, 1982, McDonald, 1991 and Pan et al., 2010), but the specificity and relevance of this input stream is not well understood. The amygdala provided considerably stronger synaptic input to the direct pathway (mean 4.3% ± 1.4% of total inputs
for D1R-Cre, versus 0.1% ± 0.1% for D2R-Cre mice, p = 0.02 by one-tailed Wilcoxon rank-sum test, z = 2.16, U = 18. Nonparametric statistical test used because D2R input is floored near zero). Amygdala inputs were manually registered via scaled rotation at 1/2 sampling density (Figure 6), MK-8776 in vitro demonstrating biased input from both basolateral and central nuclei onto direct pathway MSNs. This observation mirrors the biased limbic cortical synaptic input to the direct pathway described in Figure 4. These results suggest that the limbic system, including both limbic cortex and amygdala, may convey affective value information to the striatum, biasing action selection
preferentially through the direct pathway, consistent with a role for the direct pathway in reinforcement (Kravitz et al., 2012 and Stuber et al., 2011). Although we targeted the direct versus indirect pathway independently of striosomal organization, our results regarding preferential innervation of the direct pathway from limbic structures parallels evidence in the striosomal literature; intriguingly, limbic cortices (Gerfen, 1984, Gerfen, 1989 and Jimenez-Castellanos and Graybiel, 1987), and the amygdala (Ragsdale and Graybiel, 1988) are thought to preferentially innervate striosomal compartments, which may themselves be preferentially populated with direct-pathway-like MSNs that project to the SNc, as well as phosphatase inhibitor library to the SNr, GP, and EP (Fujiyama et al., 2011). These results, in conjunction with our own experiments, suggest that both target cell location within the striosome-matrix dichotomy
(Kincaid and Wilson, 1996) and neuronal cell type may interact to generate fine-scale Suplatast tosilate organization within the dorsal striatum. We examined the strength of synaptic dopaminergic input from the substantia nigra onto striatal projection cells. Surprisingly, we observed that a relatively small proportion of total labeled inputs arose from the substantia nigra pars compacta (SNc), but that SNc similarly innervated both direct- and indirect-pathway MSNs. When using the monosynaptic rabies virus system, only 0.8% ± 0.3% of the brain-wide inputs arose from SNc onto either direct- or indirect-pathway MSNs (Figures 3 and 7C). Figure 7A shows a representative image of substantia nigra in a D1R-Cre mouse. As expected, dense striatonigral axon fibers from direct-pathway MSNs are detectable in substantia nigra pars reticulata (SNr), but relatively few retrogradely labeled neurons are visible in pars compacta. In D2R-Cre mice, few if any fibers were detected in SNr, as expected for targeting indirect-pathway MSNs ( Figure 7B). Again, relatively few retrogradely labeled neurons were detected in SNc.