Long-range inhibitory neurons mediate cortical neurovascular coupling

Abstract

To meet the high energy demands of brain function, cerebral blood flow (CBF) parallels changes in neuronal activity by a mechanism known as neurovascular coupling (NVC). However, which neurons play a role in mediating NVC is not well understood. Here, we identify in mice and humans a specific population of cortical GABAergic neurons that co-express neuronal nitric oxide synthase and tachykinin receptor 1 (Tacr1). Through whole-tissue clearing, we demonstrate that Tacr1 neurons extend local and long-range projections across functionally connected cortical areas. We show that whisker stimulation elicited Tacr1 neuron activity in the barrel cortex through feedforward excitatory pathways. Additionally, through optogenetic experiments, we demonstrate that Tacr1 neurons are instrumental in mediating CBF through the relaxation of mural cells in a similar fashion to whisker stimulation. Finally, by electron microscopy, we observe that Tacr1 processes contact astrocytic endfeet. These findings suggest that Tacr1 neurons integrate cortical activity to mediate NVC.

Keywords: CP: Cell biology; CP: Neuroscience; cerebral blood flow; hemodynamics; pericytes; somatostatin; tachykinin.

Figure 1.. Tacr1 neurons are a conserved population of long-range GABAergic neurons

(A, B, D, and E) Representative multiplex fluorescence in situ hybridization (FISH) images demonstrating co-localization of Sst/SST (purple), Tacr1/TACR1 (red), Nos1/NOS1 (green), and Chodl/CHODL (blue) in cortical neurons in C57BL/6 mouse (A) and human (D). Scale bar, 5 μm. Quantification of neuronal density of quadruple-labeled cells in cortex and white matter (wm) in mouse (B, n = 3 mice) and human (E, n = 3 subjects). (C) Venn diagram depicting the intersectionality of these neuronal populations in mouse (see also Figure S1). (F) Cartoon depicting strategy to target Tacr1 neurons by crossing Tacr1^CreER mice to mice harboring a Cre-dependent tdTomato (tdT) fluorescent reporter. (G) Representative sagittal section of Tacr1-mediated tdT expression in the mouse cortex after tamoxifen administration. Dashed lines are approximate anatomical borders. cc, corpus callosum. Scale bar, 200 μm. (H) Quantification of the percentage of Tacr1- or Nos1-expressing neurons that expressed tdT (specificity) and the percentage of tdT⁺ neurons that co-expressed Tacr1 or Nos1 (recombination efficiency; see also Figure S2 and S3; n = 5 mice). (I) Schematic depicting experimental design. AAV-retrograde (rg) was injected into mouse somatosensory cortex. Brains were processed for whole-tissue clearing or IHC. (J) Representative contralateral projections of traced Tacr1 neurons in dorsal (scale bar, 2000 μm) and coronal (scale bar, 1,000 μm) views of cleared brains (n = 6 cells from 2 mice). (K) Quantification of Tacr1 neuron cell bodies in various brain regions following the injection AAVrg-DIO-YFP in Tacr1^CreER mice (n = 4). (L) Representative neuronal reconstructions of tdT-labeled Tacr1 neurons from cleared brains showing long-range axons (red) and dendrites (black) (n = 4 cells). (M) Inset from (L) (dashed square) to illustrate Tacr1 dendrites (see also Figure S4). Data are mean ± SEM, error bars represent SEM, and dots represent data points from individual mice or subjects.

Figure 2.. Tacr1 neurons are recruited for sensory-evoked NVC

(A) Top: schematic of chronic cranial window design and representative macroscopic image of a cranial window in mouse. Scale bar, 1 mm. Bottom: schematic depicting Ca²⁺ imaging by 2P microscopy in the S1 cortex. Sensory stimulation (air puff) to the contralateral whisker pad was performed to evoke a hemodynamic response. (B) Time course of the percentage of change in Ca²⁺ signal (ΔF/F) in mice expressing GCaMP6f in Tacr1 neurons (black) or GCaMP6s in Thy1 (pyramidal) neurons (green) following sensory stimulation (10 trials per mouse). (C) Maximum percentage of change in Ca²⁺ signal in mice expressing GCaMP6f in Tacr1 neurons before and after whisker stimulation (**p = 0.0039, n = 9 neurons in 5 mice). (D) Representative line profiles showing changes in vessel diameter (μm) of a nearby vessel and ΔF/F (%) from a GCaMP6f-expressing Tacr1 neuron. Orange bar represents whisker stimulation (1 s, 50 Hz, 50 ms). Scale bar, 2 s. (E) Left: schematic depicting strategy to express ChR2 in cortical (top) or thalamic (bottom) excitatory neurons. Areas outlined by the black dashed rectangle are shown at the right. Right: representation of ChR2-assisted circuit mapping (sCRACM) showing optogenetic stimulation of ChR2-expressing cortical or thalamic excitatory neurons and simultaneous recording from tdT-expressing Tacr1 neurons in the primary motor cortex (M1). (F) Left: example sCRACM map superimposed on a bright-field image of an M1 brain slice. Scale bar, 0.5 mm. Right: example of averaged excitatory postsynaptic current (EPSC_sCRACM) recorded from the Tacr1 neuron displayed on the grid corresponding to the light stimulus location. (G) Histogram of synaptic strength for cortical (S1, green; n = 13) and thalamic (PO, blue; n = 16) input to Tacr1 neurons in M1. (H–K) Acute optogenetic excitation of cortical ChR2-expressing Tacr1 neurons. (H) Top: IHC showing co-localization of ChR2 (green) and Tacr1 (red) protein. Scale bar, 20 μm. Middle: optogenetic stimulation protocol (1 s, 5 Hz, 30 ms pulse width, blue light). Bottom: experimental setup demonstrating continuous CBF measurement by LDF during optical excitation in awake, head-fixed mice. (I) Representative time course of the change in CBF. (J) Maximum percentage of change in CBF from light OFF to ON in control mice (n = 6, 10 trials per mouse) and mice expressing ChR2 in Tacr1 neurons (n = 5 mice, 10 trials per mouse). Activation of cortical Tacr1 neurons significantly increased CBF relative to light OFF (***p = 0.0004) compared to control mice. (K) Difference in the maximum percentage of change in CBF between light OFF and light ON is increased in mice expressing ChR2 in Tacr1 neurons compared to control mice (**p = 0.0043; n =5 mice, 10 trials per mouse; see also Figure S5). (L–O) Acute optogenetic silencing of cortical ArchT-expressing Tacr1 neurons. (L) Top: IHC showing co-localization of ArchT (green) and Tacr1 (red) protein. Scale bar, 20 μm. Middle: optogenetic silencing protocol (1 s, 5 Hz, 5 ms pulse width; yellow light). Bottom: experimental setup demonstrating continuous CBF recording by LDF during optical inhibition in the presence of sensory stimulation (air puff; 1 s, 10 Hz, 50 psi) in awake, head-fixed mice. (M) Representative time course of the change in CBF. (N) Maximum percentage of change in sensory-evoked CBF from light OFF to ON in mice expressing ArchT in Tacr1 neurons was significantly decreased compared to control mice (n = 4 mice, 10 trials per mouse, *p = 0.0038). Inhibition of cortical Tacr1 neurons significantly decreased sensory-evoked CBF relative to light OFF. (O) Difference in the maximum percentage of change in CBF between light ON and light OFF is significantly reduced in mice expressing ArchT in Tacr1 neurons compared to control mice (N, *p = 0.0143; n = 4 mice, 10 trials per mouse; see also Figure S6). (P) Top: representative macroscopic image of cranial window with chronic cannula (arrow; scale bar, 4 mm) and the position of the LDF probe (arrowhead; scale bar, 1 mm). Middle: sensory stimulation protocol (air puff; 1 s, 5 Hz, 50 psi). Bottom: experimental setup demonstrating continuous CBF recording by LDF during sensory stimulation in awake, head-fixed mice. (Q) Representative time course of the change in CBF following whisker stimulation before (baseline; black) and after (red) cortical infusion of an nNOS inhibitor (1 μL, 1 mM) through the chronic cannula. (R) Difference in the maximum percentage of change in sensory-evoked CBF 30 min after infusing L-NPA, a nNOS inhibitor, was significantly decreased relative to vehicle (1× PBS, 1 μL) (n = 7 mice, 10 trials per mouse, *p = 0.0179). Data are mean ± SEM. Statistical significance was determined by paired, nonparametric two-tailed t test (C); paired, parametric, two-tailed t test (J, N, and R); or unpaired, nonparametric two-tailed t test (K and O). ns, not significant. Error bars and shaded areas are SEM.

Figure 3.. Tacr1-neuron-evoked NVC is mediated by contractile ensheathing pericytes and arteriolar VSMCs

(A) Schematic depicting strategy to express ChR2 in Tacr1 neurons and GCaMP6f in NG2 mural cells (note: with this strategy, GCaMP6f is also expressed in Tacr1 neurons, although it is not relevant in these experiments). (B) White arrows indicate ensheathing pericytes that express GCaMP6f (green) and CD13 (pericyte marker, red). Yellow arrows indicate cells that only express GCaMP6f (green). Scale bar, 10 μm. (C) Top: representative in vivo 2P image (maximum intensity projection) showing GCaMP6f-expressing VSMCs (green) and SR-101-labeled vasculature (red). In merged image, white dashed boxes are shown enlarged at the bottom. Scale bar, 20 μm. Bottom: a VSMC (white arrow) on arteriole (i) and an ensheathing pericyte (white arrowhead) on capillary (ii). Scale bar, 10 μm. Cartoon depicts vascular tree denoting VSMCs surrounding arterioles as well as ensheathing pericytes surrounding capillaries, VSMCs/pericytes, and vascular dynamics upon optogenetic stimulation in Tacr1 neurons and whisker stimulation (n = 7–13 measurements from 5 mice). (D) Top: time course of the change in average VSMC/pericyte Ca²⁺ signal and subsequent vessel diameter upon either optogenetic stimulation (10 trials) of Tacr1 neurons (top) or whisker stimulation (bottom). (E) Maximum percentage of change in pericyte (top) and VSMC (bottom) Ca²⁺ signal (ΔF/F, ****p < 0.0001, 12 pericytes and 13 VSMCs). (F) Baseline diameter and maximum percentage of change in capillary (top) and arteriolar (bottom) diameter (ΔD/D, ****p < 0.0001, 12 capillaries and 12 arterioles). (G and H) Latency to onset of (G) mural cell relaxation or (H) vascular dilation upon either optogenetic stimulation (n = 12 pericytes, 11 VSMCs, 10 capillaries, 10 arterioles) or whisker stimulation (n = 7 pericytes, 13 VSMCs, 11 capillaries, 10 arterioles) (see also Figure S7). (I) Representative image of Tacr1 dendritic spine (purple) forming an asymmetrical synapse with a putative excitatory terminal (T+). White arrowheads point out the edges of the postsynaptic density (p). The Tacr1 distal dendrite receiving a putative excitatory terminal (T+) also contacts the perivascular astrocyte endfoot (ef; yellow), forming a suspected neuronal-astrocytic-vascular tripartite functional unit. Scale bar, 0.5 μm. (J) Tacr1 putative inhibitory axon terminal (T−) contacting astrocytic endfeet. en, endothelial cell; BV-L, blood vessel lumen. Scale bar, 0.5 μm. (K) Schematic of the neurovascular unit including proposed perivascular location of Tacr1 processes (see also Figure S8). Data are mean ± SEM. Statistical significance was determined by paired, parametric two-tailed t test (E and F) or unpaired, two-tailed Mann-Whitney test (G and H). Error bars and shaded areas are SEM.