This investigation was supported by U.S. Public Health Service Grant MH 04230.
Both marihuana (Bromberg, 1934; Keeler, 1967; Tart, 1970) and tetrahydrocannabinols (THC's) (Isbell et al., 1987; Hollister et al., 1968; Isbell and Jasinski, 1969) repeatedly have been reported to cause changes in visual, auditory, tactile and proprioceptive perception in low to moderate doses and to cause sensory illusions and hallucinations in high doses. One of the places where these drugs might be acting to cause such changes could be polysensory areas of the cortex. These areas respond both to input through several sensory modalities (Walter, 1964; Bignall and Singer, 1967) and to stimulation of primary sensory areas (Bignall and Imbert, 1969) and may be involved in an overall integration of sensory information.
The work reported here was undertaken to determine whether two THC's change the responsiveness of such cortical areas in the squirrel monkey (Saimiri sciureus). In order to avoid possible interference from anesthetic agents, responses in frontal polysensory cortex were evoked by stimulation of the postcentral trunk area of primary somatosensory cortex (Benjamin and Welker, 1957) in animals with high midbrain sections. To permit a comparison of drug effects on the response in polysensory cortex with those on the response in nonpolysensory cortex, the response at a site roughly homotopic to the stimulating electrodes was often recorded simultaneously with those in polysensory cortex. The effects of pentobarbital were determined in an attempt to differentiate between possible specific effects of the tetrahydrocannabinols on sensory integration and the less specific sedative action of the drugs (Boyd and Meritt, 1965; Hollister et al., 1968). Since the electrocorticogram (ECoG) and the blood pressure were monitored as criteria of the viability of the preparations, drug effects on these parameters are also reported.
METHODS. Male squirrel monkeys weighing between 620 and 892 g were operated on under ether anesthesia. Cannulas were placed in the trachea, in the femoral artery for monitoring blood pressure, and in the femoral vein for injecting drugs, The dorsal surface of the cortex was exposed and the superior colliculi were exposed by sucking away part of the right posterior occipital lobe, cutting the tentorium, and moving the cerebellum caudally. The brain stem was sectioned with a spatula at, or just anterior to, this point and anesthesia was terminated. Examination of the formalin-fixed brains after the experiments showed that the angles of the sections varied, the ventral edge of the section being between the rostral pons and a midhypothalamic level. Sections ranged from 90 to 100% complete, with all major sensory pathways severed in all cases.
The dura was removed and a mineral oil bath was formed over the exposed cortex with the help of a piece of rubber dam cemented to the scalp. Bath temperature was maintained between 35 and 37°C and rectal temperature between 36 and 39°C Most of the animals breathed spontaneously but a few had to be respired artificially. Stimulation of and recording from the ec.-..tical surface were by means of silver ball electrodes. The post-central trunk area of right primary somatosensory cortex was stimulated by means of two electrodes placed about 2 mm apart. The stimuli were provided by a Grass S4 stimulator and were delivered through an isolation unit and a constant current unit. The stimuli were monophasic, 0.1-msec duration pulses ranging from a fraction of a milliampere up to 8 mA. The stimulator was controlled by either a set of Tektronix timing circuits or a Devices Digitimer. Single stimuli, or pulse pairs for the study of recovery cycles, were delivered at the rate of 1/4 sec. This rate was found to be sufficiently low that repetitive stimulation (lid not produce a diminution of the responses. Monopolar recordings were made, when possible, from the ipsilateral and contralateral poly-sensory areas of postarcuate cortex and from a point on parietal lobe roughly homotopic to the stimulating electrodes. In some animals one of these three sites was not used either because a sufficiently large response could not be found or because the shock artifact interfered with the response. The responses were amplified with Grass P 511 preamplifiers, with frequency responses set at 7 Hz and 2 kHz, and recorded on magnetic tape for later analysis with an LAB-8 (Digital Equipment Corporation, Maynard, Mass.) computer. One or more monopolar electrocorticograms were recorded on a Grass polygraph. The indifferent electrode for all recording was a silver wire placed in the Gelfoam packing which replaced the removed part of the right occipital lobe. The blood pressure was recorded by means of a pressure transducer and the polygraph. At the end of each experiment the animal was sacrificed with an overdose of pentobarbital.
The recorded wave forms always consisted of an initial positive wave, usually followed by a broader negative wave. The amplitude of the positive wave was always measured from base line. The negative wave was measured as the difference between peak positivity and peak negativity. Series of 16 to 32 responses were averaged by the computer and the averaged responses were used for determinations of thresholds, amplitudes and recovery cycles. Recovery cycles were determined as reported previously, with the computer being used to separate overlapping responses to conditioning and test stimuli at short interstimulus intervals (Hurlbrink and Boyd, 1969). Interstimulus intervals of 5 to 1000 msec were tested. Average latencies, from the beginning of the shock artifact to the positive or negative peak of the response, were also made by the computer on series of 16 to 130 responses. Most, but not all, of the above measurements were made with each animal. Differences in responses before and after drugs were analyzed by Student's t test and P values less than .05 were considered to indicate statistically significant changes.
The experiments were carried out by recording control responses two to three hours after the termination of anesthesia, giving a dose of drug, and recording a similar set of responses. The ECoG of the preparations consisted primarily of synchronized activity interspersed with short periods of desynchronized activity. In agreement with the observations of Bignall and Imbert (1969) the evoked responses were diminished in amplitude, and often the wave form was widened, during periods of desynchrony. For this reason care was taken to record responses only during the periods of synchronized activity. Experiments were terminated if any marked deterioration of the ECoG was observed.
The drugs used were (—)-Y-trans-tetrahydrocannabinol (4.8-THC), (—)-Y-trans-tetrahydrocannabinol (Y-THC)' and pentobarbital. In the other commonly used system of nomenclature these tetrahydrocannabinols are A'"°) and 1', respectively. Only one drug was given to any single animal, but to gain as much information as possible several doses of the same drug were often tested in each animal. Since the interval between successive doses was between 45 and 60 minutes and all of the drugs have a relatively long duration of action (Hollister et al., 1968), the dose level being tested at any particular time was expressed in terms of the total dose at that time. The drugs were given i.v., over a period of at least 3 minutes to minimize cardiovascular effects, and the collection of data was started 10 minutes after either of the THC's and 5 minutes after pentobarbital. The THC's were administered either dissolved in polyethylene glycol 400 or as a true colloidal suspension in 1% Tween 80 in normal saline (Dagirmanjian and Boyd, 1962). Control injections of appropriate volumes of the polyethylene glycol 400 and the Tween 80 in normal saline had no consistent effects on any of the parameters measured.
RESULTS. Data were obtained from 16 monkeys. Of these, 6 received A.-THC, 5 received A.- THC and 5 received pentobarbital. There were no apparent differences in the effects of the two THC's, and they will therefore be treated together. The evoked responses were highly variable both from response to response and with time. The effects of variation from response to response were largely eliminated by computer averaging. However, it was still necessary to watch the behavior of the averaged responses throughout an experiment in order to decide whether a change was due to the administration of a drug or to the passage of time and a change in the state of the animal. The recording of evoked responses only during a constant ECoG state, i.e, high voltage activity during control periods, decreased the variability from average to average but did not eliminate the problem. An example is shown in figure 1 which depicts the changes in the average amplitude of the negative wave of the response at the parietal lobe homotopic site over a period of several hours in a typical experiment. The averages are of 25 responses and were made at about three-minute intervals throughout the experiment. At the first arrow an appropriate volume of the vehicle, Tween in normal saline, was given and produced no effect. At the second arrow 0.5 mg/kg of A.- THC was given. This was followed immediately by a sustained, highly significant (P = .003, comparing nine sets of measurements before and after drug) increase of 33% in the amplitude of the response. At the third arrow an additional 1 mg/kg of A.-THC was given. Although this was followed by a statistically significant (P = .041) decrease of 14% in amplitude, the figure shows that this change was gradual and is thus not so convincing. All changes after drug administration, even if statistically significant, have been ignored if they could not be differentiated from changes with time.
The THC's were tested at cumulative doses of 0.25 to 1.5 mg/kg. In all cases low doses increased the amplitudes of all of the evoked responses (table 1), in two cases by as much as 85 and 90%. In general, the negative wave, when present, was more influenced by all drugs tested than was the positive wave. It is known that the positive and negative wave forms of transcallosally evoked responses in the cat are separate phenomena (Peacock, 1957; Grafstein, 1959). The differential sensitivity to drugs, as well as occasional differences in thresholds, indicate that the positive and negative wave forms in the current work were also separate phenomena. Higher doses either continued to increase the response amplitudes, had no effect or, more often, decreased the response amplitudes (table 1). Typical averaged evoked responses, and the effect of two doses of A5-THC upon them, are shown in figure 2. One of the experiments in which high doses of THC depressed the responses is illustrated in figure 3. Average responses with a one standard deviation envelope, evoked in homotopic and contralateral polysensory cortex, as well as ECoG and blood pressure are shown. In this animal partial recovery of both of the responses 31/2 hours after the second dose of THC indicated that the depression of the responses by the drug was not due to a sudden deterioration of the animal.
Effects of the THC's on the thresholds for, and latencies of, the evoked responses were variable. In general, doses of THC which caused increases in amplitudes tended to decrease the threshold by 0.1 to 0.2 mA. Doses that caused decreases in amplitudes tended to increase thresholds. Significant changes in latencies in both directions, as well as no change, were also seen. These changes were not correlated with either the recording site or the dose of THC.
Although the latency for all three evoked responses often changed in the same direction, it did not always do so.
With the moderate to high stimulus intensities used, control recovery cycles at all recording sites routinely showed primarily periods of facilitation starting at about 10 msec. At the homotopic and ipsilateral sites the facilitation usually lasted several hundred milliseconds whereas at the contralateral polysensory site it was usually over in about 100 msec. Recovery cycles were either unaffected by the THC's or the facilitation was slightly reduced by the drug. Where reductions were found, they did not appear to be dose dependent. A typical result is shown in figure 4 in which the recovery cycle of the positive and negative waves in the evoked response at the homotopic (A and B) and the positive wave at the contralateral polysensory (C) sites were determined. Because of an apparent lack of dose dependence, and to smooth the curves, several recovery cycle determinations were averaged together. In this experiment the THC produced a small but consistent decrease in the facilitation of both the positive and negative wave forms at the homotopic site but had no effect on the responses recorded at the contralateral polysensory site.
Pentobarbital was tested at doses of 2 to 30 mg/kg. At low dose levels pentobarbital, like the THC's„produced increases in the amplitudes of all of the evoked responses (table 1). However, the magnitudes of the increases caused by pentobarbital were somewhat smaller than those seen after the THC's. The largest increase after pentobarbital was 69%. With higher doses of pentobarbital the amplitude of the evoked responses was decreased. The responses were usually eliminated at 30 mg/kg. Like the THC's, pentobarbital caused changes in both directions in both the thresholds for the responses and the latencies to the peaks of the responses. The results were very variable with low doses. With doses of 20 to 30 mg/kg both the threshold and the latency were usually increased.
In contrast to the small effects of doses of THC's as high as 1.5 mg/kg on recovery cycles, pentobarbital at doses of 2 to 5 mg/kg markedly decreased, or abolished, the periods of facilitation in the recovery cycle. Higher doses caused pro gressively greater decreases in the amounts of recovery at all recording sites, particularly at short interstimulus intervals. An example is shown in figure 5 where A is the recovery cycle at the ipsilateral polysensory site and B at the contralateral polysensory site. A dose of 5 mg/kg of pentobarbital (triangles) essentially eliminated the facilitation seen in the controls (circles).
Both the THC's and pentobarbital tended to decrease blood pressure in a dose-dependent manner. The THC's at doses of 0.5 mg/kg and higher abolished the extensor hypertonus of the hind limbs often seen in these decerebrate preparations. The number of experiments with pentobarbital in preparations exhibiting extensor hypertonus was too small to permit a determination of the dose needed to abolish this phenomenon. Pentobarbital had its usual effect on the ECoG, that is, it produced a dose-dependent flattening of the record with spindling (that is, bursts of activity at 12-14 waves/sec) occurring at doses of 10 to 20 mg/kg. The THC's also showed a dose-dependent reduction of high frequency activity in the ECoG, but this tended to be replaced first by single spikes and then by both slow waves and spikes, as shown in figure 3. In some experiments this progressed, at high doses, to a fairly distinct spike and wave complex. An example of this is shown in figure 6, where A is during the control period, B is 25 minutes after 0.5 mg/kg of A°-THC and C is 20 minutes after an added 1 mg/kg.
Discussion. The most dramatic effect of the THC's in the work reported here was a consistent, and usually large, increase in the ampli tudes of the responses evoked at the polysensory and homotopic recording sites after low doses, and a marked reduction in their amplitudes after higher doses. Although it is impossible from the current work to know whether these changes were due to direct cortical actions or to changes in tonic inputs, the results indicate that moderate doses of THC's do increase the responsiveness of some cortical areas. However, since low doses of pentobarbital also caused increases in the responses, these results are not necessarily related to the changes in perception caused by the THC's. Rather, the similarity of effects of THC's and pentobarbital on response amplitude, that is, an increase with low doses and a decrease with higher doses, suggests that these effects of the THC's may be a reflection of their sedative action. In addition, the failure of both the THC's and pentobarbital to produce significantly different effects on the responses in polysensory and primary somatosensory cortex indicates that the observed effects were probably related to changes in general cortical functioning, at least in the sensory sphere. These changes may or may not be related to changes in perception.
There were, however, two marked differences between the effects of the THC's and those of pentobarbital. In contrast to the typical effects of increasing doses of pentobarbital on the ECoG, the THC's consistently caused spiking in the ECoG which occasionally progressed with high doses to a spike and wave pattern, an effect never seen with the barbiturate. Similar spike and wave patterns caused by THC's in the rabbit have been reported by Lipparini et al. (1969), although Bicher and Mecoulam (1968) reported, in the same species, only an increase in cortical activity with a strong beta rhythm. The other difference was in the recovery cycles. Whereas the THC's, even in high doses, had only slight depressant effects, doses of 2 to 5 mg/kg of pentobarbital markedly decreased the amount of recovery at interstimulus intervals up to 300 trine. This difference was unexpected since pentobarbital and a THC have been shown to resemble each other in their effects on recovery cycles in medial lemniscus, thalamus and reticular formation in the cat (Boyd and Meritt, 1966). Thus, it is possible that the increased responsiveness of cortical sensory areas, without a depression of recovery, seen with the THC's may be related to their effects on sensory perception.
The decrease in blood pressure caused by both THC's in these experiments was also unexpected since A°-THC has been reported to have no effect on blood pressure in man (Isbell et al., 1967; Isbell and Jasinski, 1969), although other THC's consistently decrease blood pressure in the cat (Dagirmanjian and Boyd, 1962). The currently observed decreases in the monkey may be a species specific effect or may be related to the midbrain sections used in these experiments.
One of the problems with the pharmacological investigation of the actions of THC's has been the lack of a suitable solvent for their parenteral administration. We found many years ago that even small volumes of such solvents as ethanol and propylene glycol had sufficient effects on the nervous system to confuse the effects of THC's. Colloidal suspensions in Tween 80 have therefore been used, but they are undesirable because Tweens liberate histamine, and the suspensions are difficult to prepare. It now appears that polyethylene glycol may be a useful solvent. It has no particular pharmacological action. Although Olds and Olds (1969) have reported that it caused a small, nonstatistically significant increase in the firing rate of cells of the hippocampus, preoptic area and mesencephalie reticular formation of the rat, Lipparini et al. (1969) reported that up to 1 ml did not influence the electroencephalogram or behavior of the rabbit, cat or rat. No consistent effects of up to 1 ml/kg were seen on any of the parameters measured in the monkey in the present work.
Conclusions. The similarity of the effects of the THC's and of pentobarbital on response amplitudes and the similarity of the effects on the responses at the polysensory and homotopic recording sites may indicate that these effects reflect the sedative action of all of the drugs tested and a general change in the functioning of all sensory cortex. However, the increase in response amplitudes produced by low doses of both of the THC's, without the concomitant change in the recovery cycle produced by low doses of pentobarbital, may be related to the changes in sensory perception caused by the THC's. A second difference between the THC's and pentobarbital was seen in the ECoG. Pentobarbital produced its usual dose-dependent flattening of the ECoG, with spindling at intermediate doses, whereas the THC's caused spiking which sometimes progressed to a spike and wave pattern.
The and 0°-tetrahydrocannabinols were supplied by the National Institute of Mental Health, at 99 and 90% purity, respectively.
BENJAMIN. R. M. AND WELKER, W. I.: Somatic receiving areas of cerebral cortex of squirrel monkey (Sazmin sciurews). J. Neurophysiol. 20: 286-299.1957.
BICHER, H. I. AND MECHOULAM, R.: Pharmacological effects of two active constituents of marihuana. Arch. Int. Pharmacodyn. Ther. 172: 2431. 1968.
BIGNALL, K. E. AND SINGER, P.: Auditory, somatic and visual input to association and motor cortex of the squirrel monkey. Exp. Neurol. 13: 300- 312, 1967.
BIONALL. K. E. AND Baum M.: Polysensory and cortico-cortical projections to frontal lobe of squirrel and rhesus monkeys. Electroencephalogr. Clin. Neurophysiol. 26: 206-215, 1969.
BOYD, E. S. AND MERITT, D. A.: Effects of thiopental and a tetrahydrocannabinol derivative on arousal and recruiting in the cat. J. Pharmacol. Exp. Ther. 149: 138-145, 1965.
BOYD. E. S. AND MERITT, D. A.: Effects of barbiturates and a tetrahydrocannabinol derivative on recovery cycles of medial lemniscus, thalamus and reticular formation in the cat. J. Pharmacol. Exp. Ther. 151: 376-384, 1966.
BROMBERG, W.: Marihuana intoxication. Amer. J. Psychiat. 91: 303-330, 1934.
DAGIRMANJIAN, R. AND BOYD, E. S.: Some pharmacological effects of two tetrahydrocannabinols. J. Pharmacol. Exp. Ther. 135: 25-33, 1962. GRAFSTEIN, B.: Organization of cation' connections in suprasylvian gyrus of cat. J. Neurophysiol. 22: 504-515, 1959.
HOLLISTER. L. E., RICHARDS, R. K. AND GILLESPIE, H. K.: Comparison of tetrahydrocannabinol and synhexyl in man. Clin. Pharmacol. Ther. 9: 783- 791, 1968.
HURLBRINK, E. E. AND BOYD, E. S.: Some effects of pentobarbital and strychnine on transmission through the ventrobasal complex of the cat thalamus. J. Pharmacol, Exp. Ther. 170: 181-189, 1969.
ISBELL, H., GORODETZKY, C. W., JASINSKI, D., CLAUSSEN. U., SPULAK, F. V. AND KORTNI, F.: Effects of (—)-Y-trans-tetrahydrocannabinol in man. Psychopharmacologia 11: 184-188, 1967. ISBELL, H. AND JASINSKI, D. R.: A comparison of LSD-25 with (—)-Y-trans-tetrahydrocannabinol (THC) and attempted cross tolerance between LSD and THC. Psychopharmacologia 14: 115123, 1969.
KEELER, M. H.: Adverse reactions to marihuana. Amer. J. Psychiat. 124: 674-677, 1967.
LIPPARINI, F., SCOTTI DE CAROMS, A. AND LONGO, V. G.: A neuropharmacological investigation of some trans-tetrahydrocannabinol derivates. Physiol. Behay. 4: 527-532, 1969.
OLDs, M. E. AND OLDS, J.: Effects of anxiety-relieving drugs on unit discharges in hippocampus, reticular midbrain and pre-optic area in the freely moving rat. Int. J. Neuropharmacol. 8: 87-103, 1969.
PEACOCK, S. M.: Activity of anterior suprasylvian gyms in response to transcallosal afferent volleys. J. Neurophysiol. 20: 140-155, 1957.
TART, C. T.: Marihuana intoxication: Common experiences. Nature (London) 226: 701-704, 1970.
WALTER, W. G.: The convergence and interaction of visual, auditory, and tactile responses in human nonspecific cortex. Ann. N.Y. Acad. Sci. 112: 320-361, 1964.