This is Archived content. To view the latest NIDA Notes go here.

Dendritic spine proliferation seems to compensate for some impacts of cocaine use.

October 01, 2010
Carl Sherman, NIDA Notes Contributing Writer

NIDA-funded researchers recently were surprised to find evidence that a cocaine-induced change in the structure of brain cells represents an adaptive response that may limit the drug's impact. Previously, scientists had suspected the opposite—that the modification contributed to the tenacity of some harmful effects.

Cocaine's acute psychoactive effects, such as the rush and high, occur because the drug disrupts the normal ebb and flow of neurotransmitter molecules that carry signals between brain cells. The drug's mechanisms for producing longer-lasting effects, such as craving and altered decision-making, are unknown. However, because those long-lasting effects persist after the drug leaves the brain, scientists have inferred that alterations of the brain cell structures that receive and process neurotransmitter signals might be involved.

A leading hypothesis has proposed that a structural change that accounts for cocaine's long-lasting effects is drug-induced proliferation of dendritic spines on neurons in the nucleus accumbens (NAc), a brain region involved in reward, motivation, and addiction. Increases in numbers of these knobby, neuroreceptor-rich structures accompany the development of cocaine addiction (see figure). New dendritic spines could render neurons more responsive to neurotransmitter stimulation and theoretically might form novel, overriding, semi-permanent signaling pathways that could underlie persistent drug-related behaviors.

Images showing reduced MEF2 activity in dendritic spinesBlocking MEF2 Increases the Proliferation of Dendritic Spines: RNA interference sequences that reduce production of MEF2 proteins result in more dendritic spines (the knoblike projections in the micrographs) along neurons of the nucleus accumbens. (Graph and photograph from Neuron 59(4); Copyright Elsevier, 2008.)

In a series of experiments with laboratory animals, Dr. Christopher W. Cowan and colleagues at the University of Texas Southwestern Medical School set out to determine cocaine's mechanism for increasing the number of dendritic spines and to test the dendritic spine hypothesis. Their unexpected result suggests that scientists will need to look elsewhere for an explanation of cocaine's persistent effects.

The work also revealed cocaine's mechanism for increasing dendritic spines, and that finding suggested a potential new medication strategy for treating cocaine addiction.

Revealing the Mechanism

The Texas team's investigations centered on a small family of proteins, myocyte enhancer factor 2 (MEF2), whose many functions include regulating the number of excitatory synaptic connections in the brain. Consistent with this role, several studies have shown that in some parts of the brain, the number of dendritic spines—which are sites of excitatory synapses—is inversely related to levels of MEF2 activity.

Other studies have shown that cocaine increases the activity of an enzyme, cyclin-dependent kinase 5 (Cdk5), that inhibits MEF2 activity. Accordingly, Dr. Cowan says, "We hypothesized that chronic exposure to cocaine would inhibit MEF2, which would lead to an increase in the number of dendritic spines and alter behavioral responses to repeated cocaine use."

In an initial set of experiments, Dr. Cowan and colleagues confirmed that cocaine reduces MEF2 activity in the NAc, and they mapped out the underlying molecular pathways (see box below). They then established that MEF2 inhibition is the link between cocaine and dendritic spine proliferation by demonstrating that:

  • In and of itself, reducing MEF2 activity in the NAc resulted in an increase in the number of dendritic spines there;
  • If cocaine is prevented from inhibiting MEF2 activity, it does not increase dendritic spines.

For the first demonstration, the researchers gave mice two injections, one into each side of the NAc. One of the injections contained virus into which the researchers had incorporated two RNA interference (RNAi) sequences that together shut down most MEF2 activity; the other, a control injection, contained a similar RNAi sequence that had been modified so that it would not affect MEF2. The researchers then gave some of the animals cocaine and the others saline daily for 4 weeks, then counted dendritic spines. The cocaine-treated animals had more spines than the saline-treated animals on the side of the NAc with normal MEF2 activity, but on the side with reduced MEF2 activity, cocaine did not cause spine proliferation beyond that observed in the saline-treated animals.

For the second demonstration, the researchers repeated the previous procedures with different viruses. This time, the experimental virus led to MEF2 activity that was resistant to inhibition by Cdk5. As anticipated, cocaine given daily for 4 weeks did not result in greater dendritic spine proliferation than saline on this side of the NAc, indicating that inhibition of MEF2 is needed for cocaine-induced spine changes to occur.

Scientists Reveal Biochemical Underpinnings of Cocaine Effects

The molecular events that link cocaine and behavioral changes that the drug engenders involve the protein family called myocyte enhancer factor 2 (MEF2) and the production of knobby, receptor-rich structures called dendritic spines. To date, Dr. Christopher W. Cowan and his team at the University of Texas Southwestern Medical School have identified the following biochemical ties:

  • Two ways in which cocaine reduces MEF2 activity in the nucleus accumbens (NAc). First, chronic cocaine exposure boosts the activity of an enzyme, called cyclin-dependent kinase 5, that adds phosphate groups to MEF2 proteins. This phosphorylation inactivates the proteins, which regulate the activity of various genes. Second, cocaine increases the abundance of dopamine. Dr. Cowan and his team have found that when dopamine binds to the dopamine-1 receptor, it inhibits the activity of an enzyme, called calcineurin, that removes phosphate from MEF2 molecules. More phosphate groups remain attached to MEF2 proteins, thereby lowering their activity.
  • Eighty-two genes that cocaine may inhibit by suppressing MEF2 activity. Each of these genes was found to be associated with MEF2 and so may regulate the number of dendritic spines. The proteins encoded by these genes can now be investigated to see whether reducing their abundance increases dendritic spine density or other cocaine effects. Some of these proteins have already been linked to cocaine's effects in other studies. One, for example, influences the activity of an enzyme linked to locomotor responses affected by cocaine.

A Surprising Finding

Satisfied that cocaine causes dendritic spine proliferation by inhibiting MEF2, Dr. Cowan and colleagues addressed the hypothesis that links an increase in dendritic spines to the behavioral effects of cocaine. Experimental mice were injected with the inhibition-resistant form of MEF2 proteins into the NAc on both sides of the brain. Since under these conditions, cocaine is prevented from inhibiting MEF2 and does not produce additional dendritic spines, the scientists did not expect the mice to exhibit behaviors associated with cocaine dependence, such as preference for a place where the drug had been administered and high locomotor activity in response to repeated doses of cocaine.

But the opposite occurred. Compared with control animals with normal MEF2, the treated mice showed a more pronounced locomotor response to cocaine and increased preference for a chamber where they had previously received the drug. "I was as surprised as anyone by our finding," Dr. Cowan says. At the very least, he adds, the findings show that increases in dendritic spine density and behavioral sensitization to cocaine are "functionally uncoupled—you can block the spine increase and still get the behavioral changes." Beyond that, the results suggest that dendritic proliferation in the NAc may represent an adaptation that limits the effects of cocaine. In the absence of dendritic spine growth, Dr. Cowan notes, the cocaine-sensitized behaviors did not simply remain the same—they increased.

The production of dendritic spines, Dr. Cowan suggests, might be the brain's attempt to rebalance a system that cocaine has thrown off kilter. The NAc neurons that his team examined receive excitatory signals primarily from the prefrontal cortex, where activity is suppressed by repeated cocaine use in both humans and rodents. Dendritic spine proliferation could help offset the impact of reduced NAc activity by amplifying the weakened signals. Dr. Jonathan Pollock, chief of the Genetics and Molecular Neurobiology Research Branch at NIDA, agrees that the increase in dendritic spine density "looks like a compensatory mechanism that dampens the effects of cocaine." The findings, he says, "raise the larger question of how synapses are strengthened or weakened in response to drugs."

By exploring the influence of MEF2 on dendritic spines, scientists might unveil potential targets for therapy, Dr. Pollock notes. If further research confirms that MEF2 inhibition blocks the effect of cocaine, pharmacologists could use this action as a criterion when screening compounds for potential antidrug activity.

The findings suggest "new directions for looking at synaptic adaptability and hints of how we can manipulate the process environmentally or chemically," Dr. Pollock adds.


Pulipparacharuvil, S., et al. Cocaine regulates MEF2 to control synaptic and behavioral plasticity. Neuron 59(4): 621-633, 2008. [Full Text]