Sensory Processing in the Dorsolateral Striatum: The Contribution of Thalamostriatal Pathways

Abstract

The dorsal striatum has two functionally-defined subdivisions: a dorsomedial striatum (DMS) region involved in mediating goal-directed behaviors that require conscious effort, and a dorsolateral striatum (DLS) region involved in the execution of habitual behaviors in a familiar sensory context. Consistent with its presumed role in forming stimulus-response (S-R) associations, neurons in DLS receive massive inputs from sensorimotor cortex and are responsive to both active and passive sensory stimulation. While several studies have established that corticostriatal inputs contribute to the stimulus-induced responses observed in the DLS, there is growing awareness that the thalamus has a significant role in conveying sensory-related information to DLS and other parts of the striatum. The thalamostriatal projections to DLS originate mainly from the caudal intralaminar region, which contains the parafascicular (Pf) nucleus, and from higher-order thalamic nuclei such as the medial part of the posterior (POm) nucleus. Based on recent findings, we hypothesize that the thalamostriatal projections from these two regions exert opposing influences on the expression of behavioral habits. This article reviews the subcortical circuits that regulate the transmission of sensory information through these thalamostriatal projection systems, and describes the evidence that indicates these circuits could be manipulated to ameliorate the symptoms of Parkinson's disease (PD) and related neurological disorders.

Keywords: POm nucleus; Parkinson disease; corticostriatal; intralaminar complex; sensorimotor; superior colliculus; thalamus; zona incerta.

Figure 1

Basal ganglia processing streams involved in the expression of goal-directed and stimulus-controlled behaviors. According to the prevailing view, virtually all sensory information is transmitted to the striatum by a series of thalamocortical and corticostriatal connections. The thalamus also sends direct projections to the striatum, but its functional influence is poorly understood.

Figure 2

Circuit diagram illustrating multiple neuronal pathways for conveying sensory inputs to the striatum. The thick afferent and efferent connections of the POm indicate the most direct route for transmitting somesthetic information to the DLS. Abbreviations: DMS, dorsomedial striatum; DLS, dorsolateral striatum; MI, primary motor cortex; Pf, parafascicular nucleus; POm, posteromedial nucleus; SI, primary somatosensory cortex; SNc, substantia nigra pars compacta; VPM, ventroposteromedial nucleus. Adapted from Watson et al. (2015).

Figure 3

Patterns of striatal innervation revealed by placing an anterograde tracer in Pf (top) or POm (bottom). (A) Photomicrographs of adjacent sections depict cytoarchitecture and biotinylated dextran amine (BDA) deposits in Pf or POm of two different rats. (B) Rostrocaudal series showing spatial distribution of BDA-labeled axonal varicosities in the striatum. Each series based on superimposing reconstructions from three rats that received similar BDA injections in Pf or POm. Numbers indicate distance (mm) from bregma. Scale bars: 500 microns in (A). Abbreviations: fr, fasciculus retroflexus; GP, globus pallidus; VPM, ventroposteromedial nucleus. Adapted from Alloway et al. (2014).

Figure 4

Proportion of labeled neurons in different brain regions after injecting a retrograde tracer into the right DLS of seven rats. More than 70% of the projections originate from the two cortical hemispheres, nearly 20% from the basolateral amygdala (BLA), and 10% from the thalamus. Among thalamic neurons that project to DLS, 40% reside in the medial part of the posterior nucleus (POm). Adapted from Smith et al. (2012).

Figure 5

Neuronal responses recorded simultaneously in primary somatosensory cortex (SI; top) and DLS (bottom) during computer-controlled whisker deflections administered at 2, 5 and 8-Hz. Nearly two dozen whiskers were deflected in tandem in which each deflection consisted of back-and-forth movements in a 50-ms period. Photomicrographs illustrate the recording sites (asterisks) in SI and DLS. Peristimulus time histograms show that SI neurons adapt rapidly to repetitive whisker movements whereas DLS neurons adapt slowly. First deflection in each epoch is classified as 1-Hz because it is separated from previous deflections by at least 1 s. Adapted from Mowery et al. (2011).

Figure 6

Cumulative distributions of response latencies of regular spiking and medium spiny neurons (MSNs) recorded, respectively, in SI and DLS during whisker deflections at different frequencies. At 8-Hz, nonresponsive neurons in SI are depicted by NR. Adapted from Mowery et al. (2011).

Figure 7

Stimulus-induced responses in POm, DLS and SI cortex as a function of behavioral state. In the lightly-anesthetized state, when electrocorticographic (ECoG) activity is 4–6 Hz, neurons in POm, DLS and SI display strong phasic responses to back-and-forth whisker deflections. When ECoG activity is dominated by frequencies below 1 Hz, however, neurons in POm and DLS are unresponsive, but neurons in SI cortex still display rapidly-adapting responses. Responses in the idealized peristimulus time histograms (PSTHs) are based on studies that examined the effects of anesthesia and other behavioral states on somesthetic responses in POm (Trageser et al., ; Masri et al., ; Smith et al., ; Alloway et al., ; Watson et al., 2015), DLS (West, ; Mowery et al., ; Smith et al., ; Alloway et al., 2014) and SI cortex (Chapin and Lin, ; Mowery et al., 2011).

Figure 8

Electrical stimulation of the superior colliculus exerts opposing influences on Pf and POm. (A) Excitatory responses recorded in the Pf and centrolateral nuclei following electrical stimulation (arrows) of the superior colliculus. (B) Histology shows location of bipolar stimulating electrode in superior colliculus. Data reprinted with permission (Grunwerg and Krauthamer, 1992). Scale Bars: 10 ms (A), 1 mm (B). (C) PSTH shows inhibition of POm activity following electrical stimulation (arrow) of superior colliculus. Waveform scales: 1 ms, 100 μV. PSTH: 50 trials, 5-ms binwidths. (D,E) Photomicrographs depicting recording and stimulating sites (arrowheads) in POm and superior colliculus. Adapted from Watson et al. (2015).

Figure 9

Deep brain stimulation (DBS) in the STN can alter ZI activity. (A,B) Coronal sections from human and rat brains illustrate the close proximity of STN and ZI in myelin and Nissl-processed sections, respectively. (C) Schematic of a coronal section through the human brain shows the diencephalon with respect to the basal ganglia. Inset shows the region depicted in panel (D). (D) Schematic showing how multiple electrode contacts combined with current spread may cause stimulation of both STN and ZI in both the primate and rodent brain. Green lines, inhibitory connections; red lines, excitatory connections. Abbreviations: GPe, globus pallidus external; GPi, globus pallidus internal; POm, posteromedial nucleus; Pulv, Pulvinar; Put, Putamen; STN, subthalamic nucleus; Thal, thalamus; ZI, zona incerta.