Journal: bioRxiv

Article Title: External Globus Pallidus Arkypallidal Circuit Dynamics Gate Risk-Taking Behavior

doi: 10.64898/2026.03.20.713182

Figure Lengend Snippet: (a) Experimental diagram. Npas1-Cre-TdTomato mice received bilateral GPe injections of Cre-dependent AAV8-hSyn-DIO-hM4D(Gi)-mCherry or AAV8-hSyn-DIO-hM3D(Gq)-mCherry, with Cre-negative littermates serving as controls. All animals received C21 prior to testing, ensuring equivalent drug exposure across groups. Mice were tested on the elevated plus maze (EPM) 5 weeks post-surgery. Behavioral sessions were video-recorded for subsequent analysis. (b) Time spent in closed arms, center, and open arms during the EPM shows the expected preference for closed arms across all groups, with no effect of GPe NPAS1 manipulation on arm occupancy (two-way mixed ANOVA, maze compartment F (2,52) = 122.8, p = 2.003e-20; group F (2,26) = 1.546, p = 0.2319). (c) Number of open-arm entries does not differ across groups, indicating no effect of GPe NPAS1 manipulation on exploration of these areas (ANOVA, group F (2,26) = 1.048, p = 0.365). (d) Percent time spent in the open arms in the EPM is comparable across control, hM4D(Gi), and hM3D(Gq) mice, consistent with preserved global EPM performance (ANOVA, group F (2,55) = 0.4928, p = 0.6136). (e) Empirical cumulative distribution functions (ECDFs) of frame-wise horizontal movement in the open arms during the EPM show highly overlapping movement distributions across groups, illustrating the absence of gross shifts in locomotor behavior within high-risk regions of the maze. Pairwise Kolmogorov–Smirnov tests detected statistically significant but very small distributional differences (KS D = 0.02–0.05; FDR-corrected p < 0.001), consistent with negligible effect sizes that do not reflect meaningful differences in open-arm movement dynamics. (f) Distance traveled (two-way mixed ANOVA, group x time F (238,3094) = 0.8741, p = 0.9131), (g) speed (two-way mixed ANOVA, group x time F (238,3094) = 0.8742, p = 0.9129), and (h) acceleration (two-way mixed ANOVA, group x time F (238,3094) = 1.037, p = 0.3419) over time are comparable across control, hM4D(Gi), and hM3D(Gq) mice, indicating preserved global locomotor output across the session. Frame-wise polar histograms of heading direction during EPM show (i) control mice exhibit a modest but significant preference for a closed-arm-oriented heading (Rayleigh test, r = 0.005399, p = 0.001871). (j) hM4D(Gi) mice show a statistically significant, strong preferred closed arm heading (Rayleigh test, r = 0.01707, p = 1.225e-16). (k) In contrast, GPe NPAS1 hM4D(Gq) mice do not exhibit a statistically significant preferred heading orientation (Rayleigh test, r = 0.001264, p = 0.7505). (l) Pose features extracted from video tracking show bound box area across time was decreased for hM3D(Gq) mice compared to control mice across all EPM areas (two-way mixed ANOVA, group x time F (238,3094) = 1.180, p = 0.03497; post hoc control v hM3D(Gq) p = 0.03348). (m) Similarly, box aspect ratio over time was decreased for hM3D(Gq) mice compared to control mice (two-way mixed ANOVA, group x time F (238,3094) = 1.181, p = 0.03486; post hoc control v hM3D(Gq) p = 0.04464). (n) There were no group differences in the change in aspect ratio across time (two-way mixed ANOVA, group x time F (238,3094) = 0.08639, p = 0.06231). Dots represent individual data points, error bars or shaded bands represent standard error of the mean (SEM). For polar plots, 32 bins were computed to generate 11.25 degree bars for histogram densities.

Article Snippet: The Gi/o-coupled DREADD vector AAV8-hSyn-DIO-hM4D(Gi)-mCherry (plasmid #44362) and the Gq-coupled DREADD vector AAV8-hSyn-DIO-hM3D(Gq)-mCherry (plasmid #44361) were obtained from Addgene.

Techniques: Control

Journal: bioRxiv

Article Title: External Globus Pallidus Arkypallidal Circuit Dynamics Gate Risk-Taking Behavior

doi: 10.64898/2026.03.20.713182

Figure Lengend Snippet: (a) Experimental diagram. Npas1-Cre-TdTomato mice received bilateral GPe injections of Cre-dependent AAV8-hSyn-DIO-hM4D(Gi)-mCherry or AAV8-hSyn-DIO-hM3D(Gq)-mCherry, with Cre-negative littermates serving as controls. All animals received C21 prior to testing, ensuring equivalent drug exposure across groups. Mice were tested on the elevated plus maze (EPM) 5 weeks post-surgery. Behavioral sessions were video-recorded for subsequent analysis. (b) Time spent in closed arms, center, and open arms during the EPM shows the expected preference for closed arms across all groups, with no effect of GPe NPAS1 manipulation on arm occupancy (two-way mixed ANOVA, maze compartment F (2,52) = 122.8, p = 2.003e-20; group F (2,26) = 1.546, p = 0.2319). (c) Number of open-arm entries does not differ across groups, indicating no effect of GPe NPAS1 manipulation on exploration of these areas (ANOVA, group F (2,26) = 1.048, p = 0.365). (d) Percent time spent in the open arms in the EPM is comparable across control, hM4D(Gi), and hM3D(Gq) mice, consistent with preserved global EPM performance (ANOVA, group F (2,55) = 0.4928, p = 0.6136). (e) Empirical cumulative distribution functions (ECDFs) of frame-wise horizontal movement in the open arms during the EPM show highly overlapping movement distributions across groups, illustrating the absence of gross shifts in locomotor behavior within high-risk regions of the maze. Pairwise Kolmogorov–Smirnov tests detected statistically significant but very small distributional differences (KS D = 0.02–0.05; FDR-corrected p < 0.001), consistent with negligible effect sizes that do not reflect meaningful differences in open-arm movement dynamics. (f) Distance traveled (two-way mixed ANOVA, group x time F (238,3094) = 0.8741, p = 0.9131), (g) speed (two-way mixed ANOVA, group x time F (238,3094) = 0.8742, p = 0.9129), and (h) acceleration (two-way mixed ANOVA, group x time F (238,3094) = 1.037, p = 0.3419) over time are comparable across control, hM4D(Gi), and hM3D(Gq) mice, indicating preserved global locomotor output across the session. Frame-wise polar histograms of heading direction during EPM show (i) control mice exhibit a modest but significant preference for a closed-arm-oriented heading (Rayleigh test, r = 0.005399, p = 0.001871). (j) hM4D(Gi) mice show a statistically significant, strong preferred closed arm heading (Rayleigh test, r = 0.01707, p = 1.225e-16). (k) In contrast, GPe NPAS1 hM4D(Gq) mice do not exhibit a statistically significant preferred heading orientation (Rayleigh test, r = 0.001264, p = 0.7505). (l) Pose features extracted from video tracking show bound box area across time was decreased for hM3D(Gq) mice compared to control mice across all EPM areas (two-way mixed ANOVA, group x time F (238,3094) = 1.180, p = 0.03497; post hoc control v hM3D(Gq) p = 0.03348). (m) Similarly, box aspect ratio over time was decreased for hM3D(Gq) mice compared to control mice (two-way mixed ANOVA, group x time F (238,3094) = 1.181, p = 0.03486; post hoc control v hM3D(Gq) p = 0.04464). (n) There were no group differences in the change in aspect ratio across time (two-way mixed ANOVA, group x time F (238,3094) = 0.08639, p = 0.06231). Dots represent individual data points, error bars or shaded bands represent standard error of the mean (SEM). For polar plots, 32 bins were computed to generate 11.25 degree bars for histogram densities.

Article Snippet: The Gi/o-coupled DREADD vector AAV8-hSyn-DIO-hM4D(Gi)-mCherry (plasmid #44362) and the Gq-coupled DREADD vector AAV8-hSyn-DIO-hM3D(Gq)-mCherry (plasmid #44361) were obtained from Addgene.

Techniques: Control