Figure 1. Local neural activity controls basal and evoked arteriole diameter.

(A) Top, schematic of imaging window. Bottom, photo of mouse on spherical treadmill. (B) Top, photo of the pial vasculature of the somatosensory cortex through the PoRTs window. Cytochrome oxidase staining localized the forelimb/hindlimb (FL/HL, purple) and vibrissae cortex (green). Bottom, a maximum projection of two-photon images of vasculature within the yellow box in the top image. Scale bar 50 μm. (C) Top and middle, two-photon images (from white box in B, bottom) of the same arteriole 7 and 9 days after window implantation, both from stationary periods. Bottom, arteriole 9 days after implantation during locomotion. Scale bar 50 μm. (D) Locomotion induces rapid dilation in pial arterioles. Diameter (from region marked by white arrow in C) plotted versus time for a single locomotion event. Blue shading denotes period of locomotion. (E) Population locomotion-triggered averages following aCSF and muscimol infusions for arterioles ≤ 25 µm in basal diameter. For both cases, the individual diameters are normalized by the average basal diameter of the vessel after vehicle infusion. Note the rapid rise to peak in the control, and the lack of dilation in muscimol-infused mice, showing the locomotion-triggered response is under local neural control. Shading represents mean ± standard error. (F) LFP power spectra during stationary periods (basal) and locomotion after muscimol infusion, normalized to vehicle infusion in the same animal. Shading represents mean ± standard error. (G) Representative images of a pial arteriole during periods of no locomotion after vehicle infusion (top) and after muscimol infusion (bottom). Scale bar 50 μm. (H) Basal arteriole diameter following vehicle infusion plotted versus basal diameter after muscimol infusion. Each point is a single vessel, and the mouse identity is represented by the color. The black line shows the linear regression of aCSF vs. muscimol diameter. The point outlined in green is representative image from G. (I) Representative space-time image of linescans of the same capillary after aCSF or muscimol infusion. (J) Basal red blood cell (RBC) velocity plotted after vehicle infusion (x-axis) vs. after muscimol infusion (y-axis). The point outlined in green is the vessel shown in I.

Figure 1—figure supplement 1. Example neural responses in forelimb/hindlimb representation of somatosensory cortex during voluntary locomotion.

(A) Schematic showing experimental setup of electrophysiological measurements taken with multi-electrode array. (B) Average locomotion-evoked multi-unit spiking from a single site in the FL/HL representation in S1. (C) Average locomotion-evoked gamma-band power change from a single site in the FL/HL representation in S1. (D) Example of multi-unit activity recorded in layer five during locomotion. Top shows velocity of treadmill. (E) Locomotion-triggered spike rasters for several locomotion events.

Figure 1—figure supplement 2. Diameter changes in penetrating arterioles.

All data is taken from penetrating arterioles imaged ~100 µm below the pial surface. Statistical comparisons of locomotion triggered averages are of the control and manipulation are done with both normalized to the pre-locomotion baseline in the control condition. (A) Example of locomotion-induced dilation in a penetrating arteriole. Inset shows the vessel during a stationary period and during locomotion. Blue shading denotes periods of locomotion. Data in B and C are from penetrating arteries in mice infused with muscimol and aCSF. (B) Basal diameter of penetrating arteries after vehicle infusion (x-axis) versus muscimol infusion (y-axis). There was a significant decrease in basal diameter in the muscimol infused condition (−7.8 ± 1.4%, LME p<0.01 n = 5 mice, 13 vessels). (C) Population locomotion-triggered averages after vehicle (black) and muscimol (red) infusions (LME p<0.35, n = 5 mice, 13 arterioles in FL/HL representation). Data in D and E are from penetrating arteries in mice infected with AAV-CMV-TurboRFP-WPRE-rBG (reporter virus), and IP injected with CNO or vehicle control. (D) Basal arteriole diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) was not significantly different (−0.3 ± 14.2%, LME p<0.24 n = 5 mice, 13 vessels). (E) Population locomotion-triggered averages for arterioles ≤ 25 µm in basal diameter after vehicle (black) or CNO (red) injection (LME p<1, n = 9 mice, 13 arterioles in FL/HL representation). Data in F and G are from penetrating arteries in mice infected with AAV-hSYN-HA-hM4D(Gi)-mCherry, data in H and I is from penetrating arteries in mice infected with AAV-hSYN-HA-hM3D(Gq)-mCherry (pan-neuronal expression). (F) Basal diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) was significantly decreased by CNO (−6.7 ± 12.0%, LME p<0.03, n = 7 mice, 21 vessels). (G) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<1, n = 7 mice, 21 arterioles in FL/HL representation). (H) Basal diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis). There was no significant difference between the conditions (+2.3 ± 13.9%, LME p<0.92 n = 7 mice, 19 vessels). (I) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<1, n = 7 mice, 19 arterioles in FL/HL representation). Data in J and K are from penetrating arteries in mice infected with AAV-CaMKIIa-hM4D(Gi)-mCherry, data in L and M is from penetrating arteries in mice infected with AAV-CaMKIIa-hM4D(Gq)-mCherry (expression in pyramidal neurons). (J) Basal arteriole diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) showed no significant difference (−4.1 ± 13.5%, LME p<0.06 n = 7 mice, 20 vessels). (K) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<1, n = 7 mice, 20 arterioles in FL/HL representation). (L) Basal arteriole diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) showing significant increases (+8.5 ± 14.6%, LME p<0.04 n = 7 mice, 18 vessels). (M) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<0.83, n = 7 mice, 18 arterioles in FL/HL representation). Data in N and O are from penetrating arteries in nNOS-cre mice injected with AAV-hSyn-DIO-hM4D(Gi)-mCherry, data in P and Q is from penetrating arteries in nNOS-cre mice injected with AAV-hSyn-DIO-hM4D(Gq)-mCherry (nNOS+ expression). (N) Basal diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) showing no significant difference (−7.5 ± 12.6%, LME p<0.24 n = 9 mice, 23 vessels). (O) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<1, n = 9 mice, 23 arterioles in FL/HL representation). (P) Basal diameter of penetrating arteries after vehicle injection (x-axis) versus CNO injection (y-axis) showing no significant difference (+7.3 ± 16.4%, LME p<0.12 n = 6 mice, 16 vessels). (Q) Population locomotion-triggered averages after vehicle (black) and CNO (red) injection (LME p<1, n = 6 mice, 16 arterioles in FL/HL representation). Data in R and S are from penetrating arteries in mice after L-NAME and aCSF infusions. (R) Basal diameter of penetrating arteries after vehicle infusion (x-axis) versus L-NAME infusion (y-axis). There was no significant change in arterial diameter (−14.8 ± 10.9%, LME p<0.08 n = 6 mice, 14 vessels). (S) Population locomotion-triggered averages after vehicle (black) and L-NAME (red) infusion (LME p<0.07, n = 6 mice, 14 arterioles in FL/HL representation). Data in T and U are from penetrating arteries in mice after CP-99994 and aCSF infusions. Data in V and W is from penetrating arteries in mice after Substance P and aCSF infusions. (T) Basal arteriole diameter of penetrating arteries after vehicle infusion (x-axis) versus CP-99994 infusion (y-axis) showing no significant difference (5.4 ± 23.6%, LME p<0.53 n = 6 mice, 11 vessels). (U) Population locomotion-triggered averages after vehicle (black) and CP-99994 (red) infusion (LME p<1, n = 6 mice, 11 arterioles in FL/HL representation). (V) Basal arteriole diameter of penetrating arteries after vehicle infusion (x-axis) versus Substance P infusion (y-axis) showing diameter was significantly increased by Substance P (23.3 ± 22.3%, LME p<1.6×10⁻³ n = 6 mice, 13 vessels). (W) Population locomotion-triggered averages after vehicle (black) and Substance P (red) infusion (LME p<0.05, n = 6 mice, 13 arterioles in FL/HL representation).

Figure 1—figure supplement 3. Muscimol infusion did not significantly affect capillary diameter.

(A) Representative images of the same capillary after vehicle infusion (left) and after muscimol infusion (right). Scale bar 5 μm. (B) Capillary diameter measured after vehicle infusion (x-axis) vs. after muscimol infusion (y-axis) showed no significant changes (0.03 ± 0.85 µm LME p<0.94, n = 6 mice, 30 capillaries). The point outlined in green is representative image from A.

Figure 1—figure supplement 4. Muscimol infusion did not significantly affect basal arteriole diameter variance.

(A) Scatter plot of basal arteriole diameter variance during stationary periods following vehicle infusion (x-axis) versus muscimol infusion (y-axis). There was no significant change in arterial diameter variance. The variance measurements are from the same vessels in Figure 1H. (B) Locomotion induces rapid dilation in pial arterioles that is attenuated by muscimol infusion. Arteriole diameter measured after vehicle infusion (black) and after muscimol infusion (red). The tick marks denote periods of locomotion.

Figure 1—figure supplement 5. Effects of chemogenetic and pharmacological manipulation on locomotion behavior.

Plot showing the amount of time locomoting during 2PLSM imaging for each manipulation normalized by the amount of time locomoting with the vehicle. No manipulation resulted in a significant change in the behavior (LME, all p-values were Bonferroni corrected by the number of different viruses used or by the number of different drugs infused: Reporter Virus p<1, n = 12 mice; muscimol p<1, n = 6 mice; hSyn-G(i) DREADDs p<1, n = 7 mice; hSyn-G(q) DREADDs p<0.3, n = 7 mice; CaMKIIa-G(i) DREADDs p<1, n = 7 mice; CaMKIIa-G(q) DREADDs p<1, n = 6 mice; nNOS-G(i) DREADDs p<0.2, n = 9 mice; nNOS-G(q) DREADDs p<1, n = 6 mice; L-NAME p<1, n = 6 mice; CP-99994 p<1 n = 7 mice; Substance P p<1, n = 7 mice).