APA 2004 Alan S. Bloom, Ph.D. Medical College of Wisconsin.

APA

20

04

Alan S. Bloom, Ph.D.Medical College of Wisconsin

PROBLEM •The frequent use of marijuana by young people in our society continues to be a problem. It is the illicit drug that is most commonly used by youths today.

•Anatomically, the distribution of CB1 cannabinoid receptors in the animal and human brain is well described. High densities of receptors are located in hippocampus, cerebellum, cortex, substantia nigra, globus pallidus and other striatal regions. Receptor density is low in hypothalamus and lower brainstem.

• In general, the distribution of receptors in the brain is reasonable, if we take into account the known pharmacological actions of marijuana. They are found in areas associated with memory, coordination, attention and endocrine regulation.

• The goal of this research is to elucidate sites of action of ∆9-tetrahydrocannabinol (THC), the major psychoactive constituent of marijuana, in the human brain and to determine those sites that are related to the drug’s psychoactive properties and cognitive effects.

MARIJUANA (THC) EFFECTS

• Euphoria

• Memory Impairment

• Perceptual-Motor alterations

• Cardiovascular

• Pulmonary

• Reproductive

• Psychopathological Effects

CB1 Receptor Distribution and Functional Areas

High density in brain areas concerned with memory, cognition, motor coordination and reward

CANNABIS AND HUMAN BRAIN IMAGING

Volkow et al. (1996)- PET - iv THC -brain glucose metabolism in chronic marijuana users. Prior to drug administration, chronic users showed lower relative cerebellar metabolism than normal control subjects. THC iv (2 mg) increased relative cerebellar metabolism in both chronic users and controls, but only chronic users showed increases in orbitofrontal cortex, prefrontal cortex and basal ganglia.

Mathew - effects of THC on CBF - 15O-PET - THC (3 to 5 mg iv, over 20 min) increased CBF in the frontal regions bilaterally, insula and cingulate gyrus and subcortical regions -somewhat greater effects in the right hemisphere. The increase in blood flow correlated with the subjective sense of intoxication.

O’Leary (2000, 2002, 2003) examined the effects of smoked marijuana on task activation measured using 15O-PET. They reported that smoked marijuana produced no change in whole brain blood flow. However, significantly altered activation by a dichotic listening task in multiple brain regions was observed after smoking marijuana. In general, this occurred in the absence of significant decreases in task performance. Overall, they reported increased rCBF in anterior brain paralimbic regions and cerebellum that may be related to marijuana effects on mood and reduced rCBF in auditory and visual sensory regions and other regions that may be involved in attention. They proposed that decreases in these regions may be involved in the perceptual and cognitive effects of marijuana.

AIMS

• To quantify ∆9-THC action using BOLD imaging to determine its pharmacodynamic properties within the human brain and to relate them in time and intensity to observed physiological and behavioral actions of intravenous ∆9-THC.

• To determine the ability of ∆9-THC and frequent

marijuana use to alter functional brain activity using perceptual-motor and cognitive tasks that activate specific brain regions containing cannabinoid receptors as experimental probes.

METHODSSUBJECTS: Right-handed subjects, ages 19 to 45 years, were recruited using newspaper, radio and television advertisements and gave informed consent to this IRB-approved study prior to participation. Handedness was assessed using the Edinburgh Inventory (Oldfield, 1971). THC-using subjects were experienced heavy marijuana users with at least one year of current use (average use > 15 times/month). They were all generally healthy based upon history and physical examination and were not currently symptomatic of any major psychiatric disorder nor currently using any abuse substance other than cigarettes, alcohol or marijuana. Subjects in the control group were not current or recent marijuana users and met all other inclusion criteria.

Subjects presented to the General Clinical Research Center at the Medical College of Wisconsin, received a physical examination, a urine screen for drugs of abuse (Triage®) and blood tests for pregnancy, HIV, hepatitis, liver function and CBC. All subjects that met the inclusion criteria agreed to receive up to 3 doses (O.5, 1.0 3.0 mg.) of THC or ethanol vehicle as described by Volkow (1996) while undergoing functional MR imaging. Each injection was given on a separate day.

Subjective Ratings of High

0%

100%

200%

300%

400%

500%

600%

700%

800%

Pre-Drug Pre-Drug Post-Drug 5 Post-Drug 15 Post-Drug 25 Post-Drug 35

Time After Injection (Minutes)

Perc

ent o

f Pre

-Dru

g R

atin

g

0.0 mg

0.5mg

1.0 mg

3.0mg

Injection

Change in Heart Rate After IV THC

Dose of ∆9 -THC Increase in HR – Beats/Min Ethanol vehicle

8.5 ± 2.3

0.5 mg.

7.6 ± 0.9

1.0 mg.

18.8 ± 3.8*

3.0 mg.

27.0 ± 4.1*

Values shown are the mean ± SEM of the peak increase in heart rate in the 35 minute period after drug injection. * p≤0.01 compared to the ethanol vehicle.

Behavioral Ratings

• Administered doses are rated behaviorally effective by experienced marijuana users similar to those used in a social setting and produce a dose-related increase in HR.

• Data obtained during scanning should reflect changes similar to those experienced socially.

Pharmacokinetic Principles

• There are time-related changes in drug concentration within the circulation and body compartments. They are measurable, predictable and well known for most drugs.

• Plasma pharmacokinetics consist of a distribution phase (absorption) and an elimination phase (redistribution/metabolism) that follows first order kinetics and can be modeled as the difference of 2 exponentials, one rate constant for the absorption phase and another for the elimination phase.

INJECTION

40 MIN

INJECTION

000

tttt inout eettucy

Non-linear curve fitting model used to determine drug effects and representative voxel time courses for positive and negative change in BOLD signal (kin - 2 to 10 min; kout - 30 to 120 min)

0.5 mg 1.0 mg 3.0 mg

Subjects were injected with 0.5, 1.0 or 3.0 mg of THC. Colorized areas represent those voxels where the mean percent area under the time-effect curve was significantly different than the null hypothesis at the p≤0.01 level with a minimum activation of 0.5% and minimum cluster size of 300 µl. Areas of positive changes in areas under the curve are colored in warm colors (i.e., oranges) and negative in cool colors (blues).

Effects of THC on Brain Activity

0.5 and 1.0 mg. THC 3.0 mg THC

Negative changes in AUC Positive changes in AUC

Superior temporal gyrus right anterior cingulatemiddle frontal and precentral gyrus superior middle and inferioranterior cingulate frontal gyrusinsula bilateral posterior cingulateCerebellum caudate

N. Accumbens

Positive changes in AUCmiddle frontal gyrus Negative changes in AUCposterior cingulate superior temporal cortexglobus pallidus insulaleft caudate cerebellumN. Accumbens

POSTERIOR CINGULATE

0.12

0.44

0.55

1.36

-0.20

0.30

0.80

1.30

1.80

0 mg .5 mg 1 mg 3 mg

Dose

CH

AN

GE I

N A

UC

FOREBRAIN and STRIATUM

-0.15

0.72

0.85

1.70

-0.50

0.00

0.50

1.00

1.50

2.00

0 mg .5 mg 1 mg 3 mg

Dose

Ch

an

ge in

AU

C

3 MG. THC

M. FRONTAL GYRUS

-0.38

0.680.73

1.54

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

0 mg .5 mg 1 mg 3 mg

Dose

Ch

an

ge in

AU

C

S. TEMPORAL GYRUS

-1.20

-1.00

-0.80

-0.60

-0.40

-0.20

0.00

0.20

0.40

0 mg .5 mg 1 mg 3 mg

Dose

Ch

an

ge in

AU

C

NUCLEUS ACCUMBENS•The reinforcing (rewarding) properties of THC are controversial.

•Although it is readily self-administered by humans, this is not always the case in other species including primates and rodents.

•A body of work by Gardner has demonstrated clear reinforcing properties of THC in an appropriate rat strain.

•We focused down on THC’s effects in the n. accumbens due to this region’s association with the reinforcing properties of abuse substances.

0.5 mg 1.0 mg

3.0 mg

EFFECTS OF THC ON ACTIVITY WITHIN N.

ACCUMBENSSubjects were injected with 0.5, 1, or 3 mg of THC. Colorized areas represent those voxels in which mean percent area under the curve was greater than the null hypothesis at p<.01 level.

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0.0 mg 0.5 mg 1.0 mg 3.0 mg

Dose

Perc

en

t C

han

ge S

ign

al A

cit

ivit

y:

Are

a U

nd

er

Cu

rve

*

****

Effects of THC on activation in the n. accumbens. Activation was determined by making a combined (OR) mask of the activations produced by all three doses in the accumbens. That mask was then used to to describe a functional region in which the mean percent area under the time-effect curve was calculated for each subject. Values shown are the mean ± the SEM. ** P≤0.05 compared to vehicle, * p≤0.05 compared to other doses.

Effects of THC on Brain Activation By Cognitive Tasks

Although there is extensive literature concerning imaging task-induced brain activation, there have been few studies of the effects of cannabinoids and other psychoactive drugs on this activation (O’Leary et al.). Marijuana and its primary psychoactive constituent, delta-9-tetrahydrocannabinol (THC) produce characteristic physiological and behavioral effects in humans, including impaired short-term memory and perceptual-motor function. The purpose of this study was to examine the effects of THC administration on brain activation produced by two cognitive tasks; a one-back visuospatial working memory (VSWM) task and a concept formation (CF) task and a simple motor task, finger tapping. Frequent marijuana users were recruited from the general public. After giving informed consent and careful screening for study eligibility, subjects were trained on the tasks and placed in the scanner. Functional imaging was carried out with a GE Signa 1.5 T scanner using echo planar imaging (TE=40 msec; TR=6 sec). Imaging runs for each task were 7.5 minutes in duration. Each task was performed prior to and after the iv injection of 1 mg of THC in a counterbalanced order.

• Data from each task were analyzed using a phase shifted input function representing the on/off active-rest pattern of task presentation.

• Activation intensity was calculated for each pixel by dividing the mean signal intensity during the active periods by the mean intensity during the "rest" periods.

• The resultant functional images were averaged across all subjects for each task to produce mean activation maps for each condition.

• Areas of significant activation were determined for the group on a voxel by voxel basis using a one sample, 3-D t-test against the null hypothesis of no change and a minimum cluster size of 300µ1.

• Comparisons between pre- and post-drug activation were made by determining the mean task activation within functional and standard brain region masks (AFNI).

Task Activation Data Analysis

Based on a task described by Levine (1966), tests for concept formation and working memory by asking the subject to make choices based on rules of selection that they must deduce. Four variables are used: Color (RED vs. BLUE), Size (BIG vs. SMALL), Letter (X vs. T) and Position (LEFT vs. RIGHT). Pairs of letters with these attributes are presented sequentially to the subject who must choose one of the 2 options. The subject is then given feedback by the computer with correct or wrong appearing on the screen. This continues for 4 presentations at which time the subject should have all the information to make the correct final determination of the sort criteria. Each sequence of 4 screens can be solved in 30 sec and is followed by a 30 sec rest period. This rest-task cycle is repeated 7 times in each imaging run.

T XCorrect

X TWrong

T XWrong

x TCorrect

Red“T”

LargeLeft Red

Left

Red

Time

4 sec.

4 sec.

4 sec.

4 sec.

CONCEPT FORMATION TASK

This task is modeled after a task that was described by Cohen (1993). The 2-back VSWM task presents fifteen outline squares, arranged in a pseudorandom fashion on the screen. During each 30 sec task block, one of the squares turns black sequentially every 1.5 sec in a random order. The subject is instructed to press a button if the same square is filled in black twice, separated by any other one square (i.e. a two-back working memory task). A 30 sec 'rest' period where only a central fixation cross was presented alternates with each 30 second active period.

VISUOSPATIAL WORKING MEMORY TASK

PARTICIPANT DEMOGRAPHICS

Gender 10 Males, 3 Females

Age Mean - 23.6, S.D. - 7.0

Range - 19 to 45

Years of Use Mean - 6.6, S.D. - 3.2

Range - 3 to 12

TASK PRE-DRUG POST-DRUG

Concept Formation

Accuracy - % correct

83 ± 5 87 ± 5

Reaction Time - msec

1084 ± 153 1055 ± 200

VSWM Accuracy - % correct

88 ± 3 84 ± 3

Reaction Time- msec

592 ± 21 586 ± 25

Effects of 1 mg of ∆9-THC on task performance

CF TASK

Effects of iv THC (1 mg) on activation by a concept formation task. Colorized areas represent those voxels in which the mean task activation was greater than 0.5% and was significantly different from the null hypothesis of no activation (p≤0.01). Oranges represent positive and blues negative activation.

Pre-Drug Post-Drug

CONCEPT FORMATION TASK

Re-Drug Activations Post Drug Decreases or L & R post parietal cortex lack of significant SMA activation

Cingulate (BA 32) middle front gyrus/DEPFC

L&R premotor (BA6) cortex post-parietalL & R premotor (BA6) cortex SMAR MFG including DLPFC (BA 46, 10) cingulateL MFG thalamusR & L lateral occipital cortex (BA 18, 19) ponsR. Thalamus cerebellumL & R cerebellum including declive, culmen and vermis New area – negative activationpons posterior cingulateNegative activationant. cingulate (BA 32)

VSWM TASKPre-Drug Post-Drug

Effects of iv THC (1 mg) on activation by a visuospatial working memory task. Colorized areas represent those voxels in which the mean task activation was greater than 0.5% and was significantly different from the null hypothesis of no activation (p≤0.01). Oranges represent positive and blues negative activation.

VISUOSPATIAL WORKING MEMORY TASK

•Effects of iv THC (1 mg) on activation by bilateral fingertapping paced by a 3 Hz flashing checkerboard. Colorized areas represent those voxels in which the mean task activation was greater than 0.5% and was significantly different from the null hypothesis of no activation (p≤0.01). Orange-yellow colors represent positive and blues negative activation.

Pre-Drug 15 min Post-Drug 35 min Post-Drug

MOTOR TASKBilateral Finger Tapping

RESULTS AND CONCLUSIONS1. The 1 mg dose of THC produced a moderate marijuana-like high. Performance

(accuracy and reaction time) on the CF AND VSWM task was not affected by this dose of THC.

2. THC administration decreased brain activation induced by each of the tasks. The effect was greater on the CF task than the VSWM task. The effects of THC on the finger tapping task were time-dependent. Both motor and visual cortex activations were significantly decreased at 15 minutes after drug administration. There was a significant return of activation at 35 minutes.

3. Overall, these data suggest that THC can decrease brain activation induced by cognitive and simple motor tasks and the effect is greater on a task with higher cognitive demands. The decreases in activation were not associated with decreased task performance. However, it is possible that this could change with higher doses of THC or tasks of greater difficulty. Possible explanations for the observed decreases in task activation include receptor mediated drug-induced alterations in neurovascular coupling or drug related behavioral mechanisms.

SUMMARYWe have detected THC-induced changes in both fMRI signal (neuronal activity) and brain activation by a cognitive task. Furthermore, the fact that we observed both increases and decreases in fMRI signal as a result of THC injection and regionally specific changes in task activation strongly suggests that they are not due to either global vascular or peripheral cardiovascular drug effects. Taken as a whole, these studies demonstrate that THC produces significant effects on regional brain activity and cognitive task-induced activation at a dose that produces effects that are similar to those produced by marijuana in a social situation; and that this dose range does not produce significant global effects on brain blood flow.

Acknowledgements

Research supported by NIDA grants DA11326 and DA09465 GCRC grant 5M01RR00058

Robert Risinger, M.D.Jeffrey Benson, M.D.S. J. Li, Ph.D.Kathleen LaGraveLinda Piacentine, M.S., ACNPRaymond Hoffman, Ph.D.Elliot Stein, Ph.D.