Skip to main content

Volume 20, Number 2, June 2009

It’s been fourteen months since I walked into my first Alcoholics Anonymous meeting. It’s about thirteen months since I admitted that I am an alcoholic and I am powerless over alcohol. It’s about twelve months since I started working the Twelve Steps. [Then] why did I buy the bottle of Margaritas and drink it on December 21st? . . . . As soon as the bottle touched my lips I hated myself. How could I be doing this to myself again after all I’d been through in the past year? I wanted to throw the bottle through the window and get far away from it but I knew I would hug it near me until it was empty.
From “An Alcoholic Relapses” by Vicky S.

In alcoholism, the frequent occurrence of relapses such as the one Vicky S describes is the norm rather than the exception. Many alcoholics try to stop drinking and fail. Half to two thirds of alcoholics who attempt abstinence relapse within months of initiating treatment. Even alcoholics who have been abstinent for years are prone to relapses. Most alcoholics relapse several times before achieving complete recovery. Alcoholic relapses can devastate alcoholics and their families, cause or perpetuate serious health problems, and result in lost jobs. Although relapses often seem sudden and unpredictable, with causes inscrutable even to the alcoholic, scientific research suggests that relapse is provoked by specific, identifiable factors. Identification of these factors could help to pave the way for developing interventions that can decrease the probability of relapse. Dr. George Breese, Professor of Psychiatry and Pharmacology at the Bowles Center for Alcohol Studies, is at the forefront of efforts to elucidate the causes of relapse and to discover means of preventing it. A productive and respected leader in alcohol research with approximately 400 publications to his name, Breese uses animal models to study mechanisms and manifestations of alcohol withdrawal and relapse.

Breese Lab (left to right): Kui-Ling Huang, Man Huang, M.D., Ph.D., Tiffany Wills, Ph.D., Zhen Ming, M.D., Marc Weinberg, Ph.D., Sarah Sinnett, Buddy Whitman, M.S., Darin Knapp, Ph.D., Katie Kelm, Ph.D., Thomas McCown, Ph.D., Bonita Blake, DVM, Ph.D., Hugh Criswell, Ph.D., George Breese, Ph.D., David Overstreet, Ph.D., and Bob Angel, M.S.

Breese and his colleagues are particularly interested in the emotional factors that can lead to craving and relapse. Working with Drs. Darin Knapp and David Overstreet, Breese established several years ago that multiple withdrawals from moderate alcohol exposure causes kindling of negative affect (anxiety and dysphoria). In kindling, repeated stimulation sensitizes brain circuits such that they are more easily activated in response to subsequent stimulation. This sen-sitization is reflected in the finding that behaviors or emotional responses mediated by these brain circuits are more easily elicited. Breese found that moderate alcohol drinking was followed by a mild negative affect (anxiety-like reductions in social interaction) during alcohol abstinence that could be sensitized (kindled) by exposing rats to three 5-day regimens of alcohol-containing diet with two days of abstinence between each regimen. The Breese laboratory measures negative affect in rats with the social interaction test, which quantifies the level of social behavior between pairs of rats. Rats are normally very social animals; decreases in social interaction behavior among rat pairs in the social interaction test is interpreted as reflecting the presence of anxiety and negative affect. Rats exposed to repeated withdrawals showed more negative affect in the social interaction test than rats exposed to the same concentration of alcohol given continuously over 15 days with only one period of abstinence at the end of alcohol exposure. After the negative affect associated with repeated withdrawals had abated, re-exposure to a single 5-day treatment and withdrawal a week later resulted in the same degree of negative affect as measured after repeated withdrawals. The latter finding shows persistence of the repeated-withdrawal effect and is consistent with the presence of adaptive changes in brain circuitry.

Recently, Breese’s group has been evaluating the effects of stress in their alcohol model. Breese hypothesizes that alcoholics’ unique response to stressors is an important factor in relapse. He notes that, in human laboratory studies, exposure of alcoholics to stressful environmental or situational cues associated with alcohol elicits negative affect with reactions such as fear, sadness, anxiety, and anger. In the alcoholic, these affective responses are associated with craving for alcohol, one of the primary triggers of relapse. Unlike alcoholics, most healthy individuals do not have these negative affective responses when exposed to the same alcohol-related environmental or situational cues. Breese holds that the affective susceptibility of alcoholics to stressful environmental cues during abstinence—a primary basis for alcoholics to return to abusive drinking—reflects a persistent change in brain circuitry caused by alcohol abuse over time. Consistent with this possibility, Breese and his colleagues found that stress could substitute for the first two cycles of alcohol to sensitize withdrawal-induced negative affect. That is, repeated stresses followed by 5 days of alcohol exposure induced negative affect (as indexed by a reduction in social interaction) during withdrawal, but exposure to the repeated stresses in the absence of alcohol exposure did not induce negative affect. What’s more, they found that stress exposure after repeated alcohol withdrawals sensitized stress-induced negative affect 3 days after alcohol withdrawal. This effect was not observed without previous alcohol exposure. Like the alcohol cycles, stress appears to have caused a persistent change in brain circuitry that led to increased susceptibility to withdrawal-induced negative affect, which in the alcoholic can stimulate craving and trigger relapse. The interchangeability of stress and a withdrawal episode could help to explain why abstinent alcoholics often relapse to drinking when they are stressed.

Breese and his coworkers next investigated the mechanism of the effect of stress in the repeated withdrawal model. They first targeted corticotropin releasing factor (CRF). CRF, which occurs naturally in the body, is a peptide long known to be involved in stress responses. Breese reasoned that, if CRF is involved in mediating stress sensitization of negative affect, then administration of CRF should mimic the effect of stress in their repeated withdrawal model. In fact, when repeated CRF was substituted for stress prior to 5 days of alcohol exposure, this CRF/withdrawal protocol sensitized negative affect during alcohol withdrawal. That is, repeated administration of CRF followed by 5 days of alcohol exposure reduced social interaction behavior during withdrawal, but repeated administration of CRF in the absence of alcohol exposure did not affect social interaction behavior.

In an elegant set of experiments, Breese and his colleagues also demonstrated that a CRF antagonist—a drug that blocks the typical actions of the body’s CRF—prevented sensitization of withdrawal-induced negative affect when it was administered during the sensitization process. Remarkably, the CRF antagonist significantly reduced negative affect after the third withdrawal even though it was administered only during the first two withdrawals or during the first two repeated stresses that substituted for the withdrawals. These findings demonstrate that pharmacological reduction of negative affect during early withdrawals or early exposure to stresses can mitigate the worsening of withdrawal-associated negative affect that occurs over repeated cycles of intoxication and withdrawal. That CRF sensitized negative affect in the withdrawal paradigm while CRF antagonists prevented this sensitization provides strong evidence for involvement of CRF in kindling of negative affect during alcohol withdrawal.

Breese and his colleagues next studied cytokines in their model. Cytokines are signaling molecules that, like CRF, have been demonstrated to be important in stress responses. Breese and his colleagues found that the restraint stress that they used in their stress/withdrawal protocol increased cytokines in rat brain. When repeated cytokines (MCP-1, TNFa, IL-1b) were substituted for stress before alcohol exposure, this cytokine/withdrawal protocol (like the CRF/withdrawal protocol) sensitized negative affect during alcohol withdrawal (see Figure). Considered in aggregate, these findings are consistent with the possibility that cytokines and CRF activated by stress contribute to stress-induced negative affect during alcohol withdrawal.

“One of the most important factors in alcoholism and alcohol research is the role of stress as a precipitant of craving and relapse,” says Breese. “Stress may contribute to relapse through a kindling-like process whereby it sensitizes negative affect. The increase in negative affect may promote heavy drinking and contribute to the progression to alcoholism. But we can’t stop there. It is critical for us to understand exactly how stress interacts with chronic alcohol exposure to sensitize negative affect. Our studies show that stress causes release of cytokines and CRF in brain and that cytokines and CRF can substitute for stress to sensitize negative affect in our model. We are now beginning to look at particular brain areas where cytokines and CRF act on specific receptors to produce sensitization of alcohol-withdrawal-induced negative affect. We will be using multiple approaches, including behavioral, biochemical, and electrophysiological studies, to tackle this problem. If we can understand the specific mechanisms by which stress leads to craving and relapse, we can begin to target anti-relapse interventions to help Vicky S and millions of alcoholics like her in their quest to remain abstinent and healthy.”

Figure: Repeated MCP-1 administration (100 ng/5 ml, i.c.v.) sensitizes ethanol withdrawal-induced anxiety. Rats were injected twice at weekly intervals with either vehicle or MCP-1 (or the cytokines IL-1b or TNFa) while drinking control liquid diet (CD) and then were either continued on CD or switched to a 4.5% ethanol liquid diet (ED) for 5 days. MCP-1 was also given twice in a group exposed only to CD to test for an effect on social interaction in the absence of the ethanol exposure. MCP-1 (monocyte chemo-attractant protein-1); IL-1b (interleukin-1beta); TNFa (tumor necrosis factor-alpha); Veh (vehicle: artificial cerebrospinal fluid). *p<0.001 compared to CD-Veh or ED-Veh groups. Adapted from Breese et al. (2008)Neuropsychopharm 33:867.