Too Much of a Good Thing: Addiction as Overlearning
Addiction as Overlearning

by Brenda Patoine

November, 2006

The idea that drug addiction is a result of “learning gone wild” was bolstered by several reports at SfN revealing profound, drug-induced changes in the same neural circuitry the brain applies to learn useful behaviors.

One emerging view is that repeated drug use jacks up the brain’s plasticity, the trait that underlies our ability to learn and remember. This repeated use produces distinct and persistent chemical and structural adaptations in the brain that make it exceedingly difficult to “unlearn” the bad habit of taking drugs—and far too easy to relapse.

“Addiction usurps the brain circuitry that normally regulates learning behaviors aimed at obtaining biological rewards” such as food or mating, said Peter Kalivas, a psychologist at the Medical University of South Carolina. In a special lecture Oct. 18, Kalivas outlined the case for addiction as a form of overlearning, or, put another way, as “a disease of too much neuroplasticity.”

A cardinal feature of addiction—whether the drug of choice is nicotine, alcohol, cocaine, or methamphetamine—is the inability to stop taking the drug, even when taking it has disastrous consequences. Animal models have enabled researchers to unravel the brain pathways involved in this “compulsive relapse” phase of drug dependence to an unprecedented degree, yielding new insights about the underlying pathology. These drug-induced changes in neural circuits are now the focus of therapeutic development aimed at recalibrating the system and returning to the drug user the ability to “just say no.”

“If we can understand the neurobiology of drug relapse, we may be able to intelligently design effective treatments for addiction,” said Kalivas. “The goal of therapy would be to push the compulsive drug user back to occasional social use or to abstinence.”

At Risk for Relapse

Drug abuse is essentially a learned behavior, reinforced with each use by powerful brain chemicals that signal an intense reward. Even after long periods of abstinence, any number of triggers can cause a relapse. Psychologist Rita Fuchs of the University of North Carolina at Chapel Hill presented data from animal studies indicating that a common final neural pathway is activated during cocaine relapse triggered by exposure to the drug, drug-related cues, or environments associated with previous drug use.

This “relapse circuit” involves nerve cell connections linking the prefrontal cortex, a part of the brain associated with higher-order cognitive functions such as planning, reasoning, and decision making, to the nucleus accumbens, an evolutionarily older structure in the midbrain.

The prefrontal cortex, the brain’s cognitive control center, seems to be the lynchpin. Its normal role is to assess risk vs. reward in any given situation and inhibit impulsive behaviors that may not be in one’s best interest. Evidence from animal studies and substance abusers suggests that, over time, addictive drugs dramatically blunt neural activity in the prefrontal cortex.

“Decreased activity in the prefrontal cortex may lead to poor decision making and problems with impulse control, which in turn can predispose an individual to drug relapse,” Fuchs said.

Her team found significant reductions in the density of gray matter in parts of the prefrontal cortex of cocaine-addicted animals, suggesting a loss of neurons. Certain genes involved in frontal-lobe functions are also suppressed. One of these genes encodes for brain-derived neurotrophic factor (BDNF), a substance that supports and nourishes neurons.

“Drug-induced adaptations in the prefrontal cortex contribute to the maladaptive behavior of compulsive use, and BDNF may be one mediator of this,” Fuchs said.

Disrupting Dopamine

Other work supports the link between decreased prefrontal cortex activity and the inability to stop taking drugs. Neurophysiologist Antonieta Lavin, also of the Medical University of South Carolina, found dramatic changes in the firing patterns of prefrontal cortex neurons in live animals that had been repeatedly exposed to cocaine. The changes included slower firing overall and a destabilization of normal patterns.

Further investigation pointed to dopamine signaling as a possible cause. Dopamine, a neurotransmitter that plays a central role in learning, is also the messenger that cells in the nucleus accumbens use to talk to the prefrontal cortex. Dopamine acts on target neurons through two families of receptors, some of which activate the neuron to fire (D1 family) and some of which inhibit the neuron from firing (D2 family).

“Normally, the excitatory and inhibitory actions of dopamine signaling are perfectly balanced,” Lavin said. “But repeated exposure to cocaine throws off the balance of D1 vs. D2 receptor signaling, effectively increasing the amount of inhibitory messages that reach the prefrontal cortex.” She described the net result as akin to “having the brakes on permanently”—a persistent repression of cell activity in an area of the brain critical to decision making.

Monkeys on Meth

Using a different animal model of addiction, University of Southern California psychologist David Jentsch also discovered abnormalities in the dopamine system. His team assessed cognitive-control mechanisms in male vervet monkeys that had been given methamphetamine in a slowly escalating regimen to mimic human use of this increasingly popular street drug.

In behavioral tests, the monkeys exhibited a “massive deficit” in their ability to inhibit a conditioned response that they previously had learned—in this case, which of three cups emblazoned with different pictures held a fruit reward. It took an average of 50 trials before meth-treated monkeys were able to overcome the previously learned response and learn a new rule. This dysfunction in “response inhibition” was correlated with reductions in dopamine receptors (D2 family) and in the dopamine transporter, which helps remove excess dopamine from the synapses—the gaps between neurons—thereby ending its signal.

“Reductions in the D2-type receptors could be one mechanism by which methamphetamine impairs behavioral selection and cognitive control,” Jentsch said. “It is possible that these cognitive impairments are directly linked to the down-regulation of dopamine in this circuit.”

A Problem of Plasticity?

These studies support the central role of dopamine in the overlearning characteristic of compulsive drug use. In normal learning, Kalivas explained, dopamine “opens the brain to neuroplasticity,” enabling the brain to facilitate behavioral learning in response to a stimulus. As the stimulus is repeated and the new nerve connections stabilize, dopamine release drops off.

In contrast, drugs of abuse elicit a burst of dopamine every time they are used; there is no dropoff. This unrelenting flood of dopamine produces a potent memory of an intensely rewarding experience, which becomes more deeply ingrained with each use. After repeated exposure, benign social use of a drug progresses first to a state of “regulated relapse”—in which the user can still willfully choose to use the drug or not—and eventually to compulsive use, marked by the uncontrollable urge to get high again.

Important as it is, dopamine is only one part of the relapse story. Glutamate, the most prevalent excitatory, or neuron-activating, neurotransmitter in the brain, also plays a critical role.

“In animal models of addiction, there is a massive release of synaptic glutamate associated with the reinstatement of drug-seeking behavior,” Kalivas said. This outpouring of glutamate is markedly different than that triggered by a nondrug reward such as food or mating: it spills out of the synaptic gap between nerve cells and inundates a broader region surrounding the synapse, bathing adjacent nerve-cell fibers with “activate” messages.

Scientists revealed a decade ago that glutamate promotes the formation of dendritic spines, the bumpy outgrowths on nerve fibers that receive chemical signals from other neurons. Kalivas presented new data showing that drug-associated leakage of glutamate produces immature, relatively unstable spines at the expense of mature ones.  

These underdeveloped spines have a structure distinct from the classic, mushroom-shaped mature dendritic spine and are more adaptable. At the same time, there are profound alterations in proteins known to be involved in structural changes at nerve connections, and these alterations may contribute to a higher propensity to drug relapse.

“The big changes that are seen in … these animals seem to be evidence of an enhanced capacity for plasticity at these neural sites,” Kalivas said. “This makes glutamate transmission a prime target for therapeutic intervention.”
  If it were possible to tone down the excess glutamate release pharmacologically, the reasoning goes, this might sufficiently blunt the addict’s compulsive drive to seek out drugs.

One strategy relies on a supplement called N-acetylcysteine (NAC), a precursor to a naturally occurring amino acid (cysteine) that regulates glutamate levels in the brain. Animal studies have shown that NAC reduces relapse with cocaine and heroin. An initial study in 15 cocaine addicts found that NAC normalized blood flow in the prefrontal cortex when cocaine-associated cues were presented and significantly decreased cue-associated activity in the anterior cingulate, another structure in the relapse circuit.

“These results bode well for the clinical usefulness of NAC in addiction,” Kalivas said. His team is now initiating a second-phase clinical trial in a larger population of cocaine addicts.?