h-iPSCs were generated by introducing lentiviral transduction to fibroblasts derived from a healthy donor. The cells used in this study were generously provided by Frega et al. The complete protocol for generating and maintaining the rtTa/NgN2 positive cell line has been previously published (Chiappalone et al., 2019). Once the h-iPSCs reached 80% confluence, they were detached and seeded as individual cells in 6-well plates that were pre-coated with Matrigel solution. The cells were cultured in Essential 8 Flex Medium (Cat. A2858501, Gibco ThemoFisher) supplemented with 1% penicillin/streptomycin (Cat. 15140122, Gibco. ThermoFisher), 50 μg/mLG418 (Cat. G8168, Merck Life Science), 0.5 μg/mL puromycin (Cat. P8833, Merck Life Science), and 4 μg/mL Doxycycline. The differentiation process into neurons was induced by adding Doxycycline to the medium, and this step was designated as Day After Differentiation 0 (DAD 0). On DAD 1, the medium was replaced with DMEM/F12 (Cat. 11320074, Gibco, ThemoFisher) supplemented with 1% N2-supplement (Cat. 17502048, Gibco, ThermoFisher), 1% MEM non-essential amino acid solution (Cat. 11140050, Gibco, ThermoFisher), 1% penicillin/streptomycin, 10 μg/mL human BDNF (Cat. PHC7074, Gibco, ThermoFisher), 10 μg/mL human NT-3 (Cat. SRP312, Merck Life Science), and 4 μg/mLdoxycycline. At DAD 3, the neurons were detached using Accutase (Cat. A6964, Sigma Aldrich, Merck Life Science), collected in Neurobasal medium supplemented with 1% penicillin/streptomycin, 2% B27 supplement (Cat. 17504,044, Gibco, ThermoFisher), 1% glutamax (Cat. 35050038, Gibco, ThermoFisher), 10 μg/mL human BDNF, 10 μg/mL human NT-3, and 4 μg/mL doxycycline and centrifuged at 1,200 rpm for 5 min. The cells were then resuspended in complete Neurobasal medium and were ready for use.

Astrocytes were obtained from isolated cortices from dissection of E18 rat, and non-astrocytic cells were removed thanks to 6 h of orbital shaking after plating. The specific procedures and methods used in this process were previously described in the study by Aprile et al. (2019). Astrocytes were placed in T-75 flasks containing DMEM High Glucose supplemented (Cat. 41965039, Gibco, ThermoFisher) with 10% FBS (Cat. 10270106, Gibco Invitrogen) and 1% penicillin/streptomycin. The flasks were then incubated at 37°C with 5% CO2, and the medium was replaced every 3 days for a period of 7–10 days.

Human ERBB2+/SKBR3 breast cancer cell lines and MCF10 mammary epithelial cells were obtained from Biological Bank of IRCCS Ospedale Policlinico San Martino, Genova.

SKBR-3 were cultured and maintained in DMEM medium supplemented with 10% FBS (fetal bovine serum), 1 mM glutamine (Cat. C0550-100, Voden medical instruments), 10 μg/mL insulin (Cat. I0516, Sigma Aldrich, Merck Life Science), 10 ug/mL penicillin, and 10 µg/streptomycin. MCF10-A cells were cultured I DMEM/F12 with 5% horse serum (Cat. H1270, Sigma Aldrich, Merck Life Science) 20 ng/mL EGF (Cat. 78006.1, Voden medical instruments), 0.5 mg/mL hydrocortisone (Cat. 07925, Voden medical instruments), 100 ng/mL cholera toxin (Cat. C8052, Sigma Aldrich, Merck Life Science) 10 μg/mL insulin, 10 ug/mL penicillin, and 10 μg/mL streptomycin. Cells were maintained in humidified incubator at 37°C and 5% CO₂. When about 80% confluence was obtained, the medium has been exchanged for conditioning medium based on Neurobasal medium supplemented with 2% B27, glutamine 1 mM, 10 ug/mL penicillin, and 10 streptomycin. After 48 h, the conditioning media exposed to both cell types (referred as conditioned medium) were collected and filtered through a 0.2 µm filter and stored at −20°C until use. After removal of the conditioning medium, cells were subjected to a viability assay (NUCLEAR-ID^® Blue/Red cell viability reagent, Prod. No. ENZ-53005, Enzo Life Sciences) for 30 min of incubation, before being washed with PBS and observed qualitatively under a fluorescence microscope (Olympus IX70 wide field microscope equipped with Hamamatsu camera Orca-Flash 4.0 V3/LT+). Consequently, cells were trypsinized and detached to quantify the portion of live cells versus the apoptotic ones.

2D neuronal culture was obtained by seeding and co-culturing neurons differentiated from h-IPSCs and primary rat astrocytes onto multi-electrode array (MEA) devices. The day before seeding, MEAs were assembled with donuts-shaped Poly-dimethyl-siloxane (PDMS) structures (internal and external diameters: 5 and 22 mm respectively) to confine the active electrodes area. MEAs were sterilized in oven at 120° for 2 h. The active area of MEAs were functionalized with 1% chitosan solution pre-sterilized by autoclaving it at 120°C and left in the incubator overnight at 37°C, (Di Lisa et al., 2022). The coating solution was removed from the MEA which was then washed twice with water and left to dry under the laminar hood until seeding. Cells were seeded at a cell density of 1,500 cell/mm². After 2h, complete Neurobasal medium was added and cells were left in incubator at 37°C, 5% CO₂ and 95% humidity for 57 days. Starting from DIV11, 2.5% FBS solution has been added to the complete Neurobasal medium, and from DIV14, doxycycline has been removed from the complete Neurobasal medium (referred as iNeurons medium). iNeurons medium was based on Neurobasal medium supplemented with 1% penicillin/streptomycin, 2% B27 supplement, 1% glutamax, 10 μg/mL human BDNF, 10 μg/mL human NT-3, and 2.5% FBS.

After the recording session, 2D neuronal cultures were fixed with 4% paraformaldehyde for 20 min at room temperature. Permeabilization was achieved with phosphate buffer solution (PBS, Cat. 18912014, Gibco, ThermoFisher) containing 0.2% Triton-X100 (Cat. X100, Merck Life Science) for 15 min at room temperature. To prevent non-specific binding of antibodies, cells were then incubated in a blocking buffer solution (BBS) consisting of PBS, 0.3% bovine serum albumin (BSA, Cat. A9418, Merck Life Science), and 0.5% fetal bovine serum (FBS) for 35 min at RT. Cultures were incubated with primary antibody diluted in BBS for 1 h at room temperature. Cultures were rinsed two times with PBS and finally exposed to secondary antibodies. MAP-2 (dendritic microtubule-associated protein, Cat. 188 002 and 188,011, Synaptic System) and GFAP (glial fibrillary acidic protein, Cat. 173 002 and 173 01, Synaptic System) were used as primary antibodies to evaluate the morphology of neuronal and glial cells, respectively; Dapi (10 μg/mL, Cat. 75004, Voden medical instruments) was used to label nuclei. Alexa Fluor 488 and Alexa Fluor 549 Goat anti mouse or Goat anti rabbit were used as a secondary antibody (Cat. A11001, A11003, A11008, A11035, Gibco, ThermoFisher). An Olympus BX-51 upright microscope was used for immunofluorescence and the image acquisition was done with a Hamamatsu Orca ER II digital cooled CCD camera driven by Image ProPlus software (Media Cybernetic).

2D neuronal culture cells (DIV 57) were washed out in 0.1 M cacodylate buffer and fixed in 0.1 M cacodylate buffer containing 2.5% glutaraldehyde (Electron Microscopy Science, Hatfield, PA, United States), for 1 h at room temperature. The cells were postfixed in 1% osmium tetroxide for 2 h and 1% aqueous uranyl acetate for 1 h. Subsequently, samples were dehydrated through a graded ethanol series and flat embedded in epoxy resin (Poly-Bed; Polysciences, Inc., Warrington, PA) for 24 h at 60C. Ultrathin sections (50 nm) were cut parallel to the substrate and counterstained with 5% uranyl acetate in 50% ethanol. Electron micrographs were acquired as single snapshots and MIA (multiple image alignment) at Hitachi 7,800 120Kv electron microscope (Hitachi, Tokyo, Japan) equipped with a Megaview III digital camera and Radius software (EMSIS, Muenster, Germany).

Regarding electrophysiological characterization, to evaluate statistically significant differences between control and the conditioned cultures, statistical analysis was performed using the 2way-ANOVA’s test followed by Tukey’s multiple comparisons, since data follow a normal distribution (evaluated by the Kolmogorov-Smirnov normality test): (*) p ≤ 0.0001, significant statistical differences between different groups at specific time points; significant statistical differences within group during different time points: control group (■) p ≤ 0.0001, SKBR-3 group (•) p ≤ 0.0001 and MCF10-A (^▲) p ≤ 0.0001. For immunofluorescence characterization, to evaluate statistically significant differences between control and the conditioned cultures, statistical analysis was performed using the 1way-ANOVA’s test followed by Tukey’s multiple comparison: (*) p ≤ 0.01, (**) p ≤ 0.001 and (***) p ≤ 0.001.

Before starting this study, the effects of the conditioning medium on the MCF10-A and SKBR-3 cell lines were evaluated by a viability assay performed after 48 h of exposure to iNeurons medium. The results showed that the portion of dead cells was marginal (<4%) for both cell lines, without significant observable differences compared to the standard culture medium, demonstrating the adequacy of the conditioning protocol (Supplementary Figure S1).

The impact of SKBR-3 and MCF10-A cell conditioned medium on the activity of nervous cells was then characterized by recording the spontaneous activity of mature neuronal networks before and after their exposure to conditioned medium. To this purpose, neurons differentiated from human induced pluripotent stem cells were co-cultured with astrocytes onto MEA devices until DIV 41, when mature functional neuronal networks were formed, then the networks were exposed either to SKBR-3 or MCF10-A conditioned medium for up to 2 weeks. At the same time, neuronal networks cultured using iNeurons medium were considered as controls (Figure 1).

After recording the basal spontaneous activity at DIV41, the conditioning protocol was started, and the spontaneous activity of the neuronal networks was recorded at different timepoints to observe the effect of conditioned medium on the neural functional behavior over time. Namely, SKBR-3 conditioned networks (n° samples = 4), MCF10-A conditioned networks (n° samples = 4) and 2D control networks (n° samples = 4) were characterized in terms of spontaneous electrophysiological activity. Figure 2 shows the spontaneous activity (raw signal) and the global electrophysiological behavior of representative networks for each condition and each timepoint. Specifically, electrophysiological activity is qualitatively shown in the raster plot, where 300s of spontaneous activity are displayed. In all experimental conditions, at D0 (the first day of recording), samples showed synchronous bursts, that displayed comparable patterns typically observed in mature neuronal networks (Figure 2A). By D5, which corresponds to 5 days of conditioning protocol, networks conditioned with SKBR-3 conditioned medium began to display functional modifications, as shown in Figure 2B. The network activity significantly increased and was characterized by a high frequency of synchronous bursts with short duration. Meanwhile, the control networks and the samples conditioned with MCF10-A conditioned medium showed patterns similar to those observed at D0. Moving forward to D9, the SKBR-3 conditioned networks exhibited further evolution in activity (Figure 2C) The MCF10-A conditioned networks also exhibited a variation in firing frequency, showing a similar increase compared to the control group. Finally, raster plots generated on the final day of conditioning (D16) confirmed the functional changes observed in the SKBR-3 networks (Figure 2D). On the other hand, the MCF10-A conditioned networks and control exhibited patterns consistent with relatively minimal evolution.

Anomalies in activity patterns were also observed for SKBR-3 conditioned networks. Namely, some samples were characterized by a very rapid and synchronous network phenomena, others showed disorganized and uncoordinated spiking activity (Supplementary Figure S2).

Subsequently, a quantitative analysis of the data was performed, focusing on the evaluation of main electrophysiological parameters such as mean firing rate (MFR), mean burst rate (MBR), mean burst duration (MBD) and percentage of random spikes (%RS). These parameters were evaluated, normalized with respect to the value at D0 and averaged across the same experimental conditions, to assess the network dynamics and provide quantitative insights into the observed phenomena.

Figure 3 shows the electrophysiological parameters comparison between control samples and SKBR-3 conditioned neuronal networks. Both neuronal networks showed a similar trend of MFR during the period in culture; we can observe differences at various time points (Figure 3A). Specifically, starting from D9, SKBR-3 conditioned networks showed higher value of MFR than the control ones: 1.34 ± 0.036 (spikes/s) vs. 0.93 ± 0.038 (spikes/s), respectively (p < 0.0001). This different behavior was maintained until D16 (Ctr = 0.75 ± 0.031 spikes/s) vs. SKBR-3 = 1.16 ± 0.058 spikes/s) (p < 0.0001). Regarding the bursting behavior, both networks showed predominantly synchronous activity, but the MBR values of SKBR-3 conditioned networks increased during time (Figure 3B), showing a higher value starting from D9 (1.50 ± 0.02 bursts/min) from the control one (1.06 ± 0.02 bursts/min) (p < 0.0001). Furthermore, SKBR-3 conditioned networks exhibited a MBD (0.75 ± 0.015 s) lower than the control ones (0.82 ± 0.017 s) until D5 (p < 0.0001) (Figure 3C); after that, an increase in burst duration was observed starting from D9 until D16, control = 0.63 ± 0.019 s) and SKBR-3 = 0.76 ± 0.002 s), respectively (p < 0.0001). Consequently, during the first day after condition, SKBR-3 showed significant higher values of spikes out of burst respect to control ones (Figure 3D); then, starting from D9 a change in trend was observed, the control showed higher values than SKBR-3 ones (control = 0.94 ± 0.077 and SKBR-3 = 0.63 ± 0.037 (p < 0.0001)).

Moreover, the comparison between MCF10-A and SKBR-3 conditioned neuronal networks was carried out, as shown in Figure 4. MCF10-A conditioned networks showed higher values of firing frequency compared to SKBR-3 ones, starting from D5 ((MCF10-A = 1.38 ± 0.050 spikes/s) vs. SKBR-3 = 1.05 ± 0.052 spikes/s), p < 0.0001) with the highest values at D9 (MCF10-A = 2.66 ± 0.105 (spikes/s) vs. SKBR-3 = 1.34 ± 0.036 (spikes/s), p < 0.0001) as reported in Figure 4A. In terms of bursting behavior, the MBR of MCF10-A conditioned networks showed lower values over time in culture compared to SKBR-3 conditioned networks (p < 0.0001) (Figure 4B). Consequently, MCF10-A conditioned networks exhibited higher values of MBD, particularly at D9 (1.66 ± 0.067 s) which was significantly higher than SKBR-3 conditioned networks (0.79 ± 0.016 s) (p < 0.0001) (Figure 4C). Finally, SKBR-3 conditioned networks showed significantly higher values of random spikes compared to MCF10-A ones (Figure 4D). Moreover, also a quantitative comparison of electrophysiological behavior between the MCF10-A conditioned networks and the control networks was carried out (Supplementary Figure S3). Furthermore, based on the statistical analysis within individual groups over time, it has been observed that the control group did not exhibit statistically significant differences during the recording days. This suggests the formation of mature and stable neuronal cultures over time (Figure 3). On the other hand, a persistent statistical difference across various days was observed in different parameters (MFR, MBR, and RS) throughout the recording period for both treatments with conditioned SKBR-3 and MCF10-A media (Figure 4). The results were more evident in SKBR-3 than in MCF10-A conditioned cultures; MCF10-A conditioned networks, after an initial perturbation, showed a more stable bursting behavior and lower random spiking activity than SKBR-3. Whereas SKBR-3 conditioned networks showed an increase in MBR and RS values over time, higher than both control and MCF10-A groups. Finally, significant statistical differences were observed in both treatments regarding MFR and MBD values over the time. These results suggest that both conditioned media induced a functional perturbation, even if expressed only in some parameters, probably due to the high sensitivity of neuronal cells.

To evaluate the effect of conditioned medium produced by the cell lines SKBR-3 and MCF10-A on nervous cells, neurons (labeled with anti-Tub ßIII antibodies) and astrocytes (labeled with anti-GFAP antibodies), a qualitative morphological analysis was carried out. All samples were fixed at DIV 57 (D16). Immunofluorescence labeling showed striking differences in neuronal morphologies in the three conditions; specifically, in CTR neuronal cells displayed a fine structured and extended network, while both SKBR-3 and MCF10-A conditioned neuronal cells exhibited a coarsely structured and less extended network (Figure 5). Additionally, the conditioned neuronal networks, especially those conditioned with MCF10-A conditioned medium, showed a change in cell body morphology resembling a hypertrophic shape (Supplementary Figure S4). Ultrastructural examination of both SKBR-3 and MCF10-A conditioned neuronal cells showed the presence of a dense network with several connections and conventional synapses. No remarkable morphological differences were observed between these two conditions (Supplementary Figure S5).

Regarding the glial fraction, in the control networks, astrocytes displayed a conventional morphology (Tedesco et al., 2018), that is a highly ramified with extensive arborizations, while the conditioned cultures exhibited a modified morphology characterized by increased heterogeneity and loss of the normal structure (Figure 6A). At higher magnification, it was possible to appreciate and highlight the hypertrophic characteristics of astrocytes (Figure 6B). Indeed, the size of astrocyte bodies appeared shrinked in comparison to control. To quantify this qualitative observation, the fluorescent intensity and the percentage of covered area by GFAP-positive cells were assessed. The results confirmed the change in glial cells morphology (Supplementary Figure S6). Previous studies have reported similar results, demonstrating the modified morphological behavior of glial cells in response to the presence of metastatic breast cancer cells (Fitzgerald et al., 2008; Neman et al., 2014).

To evaluate the effects of SKBR-3 and MCF10-A cell conditioned medium (vesicles and soluble factors) on nervous cells, electrophysiological analyses were performed. The long-term electrophysiological impacts of exposing a co-culture of hiPSC-derived neurons and rat astrocytes to the conditioned medium were investigated using the MEA system. During the experiment, samples were subjected for 2 weeks to the medium conditioned by SKBR-3 and MCF10-A cells. A control group was also included, consisting of iNeurons-astrocytes co-cultured in iNeurons medium, as depicted in Figure 1. Quantitative parameters representing the behavior and functionality of the neuronal networks were derived from recorded spontaneous electrophysiological activity. These parameters were compared to determine whether conditioning induced any modifications in the physiological behavior of the network. Firstly, functional alterations were observed in all three conditions at D2, which is a common occurrence following medium changes. Therefore, the focus of characterization was primarily directed towards observing the effects 1 week after conditioning, allowing the networks to stabilize within the altered environment. The activity exhibited by SKBR-3 conditioned networks showed fluctuations influenced by the initial conditions, indicating a general increase in activity at D9. The statistical difference between SKBR-3 conditioned network and the control ones is highly significant. In the case of MCF10-A, a considerable increase in spiking activity, resulting in a firing frequency, was observed. SKBR-3 conditioned networks demonstrated an increase in bursting activity, consistent with qualitative observations of raster plots. Specifically, from D9 onwards, SKBR-3 conditioned networks exhibited a statistically significant increase in activity compared to the control ones. The control networks maintained a consistent trend with no significant changes, while MCF10-A conditioned ones, following a more critical initial adjustment, demonstrated patterns similar to those observed in the SKBR-3 conditioned samples. Burst activity is one of the main features of the development of network functionality (Wagenaar et al., 2006; Odawara et al., 2016) and efficient information conduction within the brain under normal physiological conditions, (Khazipov and Luhmann, 2006; Wagenaar et al., 2006; Egorov and Draguhn, 2013; Odawara et al., 2016); iNeuronal networks, after DIV41, should exhibit consistent and synchronous burst production frequencies. Bursting activity in vitro plays a crucial role in establishing functional connections between neurons and facilitating coordinated activity. However, uncontrolled bursting behavior or heavy strong bursting are considered a characteristic feature of the epileptic brain: in fact, they are also considered as seizure-like events in MEA recordings (Odawara et al., 2016; Chong et al., 2018; Liu et al., 2018). Specifically, breast cancer is considered a prominent factor in the development of epilepsy originating from brain tumors (Cosgrove et al., 2022). Epilepsy is characterized by abnormal neuronal activity, wherein the occurrence, spread, and cessation of epileptic seizures depend on interconnected networks of neurons that engage in synaptic and non-synaptic interactions (Wu et al., 2015). In epilepsy, the synchronized bursts of neuronal activity become excessive, unregulated, and disruptive, leading to abnormal electrical discharges and potential seizure events (Odawara et al., 2016). The increase in burst frequency, as well as the presence of disorganized and uncontrolled functional events observed in neuronal networks conditioned by SKBR-3 cells, aligns with the results reported in the existing literature regarding epilepsy events. These findings provide further support and can be explained by the previous studies conducted in this field, suggesting that epileptic events may be linked or induced by the tumor’s conditioned medium. Additionally, the analysis of mean burst duration (MBD) demonstrated a consistent decreasing trend in all samples. In the control networks and SKBR-3 conditioned ones, the values of MBD showed a tendency to stabilize over the observation period. This suggests that the electrical activity or the specific parameter being measured reached a relatively steady state in these groups. The decrease in burst duration observed confirmed the previously detected increase in burst frequency in networks conditioned by SKBR-3. However, in the MCF10-A conditioned networks higher values were observed compared to the other two groups of networks. The control ones maintained a consistent electrophysiological trend, with minor random variations in spike activity compared to D0. On the other hand, SKBR-3 conditioned networks exhibited significant oscillations around the initial value. The observed changes were described as oscillating periods, with distinct characteristics. Specifically, the neuronal activity exhibited highly synchronous periods, demonstrating a more coherent and structured pattern. These synchronous periods alternated with periods of disorganization, where the activity became more irregular. Supplementary Figure S2 (Supplementary Material) provides further evidence and visual representation of these alternating patterns. Meanwhile, MCF10-A conditioned networks demonstrated a reorganization and maintenance of network activity.

Finally, electrophysiological data indicated that the neuronal networks exhibited a substantial and consistent reaction to the conditioned medium of both the SKBR-3 cell line (a breast cancer cell line) and the MCF10-A mammary non-tumorigenic cell line. The results suggest that the secreted factors from both cell lines have a significant impact on the functional behavior of the nervous model, potentially indicating their influence on cellular communication or signaling pathways. However, differences in electrophysiological behavior between SKBR-3 conditioned networks and MCF10-A suggest that neuronal dysfunction could potentially arise from factors not only attributed to the mass effect of the metastases, but also from substances secreted by cells in the tumor microenvironment. The observed changes in excitability and electrophysiological signals resembled interictal-like epileptic activity (Holmes and Lenck-Santini, 2006; Simon et al., 2020). The correlation between tumor growth and the development of epilepsy has been widely studied in the medical field. As it has already been observed in literature, changes in electrophysiological parameters have been linked to cognitive impairment in patients with epilepsy (Pelkonen et al., 2020). This similarity suggests that our model may capture early stages of nervous tissue disorganization, which could potentially lead to the development of tumor-associated epilepsy. Altered expression and/or activity of neurotransmitter receptors and ion channels can be responsible for inducing this specific activity, as demonstrated in previous research (Lévesque et al., 2018; Simon et al., 2020). Overall, these alterations contribute to the enhancement of glutamatergic transmission while reducing GABAergic activity. Furthermore, neuronal atrophy and/or disrupted connectivity have been observed also in chronic phases of rodent epilepsy models, (Lévesque et al., 2018). However, it is important to note that further investigation and analysis are necessary to fully understand the underlying mechanisms and implications of this alteration in electrophysiological activity.

The immunocytochemistry characterization allowed us to gain information on the morphology of the 2D neuronal networks after different conditioning. In SKBR-3 and MCF10-A conditioned networks, neurons showed a hypertrophic morphology and the formation of less dense networks, suggesting unhealthy conditions. Remarkable changes in astrocyte morphology were observed in response to the presence of breast cancer-conditioned medium, showing a significant reduction in ramification compared to the control networks. These results confirmed that an alteration of normal conditions occurred, considering that under normal physiological conditions, glial cells usually exhibit a highly ramified morphology (Fitzgerald et al., 2008; Neman et al., 2014). Moreover, also astrocytes showed a hypertrophic morphology after conditioning. Interestingly, a similar result was observed in patients undergoing neurosurgical resection of breast-to-brain metastasis (Neman et al., 2014). Neman et al. demonstrated that metastatic breast cancer cells were juxtaposed to reactive astrocytes expressing glial fibrillary acidic protein (GFAP) both in the peritumoral brain and tumor regions, which showed a hypertrophic morphology comparable to astrocytes in glial scars forming in response to traumatic brain injury (Herrmann et al., 2008). Studies have demonstrated that astrocyte activation in the presence of pathological conditions like trauma, ischemia, and neurodegenerative diseases can be a protective mechanism for neurons, preventing injury-induced apoptosis (Neman et al., 2013). Furthermore, astrocytes play vital roles in various aspects of neural function, promoting nutrients transportation to neurons and oligodendrocytes. Additionally, astrocytes are involved in neural signal transduction, supporting the transmission of signals between neurons and the maintenance of the ionic balance of the extracellular matrix. In conclusion, alterations in astrocytes can be directly linked to neuronal electrophysiological activity; therefore, morphological results of this study can support, explain, and justify the electrophysiological ones.