Nf1 regulates alcohol dependence-associated excessive drinking and gamma-aminobutyric acid release in the central amygdala in mice and is associated with alcohol dependence in humans

Abstract

Background: The neurofibromatosis type 1 (Nf1) gene encodes a GTPase activating protein that negatively regulates small GTPases of the Ras family.

Methods: We assessed alcohol-related behaviors including alcohol sensitivity, dependent and nondependent drinking, and basal and alcohol-induced gamma-aminobutyric acid (GABA) release in the central nucleus of the amygdala (CeA) in Nf1 heterozygous null mice (Nf1(+/-)). We also investigated the associations of NF1 polymorphisms with alcohol dependence risk and severity in humans.

Results: Nf1(+/-) mice do not differ from wild-type mice in nondependent drinking, such as 24-hour, 2-bottle choice drinking in the dark binge drinking or limited access 2-bottle choice. However, Nf1(+/-) mice failed to escalate alcohol drinking following chronic intermittent ethanol vapor exposure (CIE) to induce dependence. Alcohol acutely increases GABA release in the CeA and alcohol dependence is characterized by increased baseline GABA release in CeA. Interestingly, GABA release in Nf1(+/-) mice is greater at baseline than wild-type mice, is not elevated by induction of dependence by CIE, and failed to show alcohol-induced facilitation both before and after CIE. Additionally, we observed that multiple variants in the human NF1 gene are associated with a quantitative measure of alcohol dependence in both African Americans and European Americans.

Conclusions: In this translational investigation, we found that Nf1 activity regulates excessive drinking and basal and ethanol-stimulated GABA release in the mouse central amygdala. We also found that genetic variation in NF1 may confer an inherent susceptibility to the transition from nondependent to dependent drinking in humans.

Keywords: Alcohol dependence; Amygdala; Electrophysiology; GABA; Genetic association; Presynaptic mechanisms.

Figure 1

Alcohol intake by Nf1 heterozygous null mice (Nf1^+/−) did not differ from wild-type (WT) mice in a 24-h, 2-bottle choice (2BC) paradigm or in the drinking in the dark (DID) paradigm. A) In the 24-h paradigm, both WT and Nf1^+/− mice consumed progressively greater amounts of ethanol at increasing alcohol concentrations (3-20% v/v). Significant main effects of ethanol dose (F_5,95 = 90.23, p < .0001), but not genotype or the interaction of ethanol dose and genotype were revealed by repeated measures two-way ANOVA. B) Both WT and Nf1^+/− mice consumed ethanol to a similar degree in the DID. There was no significant difference between genotypes (WT and Nf1^+/−) in 2-hr (day 1-3) and 4-hr (day 4) sessions.

Figure 2

Nf1 heterozygous null mice (Nf1^+/−) did not show increased alcohol intake after vapor exposure to induce dependence. A) Schematic diagram of CIE. B) Nf1^+/− mice consumed significantly less ethanol in limited access 2-hour drinking sessions after vapor treatments, compared with WT mice. The vapor treatments induced a significant increase in ethanol intake in WT mice, but not in Nf1^+/− mice. Significant main effects of genotype (F_1,15 = 9.995, p = .007) and session (F_3,45 = 3.787, p = .017), and the interaction of genotype and session (F_3,45 = 3.105, p = .036) were revealed by repeated measures two-way ANOVA. Post-hoc analysis confirmed that WT mice consumed more alcohol than Nf1^+/− mice after each alcohol vapor bout. *p < .05; ***p < .001 versus before vapor (WT mice only), ^#p < .05; ^##p < .01 Nf1^+/− versus WT mice at the same point (Bonferroni post hoc test). C) Water intake did not change through the experiment and was not different between WT and Nf1^+/− mice.

Figure 3

Nf1^+/− mice display greater baseline spontaneous inhibitory neurotransmission in CeA neurons than wild-type mice. A. Summary of baseline membrane properties of neurons from WT (n = 17) and Nf1^+/− mice (n = 14), *p < .05 as compared to WT mice. B. Representative whole-cell voltage-clamp recordings of sIPSCs in CeA neurons from WT mice (upper trace) and Nf1^+/− mice (lower trace). C. Summary of the mean sIPSC frequency in CeA neurons from WT mice (white bar) and Nf1^+/− mice (black bar). D. Summary of the mean sIPSC frequency of individual sIPSC amplitude bins in CeA neurons from Nf1^+/− mice (black bars) as compared to CeA neurons from WT mice (white bars); *p < .05 by unpaired t-test, n = 9 (WT) and n = 12 (Nf1^+/−)

Figure 4

Nf1^+/− mice display greater baseline miniature inhibitory neurotransmission in CeA neurons than wild-type mice. A. Representative whole-cell voltage-clamp recordings of mIPSCs in CeA neurons from a WT mouse (upper trace) and a Nf1^+/− mouse (lower trace). B. Summary of the mean mIPSC frequency in CeA neurons from WT mice (white bar) and Nf1^+/− mice (black bar). C. Summary of the mean mIPSC frequency of individual sIPSC amplitude bins in CeA neurons from Nf1^+/− mice (black bars) as compared to CeA neurons from WT mice (white bars); *p < .05 by unpaired t-test, n = 13 (WT) and n = 14 (Nf1^+/−).

Figure 5

Nf1^+/− mice do not display the ethanol-induced increase in inhibitory neurotransmission observed in wild-type mice. A. Representative whole-cell voltage-clamp recording from a WT CeA neuron before (Control) and during superfusion of ethanol (EtOH, 44 mM). B. Representative whole-cell voltage-clamp recording from an Nf1^+/− CeA neuron before (Control) and during superfusion of ethanol (EtOH, 44 mM). C. Summary of the change in mean mIPSC frequency and amplitude produced by ethanol superfusion; *p < .05 by one sample t-test for independent significance, ^#p < .05 by unpaired t-test; n = 8 (WT) and n = 12 (Nf1^+/−). D. Summary of the change in mean mIPSC frequency by amplitude of mIPSC produced by ethanol superfusion in CeA neurons from Nf1^+/− mice (black bars) as compared to CeA neurons from WT mice (white bars); *p < .05 by unpaired t-test, n = 8 (WT) and n = 12 (Nf1^+/−).

Figure 6

CIE WT mice display greater baseline spontaneous inhibitory neurotransmission in CeA neurons than air control WT mice but Nf1^+/− mice show no effect of CIE on baseline spontaneous inhibitory neurotransmission. A. Representative whole-cell voltage-clamp recordings of sIPSCs in CeA neurons from an air control WT mouse (upper trace), a CIE WT mouse (middle trace), and a CIE Nf1^+/− mouse (lower trace). B. Summary of the mean sIPSC frequency in CeA neurons from air control WT mice and naive Nf1^+/− mice (white bars) and CIE WT and CIE Nf1^+/− mice (black bars) *p < .05 by unpaired t-test, n = 14 (air control WT), n = 11 (CIE WT), n = 12 (naive Nf1^+/−) and n = 16 (CIE Nf1^+/−). C. Summary of the mean sIPSC frequency of individual sIPSC amplitude bins in CeA neurons from air control WT mice (white bars), CIE WT mice (gray bars) and CIE Nf1^+/− mice (black bars); *p < .05 by unpaired t-test, n = 14 (air control WT), n = 11 (CIE WT), n = 12 (naive Nf1^+/−) and n = 16 (CIE Nf1^+/−).

Figure 7

CIE WT mice display greater baseline miniature inhibitory neurotransmission in CeA neurons than air control WT mice but Nf1^+/− mice show no effect of CIE on baseline miniature inhibitory neurotransmission. A. Representative whole-cell voltage-clamp recordings of mIPSCs in CeA neurons from an air control WT mouse (upper trace), a CIE WT mouse (middle trace), and a CIE Nf1^+/− mouse (lower trace). B. Summary of the mean mIPSC frequency in CeA neurons from air control WT mice and naive Nf1^+/− mice (white bars) and CIE WT and CIE Nf1^+/− mice (black bars) *p < .05 by unpaired t-test, n = 14 (air control WT), n = 9 (CIE WT), n = 9 (naive Nf1^+/−) and n = 15 (CIE Nf1^+/−). C. Summary of the mean mIPSC frequency of individual sIPSC amplitude bins in CeA neurons from air control WT mice (white bars), CIE WT mice (gray bars) and CIE Nf1^+/− mice (black bars); *p < .05 by unpaired t-test, n = 14 (air control WT), n = 11 (CIE WT), n = 12 (naive Nf1^+/−) and n = 16 (CIE Nf1^+/−). D. Representative whole-cell voltage-clamp recording from a CIE WT CeA neuron before (Control) and during superfusion of ethanol (EtOH, 44 mM). E. Summary of the change in mean mIPSC frequency produced by ethanol in air control WT (n = 9), CIE WT (n = 8), naive Nf1^+/− (n = 12) and CIE Nf1^+/− (n = 13); *p < .05 by one sample t-test. F. Summary of the change in mean mIPSC frequency by amplitude of mIPSC produced by ethanol in CeA neurons from air control WT mice (white bars), CIE WT mice (gray bars) and CIE Nf1^+/− mice (black bars); *p < .05 by one-sample t-test, n = 9 (air control WT), n = 8 (CIE WT), n = 12 (naive Nf1^+/−), and n = 13 (CIE Nf1^+/−).