS effects at 5000 mg/m3, with 1500 mg/m3 as the no-effect level. A C10 cycloparaffinic solvent did not produce any statistically significant CNS effects at 5000 mg/m3. These studies were designed to provide data that may be useful in setting occupational exposure limits for C5-C11 isoparaffinic and cycloparaffinic hydrocarbons.
Keywords
Introduction
Hydrocarbon solvents are commercial substances that may be composed of 4 types of constituents; normal (n-) paraffins, isoparaffins, cycloparaffins, and aromatics; and with carbon numbers ranging from 5 to >15. Some of the commercial hydrocarbon solvents are complex and may contain constituents of more than one molecular type. Because of the widespread use and specific physical/chemical properties of hydrocarbon solvents there may be exposure, particularly by inhalation. Consequently, there is a need to provide consistent and appropriate occupational exposure advice. To assist in that process, it is necessary to obtain toxicological effect data that could be used in developing occupational exposure recommendations. Based on previous studies, 1 –5 it was determined that tests of acute central nervous system (CNS) effects in rodents could be used for this purpose. Ethanol was used in preliminary tests as a reference substance to demonstrate that the testing methods were appropriate. 1 To define testing needs, a matrix was developed which included representative hydrocarbons of each type and carbon number for use as a predictive tool for other, untested constituents. The group of representative substances included monoconstituent as well as more complex hydrocarbon solvents that were comprised of unique molecular types but covering a range of carbon numbers. This report describes the studies of the representative branched (ie, isoparaffinic) and saturated ring (ie, cycloparaffinic also known as naphthenic) hydrocarbons.
To identify representative test substances, the isoparaffinic and cycloparaffinic constituents of hydrocarbon solvents were examined. In principle, these constituents could range from C5 to >C15. However, hydrocarbons with carbon numbers greater than 13 have such low vapor pressures that they do not contribute greatly to the exposure. Further, there is evidence that the C10+ branched alkanes do not produce CNS effects at the highest vapor concentrations that can be stably maintained. 6 It has also been shown that the blood–brain ratios for n-alkanes decline when carbon numbers are greater than C10 7 consistent with the hypothesis that the blood–brain barrier may impede the absorption of higher molecular weight molecules. 8 Thus, for purposes of this program, efforts were concentrated on C5-C11 isoparaffins and cycloparaffins. More specifically, the acute CNS effects of several representative substances; isopentane and cyclopentane (C5 constituents), a commercial product containing C6/C7 isomers of alkyl cycloparaffinic solvent; isooctane (C8), a commercial product containing predominantly C9-C11 isoparaffinic constituents and a commercial product containing predominantly C10 cycloparaffinic isomers were assessed in rodent studies. The acute CNS effects of cyclohexane (a C6 cycloparaffin) were assessed in humans and rodents in previous studies. 4 The specific objective of the current study was, as noted above, to provide representative data on the acute CNS effects of this group of substances for use in an overall process to develop Occupational Exposure Limit (OEL) recommendations for hydrocarbon solvents. 9 A review of the literature by Caldwell et al 10 provided evidence that the majority of occupational exposures to these substances were below their respective occupational exposure limits.
Methods
Test Materials
Isoparaffinic Hydrocarbons
Isopentane, Chemical Abstract Services (CAS) number 78-78-4, was supplied by ExxonMobil Chemical Europe, Antwerp, Belgium. It had a listed purity of 95.9%, with the remainder being primarily n-pentane and was used as supplied. Isopentane has a molecular weight of 72 da, a density of 0.62 kg/L, and a boiling range of 26°C to 36°C at 105 Pa.
Isooctane, CAS number 90622-56-3, was supplied by ExxonMobil Chemical Europe, Antwerp, Belgium. This solvent is derived from an alkylation process which produces streams of essentially pure isoparaffinic hydrocarbons. These streams are then separated by distillation to produce finished solvents with the desired technical properties. This specific solvent contained approximately 97% C8 isomers, with the remaining 3% being C7 isoparaffins. The solvent had a density of 0.70 kg/L and a boiling range of 100°C to 105°C. The isooctane was used as supplied.
C9-C11 isoparaffinic solvent, CAS number 90622-57-4, was supplied by Total Solvants, Oudalle, France. It had an average molecular weight of 140 da, a density of 0.75 kg/L, and a boiling range of 161°C to 175°C. The C9-C11 isoparaffinic solvent was used as supplied.
Cycloparaffinic Hydrocarbons
Cyclopentane, a C5 cycloparaffin, Chemical Abstract Services (CAS) number 287-92-3, supplied by ExxonMobil Chemical Europe, Antwerp, Belgium, had a listed purity of 96.1% with the remaining constituents being either n-pentane or isomers of isopentane and was used as supplied. Cyclopentane has a molecular weight of 73 da, a density of 0.75 kg/L, and a boiling point of 49°C at 105 Pa.
C6/C7 cycloparaffinic solvent, CAS number 64742-89-8, was supplied by Hanf + Nelles, Dusseldorf, Germany. It had a molecular weight of approximately 99 da, a density of 0.77 kg/L, and a boiling range of 98°C to 104°C. The sample was used as supplied.
C10 cycloparaffinic solvent, CAS number 64742-48-9, was supplied by ExxonMobil Chemical Europe, Antwerp, Belgium. It had a molecular weight of 142 da, a density of 0.81 kg/L, and a boiling point of 163°C to 185°C. The sample was used as supplied.
Methods
Animal Studies
All animal studies utilized male Wistar-derived WAG/RijCrlBR rats obtained from Charles River Wiga, Sulzfeld, Germany. The rats were approximately 14 weeks of age at the time of testing. Mean body weights at study initiation ranged from 219 to 259 g. The mean body weights in the various studies are given in the online data supplements. Food (Rat and Mouse no. 3 Breeding Diet, RM3) and water were available ad libitum in the home cages, also during exposure. Food was changed after each exposure period. Food was not available during functional observational battery (FOB) and motor activity assessment or during visual discrimination performance testing. Rats that were used for visual discrimination testing were water deprived in conjunction with behavioral testing. The rats were initially housed in groups of 5 in wire mesh cages. In the tests described below, rats assessed for functional observations and motor activity were singly housed and exposed. For visual discrimination performance the rats were housed and exposed in groups of 4. In between subsequent exposures, the cages with animals were removed from the exposure chambers and placed in clean air in identical exposure chambers. The functional observation and motor activity tests were performed consecutively in the same rats and required 32 animals, divided by computer randomization with a correction for mean body weight into 4 dose groups of 8 animals. The visual discrimination tests also required 32 rats randomly divided into 4 dose groups of 8 on a body weight basis. Other matters of animal husbandry have been described in more detail previously. 2,4 The protocol was reviewed and approved by TNO’s Animal Ethics Committee. The welfare of the animals was maintained in accordance with the general principles governing the use of animals in experiments in the European Communities (Directive 86/609/EEC) and Dutch legislation (The Experiments on Animals Act, 1997).
Inhalation Exposure System
Rats were exposed by inhalation 8 h/day for 3 consecutive days in modified H 1000 inhalation chambers (Hazleton Systems Inc, Aberdeen, Maryland), with a total airflow through the chamber of approximately 20 m3/h. The exposures were repeated over several days to assess whether there was an exacerbation of effects due to repeated exposures. Target exposure concentrations were 2000, 6500, or 20 000 mg/m3 for isopentane and cyclopentane; 1400, 4200, or 14 000 mg/m3 for isooctane and for the C6/C7 alkyl cycloparaffinic solvent; 500, 1500, and 5000 mg/m3 for the C9-C11 isoparaffinic solvent; and 1000, 2500, or 5000 mg/m3 for the C10 cycloparaffinic solvent. The lower ends of the exposure ranges were chosen to approximate the recommended 9 occupational exposure levels for these solvents, that is, 1200 to 1500 mg/m3. The upper ends of the exposure range were an order of magnitude higher than the lower range to assess the potential for effects at levels above those recommended for occupational exposures. Additionally, the highest exposure levels used in the isopentane and cyclopentane studies (20 000 mg/m3) were approximately half of the lower explosive limits for these substances and the highest concentrations considered safe to test. The upper end of the exposure range for the C9-C11 isoparaffinic solvent and the C10 cycloparaffinic solvent (5000 mg/m3) was near the maximally attainable vapor concentrations for these solvents at 20°C. Temperature in the exposure chambers was monitored continuously and was maintained at 19°C to 25°C.
At the beginning of each exposure period, rats were placed in the exposure chambers when the chemical concentrations were at the defined levels. At the end of exposure periods, rats were removed from the exposure chambers for behavioral experiments while the test substance concentrations were still at the defined levels. Control animals were exposed to clean air while in the exposure chamber.
Test atmospheres were created by passing liquid material through heated water baths to create vapors. Water bath temperatures were as follows: isopentane 22°C; cyclopentane 30°C; isooctane 60°C; C6/C7 cycloparaffinic solvent 60°C; C9/C11 isoparaffinic solvent 60°C; and C10 cycloparaffinic solvent 74°C. The vapors were mixed with air and added to the main airflow systems for the inhalation chambers. Flow rates of the pumps were set to reach the target concentrations in the exposure chambers. The test atmospheres were continuously monitored by total carbon analysis (TCA, Ratfish, Germany) and converted into exposure levels by comparison to concentrations in Tedlar bags (Chrompack, Bergen op Zoom, the Netherlands) of known size and content. More specifically, sample bags were filled with 50 L air. Aliquots of test material of known weight were injected into the bags to produce standards covering a range of concentrations. These standards were then measured in the total carbon analyzer and used to convert the TCA readings into concentrations of the test substances.
Evaluation of CNS Effects
The rats were evaluated for viability and other measures of well-being, functional observations, motor activity, and visual discrimination performance. The testing procedures will be briefly described here. The reader is referred to previous publications for more detail. 2,4
Viability and physiological indicators
The assessments included daily health and viability checks. Additionally, body weights were recorded at random, on days of testing, and immediately prior to sacrifice. Body temperatures were measured by rectal probe before exposure and after the first and third 8-hour exposure periods.
Functional observations and motor activity
Neurobehavioral functioning was evaluated using selected measures from a standardized functional observational battery (FOB) and motor activity assessment protocol similar to that used in the World Health Organization/International Programme on Chemical Safety (WHO/IPCS) Collaborative Study on Neurotoxicity Assessment. 11 –13 The FOB consists of standardized observations and simple tests designed to evaluate gross changes in neurological and behavioral functioning in the rat using measures taken from different functional domains as summarized in Table 1 . Spontaneous motor activity was measured in sessions of 30 minutes each using an automated video image analysis system, with rats placed individually in a 50 × 50 × 50 cm3 (l × w × h) open-roofed cage. Rats were tested before exposure and after the first and third 8-hour exposure periods. The functional observation battery was conducted immediately after termination of exposure and required approximately 5 min/animal to complete. The motor activity testing was started immediately after completion of the FOB testing, so the animals were placed in the testing device 25 to 40 minutes after being removed from the exposure chambers. The measurements were conducted by observers who were not aware of the treatment that the rats had received.
Visual discrimination performance
A group of rats different from those used for FOB and motor activity testing was evaluated for 2-choice visual discrimination performance. The apparatus consisted of 16 operant chambers (32 × 30 × 28 cm3 [l × w × h]) and programming and recording equipment programmed with the MedState notation system (Med Associates, Inc, St Albans, Vermont). Each operant chamber was equipped with 2 levers, 2 stimulus lights, and a water dipper for delivering water as a reinforcer. In addition, a photocell assembly was mounted in the water trough to detect the entry of each rat’s head when obtaining water reinforcers. Each operant chamber was located in a ventilated, sound-attenuated cubicle. Prior to treatment, water-deprived rats were first trained to obtain water reinforcers and to lever press using autoshaping techniques. The rats subsequently received 4 weeks of training on a discrete-trial light–dark visual discrimination task to stabilize baseline responding. Animals were trained 5 d/week from Monday to Friday. During training and testing the rats were given access to water for only 15 minutes immediately after the training or testing session. On the weekends, they were given free access to water.
Test sessions consisted of 100 trials or 60 minutes, whichever came first, and were conducted at approximately the same time each day. Dose groups were counterbalanced across time of testing and testing device. Trials were initiated by the illumination of either the left or right stimulus light, and the rat’s task was to depress the lever under the illuminated light to obtain a water reward. Illumination of right and left stimulus lights was counterbalanced and occurred in a predetermined semirandom order. If the rat pressed the correct lever, the stimulus light was extinguished and a water reward was delivered. If the initial response during a trial was on the incorrect lever, the trial continued until the correct lever was pressed. A given trial remained in effect until the correct lever had been pressed; however, only the initial lever press was used to assess accuracy. Trials were separated by an intertrial interval (ITI) of 10 seconds. A response during the ITI reset the ITI timer, and the rat was required to wait a further 10 seconds before initiation of the following trial. Rats were tested on the day prior to the first exposure day and on all 3 exposure days, immediately after the exposure periods. A post-exposure test was performed the day after the last exposure period to evaluate persistence of effects.
Summary of Functional Observational Parameters Assessed in Acute Neurobehavioral Studies of Complex Hydrocarbon Solvents
Variables measured are summarized in Table 2 . For each rat, the initial response in each trial was recorded and used to calculate accuracy. If the initial trial response was correct, the latency of the lever press was also recorded. If the initial response was incorrect, the number of incorrect lever responses made by the rat before switching to the correct lever was recorded. Following a correct lever response, the water dipper was raised. The system recorded whether the rat inserted its snout to drink from the dipper to obtain the water reward, providing a measure of the number of reinforcers obtained. The latency to obtain the reinforcer in each trial was also recorded. During the intertrial period, lever responses were recorded to determine the number of ITI periods in which one or more lever presses occurred and the number of repetitive ITI lever responses.
Summary of Dependent Variables in the 2-Choice Visual Discrimination Task
Abbreviations: ITI, intertrial interval; SD, standard deviation.
Statistical Analysis
All data were analyzed using the SAS statistical software package (release 6.12). For each test measure, probability (P) values <.05 were considered statistically significant.
Body weights and body temperatures were analyzed using 1-way analysis of variance (ANOVA) conducted at each time point followed by Dunnett multiple comparison tests.
Continuous variables from the FOB were analyzed using ANOVA for pre-exposure performance in order to examine the possible preexisting differences among the groups prior to treatment. Treatment effects were analyzed using repeated measures ANOVA with test time points as the repeated factor. If either a significant effect of treatment or a significant treatment-by-time interaction was indicated, ANOVA was performed at each test time point. Group comparisons were made using Dunnett multiple comparison tests. Motor activity data were analyzed using ANOVA for pre-exposure performance. Effects of exposure on total activity or habituation were analyzed using 3-way repeated measures ANOVA with 1 treatment factor and 2 repeated factors (test time point and time blocks within each session). Each session consisted of 5 time blocks of 6 minutes each. Rank data were analyzed by Kruskal-Wallis 1-way ANOVA on each test day followed by planned multiple comparisons in case of a significant result.
Baseline visual discrimination performance prior to exposure was examined in 2 ways: (1) by examining the mean performance averaged across the 5 days in the week prior to exposure (pre-week responding) and (2) by examining the performance on the day preceding exposure (pre-day responding). One-way ANOVA was conducted on the pre-week performance and on the pre-day performance data in order to examine the possible preexisting differences among the groups prior to exposure. Treatment effects were analyzed using repeated measures ANOVA of the data recorded during the 3 exposure days. Huynh-Feldt adjustment of P values of the repeated measures factor was applied in case the assumption of sphericity of observations was violated. When a significant treatment effect was demonstrated, pairwise group comparisons were performed in order to determine which treated group significantly differed from the control group. When a significant treatment-by-time interaction was demonstrated, 1-way ANOVA was performed at each test time point followed by Dunnett multiple comparison tests. Levene's test was used to assess the equality of variances in the different groups. In case of unequal variances, 1-way ANOVA was performed on log-transformed data. If log transformation did not satisfy the assumption of equal variances, the Welch correction test was applied to the 1-way ANOVA of nontransformed data. Persistent effects were evaluated by ANOVA of post-exposure data.
Results
Exposure Levels
Isoparaffinic Hydrocarbons
The target concentrations for isopentane were 2000, 6500, and 20 000 mg/m3, and the mean analytically determined concentrations were 1942 (range 1870-2020), 6450 (range 6390-6520), and 19 570 (range 19 440-19 670) mg/m3. The target concentrations for isooctane were 1400, 4200, and 14 000 mg/m3, and the mean analytically determined concentrations were 1415 (range 1310-1690), 4205 (range 3690-4680), and 14 005 (range 13 120-14 450) mg/m3. The target concentrations for the C9-C11 isoparaffinic solvent were 500, 1500, and 5000 mg/m3, and the mean analytically determined concentrations were 500 (range 500-500), 1500 (range 1500-1500), and 5035 (range 5030-5170) mg/m3.
Cycloparaffinic Hydrocarbons
The target concentrations for cyclopentane were 2000, 6500, and 20 000 mg/m3, and the mean analytically determined concentrations were 2038 (range 1950-2070), 6412 (range 6410-6430), and 19 885 (range 19 770-20 060) mg/m3. The target concentrations for the C6/C7 cycloparaffinic solvent were 1400, 4200, and 14 000 mg/m3, and the mean analytically determined concentrations were 1405 (range 1400-1410), 4255 (range 4220-4290), and 13 955 (range 13 890-14 070) mg/m3. The target concentrations for the C10 cycloparaffinic solvent were 1000, 2500, and 5000 mg/m3 and the mean analytically determined concentrations were 1002 (range 920-1040), 2495 (range 2190-3210), and 5032 (range 4600-5430) mg/m3.
Viability and physiological indicators
Isoparaffinic Hydrocarbons
There were no remarkable clinical observations for isopentane, isooctane, or the C9-C11 isoparaffinic solvent and no significant differences in body weights in the isopentane or the C9-C11 isoparaffinic solvent–exposed groups. In the isooctane study, body weights were reduced by approximately 3% in the high-exposure group over the 3-day exposure period (P < .001; refer to the online data supplement file for body weight data). There were no effects on body temperatures at any exposure level in the studies, with any of the solvents.
Cycloparaffinic Hydrocarbons
Cyclopentane—There were no noteworthy signs of impaired viability or well-being. Similarly, there were no significant differences in body weights or body temperatures at any exposure level (body weight and temperature data are given in the online data supplements).
C6/C7 alkyl cycloparaffinic solvent—There were no remarkable signs of impaired viability or well-being. Body weights in the high-exposure group were reduced by comparison to both controls and pretest values, but the differences were not significant. There were approximately 0.5°C reductions in body temperature in the high-exposure group during the 3-day exposure period, which were statistically significant at the P < .05 level (body weight and temperature data are given in the online data supplements).
C10 cycloparaffinic solvent—In all, 1 animal from the 2500 mg/m3 group and 3 from the 5000 mg/m3 group had bloody exudates around the nose and mouth during FOB testing after the third exposure period, suggesting respiratory irritation at these levels. There were no significant effects on body weight. There were statistically significant (P < .05) reductions (approximately 1°C) in body temperature in the 2500 and 5000 mg/m3 exposure groups during the 3-day exposure period (body weight and temperature data are given in the online data supplements).
Functional Observations and Motor Activity
Isoparaffinic Hydrocarbons
Isopentane—The only difference in any of the FOB measurements was an increase in foot splay (repeated measures ANOVA, treatment-by-time F 6, 56 = 2.10, P = .0671), but the difference was not treatment related as only the lowest exposure group was statistically different from the control group and only after the first 8-hour exposure period (1-way ANOVA of data collected after the first 8-hour exposure period, F 3, 28 = 4.68, P = .0090; Dunnett t test: P < .05; FOB data are given in the online data supplements). One animal from the high-exposure group exhibited slight ataxia after the third 8-hour exposure but statistical significance was not achieved. There were no effects on the measures of motor activity (total distance run, number of movements, and mean velocity).
Isooctane—There were no significant changes in any of the FOB measurements. One rat had an abnormal body position after exposure, but this was not considered treatment related as the animal was in the low-exposure group. There were no statistically significant effects on motor activity.
C9-C11 isoparaffinic solvent—There were no significant differences in any of the FOB measurements. There were a few animals with gait abnormalities, but these were scattered across the exposure groups; and, therefore, assumed to be incidental. There were no effects on motor activity.
Cycloparaffinic Hydrocarbons
Cyclopentane—There were a few small changes in some of the FOB measures, but none of these were considered toxicologically important. These included slight changes in foot splay and arousal, but these differences were not significant when tested by 1-way ANOVA. There were also changes in click response, but these were apparently due to changes in control group responses rather than an effect of treatment. There were no effects on measures of motor activity (total distance run, number of movements, and mean velocity).
C6/C7 alkyl cycloparaffinic solvent—There were no significant changes in any of the FOB measurements, and there were also no statistically significant effects on measures of motor activity.
C10 cycloparaffinic solvent—There were some changes in gait in 3 of 8 animals from the high-exposure group which were not observed in any of the other groups. These included tiptoe walking and slight ataxia. The association with the high-exposure group suggested that these were treatment-related effects. However, there were no statistically significant differences between groups. There were statistically significant differences between groups in touch and click response, but these were not considered treatment related as the responses were similar to the pre-exposure measurements. Motor activity data were lost prior to analysis due to failure of the video analysis system. Accordingly, no motor activity data were available for this substance.
Visual Discrimination Performance
Isoparaffinic Hydrocarbons
Isopentane - Visual discrimination performance testing did not reveal any treatment-related effects (Table 3). There were no differences in the number of trials completed or discrimination ratios. Frequency of response during the ITI was reduced in the high-exposure group but the difference was not statistically significant. There were no differences in the frequency of repetitive errors. There were no differences in the overall response latency. There were no changes in the frequency of very short latency responses (ie, <1 second). Similarly, there were no differences in short (ie, <2 seconds) or long (ie, >6 seconds) latency responses. The overall conclusion was that isopentane did not affect visual discrimination performance at exposure levels up to 20 000 mg/m3.
Summarized Results of Visual Discrimination Performance Testing With Isopentane (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least 1 response was made divided by the total number of ITI (× 100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds (including the number of reponses within 1 second).
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
Isooctane—Visual discrimination performance testing did not reveal any treatment-related effects (Table 4). There were no differences in the number of trials completed or discrimination ratios. There were no differences in the frequency of repetitive errors. There were no differences in the overall response latency. There was no change in the frequency of very short latency responses (ie, <1 second). Similarly, there were no differences in short (ie, <2 seconds) or long (ie, >6 seconds) latency responses. The overall conclusion was that isooctane did not affect visual discrimination performance at exposure levels up to 14 000 mg/m3.
Summarized Results of Visual Discrimination Performance Testing With Isooctane (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least one response was made divided by the total number of ITI (×100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds (including the number of responses within 1 second).
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
C9-C11 isoparaffinic solvent—Visual discrimination performance testing did not reveal any treatment-related differences in the number of trials completed or discrimination ratios (Table 5). There was a significant effect of exposure on the response latency (repeated measures ANOVA, treatment F 3, 28 = 5.62, P = .0038), with a small but significant increase in latency to response in the high-exposure group (post hoc group comparison between high-exposure and control groups, F 1 = 12.51, P = .0014). In addition, variability in response latency was significantly increased in the high-exposure group (repeated measures ANOVA, treatment F 3, 28 = 2.97, P = .0487; post hoc group comparison between high-exposure and control groups, F 1 = 5.93, P = .0215). There was also a significant increase in the frequency of long (ie, >6 seconds) latency responses during exposure (repeated measures ANOVA, treatment F 3, 28 = 9.83, P = .0001). Post hoc group comparisons indicated significantly increased numbers of long latencies in the low-exposure group (F 1 = 4.92, P = .0348) and the high-exposure group (F 1 = 28.62, P < .0001), when compared to the control group. The frequency of long latencies was also increased in the intermediate dose group but the response was not significantly different from the control. There were no differences between groups in the post-exposure tests. The overall assessment was that C9-C11 isoparaffinic solvent had some minimal effects on visual discrimination performance at an exposure level of 5000 mg/m3 but that exposure to 1500 mg/m3 was without effect.
Summarized Results of Visual Discrimination Performance Testing With C9-C11 Isoparaffinic Solvent (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least one response was made divided by the total number of ITI (× 100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds (including the number of responses within 1 second).
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
m Group comparisons indicated a statistically significant (P < .05) difference from the control group during the 3-day exposure period.
Cycloparaffinic Hydrocarbons
Cyclopentane—Visual discrimination performance testing did not reveal any treatment-related effects. There were no differences in the number of trials completed or discrimination ratios (Table 6 ). There were no signs of behavioral disinhibition; no significant changes in ITI responding or in the frequency of repetitive errors during trial responding. There were no differences in the overall response latency. There were no changes in the frequencies of very short (ie, <1 second), short (ie, <2 seconds) or long (ie, >6 seconds) latency responses. The overall assessment was that cyclopentane did not affect visual discrimination performance at exposure levels up to 20 000 mg/m3.
C6/C7 alkyl cycloparaffinic solvent - There were no differences in the number of trials completed or discrimination ratios. There were no differences in ITI responding or in the frequency of repetitive errors. The overall latency to response was significantly different between groups (repeated measures ANOVA, treatment F 3, 27 = 4.87, P = .0078), with significantly increased latencies in the high-exposure group when compared to the control group (post hoc group comparison between high-exposure and control groups, F 1 = 8.74, P = .0064; Table 7 ). In addition, variability in response latency was significantly increased in the high-exposure group (repeated measures ANOVA, treatment F 3, 27 = 4.96, P = .0072; post hoc group comparison between high-exposure and control groups, F 1 = 8.94, P = .0059). Similarly, a nonsignificantly increased frequency of long (ie, >6 seconds) latency responses was observed. The frequencies of very short latency responses (ie, <1 second) and short latency responses (ie, <2 seconds) were reduced, but these group differences were not statistically significant. Generally, the effects on response latencies did not persist over the 3-day exposure period. There were some significant differences in the post-exposure assessment, but these seemed to be more a consequence of changes in the control performance than a treatment-related effect. Frequencies of long and short latency responses in the 14 000 mg/m3 group were significantly different from the control group in the post-exposure test, but this was due to 1 animal and therefore not judged toxicologically important. Group means were similar to those during the pre-exposure testing. The overall assessment was that C6/C7 alkyl cycloparaffinic solvent reversibly affected the response speed component of visual discrimination performance at an exposure level of 14 000 mg/m3. The no-effect level was 4200 mg/m3.
Summarized Results of Visual Discrimination Performance Testing With Cyclopentane (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least one response was made divided by the total number of ITI (× 100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds (including the number of responses within 1 second).
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
Summarized Results of Visual Discrimination Performance Testing With C6/C7 Alkyl Cycloparaffinic Solvent (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least one response was made divided by the total number of ITI (× 100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds (including the number of responses within 1 second).
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
m Group comparisons indicated a statistically significant (P < .05) difference from the control group during the 3-day exposure period..
n P < .05.
C10 cycloparaffinic solvent—Visual discrimination performance testing did not reveal any treatment-related differences in the number of trials completed or discrimination ratios. There was a small and not statistically significant reduction in the overall latency to response in the high-exposure group (Table 8 ). There was also a reduction in the frequency of very short latency responses (ie, <1 second; repeated measures ANOVA, treatment F 3, 27 = 3.51, P = .0285), with significantly fewer very short latencies in the high-exposure group when compared to the control group (post hoc group comparison between high-exposure and control groups, F 1 = 7.68, P = .0100), and an increase in the frequency of long (ie, > 6 seconds) latency responses (repeated measures ANOVA, treatment F 3, 27 = 4.12, P = .0157), with significantly increased numbers of long latency responses in the low-exposure group (post hoc group comparison between low-exposure and control groups, F 1 = 10.27, P = .0035), the mid-exposure group (post hoc group comparison between mid-exposure and control groups, F 1 = 7.58, P = .0104) and the high-exposure group (post hoc group comparison between high-exposure and control groups, F 1 = 5.75, P = .0237). However, these differences reflected pretreatment variability and were not considered treatment related. The post-exposure assessment for most groups was compromised as a number of the animals were not water deprived prior to testing. However, responses in the 5000 mg/m3 group were similar to the pre-exposure values. The overall assessment was that C10 cycloparaffinic solvent did not produce toxicologically relevant effects on visual discrimination performance at an exposure level of 5000 mg/m3.
Summarized Results of Visual Discrimination Performance Testing With C10 Cycloparaffinic Solvent (mean ± SEM)
Abbreviations: ITI, intertrial interval; SEM, standard error of the mean.
a The total number of trials completed during each session, maximum = 100.
b Number of reinforcers obtained divided by the number of reinforcers delivered (× 100).
c Number of correct trial responses divided by the number of trials responses.
d The number of ITIs in which at least one response was made divided by the total number of ITI (× 100).
e The total number of incorrect trial responses following an initial incorrect response.
f The total number of ITI responses following an initial ITI response.
g The latency (seconds) to make a correct trial response.
h Standard deviation of the response latency.
i The number of responses within 1 second.
j The number of responses within 2 seconds.
k The number of responses taking more than 6 seconds.
l The mean latency (seconds) to obtain reinforcer.
Summary of Studies of Acute CNS Effects in Rats in Comparison to Current Occupational Exposure Recommendations
Abbreviation: CNS, central nervous system; ACGIH, American Conference of Governmental Industrial Hygienists; NOEL, no observed effect level.
a ACGIH 30 Note that the list above includes both TLVs and Group Guidance Values from Appendix H.
b Lammers et al. 31
c Present report
d Lammers et al. 4
e McKee et al. 32
Discussion
The principal objective of this study was to assess the acute CNS effects of C5-C11 isoparaffinic and cycloparaffinic hydrocarbons in rodents for use in a process to calculate the occupational exposure levels for complex hydrocarbon solvents. The substances tested were isopentane and cyclopentane (C5), C6/C7 alkyl cycloparaffinic solvent, isooctane (C8), a commercial C9-C11 isoparaffinic solvent, and a C10 cycloparaffinic solvent. Hydrocarbon constituents with molecular weight >C10 may be found in some solvents, but, because of their low vapor pressures, do not contribute substantially to exposure. Further, they apparently do not produce measurable CNS effects at maximally attainable vapor concentrations. 6
It should be noted that there are some limitations to interpretation, which are imposed by the study design. More specifically, for technical reasons the rats were tested for neurobehavioral effects after rather than during the exposure period. Inhaled isopentane and cyclopentane are not absorbed efficiently. 14 Further, based on kinetic studies of n-pentane which has an elimination halftime of approximately 8 minutes, 15 isopentane and cyclopentane are rapidly eliminated, primarily by exhalation. In the present study, all neurobehavioral testing was conducted within 1 hour after the exposure period. Given the extremely rapid elimination of isopentane and cyclopentane, brain concentrations may have decreased substantially between the end of the exposure period and the initiation of neurobehavioral testing. Timing is less of an issue for the higher-molecular-weight hydrocarbons for which the halftimes for elimination from the CNS are on the order of 2 hours.
Discussion of Results of the Acute CNS Studies of Isoparaffinic Hydrocarbon Solvent Constituents
Isopentane did not produce any effects at exposure levels up to 20 000 mg/m3. This level was selected as a pragmatic upper boundary because it is equivalent to approximately 50% of the lower explosive limit. In the current study, exposures at this level did not produce any clinical effects and had no effect on body weight or body temperature. Similarly, there were no significant effects in the FOB although 1 animal from the high-exposure group exhibited gait alterations. There were no effects on motor activity. Finally, there were no significant effects on visual discrimination performance. The overall conclusion was that 20 000 mg/m3, the highest level tested, was a no-effect level for acute CNS effects.
Previous data on 2 isopentane isomers (2-methylbutane and 2,2 dimethyl propane) were recently summarized. 16,17 In a series of studies, 2-methylbutane exhibited anesthetic properties at exposure levels >90 000 ppm (265 000 mg/m3). (Note that when data were given in ppm in the original references, they were reported in that way in this article with the conversion to mg/m3 shown in parentheses.) The data for neopentane (2,2-dimethylpropane) were similar but less extensive. Light anesthesia was observed at levels of 200 000 ppm (590 000 mg/m3). Both isopentane isomers are less well absorbed than n-pentane and are rapidly eliminated by exhalation. 14 In summary, these data indicate that isopentane does not produce toxicologically important effects except at extreme exposure levels.
The results with isooctane were very similar to those with isopentane. There were no treatment-related clinical observations. Body weights were reduced in the highest exposure groups, but there were no effects on body temperature at levels up to 14 000 mg/m3, the highest concentration tested. There were no significant findings in the FOB battery, and motor activity was not affected. Finally, there were no statistically significant differences in the visual discrimination performance tests. Thus, the no-effect level for acute CNS effects was 14 000 mg/m3, the highest level tested, although the body weight changes suggested the possibility of slight physiological effects at that level.
Other studies have produced similar results. For example, Swan et al 18 reported that isooctane did not produce evidence of anesthesia at 8000 ppm (∼37 000 mg/m3), but effects were observed at levels of 16 000 ppm (∼ 75 000 mg/m3) and higher.
Schreiner et al 19 studied the neurological and other toxicological effects of a light alkylate naphtha distillate in rats. The material tested was described as being almost entirely comprised of isoparaffinic constituents with carbon numbers predominantly in the range of C5 to C8. Exposure was by inhalation at 668, 2220, and 6446 ppm (2400, 8100, and 24 300 mg/m3). Rats were exposed 6 h/d, 5 d/week for 13 weeks; neurological effects were assessed pretest and after weeks 5, 9, and 14. The only clinical finding was a report of red facial staining in the high-dose group. The neurological testing did not reveal any effects on functional observations or motor activity (note that the animals were not exposed on the day of testing, and most likely, any acute CNS effects would have been reversed), and there were no pathological findings suggestive of neurological effects. A similar study of light catalytic reformed naphtha, which contained approximately 58% isoparaffins and 30% n-paraffins with the remainder aromatic constituents, was also conducted. 20 Exposure levels were 750, 2500, and 7500 ppm (approximately 2700, 9000, and 27 000 mg/m3). There were no reports of clinical findings suggestive of acute CNS effects, and, as above, no effects on functional observations or motor activity, and no pathological changes in the nervous system. Frontali et al 21 reported that exposure to isohexanes (2-methylpentane and 3-methylpentane) for up to 14 weeks did not lead to pathologic changes in the nervous system, and Soiefer et al 22 similarly found no neurological effects in a subchronic neurotoxicity study of mixed hexane isomers.
Exposure to C9 to C11 isoparaffinic solvent at levels up to 5000 mg/m3 did not result in any physiological or clinical effects. There were no significant effects on the FOB or on motor activity. In the visual discrimination performance tests, a statistically significant finding was an increased latency to response in the high-exposure (5000 mg/m3) group, accompanied by a significantly increased frequency of long latency responses. There was also an increased frequency of long (> 6 seconds) responses in the 500 and 1500 mg/m3 groups, with the difference in the 500 mg/m3 group being significantly different from controls. A review of the individual animal data revealed that this response was observed in only a few animals and had little effect on the mean response latencies in these groups. As the differences were small and not dose responsive, they were judged to have been incidental findings. In summary, it was concluded that the overall no-effect level for acute CNS effects with the C9-C11 isoparaffinic solvent was 1500 mg/m3.
In other studies of the effects of high-molecular-weight isoalkanes on the CNS, Balster and colleagues assessed the acute CNS effects of a series of isoparaffinic solvents ranging from C7-C8 to C11-C12. The first of these studies examined the effects of increasing concentrations of a C8-C9 isoparaffinic solvent on functional observations and operant behavior in mice. 23 Effects were observed, which became progressively more pronounced over a concentration range of 2000 to 6000 ppm (∼10 000-30 000 mg/m3). In the operant behavior studies, it was also reported that there were slight but variable effects at 1000 ppm (∼5000 mg/m3). A second study assessed the effects of exposure to a range of isoparaffinic solvents on locomotor activity and operant behavior. 6 In the locomotor activity studies, C7-C8 and C8-C9 isoparaffinic solvents produced effects at levels in the range of 4000 to 6000 ppm (∼20 000-30 000 mg/m3) with 2000 ppm (∼10 000 mg/m3) as a no-effect level. A C10-C11 isoparaffinic solvent produced effects over a similar exposure range although the magnitude of response was reduced. A C11-C12 isoparaffinic solvent did not produce effects at exposure levels up to 2000 ppm, the maximum vapor level that could be stably maintained.
In the tests of operant behavior, effects were seen only with the lower molecular weight isoparaffinic solvents (C7-C11), with no-effect levels of approximately 2000 ppm (∼10 000 mg/m3). Thus, the data 23 from Balster's laboratory are quite consistent with the results of the present study. Additionally, the current study also confirmed Balster's observation 23 that the CNS effects were both subtle and reversible. The observation that C11-C12 isoparaffinic solvent did not produce acute CNS effects in studies conducted at the maximally attainable vapor concentration 6 suggested that effective brain concentrations of higher molecular weight isoparaffins cannot be achieved.
Discussion of the Results of the Acute CNS Studies of Cycloparaffinic Hydrocarbon Solvent Constituents
Cyclopentane did not produce any effects at exposure levels up to 20 000 mg/m3. This level was selected as a pragmatic upper boundary because it is equivalent to approximately 50% of the lower explosive limit. In a review of published and unpublished data on cyclopentane, Galvin and Marashi 24 reported that the LC50 for cyclopentane was >11 260 ppm (approximately 32 000 mg/m3). They also provided summarized results from a study by Virtue 25 in which it was shown that mice could be exposed for 10 minutes to levels of up to 60 000 ppm (approximately 170 000 mg/m3) without evidence of anesthetic effects. However, exposures at higher levels produced anesthesia and death in an increasing proportion of the exposed mice. Thus, the absence of effects in rats exposed for 8 hours at 20 000 mg/m3 was consistent with previous experience.
The C6/C7 alkyl cycloparaffinic solvent produced limited effects, even at the highest concentration tested (14 000 mg/m3). There were small and no significant effects on body weights, but there was a significant reduction in body temperature. There were no functional effects and no changes in motor activity. The visual discrimination studies indicated significantly increased latencies to response, not significantly reduced frequencies of very short (<1 second) and short (<2 seconds latency responses, and a nonsignificant increase in long (>6 seconds) latency responses. The no-effect level was 4200 mg/m3. These results are similar to those from previous acute studies in which cyclohexane produced minimal CNS effects at levels greater than 20 000 mg/m3 but had no effects at 8000 mg/m3. 4,26
Exposure to the C10 cycloparaffinic solvent resulted in some minor effects in the high-exposure group (5000 mg/m3). These included some evidence of irritation as evidenced by bloody exudates around the mouth and nose, as well as reduced body temperature, although body weight was not affected. In the FOB assessment, there was some evidence of altered gait, but the differences were not statistically significant. Motor activity data were not available. In the test of visual discrimination, there was no effect on latency to respond; the frequency of very short (<1 second) latency responses was reduced and the frequency of very long (> 6 seconds) latency responses was increased, but, based on a review of the pretreatment data, these differences seemed more likely to have been a reflection of the preexisting variability rather than a consequence of treatment. The overall no-effect level for acute CNS effects was 5000 mg/m3.
The tissue distributions of normal-, iso-, and cycloparaffins were investigated by Zahlsen and colleagues in a series of inhalation exposure studies. 7,27 –29 In the studies in which male Sprague-Dawley rats were exposed for 12 hours at levels of 100 ppm to n-alkanes, cycloalkanes, and aromatics, brain–blood ratios for the aliphatic molecules increased with increasing number from C6-C10 from approximately 4 to 10. The exception to this was t-butylcyclohexane (C10 cycloparaffin) for which the brain–blood ratio was less than 4. 28 In a study 29 of similar design, brain–blood ratios for isoparaffins also increased over the range of C8-C10. In a separate study, the brain concentrations of C9-C13 n-alkanes were measured following 8 hours of exposure to these compounds at their corresponding saturated vapor concentrations. 7 In this study, it was observed that the brain–blood ratios decreased with increasing carbon number, suggesting an inhibition of uptake into the CNS of hydrocarbons >C10. Hau et al 8 suggested that the blood–brain barrier effects may inhibit the uptake of aliphatic molecules >C10.
To summarize the results of the acute toxicity testing, the objectives of this specific study were to assess the effects of C5-C11 isoparaffinic and cycloparaffinic hydrocarbon solvent constituents on the CNS of rodents. There was suggestive evidence that the higher molecular weight isoparaffins and cycloparaffins may produce mild, reversible effects on gait, and/or visual discrimination at exposure levels approximating or above 5000 mg/m3. The effects seemed to increase with increasing carbon number to approximately C11, but in all cases the effects were subtle and reversible and when present were only in the highest exposure groups.
A further objective of this study was to use these data in an overall process to develop occupational exposure limits for hydrocarbon solvents. 2,4,9 The specific intent was to use the results of the acute CNS effect studies in the development of “guidance values” which could then be used to calculate occupational exposure limits for complex hydrocarbon solvents. 9 As shown in Table 9, low-molecular-weight aliphatic constituents (n-pentane, isopentane, and cyclopentane) produced no effects, on CNS or otherwise, at levels up to 20 000 mg/m3, the highest exposure levels considered safe to test. Among the C8 aliphatic constituents, the n- and isoalkanes did not produce any acute CNS effects at levels up to 14 000 mg/m3 (the highest exposure levels tested), whereas a C6/C7 cycloparaffinic solvent produced minor effects with the no-effect level at 4200 mg/m3. The C10 n- and isoalkanes produced minor effects at levels of 5000 mg/m3 with no-effect levels of 1500 mg/m3, whereas the C10 cycloaliphatic solvent did not produce CNS effects at 5000 mg/m3. The C9 aromatic molecules produced acute CNS effects at 5000 mg/m3 with no-effect levels in the range of 1000 to 1500 mg/m3, and a C10/C11 aromatic solvent produced effects at 2000 mg/m3 with 600 mg/m3 as the no-effect level. The general observations from these and previously published data are (1) that there are no strong differences between n-, iso-, and cycloaliphatic molecules, supporting the use of common guidance values for these constituents; (2) that the acute CNS effects seem to increase with increasing carbon number to approximately C10, but may then start to decline at higher carbon numbers due to declining brain–blood ratios; and (3) the aromatic molecules may be somewhat more active in producing acute CNS effects than the corresponding aliphatic species. The CNS effects also seemed to be the most sensitive indicators of effect, along with small changes in body weights and temperatures, and, in the case of the aromatic species and the C10 cycloaliphatic solvent, there may also be some respiratory irritation at levels which produce CNS effects. Finally, (4) the levels of acute CNS effects found in this program were above the current occupational exposure recommendations, 9,30 supporting the adequacy of current protective measures.
Footnotes
Two of the coauthors (RHM and DEO) are employed by companies that manufacture hydrocarbon solvents. The online data supplements are available at http://ijt.sagepub.com/supplemental.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was sponsored by the CEFIC Hydrocarbon Solvent Producers Association.