Dendritic spine changes in the development of alcohol addiction regulated by α-calcium/calmodulin-dependent protein kinase II

• Main Author: Zofia Mijakowska
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2014-03-01
• Series: Frontiers in Systems Neuroscience
• Subjects: Dentate Gyrus; Addiction; alcohol; dendritic spine; mouse model; αCaMKII; cofilin
• Online Access: http://journal.frontiersin.org/Journal/10.3389/conf.fnsys.2014.05.00045/full

Introduction <br/> <br/>Alcohol has many adverse effects on the brain. Among them are dendritic spine morphology alterations, which are believed to be the basis of alcohol addiction. Autophosphorylation of α-calcium/calmodulin-dependent protein kinase II (αCaMKII) has been shown to regulate spine morphology in vitro. Here we show that αCaMKII can also regulate addiction related behaviour and dendritic spine morphology changes caused by alcohol consumption in vivo. <br/> <br/>Method <br/><br/> 12 αCaMKII-autophosphorylation deficient female mice (T286A) and 12 wild type littermates were used in the study. T286A strain was created by Giese et al. (1998). Mice were housed and tested in two IntelliCages from NewBehavior (www.newbehavior.com). IntelliCage is an automated learning system. After 95 days of alcohol drinking interrupted by tests for motivation, persistence in alcohol seeking and probability of relapse, mice were ascribed to ‘high’ or ‘low’ drinkers group according to their performance in the tests. Additional criterion was the amount of alcohol consumed during the whole experiment. Result of each test was evaluated separately. 1/3 of the mice that scored highest in each criterion were considered ‘positive’ for this trait. ‘Positive’ animals were given 1 point, negative 0 points. Mice that were positive in at least 2 criteria were ascribed to ‘high’ drinkers (‘+’) group. Remaining mice – to ‘low’ drinkers (‘–‘). This method of behavioral phenotyping, developed by Radwanska and Kaczmarek (2012), is inspired by DSM-IV. Since the results of this evaluation are discrete (i.e. by definition all the animals score between 0 to +4), we developed also a continuous method of addiction rating, which we call ‘addiction index’. The result of the second method is a sum of the standardized (z-score) results of the above mentioned tests. We use it to examine the correlations between addiction-like behavior and spine parameters. <br/>Control group (12 WT, 8 T286A female mice) was subjected to similar training in the Intelli Cages, but had access to sweet water (5% sucrose) instead of alcohol. <br/>After the behavioral experiments brains were cut into two parts. One was stained diolistically. Images of the dendrites from upper blade of the dentate gyrus were taken using fast confocal microscope. Spines were then automatically outlined and manually corrected using custom software SpineMagick!. We performed analysis of spine density and area. We also developed a new method of spine clustering. After assigning spines into three categories (long, stubby, mushroom), we analyzed spine head and neck length and width in each category separately. <br/>The remaining half of the brain was used for immunostaining for cofilin, phosphocofilin and PSD-95. Images were taken using confocal microscope. <br/><br/>Results <br/><br/> After long-term free access to alcohol in the automated learning system, one third of wild type mice developed strong addiction-like behaviour, characterised by high alcohol consumption, high probability of relapse, increased motivation for alcohol and high persistence in alcohol seeking even when it is unavailable. Remaining 2/3 of wild type animals started to avoid alcohol. Such polarization was much less pronounced in T286A mutants (Fig 1.). Furthermore, we found that in wild type animals addiction-like behaviour correlates with the size of dendritic spines in the dentate gyrus of the hippocampus (Fig 2.). Individuals that exhibited addiction-like behaviour had smaller dendritic spines as compared with their cage-mates that did not develop motivation for alcohol and with control animals exposed to sweet water. This difference in spine area was mostly due to mushroom spines neck shortening and narrowing. Changes in spine area and other analysed parameters observed in T286A animals were more pronounced and in the opposite direction. We also found that the level of an actin depolymerizing protein, cofilin, is higher in the dendritic spines of WT high drinking animals. T286A animals do not show this effect. <br/><br/>Conslusions <br/><br/> Our data suggests that formation of a stable attitude towards alcohol depends on αCaMKII autophosphorylation. We also conclude that ethanol has a great effect on dendritic spine morphology, which is buffered by αCaMKII. We hypothesise that ‘addiction memory’ in wild type animals is coded by both morphological and functional changes. In contrast, in T286A mice a drug memory is mostly formed via changes in spine structure, which is in line with our previous findings considering context memory generation in these animals (Radwanska et al. 2011). <br/>