Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jan;144:29-36.
doi: 10.1016/j.neuropharm.2018.10.008. Epub 2018 Oct 14.

Ethanol Acts on KCNK13 Potassium Channels in the Ventral Tegmental Area to Increase Firing Rate and Modulate Binge-Like Drinking

Free PMC article

Ethanol Acts on KCNK13 Potassium Channels in the Ventral Tegmental Area to Increase Firing Rate and Modulate Binge-Like Drinking

Chang You et al. Neuropharmacology. .
Free PMC article


Alcohol excitation of the ventral tegmental area (VTA) is important in neurobiological processes related to the development of alcoholism. The ionotropic receptors on VTA neurons that mediate ethanol-induced excitation have not been identified. Quinidine blocks ethanol excitation of VTA neurons, and blockade of two-pore potassium channels is among the actions of quinidine. Therefore two-pore potassium channels in the VTA may be potential targets for the action of ethanol. Here, we explored whether ethanol activation of VTA neurons is mediated by the two-pore potassium channel KCNK13. Extracellular recordings of the response of VTA neurons to ethanol were performed in combination with knockdown of Kcnk13 using a short hairpin RNA (shRNA) in C57BL/6 J mice. Real-time PCR and immunohistochemistry were used to examine expression of this channel in the VTA. Finally, the role of KCNK13 in binge-like drinking was examined in the drinking in the dark test after knockdown of the channel. Kcnk13 expression in the VTA was increased by acute ethanol exposure. Ethanol-induced excitation of VTA neurons was selectively reduced by shRNA targeting Kcnk13. Importantly, knockdown of Kcnk13 in the VTA resulted in increased alcohol drinking. These results are consistent with the idea that ethanol stimulates VTA neurons at least in part by inhibiting KCNK13, a specific two-pore potassium channel, and that KCNK13 can control both VTA neuronal activity and binge drinking. KCNK13 is a novel alcohol-sensitive molecular target and may be amenable to the development of pharmacotherapies for alcoholism treatment.

Conflict of interest statement

Financial Disclosures:

All authors reported no related financial interests or potential conflicts of interest.


Figure 1.
Figure 1.. High extracellular calcium blocks, and isoflurane mimics, ethanol excitation of VTA neurons
A. High Ca++ medium blocks ethanol excitation. Ethanol was tested in the presence and absence of high Ca++ external medium. Increases in Ca++ concentration from normal (2.5 mM) were made by substituting Ca++ for sodium in the external medium. In normal medium, ethanol (40, 80, 120 mM) produced excitation of 3.8 ± 0.8, 9.8 ± 1.4, and 15.8 ± 1.4, respectively. High Ca++ was adjusted to maintain a regular firing rate. Despite this precaution, changing to high Ca++ medium (ranging from 5.8 to 12.5 mM total Ca++) significantly reduced the baseline firing from 2.6 ± 0.21 to 1.9 ± 0.25 Hz (paired t test, t=3.07, df=7, p<0.02, n=8 from 6 mice). In high Ca++ medium, ethanol produced changes in firing rate of −8.9 ± 5.8% (40 mM), −16.5 ± 8.7% (80 mM), and −23.6 ± 17.3% (120 mM). Ethanol excitation was significantly reduced with a significant interaction between concentration and the effect of high Ca++ (two-way RM ANOVA, F1, 7= 6.4, p<0.04 for effect of high Ca++, Tukey post-hoc comparison, * p<0.05; n=8). B. Isoflurane increases the firing rate of VTA neurons. Isoflurane (200–600 µM) was added to the superfusate for 5 minutes, and the change in firing rate was measured. Isoflurane produced a significant concentration-dependent increase in firing (one-way ANOVA, F2, 7= 10.58, p<0.002, n=8 from 4 mice).
Figure 2.
Figure 2.. KCNK13 is expressed in dopamine and non-dopamine neurons in the VTA and Kcnk13 expression is induced by acute alcohol
A. Immunohistochemistry was performed using antibodies to KCNK13 and tyrosine hydroxylase (TH). Left panel: Neurons labeled green are KCNK13-positive. Middle panel: Neurons labeled red are TH-positive. Right panel: Merging the images shows that some neurons (single arrow) are stained for both TH and KCNK13, and other neurons are TH negative but KCNK13 positive (double arrow). B. Quantitative PCR was performed in tissue from mice given a systemic injection of ethanol (EtOH, 3 g/kg, IP) or saline (Control) one hour or four hours before sacrifice. There was no change in Kcnk13 mRNA one hour after ethanol administration; four hours after ethanol treatment, there was a significant increase in Kcnk13 mRNA (unpaired t test, t = −2.25, p<0.05. n=9).
Figure 3.
Figure 3.. In vivo downregulation of Kcnk13 with lentiviral siRNA
A. Reduced expression of Kcnk13 mRNA in the VTA of mice expressing shKcnk13 compared with mice expressing the control shRNA, shScr. Effectiveness of Kcnk13 shRNA in the VTA was assessed by qPCR of RNA from dissected VTA three weeks after lentivirus infection. Kcnk13 was reduced by 22.2% (unpaired t-test, t-statistic= 1.92,* p< 0.05, n=6). B, C. Reduced expression of KCNK13 protein in the VTA of mice expressing shKcnk13. Immunohistochemistry was performed three weeks after lentiviral infection using antibodies to KCNK13 and GFP. KCNK13 immunoreactivity was compared between GFP-positive neurons and GFP-negative neurons. B. GFP-positive neurons in mice expressing shKcnk13 expressed 15.1± 3.4% less KCNK13 immunoreactivity than GFP-negative neurons (paired t-test, t= −4.88, p<0.001, n=18). In mice expressing shScr in the VTA, GFP-positive neurons expressed 0.64 ± 3.4% more KCNK13 immunoreactivity compared to GFP-negative neurons (paired t-test, t= 0.305, p>0.05, n=15). Asterisk (*) indicates significant difference between GFP-positive and GFP-negative neurons. C. Illustration of the typical location of microinjection sites (inset picture showing VTA region immunostained for GFP and TH) and immunostaining of KCNK13 and GFP in mice expressing shKcnk13 and shScr (four larger pictures). Single arrowhead indicates KCNK13-immunoreactive cells, double arrowhead indicates cells immunoreactive for both KCNK13 and GFP. Note the reduced KCNK13 immunoreactivity in the GFP-immunoreactive cells.
Figure 4.
Figure 4.. Ethanol excitation is attenuated with in vivo downregulation of Kcnk13
A, B Ratemeter graphs of the responses to ethanol of single neurons after VTA infection with lentivirus expressing either shScr (A) or shKcnk13 (B). Ethanol (40, 80, 120 mM) produced excitation of (A) 1.85, 26.5, and 37.7 %, and (B) 5.5, 9.9, and 12.8 %. C. Pooled results from extracellular recordings of VTA neurons in brain slices from mice expressing either shKcnk13 or shScr. The effect of ethanol (40, 80 120 mM) on firing rate was tested. Ethanol excitation was significantly decreased in VTA neurons from shKcnk13 mice compared with control (two-way RM ANOVA, F1, 45= 17.34, p<0.006 for effect of shRNA, n=8). Mean baseline firing between the treatment groups was not significantly different (shScr 1.90 ± 0.46 Hz, shKcnk13, 2.50 ± 0.46 Hz; unpaired t-test, t=1.08, p > 0.05, n=8)
Figure 5.
Figure 5.. Knockdown of Kcnk13 increases binge-like ethanol consumption
A. Lentivirus expressing shRNA targeting Kcnk13 (shKcnk13) or a non-targeting shRNA (shScr) was microinjected into the mouse VTA. Three weeks after injection, mice were tested for bingelike alcohol drinking. Mice injected with shKcnk13 (n=10) drank significantly more during the two-hour drinking sessions than mice injected with shScr (n=6) (two-way ANOVA, effect of shRNA, F1, 84=13.5, p<0.001). B. On day 4 and day 11 of the Drinking in the Dark protocol, mice were given access to ethanol solutions for 4 hours instead of 2 hours. There was no significant difference between the groups in the amount of ethanol consumed in 4 hours. C. Blood ethanol concentrations were assessed after the four-hour drinking session on day 11. No significant difference in blood alcohol levels between the groups was observed. D. Ethanol preference (assessed during the drinking in the dark test), is calculated as the percent of ethanol consumed over total fluid consumption. There was no significant difference in ethanol preference, but as both groups preferred ethanol to a great extent (~80%), there may have been a ceiling effect; that is, as both the shKcnk13 (n=10) and the control (n=6) groups preferred ethanol to such a great extent, there was little room to distinguish between the groups (two-way RM ANOVA, F1, 4=0.811, p>0.05). E. Following two weeks of ethanol drinking, mice that had received injections of either lentiviral-delivered Kcnk13 shRNA (shKcnk13, n=6) or scrambled control (shScr, n=6) were tested for sucrose consumption for four days, for two hours per day. There was no significant difference between the groups (two-way ANOVA, F1, 40=0.26, p>0.05).
Figure 6.
Figure 6.. Model for the role of KCNK13 in the VTA and in the response to ethanol
A. Under normal conditions, KCNK13 channels and other leak potassium channels are open, helping to maintain the negative resting membrane potential of VTA neurons. At rest there is spontaneous action potential firing due to pacemaker activity in the neuron, and this causes baseline release of neurotransmitters in target regions like the nucleus accumbens and prefrontal cortex. B. With the addition of ethanol, KCNK13 channels are blocked, causing a depolarization of the membrane and increased firing rate. The increase in the firing rate of the DA VTA neuron results in increased release of neurotransmitter in target regions.

Similar articles

See all similar articles

Cited by 4 articles

Publication types

MeSH terms