One of five deaths in the United States is attributable to smoking, which can cause an impressive list of diseases, according to the Journal of the American Medical Association, including: cancers of the lung, oropharynx, larynx, esophagus, pancreas, kidney, bladder and cervix; ischemic heart disease, arterial disease and other heart conditions; stroke; chronic obstructive pulmonary disease; and other respiratory conditions.

By and large, smokers are as aware of those statistics as anyone, and they act accordingly, to a point: The surgeon general's website states that "epidemiologic data suggest that more than 70 percent of the 50 million smokers in the United States today have made at least one prior quit attempt, and approximately 46 percent try to quit each year."

But only to a point: The success rate for a given attempt is 5 percent cold turkey, which can be raised to a still low 20 percent to 40 percent with a combination of nicotine replacement and behavioral therapy.

Ironically, many smokers are able to withstand the acute phase of nicotine withdrawal, which peaks within three days; what does them in are cravings - overwhelming urges to smoke that may suddenly occur weeks or months after quitting.

Not much is known about the brain circuits underlying drug cravings. But in the June 16, 2005, issue of Neuron, researchers from the University of Pennsylvania report that a transmitter known as the mu opioid receptor, and the consequent phosphorylation of the transcription factor CREB, is apparently involved in the cravings that can trigger relapse.

"Previous research had shown that animals without opioid receptors do not find nicotine rewarding," said Julie Blendy, associate professor in the department of pharmacology at the university. "Here, we show that a lot of the mechanisms activated by the drug are also associated by cues associated with the drug."

The researchers used an experimental approach, known as conditioned place preference, to evaluate whether blocking a specific opioid receptor subtype, the mu receptor, would also affect how rewarding the cues associated with nicotine were to the animals.

They found that repeatedly giving mice nicotine activated, via phosphorylation, CREB, a transcription factor known to control genes involved in certain other drug reward pathways. CREB levels subsequently also increased when the animals were put into the cage where they received nicotine, suggesting they had learned to associate that environment with the pleasurable effects of nicotine.

When the animals were treated with an opioid receptor antagonist prior to receiving nicotine, both the increases in CREB activation and the preference for the area they associated with nicotine were blocked. The effects were specific to nicotine; when the scientists gave the mice cocaine after blocking opioid receptors, CREB levels went up, and the mice showed a behavioral preference for the environment they associated with the cocaine.

"When we found these results with cocaine, it really emphasized the specificity of our data," Blendy told BioWorld Today. It also ruled out the idea that blocking opioid receptors was affecting memory - one of their known roles - rather than rewarding processes.

The researchers replicated the experiments using CREB knockout mice rather than the injection of opioid antagonists, and found what they termed "striking" similarities in behavioral results: The mice showed no conditioned place preference to rewarding doses of nicotine. Both the antagonist-treated and the CREB knockout mice showed a dislike for environments they associated with high doses of nicotine.

Dopamine is the reward and addiction neurotransmitter of record, and many pharmacological interventions for addiction focus on the dopamine receptor. The opioid/CREB system works somewhere upstream to fine-tune the dopamine system.

Clinical studies using opioid receptor antagonists to help people quit have been mixed. Blendy believes that might be due to the fact that "many of these studies were carried out in hospitals or in laboratories - certainly not in places where people would normally smoke." Treating in this artificial setting also means that the associations with smoking that blocking opioid receptors seem to affect also were not present. In their paper, the authors write that "clinical studies designed to evaluate administration of opioid antagonists just prior to cues associated with smoking could lead to a more promising treatment regimen."