The entopeduncular nucleus (EPN) and substantia nigra pars reticulata (SNr) constitute the output nuclei of the basal ganglia, but studies on the EPN are limited compared with those on the SNr. Both nuclei receive projections from the striatum with axons containing substance P (SP) and cannabinoid type-1 receptor (CB1R), and immunoreactivities for these substances show complementary patterns in the striatum and SNr. In this study we revealed a similar complementarity in the mouse EPN, combined it with region-specific neuronal distributions, and defined subregions of the EPN. First, the EPN was divided into two areas, one showing low SP and high CB1R (lSP/hCB1R) immunoreactivities and the other showing opposite immunoreactivities (hSP/lCB1R). The former received inputs from the dorsolateral striatum that is innervated by sensorimotor cortices, whereas the latter received inputs from the medial striatum innervated by limbic/association cortices. Then, the lSP/hCB1R area was further divided into the dorsolateral subregion in the rostral EPN and the core subregion in the caudal EPN, the latter characterized by the concentration of parvalbumin-positive neurons targeting the VA-VL thalamic nucleus. The hSP/lCB1R area was divided into the ventromedial subregion in the rostral EPN and the shell subregion in the caudal EPN, the former characterized by the concentration of nitric oxide synthase-positive neurons targeting the lateral habenula (LHb). Somatostatin-positive neurons targeting the LHb were located diffusely in three subregions other than the core. These findings illuminate structural organization inside the basal ganglia, suggesting mechanisms for sorting diverse information through parallel loops with differing synaptic modulation by CB1R. <b>Significance statement</b> The entopeduncular nucleus (EPN) in rodents corresponds to the internal segment of the globus pallidus in primates and is one of the major output nuclei of the basal ganglia. Because studies on the EPN are limited, we investigated the morphological features of the mouse EPN, focusing on the complementary immunoreactivities for substance P (SP) and cannabinoid type-1 receptor (CB1R). The EPN was first divided into two parts, one showing low SP and high CB1R immunoreactivities and the other showing opposite immunoreactivities. Each part was further divided into two subregions based on the distributions of different neuronal types. The present findings clarify the structural organization inside the EPN, suggesting mechanisms for processing diverse information with differing synaptic modulation by CB1R.

Recommended diagnostic criteria for paraneoplastic neurological syndromes