Astrocytes modulate cerebral blood flow and neuronal response to cocaine in prefrontal cortex

Abstract

Cocaine affects both cerebral blood vessels and neuronal activity in brain. Cocaine can also disrupt astrocytes, which modulate neurovascular coupling—a process that regulates cerebral hemodynamics in response to neuronal activation. However, separating neuronal and astrocytic effects from cocaine’s direct vasoactive effects has been challenging, partially due to limitations of neuroimaging techniques able to differentiate vascular from neuronal and glial effects at high temporal and spatial resolutions. Here, we used a newly-developed multi-channel fluorescence and optical coherence Doppler microscope (fl-ODM) that allows for simultaneous measurements of neuronal and astrocytic activities (reflected by the intracellular calcium changes in neurons Ca²⁺_N and astrocytes Ca²⁺_A, respectively) alongside their vascular interactions in vivo to address this challenge. Using green and red genetically-encoded Ca²⁺ indicators differentially expressed in astrocytes and neurons, fl-ODM enabled concomitant imaging of large-scale astrocytic and neuronal Ca²⁺ fluorescence and 3D cerebral blood flow velocity (CBFv) in vascular networks in the mouse cortex. We assessed cocaine’s effects in the prefrontal cortex (PFC) and found that the CBFv changes triggered by cocaine were temporally correlated with astrocytic Ca²⁺_A activity. Chemogenetic inhibition of astrocytes during the baseline state resulted in blood vessel dilation and CBFv increases but did not affect neuronal activity, suggesting modulation of spontaneous blood vessel’s vascular tone by astrocytes. Chemogenetic inhibition of astrocytes during a cocaine challenge prevented its vasoconstricting effects alongside the CBFv decreases, but it also attenuated the neuronal Ca²⁺_N increases triggered by cocaine. These results document a role of astrocytes both in regulating vascular tone and consequently blood flow, at baseline and for modulating the vasoconstricting and neuronal activation responses to cocaine in the PFC. Strategies to inhibit astrocytic activity could offer promise for ameliorating vascular and neuronal toxicity from cocaine misuse.

INTRODUCTION

In the brain cocaine directly affects cerebral vessels and neuronal activity and might additionally affect astrocytic activity [1, 2]. Cocaine’s effects in astrocytes, which modulate neurovascular coupling (NVC)—a process that regulates hemodynamic responses to neuronal activation, could also contribute to cocaine’s disruption of cerebral blood flow (CBF) [3]. Astrocytes interact with neurons and the surrounding blood vessels through their endfeet and processes that ensheath blood vessels and synapses [4]. Astrocyte processes terminate the action of glutamate released by neurons via the glutamate-glutamine cycle [5] and regulate CBF by modulating NVC [5]. Clinical studies have reported neuronal, astrocytic and vascular pathological alterations in the brain of cocaine users, including neuronal loss, reduction of glial fibrillary acidic protein (GFAP)-immunopositive astrocytes and reactive and degenerative changes of the cerebral micro vasculature [6]. In rodents astrocytes were shown to restore synaptic glutamate homeostasis in the NAcore after repeated cocaine exposure, and their manipulation attenuated relapse after cocaine withdrawal [7] indicative of cocaine’s affects in neuro-glio-vascular (NGV) processes.

Cocaine’s effects on cerebral blood vessels versus those in neurons and astrocytes are confounded by their interactions and change with chronicity of cocaine exposures. Thus, distinguishing these effects ideally requires simultaneous multi-parameter measurements performed longitudinally in vivo. Moreover, studying the role of astrocytes in brain function is difficult because their removal causes neuronal death [5]. Thus, much of what we know about astrocyte function has resulted from studies of isolated mammalian astrocytes in vitro, which cannot inform on how they interact with neurons and the surrounding vessels. Additionally astrocytes are hard to study with electrophysiological tools due to their slow changes in membrane potentials [5]. However, since astrocyte activation involves intracellular Ca²⁺ increases, the detection of astrocytic Ca²⁺ (Ca²⁺_A) signaling can be used to monitor their activity. Electrophysiological and Ca²⁺ imaging studies using rodent brain slices have led to new insights into astrocyte–neuron interactions and to astrocytes’ role in the activity of neuronal networks [8-10]. Also, with advances in in vivo imaging techniques such as two-photon microscopy (TPM), astrocyte activity and its correlation with cerebral blood flow is now being monitored in living brains [11]. However, studies on Ca²⁺_A signaling have been mostly conducted at the single-astrocyte level within a small field of view (FOV) as accessible by TPM rather than over larger astrocyte populations, from which synchronized Ca²⁺_A signaling arises [12, 13]. Indeed, the behavior of synchronized Ca²⁺_A signaling from large-scale astrocyte populations (e.g., astrocyte ensembles) is not well understood and neither is their functional role in NGV interactions.

Genetically encoded Ca²⁺ fluorescence indicators (GECIs) such as green GCaMP6f had enabled us to image cortical neuronal and astrocytic Ca²⁺ fluorescence signals in separate groups of animals [14]. Recently, jRGECO1a, a sensitive red GECI was used to image neuronal activity [15-17]. Therefore, we decided to combine red jRGECO1a with green GCaMP6f to simultaneously image neuronal and astrocytic activities in the same animal so we could study their interactions. To achieve this, we developed a spectrally resolved fluorescence imaging system capable of distinguishing two-color fluorescence emissions.

Advances in optical coherence tomography (OCT) for 3D vascular imaging have led to OCT angiography (OCTA or OCA) to visualize the vasculature [18-21] and Doppler OCT (ODT) [21, 22] for quantitative CBF velocity (CBFv) imaging. We reported simultaneous imaging with ultrahigh-resolution OCA (μOCA) and ODT (μODT) based on phase-intensity-multiplexing to concomitantly obtain 3D microangiography and quantitative CBFv measures at capillary resolution [22, 23]. 3D μODT measures intrinsic Doppler effect of moving red blood cells to image CBFv circumventing the need for a contrast agent. μODT allows 3D imaging of CBFv in arteries, veins, and capillaries [24] with high sensitivity (<20 μm/s) and a large field of view (e.g., 3 × 2.4 × 1.4 mm³). Such an ultrahigh-resolution CBFv imaging technique provides a powerful tool to study the role of astrocytes in NGV interactions and to investigate cocaine’s effect.

Here, we applied a novel multi-channel fluorescence and μOCA/μODT microscope (fl-ODM), which enabled us to concomitantly image large-scale astrocytic and neuronal Ca²⁺ fluorescence and 3D CBFv in vascular networks of the mouse prefrontal cortex (PFC). We used fl-ODM to acquire genetically encoded Ca²⁺ fluorescence images of cellular activities in neurons (Ca²⁺_N with jRGECO1a [15]) and in astrocytes (Ca²⁺_A with GCaMP6f [14]) alongside the CBFv effects in response to cocaine in the PFC of GFAP-cre mice [14] in vivo. We hypothesized that cocaine’s enhancement of Ca²⁺_A accumulation [25] in parallel to its vasoconstricting effects [26] would increase vulnerability to ischemia jeopardizing neuronal activities in the PFC. To assess the role of astrocytes in the NVC responses to cocaine, we used chemogenetics (Designer Receptors Exclusively Activated by Designer Drugs or DREADDs) to inhibit Ca²⁺_A accumulation (e.g., GFAP-DREADDs(Gi)), which we hypothesized would reduce cocaine-induced neuronal Ca²⁺_N activity and ameliorate CBF decreases in PFC.

RESULTS

Spectrally-resolved fluorescence and optical coherence Doppler microscopy for simultaneous imaging of neuronal, astrocytic and microvascular dynamics

Figure 1a illustrates a custom hybrid dual-channel fluorescence and ultrahigh-resolution optical coherence Doppler microscope (fl-ODM) which combines the two imaging modalities into an upright microscope body (FN1, Nikon) via epi-fluorescence cube turrets (C1, C2) for in vivo small animal studies. An ultrahigh-resolution optical coherence angiography (μOCA) and Doppler tomography (μODT) system in the near infrared range (λ = 1.3 μm, Δλ = 230 nm) was integrated through a dichroic mirror (λ_1DM = 1.1 μm) in C1 to provide 3D images of the microvasculature and quantitative cerebral blood flow velocity (CBFv) in vascular networks in the mouse cortex over a large field of view (FOV, e.g., 2.4 × 2 × 1.2 mm³) with capillary resolution (e.g., <5 μm). The technical details of 3D μOCA/μODT were previously reported [24] except a custom high-fidelity 2D confocal laser scanning module to interconnect the μOCT engine to fl-ODM. In parallel, a custom epifluorescence cube (C2) is used for 2-channel spectral-multiplex imaging of the synchronized intracellular calcium fluorescence changes in astrocytes (Ca²⁺_A expressed with GCaMP6f: λ_EX1 = 485 ± 12 nm, λ_DM1 = 495 nm, λ_EM1 = 520 ± 20 nm) and in neurons (Ca²⁺_N expressed with jRGECO1a: λ_EX2 = 559 ± 8 nm, λ_DM2 = 573 nm, A_EM2≥ 574 nm) over a larger FOV (e.g., 4 × 3 mm²). Pulsed high-power narrow-band blue (488 nm) and yellowish green (560 nm) LEDs from a light engine (Aura III, Lumencor) were used for time-sharing spectral excitation and synchronized with a SCMOS camera (Zyla 5.5, Andor) for sequential fluorescence image acquisition (T = 10 ms exposure per channel). Ca²⁺_N(t) and Ca²⁺_A(t) imaged at up to 80fps were quantified as the relative florescence changes (ΔF/F) vs their baselines to represent neuronal or astrocyte activities, respectively.

Fig. 1. In vivo imaging of NGV interactions.

a Hybrid fl-ODM for simultaneous 3D μOCA/μODT of microvasculature/CBFv (2.4 × 2 × 1.2 mm³) and 2-channel neuronal and astrocytic Ca²⁺ fluorescence imaging (4 × 3 mm²) of mouse cortex, modified based on a Nikon FN-1 microscope using a broadband 5x obj (e.g., LSM03, Thorlabs). See details in Methods. b Viral injection to express Ca²⁺ in astrocytes (Ca²⁺_A - GCaMP6f) and neurons (Ca²+_N - jRGECO1a) in the cortex of GFAP-cre mice. c A sketch to illustrate fl-ODM for simultaneous imaging of neuro-astroglio-vascular interactions in vivo. d1–d3 in vivo images of Ca²⁺_A (GCaMP6f) and Ca²+_N (jRGECO1a) channels and their merged images, where the inserts (d1’, d2’) show the ex vivo confirmation of cell-specific expressions; e Ex vivo confocal image to show the distribution of astrocytes (labeled by GFAP, green), neurons (labeled by NeuN, red). GFAP: antibody of fibrillary acidic protein to visualize astrocytes; NeuN: antibody for neuronal nuclei; f, g Neuron and astrocyte distributions in the dashed area in (e), indicating that astrocytes unsheathe the microvessels (dashed red lines) in the brain to likely mediate microflows.

To define the mechanisms by which astrocytic and neuronal activities are involved in NGV interactions in vivo, we used viral injection to express genetically encoded Ca²⁺ indicators in mice in a cell-specific manner. Figure 1b illustrates our approach to express astrocytic Ca²⁺_A and neuronal Ca²⁺_N in the PFC in vivo. The use of GFAP-cre mouse with delivery of two mixed viral vectors including 50% AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 (#100835, Add-gene) and 50% AAV1.Syn.NES-jRGECO1a.WPRE.SV40 (100854, Add-gene) allowed us to express GCaMP6f for Ca²⁺_A fluorescence and jRGECO1a for Ca²⁺_N fluorescence in the cortex. Prior to imaging, a cranial window was implanted above the PFC as illustrated in Fig. 1c (see Method Section for details). Figure 1d1-d3 shows simultaneous in vivo images of Ca²⁺_A (d1) and Ca²⁺_N (d2) fluorescence from PFC of a GFAP-cre mouse at ~4wks after viral injection of synapsin jRGECO1a for neurons and cre-GCaMP6f for astrocytes. Ex vivo double staining with GFAP antibody to label astrocytes and NeuN antibody to label neurons confirmed astrocyte- (d1’) and neuron- (d2’) specific Ca²⁺ expressions. Astrocytes also modulate NVC through their process endfeet forming close interactions with neurons and microvessels [5]. Our ex vivo images (Fig. 1e-g) show that astrocytes (labeled by GFAP, green) ensheathe neurons (labeled by NeuN, red) and microvessels (red dashed lines—GFAP of astrocyte endfeet) in the mouse cortex (e, g).

Cocaine increased neuronal Ca²⁺_N and astrocytic Ca²⁺_A activity and decreased CBFv

To demonstrate the technical capability of fl-ODM for tracking neuronal, astrocytic and vascular changes in real time, we simultaneously imaged the mouse PFC (A/P: +2.5;M/L: 0.5;D/V: −0.5 mm) to detect activations of Ca²⁺_N, Ca²⁺_A fluorescence and local CBFv changes elicited by an acute cocaine challenge (1 mg/kg, i.v.). Figure 2a’-c’ show representative Ca²⁺_N, Ca²⁺_A and CBFv images obtained from a GFAP-mouse at baseline (before cocaine). To assess the dynamic changes in Ca²⁺_N, Ca²⁺_A and CBFv from time of cocaine injection, five regions of interest (ROIs, i.e., white circles illustrated in a’ and b’) were selected in the PFC within the fluorescence expressing regions and five in surrounding vessels as shown in Fig. 2c’. Figure 2a-c show the time courses of Ca²⁺_N, Ca²⁺_A and CBFv in response to cocaine, respectively. Cocaine increased Ca²⁺_N and Ca²⁺_A activities (Fig. 2a, b), whereas it decreased CBFv in cerebrovascular trees (c). Specifically, cocaine transiently increased Ca²⁺_N followed by a downshoot whereas it induced a longer lasting though weaker increase in Ca²⁺_A that paralleled the long-lasting decrease in CBFv.

Fig. 2. Illustration of time-lapse images to derive Ca²+_N, Ca²⁺_A fluorescence and local CBFv changes in response to acute cocaine in mouse PFC.

Simultaneous imaging of Ca²+_N (a’) and Ca²⁺_A (b’) fluorescence along with 3D μODT of the local CBFv network (c’) in the PFC of a GFAP-mouse. Five regions of interest (ROIs) from fluorescence-expressing regions of jRGECO1a for Ca²+_N (circles in a’) and GCaMP6f for Ca²⁺_A (circles in b’) as well as 5 vessels (c’) were selected to track their dynamic changes before and after acute cocaine at t = 0 min (1 mg/kg, i.v.). a–c Time courses of mean Ca²⁺ fluorescence changes in neurons (ΔCa²⁺_N, solid line, m = 5) and astrocytes (ΔCa²⁺_A, solid line, m = 5) along with CBFv network changes in response to cocaine. Increases in ΔCa²⁺_N and ΔCa²⁺_A indicate activation of neurons and astrocytes by cocaine whereas the decrease in CBFv presumably reflects the vasoconstricting effects of cocaine. d–f Time-lapse 3D μODT images of CBFv dynamic responses of all 3 vascular components to cocaine, μODT(t), and their ratio changes over the baseline (t < 0 min), ΔμODT(t) (image size: 2 × 0.3 × 1.2 mm³); d Cocaine-induced CBFv changes in arteriolar (AF), venular (VF) and capillary (CF) flow networks, in which 4 flows in each compartment were tracked to quantify their mean variations (bold curves) shown in f). *: periods showing significant flow decreases (P* < 0.001, m = 4). ROI labels: 1-4 (red), 1-4 (light blue) and 1-4 (green) refer to arteries, veins, and capillaries.

Figure 3 summarizes the cocaine-induced mean changes in Ca²⁺_N, Ca²⁺_A and CBFv within different vascular compartments (e.g., arteries, veins and capillaries) acquired from n = 7 animals. The mean time-course changes in Fig. 3a indicate that cocaine induced a ΔCa²⁺_N increase of 2.46 ± 0.88% at t_p-N = 8.2 ± 2.1 min followed by recovery to baseline at t_r-N = 28.8 ± 3.5 min. Similarly, ΔCa²⁺_A increased to 2.97 ± 0.43% at t_p-A = 12.1 ± 2.2 min, but the effect was long-lasting and did not return to baseline until t_r-A = 59.5 ± 8.0 min. In parallel cocaine decreased mean vascular ΔCBFv to −25.1 ± 4.9% at t_p-V = 21.0 ± 2.9 min, which slowly recovered to baseline at t_r-V = 64.0 ± 7.5 min, with a similar duration to that of ΔCa²⁺_A. Statistical comparisons of the response duration to cocaine between ΔCa²⁺_N, ΔCa²⁺_A and ΔCBFv are summarized in Fig. 3b, which indicates that the response duration of Ca²⁺_A and CBFv to cocaine was significantly longer (~2 folds) than that of Ca²⁺_N (P* = 0,005 and P* = 0.002, respectively). No significant difference was found between t_r-A and t_r-V (P = 0.62).

Fig. 3. Comparisons of cocaine’s effects on neuronal Ca²⁺_N, astrocytic Ca²⁺_A fluorescence and vascular CBFv in the PFC (n = 7 mice).

a Mean ΔCa²⁺_A (green), ΔCa²⁺_N (red) and vascular ΔCBFv (black) responses to cocaine (1 mg/kg, i.v.). b Comparisons of return time to baseline between ΔCa²⁺_A, ΔCa²⁺_N and ΔCBFv, showing that the Ca²⁺_A and CBFv responses to cocaine lasted significantly longer than that of Ca²⁺_N, while there was no difference between ΔCa²⁺_A and ΔCBFv (P = 0.62, n = 7). c Temporal correlations of Ca²⁺_A and Ca²⁺_N transient changes vs CBFv changes in response to cocaine.

To assess the temporal relationship between the short-lasting dynamic change in Ca²⁺_N fluorescence (e.g., <30 min), and the long-lasting changes in Ca²⁺_A and CBFv (e.g., 50–58 min) we assessed the temporal correlations between them (Supplementary Fig. S2). The results shown in Fig. 3c revealed a strong correlation between ΔCa²⁺_A and ΔCBFv (0.765 ± 0.048, n = 7) that was significantly stronger than the significant but a weaker correlation between ΔCa²⁺_N and ΔCBFv (0.389 ±0.115, P* = 0.011, n = 7).

Cocaine effects in various vessel compartments

In addition, fl-ODM allowed us to concomitantly measure cocaine-induced transient CBFv changes in arteries, veins and capillaries. Figure 2d, e shows time-lapse μODT images and the ratio changes over the baseline of a smaller volume (2 × 0.3 × 1.2 mm³, marked by a dished box in Fig. 2c’) in the PFC acquired to quantify dynamic flow changes (2 min/volume) from baseline (t < −4 min) to t > 50 min after cocaine (1 mg/kg, i.v., t = 0 min). The 3D μODT image in Supplementary Fig. S1 shows that the selected ROIs included pial flows in layer 1 and deep flows in layers 4–5. Under isoflurane anesthesia, all three vessel compartments (f) showed flow decreases after cocaine injection, among which arteriolar (AF) and venular (VF) flows dropped −19.8% ± 6.3% (P* < 0.001, m = 4; t = 8–36 min) and −29.3 ± 4.5% (P* < 0.001, m = 4; t = 6–36 min), respectively, followed by a gradual recovery to their baselines at 49.5 ± 4.6 min and 45.5 ± 3.2 min, respectively. Although capillary flows (CF) showed an overall decrease to −13.5 ± 3.0% that peaked at 20 ± 2.9 min, individual flow changes varied, with increases over 7% in some capillaries and decreases over −35% in others. The heterogeneity in the capillary responses highlights the importance of measuring flow in multiple capillaries instead of isolated vessels when studying NGV interactions and its responses to cocaine.

Activation of GFAP-DREADDs(Gi) inhibited astrocytic Ca²⁺_A and induced vasodilation and increased CBFv during baseline

DREADDS is a chemogenic approach that enables subtype selective activation (Gq) or silencing (Gi) of cellular signaling (e.g., astrocytes) via clozapine activation [27]. A cocktail of two viruses consisting of 0.4 μl AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 and 0.4 μl AAV5.GFAP.hM4D(Gi)- mCherry was injected into the PFC of GFAP-cre mice. Figure 4 shows in vivo results using Gi-coupled DREADDS(hM4Di) expression in astrocytes (referred as GFAP-DREADDS(Gi)) into the mouse PFC. After 4–6 wks from injection, time-lapse images of Ca²⁺_A fluorescence and cerebrovascular networks in the PFC were continuously acquired before and after clozapine injection (0.1 mg/kg, i.p.) at t = 0 min for over 40 min. Clozapine instead of clozapine-N-oxide (CNO) was used to activate DREADDS(Gi) because a recent study showed that clozapine (the CNO metabolite) rather than CNO itself stimulates the DREADDS receptor [27]. Figure 4b is a representative ratio ΔCa²⁺_A image of a mouse PFC post clozapine (t = 25 min) over its baseline (t = −5 min), showing Ca²⁺_A decreases within the blue region of brain tissue along with vasodilation (red tracks in vascular trees, Fig. 4a). Details of vascular responses to clozapine activation of DREADDS(Gi) can be visualized in Supplementary movie VS1 and ratio image shown in Supplementary Fig. S3. Figure 4d summarizes the mean changes in vascular diameters [F(199,796) = 6.46, P* < 0.001] and CBFv [F(199,796) = 25.1, P* < 0.001] as a function of time post clozapine administration across animals (m = 5 ROI/parameter/animal, n = 5 mice). DREADD(Gi) activation resulted in vasodilation, with an increase of Δϕ = 9.17 ± 0.96% in mean vessel diameters (black trace) from baseline (t = −5 min, Δϕ = 0.22% ± 0.35%) and an increase in ΔCBFv of 10.51 ± 2.95% at 25 min post clozapine compared to baseline (t = −5 min, ΔCBFv = 0.21 ± 1.18%). The time course of ΔCa²⁺_A in response to astrocyte inhibition is shown in Fig. 4e, revealing a significant decrease [F(80,240) = 6.7, P* < 0.001] after 5 min post clozapine injection (P* < 0.05); at t = 25 min post clozapine, ΔCa²⁺_A decreased −3.0% ± 1.0% over its baseline (t = −5 min, ΔCa²⁺_A = 0.02% ± 0.16%).

Fig. 4. Effects of DREADDS(Gi) activation on the NGV interactions.

a–c Ratio images of cerebrovessels and of Ca²⁺_A and Ca²⁺_N fluorescence at t = 25 min after GFAP-DREADDS(Gi) activation via clozapine vs their baselines at t = −5 min, indicating vasodilation (red edges in a) and Ca²⁺_A fluorescence decrease (blue area in b) but no effects on Ca²+_N (in c) in PFC; d–f Mean time courses of Δϕ increase or vasodilation and ΔCBFv increase in d and Ca²⁺_A decrease in e but no Ca²⁺_N change in f after DREADDS(Gi) activation (n = 5 mice). g Correlation analyses of ΔCa²⁺_A(t) and ΔCa²⁺_N(t) vs Δϕ(t); h A comparison of correlation coefficients of ΔCa²⁺_A(t) vs Δϕ(t) (green bar) and ΔCa²⁺_N(t) vs Δϕ(t) (red bar), showing a significant difference (P* = 0.01).

To further characterize the role of astrocytes in modulating vascular tone, we used an excitatory DREADD (AAV5.G-FAP.hM3D(Gq)).mCherry) to evaluate the effects of astrocytic activation in an opposite strategy to the DREADD (Gi) inhibition described above. The details of virus deliveries and drug administration are summarized in Table 1 (Exp. 4). We then measured the vascular dynamics and the changes in neuronal Ca²⁺_N, and astrocytic Ca²⁺_A in response to astrocyte excitation by DREADD (Gq) through clozapine (0.1 mg/kg, i.p.).

Table 1.

Experimental design and procedures.

Experiment	Pre-Virus Injection (Animal #: Male, Female)	Drug challenge	Associated Figures
Exp 1:
Simultaneous CaA2+, CaN2+ & CBFv responses to cocaine	AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 AAV1.Syn.NES-REGCO1a.WPRE.SV40 (n = 7: M = 4, F = 3)	Cocaine (1 mg/kg, i.v.)	Figs. 2, 3
Exp 2:
Neuron, astrocytes and vascular Changes in response to astrocytic inhibition of DREADD (Gi) active by Clozapine	AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 AAV5.GFAP.hM4D(Gi)).mCherry (n = 5: M = 2, F = 3)	Clozapine (0.1 mg/kg, i.p.)	Fig. 4
	AAV1.Syn.NES-jREGCO1a.WPRE.SV40 AAV5.GFAP.hM4D(Gi)).mCherry (n = 5: M = 2, F = 3)
Exp 3:
Simultaneous CaA2+, CaN2+ & CBFv responses to cocaine before and after DREADD (Gi) activation	AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 AAV5.GFAP.hM4D(Gi)).mCherry (n = 3: M = 2, F = 1)	1st: baseline followed by cocaine (1 mg/kg, i.v.) 60 min later; 2nd: baseline followed by clozapine (0.1 mg/kg, i.p.) 30 min later; 3rd: baseline followed by cocaine (1 mg/kg, i.v.) 60 min later	Fig. 5
	AAV1.Syn.NES-jREGCO1a.WPRE.SV40 AAV5.GFAP.hM4D(Gi)).mCherry (n = 4: M = 2, F = 2)
Exp 4:
Neuronal, astrocytic and vascular Changes in response to astrocytic excitation of DREADD (Gq) by Clozapine	AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 AAV5.GFAP.hM3D(Gq)).mCherry (n = 4: M = 3, F = 1)	Clozapine (0.1 mg/kg, i.p.)	Fig. S4
	AAV1.Syn.NES-jREGCO1a.WPRE.SV40 AAV5.GFAP.hM3D(Gq)).mCherry (n = 4: M = 0, F = 4)
Exp 5:
Cocaine-induced CBFv changes with GFAP mCherry (Control) viral expression	AAV5-GFAP104-mCherry (n = 5: M = 2, F = 3)	1st: baseline followed by cocaine (1 mg/kg, i.v.) 60 min later; 2nd: baseline followed by clozapine (0.1 mg/kg, i.p.) 30 min later; 3rd: baseline followed by cocaine (1 mg/kg, i.v.) 60 min later	Figs. S8-9
Exp 6:
Test whether chemogenetic inhibition of astrocytes affects hemodynamic (CBV) changes evoked by forepaw stimulation	AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 AAV5.GFAP.hM4D(Gi)).mCherry (n = 3: M = 2, F = 1)	Clozapine (0.1 mg/kg, i.p.)	Fig. S10

Supplementary Fig. S4 summarizes the results and shows that excitation of astrocytes by DREADD (Gq) through clozapine resulted in vasoconstriction. The vascular shrinkage is presented by the blue edges of the vascular tree shown in Fig. S4a. The mean changes in vessel diameters and corresponding CBFv changes (n = 4) are shown in Fig. S4d, indicating a decrease of Δϕ = −1.0% ± 0.65% in mean vessel diameters (black trace) from baseline (t = −5 min, Δϕ = 0.10% ± 0.17%) and a decrease in ΔCBFv of −18.54 ± 13.98% at 25 min post clozapine over the baseline (t = −5 min, ΔCBFv = 0.91 ± 2.34%). Meanwhile, exciting astrocytes significantly increased astrocytic Ca²⁺ (i.e., ΔCa²⁺_A) 5 min after clozapine injection (P* < 0.05); at t = 25 min, the increase was 1.88 ± 0.36% (P* < 0.05) over the baseline (t = −5 min, ΔCa²⁺_A = 0.03 ± 0.06%). A slight decrease in neuronal Ca²⁺_N (0.91 ± 0.14%, P* ≤ 0.03) was observed at 6–8 min after clozapine followed by a gradual recovery (Fig. S4f). While astrocytes inhibition by DREADD(Gi) downregulated vascular tone (Fig. 4a, d), excitation of astrocytes by DREADD (Gq) upregulated it. The bidirectional modulation of astrocytic function confirms that astrocytic Ca²⁺_A signaling modulates vascular tone and adjusts CBFv in the brain. Taken together, these results provide evidence that astrocytic signaling modulates vascular tone at baseline, thus adjusting CBFv.

To evaluate whether astrocytic inhibition would affect neuronal activity at baseline and in response to cocaine (see subheading below), we injected a mixture of viruses to express GCaMP6f into neurons (0.4 ul AAV5.Syn.GCaMP6f.WPRE.SV40) and to express GFAP-DREADDs(Gi) (0.4 μl AAV5.GFAP.hM4D(Gi)-mCherry) into astrocytes in the PFC (n = 5). Figure 4c shows a representative ratio image of neuronal ΔCa²⁺_N fluorescence of a mouse PFC at t = 25 min after GFAP-DREADDs(Gi) activation by clozapine over the baseline (t = −5 min), exhibiting no neuronal Ca²⁺_N fluorescence changes. Figure 4f shows the mean ΔCa²⁺_N time courses across animals, indicating no significant changes [F(80,320) = 1.22, P = 0.12] before (−0.08 ± 0.06%, t = −5 min) and after (−0.64 ± 0.43%, t = 25 min) GFAP-DREADDs(Gi) activation (P > 0.05, m = 5/animal, n = 5). This result indicates that inhibition of astrocytes via clozapine activation of GFAP-DREADDs(Gi) did not influence baseline neuronal activity (Ca²⁺_N). We corroborated the specificity of DREADDs(Gi) delivery into astrocytes in immunostained brain sections that indicated AAV5.GFAP.hM4D(Gi)-mCherry expression uniquely in astrocytes (Supplementary Fig. S5).

Figure 4g plots the temporal correlations between ΔCa²⁺_A(t) and ΔCa²⁺_N(t) with Δϕ(t), and Fig. 4h shows that the cross-correlation between ΔCa²⁺_A and Δϕ (r = 0.651 ± 0.07) was significantly stronger than that between ΔCa²⁺_N and Δϕ (r = 0.411 ± 0.10, P* = 0.01). Similarly, the correlation between ΔCa²⁺_A(t) and ΔCa²⁺_N(t) with ΔCBFv(t) showed that the correlation between ΔCa²⁺_A and ΔCBFv (r = 0.535 ± 0.03) was significantly stronger than that between ΔCa²⁺_N and ΔCBFv (r = 0.347 ± 0.05, P* = 0.008) (Supplementary Fig. S6). These results indicate that inhibition of astrocytes resulted in vasodilation and CBFv increases and though the correlations between decreases in neuronal activity and vasodilation and CBFv increases were also significant, the effect was significantly smaller.

GFAP-DREADD (Gi) inhibition of astrocytic Ca²⁺_A activation attenuated cocaine-induced decreases in CBFv and the neuronal activation

To examine whether inhibition of astrocytic signaling (Ca²⁺_A) could block the CBFv decrease due to cocaine’s vasoconstricting effects, we imaged the vascular responses to cocaine in PFC without and with GFAP-DREADDS(Gi) activation by clozapine. Animal preparations were described in Table 1 (Exp 3). As shown in Fig. 5, for this experiment, two sets of imaging sessions were conducted with at least 2 h separation between two sequential cocaine injections: images were acquired from 10 min prior (baseline) to 60 min after the first cocaine injection (1 mg/kg, i.v.); followed by a no intervention period after which clozapine injection (0.1 mg/kg, 0.16 ml) was given and 30 min later the second in vivo imaging session was initiated including a 10 min baseline measure prior to and a 60 min one following a second cocaine injection (1 mg/kg, iv). Astrocytic Ca²⁺_A fluorescence and CBFv images were continuously recorded prior to and following each cocaine injection (n = 3). In an additional group of animals (n = 4) we assessed whether inhibition of astrocytes influenced the neuronal response to cocaine by imaging neuronal Ca²⁺_N fluorescence and CBFv changes in response to cocaine without and with GFAP-DREADD(Gi) activation by clozapine.

Fig. 5. Comparisons of cocaine-induced Ca²⁺_N, Ca²⁺_A and CBFv changes with vs without DREADDS(Gi) activation.

Top panel (I-III): Illustration of experimental procedure. Inhibiting Ca²⁺_A with GFAP-DREADD (Gi) blocked cocaine-induced vasoconstriction and CBFv reduction and blunted the neuronal activation in response to cocaine in PFC. a, b Cocaine-induced mean astrocytic Ca²⁺_A changes with time before and after GFAP-DREADD(Gi) activation (n = 3), showing that GFAP-DREADD(Gi) activation inhibited cocaine-induced Ca²⁺_A increase. Insets: representative ratio images of Ca²⁺_A fluorescence at=30 min after cocaine vs the baseline (t = −2 min) in one animal ΔCa²⁺_{A t=30/t=2}) illustrating the blockade of cocaine-elicited ΔCa²⁺_A increase (a) after GFAP-DREADD(Gi) activation (b). c Comparison of cocaine-induced mean ΔCa²⁺_A rates within t = 0–30 min, showing blockade of cocaine’s effects on ΔCa²⁺_A after inhibition of Ca²⁺_A with GFAP-DREADD (Gi) (P* = 0.004). d, e Cocaine-induced mean CBFv changes before and after Ca²⁺_A inhibition by GFAP-DREADD(Gi) (n = 7), showing that GFAP-DREADD(Gi) activation inhibited cocaine-induced CBFv decreases. Insets: representative ratio images of CBFv at t = 30 min after cocaine vs the baseline (t = −2 min) in one animal ΔCBFv_t=30/t=2), illustrating the blockade of cocaine-elicited CBFv decreases (d) after GFAP-DREADD(Gi) activation (e). f Comparison of cocaine-induced mean ΔCBFv rates within t = 0–30 min without and with GFAP-DREADD(Gi) activation, showing significant decreases in cocaine’s effects on ΔCBFv after inhibiting Ca²⁺_A with GFAP-DREADD (Gi) (P* = 0.01). g, h Cocaine-induced mean neuronal Ca²⁺_N changes with time before and after Ca²⁺_A inhibition by GFAP-DREADD(Gi) (n = 4), showing that GFAP-DREADD(Gi) activation blunted cocaine-induced Ca²⁺_N increases. Insets: representative ratio images of Ca²⁺_N fluorescence at=30 min after cocaine vs the baseline (t = −2 min) in one animal ΔCa²⁺_{N t=30/t=2}), showing that cocaine-induced neuronal ΔCa²⁺_N return to baseline at t = 30 min post cocaine (g) was shortened to 10 min with astrocytes’ inhibition (h). i Comparison of cocaine-induced mean ΔCa²⁺_N rates within t = 0–30 min without and with GFAP-DREADD(Gi) activation, showing significant decrease in cocaine’s effects after inhibiting Ca²⁺_A with GFAP-DREADD (Gi) (P = 0.02).

Figure 5a, b shows the temporal responses of mean astrocytic Ca²⁺_A to cocaine before and after DREADD(Gi) activation, in which the insets illustrate the corresponding representative ratio images of Ca²⁺_A fluorescence at t = 30 min after cocaine vs baseline (t = −2 min), indicating that cocaine-induced mean astrocytic Ca²⁺_A increase (ΔCa²⁺_A (a) was inhibited after GFAP-DREADD(Gi) activation (b)). Multiple ROIs (m = 5) within the PFC from each animal (n = 3) were selected to track temporal Ca²⁺_A fluorescence changes ΔCa²⁺_A/Ca²⁺_A) after cocaine and the data averaged to derive mean ΔCa²⁺_A(t) curves in Fig. 5a, b. A two-way Repeat Measurement (RM) ANOVA (Fig. 5a, b) revealed a significant treatment by time interaction [F(70,140) = 8.52, P* < 0.001], with post-hoc tests revealing that DREADD inactivated groups had a significant increase (P* < 0.001) in ΔCa²⁺_A following cocaine injection whereas with DREADD activation (astrocyte inhibition) there was no change in ΔCa²⁺_A following cocaine (P = 1). Specifically, the ΔCa²⁺_A rate, defined as the averaged per minute Ca²⁺_A increase post cocaine during t = 0–30 min was blocked from 5.55 ± 0.83%/min to 0.35 ± 0.32%/min (P* = 0.004) after GFAP-DREADD(Gi) activation (Fig. 5c), consistent with inhibition of cocaine-induced Ca²⁺_A increase in astrocytes. Similarly, Fig. 5d, e shows the mean temporal CBFv responses (ΔCBFv/CBFv) to cocaine before and after GFAP-DREADD(Gi) activation (n = 7, including animals both with astrocytic-GCaMP6f or neuronal-GCaMP6f and GFAP-DREADD(Gi) expressions) as well as their corresponding representative ratio images to illustrate the blunting of cocaine-induced ΔCBFv decrease. A two-way RM ANOVA showed a significant group by time interaction [F(70,420) = 1.86, P* < 0.001], with post-hoc tests revealing that inactivated DREADD groups had a significant reduction (P* = 0.001) in ΔCBFv following cocaine injection whereas the activated DREADD group showed no change in ΔCBFv (P = 0.86). Specifically, CBFv decreased from −0.72% ± 1.42% at baseline (t = −2 min) to −14.64 ± 4.75% at t = 30 min after cocaine (Fig. 5d), which was reduced to −1.43 ± 5.02% (t = 30 min) from its baseline (−0.47 ± 1.44% at = −2 min) with GFAP-DREADD(Gi) activation (Fig. 5e). Figure 5f indicates that cocaine-induced ΔCBFv decrease was reduced significantly from −10.1% ± 2.1%/min to −2.03 ± 1.6%/min after GFAP-DREADD(Gi) activation (P* = 0.01, n = 7). Demonstration of cocaine-induced vascular response without and with astrocytes’ inhibition via DREADDs(Gi) activation can be visualized in Supplementary movies VS2 and VS3, respectively. It showed that cocaine induced vasoconstriction, but this effect was eliminated by astrocytes’ inhibition after DREADDs(Gi) activation (Supplementary Fig. S7). These results support our hypothesis that blocking Ca²⁺_A increases would abolish cocaine-induced vasoconstriction and prevent the associated CBFv decreases.

To compare cocaine’s effects on the PFC without elective inhibition of astrocytes by clozapine, we used virus expressing specifically targeting astrocytes (AAV5-GFAP104-mCherry) as a control experiment. The experimental procedure has been described in Table 1 (Exp 5), which was similar to the experiment of GFAP-DREADD (Gi) inhibition of astrocytes (Exp 3, Table 1).

Supplementary Fig. S8 shows the time traces of mCherry fluorescence (a marker of astrocyte expression) and CBFv in PFC in response to clozapine, which was done as control for potential pharmacological effects of clozapine. One-way repeated ANOVA showed no significant time effect on mCherry fluorescence in these control animals after clozapine (0.1 mg/kg, i.p., n = 5). Quantification analysis shows no significant difference in mCherry fluorescence before (0.04 ± 0.05%/min) and after (−0.07 ± 0.23%/min) clozapine injection (P = 0.65), as well as no significant difference in CBFv (−0.62 ± 0.53%/min and −0.91 ± 1.91%/min before and after clozapine, P = 0.89), thus indicating that clozapine (0.1 mg/kg, i.p.) in controls (without DREADDs) did not affect the CBFv changes within the brain.

Supplementary Fig. S9 shows mean time-course changes in mCherry fluorescence (ΔmCherry) and in cerebral blood flow (ΔCBFv) in response to cocaine (1 mg/kg, i.v.) pre- and post-clozapine administration in these control mice (n = 5). There were no significant changes in mCherry fluorescence after cocaine over their baselines for neither pre-clozapine (P = 0.93) nor post-clozapine (P = 0.97) conditions. The mCherry fluorescence between pre- and post-clozapine did not differ from each other (F[70, 280] = 0.35, P = 1); their mean ΔmCherry_A per unit time were −0.06 ± 0.18%/min and −0.16 ± 0.13%/min, respectively (P = 0.91) (Fig. S9c). Cocaine injection significantly decreased CBFv both during the pre-clozapine (t = 11–60 min, p < 0.001) and post-clozapine conditions (t = 2–60 min, P* < 0.001). There was no significant difference in CBFv between pre- and post- clozapine (F[70, 280] = 0.39, P = 1);the mean ΔCBFv were −9.0 ± 1.63%/min and −7.99 ± 0.41%/min, respectively (P = 0.53, Fig. S9f)).

These results indicate that administration of clozapine (0.1 mg/kg, i.p.) in control mice with mCherry-expressed in astrocytes did not (1) affect the vascular physiology (measured by CBFv) or (2) modulate cocaine-induced CBFv decreases. Together with the experiments of GFAP-DREADD(Gi) inhibition of astrocytes shown in Fig. 5, it confirms that the attenuations of cocaine-induced CBFv decreases and neuronal activation were due to astrocytic inhibition activated by clozapine and not to the pharmacological effects of clozapine or to the mCherry virus.

While GFAP-DREADD(Gi) targets astrocytes and inhibits cocaine-induced Ca²⁺_A increase (Fig. 5a), it was unclear whether it could indirectly modify neuronal responses to cocaine. Figure 5g, h shows the temporal responses of mean neuronal Ca²⁺_N to cocaine before and after DREADD(Gi) activation to inhibit Ca²⁺_A. A two-way RM ANOVA showed a significant group by time interaction [F(70,210) = 5.49, P* < 0.001], with post-hoc tests showing that the DREADD inactivated group had a significant increase (P* < 0.001) following cocaine injection whereas the DREADD activated group did not (P = 0.35). Unlike the long-lasting response of astrocytes to cocaine (>60 min, Fig. 5a) the neuronal Ca²⁺_N increase returned to baseline at 30 min post cocaine (Fig. 5g). The full-width-half-maximum duration of cocaine-induced Ca²⁺_N increase was reduced from τ_g = 11.02 ± 1.85 min to τ_h = 2.24 ± 0.62 min after DREADD(Gi) activation. The ΔCa²⁺_N rate decreased from 1.11% ± 0.29%/min to 0.11 ± 0.1%/min after GFAP-DREADD(Gi) activation (Fig. 5i), thus indicating a significant reduction of cocaine-induced neuronal activation (P* = 0.02, n = 4). Taken together with the effect of DREADDS(Gi)’s blockade on cocaine-induced CBFv decreases (Fig. 5f), these results indicate that cocaine’s effects on CBFv and neuronal activity (Ca²⁺_N) were modulated by Ca²⁺_A signaling. Thus, inhibition of astrocytic activity might help alleviate cocaine associated PFC dysfunction resulting from improper tissue perfusion whereas the decrease in neuronal reactivity might help reduce compulsive drug taking.

To determine whether chemogenetic inhibition of astrocytes would affect hemodynamic changes (e.g., CBFv) evoked by a non-pharmacological process, we conducted an additional experiment to test whether DREADD(Gi) activation (i.e., inhibition of astrocytes) would influence the hemodynamic increase evoked by forepaw electrical stimulation. Forepaw stimulation, which has been widely used in brain functional studies with fMRI [28, 29] and optical imaging [30], results in increases in CBF and cerebral blood volume (CBV) in the somatosensory cortex to meet the energetic demands evoked by the stimulation as a result of neurovascular coupling (NVC). Studies assessing the role of astrocytes in NVC have measured intracellular Ca²⁺ as a marker of activity and have shown that Ca²⁺ elevation in astrocytes is associated with release of vasoactive compounds that might drive CBF/CBV changes in NVC [31-34].

For the stimulation studies, GFAP-DREADD(Gi) (AAV5.G-FAP.hM4D(Gi).mCherry) and GECI (AAV5.CAGFlex.GCaMP6f.WPRE.SV40) were pre-viral delivered into the somatosensory cortex of GFAP-cre mice (n = 3). Details of the experimental design are shown in Table 1 (Exp 6). After 4 weeks of viral injection, we imaged the stimulation-evoked changes in Ca²⁺_A signaling (ΔCa²⁺_A) and in CBV (ΔCBV) before and after DREADD(Gi) activation by clozapine. Figure S10 shows increases in both Ca²⁺_A and CBV in response to a 3 mA/5 Hz/10 s forepaw stimulation (Fig. S10a, c, respectively). However, both Ca²⁺_A and CBV responses were blocked following astrocytic inhibition via DREADD(Gi) activation at t = 30 min after clozapine injection (Fig. S10b, d, respectively). These results indicate that astrocytes inhibition blocked the stimualtion elicited vasodilation in the somatosensory cortex, thus futher corroborating that astrocytes regulate hemodynamic reponses and NVC.

DISCUSSION

Neuroimaging has advanced our understanding of the brain but there is need for tools with cellular and capillary resolutions capable of distinguishing signaling from distinct cell types alongside the dynamics of the vascular responses. In particular delineating the roles of astrocytes and neurons in NVC is relevant for understanding the physiological processes that regulates energetic needs of the brain and their disruption by drugs such as cocaine. Here we integrate fluorescence imaging and ultrahigh-resolution ODT to form multi-channel fluorescence and optical coherence Doppler microscopy (fl-ODM), and apply it to study the acute effects of cocaine on the NGV circuit in the mouse PFC. Specifically, we investigated how cocaine affects astrocytic Ca²⁺_A and neuronal Ca²⁺_N activities and CBFv, and tested the hypothesis that cocaine induces Ca²⁺_A accumulation resulting in vasoconstriction and CBFv decreases that disrupt NVC. We also investigated the role of Ca²⁺_A in cocaine-elicited Ca²⁺_N changes, and tested the hypothesis that reducing Ca²⁺_A via DREADDS(Gi) could would also decrease Ca²⁺_N increases. To ensure that our results were not due to the pharmacological effects of clozapine or to the disruption from the viral vector we conducted experiments in control mice that did not express DREADD. To determine if the role of astrocytes in hemodynamic responses to cocaine generalized to physiological stimuli we assessed the effects of astrocytic inhibition on the hemodynamic responses in somatosensory cortex to forepaw stimulation.

DREADD-induced modulation in astrocytes

Using inhibitory and excitatory DREADDs of astrocytic activation, we observed bidirectional astrocytic regulation of neurovascular CBFv in the mouse PFC. Astrocytes express multiple GPCRs [35] that coupled to Gq, such as serotonin 5-HT2a and 5-HT2B receptors [36-38] to Gi such as GABAB receptors [39], adrenergic α2-AR [36, 40], adenosine receptors A1 and A3 [36, 41], dopamine D2, D3 and D4 receptors [36, 42, 43] or to Gs as A2A and A2B receptors [36, 41].

Gi-DREADD activation in astrocytes was found to inhibit cAMP signaling [44, 45], consistent with what has been observed for multiple endogenous Gi-GPCRs [46-49]. Whereas Gq-GPCR activation in astrocytes increases cAMP signaling and Ca²⁺ increases in various brain regions [50-52], there are some inconsistencies reported on the effects of Gi-GPCR signaling in astrocytes on intracellular Ca²⁺ [53, 54]. Some observed a reduction in intracellular Ca²⁺ [53], which is in line with the canonical Gi-GPCR pathway. However, others found no apparent effect on Ca²⁺ levels [55] or increases in Ca²⁺ levels [53, 56-59]. Discrepancies on Ca²⁺ levels with Gi-DREADD activation in astrocytes of VTA area were explained by differences in duration of application of the GABAB agonist [53], or the duration and concentration of the DREADD agonist CNO/clozapine in hippocampal astrocytes [60-63]. With a clozapine dose of 0.1 mg/kg, i.p., we observed Ca²⁺_A increases concurrently with vasoconstriction with Gq-DREADD (astrocytic activation), whereas with Gi-DREADD (astrocytic inhibition), Ca²⁺_A decreased along with vasodilation. These results document that astrocytes are involved in NVC regulation at baseline.

Cocaine-induced Ca²⁺_A accumulation resulted in vasoconstriction and CBF decreases that disrupted neurovascular coupling

Although cocaine’s vasoconstricting effects are well documented [64-66], the cellular mechanisms that underlie it remain elusive. Accumulating evidence indicates that drug exposure can have dynamic and long-lasting effects on astrocytes and other glial cells. [67] To the extent that DREADD(Gi) leads to vasodilation we hypothesized that it would block the effects of cocaine.

In addition astrocytic Ca²⁺_A signaling cascades are involved in the communication between neurons and astrocytes, and astrocyte-to-astrocyte communicate via Ca²⁺ waves that propagate signaling over a large range [5]. Also activation of Ca²⁺_A may release glutamate to regulate synaptic homeostasis [7]. Thus Ca²⁺_A may regulate CBF as a function of its modulation of neuronal activity or independently of synaptic activity [68].

Prior studies from our group and others have shown that astrocytes contribute to vasodilation during NVC and to the vasoconstriction that subsequently restores vascular tone [14, 69]. However, studies on the contribution of astrocytes to cocaine’s effects on the brain have mostly focused on its synaptic and circuitry regulation associated with its rewarding and addictive effects [70]. Only few studies have investigated the effects of acute cocaine on astrocyte activity, including an in vitro study that used brain slices from the nucleus accumbens incubated with cocaine that reported increases in Ca²⁺ transients in astrocytes [71]. To our knowledge, there are no reported in vivo studies that have simultaneously imaged Ca²⁺_N and Ca²⁺_A fluorescence alongside CBFv changes in response to cocaine. The hybrid fl-ODM imaging platform reported here uniquely enabled us to characterize how astrocyte and neuronal networks interact and mediate the associated local neurovascular responses to cocaine in the PFC.

Cocaine directly affects both cerebral blood vessels and neuronal activity in the brain [26, 72-74]. Glia, in particular astrocytes, are involved in NVC, which modulates hemodynamics in response to changes in neuronal activity [75]. NVC can be disrupted by use of addictive drugs such as cocaine and by disease processes such as Alzheimer’s disease and other dementias. The ability to distinguish neuronal from vascular effects remains a challenge, partially due to technical limitations of neuroimaging techniques to differentiate vascular from neuronal and glial effects at high spatiotemporal resolutions. Here, we applied fl-ODM to study cocaine’s effects on the neurovascular network and on neuronal and astrocytic activities in the PFC. Specifically, we simultaneously imaged activations of Ca²⁺_N, Ca²⁺_A fluorescence and local CBFv changes elicited by an acute cocaine challenge (1 mg/kg, i.v.). Findings revealed a temporal association between cocaine-induced CBFv decreases (due to vasoconstriction) and the long-lasting astrocytic activation. Analysis of temporal correlations showed that cocaine-induced Ca²⁺_A increases had strong negative correlations with CBFv decreases. These findings are in agreement with our recent results obtained in separate groups of animals [1]. In that study we used single virus containing GCaMP6f (not jRGECO1a) delivered into the somatosensory cortex and reported that cocaine-induced neuronal Ca²⁺_N increases recovered by 30 min followed by a downshoot similar to our observations in this study but in the PFC (Fig. 3a above). In our previous study we also showed that cocaine-induced Ca²⁺_N changes were inversely correlated with temporal changes in tissue oxygenation [1]. The duration of neuronal activation in response to cocaine observed here and in our prior study is consistent with the pharmacokinetics of intravenous cocaine in the brain [76-78] and to the duration of striatal dopamine increases [79]. The inverse association between Ca²⁺_N and tissue oxygenation indicates that cocaine’s effects on neuronal activation and deactivation underly the changes in tissue oxygenation and might also underlie the reductions in brain glucose metabolism reported during cocaine withdrawal [80, 81]. Although we had observed that the lasting increases in Ca²⁺_A induced by cocaine were associated with vasoconstriction as assessed via quantification of vessel diameter changes [1], the measurements of Ca²⁺_A and Ca²⁺_N changes were done in separate groups of mice, whereas in the current study the fl- ODM enabled us to assess the dynamic Ca²⁺_A and Ca²⁺_N responses simultaneously in the same animals. This was crucial for assessing the role of astrocytes in mediating the neuronal responses to cocaine alongside vascular function. It also allowed us to control for group variability. For example, in our prior study using separate groups of mice, the amplitude of cocaine-induced Ca²⁺_A was 3 fold lower than that of the Ca²⁺_N response [1], whereas in the current study that measured them in the same animals there were no differences between the peak values of Ca²⁺_A and Ca²+_N (i.e., 2.97 ± 0.42% vs 2.46 ± 0.42%, P = 0.82). This discrepancy likely reflects differences between the groups (i.e., WT vs GFAP-cre mice) in our prior study. Nevertheless, this study extended our previous findings and showed for the first time that inhibition of Ca²⁺_A also attenuated cocaine-induced Ca²⁺_N increases.

Reducing Ca²⁺_A via DREADDS(Gi) relieved cocaine-induced vasoconstriction, CBFv decreases and Ca²⁺_A increases

At baseline inhibition of astrocytic activation decreased Ca²⁺_A and resulted in vasodilation and CBFv increases but did not affect neurons (Ca²⁺_N). This suggests that at baseline astrocytes, but not neurons, mediate vascular tone, which is consistent with previous reports with two-photon microscopy. [34] During a cocaine challenge, the inhibition of astrocytic activation prevented the CBFv decreases triggered by cocaine-induced vasoconstriction. Different from baseline, inhibition of astrocyte activation also blunted neuronal activation by cocaine. Together, these findings indicate the involvement of astrocytes in mediating cocaine’s effects on vasoconstriction and in modulating the neuronal responses to cocaine. Though the mechanisms by which astrocytes affect neuronal responses to cocaine are unclear it is possible that it could involve astrocytes’ role in terminating the action of glutamate released by neurons via the glutamate-glutamine cycle [5] and in restoring synaptic glutamate homeostasis after cocaine exposure [7, 25]. Moreover, manipulating astrocyte function attenuate relapse after cocaine withdrawal [7, 82, 83].

Human studies have reported neuronal, astrocytic and vascular pathology in the brain of individuals with cocaine use disorder (as well as other drugs of abuse) that encompassed neuronal loss, reduction of glial fibrillary acidic protein (GFAP)-immunopositive astrocytes and reactive and degenerative changes of cerebral microvessels [6]. These observations imply that drugs including cocaine initiate a cascade of interacting toxic processes in the NGV circuit that are likely to contribute to the cognitive and behavioral changes observed in drug users. Intracellular Ca²⁺ increases are associated with cell death [5]; thus the cocaine-induced cellular Ca²⁺ increases that we observed might be clinically relevant, particularly since they occur in parallel with CBF decreases and hypoxia [80]. The use of a new fl-ODM enabled us to separate cocaine’s effects on astrocytes, neurons and vascular networks to underpin their contributions to PFC dysfunction induced by cocaine. We focused on the PFC since clinical studies provide ample evidence of PFC dysfunction in drug users [3, 26] that is implicated on the loss of control over drug taking [84]. Relevant to our findings is a recent report that mediation of Ca²⁺_A signaling ameliorated neuronal death and reduced behavioral deficits after ischemic stroke [85]. We had also reported that nifedipine (Ca²⁺ antagonist and vasodilator [86]) prevented cocaine-induced CBF decreases and neuronal Ca²⁺ increases in PFC and reduced cocaine intake. In that study we interpreted nifedipine’s actions to indicate neuronal effects, our current findings suggest that blockade of L-type Ca²⁺ channels in astrocytes are also likely to be involved [87-90]. Our current findings provide new insights into cocaine’s effects on NGV interactions that may provide new targets for development of novel addiction treatments.

A limitation for our study was that experiments were conducted in anesthetized mice using isoflurane to avoid artifacts from animal motion during imaging. Isoflurane-induced vasodilation might have facilitated the detection of cocaine-induced vasoconstriction, especially in capillaries whereas its anesthetic effects might have attenuated the sensitivity of neurons [91] and perhaps also of astrocytes to cocaine. Another limitation was that in order to compare cocaine’s effects with and without blockade of astrocytic activation by DREADDs(Gi), each animal was administered cocaine twice, which might have resulted in tolerance. In our study we gave the second cocaine dose 2 h after the first one, for we had previously shown that the hemodynamic responses to cocaine in PFC between two cocaine doses separated by a 2 h interval did not differ from one another [92]. Also in our study the recordings in the vehicle condition always preceded those in the ‘DREADD-applied’ state, which might have influenced responses. However, the fact in our control experiments (Exp 5) we showed no significant difference in cocaine-induced CBFv changes in the pre- and post- clozapine conditions (Fig. S9) indicates that the signal differences between vehicle and ‘DREADD-applied’ conditions were not due to the order of the condition and that there was no evidence of tolerance to cocaine’s effects after 2 cocaine doses 2 h apart from each other. To shorten the experimental time, we skipped the saline injection in the vehicle condition with the assumption that potential signal fluctuations from such a small saline volume (0.16 ml) would be negligible. Indeed, we did not observe CBFv changes when an equal volume with clozapine was injected in the control mice (no DREADD) as shown in Fig. S8. Another limitation was that we only used one dose of cocaine, which did not allow us to assess if there was a dose response effect as reported for various physiological and behavioral effects of cocaine [93]. Thus, further investigation is warranted to determine if the responses we reported are also dose dependent. Another limitation was the viral delivery protocol, which only allowed delivery of two types of viruses into the PFC to minimize animal loss from the viral injections and to ensure sufficient amount of viruses (e.g., 0.4 ul per virus) to be expressed in each cell type (e.g., neuron, astrocytes) for fluorescence detection. The complexity of the studies also restricted the sample sizes of animals investigated, which precluded us to compare male and female responses to cocaine. Our findings pertain to effects of acute cocaine but further studies are needed to assess how they might change with chronic administration. Nevertheless, understanding the molecular mechanism underlying cocaine’s effects on NVC will be helpful for future putative treatments to reduce cerebrovascular pathology and neuronal toxicity from cocaine use.

In conclusion, we observed that at baseline astrocytes modulated vascular tone but did not change neuronal activity whereas during a cocaine challenge they prevented the CBFv decreases and also attenuated cocaine-induced Ca²⁺_N increases in PFC. Our findings provide further evidence for the role of astrocytes in modulating NGV interactions in responses to cocaine both via direct effects in cerebral blood vessels and indirectly via their modulation of neuronal reactivity. Strategies to inhibit astrocytic activity could be promising in addressing vascular and neuronal toxicity from cocaine misuse. In addition, our findings demonstrate the capabilities of fl-ODM for distinguishing activities of different cells (e.g., neurons, astrocytes) and its compatibility with other imaging tools for simultaneous monitoring of hemodynamics (or other processes).

METHODS

Animals

All experiments were carried out according to the National Institutes of Health guidelines and were approved by the Institutional Animal Care and Use Committee of Stony Brook University. As summarized in Table 1, a total of 40 mice were used in this study for six experiments, including 21 females (21/40 = 52.5%) and 19 males (19/40 = 47.5%). The animal numbers (including for males and females) for each experiment are described in Table 1. GFAP-cre mice, obtained from Jackson Laboratory and maintained as a heterozygous line, were used for experiments when they reached the age between postnatal 60–70 days (P60-P70). All the information regarding the generation and genotyping of this line is available at https://www.jax.org/strain/024098. The physiological conditions of the mice (Respiration, body temperature, etc.) were continuously monitored and recorded (Small Animal Monitoring and Gating system, model 1025 L, SA Instruments Inc.).

Specifically, the body temperature was maintained as 36–37.5 °C by a customized heating pad (with voltage-controlled electric circles) and the respiration was maintained 40–80 breaths/minute using 1.5 ~ 2% isoflurane in oxygen gas to ensure similar physiological status for all mice.

Viral expression of genetically encoded Ca²⁺ indicators (GECIs) in neurons and astrocytes

To simultaneously image the fluorescence from green and red GECIs (GCaMP6f, jRGECO1a) differentially expressed in astrocytes and neurons, a mixture of viral vectors containing 1:1 AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 (100835-AAV5, Addgene) and AAV1.Syn.NES-jRGECO1a.WPRE.SV40 (100854-AAV1, Addgene) was injected into the PFC (A/P: +2.5 mm; M/L: 0.5 mm; D/V: −0.5 mm) of GFAP-cre mice for astrocytic and neuronal uptake under isoflurane anesthesia. A total volume of 0.8 μl was slowly injected (e.g., ~0.1 μl/min) using a Hamilton syringe, and after the microinjection the needle was left intact for additional 5–10 min to avoid backflow or leakage of the injected viral vectors. To monitor the vascular and Ca²⁺_A responses to DREADD(Gi) activation by clozapine and examine whether Ca²⁺_A was involved in cocaine-induced vasoconstriction, a mixture of two viral vectors containing 0.4 μl AAV5.CAG.Flex.GCaMP6f.WPRE.SV40 and 0.4 μl AAV5.GFAP.hM4D(Gi)-mCherry (50479-AAV5, Addgene) was injected into the same PFC region. The details of viral delivery for each experiment have been summarized in Table 1. During viral injection, mice were anesthetized with inhalation of 2% isoflurane mixed with pure oxygen and their heads mounted on a stereotaxic frame while we monitored their physiology. After completion of the procedure, the mice were monitored daily for a few days to ensure that they fully recovered from the surgery.

Cranial window implantation

To image the brain function in prefrontal cortex (PFC) in Exp 1-5, a region of interest on the mouse PFC (A /P: +2.5; M/L: 0.5; D/V: −0.5 mm) was selected whereas for Exp 6, the somatosensory cortex [A/P: + 1.7; M/L: −1.5; D/V: −0.5 mm] was imaged. In the region of interest on the mouse cortex, the cortical bone was first thinned using a dental drill and then carefully removed, leaving the dura intact. The explored brain region was treated with dexamethasone sodium phosphate (50989-437-12, VEDCO) and then immediately covered by a 3.5 × 4.5 mm² coverslip and sealed with biocompatible glue. Dental cement was spread around the edges of the coverslip to further secure its attachment with the skull for repeated imaging.

In vivo time-lapse fl-ODM imaging

A custom hybrid fl-μODM developed in our lab was used for simultaneous 3D μOCA/μODT of microvasculature and CBFv (2.4 × 2 × 1.2 mm³) and dual-channel Ca²⁺_N and Ca²⁺_A fluorescence imaging (4 × 3 mm², 80fps) of mouse cortex in a spectral-multiplex, time-sharing mode in the mouse PFC (Fig. 1a). In fluorescence cube (C1), a dichroic mirror (λ_DM≈1.1 μm) reflected 1.3 μm μODM light and transmitted GCaMP6f (520 nm) and jRGECO1a (600 nm) Ca²⁺ fluorescence, which integrated the fluorescence microscope (Nikon FN1) and 3D μOCT. The spectral-domain μOCT engine was illuminated by an ultrabroad-band source (Super-K laser: p > 100 mW, λ = 1.3 μm, Δλ = 230 nm) and detected by fast 2k InGaAs array (GL2048R, Sensors Unlimited) at up to 140kpfs. Fluorescence cube (C2) was a spectral-multiplex epi-fluorescence filter set. For astrocytic GCaMP6f-Ca²⁺_A and neuronal jRGECO1a-Ca²⁺_N fluorescence imaging, light beams from 10ms-duration pulsed narrow-band blue LED at 488 nm and yellowish-green LED at 560 nm of a light engine (Aura III, Lumencor) were combined in a light guide to illuminate a modified fluorescence microscope (FN1 Nikon) for excitation. By spectral multiplexing, the epifluorescence cube C2 selectively allowed green Ca²⁺_A (500–540 nm) and red Ca²+_N (574–670 nm) emission from the mouse cortex (3 × 4 mm²) to be acquired by a sCMOS camera (Zyla 5.5, Andor) in a time-sharing mode synchronized with the excitation pulses at up to 80fps. Here, CM/FPC were fiberoptic collimator and polarization controller, and the Time Base was a digital I/O to synchronize 2 image workstations for fluorescence and μOCA/μODT imaging, PZT was a piezoelectric actuated focal tracking, and the μODT scan head was a custom fast 2D confocal laser scanner to interconnect FN1 and μOCT engine.

For vascular imaging in this study, a full-size 3D μODT image of mouse CBFv networks (2.4 × 2 × 1.2 mm³) was acquired in ~15 min, and time-lapse μODT images over a smaller volume (e.g., 2 × 0.3 × 1.2 mm³) were acquired per 1.2 min or less for tracking flow dynamic changes (ΔCBFv) along with Ca²⁺_A and Ca²⁺_N activations (40 fps/channel). The recorded Ca²⁺_A and Ca²+_N activity was quantified as the relative fluorescence change (ΔF/F); similarly, the flow network change was quantified as the ratio image (ΔCBFv/CBFv).

Mice were anesthetized using inhalational isoflurane (1.5% ~ 2.5%) and the head mounted onto a stereotaxic frame. To assess cocaine-induced dynamic changes in astrocytic or neuronal Ca²⁺ fluorescence and CBFv response (Exp 3, and Exp 5), we administered cocaine (1 mg/kg, i.v.) through the tail vein twice during experiments (Table 1). We selected the 1 mg/kg dose because it is a dose that is commonly used in self-administration studies, and it is equivalent to doses self-administered by cocaine misusers [94]. It also has the advantage of being a dose frequently used in animal studies [22, 80, 95, 96], which facilitates the comparisons of our results to those of others.

Immunohistochemistry

After in vivo imaging studies, the mouse was perfused transcardially with 0.1 M PBS, followed by fixation with 4% paraformaldehyde in 0.1 M PBS. The frozen brain was sliced to 40 ~ 50 μm in thickness. For GCaMP6f signal enhancement to astrocytes, the antibody [chicken anti-GFP (1:200) antibody] to green fluorescence was used as the primary antibody followed by an Alexa Fluor 488 anti-chicken for GFP (1:200) conjugated secondary antibody. To identify GFAP-DREADDs(Gi) expression into astrocytes, antibody was used to enhance green fluorescence emission in astrocytes, but no immunostaining was used for GFAP-DREADDS(Gi). The jRGECO1a expression to neurons was imaged with a confocal fluorescence microscope (A1, Zeiss) without immunostaining for fluorescence enhancement.

Data quantification and analysis

Multi-wavelength images were acquired using the time-sharing strategy and arranged to time-functional order for each channel. 3D μODT will be reconstructed by phase-intensity method (PIM) [97]. For GCaMP6f-Ca²⁺ fluorescence imaging, five regions of interest (ROIs, white circle Fig. 2a, b) were selected within the GCaMP6f/jRGECO1a-expressing cortical areas devoid of visible blood vessels to track the temporal changes of astrocytic or neuronal fluorescence. Using Time-lapse 3D μODT images, CBFv dynamic responses of all 3 vascular components (4–5 ROIs for each vascular types, arteries, veins and capillaries) to cocaine, μODT(t), and their ratio changes ΔμODT(t) over the baseline (t < 0 min) to track their changes as a function of time. To correct the absorption effects due to the hemodynamic changes, fluorescence expressed values were divided by no fluorescence expressing the cortex regions.

The neuronal or astrocytic Ca²⁺ fluorescent activity was quantified as ΔF/F (ΔF/F[Ca²⁺_N], ΔF/F[Ca²⁺_A], respectively). The relative changes of fluorescence signal over its baseline (e.g., before cocaine), $\Delta \mathsf{F}∕\mathsf{F}(\Delta [{\mathsf{Ca}}^{2+}])=([\mathsf{F}(\mathsf{t})\text{-}{\mathsf{F}}_{\mathsf{baseline}}]∕{\mathsf{F}}_{\mathsf{baseline}})\mathsf{x}100](\%)$ were calculated for each ROIs for each animal to eliminate effects of variations in GCaMP6f/jRGECO1a expression on fluorescence intensity changes. Cerebral blood flow velocity (CBFv) was quantified as relative change in $\Delta \mathsf{CBFV}=([\mathsf{CBFv}(\mathsf{t})\text{-}{\mathsf{CBFv}}_{\mathsf{baseline}}]∕{\mathsf{CBFv}}_{\mathsf{baseline}})\mathsf{x}100](\%)$.

Specifically, for Experiment 1 (Table 1, Exp.1), three data sets were obtained from the images of each mouse, including (1) [Ca²⁺] fluorescence intensity change from astrocytes (ΔF/F[Ca²⁺_A]); (2) from neurons (ΔF/F[Ca²+_N]), which reflected astrocytic or neuronal activation to the cocaine; (3) cerebral blood flow change (ΔCBFv) in response to cocaine (1 mg/kg, i.v.). For Experiment 2-5 (Table 1, Exp.2-5), four data sets were obtained from each mouse: (1) [Ca²⁺] dependent fluorescence intensity change from astrocytes (ΔF/F[Ca²⁺_A]) or (2) neurons (ΔF/F[Ca²+_N]); (3) cerebral blood flow change (ΔCBFv), (4) changes in vessel diameter (Αφ) before and after DREADD(Gi) (or (Gq)) activation by clozapine (0.1 mg/kg, i.p.) or/and in response to cocaine (1 mg/kg, i.v.). For Experiment 6, the [Ca²⁺] dependent fluorescence intensity change from astrocytes (ΔF/F[Ca²⁺_A]) and cerebral blood volume changes (ΔCBV) before and after electrical forepaw stimulation.

Each ROI was selected independently depending on the local expression of each subject for astrocyte, neuron, and vessels. Data from different animals in each group were averaged and the mean response of each measurement from each mouse (averaged all ROIs response) was used for statistical comparison and for calculation of correlation coefficients (r values). Data was analyzed with mix-sex (i.e., no separation of females from males due to small samples).

Statistics

All data are presented as mean ± s.e.m. Data were analyzed by one-way or two-way mixed model analysis of variance (ANOVAs) and the Holm-Sidak method was used for post-hoc analysis. Comparisons made between two different groups (e.g., ΔCa²⁺_N-Δϕ vs ΔCa²⁺_A-Δϕ, cocaine-induced ΔCa²⁺_A without vs with DREADDS(Gi)/(Gq) activation) were analyzed using Student’s t-tests. If P-values are less than 0.001, then they are reported as P* < 0.001; otherwise, precise P-values are provided. For correlation coefficients (r values), Pearson correlations were used and t-test were applied to compare between neuron vs. CBFv and astrocyte vs. CBFv. All statistical tests were performed using SigmaStat software (Systat Software Inc), with alpha levels set at 0.05 to report significance.

DATA AVAILABILITY

Correspondence and requests for additional materials should be addressed to Yingtian Pan.