A prefrontal cortex-hypothalamus circuit for heart rate control in non-human primates

Figures

• 

  Figure 1:. Identification of a prefrontal–hypothalamic pathway for rapid influence over cardiac dynamics. a, Schematic illustration of the putative primate PFC-hypothalamus system that we study in the context of heart rate modulation. b, Stimulation mapping of hypothalamus. Left: T1-weighted MRI showing stimulation sites near LHA (cc, corpus callosum; lv, lateral ventricle; 3v, third ventricle; PVN, paraventricular nucleus; VMH, ventromedial hypothalamus; LHA, lateral hypothalamus). Right: Representative photoplethysmograph (PPG) traces as a function on depth in LHA. Yellow bar indicates electrical stimulation for 500 ms (-300 μA). Asterisks indicate significant decrease in the heart rate averaged 0–1.5 s after stimulation onset. n = 7, 5, 5, 5,10 stimulations as a function of depth (dorsal to ventral); P = 0.3125, 0.4375, 0.3125, 0.0625, 0.0371 (two-tailed signed-rank test against a hypothetical median of 0). c, Reconstruction of recording sites in LHA. Left: Schematic illustration of monkey brain in the coronal plane at interaural A15. Circles indicate stimulation sites. Red indicates significant heart rate decrease, blue indicates significant heart rate increase, and un-filled indicates…

• 

  Figure 2:. a12 neuronal activity tracks spontaneous fluctuations in heart rate. a, An example a12 neuron during resting state session with spiking activity (red, smoothed with Gaussian kernel σ = 2 s) correlated with heart rate (black, same). Bottom left: detailed view of correlated changes in spike and heartbeat trains (red and black; dots are mean rates in 20 s bins, error bars represent s.e.m., too small to see). Bottom right: spike rate vs. heart rate correlation (quantified using 3 s time bins; text indicates rank correlation). b, a12 neurons reflect heart rate rather than individual heartbeats. Left: Power spectral density of cross-correlation between spike train and heartbeat train for the example neuron (red) compared to n=100 randomly time-shifted controls representing the null hypothesis that spikes are unrelated to heartbeats (dark and light gray are central 68% and 95% of the control distribution). * indicates the mean power spectral density in a frequency band is significantly different between real vs. controls (p < 0.05, two-tailed circular shift tests). Right: % of a12 neurons with significantly different power spectral density of spike-heart cross-correlation in each…

• 

  Figure 3:. Dissecting the role of a12→LHA to modulate cardiac output. a, Left: Representative PPG traces as a function of current in the effective sites of a12 (red) and LHA (green). Conventions are the same as in Figure 1. Asterisks indicate significant decrease in the heart rate for each site averaged 0–1.5s after stimulation onset. For a12, n = 72, 46, 22, 48 stimulations for -300, -200, -100, -50 μA, respectively; P = 7.11e-13, P = 4.9e-09, P = 4.01e-05, P = 0.112. For LHA, n = 232, 29, 30, 30 stimulations; P = 4e-36, 3.17e-06, 2.13e-06, 1.8e-05 (two tailed signed-rank test against a hypothetical median of 0). Right: Average heart rate responses from a12 (red) and LHA (green) as a function of current. Data are mean ± s.e.m. P = 0.449, 0.443, 0.0507, 0.00176 for -300, -200, -100, -50 μA (two-tailed rank-sum test; a12 vs. LHA at each current). b, Cumulative distribution function plot of electrical stimulation-evoked heart rate response latencies in a12 (red) and LHA (green). n = 72, 232 for a12 and LHA stimulations, respectively. Asterisks indicates p-value when comparing all stimulations from a12 and LHA stimulation, while Pfastest 50 indicates p-value when comparing stimulation response…

• 

  Figure 4:. Sonogenetic perturbation of a12 modulates heart function. a, Schematic illustration of the sonogenetic experimental procedure performed in non-human primates, which first involves an injection of AAV encoding TRPV1 into a12 followed by focused ultrasound (FUS) stimulating the same brain region. b, Representative fluorescent in situ hybridization images in a12 visualizing DAPI (blue), Vglut2 (yellow), and Trpv1 (magenta) expression in both the injected and non-injected hemispheres. c, Top: representative MR thermometry map showing localized heating in a12. Bottom: Temperature change time course computed with an ROI within a12. The yellow bar indicates FUS stimulation period. Shaded region represents s.e.m. d, Top: Representative PPG traces shown for a12 TRPV1+ (red), a12 TRPV1- (blue), and off-target (dorsal prefrontal cortex, grey) groups. Bottom: Time course of the average heart rate change computed in 10 s time bins for the same three groups. Yellow bar indicates FUS stimulation period. Data are mean ± s.e.m. * denotes comparison between TRPV1+ and TRPV1- groups. # denotes comparison between TRPV1+ and off-target groups. n = 87, 55, 135 stimulations for TRPV1+, TRPV1-, and…

• 

  Figure 5:. Sonogenetic and chemogenetic perturbation of a12 bidirectionally modulate motivated behavior. a, Decision task motivated by reward magnitude, delay, and uncertainty. Animals chose between offers which provided different reward magnitudes, delay durations, and orders of reward and delay (reward-first or delay-first). Both rewards and delays could be either certain or uncertain. b, Sequence of events after choosing reward-first vs. delay-first offers. c, Animals were highly motivated to perform the task, shown by frequently choosing offers that were reward-first, high E[R], low E[T], and low Unc[R]. *** indicates p < 0.001 (two-tailed signed-rank test). d, In different sessions, animals performed the task during sonogenetic on-target stimulation of a12 (red), off-target stimulation of a13 (gray), or DCZ to silence a12 (blue). e, On-target stimulation of a12 shifts choice toward uncertain immediate rewards. Probability of choosing an offer as a function of reward magnitude (x-axis: small, uncertain, medium, or large) and attribute order (reward-first, left; delay-first, right), separately for stimulation trials (red) and control trials (black). ** indicates significant difference (p…