mTORC2 in the dorsomedial striatum of mice contributes to alcohol-dependent F-Actin polymerization, structural modifications, and consumption

Abstract

Actin is highly enriched at dendritic spines, and actin remodeling plays an essential role in structural plasticity. The mammalian target of rapamycin complex 2 (mTORC2) is a regulator of actin polymerization. Here, we report that alcohol consumption increases F-actin content in the dorsomedial striatum (DMS) of mice, thereby altering dendritic spine morphology in a mechanism that requires mTORC2. Specifically, we found that excessive alcohol consumption increases mTORC2 activity in the DMS, and that knockdown of Rictor, an essential component of mTORC2 signaling, reduces actin polymerization, and attenuates the alcohol-dependent alterations in spine head size and the number of mushroom spines. Finally, we show that knockdown of Rictor in the DMS reduces alcohol consumption, whereas intra-DMS infusion of the mTORC2 activator, A-443654, increases alcohol intake. Together, these results suggest that mTORC2 in the DMS facilitates the formation of F-actin, which in turn induces changes in spine structure to promote and/or maintain excessive alcohol intake.

Fig. 1

Binge drinking of alcohol increases F-actin content in the DMS. Mice underwent IA20%-2BC for 8 weeks. The DMS and DLS were dissected at the end of the last 4 h-binge (B, black) alcohol drinking session. Control animals had access to water only (W, white). a Schematic representation of the experimental timeline, and DMS and DLS dissection. b, c Filamentous (F-) and globular (G-) actin content in the DMS (b) and DLS (c) were determined by western blot analysis. Data are presented as the mean ratio of F-actin or G-actin/total (F + G) actin ± S.E.M and expressed as percentage of water control. b DMS: F-actin t₍₁₂₎ = 2.79, p = 0.016; G-actin t₍₁₂₎ = 3.66 p = 0.003; n = 7 water, alcohol. c DLS: F-actin t₍₁₄₎ = 0.35, p = 0.734; G-actin t₍₁₄₎ = 0.48 p = 0.637; n = 8 water, alcohol. *p < 0.05, **p < 0.005

Fig. 2

Binge drinking of alcohol activates mTORC2 in the DMS. Mice drinking paradigm and dissection scheme are outlined in Fig. 1a. The DMS and DLS were dissected at the end of the last 4 h-binge (B, black) alcohol drinking session. Control animals had access to water only (W, white). a, b Ser⁹GSK3β phosphorylation (pGSK3β) was measured by western blot analysis. GAPDH immunoreactivity was used as an internal loading control. Data are presented as the mean ratio of Ser⁹GSK3β to total GSK3β ± S.E.M, and expressed as percentage of water control. a DMS: t₍₁₅₎ = 3.87, p = 0.0015; n = 8 water, 9 alcohol. b DLS: t₍₇₎ = 0.81, p = 0.444; n = 4 water, 5 alcohol. c, d Ser⁴²²SGK1 phosphorylation (pSGK1) was measured by western blot analysis. GAPDH immunoreactivity was used as an internal loading control. Data are presented as the mean ratio of Ser⁴²²SGK1 to total SGK1 ± S.E.M, and expressed as percentage of water control. c DMS: t₍₁₅₎ = 4.87, p = 0.0002; n = 8 water, 9 alcohol. d DLS: t₍₁₅₎ = 0.31, p = 0.762; n = 8 water, 9 alcohol. **p < 0.01, ***p < 0.001

Fig. 3

Downregulation of mTORC2 in the DMS reduces F-actin content. Lentivirus (ltv)-expressing shRictor or SCR and GFP reporter was bilaterally infused into the DMS. Striatal tissues were collected 4 weeks after virus infusion and used for IHC (a) or western blot (b, c) analysis. a DMS-containing slices were co-stained with anti-GFP and NeuN antibodies. Left image (×5 magnification) depicts the specificity of the infection site; scale bar 1 mm. Right images (×20 magnification) depicts DMS-infected neurons; scale bar 40 μm. b Levels of Rictor, AKT and Ser⁴⁷³AKT phosphorylation (pAKT) were measured by western blot analysis. GAPDH immunoreactivity was used as an internal loading control. Data are presented as the mean ratio of Rictor/GAPDH or pAKT/AKT ± S.E.M. b Rictor knockdown t₍₁₂₎ = 4.10, p = 0.0015. b Ser⁴⁷³AKT phosphorylation t₍₁₂₎ = 3.19, p = 0.008, n = 7 per virus treatment. c The filamentous (F) and globular (G) actin contents were determined by western blot analysis. Data are presented as the mean ratio of F-actin or G-actin /total (F + G) actin ± S.E.M, and expressed as the percentage of ltv-SCR. F-actin, t₍₉₎ = 3.36, p = 0.0084; G-actin, t₍₉₎ = 3.4, p = 0.008. n = 5–6 per virus treatment; **p < 0.01

Fig. 4

Alcohol regulates MSN dendritic spine morphology via mTORC2. Mice underwent 4 weeks of IA20%-2BC. Mice consuming water only were used as control. Low titer of Ltv-shRictor or ltv-SCR (1 × 10⁵ pg/ml) was infused bilaterally into the DMS, and after 1 week of recovery, mice were given 4 more weeks of IA20%-2BC access. Four hours after the beginning of the last drinking session, mice were perfused, and MSNs morphology was analyzed. a Schematic representation of the experiment timeline. b Left, sample image of a GFP-positive DMS MSN; Right, semi-automated two-dimensional reconstruction of the GFP-positive neuron. Scale bar, 100 μm. c, d Analysis of neuronal dendritic arborization. Sholl analysis was performed on reconstructed neurons and the area under the curve was calculated. Two-way ANOVA showed a significant main effect of alcohol (F_(1,46) = 27.24, p < 0.001) but no effect of virus (F_(1,46) = 0.007, p = 0.935) and no interaction (F_(1,46) = 0.17, p = 0.683) between the four conditions (water/SCR (blue), water/shRictor (hatched blue), alcohol/SCR (orange), alcohol/shRictor (hatched orange). The number of intersections was analyzed for each point, and two-way ANOVA showed a significant main effect of alcohol at 50, 60, 70, 80, 90, 110, and 120 μm from the soma, but no effect of virus or interaction. e Total dendritic length. Data are presented as the average total dendritic length ± S.E.M and expressed in μm. Two-way ANOVA showed a significant main effect of alcohol (F_(1,47) = 26.35, p < 0.001) but no effect of virus (F_(1,47) = 0.16, p = 0.69) and no interaction (F_(1,47) = 0.18, p = 0.677). f Number of branching points. Data are presented as the average number of branching points ± S.E.M. Two-way ANOVA showed a significant main effect of alcohol (F_(1,47) = 11.98, p = 0.0012) but no effect of virus (F_(1,47) = 2.81, p = 0.1), and no interaction (F_(1,47) = 1.51, p = 0.225). g Number of ending points. Data are presented as the average number of ending points ± S.E.M. Two-way ANOVA showed a significant main effect of alcohol (F_(1,47) = 10.26, p = 0.0024) but no effect of virus (F_(1,47) = 0.81, p = 0.372) and no interaction (F_(1,47) = 1.19, p = 0.28). h Representative x100 confocal z-stack images of dendritic segments bearing spines for the four experimental conditions (water/SCR, water/shRictor, alcohol/SCR, alcohol/shRictor). Scale bar, 5 μm. i Average spine head width. Data are presented as the average of spine head width ± S.E.M and are expressed in μm. Two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 24.43, p < 0.001) and virus (F_(1,44) = 49.16, p < 0.001) but no interaction (F_(1,44) = 1.03, p = 0.316). post hoc Turkey’s test detected a significant difference between water and alcohol within the SCR group (p < 0.001), a significant difference between SCR and shRictor within the water group (p < 0.001) and within the alcohol group (p < 0.001). j Spine area of DMS MSNs. Data are presented as the average spine area ± S.E.M and are expressed in μm². Two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 8.94, p = 0.005) and virus (F_(1,44) = 33.9, p < 0.001), and a significant interaction (F_(1,44) = 9.27, p = 0.004). Post hoc Turkey’s test detected a significant difference between water and alcohol within the SCR group (p < 0.001), a significant difference between SCR and shRictor within the alcohol group (p < 0.001) but no difference between SCR and shRictor within the water group (p = 0.217). k Average length to width ratio of DMS MSNs. Data are presented as the average of spine length to width ratio ± S.E.M. Two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 111.6, p < 0.001) and virus (F_(1,44) = 93.83, p < 0.001), and a significant interaction (F_(1,44) = 14.75, p < 0.001). Post hoc Turkey’s test detected a significant difference between water and alcohol within the SCR group (p < 0.001), a significant difference between SCR and shRictor within the water group (p < 0.001) and within the alcohol group (p < 0.001). l Average spine length of DMS MSNs of mice. Data are presented as the average spine length±S.E.M and are expressed in μm. Two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 41.68, p < 0.001) and virus (F_(1,44) = 13.86, p < 0.001) and a significant interaction (F_(1,44) = 25.73, p < 0.001). Post hoc Turkey’s test detected a significant difference between SCR and shRictor within the water group (p < 0.001) but no difference within the alcohol group (p = 0.776) and no difference between water and alcohol in the SCR group (p = 0.763). m Spine density of DMS MSNs. Data are presented as the average spine density per 10 μm±S.E.M. Two-way ANOVA showed no main effect of alcohol (F_(1,44) = 2.55, p = 0.118) or virus (F_(1,44) = 0.001, p = 0.973) and no interaction (F_(1,44) = 0.13, p = 0.721). n GFP intensity. Data are presented as the average GFP intensity±S.E.M and are expressed as percentage of the water/SCR control. Two-way ANOVA showed no main effect of alcohol (F_(1,44) = 0.0004, p = 0.995) or virus (F_(1,44) = 0.89, p = 0.349) and no interaction (F_(1,44) = 0.23, p = 0.631. o Percentage of filopodia, thin, stubby, and mushroom-type spines of DMS MSNs. Data are presented as the average % of each type of spines across the four conditions±S.E.M. Filopodia-type spines, two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 124.4, p < 0.001) and virus (F_(1,44) = 89, p < 0.001) but no significant interaction (F_(1,44) = 3.27, p = 0.077). Post hoc Turkey’s test detected a significant difference between water and alcohol in the SCR group (p < 0.001), and a significant difference between SCR and shRictor within the water group (p < 0.001), and within the alcohol group (p < 0.001); Thin spines, two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 34, p < 0.001) and virus (F_(1,44) = 25.4, p < 0.001) but no interaction (F_(1,44) = 0.26, p = 0.613). Post hoc Turkey’s test detected a significant difference between water and alcohol within the SCR group (p < 0.001), and a significant difference between SCR and shRictor within the water group (p = 0.013) and the alcohol group (p = 0.0017); Stubby spines, two-way ANOVA showed no significant main effect of alcohol (F_(1,44) = 1.41, p = 0.242) or virus (F_(1,44) = 1.190, p = 0.282) and no interaction (F_(1,44) = 0.94, p = 0.337); Mushroom spines, two-way ANOVA showed a significant main effect of alcohol (F_(1,44) = 178.8, p < 0.001) and virus (F_(1,44) = 130.8, p < 0.001), but no interaction (F_(1,44) = 0.61, p = 0.44). Post hoc Turkey’s test detected a significant difference between water and alcohol within the SCR group (p < 0.001), and a significant difference between SCR and shRictor within the water group (p < 0.001) and within the alcohol group (p < 0.001). n = 12–14 neurons, 5 mice per condition. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 5

Downregulation of mTORC2 in the DMS reduces binge-like alcohol drinking whereas activation of mTORC2 promotes it. a Schematic representation of experimental timeline of experiment Mice were infused bilaterally with ltv-shRictor or ltv-SCR in the DMS. Four weeks later, animals underwent an IA20%-2BC for 6 sessions. Following 2 weeks of alcohol deprivation in which mice had access to water only, they were subjected to an IA0.01% sac for six sessions. Fluid intake was measured 4 h after the beginning of each drinking session. b, c Black circle and bar depicts ltv-shRictor and white circle and bar depicts ltv-SCR. b Alcohol intake (g/kg/4 h) across drinking sessions (left, main effect of virus F_(1,13) = 14.18, p = 0.0024) and averaged (sessions 1 to 6) alcohol intake (right, t₍₁₃₎ = 3.77, p = 0.0024). c Water (ml/kg/4 h) intake on session 1 to 6 (s1–6) in the IA20%-2BC paradigm t₍₁₃₎ = 0.72, p = 0.484. d Schematic representation of timeline of experiment. Bilateral guide cannula were surgically implanted in the DMS. Two weeks later, mice underwent IA20%-2BC for 3 weeks and Vehicle or A-443654 (1 μg/μl) was infused in the DMS 15 min before the beginning of a drinking session. Fluid intake was measured 1 h after the beginning of the drinking session. Following 2 weeks of access to water only, mice were subjected to IA0.01% sac for 2 weeks prior to Vehicle or A-443654 (1 μg/μl) administration e Alcohol (g/kg/1 h) intake following intra-DMS infusion of vehicle (white) or A-443654 (black) t₍₈₎ = 2.38, p = 0.0449. f Water (ml/kg/1 h) intake t₍₈₎ = 1.23, p = 0.254. Data are expressed as mean ± S.E.M. b–c n = 7–8 per treatment, **p < 0.01 vs. ltv-SCR; e–f n = 9 per treatment, *p < 0.05 vs. vehicle