Page 1/36

Novel Ca2+-modulated Photoactivatable ImagingReveals Neuron-Astrocyte Glutamatergic Circuitrieswithin the Nucleus AccumbensIrene Serra

Instituto Cajal, CSICJulio Esparza

Instituto Cajal, CSICLaura Delgado

Instituto Cajal, CSICCristina Martin

Instituto Cajal, CSIC https://orcid.org/0000-0001-6170-0615Margalida Puigròs

Institute for Biomedical Research of Barcelona, CSIC, CIBERNEDPetar Podlesniy

Institute for Biomedical Research of Barcelona, CSIC, CIBERNEDRamon Trullas

Spanish National Research Council https://orcid.org/0000-0001-7951-9881Marta Navarrete ( mllinas@cajal.csic.es )

Instituto Cajal, CSIC https://orcid.org/0000-0003-2097-4788

Article

Keywords: neuron-astrocyte communication, astrocyte heterogeneity, nucleus accumbens, glutamatergicafferents, mGluR5

Posted Date: October 22nd, 2021

DOI: https://doi.org/10.21203/rs.3.rs-956625/v1

License: This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License

Page 2/36

AbstractAstrocytes are key elements of brain circuits that are involved in different aspects of the neuronalphysiology relevant to brain functions. Although much effort is being made to understand how thebiology of astrocytes affects brain circuits, astrocytic network heterogeneity and plasticity is still poorlydened. Here, we have combined structural and functional imaging of astrocyte activity using the Ca2+-modulated photoactivatable ratiometric integrator and specic optostimulation of glutamatergicpathways to map the functional neuron-astrocyte circuitries in the nucleus accumbens (NAc). We showedpathway-specic astrocytic responses induced by selective optostimulation of main inputs from theprefrontal cortex, basolateral amygdala, and ventral hippocampus. Furthermore, the differences in basalCa2+ dynamics between the NAc shell and core astrocytes were associated with differences inmitochondrial DNA copy number, exhibiting molecular heterogeneity in the regulation of theirmitochondrial genomes. Finally, co-stimulation of glutamatergic pathways induced non-linear Ca2+-signaling integration, revealing integrative properties of NAc astrocytes. All these results demonstrate theexistence of specic neuron-astrocyte circuits in the NAc, a critical insight to the understanding of howthe NAc integrates information.

IntroductionAlthough astrocytes have been traditionally regarded as a homogeneous population of cells,accumulating evidence indicates that there is functional heterogeneity in the astrocyte subpopulations1–4

from different brain regions based on their morphology, functionality, physiological properties,developmental origin and response to pathologies5–8. Neuron-astrocyte communication has been shownto be a key element in brain physiology, yet the heterogeneity in the functional interaction betweenastrocytes and synapses in different brain circuitries remains poorly understood.

The nucleus accumbens (NAc) is a part of the basal forebrain that plays a relevant role in the rewardsystem in numerous neurological and psychiatric disorders9. One remarkable feature of NAc is that itreceives and integrates different glutamatergic signals from several brain regions including the medialprefrontal cortex (mPFC), basolateral amygdala (Amyg), and ventral hippocampus (vHip). The NAc hasbeen divided into two different structures, the so-called shell (AcbSh) and core (AcbC), distinguished bydifferential expression of neuropeptides, morphology, membrane properties, and synaptic inputs fromafferent structures. The AcbC is related to guiding behavior towards a specic goal based on learning,whereas the AcbSh is involved in unconditioned reward-seeking behaviors10.

The particular pathway-specic activation of glutamatergic bers to the NAc has been demonstrated toevoke diverse physiological and behavioral responses10,11. For example, the mPFC-NAc pathway interactswith behaviors that require not only visual attention, but also attention in the action, taking into accountthe outcomes of expected/anticipated reward12,13, working memory14 or the integration of differentinputs15. vHip afferents regulate rewards related to contextual cues or reward-related memories15–17,

Page 3/36

which are relevant to disorders that alter reward processing, e.g. addiction or anhedonia, and are involvedin susceptibility to stress disorders16,18. Lastly, afferents from Amyg are involved in value and emotionaldecisions and reward seeking, focused on positive valence19–21. However, all these studies were centeredmainly on the innervation patterns of distinct input and/or output features of different types of neuron9,22,23, leaving the functional intersection of astrocytes in these hallmarks underappreciated for decades.

Considering astrocytes as integrative space-time detectors of neuronal signaling24,25, we exploredwhether astrocytes establish segregated neuron-astrocyte networks within the NAc that have intrinsicproperties and functional consequences to the neuronal circuit´s output. To this end, we optogeneticallymanipulated the specic synaptic afferents to the NAc in combination with a new adapted technique,calcium-modulated photoactivatable ratiometric integrator under GFAP promoter (CaMPARIGFAP), to

selectively dissect Ca2+ astrocyte signaling. We demonstrated that selective optostimulation of mainglutamatergic inputs (i.e. mPFC, Amyg, and vHip) induces astrocytic Ca2+ activities mediated by mGluR5that do not coincide with glutamatergic innervation patterns, suggesting unexpected neuron-astrocytecircuitries

ResultsCaMPARI GFAP as a new functional tool to monitor astrocyte activity.

To study functional neuron-astrocyte circuitries in the NAc, we used a uorescent technique based onCaMPARI26,27, a genetically encoded Ca2+ indicator (GECI) that undergoes irreversible green-to-reduorescence conversion upon coincident elevated intracellular Ca2+ and ultraviolet (UV, λ = 405 nm) lightillumination28. The GECI properties of the molecule allow real-time monitoring of astrocytic Ca2+

dynamics, while its irreversible photoconversion properties enable large-scale spatial analysis ofastrocytic activation with precise temporal resolution. We selectively expressed CaMPARI in NAcastrocytes by injecting the adeno-associated virus AAV2/9-GFAP-CaMPARI (CaMPARIGFAP), whichcontains the specic astroglial promoter glial brillary acidic protein (GFAP) in the NAc (Figs. 1A and 1B).Specic expression of CaMPARIGFAP in NAc astrocytes was conrmed for both green and reduorescence signals by its colocalization with astrocytic marker S100β, but negligibly with the neuronalmarker NeuN. 95.7 ± 0.7% of CaMPARIGFAP Green cells and 94.1 ± 1.7% of CaMPARIGFAP Red cells wereS100β-positive (p < 0.001; n = 2861 cells, 8 elds, 2 mice) (Fig. 1C and D). Afterward, we studied both theability to track real-time Ca2+ astrocytic activity and photoconversion properties28. Using local applicationof ATP (20 mM) through a micropipette, which reliably elevates intracellular Ca2+ levels in NAcastrocytes29, we showed a transient decrease in CaMPARIGFAP uorescence in astrocytes (Fig. 1E1−3),

demonstrating the ability of the molecule to monitor Ca2+ dynamics. In parallel, application of UV lightduring a xed temporal window led to green-to-red photoconversion in those astrocytes that were activeat the moment of illumination due to the ATP stimulation (Fig. 1F1). CaMPARIGFAP Red was measuredpost-hoc after PFA xation. The CaMPARIGFAP Red/Green uorescence ratio (FRed/FGreen) changed

Page 4/36

according to the distance from ATP application (Fig. 1F2) showing that red uorescence intensity

correlated with Ca2+ activity conrming that this parameter can be used to quantify the astrocyticactivation28.

Although it is being reported that CaMPARI uorescence signal is reduced upon chemical xation28,30,photoconversion turnover was robust in astrocytes, and this decrease did not compromise CaMPARIGFAP

post-hoc signal detection (Fig. 1F). Furthermore, no signicant photoconversion was detected whentissue was not exposed to UV light, conrming that background CaMPARIGFAP spontaneous turnover isresidual and does not affect the reported measurements (Fig. S1). Overall, these observations indicatethat astrocytic Ca2+ activity can be studied using both CaMPARIGFAP approaches, real-time monitoring ofcalcium dynamics and large-scale activation proles given by uorescence turnover analysis.

Optostimulation of mPFC axons induces specic activation of a subpopulation of astrocytes in the NAc.

To study neuron-astrocytic circuitry heterogeneity in the NAc, we rst analyzed the neural element of thecircuit, characterizing the anatomical and functional patterns of the mPFC glutamatergic afferents to theNAc (Fig. 2A and B). To this end, we injected AAVs expressing the opsins channelrhodopsin (ChR2) orChrimsonR with mCherry or tdTom uorescent reporters (AAV-CaMKIIa-hChR2(H134R)-mCherry; AAV-hSync-ChrimsonR-tdTom) in the mPFC. Four weeks after injection, we analyzed the projectinguorescence signals from mPFC axons specically in the NAc. Medium spiny neurons (MSNs) from boththe AcbC and AcbSh subregions were analyzed to characterize both density and synaptic strength ofmPFC afferent inputs. In agreement with previous studies26,31, we found non-uniform distribution ofmPFC afferents in the NAc, with higher innervation density in the AcbC (0.25 ± 0.04 ΔF/F0) than theAcbSh (0.14 ± 0.03 ΔF/F0) (p = 0.02; n = 11 mice; Fig. 2A2). Optogenetic stimulation of the mPFC witheither ChR2 or ChrimsonR (Fig. S2) evoked glutamatergic excitatory postsynaptic currents (EPSCs; asindicated by their sensitivity to 20 µM CNXQ and 50 µM AP5; Fig. S2C). The amplitude of EPSCs recordedin AcbC (208.6 ± 37.4 pA; n = 8 cells, 4 mice) and AcbSh neurons (33.1 ±12.3; n=15, 4 mice: p = 0.002;Fig. 2B3) correlated positively with the afferent innervation density (Pearson r = 0.73, p < 0.001; Fig. 2B2).To avoid differences derived from stimulation intensity, light stimulation parameters remained constantfor all the recordings (70% intensity, 1 ms, Fig. S2B see Materials and Methods). These results suggestthat mPFC neurons preferentially innervate the AcbC subregion and that the opto-evoked EPSC amplitudein NAc MSNs was dependent on the density of innervating glutamatergic bers.

We next investigated the astrocytic Ca2+ dynamics associated with the mPFC glutamatergic pathway(Fig. 2C), focusing on the post-synaptic element in the neuron-astrocyte circuitry under study. To this end,we monitored in both NAc subregions the Ca2+ events using CaMPARIGFAP Ca2+-dependent greenuorescence, before and after optogenetic stimulation (10 pulses at 4 Hz, 200 ms interval - 4 times, 5 sinterval). This protocol increased the astrocyte Ca2+ spike frequency in both the AcbC (1.62 ± 0.23 changefrom basal, p = 0.02) and AcbSh (2.62 ± 0.52 change from basal, p = 0.003) (n = 25 slices, 8 mice; Fig.2C3, left) without variations in the Ca2+ amplitude of the responses (Fig. S3A). Surprisingly, these Ca2+

Page 5/36

dynamics in response to optostimulation did not show the same spatial pattern of activity across NAcregions as observed by the glutamatergic inputs’ innervation, indicating that astrocytes were notresponding exclusively in areas within the direct reach of the glutamatergic afferents. Although the AcbCshowed a higher density of mPFC projections, higher astrocytic activity was detected in the AcbSh (Fig.2A2 and 2C3). This increase in Ca2+ response was mediated by group I metabotropic glutamate receptor,as it was blocked by the selective mGluR5 antagonist MPEP (50 µM) (0.83 ± 0.10 change from basal, p =1.67) (n = 8 slices, 2 mice; Fig. 2C3). Overall, these results showed that mPFC inputs induce astrocyticactivity in a pattern across subregions opposite from the innervation prole.

To investigate this, we performed a spatial analysis of the astrocytic response in the whole NAc, takingadvantage of the photoconversion of CaMPARIGFAP in a Ca2+-dependent manner (Fig. 2D-2E). To avoidcrosstalk between photoconverted CaMPARIGFAP Red signals and mCherry from ChR2-expression axons,these experiments were performed using a different reporter (EYFP) fused to ChR2 (AAV-CaMKIIa-hChR2(H134R)-EYFP). We developed an analysis method based on the alignment of every NAc slice to areference mask followed by the regular division of the area in 50 µm squares (Fig. S4, see Materials andMethods). By this method, called “partition in regular quadrants” (PRQ), we rst obtained the spatialproles of mPFC afferents to the NAc (Fig. 2D). In agreement with our previous characterization(Fig. 2A2), PRQ analysis detected higher density of projections in the AcbC (1.88 ± 0.03 normalizedΔF/F0) compared to the AcbSh (0.98 ± 0.06 normalized ΔF/F0) (p < 0.001; n = 9 slices, 6 mice; Fig. 2D4),supporting PRQ as a useful method to analyze uorescence signals in broad areas. Secondly, wespatially analyzed NAc astrocytic responses to mPFC afferent optostimulation (Fig. 2E). To trigger the reduorescence turnover of CaMPARIGFAP, 40 s of UV was delivered after optogenetic stimulation.Optostimulation of mPFC afferents led to a signicant astrocytic photoconversion in both NAcsubregions, AcbC (1.64 ± 0.16 ΔF/F0 change from basal, p = 0.0035) and AcbSh (1.92 ± 0.19 ΔF/F0

change from basal, p = 0.0014) (n = 18 slices, 6 mice; Fig. 2E4), showing the same activation proles as

those registered by Ca2+ real-time imaging analysis (Fig. 2C3). In line with the recorded Ca2+ signals, thisphotoconversion was mediated by mGluR5 receptors, since both regions reduced their activity to basallevels in the presence of MPEP (0.97 ± 0.14 ΔF/F0 change from basal, p = 0.8354) (n = 8 slices, 2 mice;Fig. 2E4).

To further ascertain the spatial overlap between the afferents and the activated astrocytic area, weanalyzed these two PRQ spatial proles (mPFC innervation pattern (Fig. 2D) and NAc astrocytic activitypattern (Fig. 2E)), dening the active areas by k-mean clustering (Fig. S5, see Materials and Methods).Using this approach, we identied the activation threshold for each uorescence signal, generating onemask containing the area with increased glutamatergic innervation and another containing the NAc areashowing robust astrocytic activation. As shown in Fig. 2F, the overlapping region between these twomasks (16.2 ± 2.94% area of NAc) was signicantly smaller when compared to the afferents area (30.8 ±0.75% area of NAc, p = 0.04) or to the activated astrocytic regions (40.9 ± 6.66% area of NAc, p = 0.001)(Fig. 2F2), indicating a low rate of spatial overlap between both signals. Furthermore, we studied the

Page 6/36

overlapping degree and spatial distributions within the AcbC and AcbSh using a bivariate similarityindex32,33, and found that astrocytes in both subregions were interacting in the same way with theafferents, showing in both cases low overlapping rate (MANOVA, d = 0, p = 0.586; Fig. 2G).

Taken together, these results demonstrate that astrocytes from both NAc subregions do not respond inareas under the direct reach of the glutamatergic afferents, showing pathway-specic astrocyte activitytriggered by mPFC inputs.

Optostimulation of the Amyg reveals pathway-specic activity patterns in NAc astrocytes.

The NAc receives a broadly distributed glutamatergic projection originating in the Amyg that promotesreward-seeking behaviors34. To characterize the NAc astrocyte responses to Amyg inputs, the sameapproach was used as above. In agreement with previous studies26,31, histological characterization ofglutamatergic afferent patterns conrmed that both the AcbC and AcbSh were similarly innervated byAmyg (AcbC 0.37 ± 0.06 ΔF/F0, AcbSh 0.33 ± 0.06 ΔF/F0) (p = 0.63, n = 13 mice; Fig. 3A2). PRQ analysisof the spatial distribution of the afferents evidenced that although innervation was similar betweensubregions, the AcbC (1.34 ± 0.08 ΔF/F0 of change) was more innervated than the AcbSh (1.14 ± 0.05ΔF/F0 of change) (p = 0.04, n = 9 slices, 6 mice; Fig. 3D4). When studying neuronal responses to theoptostimulation of those axons, we registered a positive correlation (Pearson r = 0.57, p = 0.03) betweenthe amplitude of the EPSCs and the density of afferent projections in the recording area (Fig. 3B2).Accordingly, the EPSC amplitudes in each subregion were similar (AcbC neurons 99.6 ± 35.2 pA, n = 11cells, 3 mice; AcbSh neurons 79.4 ± 15.4 pA, n = 25 cells, 4 mice; p = 0.54, Fig. 3B3).

The study of astrocytic responses to optostimulation of those afferents showed that astrocytes from theAcbSh (2.04 ± 0.38 change from basal, p = 0.015) displayed higher activity compared to those from theAcbC (1.45 ± 0.21 change from basal, p = 0.049) (n = 25 slices, 7 mice; Fig. 3C3 and S3B). Theseresponses were mediated by mGluR5, as it was blocked by bath-application of MPEP (50 µM; 1.56 ± 0.39change from basal, p = 0.18; n = 5 slices, 2 mice; Fig. 3C3). Furthermore, spatial PRQ analysis showed thesame prole, with stronger astrocytic activation in the AcbSh (1.78 ± 0.21 ΔF/F0 change from basal, p =0.005) compared to the AcbC (1.49 ± 0.18 ΔF/F0 change from basal, p = 0.026) (n= 18 slices, 6 mice; Fig.

3E4). In line with Ca2+ imaging data, bath application of the mGluR5 antagonist reduced those responsesto basal levels (0.93 ± 0.10 ΔF/F0 change from basal, p = 0.5) (n= 8 slices, 2 mice; Fig. 3E4). Like for themPFC, these results showed that Amyg optostimulation gave rise to different NAc astrocyte activationproles than the Amyg glutamatergic innervation pattern.

Afterward, we compared the area of overlap, calculating masks containing the spatial innervation patternof Amyg afferents and the activation pattern of NAc astrocytes 35. We found that the areas withincreased astrocytic activity in response to Amyg optostimulation (34.5 ± 5.34% area from NAc, p =0.003) did not spatially match the areas containing more density of glutamatergic projections (36.8 ±1.32% area from NAc, p < 0.001), as shown by the low rate of overlap between them (17.5 ± 2.05% area

Page 7/36

from NAc) (n= 9 slices, 6 mice; Fig. 3F2). Interestingly, astrocytes displayed bigger responses in dorsalregions of the NAc, while axons preferentially innervated ventral areas of the nucleus (Fig. 3F1). We didnot observe signicant differences in this interaction between AcbC and AcbSh astrocytic responses(MANOVA, d = 0, p = 0.482; Fig. 3G). Altogether, these results characterize the neural-astrocyte circuit ofthe Amyg-NAc pathway, showing different astrocytic activity patterns than the innervation ones, withastrocytes in the dorsal areas of the NAc with a preferentially low degree of innervation showing higheractivity.

Astrocytes from the NAc show pathway-specic activation motifs in response to vHip glutamatergicafferents.

After characterizing astrocytic activity in response to the Amyg and mPFC, we selectively activatedanother principal glutamatergic input implicated in anxiety-like behavior and social interaction: the vHip.Using the same optogenetic approach, the histological evidence showed stronger innervation comingfrom the vHip to the AcbSh (0.80 ± 0.15 ΔF/F0) with respect to the AcbC (0.31 ± 0.06 ΔF/F0) (p = 0.01; n =6 mice; Fig. 4A2). Functional characterization of these afferents showed a positive correlation (Pearson r= 0.56, p = 0.005; Fig. 4B2), as observed on the other two glutamatergic projection nuclei (mPFC andAmyg). Neurons in the NAc displayed bigger EPSC events in areas with higher density of vHip axons, andthe synaptic neuronal strength was substantially stronger in the AcbSh (303.4 ± 62.5 pA; n = 21 cells, 6mice) than in the AcbC (24.1 ± 14 pA; n = 4 cells, 3 mice; p < 0.001; Fig. 4B3). These results conrm thatthe EPSC amplitude was dependent on the density of glutamatergic bers innervating each MSN, being acommon characteristic of the three glutamatergic pathways.

Next we studied the astrocytic responses to those afferents. Real-time imaging from astrocytes in theAcbC showed increased frequency (2.6 ± 0.63% of change from basal, p = 0.018) of Ca2+ spikes inresponse to optostimulation with respect to astrocytes from the AcbSh (1.51 ± 0.21 change from basal, p= 0.031) (n = 13 slices, 5 mice) (Fig. 4C3 and S3C). As discovered for the other pathways, NAc astrocytesdid not respond to vHip optostimulation with the same activation prole as the one displayed by theafferents. The astrocytic activity pattern was stronger in the AcbC, in opposition to the innervation prolein which the AcbSh gathered the majority of incoming afferents (Figs. 4A2 and 4C3). As we found forother glutamatergic inputs, astrocytic responses triggered by vHip axons were mediated by mGluR5receptors (MPEP, 50 µM; 0.66 ± 0.51 change from basal, p = 0.055; n = 6 slices, 2 mice; Fig. 4C3). Theseresults demonstrated that although NAc astrocytes respond differently depending on the pathway, theglutamate-dependent activity in the nucleus is, in all cases, mediated mainly by mGluR5 receptors.

Analysis of the spatial prole of astrocytic activation using PRQ conrmed the strong astrocyticresponses in the AcbC (2.46 ± 0.32 ΔF/F0 change from basal, p = 0.003) and AcbSh (2.05 ± 0.20 ΔF/F0

change from basal, p = 0.0012; n= 16 slices; 6 mice), which were blocked in presence of mGluR5antagonist (MPEP, 50 µM; 1 ± 0.23 ΔF/F0 change from basal, p = 0.99; n = 8 slices, 2 mice; Fig. 4E).Overall, the spatial location of astrocytes activated by the vHip glutamatergic afferents differed from the

Page 8/36

ones triggered by the other glutamatergic pathways (mPFC and Amyg), suggesting that astrocytes in theNAc respond differently to the diverse glutamatergic nuclei showing synapse-specic activity.

Interestingly, in contrast to the results found after stimulation of the mPFC and Amyg, PRQ analysis ofspatial overlap (23.6 ± 2.50% area of NAc) between glutamatergic afferents from vHip (34.6 ± 1.51% areaof NAc, p = 0.12) and astrocytic activation (46.8 ± 6.97% area of NAc, p = 0.003) showed colocalization inthe AcbSh (n= 8 slices, 6 mice; Fig. 4F2). This interaction was present in the AcbSh sub-region but not inthe AcbC (MANOVA, d = 1, p = 0.015; Fig. 4G).

Astrocytes in the NAc show synapse-specicity in response to principal glutamatergic nuclei inputs.

Next, we focused on NAc-intrinsic properties that could impact the previously recorded astrocytic calciumresponses, since the integration of glutamatergic inputs in the NAc is an essential operation for diversefunctions (e.g., reward, motivation and goal-directed behaviors) 9,26,36. First, we conducted a structuralanalysis of astrocytic density, followed by analysis of the Ca2+ signaling evoked by a group I mGluRagonist (DHPG, 50 µM) (Fig. S6). No signicant differences were detected, neither in density (p = 0.61)nor spatial distribution (p = 0.84) of astrocytes, which showed a homogeneous dispersion across thewhole nucleus (Fig. S6A2 and S6B3). Furthermore, the astrocyte response to DHPG bath applicationshowed no differences across subregions (p = 0.64; Fig. S6C3), dismissing the existence of specic areasin the nucleus more sensitive to mGluR activation.

Next we explored the differences among the three glutamatergic pathways, analyzing the anatomicallocation of each astrocytic activity pattern and comparing the strength of astrocytic activation betweenpathways (Fig. 5). Using a spatial correlation test (Fig. 5B, see Materials and Methods), we observedpositive spatial correlations; mPFC vs Amyg (Pearson r = 0.5122, p < 0.001), Amyg vs vHip (Pearson r =0.5877, p < 0.001), mPFC vs vHip (Pearson r = 0.705, p < 0.001) indicating that the three neuron-astrocytecircuitry responses encompass similar areas of the NAc. When focusing on the intensity of the astrocyticresponses (Fig. 5C), we detected that vHip afferents signicantly triggered stronger activity in the AcbCcompared to the mPFC or Amyg (one-way ANOVA, p = 0.03), while in the AcbSh, the three projectionnuclei activated NAc astrocytes to a similar degree (one-way ANOVA, p = 0.36). This result demonstratesa different level of astrocytic activation depending on the pathway, showing synapse-specicity andsupporting the notion that NAc astrocytes can discern the origin from each of the glutamatergic inputs.

Moreover, the differences in AcbC and AcbSh astrocytic responses for each pathway (Fig. 5C2) suggestedthe existence of intrinsic processing for each NAc region. To further explore this idea and to understandthe molecular mechanisms that underlie the physiological heterogeneity between astrocyte Ca2+ activityfrom the AcbC and AcbSh (Fig. 6A and B), we measured the number of copies of mitochondrial genomesin single astrocytes from each region. Mitochondrial DNA (mtDNA) encodes essential genes for energygeneration and is a key regulator of mitochondrial function. Mitochondria mediate energy supply andmetabolism37 and have been shown to regulate intracellular Ca2+ dynamics in astrocyticmicrodomains38. We measured mtDNA copy number using a novel approach based on single-cell mtDNA

Page 9/36

amplication by droplet digital PCR (patch-dPCR)39. Unlike massively parallel sequencing techniques thatprovide relative DNA quantication based on a control DNA or number of reads, patch-dPCR allowsaccurate quantication of absolute values of DNA copies at single-cell resolution39. The mitochondrialDNA quantication for AcbSh (103.1 ± 14.9 mtDNA copies/cell; n = 32 cells, 11 mice) and AcbC (163.1 ±32.9 mtDNA copies/cell; n = 32 cells, 11 mice) astrocytes revealed higher mtDNA copy number in AcbCwith respect to AcbSh (p < 0.04). This difference was also present comparing AcbC (p < 0.001) astrocyteswith respect to control (in which the pipette was internalized into the bath and no mtDNA collection wasperformed) compared to AcbSh (p = 0.004) (Figure 6B2), suggesting molecular heterogeneity between the

astrocytes from both sub-regions. These results, in concordance with the Ca2+ signaling observations(Figure 6A), reveal different functional dynamics among sub-regions, indicating an increased activitycoupled with higher mtDNA copy number in AcbC astrocytes.

Overall, these results demonstrate that NAc astrocytes respond specically to each glutamatergic inputwith synapse-specicity, suggesting physiological and molecular heterogeneity between AcbC and AcbShastrocytes.

Pathway-specic neuron-astrocyte networks in NAc circuitries

The above results showed tethered spatial overlap between high-density afferent areas and regions withan increased evoked astrocytic activity (Figs. 2F, 3F and 4F). To further ascertain these regions, wecompared the astrocyte’s interaction with the afferents in each of the three pathways using a pixel-by-pixel spatial correlation analysis (Fig. 7A): mPFC vs Amyg (Pearson r = 0.0985, p = 0.009) and Amyg vsvHip (Pearson r = 0.3158, p < 0.001) showed no correlation, whereas mPFC vs vHip (Pearson r = -0.2866,p < 0.001) showed a negative correlation. We conclude that the interaction between astrocytes andafferents showed spatial segregation, which could point to the existence of hotspots in the NAc for eachglutamatergic input.

Furthermore, we wanted to ascertain whether the NAc astrocytes exhibit these pathway-specic featuresin response to other neurotransmitters rather than the glutamatergic system, so we focused on the ventraltegmental area (VTA) inputs to the NAc (Fig. S5). The VTA contains different neurons that co-releasedopamine and glutamate, but it has been classically described as a predominantly dopaminergicnucleus40. Innervation of this nucleus to the NAc was homogeneous across subregions, as shown by theanalysis of afferent’s histology (0.46 ± 0.1 and 0.54 ± 0.11 ΔF/F0 in the AcbC and AcbSh respectively, p =

0.6; Fig. S7A3). Afterward, we registered the astrocytic Ca2+ dynamics in response to optostimulation ofVTA afferents in control conditions and after bath perfusion of dopamine receptor antagonists (10 µMhaloperidol and 10 µM SCH 23390; Fig. 7B). In agreement with previous studies which showed that NAcastrocytes respond to dopamine released from the VTA27, the optostimulation of these axons increasedthe frequency of Ca2+ spikes in both NAc subregions AcbC (2.06 ± 0.2, change from basal; p = 0.001) andAcbSh (2.04 ± 0.23, change from basal; p = 0.003) (n = 8 slices, 5 mice) which was blocked in presence ofantagonists (0.99 ± 0.1, change from basal; p = 0.89; n = 9 slices, 2 mice) (Fig. S7B2), conrming that this

Page 10/36

neuron-astrocyte communication was mediated by dopamine. Using CaMPARIGFAP photoconversion andPRQ spatial analysis, we observed increased calcium activity in wide areas of the NAc (n = 6 slices, 3mice; Fig. S7D4) and spatial analysis revealed that NAc astrocytes respond in areas directly innervated byVTA axons (p = 0.07, Fig. S7E). In fact, when the % overlap area was compared between pathways (p <0.001, Fig. 7B2), our data show that restricted interaction between intense astrocytic Ca2+ activity andhighly innervated areas of the NAc is an intrinsic feature of glutamatergic transmission not present indopaminergic afferents.

Simultaneous activation of different glutamatergic pathways reveals integrative properties of NAcastrocytes.

Finally, we investigated how incoming synaptic signals from various glutamatergic pathways wereintegrated by NAc astrocytes. The ability to integrate multiple synaptic inputs from different entries is afundamental property of neurons25. Similarly for individual astrocytes, there is evidence of synaptic inputintegration through Ca2+ spike mediation25,41. To test for the existence of integrative capacities inastrocytes of the NAc, we studied Ca2+ responses triggered by the combined stimulation of two or threeglutamatergic pathways simultaneously. For that, we expressed the ChR2-EYFP opsin in combinations of2 or 3 glutamatergic nuclei, stimulated all the NAc projecting bers and measured the astrocytic responsethrough CaMPARIGFAP photoconversion in the AcbC and AcbSh (Fig. 7).

Co-stimulation of the mPFC and Amyg afferents did not trigger large astrocytic activity in the AcbC (1.23± 0.26 ΔF/F0 change from basal, p = 0.42) nor in AcbSh (1.18 ± 0.09 ΔF/F0 change from basal, p = 0.11),the latter being signicantly lower than the individual responses (p = 0.05; n= 10 slices, 3 mice; Fig. 7A3).A similar outcome resulted from the co-stimulation of Amyg and vHip axons (Fig. 7B3), with recordedvalues close to baseline in the AcbC (1.36 ± 0.32 ΔF/F0 change from basal, p = 0.3) and AcbSh (1.14 ±0.22 ΔF/F0 change from basal, p = 0.55; n= 14 slices, 4 mice). Therefore, we found that co-activation of

mPFC+Amyg or Amyg+vHip, resulted in weaker responses of astrocytic Ca2+ signaling compared tostimulation of single pathways. Conversely, co-stimulation of the mPFC and vHip afferents (Fig. 7C)induced strong astrocytic activity in both the AcbC (2.11 ± 0.4 ΔF/F0 change from basal, p = 0.039) and

AcbSh (2.24 ± 0.47 ΔF/F0 change from basal, p = 0.046), resulting in a similar Ca2+ activity as thattriggered by individual pathway optostimulation (AcbC, p = 0.12; AcbSh, p = 0.73; n= 12 slices, 5 mice;Fig. 7C3). These different outcomes in astrocytic dynamics depending on the combination ofglutamatergic inputs conrm that astrocytes are able to differentially respond according to the pathwayfrom which the inputs are received.

Moreover, when all three pathways were stimulated simultaneously (mPFC + Amyg + vHip), we did notnd any increase from basal levels in astrocytes of the AcbC (1.03 ± 0.16 ΔF/F0 change from basal, p =0.84), and remarkably, astrocytes from the AcbSh reduced their activity even below basal values (0.69 ±0.09 ΔF/F0 change from basal, p = 0.018; n = 12 slices, 3 mice; Fig. 7D3). These differences in Ca2+

Page 11/36

processing between subregions further suggest the existence of neuron-astrocyte circuitries that work tocoordinate responses to glutamatergic afferents.

DiscussionNumerous studies have demonstrated distinct functional and structural domains within the NAc9,42−45.Thus, two primary subregions have been identied that occupy the dorsolateral and ventromedialportions of the NAc, known as the AcbC and AcbSh, respectively. Their different roles are commonlyviewed as stemming from differences in neural circuitry between the regions. Indeed, these roles havetraditionally been studied together with structural connectivity patterns determined by axonal tracers9,46.By using a newly developed and highly specic CaMPARIGFAP approach, our data reveal for the rst timethe existence of synapse-specic neuron-astrocyte circuitries within the NAc.

Simultaneous imaging of large numbers of astrocytes is required for higher spatial and temporal imagingstudies, which makes CaMPARIGFAP a powerful tool for unraveling the dynamics of the array ofastrocytes and their relationship with the circuits to which they belong. Using this new approach, we havepresented evidence supporting the existence of unique neuron-astrocyte circuits in each NAc region,which receive and integrate glutamatergic innervation differentially from the mPFC, Amyg, and vHip. Ourresults reveal a functional and mtDNA copy number heterogeneity between the two subpopulations thatleads to differential input integration of the excitatory pathways. Whether these different outcomes of theAcbSh and AcbC Ca2+ dynamics are due to particular astrocyte subpopulation heterogeneity, specicproperties of the glutamatergic afferents, the total amount of glutamate released into the NAc by thereported pathways, or rather a combination of these, must be further explored. Moreover, considering thecritical role of astrocytes for synaptic function, the present results support the idea that different neuron-astrocyte circuitries are associated with discernible function in NAc; i.e., an astrocyte subpopulation fromthe AcbSh could be related to the limbic system and another from the AcbC could be preferentially linkedto the motor systems.

Our results suggest that NAc astrocytes respond to excitatory inputs in a pathway-specic way, sinceoptostimulation of mPFC inputs triggered astrocytic activation in both NAc subregions. (Fig. 2), whileastrocytes responding to the Amyg seemed to be preferentially concealed within dorsal regions of theNAc (Fig. 3). Further, vHip and VTA activated a broad astrocyte network covering most of the NAc (Figs. 4and S7). Likewise, we observed spatial differences in the way astrocytes interact with highly denseafferent areas among the pathways 8,47−49. Moreover, astrocytes did not appear to respond solely inhighly innervated regions, which is a common characteristic of each discrete glutamatergic pathway.However, we do not know the implications of activating a prole of astrocytes preferentially in less-innervated areas in response to glutamatergic stimuli, as the existence of networks within a system easesits capacity to autoregulate. Since the functions of astrocytes are mainly homeostatic, it seems plausiblethat autoregulation of astrocytic proles favors communication under dynamic equilibrium50.

Page 12/36

This picture of several structurally independent networks entangled points to the existence of a modularsystem, in which modules connect internally with each other to achieve robustness in the integration andrelease of signals. Modules may be useful when outputs change over time, since reconnections amongalready existing ones remove the need to form them from scratch51,52, adding complexity to the networkstructure with features such as plasticity, capacity to integrate or adaptability53. Additionally, modularitymay provide a means to nely achieve tuned temporal (such as synchronization) and spatial resolutionof signals (discerning among inputs to integrate outputs accordingly). For a holistic understanding of thecircuits that mediate diverse behaviors in the NAc, it would probably be necessary to understand themechanisms by which astrocytes integrate information.

The NAc integrates convergent glutamatergic inputs and has been proposed that the membraneproperties of the GABAergic MSN (the primary neurons of the NAc) are responsible for mediating thisafferent integration9,22,23,26,31,54,55. However, although our results suggest that NAc astrocytes respond tothe excitatory inputs in a pathway-specic way, we did not detect individual-specic clusters ofphotoconversion for each afferent, supporting our hypothesis that astrocytes share and integrateinformation from different glutamatergic inputs. Furthermore, the fact that co-stimulated responsesshowed non-linear activity when compared to individual ones, meaning that the resulting co-stimulatedresponse is not the sum of the individual responses, indicates that astrocytes are able to performcomplex computations on the information coming from different sources. This lack of linearity showsthat the integration property orchestrated by astrocytes in the NAc could mechanistically explain thedivergent physiological and behavioral responses produced by the activation of different glutamatergicinputs to the NAc, revealing the neuron-astrocyte network as a critical center for the integration propertiesof the NAc 9,22,23,26,31,54,55.

Taken together, the results presented here demonstrate the value of creating comprehensive functionalastrocyte-network maps, providing a potential explanation for how the NAc integrates information frommultiple glutamatergic regions.

Materials And MethodsMice. All the biosafety procedures for handling and sacricing animals were approved by the bioethicscommittee from the Consejo Superior de Investigaciones Cienticas (CSIC), and followed the EuropeanCommission guidelines for the welfare of experimental animals (2010/63/EU, 86/609/EEC). C57BL/6wild-type mice of both genders were used, and were housed in standard laboratory cages with ad libitumaccess to food and water, under a 12:12 h dark-light cycle in temperature-controlled rooms.

CaMPARI GFAP viral vector. pAAV-hSyn-CaMPARI-WPRE (Ca2+-modulated photoactivatable ratiometricintegrator) plasmid was acquired from UNC Vector Core. The hSyn promoter was substituted by the GFAPpromoter short version GFAP104 and cloned into AAV particles by Unitat de Producció de Vectors (UPV)(AAV2/9-GFAP104-CaMPARI; viral titer 1.05 × 1013 gc/ml).

Page 13/36

Viral injection. The following constructs were used: AAV5/CaMKIIahChR2 (H134R)-mcherry, viral titer 5.6x 1012 gc/ml; pAAV-Syn-ChrimsonR-tdTomato, viral titer 4.1 x 1012 gc/ml; AAV-CaMKIIa-hChR2(H134R)-EYFP, viral titer 6.1 x 1012 gc/ml; AAV2/9-GFAP104-CaMPARI; viral titer 1.05 × 1013 gc/ml. P15-P20 micewere anesthetized with 2% isourane in oxygen and placed in a custom adapted stereotaxic apparatus.Stereotaxic bilateral injections (300-350 nl; 60 nl/min) were made in NAc (AP: 1.45 mm ML: ± 0.65 mmDV: - 4.3 mm) and/or mPFC (AP: 1.7 mm ML: ± 0.3 mm DV: - 2.3 mm), Amyg (AP: -1.3 mm ML: ± 2.95mm DV: - 4.4 mm), vHip (AP: -3.1 mm ML: ± 2.95 mm DV: - 4.2 mm) and VTA (AP: -3 mm ML: ± 0.5 mmDV: - 4.3 mm). After injection, the pipette was held in place for 5 min prior to retraction to avoid leakage,then removed and skin sutured. Animals were allowed to recover from anesthesia with the help of heatingpads and returned to the cage once they showed regular breathing and locomotion. 3-4 weeks after viralinjection, the location of the virus was conrmed on the basis of CaMPARIGFAP, mCherry, tdTom and EYFPexpression (Figs. 2A, 3A, 4A and S7 upper panels).

Slice preparation. Experiments were performed on slices from mice (1-3 months, both sexes) unlessotherwise indicated. 3-4 weeks after viral infection, animals were anaesthetized and prior to decapitation,transcardially perfused with ice-cold protective articial cerebrospinal uid (NMDG ACSF) containing (inmM): N-methyl-D-glucamine (NMDG) 93, KCl 2.5, NaH2PO4 1.25, NaHCO3 30, HEPES 20, glucose 25,thiourea 2, Na-ascorbate 5, Na-pyruvate 3, CaCl2 0.5, MgCl2 10. The brain was rapidly removed andplaced in ice-cold NMDG ACSF, and slices (300 µm thick) were obtained with a vibratome (LeicaVibratome VT1200S, Germany) and maintained at 34 ºC in NMDG ACSF for 10 min and transferred for >1h incubation at room temperature (RT) in standard ACSF containing (in mM): NaCl 119, KCl 2.5, NaH2PO4

1, MgCl2 1.2, NaHCO3 26, CaCl2 2.5, and glucose 11, and gassed with 95% O2/5% CO2 (pH = 7.3). Slicescontaining the NAc region were placed into the microscope recording chamber superfused with gassedACSF. Unless otherwise indicated, ACSF was supplemented with 0.05 mM picrotoxin to block GABAA

receptors.

For some experiments, for anatomical density of projection afferents (Figs. 2A2, 3A2, 4A2 and S7A3) andcontrol viral infection measurements, injection site and NAc slices were kept and xed with 4%paraformaldehyde (PFA), 4% sucrose in PBS for 1 h at RT. After xation, slices were incubated with DAPI(1.5 µg/ml, Sigma-Aldrich) for 10-15 min and mounted in Vectashield antifading mounting medium(Vector Laboratories, Burlingame, CA). Fluorescence images were acquired with a Leica SP-5 invertedconfocal microscope using Leica LAS AF software. Image analysis was carried out with Image Jsoftware (public domain software developed at the US National Institutes of Health [NIH]). AcbC andAcbSh regions of interest (ROIs) were delimited manually, and average uorescence inside the regionswas expressed as the absolute change (|∆F|/F0|) between the averaged ROI value (F) and a backgroundROI (F0) set in reference point of the tissue without afferent's uorescent signal.

Electrophysiology. Recordings from NAc medium spiny neurons (MSNs) were made using the whole-cellpatch-clamp technique. Patch electrodes had resistances of 4-10 MΩ when lled with an internal solutionthat contained (in mM): potassium gluconate 135, KCl 10, HEPES 10, MgCl2 1, ATP-Na2 2, titrated with

Page 14/36

KOH to pH 7.3. Recordings were obtained with PC-ONE ampliers (Dagan) and pClamp software(Molecular Devices). MSNs were voltage clamped at -75 mV, unless otherwise noted. Electrophysiologicalproperties were monitored before and at the end of the experiments. Series and input resistances weremonitored throughout the experiment using a −1 mV pulse. Recordings were considered stable when theseries and input resistances and stimulus artifact duration did not change >20%. Cells that did not meetthese criteria were discarded. Cells were visualized on an Olympus BX50WI microscope (Olympus Optical,Tokyo, Japan) under a 40x water immersion objective.

Optogenetic stimulation. Optically evoked excitatory postsynaptic currents (EPSCs) were obtained every 5s with light pulses (0.1 – 1 W /cm2) using CoolLED’s pE-300white system through the microscopeobjective for full-eld optostimulation. Light stimulation strength and pulse duration remained constantfor all the recordings (70% LED intensity, 1 ms) in order to compare evoked EPSC responses between NAcsubregions and different glutamatergic afferents (Figs. 2B, 3B and 4B). Two different wavelengths wereused, 470 nm and 565 nm, to stimulate afferents expressing ChR2 or ChrimsonR opsins, respectively.Dose-response curves (Fig. S2) were calculated measuring evoked EPSC amplitudes at different lightintensities (10, 30, 50, 70 and 90% LED intensity).

Ca 2+ imaging. Ca2+ levels in astrocytes located in the NAc were monitored by uorescence microscopyusing the genetically encoded Ca2+ indicator CaMPARIGFAP. Astrocytes were imaged with a CCD camera(Hamamatsu C474-95) attached to an Olympus BX50WI microscope (Olympus Optical, Tokyo, Japan),coupled with a 10x water immersion objective. Cells were illuminated at 300 ms exposure with a LED at470 nm using CoolLED’s pE-300white system through the microscope objective, and images wereacquired every 1 s at RT. The CoolLED and CCD camera were controlled and synchronized by Nis-Elements Advance Research software (Nikon Instruments Europe B.V.). After 3 min basal activityrecording, optostimulation of glutamatergic afferents was full-eld delivered at a xed intensity (1 - 5W/cm2) by an external LED (Thorlabs M590F3, λ = 590 nm), followed by 3 min recordings of theastrocytic responses.

Preliminary analysis was carried out using Image J Software (public domain software developed at theUS NIH). Minor drift in the XY plane of image stacks was post hoc corrected using TurboReg (Image Jplugin). For each slice recorded, 50 ROIs of 15 µm2 were selected for analysis each in the AcbC andAcbSh, and raw uorescence values within each ROI were acquired. MATLAB software (R2018a;Mathworks, Natick, MA, USA) was used to analyze Ca2+ parameters from the raw values. The rst 5 s ofthe ROI signals was removed (corresponding to highly non-linear bleaching) and then the signals werelow-pass ltered (Chebyshev type II). Each signal was discretized in segments of a xed time length, andthe 80th percentile of each segment was found. A line was tted to the found points and was dened asthe basal line (F0). Each point of each ROI was expressed as the absolute change (|∆F|/F0|) between theROI point and the corresponding point in the basal line of that ROI.

An event threshold (event_th) for each ROI was calculated as

Page 15/36

event_thp=3n∑i=2n|xpi−xpi−1|

where xpi is the ith point of the pth ROI signal. Any event threshold below 0.004 (0.4% change) wasautomatically set to that value to avoid false positive in silent signals due to stochastic noise. Eventpeaks were dened as the local maxima that are above the given event threshold with a prominence of atleast 20% the threshold value, and that are separated from the peak of any other event for at least 3 s.MATLAB’s built-in function “ndpeaks” was used for the event detection step. Parameters of % response,frequency and amplitude were measured for every ROI. Each stimulated slice response was compared toits basal activity control recording.

CaMPARI GFAP photoconversion. NAc slices (300-µm thick) with astrocytes expressing CaMPARIGFAP andglutamatergic bers expressing ChR2-EYFP were placed into a microscope recording chamber (OlympusBX50WI; Olympus Optical, Tokyo, Japan) superfused with gassed ACSF in presence of picrotoxin (0.05mM). After 5 min to allow picrotoxin to act on the GABAA receptors, light stimulation protocols wereapplied: optogenetic stimulation followed by 40 s UV (λ = 405 nm) light or basal condition, in which only40 s UV light were applied. Full-eld optogenetic stimulation was delivered using a 10x microscopeobjective. UV light was delivered obliquely at a xed intensity (1 - 5 W/cm2) by an external LED (ThorlabsM405F1). Since photoconversion turnover is linearly dependent on UV light intensity 27,28, all slices wereplaced at the same distance and orientation to the light beam in order to avoid variation in the scatteringpattern. NAc slices at the coordinates AP +1.3 and AP +0.98 were considered, alternating UV light basal orUV light TBS protocol AP coordinates in different hemispheres.

Slices were xed with 4% paraformaldehyde (PFA), 4% sucrose in PBS for 1 h at RT. After xation, sliceswere incubated with DAPI (1.5 µg/ml, Sigma-Aldrich) for 10-15 min and mounted in Vectashieldantifading mounting medium (Vector Laboratories, Burlingame, CA). Fluorescence images were acquiredwith a Leica SP-5 inverted confocal microscope using Leica LAS AF software. Z-stack (10 plane / 10 µmthickness) mosaics were collected for the three uorescence signals: CaMPARIGFAP Green (488 nmexcitation laser), EYFP (510 nm excitation laser) and CaMPARIGFAP Red (561 nm excitation laser). Tocompare uorescence intensity between different samples, all images were acquired under identicalconditions. Since photoconversion turnover is robust in astrocytes, the decrease of CaMPARIGFAP

uorescence after chemical xation 28,30 did not compromise red signal detection, and no posteriorimmunostaining was needed.

Image pre-processing was performed in the following manner: all images were registered to a referencemask using bUnwarpJ (Image J plugin) to align the same anatomical location across them. MATLABsoftware (R2018a; Mathworks, Natick, MA, USA) was used for uorescence analysis. An average imagefrom each stack was calculated and images were divided into 50 µm x 50 µm square grids, and each gridwas assigned a value equal to the mean uorescence signal inside it. A reference autouorescence signalfor each grid was computed as the mean signal of that grid in control images (n = 9 slices, 3 mice) inwhich no virus was injected. This reference was subtracted from the images analyzed to account for the

Page 16/36

heterogeneity in the autouorescence signal across the NAc. Moreover, each image was normalized to itsbackground signal computed from an ROI located outside of the NAc. That is,

Gi , j =∑ i , jF − Ri , j

F0

where Gi,j is the value assigned to grid [i,j], ∑ i , jF corresponds to the mean uorescence signal inside thegrid, Ri,j is the reference autouorescence signal of grid [i,j], and F0 is the background signal outside theNAc of the image (Fig. S4). Lastly, to compare basal-optostimulated CaMPARIGFAP pairs of images, eachpair was normalized to the mean uorescence in the NAc in the basal condition.

An activation threshold based on uorescence signal was used to create binary activation masks fromthe images (CaMPARIGFAP Red and glutamatergic or dopaminergic afferents) by dening pixels as active(1, ≥ threshold) or inactive (0, < threshold). The activation thresholds were determined automatically by ak-mean clustering method using k=5 clusters for glutamatergic and dopaminergic afferents, and k=6clusters for CaMPARIGFAP Red images (Fig. S5). In each stimulated slice, afferent-astrocyte pairedactivation masks were used to determine spatial colocalization (% overlap; Figs. 2F, 3F, 4F and S7E. X-and y-axis include the area relationship between overlap/astrocytes and overlap/glutamatergic afferents,respectively; values close to 0 indicate no overlap while values close to 1 indicate that all of the afferent'sarea overlaps. Bivariate representation of the two allows dissection of the spatial distribution of theseoverlapped areas; values close to the diagonal (45º) indicate that the overlap area is located equallybetween afferents and astrocytes areas, while values close to the x or y axis indicates that colocalizationis embedded respectively within one of the two. Multivariate analysis of variance (MANOVA) was used todetermine differences between AcbC and AcbSh distributions.

To study spatial pattern differences among the distinct unpaired experimental conditions, a spatialcorrelation approach was used (Fig. 5). For astrocytic activity analysis, only pixels above the activationthreshold were considered for each individual slice. Using the ltered data, average PRQ imagescontaining spatial information of the active regions were generated (Fig. 5A). For overlap domainanalysis, binary masks containing colocalization pixels between glutamatergic afferents and astrocyticactive masks were used for each individual slice. Afterwards, an overlap average PRQ image for eachexperimental condition was generated (Fig. 5B). Pixel-by-pixel Pearson correlation was used to analyzesimilarities between pairs of average PRQ images in both analyses.

Immunohistochemistry. C57BL/6 wild-type mice transfected with CaMPARIGFAP viral vector in the NAcwere euthanized with sodium pentobarbital and transcardially perfused with PBS followed by ice-cold 4%paraformaldehyde (PFA) and 4% sucrose in PBS. Brains were removed and postxed overnight (o/n) inthe same xative solution. Coronal brain slices of 50 µm were obtained in a VT100S vibratome (Leica)and collected as oating sections. Slices were permeabilized with 1% Triton in PBS, and non-specicbinding was blocked for 2 h at room temperature (RT) with 0.3% goat serum, 0.1% Triton in PBS. Afterblocking, sections were incubated with the corresponding primary antibodies at 4 oC o/n, and with

Page 17/36

secondary antibodies for 1 h at RT. The primary antibodies used were rabbit anti-S100 (1:200 dilution;ab868, Abcam) and mouse anti-NeuN (1:500; MAB377, Merck). Secondary antibodies were anti-mouseAlexa 647 (1:1000; A21236, Thermo Fisher Scientic) and goat anti-rabbit Alexa 405 (1:1000, A31556,Thermo Fisher Scientic). Finally, sections were washed with 0.1% Triton in PBS and mounted forimaging in Vectashield antifading mounting medium (Vector Laboratories, Burlingame, CA).

Fluorescence images were acquired with a Leica SP-5 inverted confocal microscope using Leica LAS AFsoftware. Image analysis was carried out with Image J software (public domain software developed atthe US NIH).

Patch-dPCR single-cell. C57BL/6 wild-type mice (P20 – P30) with no viral infection were used. Tofacilitate astrocyte detection, cells were stained with an astrocytic marker, the red uorescence dyesulforhodamine (SR101). NAc slices (300 µm thick) were incubated with SR101 at low concentration (0.5µM) at 34 oC for 20 min in ACSF, and then washed 10 min at 34 oC in ACSF 56. After incubation, sliceswere transferred to an immersion recording chamber and perfused with normal gassed ACSF.

Recordings from NAc astrocytes were made using the whole-cell patch-dPCR technique. The patch pipettewas lled with minimum volume (xed at 1.5 µl) of internal solution to prevent loss of sample DNAduring extraction. Astrocyte identication was conrmed by their electrophysiological properties, showingno electrical excitability when depolarizing current steps were applied (Fig. 6B). Depolarizing steps weremaintained while applying negative pressure until major soma contents were suctioned into the patch-pipette. Tip content was ejected into an Eppendorf tube containing 4 µl buffer lysis solution (100T DNA/RNA/Protein Solubilization Reagent #DCQ100T, DireCtQuant, Lleida, Spain). Each sample was incubatedat 90 oC for 3 min, cooled to room temperature and centrifuged for 10 min at 10,000 rcf before storing at-20 oC. mtDNA was amplied following a modication of a previously described dPCR procedure 39.Briey, mtDNA copy number was measured simultaneously for each sample in a multiplex assay withtwo different primer pairs: mmt60-ND4, which targets a region of the ND4 gene corresponding to bases11161-11220; and mmt67-ND5, which targets a region of the ND5 gene corresponding to bases 12216-12282 of the mouse mtDNA NC_005089.1 reference sequence. These two primer pairs target adjacentregions and produce amplicons of 60 and 67 base-pairs in length, respectively were chosen to amplifymtDNA molecules without deletions. The sequences of the primers are: mmt60-ND4, forward 5’-CCAACTACGAACGGATCCACA-3’, reverse 5’-AGACCATTTGAAGTCCTCGGG-3’; mmt67-ND5, forward 5’-TACGGACGAACAGACGCAAA-3’, reverse 5’ CGATGTCTCCGATGCGGTTA-3’. Amplication specicity andabsence of non-specic amplication were also veried by agarose gel electrophoresis. Analyses wereperformed using amplitude multiplex with EvaGreen on a QX200 Digital Droplet PCR platform (Bio-RadLaboratories, RRID:SCR_008426). The dPCR reaction consisted of 1X QX200™ ddPCR™ EvaGreenSupermix (1864033, Bio-Rad, RRID:SCR_008426), 1.3 µl of sample and primer concentrations of 120 nMof mmt67-ND5 and 80 nM of mmt60-ND4 DNA. Restriction enzyme digestion was performed with 1U ofFast Digest AluI enzyme for 15 min at 37 ºC before partition in droplets. Non-template controls wereincluded in each analysis plate to monitor possible reaction contamination. Data analysis was performedwith QuantaSoft Analysis Pro v1.0 using thresholds to distinguish single and double-positive droplet

Page 18/36

populations. PCR amplication was performed using the following thermal cycling program: 95 ºC for 5min; 40 repeats of 95 ºC 30 s, 60 ºC 1 min; 4 ºC 5 min; 90 ºC 10 min using a C1000 Touch Thermal Cycler(Bio-Rad). The total number of mtDNA copies per cell was calculated by adding the number of copies ofmmt60-ND4 plus mmt67-ND5 and dividing the result by two.

Statistical analysis. All data are expressed as mean ± SEM. Statistical analysis of the differencesbetween groups were determined by two-tailed unpaired or paired t-test, one-way ANOVA or two-wayANOVA followed by Holm-Sidak test for multiple comparisons, unless otherwise indicated. Differenceswere considered to be signicantly different when p < 0.05. Signicant changes from basal, in the case ofnormalized data (basal = 1), were determined by one-sample t-test. Correlation analysis was performedby Pearson's test. Statistical differences among XY distributions were determined by multivariate analysisof variance (MANOVA). Statistical differences were calculated using GraphPad Prism 7 and MATLAB(R2018a; Mathworks, Natick, MA, USA).

DeclarationsAcknowledgments

We thank Alfonso Araque, Eduardo Martín, Juliana M Rosa and Gertrudis Perea for expert advice andcritical reading of the manuscript. This work was supported by grants from the Spanish Ministry ofScience and Innovation to MN (Ramón y Cajal RYC-2016-20414 and RTI2018-094887-B-I00) and FondoEuropeo de Desarrollo Regional (FEDER) and PID2020-115091RB-10 to RT. The professional editingservice NB Revisions was used for technical preparation of the text prior to submission.

Author contributions

IS carried out most of the experimental work. JE developed PRQ analysis method. PP, MP and RT carriedout the patch-dPCR single-cell analysis of mtDNA copy number. CM-M carried out someimmunohistochemical analysis. LD helped in analysis data. MN designed research and wrote the paper.

Declaration of interests

The authors declare no competing nancial interests.

Data availability statement

The data that support the ndings of this study are available from the corresponding author uponreasonable request.

References1. Haim, L. ben & Rowitch, D. H. Functional diversity of astrocytes in neural circuit regulation. Nature

Reviews Neuroscience 18, (2017).

Page 19/36

2. Khakh, B. S. & Deneen, B. The Emerging Nature of Astrocyte Diversity. Annual Review ofNeuroscience 42, (2019).

3. Khakh, B. S. & Sofroniew, M. v. Diversity of astrocyte functions and phenotypes in neural circuits.Nature Neuroscience 18, (2015).

4. Martin, R., Bajo-Graneras, R., Moratalla, R., Perea, G. & Araque, A. Circuit-specic signaling inastrocyte-neuron networks in basal ganglia pathways. Science 349, 730–734 (2015).

5. Chai, H. et al. Neural Circuit-Specialized Astrocytes: Transcriptomic, Proteomic, Morphological, andFunctional Evidence. Neuron 95, (2017).

. Morel, L. et al. Molecular and Functional Properties of Regional Astrocytes in the Adult Brain. TheJournal of Neuroscience 37, (2017).

7. John Lin, C.-C. et al. Identication of diverse astrocyte populations and their malignant analogs.Nature Neuroscience 20, (2017).

. Batiuk, M. Y. et al. Identication of region-specic astrocyte subtypes at single cell resolution. NatureCommunications 11, (2020).

9. Salgado, S. & Kaplitt, M. G. The Nucleus Accumbens: A Comprehensive Review. Stereotactic andFunctional Neurosurgery 93, 75–93 (2015).

10. Goto, Y. & Grace, A. A. Dopamine Modulation of Hippocampal-Prefrontal Cortical Interaction DrivesMemory-Guided Behavior. Cerebral Cortex 18, (2008).

11. Sesack, S. R. & Grace, A. A. Cortico-Basal Ganglia Reward Network: Microcircuitry.Neuropsychopharmacology 35, (2010).

12. Christakou, A., Robbins, T. W. & Everitt, B. J. Functional disconnection of a prefrontal cortical–dorsalstriatal system disrupts choice reaction time performance: Implications for attentional function.Behavioral Neuroscience 115, (2001).

13. Christakou, A. Prefrontal Cortical-Ventral Striatal Interactions Involved in Affective Modulation ofAttentional Performance: Implications for Corticostriatal Circuit Function. Journal of Neuroscience24, (2004).

14. Cools, R., Sheridan, M., Jacobs, E. & D’Esposito, M. Impulsive Personality Predicts Dopamine-Dependent Changes in Frontostriatal Activity during Component Processes of Working Memory.Journal of Neuroscience 27, (2007).

15. Floresco, S. B., Braaksma, D. N. & Phillips, A. G. Thalamic–Cortical–Striatal Circuitry SubservesWorking Memory during Delayed Responding on a Radial Arm Maze. The Journal of Neuroscience19, (1999).

1. LeGates, T. A. et al. Reward behaviour is regulated by the strength of hippocampus–nucleusaccumbens synapses. Nature 564, (2018).

17. Ito, R., Robbins, T. W., Pennartz, C. M. & Everitt, B. J. Functional Interaction between the Hippocampusand Nucleus Accumbens Shell Is Necessary for the Acquisition of Appetitive Spatial ContextConditioning. Journal of Neuroscience 28, (2008).

Page 20/36

1. Bagot, R. C. et al. Ventral hippocampal afferents to the nucleus accumbens regulate susceptibility todepression. Nature Communications 6, (2015).

19. Beyeler, A. et al. Organization of Valence-Encoding and Projection-Dened Neurons in the BasolateralAmygdala. Cell Reports 22, (2018).

20. Beyeler, A. et al. Divergent Routing of Positive and Negative Information from the Amygdala duringMemory Retrieval. Neuron 90, (2016).

21. Hsu, C.-C., Madsen, T. E., O’Gorman, E., Gourley, S. L. & Rainnie, D. G. Reward-related dynamicalcoupling between basolateral amygdala and nucleus accumbens. Brain Structure and Function 225,(2020).

22. Tye, K. M. Glutamate Inputs to the Nucleus Accumbens: Does Source Matter? Neuron 76, 671–673(2012).

23. Papp, E. et al. Glutamatergic input from specic sources inuences the nucleus accumbens-ventralpallidum information ow. Brain Structure and Function 217, 37–48 (2012).

24. Gordleeva, S. Yu., Ermolaeva, A. v., Kastalskiy, I. A. & Kazantsev, V. B. Astrocyte as SpatiotemporalIntegrating Detector of Neuronal Activity. Frontiers in Physiology 10, (2019).

25. Perea, G. & Araque, A. Properties of synaptically evoked astrocyte calcium signal reveal synapticinformation processing by astrocytes. J Neurosci 25, 2192–2203 (2005).

2. Britt, J. P. et al. Synaptic and Behavioral Prole of Multiple Glutamatergic Inputs to the NucleusAccumbens. Neuron 76, (2012).

27. Zolnik, T. A. et al. All-optical functional synaptic connectivity mapping in acute brain slices using thecalcium integrator CaMPARI. The Journal of Physiology 595, (2017).

2. Fosque, B. F. et al. Labeling of active neural circuits in vivo with designed calcium integrators.Science 347, (2015).

29. Centemeri, C. et al. Characterization of the Ca 2+ responses evoked by ATP and other nucleotides inmammalian brain astrocytes. British Journal of Pharmacology 121, (1997).

30. Moeyaert, B. et al. Improved methods for marking active neuron populations. NatureCommunications 9, (2018).

31. Li, Z. et al. Cell-Type-Specic Afferent Innervation of the Nucleus Accumbens Core and Shell.Frontiers in Neuroanatomy 12, (2018).

32. Pastore, M. & Calcagnì, A. Measuring Distribution Similarities Between Samples: A Distribution-FreeOverlapping Index. Frontiers in Psychology 10, (2019).

33. Schmid, F. & Schmidt, A. Nonparametric estimation of the coecient of overlapping—theory andempirical application. Computational Statistics & Data Analysis 50, (2006).

34. Ambroggi, F., Ishikawa, A., Fields, H. L. & Nicola, S. M. Basolateral Amygdala Neurons FacilitateReward-Seeking Behavior by Exciting Nucleus Accumbens Neurons. Neuron 59, (2008).

35. Rana, S., Jasola, S. & Kumar, R. A review on particle swarm optimization algorithms and theirapplications to data clustering. Articial Intelligence Review 35, (2011).

Page 21/36

3. Floresco, S. B., Blaha, C. D., Yang, C. R. & Phillips, A. G. Modulation of Hippocampal and Amygdalar-Evoked Activity of Nucleus Accumbens Neurons by Dopamine: Cellular Mechanisms of InputSelection. The Journal of Neuroscience 21, (2001).

37. Jackson, J. G. & Robinson, M. B. Regulation of mitochondrial dynamics in astrocytes: Mechanisms,consequences, and unknowns. Glia 66, (2018).

3. Agarwal, A. et al. Transient Opening of the Mitochondrial Permeability Transition Pore InducesMicrodomain Calcium Transients in Astrocyte Processes. Neuron 93, (2017).

39. Podlesniy, P. & Trullas, R. Absolute measurement of gene transcripts with Sele-digital PCR. ScienticReports 7, (2017).

40. Björklund, A. & Dunnett, S. B. Dopamine neuron systems in the brain: an update. Trends inNeurosciences 30, (2007).

41. Shigetomi, E., Bowser, D. N., Sofroniew, M. v & Khakh, B. S. Two forms of astrocyte calciumexcitability have distinct effects on NMDA receptor-mediated slow inward currents in pyramidalneurons. J Neurosci 28, 6659–6663 (2008).

42. Cartmell, S. CD. et al. Multimodal characterization of the human nucleus accumbens. NeuroImage198, (2019).

43. Castro, D. C. & Bruchas, M. R. A Motivational and Neuropeptidergic Hub: Anatomical and FunctionalDiversity within the Nucleus Accumbens Shell. Neuron 102, (2019).

44. Kardos, J. et al. Molecular Plasticity of the Nucleus Accumbens Revisited—Astrocytic Waves ShallRise. Molecular Neurobiology 56, (2019).

45. Parent, A. Handbook of Basal Ganglia Structure and Function. vol. 24 (2016).

4. Floresco, S. B. Dissociable Roles for the Nucleus Accumbens Core and Shell in Regulating SetShifting. Journal of Neuroscience 26, (2006).

47. Herrero-Navarro, Á. et al. Astrocytes and neurons share region-specic transcriptional signatures thatconfer regional identity to neuronal reprogramming. Science Advances 7, (2021).

4. Matias, I., Morgado, J. & Gomes, F. C. A. Astrocyte Heterogeneity: Impact to Brain Aging and Disease.Frontiers in Aging Neuroscience 11, (2019).

49. Clarke, B. E., Taha, D. M., Tyzack, G. E. & Patani, R. Regionally encoded functional heterogeneity ofastrocytes in health and disease: A perspective. Glia 69, (2021).

50. Verkhratsky, A., Rodriguez, J. J. & Parpura, V. Calcium signalling in astroglia. Mol Cell Endocrinol 353,45–56 (2012).

51. Wagner, G. P. & Altenberg, L. Perspective: Complex Adaptations and the Evolution of Evolvability.Evolution 50, (1996).

52. Lipson, H., Pollack, J. B. & Suh, N. P. ON THE ORIGIN OF MODULAR VARIATION. Evolution 56, (2002).

53. Parpura, V. & Verkhratsky, A. Astrogliopathology: could nanotechnology restore aberrant calciumsignalling and pathological astroglial remodelling? Biochim Biophys Acta 1833, 1625–1631 (2013).

Page 22/36

54. Gipson, C. D., Kupchik, Y. M. & Kalivas, P. W. Rapid, transient synaptic plasticity in addiction.Neuropharmacology 76, (2014).

55. Stuber, G. D., Britt, J. P. & Bonci, A. Optogenetic Modulation of Neural Circuits that Underlie RewardSeeking. Biological Psychiatry 71, (2012).

5. Katz, K. W., Meier, S. D., Stephan, J. & Rose, C. R. Developmental prole and properties ofsulforhodamine 101—Labeled glial cells in acute brain slices of rat hippocampus. Journal ofNeuroscience Methods 169, (2008).

Figures

Page 23/36

Figure 1

Development of CaMPARIGFAP as a functional tool to study astrocytic networks. (A) The geneticallyencoded Ca2+ indicator CaMPARI was expressed within an adeno-associated virus (AAV) vector system,under control of a specic astroglial promoter (glial brillary acidic protein [GFAP]). (B) CaMPARI wasselectively expressed in astrocytes of the nucleus accumbens (NAc) by intracranial viral vector injectionfor acute slice imaging experiments 2-4 weeks later. (C) Representative confocal images of NAc cortical

Page 24/36

slices showing the expression of CaMPARIGFAP in astrocytes. Fluorescence signals show cell expression,with partial red-to-green photoconversion after 40 s of total illumination with 405 nm light. Magenta andcyan show immunohistochemically labelled cells with the astrocytic and neuronal markers (S100 andNeuN, respectively). Merged image of the four signals (right; scale bar = 100 µm).Inset for every image(bottom; scale bar = 30 µm). Note the selective expression of CaMPARIGFAP in S100-positive cells(astrocytes). (D) S100 (astrocyte) and NeuN (neuron) cell populations had different percentages ofCaMPARIGFAP Green and Red positive cells (n = 2861 cells; 8 elds; 2 mice), showing the selectiveexpression of CaMPARI. Two-way ANOVA, Holm-Sidak test for multiple comparisons, ***: p < 0.001. Allerror bars express SEM. (E1) Time-lapse of astrocytic activity obtained by real-time imaging ofCaMPARIGFAP Green uorescence changes in acute slices, and in response to ATP (20 mM) stimulus.(E2) Fluorescence traces showing astrocyte activities of the ve circular regions marked in the time-lapse.(E3) Heatmap showing changes in uorescence signal vs time obtained from 100 regions of interest(ROIs; within the same eld) taken from the time-lapse. (F1) CaMPARIGFAP Green, Red and mergeduorescences of astrocytes after 40 s UV (λ = 405 nm) pulse and 20 mM ATP local stimulation. (F2)CaMPARIGFAP Green and Red uorescence ratio as a function of distance from ATP stimulus; the ATP-lled pipette was located at 0 mm distance.

Page 25/36

Figure 2

Pathway-specic nucleus accumbens (NAc) astrocyte activity in response to medial prefrontal cortex(mPFC) afferents. (A1) Scheme and representative coronal brain slices showing opsin expression in themPFC (top; scale bar = 1 mm) and their associated axons in the NAc (bottom; scale bar = 500 µm). (A2)Quantication of mPFC projection uorescence intensity (ΔF/F0) in the NAc core (AcbC; slashed bar) andshell (AcbSh; solid bar) regions, showing the specic innervation prole of the mPFC. Two-tailed unpaired

Page 26/36

t-test; n = 11 infections, 6 mice; *: p < 0.05. (B1) Experimental scheme based on electrophysiologicalrecordings of NAc medium spiny neurons (MSNs) during specic optostimulation of mPFC axons. (B2)Correlation analysis of individual MSN excitatory postsynaptic current (EPSC) amplitudes and theafferents’ uorescence in the area of register, showing a positive correlation between the amplitude of theresponses and density of afferents (n = 20 cells, 3 mice). Pearson r correlation, p-value (two-tailed). (B3)EPSC amplitude recordings from neurons of the AcbC (n = 8 cells, 4 mice) and AcbSh (n = 15 cells, 4mice) and representative traces (inset) showing the synaptic neuronal strength in each NAc subregion inresponse to mPFC afferent optostimulation. Two-tailed unpaired t-test using Welch correction; **: p <0.01. (C1) Experimental scheme for monitoring astrocytic Ca2+ dynamics using the CaMPARIGFAP Ca2+indicator while specically optostimulating the mPFC axons. Inset, image shows in red mPFC afferentscoming to the NAc, with expression of CaMPARIGFAP in green. Scale bar = 500 µm. (C2) Left,representative heatmaps (ROIs pooled from 3 different experiments) and Right, representative Ca2+traces at the AcbC and AcbSh in response to mPFC optostimulation. (C3) Left, proportion of ROIsresponding to mPFC-afferent optostimulation (2420 ROIs) at the AcbC (slashed bar) and AcbSh (solidbar) in control condition (n = 25 slices, 8 mice). After the control experiment, a subset of those ROIs wereregistered in presence of a selective antagonist of the metabotropic glutamate receptor subtype mGluR5(MPEP, 50 µM) (magenta bar; n = 4 slices, 2 mice). Right, average change of astrocytic Ca2+ spikefrequency in response to optostimulation in control condition at the AcbC (slashed bar) and AcbSh (solidbar) (n = 25 slices, 8 mice) and in presence of MPEP (magenta bar; n = 4 slices, 2 mice). One-sample t-test, #: p < 0.05; ##: p < 0.01; one-way ANOVA, Holm-Sidak test for multiple comparisons, p > 0.05. (D1)Experimental scheme for the analysis of mPFC axon uorescence using the Partition in RegularQuadrants (PRQ) method. (D2) Average PRQ image showing the mPFC glutamatergic innervation patternin the NAc. Lines starting from pixel 0 in each subregion were used for quantication (pixel = 50 µm).(D3) Quantication of mPFC afferent uorescence changes across space in the AcbC and AcbSh. (D4)Average astrocytic spatial uorescence variation in the AcbC (slashed bar) and AcbSh (solid bar) (n = 9slices, 6 mice). One-sample t-test, ###: p < 0.001; two-tailed unpaired t-test using Welch correction, ***: p< 0.001. (E1) Experimental scheme for the analysis of astrocytic Ca2+ uorescence turnover in responseto mPFC optostimulation using the PRQ method. (E2) Average PRQ image showing astrocytic activationpattern in the NAc in basal and optostimulated conditions. Lines starting from pixel 0 in each subregionwere used for quantication (pixel = 50 µm). (E3) Quantication of the astrocytic photoconverteduorescence changes across space in the AcbC and AcbSh. (E4) Average spatial uorescence change inoptostimulated condition with respect to basal, in control experiments at the AcbC (slashed bar) andAcbSh (solid bar; n = 18 slices, 6 mice) and in presence of MPEP (magenta bar; n= 8 slices, 2 mice). One-sample t-test, #: p < 0.05; ##: p < 0.001; one-way ANOVA, Holm-Sidak test for multiple comparisons, *: p <0.05, **: p < 0.01. (F1) Masks of mPFC glutamatergic afferents area (red) and astrocyte activation area(yellow) dened by a k-mean cluster thresholding method. In orange, the overlap area between the two.(F2) Quantication of the spatial overlap (orange bar) between mPFC afferents (red bar) and activeastrocytes (black bar) responding to optostimulation (n= 9 slices, 6 mice). One-way ANOVA, Holm-Sidaktest for multiple comparisons, *: p < 0.05; **: p < 0.001. (G) Bivariance similarity index showing the spatial

Page 27/36

overlap in the AcbC and AcbSh. Note that there was no difference between subregions regarding theastrocytic interaction with mPFC afferents. MANOVA, d = 0, p = 0.586. All error bars express SEM.

Figure 3

Pathway-specic nucleus accumbens (NAc) astrocyte activity in response to basolateral amygdala(Amyg) afferents. (A1) Scheme and representative coronal brain slices showing opsin expression in theAmyg ( top; scale bar = 1 mm) and their associated axons in the NAc (bottom; scale bar = 500 µm). (A2)

Page 28/36

Quantication of Amyg projections uorescence intensity (ΔF/F0) in the NAc core (AcbC; slashed bar)and shell (AcbSh; solid bar) region, showing the specic innervation prole of the Amyg. Two-tailedunpaired t-test, n = 13 infections, 7 mice; *: p < 0.05. (B1) Experimental scheme for electrophysiologicalrecordings of NAc medium spiny neurons (MSNs) during specic optostimulation of Amyg axons. (B2)Correlation analysis of individual MSN excitatory postsynaptic current (EPSC) amplitudes and theafferents’ uorescence in the area of register, showing a positive correlation between the amplitude of theresponses and density of afferents (n = 15 cells, 2 mice). Pearson r correlation, p-value (two-tailed). (B3)EPSC amplitude recordings from neurons of the AcbC (slashed bar) and AcbSh (solid bar). Inset,representative traces showing the synaptic neuronal strength in each NAc subregion in response to Amygafferent optostimulation. Two-tailed unpaired t-test, p > 0.05. (C1) Experimental scheme for monitoringastrocytic Ca2+ dynamics using the CaMPARIGFAP Ca2+ indicator while specically optostimulating theAmyg axons. Inset, image shows in red Amyg afferents coming to the NAc, with expression ofCaMPARIGFAP in green. Scale bar = 500 µm. (C2) Left, representative heatmaps (ROIs pooled from 3different experiments) and Right, representative Ca2+ traces at the AcbC and AcbSh in response to Amygoptostimulation. (C3) Left, proportion of ROIs responding to Amyg-afferent optostimulation (1550 ROIs)in control condition at the AcbC (slashed bar) and AcbSh (solid bar; n = 17 slices, 7 mice). After thecontrol experiment, a subset of those ROIs were registered in presence of a selective antagonist of themetabotropic glutamate receptor subtype mGluR5 (MPEP, 50 µM) (magenta bar; n = 5 slices, 2 mice).Right, average change of astrocytic Ca2+ spike frequency in response to optostimulation in controlcondition at the AcbC (slashed bar) and AcbSh (solid bar; n=17 slices, 7 mice) and in presence of MPEP(magenta bar; n = 5 slices, 2 mice). One-sample t-test, #: p < 0.05; ##: p < 0.01; one-way ANOVA, Holm-Sidak test for multiple comparisons, p > 0.05. (D1) Experimental scheme for the analysis of Amyg axonuorescence using the Partition in Regular Quadrants (PRQ) method. (D2) Average PRQ image showingthe Amyg glutamatergic innervation pattern in the NAc. Lines starting from pixel 0 in each subregion wereused for quantication (pixel = 50 µm). (D3) Quantication of Amyg afferent uorescence changesacross space in the AcbC and AcbSh. (D4) Average spatial uorescence variation in the AcbC (slashedbar) and AcbSh (solid bar) (n = 9 slices, 6 mice). One-sample t-test, #: p < 0.05, ##: p < 0.01; two-tailedunpaired t-test, *: p < 0.05. (E1) Experimental scheme for analysis of astrocytic Ca2+ uorescenceturnover in response to Amyg optostimulation using the PRQ method. (E2) Average PRQ image showingastrocytic activation pattern in the NAc in basal and optostimulated conditions. Lines starting from pixel0 in each subregion were used for quantication (pixel = 50 µm). (E3) Quantication of the astrocyticphotoconverted uorescence changes across space in the AcbC and AcbSh. (E4) Average spatialuorescence change in optostimulated condition with respect to basal, in control experiments at the AcbC(slashed bar) and AcbSh (solid bar; n = 18 slices, 6 mice) and in presence of MPEP (magenta bar; n= 8slices, 2 mice). One-sample t-test, #: p < 0.05, ##: p < 0.001; one-way ANOVA, Holm-Sidak test for multiplecomparisons, **: p < 0.01. (F1) Positive masks of Amyg glutamatergic afferents area (red) and astrocyteactivation area (yellow) dened by a k-mean cluster thresholding method. In orange, the overlap areabetween the two. (F2) Quantication of the spatial overlap (orange bar) between Amyg afferents (red bar)and active astrocytes (green bar) responding to optostimulation (n= 9 slices, 6 mice). One-way ANOVA,Holm-Sidak test for multiple comparisons, **: p < 0.01. (G) Bivariance similarity index showing the spatial

Page 29/36

overlap in the AcbC and AcbSh. Note that there was no difference between subregions regarding theastrocytic interaction with Amyg afferents. MANOVA, d = 0, p = 0.482. All error bars express SEM.

Figure 4

Pathway-specic nucleus accumbens (NAc) astrocyte activity in response to ventral hippocampus (vHip)afferents. (A1) Scheme and representative coronal brain slices showing opsin expression in the vHip (top;scale bar = 1 mm) and their associated axons in the NAc (bottom; scale bar = 500 µm). (A2)

Page 30/36

Quantication of vHip projection uorescence intensity (ΔF/F0) in the NAc core (AcbC; slashed bar) andshell (AcbSh; solid bar) regions. Two-tailed unpaired t-test, (n = 10 infections, 6 mice); **: p < 0.01. (B1)Experimental scheme based on electrophysiological recordings of NAc medium spiny neurons (MSNs)during specic optostimulation of vHip axons. (B2) Correlation analysis of individual MSN excitatorypostsynaptic current (EPSC) amplitudes and the afferents’ uorescence in the area of register, showing apositive correlation between the amplitude of the responses and density of afferents (n = 23 cells, 5mice). Pearson r correlation, p-value (two-tailed). (B3) EPSC amplitude recordings from neurons of theAcbC (slashed bar) and AcbSh (solid bar) (n = 21 cells, 6 mice). Inset, representative traces showing thesynaptic neuronal strength in each NAc subregion in response to vHip afferents optostimulation. Two-tailed unpaired t-test using Welch correction, **: p < 0.01. (C1) Experimental scheme for monitoringastrocytic Ca2+ dynamics using the CaMPARIGFAP Ca2+ indicator while specically optostimulating thevHip axons. Inset, image shows in red vHip afferents coming to the NAc, with expression ofCaMPARIGFAP in green. Scale bar = 500 µm. (C2) Left, representative heatmaps (ROIs pooled from 3different experiments) and Right, representative Ca2+ traces in the AcbC and AcbSh in response to vHipoptostimulation. (C3) Left, proportion of ROIs responding to vHip-afferent optostimulation (1450astrocytes) in control condition at the AcbC (slashed bar) and AcbSh (solid bar) (n = 13 slices, 5 mice).After the control experiment, a subset of those astrocytes were registered in presence of 50 µM MPEP(magenta bar; n = 6 slices, 2 mice). Right, average change of astrocytic Ca2+ spike frequency in responseto optostimulation in control condition in the AcbC (slashed bar) and AcbSh (solid bar) (n = 13 slices, 5mice) and in the presence of MPEP (magenta bar; n = 5 slices, 2 mice). One-sample t-test, #: p < 0.05, ##:p < 0.01; one-way ANOVA, Holm-Sidak test for multiple comparisons, *: p < 0.05; **: p < 0.01. (D1)Experimental scheme for the analysis of vHip axon uorescence using the Partition in Regular Quadrants(PRQ) method. (D2) Average PRQ image showing the vHip glutamatergic innervation pattern in the NAc.Lines starting from pixel 0 in each subregion were used for quantication (pixel = 50 µm). (D3)Quantication of vHip afferent uorescence changes over space in the AcbC and AcbSh. (D4) Averagespatial uorescence variation in the AcbC (slashed bar) and AcbSh (solid bar) (n = 8 slices, 6 mice). One-sample t-test, ###: p < 0.001; two-tailed unpaired t-test, ***: p < 0.001. (E1) Experimental scheme for theanalysis of astrocytic Ca2+ uorescence turnover in response to vHip optostimulation. (E2) Average PRQimage showing astrocytic activation pattern in the NAc in basal and optostimulated conditions. Linesstarting from pixel 0 in each subregion were used for quantication (pixel = 50 µm). (E3) Quantication ofthe astrocytic photoconverted uorescence changes over space in the AcbC and AcbSh. (E4) Averagespatial uorescence change in optostimulated condition with respect to basal, in control experiments inthe AcbC (slashed bar) and in AcbSh (solid bar) (n= 16 slices 8 pairs, 6 mice) and in presence of MPEP(magenta bar; n= 8 slices, 2 mice). One-sample t-test, ##: p < 0.001; one-way ANOVA, Holm-Sidak test formultiple comparisons, *:p < 0.05; **: p < 0.01. (F1) Masks showing the activated area of vHipglutamatergic afferents (in red), of astrocytes (in yellow), and the overlap area (in orange). Masks weredened by a k-mean cluster thresholding method. (F2) Percentage of area of vHip afferents (red bar), ofthe spatial overlap (orange bar), and of active astrocytes (blue bar) responding to optostimulation (n= 8slices, 6 mice). One-way ANOVA, Holm-Sidak test for multiple comparisons, ***: p < 0.001. (G) Bivariancesimilarity index showing the spatial overlap in the AcbC and AcbSh. Note that only in the AcbSh,

Page 31/36

astrocytes responded to optostimulation in areas with strong vHip glutamatergic innervation. MANOVA, d= 1, *: p < 0.05. All error bars express SEM.

Figure 5

Synapse-specicity of nucleus accumbens (NAc) astrocytes. (A) Average Partition in Regular Quadrants(PRQ) images (top) of photoconverted astrocytic activity showing isolated astrocytic active areas (aboveactivation threshold) in response to each glutamatergic input. Lines starting from pixel 0 in each

Page 32/36

subregion were used for quantication (pixel = 50 µm). (B) Spatial comparison of the three pathway-specic astrocytic responses by pixel-to-pixel correlation values of the different average PRQ image pairs.Pearson r correlation, p-value (two-tailed). (C1) Astrocytic photoconverted uorescence changes overspace in the NAc core (AcbC) and shell (AcbSh) regions. (C2) Average spatial uorescence changes inresponse to optostimulation of the three pathways (mPFC, black; Amyg, green; vHip, blue) in the AcbCand AcbSh. One-way ANOVA, Holm-Sidak test for multiple comparisons; n = 26 slices, 18 mice; *: p < 0.05.

Figure 6

Page 33/36

Characterization of calcium activity and mtDNA copy number in nucleus accumbens (NAc) astrocyticsubpopulations. (A1) Experimental scheme for analysis of astrocytic Ca2+ uorescence dynamics usingreal-time imaging and partition in regular quadrants (PRQ) spatial analysis of astrocytic activity. (A2)Real-time imaging representative Ca2+ traces of astrocytic basal activity in the NAc core (AcbC) and shell(AcbSh) subregions. (A3) Left, proportion of ROIs showing basal Ca2+ events; Right, basal astrocyticCa2+ spiking frequency (n= 2420 ROIs, 36 slices, 20 mice). (A4) Average PRQ image showing astrocyticbasal activation pattern in the NAc. Lines starting from pixel 0 in each subregion were used forquantication (pixel = 50 µm). (A5) Quantication of the astrocytic photoconverted uorescence signalacross space in the AcbC and AcbSh. (A6) Average spatial uorescence signal of basal activity in theAcbC (red bar) and AcbSh (blue bar) (n= 20 slices, 12 mice). One-sample t-test, ##: p < 0.01; two-tailedunpaired t-test, *: p < 0.05; ***: p < 0.001. (B1) Patch-dPCR astrocytic single-cell mtDNA collection (left) fordigital PCR showing representative I-V registers (right). (B2) NAc-representative areas of RNA collection(left) and per cell mitochondrial DNA quantication (right) for astrocytes of the AcbC (red bar; n = 32) andAcbSh (blue bar; n = 32) compared to control (n = 37 slices, 11 mice). One-way ANOVA, Holm-Sidak testfor multiple comparisons, **: p < 0.01, ***: p < 0.001. All error bars express SEM.

Page 34/36

Figure 7

Glutamatergic transmission shows spatial segregation of afferent-astrocyte overlap regions (A1) AveragePartition in Regular Quadrants (PRQ) images showing overlap colocalization between high-densityafferent regions and increased astrocytic activity for each glutamatergic input. (A2) Spatial comparisonof the three overlap domains by pixel-to-pixel correlation of different PRQ average pairs. Pearson rcorrelation, p-value (two-tailed). (B1) Average PRQ images showing overlap colocalization between high-

Page 35/36

density afferent regions and increased astrocytic activity for ventral tegmental area (VTA) input. (B2)Statistical comparison of the % overlap area among VTA and glutamatergic inputs. One-way ANOVA,Holm-Sidak test for multiple comparisons, ***: p < 0.001. All error bars express SEM.

Figure 8

Integrative properties of nucleus accumbens (NAc) astrocytes in response to co-stimulation ofglutamatergic pathways. (A1) Average Partition in Regular Quadrants (PRQ) image showing astrocytic

Page 36/36

activation pattern in the NAc in response to co-stimulation of mPFC and Amyg afferents. (A2)Quantication of astrocytic photoconverted uorescence changes over space in the NAc core (AcbC) andshell (AcbSh) subregions. (A3) Average spatial uorescence change in co-stimulated condition withrespect to basal in the AcbC and AcbSh (mPFC + Amyg, orange bars; n = 10 slices, 3 mice). Data of co-stimulation is plotted and compared to individual signals from the mPFC (black bars) and Amyg (greenbars). (B1) Average PRQ image showing astrocytic activation pattern in the NAc in response to co-stimulation of Amyg and vHip afferents. (B2) Quantication of astrocytic photoconverted uorescencechanges over space in the AcbC and AcbSh. (B3) Average spatial uorescence change in co-stimulatedcondition with respect to basal in the AcbC and AcbSh (Amyg + vHip, purple bars; n = 14 slices, 4 mice).Data of co-stimulation is plotted and compared to individual signals from the Amyg (green bars) andvHip (blue bars). (C1) Average PRQ image showing astrocytic activation pattern in the NAc in response toco-stimulation of mPFC and vHip afferents. (C2) Quantication of astrocytic photoconverteduorescence changes across space in the AcbC and AcbSh. (C3) Average spatial uorescence change inco-stimulated condition with respect to basal in the AcbC and AcbSh (mPFC + vHip, dark green bars; n=12 slices, 5 mice). Data of co-stimulation is plotted and compared to individual signals from the mPFC(black bars) and vHip (blue bars). (D1) Average PRQ image showing astrocytic activation pattern in theNAc in response to mPFC, Amyg and vHip co-stimulation. (D2) Quantication of astrocyticphotoconverted uorescence changes across space in the AcbC and AcbSh. (D3) Average spatialuorescence change in co-stimulated condition with respect to basal in the AcbC and AcbSh (mPFC +Amyg + vHip, grey bars; n= 12 slices, 3 mice). Note that in the AcbSh, astrocytic activity was signicantlyinhibited below basal levels in response to optostimulation (#: p < 0.05). Data of co-stimulation is plottedand compared to individual signals triggered by mPFC (black bars), Amyg (green bars) and vHip (bluebars). Lines starting from pixel 0 in each subregion were used for quantication (pixel = 50 µm).Statistics used are one-sample t-test, #:p < 0.05; ##: p < 0.01 and one-way ANOVA, Holm-Sidak test formultiple comparisons, *: p < 0.05; ***: p < 0.001. All error bars express SEM.

Supplementary Files

This is a list of supplementary les associated with this preprint. Click to download.

SupFigure1.jpg

SupFigure2.jpg

SupFigure4.jpg

SupFigure5.jpg

SupFigure6.jpg

SupFigure7.jpg