Richard A. Deitrich and Allan C. Collins
Department of Pharmacology, School of Medicine, University of Colorado, Denver, Colorado, 80220 and School of Pharmacy and Institute for Behavioral Genetics, University of Colorado, Boulder, Colorado, 80309.

INTRODUCTION

It has long been recognized among the laity that vast differences in response to ethanol exist among humans. For many years, different individual responses to ethanol were considered to be accidental, their occurrence in groups was not reported, and there was no suggestion that a rational explanation for individualized response might be possible. New light was thrown on the multiplicity and diversity of responses to drugs in general when clinical and research attention was directed toward the incidence of individual differences in response to drugs within families and other groups. Numerous investigations have clearly established a genetic influence on variability in drug response. The field of study which deals with this phenomenon is called pharmacogenetics. Evidence is now available to demonstrate that an individual's response to alcohol, like that to other drugs, is influenced by genetic factors. In this chapter, we shall present selected studies from the literature and our own findings with respect to the pharmacogenetics of alcohol related behaviors in humans and other animals. We shall also suggest how genetically influenced differences in these behaviors may be used as a tool to aid in the elucidation of the underlying mechanisms by which alcohol exerts its actions. In general, those studies which provide biochemical explanations will be emphasized.

DEFINITIONS

The following definitions are provided for those readers who may not be acquainted with the genetic and behavioral terms or the biochemical pathways under consideration. They are intended specifically for this presentation, and more technical definitions may be found in appropriate texts.

Alcohol Acceptance

For this method of assessing alcohol consumption, the usual procedure is to deprive animals of drinking fluid for 24 hours and then present them with a specified ethanol solution. The amount consumed in a certain period of time is compared to predeprivation water consumption to provide a measure of acceptance.

Alcohol Preference

There are numerous variations of this basic procedure for assessing alcohol consumption. Usually, a two-bottle choice situation is used--one bottle contains a specified ethanol solution, and the other contains either water or a particular control solution. Ths positions of the bottles are changed daily according to a prearranged schedule. The amount consumed from each bottle is recorded each day to obtain a measure of alcohol preference. Preference may be expressed in terms of volume of ethanol solution consumed, volume consumed relative to body weight, or ratio of volume consumed to total liquid consumption.

Allele

One of two or more alternate forms of a gene occupying a locus.

Biochemical Pathways

Ethanol Metabolism: CH3CH20H+NAD+ ADH - CH3CHO+NADH+H+

Acetaldehyde Metabolism: CH3CHO+NAD+ ALDH CH3C00H+NADH+H+

Amine Metabolism: R-CH2NH2+02+H20 MAO R-CHO+H202+NH3

Enzymes

Alcohol Dehydrogenase (ADH). A group of isozymes occurring primarily in the cytosol of the liver and responsible for most of the metabolism of ethanol. Pyrazole is an inhibitor of these enzymes.

Aldehyde Dehydrogenase (ALDH). A group of isozymes occurring primarily in the liver but also in other tissues. There are at least three isozymes in the cytosol of the liver. One of these enzymes can be induced in mice and in some lines of rats by phenobarbital and DDT. Another liver cytosolic enzyme can be induced in rats by tetrachlorodibenzo-p-dioxin

(TCDD). Liver mitochondria contain at least two isozymes. There is a matrix enzyme which has a high affinity for acetaldehyde (low Km) and is primarily responsible for acetaldehyde metabolism. This enzyme is inhibited by pargyline and by some other monoamine oxidase inhibitors. Liver mitochondria also contain an enzyme between the inner and outer membranes that has a high Km for acetaldehyde. In addition, liver microsomes contain an aldehyde dehydrogenase. All of these enzymes are inhibited by disulfiram (Antabuse).

Monoamine Oxidase (MAO). An enzyme located on the outer mitochondrial membrane that is responsible for oxidizing many biogenic amines (such as epinephrine, norepinephrine, dopamine and serotonin). It is inhibited by pargyline, nialamide and several other compounds.

Fl Generation

The offspring of a cross between two inbred parental strains. These animals are heterozygous at each locus at which the parental strains differed. They are all genetically uniform.

F2 Generation

Progeny produced by the mating of two individuals of an Fl generation. Alleles of genes that are heterozygous in the Fl generation can segregate in the F2, and each individual F2 animal is a genetically unique event. However, genetic variability is a function of the particular parental strains that produced the Fl generation.

Genotype

The genetic constitution of an organism, as distinguished from its observable characteristics (phenotype).

Heritability

A measure of the proportion of phenotypic variance that is due to genetic variability among individuals in a population. The measure is descriptive of a particular population in a particular environmental milieu.

Heterogeneous Stock

Genetically heterogeneous populations can be established in various ways. A commonly used method is to intermate a number of inbred strains, with subsequent maintenance of the stock through random mating or some other outbreeding scheme. A heterogenous stock is useful for normative studies; also, such stock is often used as the foundation population for a selective breeding program.

Heterozygous

Having two different alleles at a locus.

Homozygous

Having identical alleles at a locus.

Inbred Strain

Offspring of individuals that are more closely related than would occur if mating were at random in a population; usually, inbreeding denotes the mating of siblings, and a strain that has been propagated by brother-sister mating over 20 consecutive generations is commonly designated as inbred. All individuals are descendants of one female and her brother. They are homozygous at all loci. The result of such inbreeding is a reduction of genetic variability to the point where, for practical purposes, each individual can be regarded as genetically identical to each other individual. It is important to note that the process of inbreeding is nondirectional; that is, the particular genetic configuration that is fixed in a strain is a matter of chance, except for characters related to reproductive fitness. See McClearn (1) for a discussion of inbreeding and other mating procedures used in alcohol research.

Locus

Position on a chromosome occupied by allelic forms of a particular gene.

Phenotype

The observable properties of an organism. These properties are produced by the genotype in conjunction with the environment.

Selective Breeding

A mating procedure which may produce populations that differ in a particular phenotypic attribute. In other words, it is a directional process. All animals in a genetically heterogeneous population are measured for the trait in question; females displaying high levels of the trait are mated with males showing high levels, and females displaying low levels are mated with males showing low levels. Thus, two lines are begun, and they are maintained by mating the highest males and females of the high line and the lowest of the low line in subsequent generations. To the extent that genetic factors influence the variability of the trait in the foundation population, the lines become increasingly divergent with respect to the trait in succeeding generations. Continued selection may lead to increasing homozygosity within lines for the locus influencing the phenotype; however, if heterozygosity at a locus which influences the phenotype contributes to a high or low level of the trait, homogeneity will not be achieved within lines.

ANIMAL MODELS OF ALCOHOL-RELATED BEHAVIORS

Genetic influences upon several alcohol-related behaviors are especially evident. Specifically, preference for alcohol, initial (first dose) sensitivity to alcohol, and the development of physical dependence upon alcohol all appear to be influenced by genetic factors. Direct evidence to support the contention that heredity influences these behaviors has been obtained from both animal and human studies. The use of animal models has been criticized on the grounds that they do not completely mimic the human condition. Nevertheless, there is no doubt that animal models of various aspects of human alcoholism do provide much important information.

Preference for Alcohol

Studies concerned with alcohol preference in humans are relatively few [see Varela (2) for a discussion of human appetites for alcohol). However, preference for alcohol is the most intensively studied of the genetically influenced alcohol-related behaviors in laboratory rodents. Two approaches, typifying much of the work in pharmacogenetics, have been used to demonstrate a genetic influence on alcohol preference. In the first approach, the investigator takes advantage of existing laboratory strains (or of different species that exist in nature). The second approach is to apply the selective breeding procedure to a particular behavioral or biochemical trait of interest. The latter method (see definitions) is much more tedious and time consuming. When properly carried out, however, it provides an extremely useful tool for the study of behavioral or biochemical traits.

McClearn and Rodgers (3) utilized the first of these approaches to assess the alcohol preference of several inbred mouse strains, whereas Eriksson (4,5) applied the second approach and bred rats selectively for alcohol preference. These investigators, the former in the United States and the latter in Finland, assessed alcohol preference in a similar fashion (see definitions). McClearn and Rodgers observed a substantial difference in preference for 10% alcohol solution among various inbred mouse strains. The C57BL/Crgl strain showed significant preference for alcohol, while A/Cal, BALB/c and DBA/2 animals preferred water and avoided alcohol. Intermediate preference was shown by the C3H/HeCrgl strain.

Major advantages of using inbred strains of mice or rats are their general availability and their constancy over time and in different laboratories. For example, many studies have replicated some of the strain differences in alcohol preference originally noted by McClearn and Rodgers. These include investigations by Rodgers et al. (6), Fuller (7), McClearn and Rodgers (8), Eriksson and Pikkarainen (9) and Pickett and Collins (10). The disadvantage of the use of inbred strains is that the process of inbreeding is non-directional (see definitions). The strains will differ with respect to many characters, some of which may be related to the preference difference and some of which may not. As a' result, relationships between preference and any other behavioral, biochemical or pharmacological trait are likely to be fortuitous. Only by other means of genetic analysis can such correlations be tested.

Further evidence that alcohol preference is genetically influenced comes from studies in which inbred strains were crossed to produce filial generations. McClearn and Rodgers (8) obtained an Fl generation from a C57BL x DBA cross and found that these animals exhibited intermediate preference. Thomas (11) observed that the preference of animals in the Fl generation produced by a C57BL x DBA cross was more similar to that of the parental DBA strain. A subsequent F2 generation showed a wide range of alcohol preference, with most of the animals exhibiting high preference. However, when Pickett and Collins (10) developed an F2 generation from a C57BL x DEA cross, they obtained a majority of low-preferring animals. In addition to providing further evidence of a genetic influence on alcohol preference, these studies also suggest that a number of genes, perhaps working in an additive fashion, are involved in this behavior. Whitney et al. (12) reanalyzed much of the early preference data and concluded that early estimates of heritability were erroneously high. These authors suggest that a value for heritability that ranges between 0.10 and 0.15 may be a more realistic estimate.

As previously noted, Eriksson (4) developed two selectively bred lines of rats which differ in alcohol consumption.

The high-preferring line is referred to as AA (alcohol addicted), and the low-preferring line is designated ANA (alcohol non-addicted). It should be noted that the original designation "alcohol addicted" is inappropriate according to more recent usage of the term "addicted", since it implies that these animals are inherently addicted, which is clearly not the case, or that they are easily addicted upon exposure to alcohol, which also has not been demonstrated. Still, because the designations AA and ANA are used by Finnish researchers, they will be used here to avoid unnecessary confusion. These animals were selected for alcohol preference through a breeding program of more than 20 generations. Eriksson (13) conducted a genetic analysis of the AA and ANA lines and concluded that alcohol consumption is a polygenic, additive trait and that its heritability is low. The fact that the selection process required many generations to achieve separation of the two lines also indicates that the trait is complex and polygenic.

Eriksson is not the only investigator to breed rats successfully for differences in alcohol consumption. For example, Mardones (14) bred for alcohol preference by starting with a single mating pair of high-preferring and a single pair of low-preferring rats. Succeeding generations were obtained by brother-sister matings to produce inbred strains that exhibit a clear-cut difference in alcohol preference. Eriksson (5) points out that the approach used by Mardones has a major deficiency in that there is a strong possibility that any other strain differences were determined largely by the original mating pairs and are merely coincidental (or perhaps dependent on only one usually strong metabolic factor). This is essentially the same caution expressed by McClearn (1) with respect to the interpretation of data derived from inbred strains.

An interesting approach to the separation of genetic and environmental influences on alcohol preference was used by Randall and Lester (15). These investigators placed weanling mice of the C57BL (alcohol-preferring) strain with DBA (alcohol-avoiding) adults and vice versa. A choice of alcohol or water was available for seven weeks, at which time all mice were individually tested for alcohol preference. The C57BL offspring housed in a "non-drinking" environment drank about half as much alcohol as their siblings who were left with alcohol-preferring adults. Similarly, the normally alcoholavoiding DBA offspring that were housed with "drinkers" drank almost twice as much alcohol as their siblings who were left with DBA adults. In spite of these marked changes in drinking behavior, the strain difference persisted, i.e., the "adopted" C57BL mice still drank more alcohol than the "adopted" DBA mice. Somewhat different results were obtained when fertilized ova were transferred between alcohol-preferring and alcohol-avoiding mothers (16). In that study, C57BL offspring nurtured by DBA surrogate mothers actually increased their alcohol intake in spite of being born and raised by an alcohol-avoiding adult. DBA offspring born of C57BL surrogate mothers also drank more alcohol than DBA mice raised by DBA mothers.

Most recently, McClearn and Anderson (17) initiated a selective breeding program in mice for an alcohol consumption behavior which they called acceptance (see definitions). The lines have separated at the 10th generation. Calculated heritability is 0.15 in the high line and 0.28 in the low line. The gradual separation of the lines suggests that alcohol acceptance, like preference, is influenced by more than one gene. In addition, these investigators conducted tests to determine whether alcohol acceptance and alcohol preference are identical behaviors. Correlations were calculated among the following measures: (1) alcohol preference, as measured by the standard two-bottle choice method; (2) alcohol acceptance with thirst motivation; and (3) alcohol acceptance without thirst motivation. Although correlations between alcohol preference and the two acceptance measures were statistically significant, they were low. This observation led McClearn and Anderson to suggest that different methods of assessing alcohol consumption are not equivalent and that investigators studying mechanisms underlying alcohol consumption must be cautious in generalizing their results.

Sex Differences in Alcohol Preference

An influence of heredity on alcohol preference has also been seen when drinking behaviors of the two sexes are compared. Eriksson and Pikkarainen (9,18) noted a significantly greater preference for alcohol in female than in male C57BL mice. No sex difference was observed in the alcohol-avoiding DBA strain. Eriksson and Malmstrom (19) reported that female albino rats consume more alcohol than do males, and Eriksson (20) found that females of both the AA (high-preferring) and ANA (low-preferring) rat lines consume more alcohol than their male counterparts. Similarly, Brewster (21) observed a greater preference for alcohol in females than in males of the Maudsley Reactive and Maudsley Non-Reactive rat strains. Russell and Stern (22) noted a greater preference in females of the Wistar, Hooded, Tryon Maze Bright and Tryon Maze Dull rat strains. Taken together, these data argue strongly that, at least in laboratory mice and rats, female animals manifest a significantly greater preference for alcohol than do males.

Relationships Between Preference and Other Measures

While genetic influences on alcohol preference are of interest in themselves, these demonstrated effects are of value in testing hypotheses concerning the underlying biochemical and physiological mechanisms which might account for differing preferences for alcohol. The genetic method has already been used to generate and test a few such hypotheses. Specifically, the roles of differential ethanol or acetaldehyde metabolism of neurotransmitter function have received the greatest attention.

Alcohol Dehydrogenase. Several studies have suggested that mouse strains with higher alcohol preference have greater hepatic alcohol dehydrogenase (ADH) activity as determined in vitro. One of the first suggestions that hepatic ADH may influence alcohol preference came from the studies of Rodgers et al. (6), who found a positive relationship between alcohol preference and in vitro hepatic ADH activity. Other studies (8,9,18) have confirmed this finding. However, the precise manner in which an elevated ADH activity influences alcohol preference remains a question. Schlesinger (23) observed that the alcohol-preferring C57BL strain, which has high hepatic ADH activity, attains blood alcohol concentrations differing only slightly from those of the alcohol-avoiding DBA strain following administration of the same dose of alcohol.

Koivula et al. (24), using Eriksson's AA and ANA lines, obtained evidence suggesting that alcohol preference is not directly related to hepatic ADH activity in rats. These investigators detected greater ADH activity in both sexes of the low-preferring ANA line than in the AA animals. Ethanol disappearance from the blood of AA females is faster than in ANA females, but there is no difference in disappearance rate between AA and ANA males (25). Thus, the data obtained from rats and mice are not in agreement. The mouse data suggest that alcohol preference increases with an increase in hepatic ADH activity, whereas the rat data suggest that the opposite situation may pertain. This discrepancy could be accounted for by assuming that the amount of ADH is not the rate-limiting factor in ethanol metabolism in vivo. Apparent relationships between ADH activity and ethanol preference or metabolism may be entirely coincidental.

One way to further test an hypothesized relationship between alcohol preference and hepatic ADH activity is to calculate correlations between the two variables in hybrid generations that are obtained by crossing two inbred strains that are known to differ in preference. The F2 generation provides a suitable group for testing correlations in that segregation of genes, and therefore of genetically influenced traits, should have occurred. Only if genes which control the two traits are on the same chromosome may a spuriously high correlation be found between ADH activity and alcohol preference.

Another way to test an hypothesized relationship between behavior and biochemistry is to calculate correlations in a heterogeneous stock of animals (see definitions). For purposes of such genetic analyses, these animals can be used in the same way as an F2 generation, and they display even greater genetic variability. McClearn (26) tested the correlation between alcohol preference and hepatic ADH activity in an F2 generation obtained from a C57BL x DBA cross and in HS/Ibg mice (a heterogeneous stock derived from the crossing of eight inbred strains). No significant correlation was seen between preference and ADH activity in the F2 animals, while a low but significant correlation (.29) was found when the HS data were analyzed. These results may have been influenced by the fact that the F2 sample was very small (20 animals) in comparison with the larger number of animals (60) used in the HS study. Had a higher correlation been found between alcohol consumption and enzyme activity in these genetically heterogeneous groups, stronger evidence for a role of ADH activity in preference would have been obtained. Nevertheless, the correlation of .29 appears to be meaningful, and it indicates that the magnitude of the genetic influence on ADH activity is much smaller than might have been concluded on the basis of the earlier strain comparisons. Again, these results illustrate the potential hazards of inferring relationships between characteristics from their association in inbred strains. Even though relationships observed when inbred strains are compared provide only suggestive evidence for a causal connection between a given biochemical mechanism and some behavior, this approach has been utilized far more often than the more accurate method of examining correlations in genetically heterogeneous populations.

Aldehyde Dehydrogenase. The data are less ambiguous with respect to the second step in alcohol metabolism. Ethanol is converted by alcohol dehydrogenase to acetaldehyde, which is subsequently converted by hepatic aldehyde dehydrogenase (ALDH) to acetate (see definitions of biochemical pathways). One of the first observations which suggested that acetaldehyde levels may influence alcohol preference came from the studies of Schlesinger et al. (27), who noted that C57BL mice accumulate considerably less acetaldehyde in the blood following ethanol administration than do mice of the DBA strain. Furthermore, these investigators found that disulfiram (Antabuse), an inhibitor of ALDH, served to elevate blood acetaldehyde and caused a significant decrease in alcohol preference in the C57BL strain. Additional evidence in support of the acetaldehyde hypothesis was reported by Sheppard et al. (28). When hepatic ALDH activity and alcohol preference were measured in the C57BL and DBA strains and in their Fl offspring, the C57BL's exhibited significantly greater ALDH activity than the DBA's and the Fl generation was intermediate to the parents in both preference and enzyme activity. Eriksson and Pikkarainen (9) examined alcohol preference and hepatic aldehyde oxidizing capacity in C57BL's and CBA's (similar to DBA's in that they are alcohol-avoiding) and in F2 offspring of a C56BL x CBA cross. Their data do not support the findings of Sheppard at al. (28), since no differences in liver cytosolic ALDH activity were detected among the groups. These results should be viewed cautiously, however, because recent data (29,30,31,32) suggest that cytosolic ALDH is of less importance with respect to total acetaldehyde metabolism than are the mitochondrial enzymes. Mitochondrial ALDH activity was not determined in the study by Eriksson and Pikkarainen (9).

It has previously been noted that females of the C57BL strain show a significantly greater preference for alcohol than do males (9,18). When Redmond and Cohen (33) compared acetaldehyde exhalation following ethanol injection in C57BL males and females, they found that males exhaled significantly more acetaldehyde. This finding suggests that the sex difference in alcohol preference in the C57BL strain may be related to differing acetaldehyde levels in males and females. A study by Koe and Tennen (34) demonstrated that butyraldoxime inhibited liver ALDH in vivo, increased blood acetaldehyde levels after ethanol treatment, and reduced alcohol consumption by C57BL mice.

Studies similar to those of Sheppard et al. (28) with mice have been carried out with the AA and ANA rats (24). ALDH activity was assayed with 0.5 mM acetaldehyde in various lever fractions of these animals. No difference was found in the enzyme activity in the mitochondrial fraction, although activity in the soluble fraction was greater in the AA line. This concentration of acetaldehyde approximates the concentration in the liver during ethanol oxidation (35). Eriksson (25) detected lower acetaldehyde concentrations in ethanolperfused livers from AA rats than in those from ANA rats. A greater mitochondrial and microsomal ALDH activity was found by Koivula et al. (24) in the AA than in the ANA line when a high-concentration (4.5 mM) propionaldehyde was used as substrate. The use of this concentration of aldehyde measures the total amount of ALDH that is present. These data suggest that the AA and ANA lines have similar activities of the low-Km mitochondrial enzyme (measured with 0.5 mM acetaldehyde), while the activity of the high-Km mitochondrial enzyme (measured with 4.5 mM propionaldehyde) is greater in the AA line. When acetaldehyde (0.4 mM) metabolism was measured in a homogenate system, the rate of acetaldehyde elimination was slightly greater in the AA animals. These findings may be compared to reports from a number of laboratories (29,30,31, 32) that the low-Km enzyme in rat liver mitochondria is primarily responsible for acetaldehyde metabolism. Collectively, these data indicate that higher Km enzymes may play a role in the control of blood acetaldehyde level. A factor not considered here, however, is the possibility that ALDH activity in other tissues may be of importance and may be different in the AA and ANA lines. In any event, it appears that high blood acetaldehyde is a deterrent to alcohol consumption in rats and mice. This was confirmed serendipitously by the finding that a change in diet caused a marked reduction in ethanol preference and a corresponding decrease in rat liver acetaldehyde metabolism (35). Added confirmation that acetaldehyde is a deterrent to alcohol consumption comes from the observation that a diet low in protein and high in carbohydrates results in elevated blood acetaldehyde, a decrease in rate of ethanol metabolism, and reduced alcohol consumption (36).

Another means of altering acetaldehyde metabolism has been to utilize the induction of a liver cytosolic ALDH (37). induction is brought about by phenobarbital, it is known to be controlled by a single gene, it is observed only in the liver cytosol, and it is dependent upon genotype in certain rat strains    (38).        A    rat liver cytosolic enzyme, called the (1) enzyme, is induced from ten to thirty-fold by phenobarbital treatment in homozygous reactor (RR) animals; animals that do not show induction following phenobarbital treatment are designated rr. The (I) enzyme is present in non-treated animals (39), but the genetically determined induction effect is present only in some rat strains such as Long-Evans, SpragueDawley, Wistar and ACIF (38). The induced enzyme has a high Km (mmolar range) for acetaldehyde, and, when induced, it does alter blood acetaldehyde in ethanol-treated animals but not to a degree consistent with the large increase in enzyme activity (40,41). Similarly, the metabolism of dopamine or norepinephrine in rat liver slices is not markedly altered by induction of this enzyme (42).

Induction of another ALDH isozyme (t enzyme) in the cytosol of rat liver by tetrachlorodibenzo-p-dioxin (TCDD) has recently been reported (43). In this case, a genetic effect has not been found. The induction is approximately a hundredfold, but the Km for acetaldehyde is even higher than that of the phenobarbital-induced enzyme (44). Petersen et al. (40) found that even such a large induction had no effect on blood acetaldehyde after ethanol treatment or on dopamine metabolism in rat liver slices. Basic principles of enzymology adequately explain such results, since it is known that liver acetaldehyde levels during ethanol metabolism rarely exceed 0.5 mM (45). From the Michaelis-Menten equation (v/Vmax = S/Km+S) and the Km values for acetaldehyde of the • enzyme (2.7 mM) and the t enzyme (22 mM), one can easily calculate that the v/Vmax ratio for the 0 enzyme will be 0.16 (or 16% of Vmax), while that for the T enzyme will be 0.019 (or 1.9% of Vmax). In the case of the 0 enzyme, operating at 16% of its Vmax (which is achieved at 0.5 mM acetaldehyde), a thirty-fold induction should increase the cytosolic contribution to acetaldehyde metabolism by a factor of 4.8. If the cytosol contributes 15% of the total acetaldehyde oxidizing capacity of rat liver, the total increase in this capacity will be 0.72 times. In other words, a gram of liver in which the 0 enzyme has been induced will be operating at 172% of the capacity of a gram of control liver. When the t enzyme is induced a hundred-fold, the contribution of the cytosol to acetaldehyde (0.5 mM) oxidation will increase by a factor of 1.9. Thus, liver in which the t enzyme has been induced will be operating at 129% compared to control liver. These caclulations agree with our finding that induction of the 0 enzyme decreased blood acetaldehyde concentration, while induction of the t enzyme was without effect. Eriksson et al. (41) obtained similar results following induction of the 0 enzyme, although these authors found no difference in acetaldehyde oxidation rate per gram liver in vivo when a correction for blood flow through the liver and ethanol oxidation rate was taken into account.

Marcelos et al. (46) tested AA and ANA rats for alcohol preference thirty days after treatment with phenobarbital. Although both lines manifest a slit induction of cytosolic ALDH by phenobarbital, the treatment had no effect on preference in either line. Also, no effect of treatment on preference was found in two other rat strains, one which showed an induction of cytosolic ALDH and one which did not. It should be emphasized that this study does not prove that an increase in cytosolic ALDH activity has no influence on alcohol preference, because preference was not measured during the time when enzyme activity was increased. The findings of Marselos et al. demonstrate only that alcohol preference is not influenced by inducibility of ALDH activity. These results are not surprising, since there is no reason to believe that potential for induction, a single-gene effect, should be related to a polygenically determined behavior such as alcohol preference. Nevertheless, studies involving enzyme induction may be useful in examining the effects of acetaldehyde on preference.

Marselos and Pietikainen (47) conducted a direct test of the effect of ALDH induction on preference by using DDT to induce rat liver cytosolic ALDH. No difference in alcohol preference between rats that showed induction (RR) and those that did not (rr) was observed under these conditions. Results from our laboratory (44) indicate that there are a number of enzymes induced by phenobarbital, as well as the specific enzyme induced by TCDD. There is no information as to which of these multiple isozymes is induced by DDT; if it is an enzyme with a very high Km, no effect on blood acetaldehyde would be expected. Blood acetaldehyde concentrations were not measured by Marselos and Pietikainen (47).

It must be pointed out that blood acetaldehyde is customarily measured after administration of rather large doses of ethanol, whereas the amount consumed by animals in an alcohol preference study is relatively small. Nevertheless, an overall view of the accumulated data does support the hypothesis that alcohol preference is influenced by the concentration of acetaldehyde in the blood. Although this variable appears to be controlled principally by the low-Km mitochondrial ALDH, activity of high Km isozymes, such as the 0 enzyme, may be critical in the fine control of blood acetaldehyde concentration.

Biogenic Amines. Another major hypothesis concerning underlying mechanisms which might account for differences in alcohol preference arose from the observation that drugs which influence brain serotonin level also appear to have an effect upon preference. However, no clear picture has emerged from the results of studies in this area. For example, Hill (48) found a reduction in alcohol consumption following intraventricular injection of serotonin, and similar results were obtained when 5-hydroxytryptophane (the metabolic precursor of serotonin) was administered (49,50). On the other hand, drugs such as parachlorophenylalanine and parachloroamphetamine, which decrease brain serotonin level, have also been observed to reduce alcohol preference (51,52,53), although some investigators (49) have found an increase in preference following treatment. Sanders et al. (53) observed that pargyline, an inhibitor of monoamine oxidase, also serves to decrease alcohol preference. These inconsistencies may be explained by the suggestion of Nachman et al. (54) that those drugs which decrease preference may be acting as an aversive stimulus. Dembiec et al. (55) found that pargyline inhibits hepatic ALDH and causes an elevation in blood acetaldehyde level. Peterson et al. (40) observed that pargyline inhibits only the low-Km ALDH found in liver mitochondria. This suggests that pargyline may be acting in a manner similar to disulfiram. Results of studies recently completed in our laboratories support this suggestion. In these studies, utilizing C57BL/Ibg mice, a decrease in alcohol preference was detected following administration of either pargyline or Lilly 51641 (another monoamine oxidase inhibitor), while nialamide, which inhibits monoamine oxidase but not ALDH, did not influence preference. Dembiec et al. (55) noted that pargyline and Lilly 51641 elevate blood acetaldehyde and that nialamide has a minimal effect. Thus, inconsistencies in these data could be explained by the possibility that acetaldehyde may be acting as an aversive stimulus. Ambiguities in the results of studies intended to examine effects of alterations in brain amine content on alcohol preference may be attributable to such confounding effects.

Because genetic analyses do not require the use of drugs, which generally have a multitude of actions, such problems are less likely to arise. Unfortunately, genetic methods have seldom been used in studies of the relationship between neurotransmitters and alcohol preference. Ahtee and Eriksson (56) measured whole brain concentrations of serotonin and 5-hydroxyindole acetic acid (5HIAA) in AA and ANA rats. The concentrations of both of these compounds were 15-20% higher in the AA line, but the difference was not statistically significant. When 5HIAA removal from the brain was inhibited by probenecid, the 5HIAA content of AA brain was significantly greater than that of ANA brain. This finding indicates that turnover, rather than static levels of amines, may play an important role in alcohol preference. Furthermore, ethanol consumed in the drinking water elevated serotonin in the AA but not in the ANA line. This implies that there is an interaction between alcohol treatment and one or more of the processes which influence serotonin level (i.e., rate of synthesis, release, reuptake or metabolism). Recently, the same investigators (57) measured brain catecholamine content in the AA and ANA animals. A greater concentration of dopamine, but not norepinephrine, was found in the AA line.

Pickett and Collins (10) examined whole brain concentrations of serotonin in the alcohol-preferring C57BL and the alcohol-avoiding DBA mouse strains and in their Fl and F2 offspring. There was no difference in serotonin content between the parental strains, and there was no apparent relationship bewteen serotonin level and alcohol preference in the F2 generation. These data are in agreement with the findings of Ahtee and Eriksson (56) in suggesting that alcohol preference is not influenced by static brain serotonin levels. If this biochemical variable had an influence on preference, a clear-cut relationship should have been observed in the F2 generation in the study by Pickett and Collins (10). Correlational studies using other genetically heterogeneous populations would provide a more rigorous test of the hypothesis that neurotransmitter function is involved in alcohol preference.

Initial Sensitivity to Alcohol

In comparison with alcohol preference research, far fewer studies indicate that initial central nervous system sensitivity to alcohol is influenced by genetic factors. Most of the studies have utilized inbred or selectively bred mice. As in preference research, the conclusion that there is a genetic influence on sensitivity is based upon results showing that different inbred strains respond differently to alcohol and that differential sensitivity between lines may be achieved by a selective breeding program.
A study from McClearn's research group (58) provided information as to the cause of differing sensitivities to alcohol. In this study, mice of the C57BL and BALB/cCrgl strains were injected with a hypnotic dose of ethanol and duration of loss of the righting reflex (alcohol-induced "sleep time") was measured. Animals of the BALB strain slept over 3.5 times as long as did mice of the C57BL strain. A portion of the difference in sleep time could be explained by the observation that the C57BL strain regained the righting response ("time of awakening") at a blood alcohol concentration considerably, higher than that seen in the BALB strain at time of awakening. These data, and results of an earlier preliminary study (59), indicate the genotype influences neuronal sensitivity to alcohol.

Evidence which suggests that enzymes involved in alcohol metabolism influence duration of alcohol-induced sleep time has come from several studies. Belknap et al. (60) examined the correlation between duration of sleep time and hepatic ADH and ALDH activities in the heterogeneous stock of mice (HS/Ibg) developed by McClearn (61). A correlation of -0.57 was found between ADH activity/gram body weight and sleep time, accounting for approximately one-third of the variance in sleep-time scores. A smaller correlation (-0.39) between ALDH activity/ gram body weight and sleep time was obtained. Genetic influences on sleep time can, therefore, be explained at least in part, by differential activities of those enzymes which are of primary importance in alcohol metabolism. Damjanovich and Maclnnes (62) studied the effect of genotype on fall time (the time from ethanol injection to loss of ability to cling to a wire mesh) and on sleep time using three inbred mouse strains (C57BL/6J, DBA/2J and BALB/cJ). The results showed differences in fall time, which could be explained by differential rates of alcohol absorption; there were also differences in sleep time, which the authors suggested might be explained, at least in part, by differing rates of alcohol metabolism.

Several other studies, also utilizing inbred mouse strains, have added to the evidence that genotype interacts with alcohol to influence behavior. For example, Randall et al. (63) examined the effect of various alcohol doses (0.75, 1.50 and 2.25 grams/kilogram body weight) on locomotor activity in C57BL and BALB/c.3 mice. The C57BL strain showed a dose-dependent decrease in locomotor activity, whereas the BALB/cJ strain showed an increase in activity over this dose range. These data may be explained by the hypothesis that C57BL mice are more sensitive to the depressant action of alcohol, while BALB mice are more sensitive to the activating effect of this drug. Comparison of these findings with those of the Kakihana et al. study (58) suggest that the factors which result in a decrease in locomotor activity may not be the same as those involved in hypnosis, since the previous investigation found less neuronal sensitivity to the hypnotic actions of alcohol in C57BL than in BALB mice.

Alcohol may also influence learning and memory, and the effects seem to be related to genotype. This suggestion is based upon the results of a study by MacInnes and Uphouse (64), who measured the effects of several doses of alcohol (ranging from 0.5 to 3.0 grams/kilogram) on the acquisition and retention of a passive-avoidance task in C57BL and DBA mice and in the Fl generation produced by crossing these two strains. In the acquisition of the task, C57BL mice were relatively unaffected by doses of alcohol that seriously interferred with the performance of both DBA and Fl animals. On the other hand, when retention was measured the following day, the F1's showed better retention after higher doses during acquisition than either parental strain.

Perhaps the best evidence indicating the initial sensitivity to alcohol is under genetic control comes from the studies of McClearn and Kakihana (65). These investigators selectively bred mice for long or short sleep time following a hypnotic dose of ethanol (4.1 gram/kilogram). The foundation population was the heterogeneous stock of mice (HS/Ibg) developed by McClearn (61). After eighteen generations of selective breeding (see definitions), there was virtually no overlap in sleep time between the long-sleep (LS) and short-sleep (SS) lines. The mean sleep time of the SS line was eleven minutes, and that of the LS line was approximately 140 minutes. Heston et al. (66) measured the activity of hepatic ADH and ALDH, the rate of in vivo ethanol elimination, and the ED50 for loss of the righting response in the two lines. No differences in ADH or ALDH activity were detected, and in vivo rates of ethanol metabolism were similar in the two lines. ED50 for loss of the righting response, however, was nearly two times as great in the SS as in the LS line. These data suggest that the two lines differ principally in neuronal sensitivity to the hypnotic effects of ethanol. Erwin et al. (67) also noted that the lines differ in sensitivity to hypnotic doses of several alcohols, while they detected no differences in sleep time following administration of pentobarbital, diethylether, chloral hydrade, paraldehyde or trichloroethanol. Thus, these animals may be very valuable in ascertaining the specific mechanisms by which alcohol elicits its hypnotic effects. For example, they have already proved to be of value in testing one of the more controversial hypotheses concerning alcohol's mechanism of action. Church et al. (68) assessed the sensitivity of the LS and SS mice to intraventricular injection of salsolinol, the tetrahydroisoquinoline alkaloid which Cohen and Collins (69) have suggested may form in vivo as a consequence of the condensation of acetaldehyde with dopamine. The LS mice were found to be nearly twice as sensitive as the SS to the effects of salsolinol as measured by depression of locomotor activity. Because the LS and SS lines differ in sensitivity to alcohols but not to other hypnotics (67), these data support the hypothesis that salsolinol, or related compounds, may play a role in the depressant action of alcohol.

The hypothesis that alcohol's depressant effect is influenced by interactions with neurotransmitters has also been tested using LS and SS mice. Chan (70) detected significSntly higher GABA levels in cerebral cortex and pons-medulla of LS than SS mice, whereas no difference in brain glutamic acid content was observed. Collins et al. (71) found significantly greater concentrations of both dopamine and norepinephrine in the brain of SS mice. Administration of ethanol (4.1 gram/ kilogram) decreased dopamine turnover in both lines, but to a significantly greater extent in the LS. It should be emphasized, however, that finding differences between selected lines does not provide unambiguous evidence to support a hypothesis. The hypothesis should be further tested by examining the correlation between the behavioral trait and the mechanism in question in a second or perhaps a third generation derived by crossing the two lines.

One other point should be made concerning the LS and SS mice. These animals were selectively bred for differential sensitivity to the hypnotic effects of alcohol, and there is no reason to expect that they would differ in the same fashion with respect to other actions of this drug. In fact, Sanders (72) recently reported that the LS mice are less sensitive than the SS to the activating effects of low doses of ethanol. Interestingly, Sanders also found that the LS animals are less sensitive to the activating effects of pentobarbital.

An overall view of the above studies indicates that the results have tended to vary as a function of the behavior that is measured and the alcohol dose that is administered. These observations suggest that alcohol may exert its influences via a number of different mechanisms. Thus, an understanding of how alcohol affects a specific behavior may be best obtained by utilizing animals which differ in sensitivity to alcohol as measured by that particular behavior. Furthermore, since many differences appear to be dose-dependent, it is probably necessary to use various alcohol doses in testing any hypothesis.

The observations that inbred strains of mice differ in sensitivity to various effects of alcohol and that selective breeding for alcohol's hypnotic actions has been successful are strong arguments in favor of a genetic influence. However, only a relatively few studies have utilized a genetic approach in attempts to determine the mechanisms which underlie differences in sensitivity. Hopefully, further application of genetic methods will provide critical evidence as to biochemical and physiological mechanisms which might account for differing initial sensitivities to alcohol.

Physical Dependence Upon Alcohol

Only two studies have attempted to determine whether the development of physical dependence upon alcohol is influenced by genetic factors. In the first of these investigations, Goldstein (73) carried out a modified selective breeding program. Male and female Swiss-Webster mice were subjected to two cycles of intoxication and withdrawal. Intoxication was achieved by Goldstein's established procedure of injecting animals with pyrazole to inhibit alcohol metabolism and administering alcohol by inhalation. When withdrawal was evaluated by scoring the convulsions elicited by handling, a wide range of withdrawal scores was observed. Two pairs of high-scoring and two pairs of low-scoring mice were bred, and their offspring were tested for severity of withdrawal according to the same procedure. A significant difference in mean withdrawal score between offspring of high-scoring and low-scoring parents was evident even in this first generation. A second generation was then selectively bred, and the between-line difference was greater than that observed in the first generation. Additional evidence of a genetic effect on withdrawal severity was found in this study in that male mice in all three generations exhibited a more marked withdrawal syndrome than did females. Goldstein suggests that this sex difference may be explained by the fact that the males attained higher blood alcohol concentrations. This observation implies that the sex difference in withdrawal severity may be dependent upon blood alcohol level rather than differential central nervous system sensitivity to alcohol. If the development of physical dependence upon alcohol is facilitated by maintenance of higher blood alcohol concentrations for a longer period of time, males, which metabolize alcohol more slowly than females, should become dependent more rapidly.

One additional observation should be made concerning this study. The rapid separation between the lines obtained by the selective breeding procedure suggests that withdrawal severity may be influenced by only a few genes or, if many genes are involved, that only a few are of major importance. In any event, these results clearly indicate that withdrawal severity is influenced by genetic factors.

The only other published study of genetic influences on withdrawal was carried out by Goldstein and Kakihana (74). When these investigators used Goldstein's standard procedure to assess severity of withdrawal in LS and SS mice, the SS animals showed a more intense withdrawal reaction. The investigators then attempted to ascertain whether this difference in withdrawal severity was related to differential sleep time by comparing sleep-time and withdrawal scores in mice of the HS/Ibg stock. Because no correlation between these measures was found, it was concluded that no relationship exists between sleep time and withdrawal severity. However, it should be pointed out that the LS and SS lines differ only in central nervous system sensitivity to alcohol and have similar rates of alcohol metabolism (66). The HS mice, on the other hand, undoubtedly vary in both neuronal sensitivity and metabolism rate. It is possible that the SS mice, which have less initial sensitivity (greater tolerance) to alcohol, show a more marked withdrawal syndrome than the LS mice because those neurochemical factors which contribute to greater tolerance also contribute to the development of physical dependence. A study utilizing HS mice and designed to measure the correlation between central nervous system sensitivity, as indicated by blood alcohol level at time of regaining the righting response, and withdrawal severity might determine whether a greater initial tolerance to alcohol facilitates dependence development.

LOWER ORGANISMS AS MODELS

The long history of the utility of Drosophila melanogaster (fruit flies) in genetic research is well known. Some of these studies are relevant to our discussion of alcohol-related behaviors. For example, McKenzie and Parsons (75) reported that flies derived from a vineyard or from a wine cellar were more resistant to the effects of alcohol than those derived from an area remote from the vineyard. Although the presence of ADH in normal amounts is necessary if the flies are to use ethanol as a carbon source (76), the difference in resistance did not appear to be related to ADH level. Another species, Drosophila simulans, was never found in the wine celler and might, therefore, be called a non-preferring species. Actually, D. simulans does reject media containing 9% alcohol for purposes of oviposition, and this species is also more sensitive to alcohol, both as adults and as larvae, than is D. melanogaster (75,77). While fruit flies are even further removed from the problems of human alcoholism than are laboratory rodents, their short generation time and relatively simple genetic structure might make them an interesting model for further studies on the pharmacogenics of alcoholism.

Genetic influences on responses to alcohol have also been studied in microorganisms. The interest in yeast is obvious, from a commercial as well as a scientific viewpoint, and it has a long history that extends back to Pasteur (78,79). Even bacteria (80,81) and a fungus (82) have not been ignored in the study of the effects of alcohol. Of particular interest is the study by Fried and Novick (81) utilizing mutants of Escherichia coli which are resistant to ethanol. The genetic alteration appears to affect the cell membrane. As the authors point out, this may provide a useful tool for study of the physical biochemistry of the action of ethanol on cell membranes in general.

HUMAN STUDIES

A number of investigators have observed genetic differences in response to ethanol in humans. One of the most frequently studied of these genetically influenced responses is the flush reaction which occurs in Orientals (83,84) and American Indians (85,86). An unpleasant flush reaction occurs in the blush region almost immediately upon ingestion or intravenous administration of ethanol. Wolff (86) has postulated that the response is due to a genetically influenced variation in responsiveness of the vasomotor system. Another possible explanation involves the metabolism of ethanol. During ethanol metabolism, the concentration of NAD drops and there is a corresponding rise in NADH level. It is now widely accepted that, given a normal amount of ADH, the rate-limiting step in ethanol metabolism lies in the reoxidation of NADH to NAD. Under normal conditions, the absolute amount of ADH is not the rate-limiting factor. However, some individuals, notably a large percentage of Japanese, have an "atypical" liver ADH (87,88). Because this enzyme is much more active than the "typical" enzyme, there may be an initial "burst" of ethanol metabolism before the NAD/NADH ratio has been markedly lowered by ethanol, and this phenomenon may overwhelm the ability of the liver to metabolize the resultant acetaldehyde. After a few minutes, the NAD/NADH ratio would drop and reoxidation of NADH and NAD would become rate limiting. This suggestion is supported by the fact that individuals with the "atypical" enzyme do not carry out the overall metabolism of ethanol any more rapidly than do those with the "typical" liver ADH (89,90). Since it is well known that acetaldehyde causes unpleasant reactions, as many disulfiram (Antabuse) users will attest, this product of ethanol oxidation may be responsible for the flush seen in Orientals who ingest ethanol.

The supposed Asiatic origin of the American Indian (85, 86) leads to the speculation that this group may also show a high incidence of the "atypical" liver ADH. A recent paper by Bennion and Li (85), however, casts doubt on this hypothesis. While the flush reaction to ethanol does occur in Indians (86), liver biopsy of seven individuals failed to demonstrate the presence of the "atypical" ADH in any case. Unfortunately, the individuals on whom the biopsies were performed were not tested for the flush reaction, so the relationship between the two variables could not be determined. If it is true that the "atypical" enzyme is involved in the flush reaction, and that this reaction in turn is important in rejection of alcohol by the Japanese, then a paradox arises. The incidence of alcoholism among Japanese is comparatively low, but there is a high incidence among American Indians. It has been suggested (88) that the increased acetaldehyde may in one case be a deterrent to excessive alcohol intake, while it may lead to greater addiction liability in a different culture. (Also, see material presented earlier in this chapter and reference 92 for a discussion of the role of acetaldehyde in the actions of ethanol.)

The question of rate of alcohol metabolism in Indians as compared to Caucasians has been the subject of considerable controversy. Fenna et al. (93) found that Canadian Indians and Eskimos metabolized alcohol more slowly than Caucasians and attributed this difference to racial variation. Their interpretation has been challenged on theoretical grounds by Lieber (94). The study of Bennion and Li (85) found no difference in rate of ethanol metabolism between Southwestern Indians and Caucasians. Despite this controversy, there is no doubt that genetic differences in overall alcohol metabolism do exist among humans. Vessel et al. (95) used the twin method, a common procedure for distinguishing between genetic and environmental effects, to investigate factors affecting ethanol metabolism. A genetic influence was suggested by the finding that dizygotic twins exhibited different rates of metabolism, while monozygotic twins did not. It is possible, however, that an effect due to differential volumes of distribution may also have been involved in these results.

It should be noted that at least two precautions are necessary, but not always taken, in studies of human response to ethanol. First, rate of ethanol metabolism must be expressed in terms of body weight and corrected for volume of distribution. Second, it is important to distinguish between genetic and environmental influences when ethnic groups are compared. While it is relatively easy to do this in twin studies by establishing the zygosity of twin pairs, it may be nearly impossible to determine genotypes of individuals in different ethnic groups. There is no single, objective method of verifying genotype, and family records are seldom detailed enough to indicate the genetic background of an individual.

Much work has been carried out in an attempt to find a genetic marker associated with alcoholism. Needless to say, such a marker would be most useful in counseling and in prevention programs. Positive associations between certain blood groups (A, notably) or color blindness and alcoholism have been reported (see reference 5 for a discussion). More recently, a relationship between other blood groups (S and D, as well as complement C3) and alcoholism has been found (96).

Many studies have shown that alcoholism in humans has a genetic component (97,98,99,100,101). Furthermore, many studies have demonstrated an association between alcoholism and serological markers (e.g., 96). However, the results of these genetic investigations have not provided a biochemical theory to explain human variability in response to alcohol. The development of such a theory, rather than continued confirmation of the fact that there is a genetic influence on human alcoholism, would seem to be a more profitable focus for future research. To date, only the occurrence of an "atypical" ADH in some humans has provided a biochemical clue to human alcoholism and it has not yet been proven that this condition is predictive.

SUMMARY

The literature cited in this review provides unequivocal evidence to support the notion that several alcohol-related behaviors are influenced by genetic factors. Our progress toward and understanding of biochemical mechanisms underlying preference for alcohol, initial sensitivity to alcohol, and the development of physical dependence upon alcohol has been facilitated by the use of animal models that differ genetically in one or more of these behaviors. For example, use of genetic methods has provided a great deal of evidence in support of the hypothesis that high blood acetaldehyde levels are aversive and result in decreased preference for alcohol. The results of studies with disulfiram (Antabuse) lead to the same conclusion. However, any study in which a drug is utilized must be interpreted with caution, since few, if any, drugs have only one effect on a living organism.

It is our hope that readers of this review will not only have been familiarized with the evidence that variability in alcohol-related behaviors is influenced by genetic factors, but will also have become acquainted with the use of genetic analysis as a means of testing hypotheses concerning the actions of alcohol and other drugs. This little-used technique, like the administration of a drug which serves to stimulate or inhibit a biochemical pathway or a physiological process, can be applied to the investigation of underlying mechanisms which might account for variations in a particular behavior. Genetic analyses have the advantage of being less susceptible to the confounding effects which may occur when drugs are used, while they often have the disadvantage of requiring more time and effort. Hopefully, a combination of approaches will provide the tools necessary to elucidate the mechanisms which underlie human alcoholism.

ACKNOWLEDGEMENTS

The authors would like to thank Drs. Gerald McClearn and David Jensen for helpful discussions and Rebecca Miles for assistance in preparation of the manuscript. Studies cited in this review that were carried out in the laboratories of #le authors were supported by NIAAA grant AA-00263 to R.A.D. and a grant from the National Council on Alcoholism to A.C.C.

REFERENCES

1. McClearn, G.E.: Genetics as a tool in alcohol research. Ann. N.Y. Acad. Sci. 197:26-31 (1972).
2. Varela, A.: The role of personality in alcohol abuse, Alcohol and Alcoholism. Edited by Popham, R.E., University of Toronto Press, Toronto, Canada, p. 160-164 (1970).
3. McClearn, G.E. and Rodgers, D.A.: Differences in alcohol preference among inbred strains of mice. Quart. J. Stud. Alc. 20:691-695 (1959).
4. Eriksson, K.: Genetic selection for voluntary alcohol consumption in the albino rat. Science 159:739-741 (1968).
5. Eriksson, K.: Alcohol inhibition and behavior: a comparative genetic approach, Psychopharmacogenetics. Edited by Eleftheriou, B.E., Plenum Press, New York, p. 127-168 (1975).
6. Rodgers, D.A., McClearn, G.E., Bennett, E.L. and Hebert, M.: Alcohol preference as a function of its caloric utility in mice. J. Comp. Physiol. Psychol. 56:663-672 (1963).
7. Fuller, J.L.: Measurement of alcohol preference in genetic experiments. J. Comp. Physiol. Psychol. 57:85-88 (1964).
8. McClearn, G.E. and Rodgers, D.A.: Genetic factors in alcohol preference of laboratory mice. J. Comp. Physiol. 54:116-119 (1961).
9. Eriksson, K. and Pikkarainen, P.H.: Strain and sex differences in voluntary alcohol consumption, liver ADH activity and aldehyde-oxidizing capacity in inbred strains of mice. Jap. J. Stud. Alc. 5:1-7 (1970).
10. Pickett, R.A., II and Collins, A.C.: Use of genetic analysis to test the potential role of serotonin in alcohol preference. Life Sci. 17:1291-1296 (1975).
11. Thomas, K.: Selection and avoidance of alcohol solutions by two strains of inbred mice and derived generations. Quart. J. Stud. Alc. 30:849-861 (1964).
12. Whitney, G., McClearn, G.E. and DeFries, J.C.; Heritability of alcohol preference in laboratory mice and rats. J. Nered. 61:165-169 (1970).
13. Eriksson, K.: The estimation of heritability for the self-selection of alcohol in the albino rat. Ann. Med. Exp. Biol. Fenn. 47:172-174 (1969).
14. Mardones, J.: Experimentally induced changes in the free selection of ethanol. Int. Rev. Neurobiol. 2:41 (1960).
15. Randall, C.L. and Lester, D.: Social modification of alcohol consumption in inbred mice. Science 189:149-151 (1975).
16. Randall, C.L. and Lester, D.: Alcohol selection of DBA and C57BL mice arising from ova transfers. Nature 255:147-148 (1975).
17. McClearn, G.E. and Anderson, S.M.: Ethanol consumption: selective breeding in mice. In preparation (1976).
18. Eriksson, K. and Pikkarainen, P.H.: Differences between the sexes in voluntary alcohol consumption and liver ADH activity in inbred strains of mice. Metabolism 17:10371042 (1968).
19. Eriksson, K. and Malmstrom, K.K.: Sex differences in consumption of alcohol in albino rats. Ann. Med. Exp. Biol. Fenn. 45:389-392 (1967).
20. Eriksson, K.: Rat strains specially selected for their voluntary alcohol consumption. Ann. Med. Exp. Biol. Fenn. 49:67-72 (1971).
21. Brewster, D.J.: Ethanol preference in strains of rats selectively bred for behavioral characteristics. J. Genet. Psychol. 11:217-227 (1969).
22. Russell, K.E. and Stern, M.H.: Sex and strain factors in voluntary alcohol intake. Physiol. Behay. 10:641-642 (1973).
23. Schlesinger, K.: Genetic and biochemical correlates of alcohol preference in mice. Amer. J. Psychiat. 122:767-773 (1966).
24. Koivula, T., Koivulsalo, M. and Lindros, K.O.: Liver aldehyde and alcohol dehydrogenase activities in rat strains genetically selected for their alcohol prefererence. Biochem. Pharmacol. 24:1807-1811 (1975).
25. Eriksson, C.J.P.: Ethanol and acetaldehyde metabolism in rat strains genetically selected for their ethanol preference. Biochem. Pharmacol. 22:2283-2292 (1973).
26. McClearn, G.E.: Genetics and motivation of the mouse, Nebraska Symposium on Motivation. Edited by Arnold, W. J., University of Nebraska Press, Lincoln, Nebraska, p. 47-83 (1968).
27. Schlesinger, K., Kakihana, R. and Bennett, E.L.: Effects of tetra-ethylthiuramdisulfide (Antabuse) on metabolism and consumption of ethanol in mice. Psychosomat. Med. 28:514-520 (1966).
28. Sheppard, J.E., Albersheim, P. and McClearn, G.E.: Enzyme activities and ethanol preference in mice. Biochem. Genet. 2:209-212 (1968).
29. Grunnet, N.: Oxidation of acetaldehyde by rat liver mitochondria in relation to ethanol oxidation and the transport of reducing equivalents across the mitochondrial membrane. Eur. J. Biochem. 35:236-243 (1973).
30. Tottman, S.O.C., Pettersson, H. and Kiessling, K.H.: The subcellular distribution and properties of aldehyde dehydrogenases in rat liver. Biochem. J. 135:577-586 (1973).
31. Corrall, R.J.M., House, P., Margolis, J., Kong, M. and Landau, B.R.: Subcellular site of acetaldehyde oxidation in rat liver. Biochem. Pharmacol. 25:17-20 (1976).
32. Siew, C., Deitrich, R.A. and Erwin, V.G.: Localization and characteristics of rat liver mitochondrial aldehyde dehydrogenases. Arch. Biochem. Biophys., in press (1976).
33. Redmond, G.P. and Cohen, G.: Sex difference in acetaldehyde exhalation following ethanol administration in C57BL mice. Nature 236:117-119 (1972).
34. Koe, K.B. and Tennen, S.S.: Inhibiting action of n-butyraldoxime on ethanol metabolism and on natural ethanol preference of C57BL mice. J. Pharmacol. Exp. Therap. 174:434-449 (1970).
35. Lindros, K.O., Koivula, T. and Eriksson, C.J.P.: Acetaldehyde levels during ethanol oxidation: a diet-induced change and its relation to liver aldehyde dehydrogenase and redox states. Life Sci. 17:1589-1598 (1975).
36. Eriksson, K., Pekkanen, L., Forsander, 0. and Ahtee, L.: The effect of dietary factors on voluntary ethanol intake in the albino rat, The Effects of Centrally Active Drugs on Voluntary Alcohol Consumption. Edited by Sinclair, J.D. and Kiianmaa, K., Finnish Foundation for
Alcohol Studies, Helsinki, Vol. 24, p. 15-26 (1975).
37. Redmond, G. and Cohen, G.: Induction of liver acetaldehyde dehydrogenase: possible role in tolerance after exposure to barbiturates. Science 171:387-389 (1971).
38. Dietrich, R.A., Collins, A.C. and Erwin, V.G.: Genetic influence upon phenobarbital-induced increase in rat liver supernatant aldehyde dehydrogenase activity. J. Biol. Chem. 247:7232-7236 (1972).
39. Roper, M. and Deitrich, R.A.: Preparation of antibodies to the phenobarbital-inducible aldehyde dehydrogenase. In preparation (1976).
40. Petersen, D., Collins, A.C. and Deitrich, R.A.: Influence of induction of aldehyde dehydrogenase on blood acetaldehyde levels. In preparation (1976).
41. Eriksson, C.J.P., Marselos, M. and Koivula, T.: Role of cytosolic rat liver aldehyde dehydrogenase in the oxidation of acetaldehyde during ethanol metabolism in vivo. Biochem. J. 152:709-712 (1975).
42. Tank, W., Weiner, H. and Deitrich, R.A.: Metabolism of dopamine and norepinephrine in liver of animals treated with aldehyde dehydrogenase inducing agents. In preparation (1976).
43. Roper, M., Stock, R. and Deitrich, R.A.: Phenobarbital and tetrachlorodibenzo-p-dioxin induce different isoenzymes of aldehyde dehydrogenase. Fed. Proc. 35:282 (1975).
44. Deitrich, R.A., Bludeau, P.A. and Roper, M.: Induction of different rat liver supernatant aldehyde dehydrogenases by phenobarbital and tetrachlorodibenzo-p-dioxin. In preparation (1976).
45. Forsander, 0.A., Hillbom, M.E. and Lindros, K.O.: Influence of thyroid function on the acetaldehyde level of blood and liver of intact rats during ethanol metabolism. Acta Pharmacol. Toxicol. 27:410-416 (1969).
46. Marselos, M., Eriksson, K. and Hanninen, O.: Aldehyde dehydrogenase inducibility and ethanol preference in rats. Med. Biol. 53:224-230 (1975).
47. Marselos, M. and Pietikainen, S.: Voluntary consumption of ethanol in rats after induction of the soluble aldehyde dehydrogenase in the liver, The Effects of Centrally Active Drugs on Voluntary Alcohol Consumption. Edited by Sinclair, J.D. and Kiianmaa, K., Finnish Foundation for Alcohol Studies, Helsinki, Vol 24, p. 27-34 (1975).
48. Hill, S.Y.: Intraventricular injection of 5-hydroxytryptamine and alcohol consumption in rats. Biol. Psychiat. 8:151-158 (1974).
49. Geller, I.: Effects of parachlorophenylalanine and 5-hydroxytryptophane on alcohol intake in the rat. Pharmacol. Biochem. Behay. 1:361-365 (1973).
50. Myers, R.D., Evans, J.E. and Yakish, T.L.: Ethanol preference in the rat: interactions between brain serotonin and ethanol, acetaldehyde, paraldehyde, 5-THP, and 5-HTOL. Neuropharmacology 11:539-549 (1972).
51. Cicero, T.J. and Hill, S.Y.: Ethanol self selection in rats: a distinction between absolute and 95% alcohol. Physiol. Behay. 5:787-791 (1970).
52. Frey, H.H., Magnussen, M.P. and Nielsen, C.K.: The effect of p-chloroamphetamine on the consumption of ethanol by rats. Arch. Int. Pharmacodyn. 183:165-172 (1970).
53. Sanders, B., Collins, A.C. and Wesley, V.H.: Reduction of alcohol selection by pargyline in mice. Psychopharmacologia 46:159-162 (1976).
54. Nachman, M., Lester, D. and LeMagnen, J.: Alcohol aversion in the rat: behavioral assessment of noxious drug effects. Science 168:1244-1245 (1970).
55. Dembiec, D., MacNamee, D. and Cohen, G.: The effect of pargyline and other monoamine oxidase inhibitors on blood acetaldehyde levels in ethanol-intoxicated mice. J. Pharmacol. Exp. Therap. 197322-339 (1976).
56. Ahtee, L. and Eriksson, K.: 5-hydroxytryptamine and 5-hydroxyindoleacetic acid content in brain of rat strains selected for their ethanol intake. Physiol. Behay. 8:123-126 (1972).
57. Ahtee, L. and Eriksson, K.: Dopamine and doradrenaline content in the brain of rat strains selected for their alcohol intake. Acta Physiol. Scand. 93:563-565 (1975).
58. Kakihana, K., Brown, D.R., McClearn, G.E. and Tabershaw, I.R.: Brain sensitivity to alcohol in inbred mouse strains. Science 154:1574-1575 (1966).
59. McClearn, G.E.: Genetic differences in the effect of alcohol upon behavior of mice, Proceedings of the Third International Conference on Alcohol and Road Traffic. Edited by Havard, J.D.J., British Medical Association, London, p. 153-155 (1962).
60. Belknap, J.K., Maclnnes, J.W. and McClearn, G.E.:
Ethanol sleep times and hepatic alcohol and aldehyde dehydrogenase activities in mice. Physiol. Behay. 9:453457 (1972).
61. McClearn, G.E., Wilson, J.R. and Meredith, W.: The use of isogenic and heterogenic mouse stocks in behavioral research, Contributions to Behavior-Genetic Analysis: The Mouse as a Prototype. Edited by Lindzey, G. and Thiessen, D.D., Appleton-Century-Crofts, New York, p. 322 (1970).
62. Damjanovich, R.P. and Maclnnes, J.W.: Factors involved in ethanol narcosis: analysis in mice of three inbred strains. Life Sci. 13:55-65 (1973).
63. Randall, C.L., Carpenter, J.A., Lester, D. and Friedman, H.J.: Ethanol-induced mouse strain differences in loco-motor activity. Pharmacol. Biochem. Behay. 3:533-535 (1975).
64. Maclnnes, J.W. and Uphouse, L.L.: Effects of alcohol on acquisition and retention of passive avoidance conditioning in different mouse strains. J. Comp. Physiol. Psychol. 81:398-410 (1973).
65. McClearn, G.E. and Kakihan, R.: Selective breeding for ethanol sensitivity in mice. Behay. Genet. 3:409-410 (1973).
66. Heston, W.D.W., Erwin, V.G., Anderson, S.M. and Robbins, H.: A comparison of the effects of alcohol on mice selectively bred for differences in ethanol sleep time. Life Sci. 14:365-370 (1974).
67. Erwin, V.G., Heston, W.D.W., McClearn, G.E. and Deitrich, R.A.: Effects of hypnotic agents on mice with differing central nervous system sensitivity to ethanol. Pharmacol Biochem. Behay., in press (1976).
68. Church, A.C., Fuller, J.L. and Dudek, B.C.: Salsolinol differentially affects mice selected for sensitivity to alcohol. Psychopharmacology 47:49-52 (1976).
69. Cohen, G. and Collins, M.A.: Alkaloids from catecholamines in adrenal tissue: possible role in alcoholism. Science 167:1749-1751 (1970).
70. Chan, A.W.K.: Putative neurotransmitters in inbred mice: effect of ethanol. Pharmacologist 17:198 (1975).
71. Collins, A.C., Lebsack, M.E. and Yeager, T.N.: Mechanisms that underlie sex-linked and genotypically determined differences in the depressant actions of alcohol. Ann. N.Y. Acad. Sci. 273:303-316 (1976).
72. Sanders, B.: Sensitivity to low doses of ethanol and phenobarbital in mice selected for sensitivity to hypnotic doses of ethanol. J. Comp. Physiol. Psychol. 90:394-398 (1976).
73. Goldstein, D.B.: Inherited differences in intensity of alcohol withdrawal reactions in mice. Nature 245:154-156 (1973).
74. Goldstein, D.G. and Kakihana, R.: Alcohol withdrawal reactions in mouse strains selectively bred for long or short sleeptime. Life Sci. 17:981-986 (1975).
75. McKenzie, J.A. and Parsons, P.A.: Micro-differentiation in a mutual population of Drosophila melanogaster to alcohol in the environment. Genetics 77:385-394 (1974).
76. Sofer, W. and Vigue, C.: Adhn5: a temperature-sensitive mutant at the Adh locus in Drosophila. Biochem. Genet.//:387-396 (1974).
77. Parsons, P.A.: Genetics of resistance to environmental stress in Drosophila populations. Ann. Rev. Genet. 7: 239-265 (1973).
78. Wilks, C.: Production of yeast alcohol dehydrogenase isoenzymes by selection. Nature 261:26-29 (1976).
79. Ismail, A.A. and Ali, A.M.M.: Selection of high ethanol-yielding Saccharomyces. II. Genetics of alcohol tolerance. Fol. Microbiol. 16:350-354 (1971).
80. Milhaud, P. and Balassa, G.: Biochemical genetics of bacterial sporulation. IV. Sequential development of resistance to chemical and physical agents during sporulation of Bacillus subtilis. Molec. Gen. Genet. 125:241250 (1973).
81. Fried, V.A. and Novick, A.: Organic solvents as probes for the structure and function of the bacterial membrane: effects of ethanol on the wild type and an ethanol resistant mutant of Escherichia coli. J. Bacteriol. 114: 239-248 (1973).
82. Page, M.M. and Cove, D.J.: Alcohol and amine catabolism in the fungus Aspergillus nidulans. Biochem. J. 127:17P (1972).
83. Wolff, P.H.: Ethnic differences in alcohol sensitivity. Science 175:449-450 (1972).
84. Ewing, J.A., Rouse, B.A. and Pellizzari, E.D.: Aldohol sensitivity and ethnic backg round. Amer. J. Psychiat. 131 :206-210 (1974).
85. Bennion, L.J. and Li, T.K.: Alcohol metabolism in American Indians and Whites: lack of racial differences in metabolic rate and liver alcohol dehydrogenase. New Eng. J. Med. 294:9-12 (1976).
86. Wolff, P.H.: Vasomotor sensitivity to alcohol in diverse mongoloid populations. Amer. J. Hum. Genet. 25:193-199 (1973).
87. von Wartburg, J.P., Papenberg, J. and Aebi, A.: An atypical human alcohol dehydrogenase. Canad. J. Biochem. 43:889-898 (1965).
88. Stamatoyannopulos, G., Chen, S.H. and Fukui, M.: Liver alcohol dehydrogenase in Japanese: high population frequency of atypical form and its possible role in alcohol sensitivity. Amer. J. Hum. Genet. 27:789-796 (1975).
89. von Wartburg, J.P. and Schurch, P.M.: Atypical human liver alcohol dehydrogenase. Ann. N.Y. Acad. Sci. 151: 936-946 (1968).
90. Edwards, J.A. and Price-Evans, D.A.: Ethanol metabolism in subjects possessing typical and atypical liver alcohol dehydrogenase. Clin. Pharmacol. Therap. 8:824829 (1967).
91. Deitrich, R.A.: Biochemistry of alcoholism. Psychoneuroendocrinology, in press (1976).
92. Deitrich, R.A. and Erwin, V.G.: Involvement of biogenic amine metabolism in ethanol addiction. Fed. Proc. 34: 1962-1968 (1975).
93. Fenna, D.M.L., Schaefer, O. and Gilbert, F.A.L.: Ethanol metabolism in various racial groups. Canad. Med. Assoc. J. 105:472-475 (1971).
94. Lieber, C.S.: Metabolism of ethanol and alcoholism: racial and acquired factors. Ann. Intern. Med. 76:326327 (1972).
95. Vesell, E.S., Page, J.G. and Passananti, G.T.: Genetic and environmental factors affecting ethanol metabolism in man. Clin. Pharmacol. Therap. 12:192-201 (1971).
96. Hill, S.Y., Goodwin, D.W., Cadoret, R., Osterland, C.K. and Doner, S.M.: Association and linkage between alcoholism and eleven serological markers. J. Stud. Alc. 36: 981-992 (1975).
97. Goodwin, D.W., Schulsinger, F., Hermanson, L., Guze, S.B. and Winokur, G.: Alcohol problems in adoptees raised apart from alcoholic biological parents. Arch. Gen. Psychiat. 28:238-243 (1973).
98. Amark, C.: A study in alcoholism: clinical, social-psychiatric and genetic investigations. Acta Psychiat. Neurol. Scand., Suppl. 70:108-120 (1951).
99. Winokur, G., Reich, T., Remmer, J. and Pitts, F.N.J.: Alcoholism. III. Diagnosis and familial psychiatric illness in 259 alcoholic probands. Arch. Gen. Psychiat. 23:104-111 (1970).
100. Partanen, J., Bruun, K. and Markkanen, T.: Inheritance of Drinking Behavior: A Study on Intelligence, Personality, and Use of Alcohol by Adult Twins, Finnish Foundation for Alcohol Studies, Helsinki (1966).
101. Mardones, J.: Evidence of genetic factors in the appetite for alcohol and alcoholism. Ann. N.Y. Acad. Sci. 197:138-142 (1972).