This month, we are going to talk about the world’s most used and abused drug, alcohol. Alcohol is so much a part of life in the Western world, that people often do not even notice that it is a drug. Given that alcohol is one of the most abused drugs in the world, it is of great importance to understand how it affects the brain. In this Dose of Science we are taking a look at alcohol’s effect on monkey brains.
A range of studies indicates that alcohol causes brain volume shrinkage in humans. In these studies there is a range of potential confounders such as dietary deficiency, comorbidity, stress, etc. which (for practical reasons) can not be controlled for. These potential confounding factors, however, can be controlled in animal experiments, in order to directly observe the neurotoxic effect of alcohol. This approach was taken by Christopher Kroenke and colleagues, who conducted a longitudinal MRI study to observe how the volume of a monkey's brain changes with exposure to alcohol.
Most neuroscience experiments are done with mice; there are certain advantages, however, to using monkeys to study the effects of alcohol. For example, they will take alcohol voluntarily, unlike rodents who must be forced into alcohol exposure. Furthermore, just like humans, monkeys show significant variation in their voluntary alcohol intake, with a not insignificant portion drinking to a point of physical dependence. These differences justify the use of monkeys over rodents in alcohol research. The downside of using monkeys is the much increased cost, and as a consequence, the lower sample size. In this study, only 18 rhesus monkeys participated. While this is a respectable sample size for studies using monkeys, it is still a very low sample size, associated with large statistical uncertainties.
Dosing of the monkeys was as follows: For an initial induction phase (3 months), monkeys took 1g/kg alcohol, which roughly corresponds to four drinks in human equivalent doses. In the next phase of the experiment, the monkeys had access to diluted ethanol, of which they consumed as much as they wished. During this free access period, all the monkeys consistently maintained a constant alcohol intake – that is, each monkey established his/her own drinking routine. The daily alcohol intake was 2.4±1.1 g/kg, within an overall range of 0.2 to 4.3g/kg.
For the purposes of analysis, the monkeys were divided into two groups: `heavy’ (> 3g/kg/day ethanol) and `moderate’ drinkers (≤ 3g/kg/day ethanol), with 6 and 12 monkeys in the two groups respectively. Note that the separation threshold between heavy and the moderate drinkers is arbitrary. The authors say that the 3g/kg ethanol is roughly equivalent to 12 standard US drinks (one US alcohol unit is 14g of ethanol; note that the UK alcohol unit is 8g).
To estimate the brain volume of monkeys, images were taken using magnetic resonance imaging (MRI) at the beginning of the experiments, again after 6 months, and after 12 months of free access to alcohol. To account for individual differences, volume measurements were expressed as percentages of the baseline volume (that is the volume before the monkeys started drinking).
For cerebral cortical grey matter, i.e. neuronal cell bodies, significant volume reductions were observed in the heavy drinking group, but not among the non-heavy drinkers. In contrast, the average volume of white matter, composed of the axon cables which connect neurons, was unchanged for both groups. Since the division between heavy and moderate drinkers was set at an arbitrary threshold, it is worthy examining whether the data is pooled together (I don't know what the text in italics is intended to mean – jld). In this case, after 6 months, the volume of numerous brain regions, including the cerebral cortex and the hippocampus, had shrunk. Somewhat confusingly, after 12 months, the shrinkage was not as significant in many areas as it was after 6 months of free access drinking. This fluctuation could feasibly be due to the low sample-size of the study.
To further examine the graded relationship between mean daily alcohol intake and volume reductions, correlation analysis was performed. A statistically significant correlation was found, with higher ethanol intake predicting larger volume reductions. This correlation strengthens the argument that alcohol is at least partially responsible for the volumetric effects. Correlation analysis can not, however, take into account nonlinear effects, therefore it remains a possibility that alcohol is not neurotoxic at low doses.
The monkeys participating in the experiments were young adults, or adults entering middle age. Age is a possible confounding factor for changes in brain volume, but multiple regression analysis indicated that age does not directly explain to the volume shrinkage observed here. The authors note, however, in line with human epidemiological studies, that younger monkeys had a tendency to exhibit the higher daily alcohol intake associated with larger volume reductions.
So what was learned from the paper, and what further research does it suggest? The paper's strength consists in its control for many potential confoundin factors which can not be addressed in human studies, while its weakness is the small sample size. The paper supports the hypothesis that alcohol is neurotoxic, and that this toxicity is responsible for the observed grey matter volume reduction. Given that this hypothesis is also supported by a number of human studies, it is likely that the observed neurotoxicity also translates to us.
The observed differences between heavy and moderate drinkers, plus the correlation between alcohol intake and volume reduction, argue that alcohol’s impact is dose-dependent. This finding is unsurprising, since the effects (both harmful and potentially beneficial) of all psychoactive substances change as a function of the dose.
Where the paper falls short, is in failing to identify the precise dose at which alcohol consumption leads to observable differences in volume. This is not the researcher’s fault, however, as such an investigation would require a higher sample size, unlikely to be obtained in the current regulatory climate. An analysis is also required, of whether the volumetric changes can be linked to any cognitive deficiencies. Monkeys are very intelligent, so a range of tests could have been done, to reveal whether the observed volumetric changes correlate with a decline in cognitive abilities.
[1]: Kroenke, Christopher D., et al. "Monkeys that voluntarily and chronically drink alcohol damage their brains: a longitudinal MRI study."Neuropsychopharmacology 39.4 (2014): 823-830.
