To verify that the small differences in injection site were not responsible for the observed differences in cortical input, we examined the degree of correlation between the anterior-posterior position of the center of the striatal injection site and the anterior-posterior center of gravity of cortical input across all cell types (n = 19). We determined that injection site location predicted less than 5% of the variance in cortical input location (Figure S2). As expected, the cortical center

of gravity for D1R-Cre mice fell below the best-fit line for 7 of 9 animals, whereas cortical center of gravity for D2R-Cre mice fell above the best-fit line for Caspase activation 7 of 10 animals. These observations indicate that cell type identity is much more likely to be the major contributor to cortical input specificity. It is known that two morphologically distinct types of corticostriatal pyramidal cells exist, which have been proposed to differentially innervate striatal projection neuron subtypes (Lei et al., 2004 and Reiner et al., 2003). Intratelencephalic-type (IT-type) pyramidal neurons project to both ipsilateral and contralateral striatum, whereas another type of corticostriatal neuron only projects to ipsilateral striatum but also sends projections along the pyramidal tract (PT-type). There is some evidence to suggest that

these two cell types may preferentially reside in different cortical layers in rats (Lei AZD2281 concentration et al., 2004 and Reiner et al., 2003), although there are also studies in both mice and rats suggesting that PT and IT neurons largely Vasopressin Receptor inhabit the same

cortical layers (McGeorge and Faull, 1987 and Sohur et al., 2012). To determine whether different cortical layers preferentially targeted the direct or indirect pathway, we documented the levels of layer 2/3, superficial layer 5, and deep layer 5 monosynaptic inputs onto either D1R or D2R-expressing MSNs. When examined across the four cortical regions that provided the greatest input to dorsal striatum (Figure 4I), direct- and indirect-pathway MSNs received similar proportional levels of input from each cortical layer (S1, primary somatosensory cortex; M1, primary motor cortex; M2, secondary motor cortex; PFC, insular and orbitofrontal cortices; p > 0.15 for all individual cortical region/layer D1R versus D2R comparisons by two-tailed t test). Furthermore, there was no significant difference in terms of overall cortical input strength from any specific input layer. For layer 2/3, inputs were 19.3% ± 2.5% versus 23.3% ± 2.1% of overall cortical inputs from D1R versus D2R, mean ± 1 SEM, p = 0.2 by two-tailed t test. For superficial layer 5, including all layer 5 input from prefrontal regions, inputs were 56.7% ± 2.6% versus 55.7% ± 2.7%, p = 0.8.