Article Figures & Data
Figures
- Figure 1.
QR2 inhibition in the aIC rescues scopolamine-induced amnesia. a, Rats were given a novel or familiar taste and killed 3 h later. b, A significant reduction in QR2 mRNA was measured in the aIC of rats following novel taste. c, QR2 mRNA expression was significantly reduced in the aIC of mice following novel taste. d, Mice injected with shQR2 or scrambled lentivirus to the aIC underwent novel taste learning, and 2 d later they were given a choice test and preference was assessed. e, Mice injected to the aIC with shQR2 showed significantly improved novel taste memory compared with controls. f, Mice injected to the aIC with shQR2 showed significantly reduced QR2 mRNA levels compared with controls. g, Cannula position validation following behavioral tests. h, Rats were given a novel taste after receiving an intracranial infusion of vehicle or S29434 20 min prior, or received a novel taste and then received S29434 5 min later. The next day, the preference for the novel taste was measured. i, S29434-mediated inhibition of QR2 significantly improved novel taste memory, whether given before or after novel taste presentation. j, Rats were given a novel taste after having first received an intracranial infusion of vehicle or S29434 20 min prior and an intraperitoneal injection of vehicle or scopolamine 1 h before. The following day taste memory was tested. k, Scopolamine significantly impedes novel taste memory, while S29434 prevents scopolamine-induced memory deficit. Data are shown as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001. See supporting data in Extended Data Figure 1-1 and Extended Data Figure 1-2.
- Figure 2.
Local GABAA receptor activity in the aIC is upstream of muscarinic AChR-dependent reduction in QR2 mRNA levels. a, Rats were given a novel or familiar taste 20 min after having received injections of vehicle, bicuculline, or muscimol in the aIC and were killed 3 h later. b, A significant reduction of QR2 mRNA was measured in the aIC of rats that received injections of a novel taste and vehicle in the aIC. Animals that received bicuculline all showed a reduction in QR2 mRNA, while no reduction in QR2 mRNA was measured in animals receiving muscimol. *Main effect (treatment); #effect within treatments. c, Implanted cannula positions were validated following behavioral tests. d, Rats received either scopolamine or vehicle and 40 min later received local aIC infusion of bicuculline. After 20 min, the animals were given a familiar taste and were killed 3 h later. e, Scopolamine prevents the local, bicuculline-dependent reduction in QR2 mRNA in rat aIC. f, Rats received injections of either bicuculline or vehicle locally to aIC, and 20 min later were given a novel taste. The next day, the animals were given a choice test between the novel taste and water, and preference for the newly learned taste was measured. g, Rats that received bicuculline had significantly better memory of the novel taste compared with controls. Data are shown as the mean ± SEM. #p < 0.05; *p < 0.05; **p < 0.01. See supporting data in Extended Data Figure 2-1.
- Figure 3.
miR-182 is upregulated locally in the aIC following novel taste consumption in a GABAAR-mediated, muscarinic AChR-dependent manner. a, QR2 pre-mRNA levels were unchanged in the aIC of mice following novel taste. b, miR-182 expression was measured 1 and 3 h following familiar or novel taste consumption in the aIC of mice. miR-182 expression was elevated 1 h, but not 3 h, following novel taste. c, Rats were given intraperitoneal injections of scopolamine or vehicle 1 h—and then aIC infusions of vehicle or bicuculline 20 min—before familiar taste. Animals were killed 1 h later, and miR-182 was measured. d, miR-182 expression was significantly increased in the aIC after 1 h, in rats that received bicuculline locally to the aIC and vehicle intraperitoneally. e, Infusions of vehicle or Eserine then were given locally to the aIC of rats 20 min before familiar taste. Animals were killed 1 h later, and miR-182 was measured. f, miR-182 expression significantly increased in rat aICs that received Eserine locally, compared with vehicle. g, One of four predicted miR-182 seeds to QR2 mRNA binding sites (miR-182, green; QR2 mRNA, red) in the rat. Data are shown as the mean ± SEM. *p < 0.05; **p < 0.01. See supporting data in Extended Data Figure 3-1 and Extended Data Figure 3-2.
- Figure 4.
Increased miR-182 expression in the aIC reduces local QR2 mRNA levels and improves novel taste learning. a, Lentiviral vectors containing either reporter GFP and miR-182, or GFP alone, under the CMV promoter were generated. b, Transfected HEK293FT cells showed significantly increased levels of miR-182 with the overexpression plasmid, compared with that of GFP. c, QR2 mRNA measured showed significantly reduced expression in cells transfected with the miR-182 plasmid. d, Mice were injected with an miR-182 or GFP construct harboring lentivirus in the aIC and then underwent novel taste learning, and were then given a choice test to assess taste memory. Mice were killed to assess infection efficacy. e, Mice injected with miR-182 showed significantly improved memory for the novel taste compared with GFP-injected controls. f, aIC samples taken from miR-182 overexpression vector-injected mice following completion of behavioral experiments show a significant increase in miR-182 expression compared with GFP-injected controls. g, QR2 mRNA was significantly reduced in the aIC of mice injected with the miR-182 virus, compared with controls. h, Schematic representation of the train of events on novel taste exposure, leading to reduced QR2 expression and improved novel taste memory in the rodent aIC. Data are shown as the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. See supporting data in Extended Data Figure 4-1.
- Figure 5.
Endogenous QR2 activity generates cellular ROS and affects Kv2.1 channel redox state. a, QR2 activity is strongly inhibited by S29434. b, NQO1 activity is not affected by S29434, but is sensitive to dicoumarol. c, Endogenous mouse brain NQO1 and QR2 activity can be measured with BNAH and is sensitive to QR2 inhibitor S29434. d, QR2 uses small molecules of the mouse brain cytoplasm as cofactors to reduce menadione to menadiol in a S29434-sensitive manner. e, HEK293FT cells given reductase cofactor BNAH show reduction in the ROS signal, which is sensitive to dicoumarol, but not to S29434. f, Basal ROS levels in HEK293FT cells are reduced by S29434 in a dose-dependent manner. g, Kv2.1 clumping in the mouse aIC is significantly reduced by infection with shQR2 compared with scrambled control lentivirus. h, Kv2.1 clumping in the mouse aIC trends toward significant reduction by S29434 compared with vehicle. i, Schematic representation of the molecular events following QR2 downregulation. Upon reduced QR2 expression or activity, either naturally or experimentally, there is a reduction in cellular ROS. In the mouse brain, this is exemplified by the redox state of ROS-sensitive Kv2.1 potassium channels. Although Kv2.1 channels are likely not the only molecules affected by QR2 ROS modulation, functional Kv2.1 conductance has been shown to be important for proper memory formation. Data are shown as the mean ± SEM. #p = 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. See supporting data in Extended Data Figure 5-1 and Extended Data Figure 5-2.
Extended Data
Table 1-1
Detailed Statistical Analysis. Download Table 1-1, DOCX file.
Figure 1-1
QR2 mRNA expression is unchanged in brain areas not associated with taste memory following novel taste consumption. a, QR2 mRNA is unchanged in rat OC following novel taste consumption. b, QR2 mRNA is unchanged in mouse CA1 following novel taste consumption. Download Figure 1-1, TIF file.
Figure 1-2
Lentivirus containing either shRNA targeting QR2 or a scrambled control injected to aIC did not alter QR2 mRNA expression in CA1. a, Diagram of lentivirus containing shRNA targeting QR2, or a scrambled control, used to reduce QR2 expression in mice. b, QR2 expression in CA1 remains unaffected by local aIC infection with lentivirus harboring shRNA targeting QR2. Download Figure 1-2, TIF file.
Figure 2-1
Local aIC GABAA receptor antagonism and scopolamine injections do not affect QR2 mRNA in the occipital cortex. a, QR2 mRNA expression is unchanged following novel taste or antagonism of GABAAR with bicuculline locally in the aIC. b, QR2 expression remains unchanged in the OC of rats, following local antagonism of GABAAR in the aIC, with or without prior injections of scopolamine. Download Figure 2-1, TIF file.
Figure 3-1
miR-182 expression does not increase in the mouse CA1 following novel taste consumption or in the rat OC following pharmacological manipulation locally to the rat aIC. a, miR-182 levels remain unaltered in the CA1 of mice following novel taste consumption. b, miR-182 levels remain unchanged in the OC of rats, following local aIC antagonism of GABAAR with bicuculline, with or without prior injection of scopolamine. Download Figure 3-1, TIF file.
Figure 3-2
Predicted hybridization sites of miR-182 to QR2 mRNA in the human, mouse, and rat genome. Download Figure 3-2, TIF file.
Figure 4-1
Mice injected in IC with a lentivirus harboring shRNA targeting QR2 do not show changes in QR2 mRNA or miR-182 expression in the hippocampus CA1. a, miR-182 levels in CA1 were not elevated following local aIC infection with a lentivirus overexpressing miR-182. b, QR2 mRNA levels in CA1 were not changed following local aIC infection with a lentivirus overexpressing miR-182. Download Figure 4-1, TIF file.
Figure 5-1
Time-course of menadiol formation by QR2 activity with or without S29434, using endogenous brain cytoplasmic small molecules as cofactors. Download Figure 5-1, TIF file.
Figure 5-2
Cysteine redox in Kv2.1 is sensitive to ROS generated by QR2 activity. a, Kv2.1 blot following nonreducing gel electrophoresis of mouse aIC. Top, bottom, Two separate experiments in which either lentivirus containing shRNA targeting QR2 mRNA (top) or a scrambled control (bottom) was injected into the mouse aIC. b, Kv2.1 blot following nonreducing gel electrophoresis of mouse aIC samples. Top, bottom, Two separate experiments in which either QR2 inhibitor S29434 (top) or vehicle (bottom) was injected intraperitoneally. c, Kv2.1 oligomerization due to aIC redox state, as seen in a (top) is abolished in the blot following the addition of β-mercaptoethanol. d, Kv2.1 oligomerization due to aIC redox state, as seen in b (top), is abolished in the blot following the addition of β-mercaptoethanol. Download Figure 5-2, TIF file.
In this issue
