Dietary influences on alcohol intake: a review.

Alcohol intake is affected by both environmental and inherited biological mechanisms. In the early history of alcohol research, caloric interactions were among the most intensely studied of the environmental factors, and a large body of evidence was obtained that indicated that nutrition can affect alcohol drinking. Much of this evidence still stands today but appears to have been largely overlooked and to be in danger of being forgotten. Nevertheless, alcohol is not only a pharmacological substance but also a nutrient. As a nutrient it influences the metabolism of most tissues in the body, with especially marked effects on glucose homeostasis. Alcohol has a high energy content, and this energy is utilized by the body as efficiently as the energy in normal food. Ethanol has such good properties as a substrate for energy production that we are faced with the problem of explaining, not why it is consumed, but why it is not consumed in still larger quantities by nonalcoholic humans or by animals. When alcohol is consumed by animals, the intake of food decreases in relation to the caloric content of the alcohol; if a choice of macronutrients is possible, alcohol decreases the consumption of carbohydrates most. The interaction between alcohol and food intake goes both ways, however, with the intake of different foods also influencing alcohol consumption. For example, a high carbohydrate/low protein diet depresses alcohol intake, whereas a low carbohydrate/high protein food increases it. If such specific diets can help to depress alcohol intake, nutritional therapy might be useful in the treatment of alcohol abusers, probably not as the primary treatment, but perhaps as an adjunct to other forms of treatment.

Hypothesis: factors involved in the mechanisms regulating food intake affect alcohol consumption.

In addition to being a pharmacological agent, alcohol (ethanol) can also be considered a food; the body can utilize effectively its calorific contribution. The consumption of alcohol has in many respects the same characteristics as the intake of food. In animal experiments, food intake decreases in relation to the calorific value of the alcohol consumed; in human studies, various results have been obtained from no compensation to full compensation for the contribution from alcohol. If a choice is possible, primarily the carbohydrate part of the diet is decreased for compensation to ethanol intake. Food, in turn, can influence alcohol consumption. There is evidence that it is the carbohydrate content of the diet which influences alcohol drinking. The daily intake of alcohol is equally or even better regulated than intake of carbohydrate, protein, or fat. There is a circadian rhythm of alcohol consumption which resembles the rhythm of food intake. However, the mechanism which regulates alcohol intake does not have a very strong influence on the behaviour of most animals or humans, and its control is often overpowered by outside factors. Nevertheless, it is important to identify the precise mechanism because a better understanding should provide valuable information for use in the search for risk markers and pharmacological treatments for alcoholism.

Alcohol elimination and the regulation of alcohol consumption in AA and ANA rats.

Previous studies have usually found that animals with either higher alcohol elimination rates or ADH (alcohol dehydrogenase, EC1.I.I.I) activities have higher voluntary intakes of alcohol than ones with lower elimination rates. This relationship has now been studied in the AA and ANA rat lines genetically developed, respectively, for high and low alcohol consumption. Female AA and ANA rats had their alcohol elimination rate measured before being given a free choice between 10% (v/v) alcohol and water for 3 weeks. The elimination rate was then measured again and liver ADH activity was determined. The alcohol elimination rate was significantly higher in AA than ANA rats before drinking and was increased by alcohol drinking in AA but not ANA rats. ADH activity was similar in both lines and unrelated to either alcohol drinking or elimination rates, suggesting that the enzyme activity is not a rate-limiting factor in the alcohol metabolism of these two lines. The present results support the conclusion that alcohol elimination and alcohol consumption are partially determined by genetics. Furthermore, although alcohol elimination itself probably does not have direct control over drinking, some factor related to the alcohol elimination rate appears to be among the mechanisms influencing the level of alcohol drinking.

Dietary regulation of voluntary alcohol consumption in rats. Influence of a high protein diet and a methylene blue diet.

The importance of glucose homeostasis for high voluntary alcohol consumption was studied in alcohol-preferring (AA) and alcohol-avoiding (ANA) rats fed either a control diet, a protein-rich diet or a control diet supplemented with methylene blue. AA rats on the control diet were found to receive 13.6% of their daily energy intake from alcohol. On the high-protein or methylene blue diet, the alcohol consumption of the AA rats was respectively 40% and 48% higher than on the control diet. The voluntary alcohol consumption of ANA rats corresponded to 0.8-2.3% of their daily energy intake irrespective of diet. The protein diet increased the blood glucose concentration of AA rats by 20% but no increase was observed after the methylene blue diet. The diets had no effect on the blood glucose levels of ANA rats. In AA rats, the protein diet reduced the hepatic concentration of the three major glucogenic amino acids (serine, glycine, alanine) on average by 24%, suggesting an increased utilization for gluconeogenesis. No such reduction was observed in AA rats on the methylene blue diet or in ANA rats on any diet. The utilization of amino acids for maintenance of glucose balance in AA rats is further supported by the observed negative correlation between plasma concentration of urea, the end product of amino acid catabolism, and the sum of the concentrations of the three glucogenic amino acids in the liver, and by the positive correlation between plasma urea and blood glucose concentration. Furthermore, in AA rats, but not in ANA rats, the concentration of alanine, the main amino acid used in gluconeogenesis, correlated negatively with the amount of alcohol consumed. These findings indicate that the maintenance of glucose homeostatis is important for high voluntary alcohol consumption.

Hepatic carbohydrate metabolism in rats bred for alcohol preference.

The influence of ethanol on the carbohydrate metabolism was studied in two strains of rats: the AA strain with an inherited preference for alcohol and the ANA strain with an aversion to alcohol. In both strains, a single intraperitoneal dose of ethanol (1.5 g/kg body wt.) slightly increased the blood glucose concentration. In AA rats alcohol increased the rate of gluconeogenesis from alanine and had no effect on the liver glycogen stores, whereas in ANA rats the rate of gluconeogenesis remained unchanged and the glycogen stores decreased. It thus appears that the two rat strains maintain their blood glucose concentration by different mechanisms; the ANA rats utilise both glycogenolysis and gluconeogenesis but the AA rats only gluconeogenesis.

Protein, carbohydrate, and ethanol consumption: interactions in AA and ANA rats.

Female rats of the alcohol-preferring AA line were given a choice between protein, carbohydrate, and fat sources. Subsequent free access to alcohol decreased their intake of carbohydrate but not of protein or fat. The individual alcohol intakes were negatively correlated with the carbohydrate intakes (-0.76) and with the alcohol-induced changes in carbohydrate intake (-0.83), but positively correlated with the concurrent protein intakes (0.78). The alcohol consumption and ethanol elimination rate of AA rats were subsequently found to vary directly with the protein content (5, 10, 20 and 40% by energy) of the premixed diet given to them--they drank more than 9 times more ethanol on the 40% than on the 5% protein diet--but the alcohol intake of the alcohol-avoiding ANA rats was not affected by protein intake. The results clearly indicate that the three macronutrients are related to alcohol consumption in different ways, suggesting that protein intake affects the selection of alcohol of AA rats through a factor correlated to ethanol elimination (but not through a ceiling effect from the rate of elimination) and that the alcohol in turn reduces carbohydrate consumption.

The interaction between voluntary alcohol consumption and dietary choice.

The influence of forced and voluntary alcohol intake on the choice by AA (heavy drinking) and ANA (light drinking) rats between carbohydrate, protein and fat was studied. Alcohol consumption reduced the carbohydrate intake but did not influence the total energy consumption. In AA rats on free choice, voluntary alcohol intake correlated negatively with carbohydrate consumption but positively with protein and water intake. It was also found that the day to day variation in alcohol intake was of the same order as those in carbohydrate, protein and water intake and smaller than that in fat consumption. These findings in addition to earlier observations point to a regulation of alcohol intake in rats which in many respects resembles the regulation of dietary choice.

Is carbohydrate metabolism genetically related to alcohol drinking?

Ethanol increased significantly the rate of gluconeogenesis from alanine in rats from the AA strain (alcohol-preferring animals), while no change was observed in rats from the ANA strain (alcohol-avoiding animals). The glycogen stores after alcohol administration were depleted in the livers of the ANA rats but remained unaffected in the AA animals. The results indicate that alcohol disturbs the carbohydrate homeostatis in the ANA rats, but not in the AA animals.

Liquid/air partition coefficients of acetaldehyde: values and limitations in estimating blood concentrations from analysis of breath.

We report liquid/air partition coefficients for dilute solutions of acetaldehyde in water, saline (0.9% wt/vol NaCl), human plasma, and corn oil. Equilibrium was studied at 34 degrees C and 37 degrees C with various concentrations of acetaldehyde in the liquid phases. At 37 degrees C, the liquid/air partition ratios for water, saline, plasma, and oil were 143 +/- 7.1, 132 +/- 6.6, 183 +/- 3.6, and 64 +/- 9.9 (mean +/- SE), respectively. At 34 degrees C, all the values were higher and the temperature coefficients of solubility were 3.7%, 3.3%, and 4.0% per 1 degree C for water, saline, and plasma solutions of acetaldehyde, respectively. The partition coefficients were independent of the concentration of acetaldehyde in the liquids, and the solubility was higher in water than in oil. The results are discussed with emphasis on the usefulness of expired breath as a biological specimen for the quantitative determination of acetaldehyde.

Diabetes and the alcohol consumption of AA and ANA rats.

Some earlier results have suggested a link between diabetes and voluntary ethanol consumption. Measurement of blood glucose levels now showed, however, that the AA line of rats developed for high alcohol intake is not diabetogenic. The alcohol-avoiding ANA line was somewhat more susceptible to streptozotocin-induced diabetes. Streptozotocin greatly increased blood glucose levels, water intake, and food consumption, but had no effect on the voluntary alcohol drinking of either AA or ANA rats. Normalization of blood glucose levels with insulin from osmotic minipump implants also failed to change their ethanol drinking. Thus, although the evidence from some other lines of rats and mice suggests that alcohol drinking is related to some forms of diabetes, the present results indicate that it is independent of the diabetes induced by streptozotocin and also of water and food intake regulation in this situation.

A high hepatic concentration of free proline does not induce collagen synthesis in rat liver.

It has been reported that the concentration of free proline in the liver can be a limiting factor in the synthesis of hepatic collagen, and there has also been found to be a good correlation between the free proline and the amount of collagen in cirrhotic human livers. Since ethanol retards the breakdown of proline, it might be expected that ethanol-induced liver cirrhosis could be produced by the effect of ethanol on the hepatic proline level. In the present study the hepatic free proline level was increased more than three-fold by the administration of proline-rich diet to the rats used in the experiment. Administration of ethanol to the animals did not further increase the concentration of free proline of the liver. The high free proline level had no effect on the collagen formation, nor on the structure of the liver. It can therefore be assumed that the increased free proline levels observed in cirrhotic livers did not induce an increased collagen accumulation, and are a consequence of an altered proline metabolism.

Influence of extracellular proline on collagen synthesis in rat liver slices.

The synthesis of collagen was measured in incubated rat liver slices. The concentration of proline in the medium was varied, but the specific radioactivity of proline was kept constant. The synthesis of total protein was found to be about 500-fold that of collagen. The incorporation into hydroxyproline--that is, the synthesis of collagen--and the total incorporation of proline into protein were dependent on the concentration of proline in the medium, by Michaelis-Menten kinetics, with an apparent Km of 0.40 mM for the synthesis of collagen and 0.43 mM for the synthesis of total protein.

Influence of ethanol oxidation rate on the lactate/pyruvate ratio and phosphorylation state of the liver in fed rats.

The effect of the ethanol oxidation rate on the interaction between the phosphorylation state (the [ATP]/[ADP] X [HPO4]2- ratio) and the redox state (the free [NAD+]/[NADH] ratio) of the liver cytosol was studied in intact fed rats. The rate of ethanol oxidation was inhibited to different degrees with pyrazole. The ethanol oxidation rate had no influence on the liver lactate level but correlated significantly with the pyruvate level. Accordingly, a significant correlation was also found between the ethanol oxidation rate and the lactate/pyruvate ratio. The rate of ethanol oxidation correlated significantly with the liver 3-phosphoglycerate level. No change in the glyceraldehyde-3-phosphate level was found. No correlation was found between the ethanol oxidation rate and the glyceraldehyde-3-phosphate/3-phosphoglycerate redox couple. Ethanol administration slightly increased the liver ATP level, but the simultaneous administration of pyrazole eliminated this effect. Other adenine nucleotides and HPO4 2- were not changed. The changes in the rate of ethanol oxidation had no effect on the phosphorylation state in the fed liver. It is assumed that in the fed liver the phosphorylation state is so well stabilized that the redox level has no influence.