Nitric oxide is not responsible for initial sensory-induced neurovascular coupling response in the barrel cortex of lightly anesthetized mice

Abstract

Significance: Neurovascular coupling matches changes in neural activity to localized changes in cerebral blood flow. Although much is known about the role of excitatory neurons in neurovascular coupling, that of inhibitory interneurons is unresolved. Although neuronal nitric oxide synthase (nNOS)-expressing interneurons are capable of eliciting vasodilation, the role of nitric oxide in neurovascular coupling is debated.

Aim: We investigated the role of nitric oxide in hemodynamic responses evoked by nNOS-expressing interneurons and whisker stimulation in mouse sensory cortex.

Approach: In lightly anesthetized mice expressing channelrhodopsin-2 in nNOS-interneurons, 2D optical imaging spectroscopy was applied to measure stimulation-evoked cortical hemodynamic responses. To investigate the underlying vasodilatory pathways involved, the effects of pharmacological inhibitors of NOS and 20-HETE were assessed.

Results: Hemodynamic responses evoked by nNOS-expressing interneurons were altered in the presence of the NOS inhibitor LNAME, revealing an initial 20-HETE-dependent vasoconstriction. By contrast, the initial sensory-evoked hemodynamic response was largely unchanged.

Conclusions: Our results challenge the involvement of nNOS-expressing interneurons and nitric oxide in the initiation of functional hyperemia, suggesting that nitric oxide may be involved in the recovery, rather than initiation, of sensory-induced hemodynamic responses.

Keywords: inhibitory interneuron; neuronal nitric oxide synthase interneuron; neurovascular coupling; nitric oxide; optogenetics; vasomotion.

Fig. 1

NOS-dependence of hemodynamic responses evoked by 2s nNOS-IN activation or whisker stimulation. Data from representative mouse showing hemodynamic response to 2s optogenetic stimulation of nNOS-INs (a)–(c) or 2 s whisker stimulation (d)–(e). (a), (d) Trial-averaged stimulation-evoked changes in Hbt, Hbo, and Hbr, compared with the baseline, pre-LNAME (upper) and post-LNAME (lower) injection. Color bars represent fractional change. (b): Thinned cranial window (upper) with cortical surface vasculature visible (imaged at 575 nm illumination), optic fiber and electrode can be seen. White ROI indicates whisker barrel cortex and purple ROI indicates arteriole region, from which timeseries data are extracted. Scalebar represents 1 mm. Diagram (lower) indicating surface arteries (red) and veins (blue) visible through thinned skull window. (c), (e) Evolution of trial-averaged changes in Hbt. Stimulation of nNOS-INs (c) or whiskers (e) occurs at 0 to 2s. Color bar represents fractional change. Time points relative to stimulation onset (0 s) are indicated above the images. (f) Group data ($n=12$ mice). Mean fractional change in Hbt, Hbo, and Hbr in arteriolar ROI before (left) and after (right) LNAME. Blue shading indicates photostimulation period (upper), grey shading indicates whisker stimulation period (lower). Data: mean ± SEM (g) Hbt time series from example mouse illustrating three metrics used for analysis: initial minima (h), peak (i), peak-to-peak amplitude (change between peak and initial minima), (j). (h)–(j) Fractional change in Hbt, Hbo, and Hbr in response to optogenetic and whisker stimulation with (+) and without (−) LNAME. Darker lines represent group mean ± SD, lighter lines indicate trial-averaged mean for individual animals. $n=12$ mice. *$p<0.025$, **$p<0.005$, ***$p<0.0005$. (h) Initial fractional change (“initial minima”). (i) Peak fractional change. (j) Maximum stimulation-evoked net change (minima to maxima).

Fig. 2

Development of effect of LNAME. Mean fractional change in Hbt, Hbo, and Hbr in arteriolar ROI in response to 2 s optogenetic stimulation of nNOS-INs (top row) or 2 s whisker stimulation (lower row) at different time points relative to injection of LNAME. Column 1: experiment immediately prior to LNAME injection; Column 2: experiment immediately following LNAME injection; Column 3: experiment commencing 60 to 70 min post LNAME injection; Column 4: experiment commencing 95 to 135 min post LNAME injection. Blue shading indicates photostimulation period (top row), grey shading indicates whisker stimulation period (lower row). Data are mean ± SEM, $n$ indicates number of mice. Note: the following data are also included in Fig. 1: pre-LNAME; post-LNAME (95 to 135 min $[n=9]$ and 60 to 70 min $[n=3]$).

Fig. 3

Stimulation-evoked multi-unit activity (MUA) is unchanged in presence of LNAME. Neural responses to (a) 2 s optogenetic stimulation of nNOS-INs or (b) 2 s whisker stimulation (stimulations start at 0 s). Top row: mean change in MUA compared with baseline throughout cortical depth (as indicated by electrode channel number). Color bar represents fractional change. Left: pre-LNAME injection. Right: post-LNAME injection. Bottom row: mean time series of response taken from channels 3 to 6 of electrode (mean ± SEM). Data were collected concurrently with subset of hemodynamic responses displayed in Fig. 1. (c) Peak and mean MUA during 2 s optogenetic or whisker stimulation with (+) and without (−) LNAME. Darker lines represent group mean ± SD, lighter lines indicate trial-averaged means for individual animals. $n=4$ mice for all panels.

Fig. 4

Time-matched experiments with no inhibitor applied. Mean fractional change in Hbt, Hbo, and Hbr in arteriolar ROI in response to 2 s optogenetic activation of nNOS-INs (top row) or 2 s whisker stimulation (bottom row) in anaesthetized mice in which no pharmacological inhibitor was applied. The “pre” (left) and “post” (right) measurement points were time-matched to the pre and post time points in the experiments in which LNAME was administered (Fig. 1). Blue shading indicates photostimulation period (top row), grey shading indicates whisker stimulation period (bottom row). Data are mean ± SEM, $n$ represents number of mice.

Fig. 5

Hemodynamic responses evoked by nNOS-IN activation or whisker stimulation in presence of simultaneous inhibition of NOS and 20-HETE synthesis. Group data ($n=6$ mice). (a): Mean fractional change in Hbt, Hbo, and Hbr in arteriolar ROI before (left) and after (right) LNAME and HET0016 injection. Blue shading indicates photostimulation period (upper panels), grey shading indicates whisker stimulation period (lower panels). Data are mean ± SEM. (b), (c) Darker lines represent group mean ± SD, lighter lines indicate trial-averaged mean for individual animals. (b) Initial fractional change (“initial minima”) in Hbt in response to optogenetic and whisker stimulation, with (+) and without (−) inhibitors (LNAME+HET0016). (c) Maximum fractional change in Hbt (“peak”) evoked by optogenetic and whisker stimulation, with (+) and without (−) LNAME+HET0016. *$p<0.025$.

Fig. 6

Change in Hbt evoked by simultaneous presentation of whisker and optogenetic stimulation is similar to that predicted by summing changes in Hbt evoked by independent optogenetic and whisker stimulation. Group data ($n=5$ mice): Mean fractional change in Hbt in arteriolar ROI before (a) and after (b) NOS inhibition with LNAME. Blue shading indicates stimulation period. Continuous green line indicates hemodynamic response to 2 s simultaneous stimulation, and dash and dot green line indicates the predicted response calculated by summing the responses to separate 2 s optogenetic (grey dotted line) and 2 s whisker (grey dashed line) stimulations. For visual clarity, error bars are not shown. (Separate optogenetic and whisker stimulation data are also included in Fig. 1.)

Fig. 7

LNAME enhances vasomotion. (a), (c) Example 200 s Hbt time series from experiments occurring before (upper) and after (lower) LNAME injection (a) or matched timepoints with no inhibitor (c). Responses to individual stimulations can be seen. Grey shaded region indicates whisker stimulation, and blue shaded region indicates optogenetic stimulation. (b), (d) Example power spectrum of Hbt data (1000 s duration analyzed) from same experiments as (a), (c), before (upper, black) and after (lower, grey) LNAME injection (b) or matched timepoints with no inhibitor (d). Peaks are observed at frequency of stimulation (ISI 25s: 0.04 Hz) and its harmonics (black arrowheads). After LNAME injection, a peak at 0.1Hz (vasomotion, grey arrow) is observed and the apparent stimulation frequency is reduced (white arrowhead) (e): Mean power spectrum of oscillations in Hbt before (black) and after (grey) LNAME (left, $n=12$) or no inhibitor (right, $n=8$). Inset: highlight of 0.07 to 0.15 Hz. Data from arteriolar ROI in experiments as shown in Figs. 1 and 4. For visual clarity, error bars are not shown. $n$ indicates the number of mice.

Fig. 8

Optogenetic nNOS-IN stimulation produces nonconducted increases in Hbt across the arterial vascular network. (a) Reference images from two representative animals. Scalebar represents 1 mm. (b) Vessel map diagram and location of optical probe (black rectangle) and recording electrode (grey rectangle). (c)–(f) Average Hbt image taken 2 to 5s after the start of optogenetic nNOS-IN stimulation (c)–(d) or whisker stimulation (e)–(f), pre (c) and (e) and post (d) and (f) injection of LNAME. Color bars represent fractional change as compared with baseline. Black circles indicate local, nonconducted blood volume increases.