Until very recently it was thought that the
addictive drugs (
coke,
heroin, etc.) were
addictive because of their effects on the
brain's
dopamine pathways. They both -- in different ways -- increase the level of
dopamine available to stimulate
neurons in the
nucleus accumbens, one of the brain's so-called "
pleasure centers." Theory said that when the
addictive drugs were no longer being taken, the neurons in question didn't have enough stimulation, which directly led to the user's
craving of the
drug. Perceptively, the user feels horrible, and the
only thing that could make him or her feel better is more of their
drug of choice. This view is somewhat true, as the degree of
stimulation given by these drugs is unmatchable by a
sober state, but it now appears that there's more to
addiction than just that.
Cocaine, nicotine, the opiates, and other addictive drugs all cause heightened expression of an enzyme named Delta-FosB (pronounced fohz-bee). Delta-FosB is a transcription factor; it doesn't catalyze any reactions itself, but instead causes the expression of other genes. Among those genes is one specific to the nucleus accumbens, a glutamate receptor named GluR2 that is also sensitive to dopamine. Glutamate receptors are often responsible for learning -- when open they allow in Ca+ ions, which make the neuron more sensitive to whatever state it was in when the receptors were opened (for more on this, see NMDA). GluR2's sensitivity to dopamine means it's responsible for learning what brings pleasure. Another enzyme made to express by Delta-FosB is named Cdk5, and is responsible for desensitization of the neuron to dopamine. The more Cdk5 present in a neuron, the more dopamine needed to elicit a response.
The net effect is that the more of an addictive drug you take, the better you learn to like the surroundings, situations, and feelings of the drug, and the more you have to take of it (in the short term) to get the same feeling again. This wouldn't be so bad, except that Delta-FosB is an incredibly durable enzyme, and when its presence is jacked up into the stratosphere with drugs, it takes a long time to go away. With the Delta-FosB still around, the memory of the drug euphoria stays strong, as does the changed sensitization to the drugs effects. It's a recipe for addiction.
Backing up this theory is a study in which the gene for the dopamine reuptake channel of mice is knocked out. Cocaine blocks the dopamine reuptake channel, leaving an excess of dopamine in the synaptic gap, which makes cocaine use pleasurable. The researchers theorized that without a dopamine channel to block, the cocaine wouldn't cause pleasure, so the mice wouldn't self-administer it. Interestingly, even though there was no measured change in the mice's dopamine levels, they persistently self-administered it and became conditioned to the particular environment it was given in. As a caveat, cocaine also blocks serotonin and norepinepherine uptake channels, so these may have been responsible for the addiction.
Another study, done by the discoverers of the Delta-FosB - GluR2 link, inserted an extra gene that produced double the regular amount of GluR2 proteins specifically in the nucleus accumbens. These mice acted perfectly normal, except when given cocaine. On coke, they had 50% more motor activity, showing that they were dramatically more sensitive to it's effects. They also had a much higher propensity to remember (and seek out) the conditions and actions that had coincided with earlier cocaine exposure.
Knowing about all of this is cool, because it points to a way to possibly cure addiction diseases in the future. If there was a protein or other agent that blocked or destroyed Delta-FosB, the comparatively short-lived GluR2 and Cdk5 proteins would die back, leaving the addict without an addiction. While this wouldn't help the more physical symptoms of addiction -- heroin and alcohol withdrawal, etc. -- it would almost entirely end the craving, which is always the hardest part of an addiction to get through.