Motor coordination deficits in mice lacking RGS9

Abstract

RGS9-2 is a striatum-enriched protein that negatively modulates dopamine and opioid receptor signaling. We examined the role of RGS9-2 in modulating complex behavior. Genetic deletion of RGS9-2 does not lead to global impairments, but results in selective abnormalities in certain behavioral domains. RGS9 knockout (KO) mice have decreased motor coordination on the accelerating rotarod and deficits in working memory as measured in the delayed-match-to-place version of the water maze. In contrast, RGS9 KO mice exhibit normal locomotor activity, anxiety-like behavior, cue and contextual fear conditioning, startle threshold, and pre-pulse inhibition. These studies are the first to describe a role for RGS9-2 in motor coordination and working memory and implicate RGS9-2 as a potential therapeutic target for motor and cognitive dysfunction.

Figure 1. RGS9 KO Mice Exhibit Deficits in Motor Coordination but Normal Locomotor Activity

(A) RGS9 KO mice have deficits in motor coordination as measured on the accelerating rotarod. ANOVA revealed a main effect of genotype [F_1,20 = 5.64, p = 0.028], a main effect of trial, [F_26,520 = 4.82, p < 0.000001 ] but no interaction between genotype and trial, [F_26,520 = 1.26, p=0.18]. (B) Mean time to fall off rotarod over trails 8-27 in RGS9 KO and WT (p=5.05×10^-12). (C)RGS9 KO mice show similar locomotor activity as WT in a 10 min open field test (p = 0.76).

Figure 2. RGS9 KO Mice Exhibit Impaired Working Memory in the DMP Task of the Water Maze

(A) RGS9 KO mice exhibited slower swim speeds than WT (RGS9 KO vs. WT, p > 0.05). (B) Block 1: RGS9 KO and WT showed no savings over the first 4 days (Days 1-4) of DMP training as measured by distance traveled to reach the submerged platform (RGS9 KO: Trial 1 vs. Trial 2, p = 0.28; WT: Trial 1 vs. Trial 2, p = 0.76). RGS9 KO mice traveled the same as WT to reach the platform during both Trial 1 and Trial 2 (Trial 1, p = 0.44; Trial 2; p = 0.98). (C) Block 2: RGS9 KO and WT did not exhibit savings over the second 4 days (Days 5-8) of DMP training (RGS9 KO: Trial 1 vs. Trial 2, p = 0.89; WT: Trial 1 vs. Trial 2, p = 0.77). RGS9 KO mice traveled the same as WT to reach the platform during both Trial 1 and Trial 2 (Trial 1, p = 0.33; Trial 2, p = 0.13). (D) Block 3: Similarly, RGS9 KO and WT did not exhibit savings over the third 4 days (Days 9-12) of DMP training (RGS9 KO: Trial 1 vs. Trial 2, p = 0.25; WT: Trial 1 vs. Trial 2, p = 0.17). RGS9 KO mice traveled the same as WT to reach the platform during both Trial 1 and Trial 2 (Trial 1, p = 0.23; Trial 2, p = 0.12). (E) Block 4: Unlike RGS9 KO mice, WT showed a savings in the final 4 days of testing in the DMP as measured by distance traveled to reach the platform (RGS9 KO: Trial 1 vs. Trial 2, p = 0.29; WT: Trial 1 vs. Trial 2, p < 0.05). RGS9 KO mice traveled the same distance as WT during Trial 1 (p = 0.40), but traveled more during Trial 2 (p = 0.0060).

Figure 3. RGS9 KO Mice are Impaired in the Visible Platform Task

(A) RGS9 KO mice displayed an abnormal learning curve compared to WT during 12 trials as measured by latency to reach the visible platform in the water maze. ANOVA with repeated measures revealed a main effect of Genotype [F_1,20 = 18.85, p = 0.00032], a main effect of Trial [F_11,220 = 2.27, p = 0.012] but no interaction between Genotype and Trial [F_11,220 = 0.55, p = 0.87]. However, there was no difference between RGS9 KO and WT in duration to the platform on trial 12 (p > 0.05). Legend in A applies to Panels A-C. (B) Swim speed was unaffected by genotype (F_1,11 = 0.82, p = 0.38). (C) Similar to A, RGS9 KO mice displayed an abnormal learning curve compared to WT during 12 trials as measured by distance traveled to reach the visible platform in the water maze. ANOVA with repeated measures revealed a main effect of Genotype [F_1,20 = 14.42, p = 0.0011], a main effect of Trial [F_11,220 = 2.07, p = 0.023] but no interaction between Genotype and Trial [F_11,220 = 0.53, p = 0.88]. However, there was no difference between RGS9 KO and WT in distance traveled during trial 12 (p>0.05).

Figure 4. RGS9 KO Mice Exhibit Normal Anxiety-like Behavior, Pre-Pulse Inhibition and Startle Amplitude

(A) Both context- and cue-dependent fear learning was normal in RGS9 KO mice (context: RGS9 KO vs. WT, p = 0.52, cue: RGS9 KO vs. WT, p = 0.90). Legend in A applies to Panels A-D. (B) Anxiety-like behavior in the RGS9 KO mice does not differ from wt as measured by time spent in the center of an open field (p = 0.372). (C) In a pre-pulse inhibition test, RGS9 KO mice showed normal sensorimotor gating (4db, p = 0.49; 8db, p = 0.12; 16db, p =0.97). (D) Finally, startle amplitude did not differ across genotype (p = 0.65).