Hybrid C57BL/6J ?? FVB/NJ Mice Drink More Alcohol than Do C57BL/6J Mice

UNLIKE WITH HUMAN subjects, assessment of the reinforcing effects of ethanol in rodents is necessarily a matter of inference. The two basic approaches are to allow the animal to self-administer ethanol or to use principles of Pavlovian conditioning to pair ethanol's effects with a specific cue and assess approach or avoidance of that cue in a subsequent drug-free test (Cunningham and Phillips, 2003). The earliest studies of voluntary self-administration (Richter and Campbell, 1940) were adapted from the nutrition field and offered rats two bottles, one containing an alcohol solution in tap water and the other tap water alone. These studies reported that self-administration was dependent on concentration of ethanol offered. Individual differences in ethanol preference among rats also were noted. This basic procedure is usually termed a “two-bottle ethanol preference” test, which has many variants. The earliest attempts to document a genetic contribution to individual differences in preference drinking were successful. Mardones and Segovia–Requelme (1983) successfully bred rats to prefer or avoid ethanol-containing solutions, and McClearn and Rodgers (1959) demonstrated that C57BL/6 (B6) inbred mice had nearly absolute preference for 10% ethanol over tap water, that DBA/2 mice were near-teetotalers, and that other inbred strains showed intermediate preference. Fairly large surveys of 14 (Rodgers, 1972) or 15 (Belknap et al., 1993) inbred strains reinforced the primacy of B6 but revealed no other strains with higher preference. The genetic contribution to the trait was further evidenced by the relatively high preference of other strains (C57 L/J, C57BR/cdJ) from the C57BL lineage (Belknap et al., 1993). Subsequent analyses of even larger panels of inbred mouse strains (28 strains: Bachmanov et al., personal communication; 22 strains: Finn et al., unpublished) also have failed to reveal a more extreme preferrer of alcohol solutions than B6. Ethanol intake is often reported using two indices. One is the “preference ratio,” or the proportion of total fluid intake that is taken from the ethanol bottle in a two-bottle ethanol preference test. As already noted, preference ratio depends on the concentration offered, and the usual finding is that preference increases as concentration increases, up to a point, but thereafter, preference ratios decline. That is, the relationship is an inverted-U shaped curve. The second index, g of ethanol consumed per kg body weight, gives a direct measure of the amount of ethanol consumed and relates directly to blood ethanol concentration. In the studies reported below, we report both indices but base our interpretations of the data only on g/kg intake. From the strain surveys reported above, and from the data reported in many (>100) other published studies of 24-hr two-bottle ethanol preference, it has been well-documented that male B6 mice will self administer ethanol in the range of 10–14 g/kg/day, while female B6 mice will self-administer in the range of 12–18 g/kg/day, when the choice is between a 10% ethanol solution and water. Many null mutants and transgenics have been tested for ethanol preference. Drinking in these studies necessarily compares intake of the null mutant versus the wild-type onto which the mutation has been bred. While some of the wild-types in these studies are B6, many are not, and often the wild type is itself a hybrid of two or more inbred strains, a segregating intercross, or a partial backcross to an inbred strain. A survey of many such studies reported that about 1/3 found increased ethanol preference versus wild-type in the null mutants, 1/3 found a reduction, and about 1/3 no effect (Cunningham and Phillips, 2003). However, there are no reports of which we are aware where substantially greater g/kg doses of ethanol than those self-administered by B6 are reported. In one of our laboratories (Texas), we have been screening null mutants for ethanol preference. The background strain on which the mutant α1 S267Q glycine receptor transgenic was held was the F1 hybrid of B6 and FVB/NJ (FVB). As reported in Experiment 1, we noted a very high intake of ethanol in the B6×FVB F1 hybrids. To substantiate this finding, and to test the hypothesis of an effect of maternal genotype, we attempted to repeat the finding in another laboratory (Oregon). As we are also exploring other methods for inducing high levels of ethanol self-administration (Finn et al., 2005; Rhodes et al., 2005), we also tested B6×FVB F1 mice in these newer, limited access procedures. We found that B6×FVB F1 mice indeed self administer extremely high levels of ethanol in two-bottle preference tests. They appeared to drink more than B6 mice in the standard two-bottle preference test, and resembled B6 in the newer assays that lead to higher brain concentrations of ethanol.