henry wagner lectureship imaging the addicted brain:...

H E N R Y W A G N E R L E C T U R E S H I P

Imaging the Addicted Brain:From Molecules to Behavior

It is a pleasure and an honor for me to deliver this year’sHenry Wagner lecture. For most of my professionalresearch career, I’ve made use of nuclear medicine

imaging technologies to investigate the effects of drugs inthe human brain. Today, as director of the National In-stitute on Drug Abuse (NIDA), I often start my talks bynoting the dramatic escalation and wide-ranging tolls ofdrug abuse and addiction. In this country alone, the costsassociated with the consequences of drug abuse includingalcohol are estimated to be $486 billion a year. Addictionis one of the medical consequences of drug abuse, butdrug abuse also contributes significantly to the burden ofmany medical diseases. In addition, drug abuse and ad-diction have devastating social consequences that rangefrom loss of work, poor school performance, and familydisintegration to criminal behaviors.

Drug addiction is a disease for which we can targetthe vector: the drug. It is clear, however, that althoughchronic drug administration is a requirement for produc-ing addiction, drugs themselves are not the only variablesinvolved. Other variables are at work, some enabling andsome protecting against the process of addiction. Weknow that biology is extremely important––genetics canmake an individual more vulnerable or alternatively moreresilient to the effects of drugs. We also know that pastdevelopmental history, such as conduct problems whilegrowing up, may make some individuals more vulnerable.And we have come to understand that environmentalfactors––particularly stress––play an extremely importantrole in facilitating drug addiction.

Why do people take drugs? People take drugs becausethey want to change their mental state; they want to feelgood; or they want to feel better. Why is it that somedrugs may be expected to have such an effect? Years ofresearch, initially in laboratory animals, have shown usthat one of the characteristics that is indispensable fordrugs to have this effect is their ability to increase braindopamine concentration in areas that form part of thelimbic circuit. Research has shown that dopamine is aneurotransmitter that is involved in the regulation andmotivation of behaviors that are indispensable for sur-vival. So, for example, food, which we need in order tosurvive, increases dopamine, which in turn motivates anddrives us to learn that it is salient and to engage inbehaviors that result in obtaining food. Similarly, dopa-

mine provides the drive for sexual behaviors that areultimately necessary for the reproduction of the species. Italso drives the motivation for our gregarious behavior thatresults in the social interactions that also facilitate thechances of our species’ survival. Drugs of abuse target thesame mechanisms; that is, increasing dopamine, but theydo it at a greater magnitude and duration than naturalreinforcers. This ability of drugs to directly increase do-pamine––the same mechanism that nature uses to enhanceand motivate our behavior––is considered to be crucialfor their reinforcing effects. That is, drugs of abuse en-gage the neurobiological mechanisms by which natureensures that behaviors that are indispensable for survivalmotivate the procurement of more drugs and, with re-peated administration, can result in addiction.

Although increases in dopamine are important in re-inforcing the effects of drugs, by themselves such in-creases do not adequately explain addiction. Addiction, asdefined by the Diagnostic and Statistical Manual of Men-tal Disorders (IV), is the condition that emerges fromchronic drug use that leads to the compulsive administra-

Nora Volkow, MD, Director of the National Institute onDrug Abuse, delivered the Henry Wagner lecture at aplenary session on June 20 at the SNM Annual Meeting inPhiladelphia, PA.

NE

WS

LI

NE

Newsline 13N

by on June 5, 2020. For personal use only. jnm.snmjournals.org Downloaded from

http://jnm.snmjournals.org/

tion of the drug despite the fact that the subject may nolonger want to take it and even at the expense of seriouslyadverse consequences. However, if you were to givedrugs to a person who is not addicted, you would see anincrease in dopamine in his or her brain that would beequal to or even larger than that seen in the brain of anaddicted individual. So, the ability of drugs of abuse toincrease dopamine by itself does not explain the processof addiction.

What is the role of dopamine in producing the loss ofcontrol seen in the addicted person? My colleagues and Iat Brookhaven National Laboratory (BNL) have usedPET to investigate the involvement of the brain dopaminesystem in a wide variety of drug addictions. In this talk,I will concentrate on one of the proteins we have inves-tigated that is involved in dopamine neurotransmission;namely the dopamine D2 receptors. The dopamine cellsreside in the mesencephalon (Fig. 1), from which theysend projections to the areas of the brain they modulate.When dopamine cells fire, they release dopamine, and thismessage is transmitted by postsynaptic dopamine recep-tors. One of the dopamine receptors that has been shownto be important in the reinforcing effects of drugs of abuseis the dopamine D2 receptor. We have used: [18F]N-methylspiroperidol as well as 11C-raclopride as dopamineD2 receptor ligands to measure dopamine D2 receptors inaddicted individuals. We have shown that dopamine D2

receptor availability is significantly decreased across awide variety of types of drug addictions. Moreover, these

decreases are observed both during early drug withdrawaland after protracted drug detoxification, as shown in thebrain images obtained from a study that measured D2

receptor availability in cocaine abusers at 1 month and at4 months after last cocaine use (Fig. 2). The individualmeasures for D2 receptor availability are shown in thisgraph (Fig. 3) in green for the cocaine abusers and in pinkfor the controls. The measure of D2 receptor availability(obtained with [18F]N-methylspiroperidol) is plotted as afunction of age since dopamine D2 receptors in the braindecrease at a rate of approximately 4%–6% per decade.The graph clearly shows that, as a group, cocaine abusershave significant reductions in dopamine D2 receptors. Wehave found that these reductions exist whether subjectsare tested 1 week after their last use of cocaine or 4–6months after last utilization. Reductions in dopamine D2

receptors, then, are long lasting. Such reductions are byno means specific to cocaine. We and others have alsodocumented dopamine D2 reductions in alcoholic individ-uals with family histories of alcoholism. Similar findingshave been reported in heroin, and we recently reported thesame pattern for methamphetamine addiction (Fig. 4).

FIGURE 1. (A) Anatomy; (B) dopamine (DA) synapse; (C) DAD2 receptors.

FIGURE 2. Effect of cocaine abuse on dopamine D2 recep-tors. Top row: normal subject; middle row: cocaine abuser (1month after cessation of use); bottom row: cocaine abuser (4months after cessation of use).

FIGURE 3. Dopamine D2 receptors in controls and in co-caine abusers. Pink � normal controls; green � cocaineabusers.

FIGURE 4. Dopamine D2 receptors are lower in individualswith addictions.

NE

WS

LI

NE

14N THE JOURNAL OF NUCLEAR MEDICINE • Vol. 45 • No. 11 • November 2004



What does this reduction in dopamine D2 receptorstell us about their involvement in drug addiction? Con-sider the physiological function of dopamine cells. Fig. 5represents a dopamine synapse in a normal and in anaddicted individual. Dopamine cells function as a mech-anism to signal what is salient, so that we can shift ourattention to those behaviors that are extremely importantfor survival. What things are salient? Pleasure is salient,because that’s the way nature promotes and motivates usto eat, for example. Aversive things are salient, becausethat’s the way nature teaches us that certain objects,smells, tastes, or actions may be harmful. Things that arenovel or unexpected are salient, because by separating outthe novel from the routine we can optimize our resourcesand be on guard for discontinuities rather than constantlyscanning the environment. Dopamine self-fires to signalthese salient factors. The dopamine that is released intothe synapse has a limited time to interact with the recep-tors, because it is rapidly brought back into the terminalby the dopamine transporters. Mathematical modelingsuggests that any given dopamine molecule will stay inthe synapse less than 50 milliseconds. The probability ofdopamine interacting with a receptor in this brief timeperiod is a function of 2 factors: how much dopamine isliberated in the synapse and how many receptors areavailable. In a person who is addicted, the dopamine cellsmay fire, but the probability of an interaction is going tobe significantly reduced because receptors are signifi-cantly lower in such an individual. So, it is much lesslikely that the individual will experience an activation bya salient stimuli. In this way, the addicted person learnsthat natural reinforcers are no longer exciting or motivat-ing (the changes in dopamine are not large enough tosignal them as salient stimuli).

What about drugs? Drugs increase dopamine in bothquantitatively and qualitatively different ways from nat-ural stimuli. Increases in dopamine in the synapse are5–10 times greater with drugs than with natural reinforc-ers. Moreover, drugs such as cocaine, amphetamine, andmethamphetamine block the transporter that quickly re-cycles dopamine back into the terminal. The result is thatwith drugs, dopamine stays in the synapse for a longerperiod of time than for natural reinforcers. Thus, despite

the fact that the number of receptors is decreased in a drugabuser, the probability of interaction from a drug is veryhigh, not only because the dopamine concentration is verylarge but also because dopamine’s residence time in thesynapse is long. The drug abuser learns that while naturalreinforcers are no longer able to produce a signal ofsaliency, drugs of abuse do––a fact that drives and moti-vates subsequent behavior.

How do we know the extent to which these decreasesin receptors are the consequences of chronic drug admin-istration? How do we know that they were not therebefore the subject became addicted and, hence, renderedhim or her more vulnerable to addiction? The answer isthat we don’t know. The only way to answer this questionis to test the subjects before they become addicted––anextremely expensive and challenging undertaking. Yet,such an effort is important, because it addresses the issueof what makes a person more vulnerable to drugs ofabuse. We decided to investigate this question by lookingat the data from another perspective. Here I show theresults from another study where we measured dopamineD2 receptors with 11C-raclopride PET in cocaine abusersand in controls (Fig. 6). The results are the same; cocaineabusers (represented in green) have lower D2 receptoravailability than controls (represented in purple). In thiscase, however, focus on the normal controls. As notedpreviously, we lose D2 receptors as we age. But look atthe variability in the levels of dopamine D2 receptorsamong control subjects regardless of their ages. Individ-uals in their early 30s, for example, showed an almost50% variability in the availability of receptors.

The questions that follow from such an observationare: If low dopamine D2 receptors are associated withdrug addiction, what does it mean to be a person who isnot addicted but has low levels of these receptors? Howdoes that low expression affect an individual’s responsesto drugs of abuse? To address these questions, we con-ducted a relatively straightforward study. We took 23healthy controls and measured dopamine D2 receptor

FIGURE 5. (A) Non-drug abuser; (B) drug abuser.

FIGURE 6. Dopamine D2 receptors in controls and cocaineabusers (11C-raclopride). Purple � normal controls; green �cocaine abusers.

NE

WS

LI

NE

Newsline 15N



availability. We then administered intravenous methyl-phenidate, which is a stimulant drug that, like cocaine,increases dopamine by blocking the dopamine trans-porter, to each of the individuals and asked a simplequestion: how did they like the effects experienced withthis stimulant? Approximately 50% of the individualsliked the way that the injected methylphenidate madethem feel, and approximately 50% did not. This contrastswith the results we have obtained in cocaine abusers, allof whom (expect for 1 subject) reported the effects ofintravenous methylphenidate to be extremely pleasurable.Why then the variability among the normal controls? Wefound that this variability was in part related to the ex-pression of dopamine D2 receptors in these individuals(Fig. 7). In the upper panel is an image representingdopamine D2 receptor availability measured in a controlsubject who reported the effects of the drug as unpleasant,and in the lower panel is an image from a control subjectwho reported these effects as pleasant. The subject whoreported the drug as pleasant had lower receptor avail-ability than the subject who reported it as unpleasant. Tothe right of the figure is the plot of the individual datashowing that subjects that reported the effects of methyl-phenidate as pleasurable had significantly lower levels ofD2 receptors than individuals who reported the effects asunpleasant. (I might note that there are 2 outliers who didnot feel the effects of the drug that are extremely inter-esting but beyond the scope of this lecture.)

Individuals who reported the effects as unpleasant hadsignificantly higher levels of dopamine D2 receptor avail-ability than those who reported it as pleasant. Indeed, insome subjects in the latter group, receptor availabilitymeasures were indistinguishable from those that we haveseen in addicted individuals. Why is this an interestingfinding? First, it demonstrated the obvious fact that theeffects of a drug are not merely a function of the phar-macologic action of the drug but of the unique interactionbetween the drug and the biochemical characteristics of

the subject’s brain. In this case, the biochemical targetthat appears to modulate methylphenidate’s reinforcingeffects (as experienced by pleasurable responses) is thedopamine D2 receptors.

But why would subjects with high levels of receptorsreport the drug as unpleasant whereas those with lowerreceptor levels tended to report it as pleasant? The sim-plest of explanations is derived from knowledge gainedby studies in laboratory animals, where electrical stimu-lation of different pleasure centers in the brain has beenwidely used. One of those pleasure centers is the posteriorhypothalamus. It has been demonstrated that animals willpress a lever to deliver current into the posterior hypo-thalamus. What’s interesting, however, is that if the cur-rent is too weak, the animal will not press the lever,because the current is not sufficiently reinforcing to mo-tivate the behavior. If the current is too high, the animalstops pressing the lever, because the current becomesaversive. So, there appears to be an optimal window forelectrical stimulation to be perceived as rewarding. Toolittle and the sensation is not sufficient; too much and itbecomes aversive. Using these observations, it is possibleto arrive at a similar explanation for why human subjectswith high levels of dopamine D2 receptors experience theeffects of methylphenidate intravenously as unpleasant.For individuals with high levels of receptors, intravenousmethylphenidate produces a significant increase in syn-aptic dopamine, pushing them to a threshold at which theexperience is aversive. On the other hand, in individualswith low levels of D2 receptors, the low levels willattenuate the large dopamine increases induced by meth-ylphenidate, bringing it into the “pleasurable” windowlevel.

In human imaging studies, we often make associa-tions that guide us to the answers to significant questions.In this case, we’re postulating that perhaps what’s goingon is that high levels of D2 receptors make the experienceaversive because they exceed a specific threshold. Howdo we test this? Such an experiment should be simple.Would subjects who perceive the effects as aversive stilldo so if they were given only 1/10 of the original dose?Would lowering the dose significantly make the experi-ence pleasant? To test this hypothesis, we called theindividuals in our original study back for a low dose ofintravenous methylphenidate. But they refused, preciselybecause the original experience had been so aversive. Sowe don’t have an answer to this question. For now, wecan say that we believe that low levels of receptors eithermake individuals more vulnerable to taking drugs, be-cause the experience is pleasant and hence the probabilityof trying it again is going to be much higher, or, alterna-tively, that high levels of receptors may protect againstdrug abuse, since the reaction to the drug will tend to beaversive, decreasing the probability of taking the drugagain. One could interpret these results either way. Eitherlow levels of receptors make the individual more vulner-

FIGURE 7. Dopamine D2 receptors and response to intra-venous methylphenidate (MP) in controls. Subjects with lowreceptors report MP as pleasant and those with high recep-tors as unpleasant.

NE

WS

LI

NE




able or, alternatively, high levels of dopamine D2 recep-tors are protective against self-administration of highdoses of drugs of abuse (the hypothesis I favor).

How do you test this hypothesis? One method wouldbe to increase dopamine D2 receptors in the subjects whoreported the effects as pleasant. If the hypothesis is cor-rect that high levels of receptors are protective, then thesubjects would now experience the effects as unpleasant.The problem is that we don’t know how to noninvasivelyincrease dopamine D2 receptors in the human brain. Butwe can do this in animals. This is a perfect example of asituation in which an imaging technology such as PETcan provide information that can guide us to experimentsin animals that will help find causal associations fromfindings in humans. In this study from BNL, Dr. PeterThanos began by training rats to self-administer alcohol.He then stereotaxically injected into the striatum an ade-novirus into which he had introduced a dopamine D2

receptor gene. The adenovirus increased dopamine D2

receptors in the rat brain by 50%. The effect was not longlasting, because the adenovirus does not incorporate thegene into the chromosome. By day 10, the dopamine D2

receptor levels were back at baseline. On day 20, he againadministered the adenovirus with the receptor gene, andagain the receptors went up (Fig. 8). The question, ofcourse, is whether these changes in receptor levels mod-ified the levels of alcohol intake. The answer was adramatic “yes.” On day 4 after the initial administration,when the dopamine D2 receptors were at their highestlevels, alcohol intake was reduced by almost 70%. It wasnot eliminated entirely, but it was dramatically reduced.As the receptors returned to baseline, alcohol consump-tion returned to its previous levels. After the secondinjection of the adenovirus, alcohol intake was againdramatically reduced. (You can see in blue in Fig. 8 theresults from the animals that were treated with an adeno-virus that did not contain a gene, which was done to

ensure that the modification in behavior was not the resultof another factor, for example, an inflammatory reactionto the viral injection.) Over all, Thanos’ results showedthat increasing dopamine D2 receptors does, in fact, reg-ulate the self-administration of alcohol. He has recentlyreplicated these findings in cocaine administration. Inaddition, he has verified these findings in rats that havebeen inbred for a predisposition to self-administer alco-hol. In these animals with genes that make them vulner-able to consuming alcohol, increasing dopamine D2 re-ceptors protects them against consuming large quantitiesof alcohol. It appears that dopamine D2 receptors areactually interfering with the administration of high dosesof drugs––which is what drug addiction and alcoholismare all about.

What causes an individual to have high or low levelsof dopamine D2 receptors? Genes are the most obviousanswer. We all know this, and it is an explanation that hasbeen given for a number of years. I still believe that genesare an important contributor in determining the levels ofexpression of dopamine D2 receptors in the brain, butother factors must also be taken into account. The envi-ronment, for example, plays an extremely important role.Imaging technologies have allowed us for the first time tobegin to look at the significant interactions between en-vironment, brain neurobiology, and behavior. The exam-ple that I’m going to show is from Drake Morgan and hiscolleagues at Wake Forest. They measured dopamine D2

receptors in a group of macaque monkeys and then at-tempted to determine to what extent change in environ-mental surroundings could affect expression of these re-ceptors and subsequent self-administration of drugs. Whyare such questions important? Because epidemiologicstudies have indicated again and again that one of theenvironmental predictors of risk for drug abuse and ad-diction is poverty and its attendant social stressors.

Morgan’s study looked first at questions of socialhierarchy and dopamine D2 receptors and then at relatedquestions about drug self-administration. For primates,one of the most powerful drives is social reinforcement.In this study, each of the monkeys was raised in isolation,and dopamine D2 receptor levels were measured. Themonkeys were then placed in a group, where a naturalsocial hierarchy was allowed to form. Dopamine D2 lev-els were then measured again. The researchers were cu-rious about whether receptor levels in any way couldpredict placement in the social hierarchy. A previousstudy by the same group showed that dominant monkeyshad higher levels of dopamine D2 receptors than subor-dinates. The question here was whether hierarchical po-sition was somehow predetermined by levels of dopamineD2 receptors in the brain. The somewhat surprising an-swer was that they could not predict which animals wouldbe dominant or subordinate based on the receptor levelsmeasured when the monkeys still lived in isolation. In-stead, once the animals were placed in a group situation in

FIGURE 8. Effects of treatment with an adenovirus carrying adopamine (DA) D2 receptor gene into NAc in DA D2 receptors.Reprinted, with permission, from Thanos PK, Volkow ND, Frei-muth P, et al. Overexpression of DA D2 receptors reduces alco-hol self-administration. J Neurochem. 2002;78:1094–1103.

NE

WS

LI

NE

Newsline 19N



which a hierarchy evolved, dominant animals showedsignificant increases in dopamine D2 receptors in theirbrains, whereas subordinate animals did not (Fig. 9).Thus, if you compared these dominant and subordinateanimals in this group with no a priori knowledge, youwould conclude that dominant animals have significantlyhigher levels of dopamine D2 receptors. But this higherlevel was not there before they became dominant but wastriggered instead by an environmental social intervention.

What about their drug taking? The investigators tookthe same animals and put them into a cocaine self-administration protocol in which the animal pressed alever to get cocaine. Different doses of cocaine wereadministered. In Figure 9, the dominant animals are ingreen. None of the lever presses among these animalsreached statistical significance (i.e., none of the animalsself-administered cocaine to any significant level). Thehighest level chosen among these animals was actuallythe lowest dose available. Compare that with the actionsof the subordinate animals, which had low levels of D2

receptors. These animals readily self-administered rela-tively high doses of cocaine. This study showed not onlythe clear association between low levels of dopamine D2

receptors and vulnerability to drug abuse, but the apparentrole of high levels of dopamine D2 receptors as a protec-tion against the self-administration of high doses of drugs.This brings to light the very interesting suggestion that theenvironment, in this case a social variable, produces neu-robiologic changes in the brain. In this case, thosechanges occur in mechanisms that we know are importantin regulating the administration of drugs.

The effects of drugs are not limited to the devastatingconsequences of addiction. They also are significant con-tributors to other diseases. They can produce brain tox-icity either by their deleterious effects on blood flow inthe brain, or through direct neurotoxic effect to neurons.Drugs are currently one of the main contributors to the

AIDS epidemic across the world. Drugs also contribute tocancer, cardiopulmonary diseases, and mental illness, andnew data are showing that they may contribute to obesity.Brain imaging plays an extremely important role in tryingto understand the mechanism by which drugs of abusecontribute to these diseases and outcomes.

Let’s look at neurotoxicity and methamphetamine asan example. Methamphetamine has caused a great deal ofconcern, because animal studies have revealed that itdamages dopamine terminals and in some instances it canlead to dopamine cell death. We engaged in an investi-gation to determine to what extent the same types ofdamage were occurring in the brains of people abusingmethamphetamine. We also wanted to determine whetherdamage to dopamine terminals resulted in functional con-sequences. We found (and it has been validated by others)that by using a ligand that binds to dopamine transporteras a marker of terminals, in this case 11C-d threo meth-ylphenidate, that methamphetamine abusers experiencedsignificant reductions in dopamine transporters. This isvisible in the striatum in Figure 10. Moreover, we showedthat these reductions in dopamine transporters were func-tionally significant. The lower the transporters, the worsethese subjects performed in tasks related to motor speed,whether it was gross motor speed (how quickly they couldwalk a line) or very fine motor tasks (placing pins in littleholes). It was also associated with impaired memory; thelower the level of transporters, the harder it was forsubjects to remember words they had heard before. Meth-amphetamine, then, was neurotoxic to the human brain,and this was producing dysfunction in behavior.

What was also very intriguing and worrisome was thatthese types of abnormalities are also tied to those reportedin Parkinson’s disease, where you also see a reduction indopamine transporters (although the region in the striatumis slightly different). Studies of patients with Parkinson’sdisease show reductions in dopamine transporters linkedwith motor slowing and memory impairment. The ques-tion that follows is whether methamphetamine users are

FIGURE 9. Left: animals housed individually; right: animalshoused in a group. Reprinted, with permission, from MorganD, Grant KA, Gage HD, et al. Social dominance in monkeys:dopamine D2 receptors and cocaine self-administration. NatNeurosci. 2002;5:169–174.

FIGURE 10. Dopamine transporters in methamphetamineabusers. Top: normal control; bottom: methamphetamineabuser.

NE

WS

LI

NE




putting themselves at higher risk of developing neurode-generative diseases such as Parkinson’s when, as part ofaging, they will lose more dopamine terminals. Onewould predict that methamphetamine abusers would be athigher risk for Parkinsonism if the changes are irrevers-ible. However, if the changes in the dopamine terminalsrecover after cessation of abuse, perhaps these individualsare not inevitably at risk.

This, again, is something that could be evaluated withimaging. It is extremely difficult to get methamphetamineabusers to stop taking this very addictive drug. We lookedat a group of 15 parolees, for whom abstinence from druguse was a monitored condition of continued freedom fromjail. Only 5 individuals were able to stay clean, despite theserious consequences––a powerful indicator of how ma-lignant drug addiction can be. These 5 individuals whostayed methamphetamine-free underwent PET imaging,as did a group of controls. To my surprise, we sawsignificant recovery in the expression of dopamine trans-porters in the brains of these individuals after 9 months ofdetoxification (Fig. 11). All 5 of the subjects showed anincrease in transporters in striatum. However, the recov-ery of motor function did not reach this level of sig-nificance. Recovery in the biochemical parameter wassignificant, with a somewhat less marked trend towardimprovement in motor function.

What does this mean? This evidence suggests thatrecovery of normal biochemistry may not be sufficient torestore previous normal function. Indeed, although thedopamine transporters showed significant recovery, brainglucose metabolism in many areas remained deficient.This was measured in a crossover study designed toattempt to get information on a larger sample of detoxi-fied methamphetamine abusers. This is shown in theseimages of the areas of the brain in which methamphet-amine abusers had significantly lower metabolic activitythan comparable controls (Fig. 12). The top panel showssubjects who had been detoxified for less than 6 months,and the bottom panel, subjects that had been detoxifiedfor 6–36 months. In the early detoxification group, met-

abolic activity decreases are seen throughout the caudate,nucleus accumbens, thalamus, and mesencephalon. In theprotracted detoxification group, significant recovery canbe seen in the thalamus, but activity in nucleus accum-bens remains significantly decreased, even in individualsas much as 3 years into detoxification. These reductionsin activity in the nucleus accumbens are likely to play avery important role in the dysphoria, depression, andanhedonia that methamphetamine abusers continue to ex-perience even after months of detoxification.

However, I’m going to discuss results that I normallywould not show, because I do not know yet their func-tional significance. Fig. 13 shows, in the same group ofmethamphetamine abusers, the areas of the brain that hadincreased metabolic activity when compared with controlsubjects both during early and protracted detoxification.During early detoxification, we see significant increasesthroughout the whole posterior parietal cortex, includingthe precuneus, and the hyperactivity in the precuneusremains, even after protracted detoxification. This poses achallenge––it’s much easier to interpret decreases in ac-tivity than increases. The precuneus is an area that basi-cally has been found to regulate the levels of arousal.

FIGURE 11. Effects of methamphetamine detoxification ondopamine transporters and on motor activity.

FIGURE 12. Effects of methamphetamine detoxification onbrain metabolism.

FIGURE 13. Comparison of metabolic activity in brain im-ages from methamphetamine abusers during early and pro-tracted detoxification with brain images of control subjects.

NE

WS

LI

NE




When an individual is in a coma, for example, this areabecomes extremely hypoactive. The greater the alertness,the greater the activity in this brain region. Indeed, anes-thesiology studies have shown that the depth of anesthesiavaries inversely with the activity in the precuneus. Thegreater the depth of anesthesia, the lower the activity inprecuneus; the shallower the anesthesia, the greater theactivity in this brain region. Looking at the seeminglyparadoxical images of increased activity in the precuneusled me to speculate that if this area is important forarousal, then these individuals should have serious diffi-culty falling asleep and significant sleep disturbances. Dothey? I don’t know. No one has looked at this question.Clinicians have made anecdotal comments about meth-amphetamine abusers reporting sleep disturbances, but noone has systematically looked at this issue. I’m presentingthis possibility, not because I know the answer to thequestion, but because it is an example of one of the manyways in which imaging can alert us to the consequencesof drugs of abuse and point toward consequences thatmay have not been recognized.

Drugs target not only the human brain but are distrib-uted throughout the entire body. As a result, they contributeto morbidity and mortality in a wide variety of diseases.Of utmost concern has been nicotine, because it’s a majorcontributor to lung cancer. But it’s also recognized thatsmokers have a much higher probability of developingcancers in many other areas of the body. This study byJoanna Fowler and colleagues at BNL illustrates thepower of imaging technology, in this case PET technol-ogy, to provide us with mechanisms that can help usunderstand the adverse consequences of drugs. Thesestudies assessed the effects of cigarette smoke on theconcentration of monoamine oxidase in the human body.Monoamine oxidases are important in the brain to metab-olize to catecholamines and regulate their concentrations.In the body, monoamine oxidases are important in help-ing to detoxify substances. In Fig. 14, you see the con-

centration of monoamine oxidase B in a normal person:high levels in the brain, kidney, heart, and lower levels inthe lung. On the right is the concentration of monoamineoxidase B in a cigarette smoker––the image speaks foritself. Cigarette smokers have a dramatic reduction in theconcentration of this enzyme not only in their brains butthroughout their bodies. Nicotine is not the cause of thiseffect; instead, it is brought about by one of the chemicalsin cigarette smoke. It is widely believed that the reasonthat cigarette smokers are at high risk for cancer of thelung is because carcinogenic compounds in smoke accu-mulate in the lung. These data indicate that the chemicalsin smoke end up not only in the lungs but throughout theentire body, providing an explanation as to why a widevariety of cancers (bladder, pancreas, kidney) are muchmore frequently present in cigarette smokers than in non-smokers.

ConclusionIn the field of drugs of abuse, multiple studies have

shown that drugs of abuse affect genes, protein expres-sion, neuronal circuits, and behavior, and, of course, havesocial consequences. However, these studies by them-selves cannot explain the process of addiction. Part of thechallenge is to integrate all that we know and havelearned across the different levels of analysis. I haveprovided you with some examples that indicate that nu-clear medicine imaging technologies are tools that allowus to do just that. We can now use imaging technologiesto investigate how genes affect protein expression, howprotein expression affects neurobiology, how neurobiol-ogy affects behavior, and how that, in turn, affects socialinteractions. But we can also begin to use imaging tech-nologies as I illustrated in the study by Morgan to reversethe investigation and study how the environment, such associal factors, can modify elements of neurobiology, suchas protein expression. This knowledge is extremely im-portant. As we gain more information about how genesmake individuals more vulnerable or more protected, thisknowledge will allow us to perform interventions to pro-tect those most at risk. Ultimately, predetermination doesnot mean predestination. And, it is in our capacity now todevelop the knowledge that will allow us to identify andprotect those who are most vulnerable.

I want to thank my colleagues at Brookhaven NationalLaboratory without whom this work would not have beenpossible, in particular Joanna Fowler, Gene-Jack Wang,Yu-Shin Ding, Peter Thanos, John Gatley, and Christo-pher Wong. I also want to thank the Department ofEnergy and the National Institute on Drug Abuse forsupport.

Nora D. Volkow, MDDirector, National Institute on Drug Abuse

National Institutes of Health

FIGURE 14. Cigarette smokers have a dramatic reductionin the concentration of monoamine oxidase not only in theirbrains but throughout their bodies.

NE

WS

LI

NE




2004;45:13N-24N.J Nucl Med. Nora D. Volkow Imaging the Addicted Brain: From Molecules to Behavior

http://jnm.snmjournals.org/content/45/11/13N.citationThis article and updated information are available at:

http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 2004 SNMMI; all rights reserved.


http://jnm.snmjournals.org/content/45/11/13N.citation

http://jnm.snmjournals.org/site/misc/permission.xhtml

http://jnm.snmjournals.org/site/subscriptions/online.xhtml


henry wagner lectureship imaging the addicted brain:...

Documents