Distinctive physiology of molecularly identified medium spiny neurons in the macaque putamen

Abstract

The distinctive physiology of striatal medium spiny neurons (MSNs) underlies their ability to integrate sensory and motor input. In rodents, MSNs have a hyperpolarized resting potential and low input resistance. When activated, they have a delayed onset of spiking and regular spike rate. Here, we show that in the macaque putamen, latency to spike is reduced and spike rate adaptation is increased relative to mouse. We use whole-cell brain slice recordings and recover single-cell gene expression using Patch-seq to distinguish macaque MSN cell types. Species differences in the expression of ion channel genes including the calcium-activated chloride channel, ANO2, and an auxiliary subunit of the A-type potassium channel, DPP10, are correlated with species differences in spike rate adaptation and latency to the first spike, respectively. These surprising divergences in physiology better define the strengths and limitations of mouse models for understanding neuronal and circuit function in the primate basal ganglia.

Keywords: CP: Cell biology; CP: Neuroscience; Patch-Seq; cross-species divergence; latency to spike; macaque; medium spiny neurons; mouse; physiology; putamen; single-cell gene expression; spike rate adaptation; spike rates; striatum.

Figure 1.. Whole-cell physiology recordings from macaque putamen MSNs

(A) Exemplar biocytin fills from mouse and macaque medium spiny neurons (MSNs) in acutely prepared brain slices. Scale bar: 100 μm (applies to both images). (B) Resting membrane potential is equivalent between mouse and macaque MSNs (n = 81 mouse; n = 47 macaque neurons; Student’s t test, p > 0.89). (C and D) Exemplar spiking and frequency-current (F-I) curves reveal equivalent excitability between mouse and macaque MSNs (n = 81 mouse; n = 47 macaque neurons; two-way ANOVA, p > 0.85). (E) Exemplar subthreshold current injection steps to examine input resistance. (F and G) Macaque MSNs have modestly higher input resistance than mouse at hyperpolarized membrane potentials (n = 72 mouse neurons; n = 39 macaque neurons; Student’s t test, p < 10⁻³ at −80 mV, p > 0.15 at −60 mV). Center point shows mean, and error bars are the standard error of the mean.

Figure 2.. Firing rate adaptation in macaque but not mouse MSNs

(A) Exemplar traces for measurement of firing rate adaptation in mouse and macaque MSNs. (B and C) Spike rate adaptation is greater in macaque than in mouse. Adaptation is calculated as ratio of 9^th ISI to first ISI using the first trace with at least 10 spikes (n = 81 mouse neurons; n = 47 macaque neurons; Student’s T test, p < 10⁻¹⁰). (D and E) Inter-spike interval monotonically increases throughout the spike train (n = 81 mouse; n = 47 macaque neurons). (F) Degree of adaptation decreases over the spike train. (G) Greater spike rate adaptation in macaque is associated with elevated expression of ANO2-type calcium-activated chloride channels (t test, p < 10⁻⁴). (H) SK-type calcium-activated potassium channels are elevated in macaque relative to mouse, particularly KCNN1 (n = 27 mouse; n = 46 macaque neurons for Patch-Seq analysis; ANOVA p < 10⁻⁵). Center point shows mean, error bars and boxplots are the standard error of the mean.

Figure 3.. Expression of subthreshold potassium channel subunits is correlated with reduced latency to the first spike in macaque MSNs

(A) Exemplar traces reveal latency to first spike is reduced in macaque relative to mouse MSNs. (B) Mouse neurons exhibit a slow, progressive depolarization (V_memb slope) preceding the first spike that is absent in macaque. Slope is calculated from the voltage response to the largest current injection in a family of current steps (50 pA intervals) that does not contain an action potential (p < 10⁻⁷). (C) First spike latency is reduced in macaque relative to mouse. Latency is calculated from the onset of current injection in the first trace with at least 10 spikes (p < 0.001). (D) Spike threshold is reduced in macaque relative to mouse (p < 10⁻⁴). (E) Species differences in A-type potassium currents do not readily explain progressive depolarization or latency to spike in physiology. (F) Elevated expression of regulatory subunits, especially KCNIP1 and DPP10, is consistent with rapid inactivation of A-type potassium currents in macaque, underlying reduced latency to first spike. (G) Exemplar action potential waveforms from mouse and macaque MSNs. (H) Spike peak is reduced in macaque relative to mouse (p < 10⁻³). (I) Spike width is greater in macaque (p < 10⁻⁷). (J) Spike upstroke unchanged across species (p > 0.9). (K) Spike downstroke is faster in mouse than macaque (p < 10⁻²¹) (for physiology: n = 81 mouse; n = 47 macaque neurons; Student’s t test for all tests; for Patch-seq analysis: n = 27 mouse neurons; n = 46 macaque neurons; ANOVA; KCNIP1 p < 10⁻⁵, DPP10 p < 10⁻²). Boxplots show the mean and standard error.

Figure 4.. Select spike properties vary by transcriptomically defined cell types in macaque putamen

(A) Uniform manifold approximation and projection (UMAP) based on Patch-seq single-cell RNA sequencing (scRNA-seq) data reveals separation of direct and indirect pathway neurons as well as co-clustering of mouse and macaque neurons across species. (B) Heatmap of marker gene expression per cell; genes are grouped by markers for all MSNs, indirect pathway MSNs (iMSNs), and direct pathway MSNs (dMSNs). (C) Aggregated data describing marker gene expression across MSNs. (D and E) Excitability of iMSNs is greater than dMSNs in mouse (D) and macaque (E) (mouse dMSNs versus iMSNs, two-way ANOVA p < 10⁻²⁴; macaque dMSNs versus iMSNs; two-way ANOVA p < 10⁻¹¹). (F) First spike latency and firing rate adaptation are different across dMSNs and iMSNs in mouse (one-way ANOVA p < 10⁻⁸ across groups; Student’s t test p < 0.01) but not macaque (Student’s t test p > 0.4). (G) Spike rate adaptation is different across direct and indirect pathways in mouse (one-way ANOVA p < 10⁻¹³ across groups; Student’s t test p < 0.01) but not macaque (Student’s t test p > 0.1) (for Patch-seq comparisons: n = 21 mouse dMSNs; n = 6 mouse iMSNs; n = 22 macaque dMSNs; n = 24 macaque iMSNs; for physiology comparisons: n = 30 mouse dMSNs; n = 33 mouse iMSNs; n = 22 macaque MSNs; n = 19 macaque iMSNs). Center point shows mean, and error bars and boxplots are the standard error of the mean.