Differential expression of genes related to Ca2+ signal pathways in TSCSection 1: quality control of RNA sequencing of TSC brain tissue

In the present study, we conducted RNA-Seq analysis of datasets consisting of tissue samples from TSC cortical tuber (n = 21) and age- and tissue-matched control cortex tissues (n = 15) to elucidate differentially expressed genes, enriched pathways, and putative molecular mechanisms associated with TSC.

A PCA was performed to explore the underlying dimensions on which most variance is observed within the dataset, specifically examining whether the variance of the data was driven by the diagnosed pathology and/or the area of resection. The scatter plots of the samples based on the scores of PC1 and PC2 for pathology diagnosed (Online Resource 3a), and area of resection (Online Resource 3b) did not exhibit distinct divergent clustering patterns.

Section 2: differential expression and GO pathway analysis

Exploring the transcriptional profile of TSC through differential expression analysis revealed significant alterations in gene expression between TSC and control cortex. A total of 3772 genes exhibited statistically significant differential expression (Log2FC ≥  ≤  ± 0.5; p.adj < 0.05), including 2104 upregulated genes and 981 downregulated genes (Online Resource 3c).

To gain insights into the functional role of the 3772 differentially expressed genes identified in our study, a Gene Ontology (GO) pathway enrichment analysis was performed on the upregulated and downregulated sets of genes separately. Among the 2104 upregulated genes, 764 GO Biological Processes (GO_BP), 43 GO Molecular Function (GO_MF), and 91 GO Cellular Component (GO_CC) terms exhibited significant enrichment (Online Resource 4, 5 and 6). The analysis revealed that most significantly enriched pathways include neutrophil activation and degranulation involved in immune response, I-kappaB kinase/NF-kappaB signaling, positive regulation of cytokine production, T cell activation, gliogenesis and glial cell differentiation supporting the strong inflammatory phenotype and astrogliosis observed in TSC. Subsequently, a GO pathway enrichment analysis was performed on the 981 downregulated genes. Enriched terms within GO_BP included pathways involved in neurotransmitter secretion and transport, regulation of membrane potential, and regulation of cation channel activity (Fig. 1a, shows the top 20 enriched pathways). The enriched terms within GO_MF involved in voltage-gated Ca2+ channel activity, mitochondrial protein-containing complex, and neurotransmitter receptor complex (Fig. 1b shows the top nine enriched pathways). The enriched compartments (GO_CC) among the downregulated genes involved mitochondrial protein-containing complex, postsynaptic density, ion channel complex and Ca2+ channel complex (Fig. 1c shows the top 20 enriched pathways). The complete list of significantly enriched GO pathways is reported in Online Resource 7, 8, and 9.

Fig. 1

Bulk RNASeq analysis of control cortex and TSC cohort. a. Visualization of the significantly enriched GO Biological Processes (BP) pathways (p.adj < 0.05) among the 981 downregulated genes. The size of each point is proportional to the number of differentially expressed genes present in the pathway, while the intensity of the color indicates the p.adj of the pathway enrichment. b. Visualization of the significantly enriched GO Molecular Function (MF) pathways (p.adj < 0.05) among the 981 downregulated genes. The size of each point is proportional to the number of differentially expressed genes present in the pathway, while the intensity of the color indicates the p.adj of the pathway enrichment. c. Visualization of the significantly enriched GO Cellular Components (CC) pathways (p.adj < 0.05) among the 981 downregulated genes. The size of each point is proportional to the number of differentially expressed genes present in the pathway, while the intensity of the color indicates the p.adj of the pathway enrichment. d. Supervised clustering heatmap visualizing the expression of the significant Ca2+-related genes in control tissues and TSC cortical tubers. Adjusted p value: *p.adj ≤ 0.05; **p.adj ≤ 0.01; ** p.adj ≤ 0.001; ****p.adj ≤ 0.0001

Section 3: differential expression of genes related to Ca2+ signaling pathways

The GO pathway analysis of differentially expressed genes in TSC provided valuable insights into their functional implications, emphasizing the potential involvement of cellular respiration, mitochondria, Ca2+ regulation, and neurotransmitter signaling. To further investigate these findings, we examined the expression of specific genes related to Ca2+ homeostasis, including Ca2+ buffers, transducers, regulators, and channels, as well as genes involved in mitochondrial and ER Ca2+ homeostasis, Ca2+ regulated inflammation, and G-alpha-q/11 coupled GPCRs (Online Resource 10). These investigations aimed to gain a deeper understanding of the molecular mechanisms underlying the observed functional alterations in TSC (Fig. 1d; Online Resource 3d). None of the genes, involved in the general Ca2+ homeostasis, were differentially expressed in TSC compared to controls, however the expression of multiple genes encoding for proteins involved in Ca2+ release from the ER and SOCE were altered. ITPR1 and ITPR2 encode for two of the Inositol 1,4,5-Trisphosphate Receptor Type 1 and 2 (IP3R1 and IP3R2) and the latter was upregulated (Log2FC = 1.25; p.adj = 0.001). IP3Rs play a crucial role in mediating Ca2+ release from the ER in response to inositol-1,4,5-trisphosphate-mediated signaling, thereby generating cytoplasmic Ca2+ signals and facilitating Ca2+ influx into the mitochondrial matrix to regulate oxidative metabolism and cell survival [10, 33, 85]. Ca2+ release from the ER into the cytoplasm is mediated by Ryanodine Receptor 1–3 (RyR1-3), a Ca2+ channel encoded by RYR1 (not differentially expressed), RYR2 (Log2FC = − 0.65; p.adj = 0.015), and RYR3 (Log2FC = 0.64; p.adj = 0.045). In TSC, differential expression of these genes combined with the altered expression of IP3R, suggests an impairment in the regulatory mechanism controlling Ca2+ release from the ER [92].

The mechanism regulating Ca2+ entrance in the cytoplasm is mediated by the ORAI Calcium Release-Activated Calcium Modulator 1–3 (ORAI1, ORAI2, ORAI3), a membrane Ca2+ channel that is activated by the Ca2+ sensor STIM1-2 (STIM1, STIM2) when Ca2+ stores are depleted [58]. Furthermore, Ca2+ ions influx in the cytoplasm is also mediated by Transient Receptor Potential Cation Channel Subfamily C Member 1, 3, and 4 (TRPC1, TRPC3, TRPC4), a membrane protein that can form a non-selective channel permeable to Ca2+ [58]. From the differential expression analysis, the data suggest a mild alteration at transcriptional level of the mechanism of Ca2+ entry in the cytoplasm and subsequently in the ER as only TRPC3 (Log2FC = -0.65; p.adj = 0.037) and ORAI3 (Log2FC = 0.71; p.adj = 0.013) expression was impaired. In contrast to the possible alteration in the SOCE, genes involved in the [Ca2+]ER homeostasis such as calreticulin (CALR) and calnexin (CANX) were not differentially expressed. In the mitochondria, Ca2+ uptake occurs through voltage-dependent anion-selective channel proteins (VDAC1) on the outer mitochondria membrane (OMM) and the mitochondrial Ca2+ uniporter (MCU, MICU1, MICU2, MICU3, MCUB) complex localized on IMM [41]. Subsequently, Ca2+ is released back from the mitochondrial matrix to the intermembrane space (IMS) via NCLX (SLC8B1) and Leucine zipper EF-hand- containing transmembrane protein 1 (LETM1). Our transcriptomic data showed the upregulation of MCUB (Log2FC = 1.38; p.adj = 0.001) and SLC8B1 (Log2FC = 0.82; p.adj = 0.0001) suggesting a reduction in Ca2+ entry into the mitochondria with a potential increase in Ca2+ efflux, resulting in reduced-Ca2+ dynamics in the mitochondrial matrix.

Further, the expression of most relevant genes encoding for Ca2+ buffer regulators was compromised. Protein Phosphatase 3 Catalytic Subunit Alpha (PPP3CA, Log2FC = − 0.72; p.adj = 0.026) enables several functions, including ATPase binding activity, CaM binding activity, and CaM -dependent protein phosphatase activity, calcineurin-NFAT signaling cascade, peptidyl–serine dephosphorylation and response to Ca2+ ion. RCAN3 (Log2FC = 0.73; p.adj = 0.004) encodes for RCAN Family Member 3 and is involved in Ca2+-mediated signaling, inhibiting calcineurin-dependent transcriptional responses by binding to the catalytic domain of calcineurin A [72, 76].

Given the established involvement of the immune response in the epileptogenesis of TSC corroborated by the significant enrichment of inflammatory response pathways in our study, the expression of genes associated with Ca2+ regulated inflammation was investigated. Nuclear Factor of Activated T Cells 1–4 (NFATC1, NFATC2, NFATC3, NFATC4) are DNA-binding transcription complexes which play a central role in inducible gene transcription during immune response. The upregulation of NFATC1 (Log2FC = 1.66; p.adj = 0.001), NFATC2 (Log2FC = 1.8; p.adj = 4.233e–05) suggests a possible interplay between the impairment of Ca2+ pathways and inflammatory response in TSC.

Purinergic Receptor P2Y2 (P2RY2) is a G protein-coupled receptor that triggers intracellular Ca2+ release and influx, leading to changes in cytosolic Ca2+ concentration. Activation of P2RY2 can trigger Ca2+ influx from the extracellular space, mediated by TRPC and further modulate Ca2+ entry pathways, such as SOCE [108]. In our TSC cohort, P2RY2 was upregulated with a Log2FC = 3.35 and p.adj = 0.0004, suggesting increased intracellular Ca2+ mobilization.

In conclusion, the comprehensive transcriptional analysis of TSC unveiled substantial alterations in pathways associated with cellular respiration, ER and mitochondria, Ca2+ regulation, and neurotransmitter signaling, in addition to the well-established dysregulated immune response in TSC. Specifically, the expression of genes involved in general Ca2+ homeostasis indicated a potential impairment in Ca2+ handling. Furthermore, the dysregulation of SOCE revealed a disturbance in Ca2+ entry mechanisms, while the impairment in the regulatory mechanism controlling Ca2+ release from the ER suggested an imbalance in intracellular Ca2+ dynamics accompanied by alterations in the Ca2+ influx and efflux mechanisms in the mitochondria. In addition, our data indicated the possible interplay between disturbances in Ca2+ signaling and inflammatory response in TSC. Collectively, these findings contribute to our understanding of the complex Ca2+ dysregulation and its potential involvement in the pathophysiology of TSC.

Section 4: differential expression of genes related to Ca2+ signaling pathways in astrocytes

In the field of epilepsy research, investigations into Ca2+ signaling dynamics have predominantly concentrated on neurons, providing insights into the complexities of neuronal dysfunction. Bulk sequencing approaches have been utilized in revealing alteration in Ca2+-related processes within these neuronal population and tissue of patients as mentioned in Sect. 3. However, to increase our knowledge on epilepsy and seizures, we should investigate beyond just neurons and consider a wider perspective. Astrocytes play an important role in maintaining brain homeostasis, but despite their significance, the exploration of Ca2+signaling in astrocytes has remained less explored in the context of epilepsy. Therefore, by investigating these glial cells, we aim to unveil more specific glial dysregulation of calcium signaling, shedding light on the contribution of astrocytes in epilepsy. Using single-cell RNA sequencing (scRNA-seq), we identified seven different cell types in our frontal cortex samples (Online Resource 11). To investigate the Ca2+signaling in astrocytes, we extracted the astrocytes from the dataset, and we looked at the expression profiles of Ca2+-related genes. By comparing control (n = 3) and TSC (n = 5) samples, we identified eight differentially expressed genes that were related to calcium signaling as mentioned in Sect. 3 (Fig. 2). ATP2B4 (Log2FC = 0.789; p.adj = 0.0229), SLC8A3 (Log2FC = 1.232; p.adj = 0.006), ITPR3 (Log2FC = 3.641; p.adj = 0.003), CANX (Log2FC = 1.636; p.adj = 0.002), MCUB (Log2FC = 2.907; p.adj = 1,55E-05), SARAF (Log2FC = 0.568; p.adj = 0.037) and P2RY2 (Log2FC = 2.897; p.adj = 0.005) were found to be upregulated, while NFATC4 (Log2FC = − 0.699; p.adj = 0.022) was found to be downregulated in TSC.

Fig. 2

Differential expression of calcium related genes using single-cell RNA sequencing. Control and TSC tissues were sequenced using the 10 × genomics Chromium Single Cell Gene Expression Flex protocol. From all cells, astrocytes were extracted computationally, and differential expression analysis was performed using DESeq2 between control (n = 3) and TSC (n = 5) samples. Data are expressed as mean ± SEM. Adjusted p value: *p.adj ≤ 0.05; **p.adj ≤ 0.01; ****p.adj ≤ 0.0001)

Proteomics

We conducted transcriptomic analysis at both bulk and single-cell levels; however, recognizing the potential disparities between transcriptomic and proteomic profiles, we aimed to complement our finding by performing proteomic analysis specifically in primary astrocyte cultures. Previous studies were conducted to verify the maturity of our control astrocytes. Despite being derived from younger individuals compared to the TSC astrocytes, they still exhibited a higher level of maturity[62] and similar levels of maturity to cultures used in previous studies obtained from adult individuals[5, 6]. Control and TSC primary astrocyte were cultured to conduct proteomic analysis to elucidate protein abundance and enriched pathways associated with TSC. LogFoldChange ≥  ≤  ± 0.5 and p value ≤ 0.05 were used as the screening criteria for significantly differentially expressed proteins (DEPs). A total of 1660 proteins were quantified with the mass spectrometry analysis (Online Resource 12) among which 147 were differentially expressed, 76 upregulated while 71 downregulated in TSC compared to controls (Fig. 3a). The functional significance of all identified DEPs in the TSC primary cultures was explored with the ingenuity pathway analysis (IPA) that predicted a total 705 canonical pathways (Online Resource 13) of which 209 were significantly enriched. Among these, ten were predicted activated (positive zScore), while 67 were predicted inhibited (negative zScore). The remaining pathways had zero activity pattern predicted (zScore = 0, n = 6) or the activity pattern was not available (n = 113). Of relevance to TSC, we found enriched pathways involved in the extracellular matrix organization (ECM), immune response and cytokines signaling, autophagy, ferroptosis signaling pathway, Golgi-to-ER traffic, regulation of mitotic cell cycle, Slit/Robo pathways and collagen biosynthesis pathways (Fig. 3b, shows the top 20 enriched pathways). Lastly, the 1660 quantified proteins were next imported into the STRING database (STRING 12.0) to perform network interaction analysis of protein–protein relationships. The GO pathway analysis revealed strong enrichment of pathways involved in the calcium signaling and ATP metabolism (Online Resource 14).

Fig. 3

Ingenuity pathway analysis (IPA) on control and TSC primary astrocytes cultures. a. Volcano plot showing the differentially expressed proteins (DEPs) (p value ≤ 0.05) between control (n = 6) and TSC (n = 6) primary culture astrocytes. A total of 76 proteins were found to be upregulated and 71 downregulated in TSC compared to control astrocytes. b. Top-ranked 40 enriched canonical pathways predicted by IPA (p value ≤ 0.05). The size of each point is proportional to the number of differentially expressed proteins present in the pathway, while the intensity of the color indicates the p value of the predicted canonical signaling pathways

Reduced response to environmental changes and stimuli in TSC astrocytes

To explore the intracellular calcium dynamics in TSC we first explored the concentration of calcium in the cytosol of TSC primary astrocytes given the strong relationship between mTOR activation and CaM activity. Cytosolic Ca2+ responses upon stimulation were followed via Fura-2 probe, and no differences in basal [Ca2+]c level were reported (Fig. 4a). Both control and TSC astrocytes have been challenged with: DHPG (200 µM), a selective agonist of group I mGluRs, glutamate (200 µM) and ATP (200 µM). Upon all the applied stimuli, TSC astrocytes displayed a reduced responsiveness compared to controls. Indeed, upon DHPG 28% of control astrocytes responded with increased [Ca2+]c while the responding cells in TSC group were only the 20% (p value: 0.0354) (Fig. 4 b and e; Online Resource 15a). The difference in the percentage of responding cells is higher considering glutamate stimulation (31% control astrocytes; 11% TSC astrocytes, p value: 0.0135) (Fig. 4c, f; Online Resource 15b) and ATP stimulation (21% control astrocytes; 5% TSC astrocytes, p value: 0.0385) (Fig. 4d, 4g; Online Resource 15c). Taking into account only the responding cells in both the experimental groups, the [Ca2+]c peak (Fig. 4h, i, l) was not significatively different upon the applied stimuli (DHPG; Glutamate and ATP), and neither the curve profiles displayed any difference (Fig. 4e–g).

Fig. 4

Ca2+ imaging in the cytoplasm of control and TSC astrocytes. Control and TSC astrocytes, previously loaded with Fura-2/AM, were stimulated with 200 μM DHPG, 200 μM Glutamate or 200 μM ATP in Ca2+-containing KRB solution (n = 216 control cells and n = 140 TSC cells, form 12 independent coverslip). a. No difference in basal Ca2+ level could be observed. b. c. d. The percentage of responding astrocytes (peak upon stimulation > 0.05), has been investigated, in both experimental group (light green: not responding cells; blue: responding cells). e. f. g. Representative curves of cytosolic Ca2+ responses upon the applied stimulus. The blue curve showed the Ca2+ response of control astrocytes, while the dark orange curve showed the Ca2+ response of TSC astrocytes. h. i. l. The amplitude of cytosolic Ca2+ response, upon DHPG, Glutamate and ATP stimulation has been measured, considering only responding cells in both control and TSC astrocytes. Data are expressed as mean ± SEM; p value: *p value ≤ 0.05; **p value ≤ 0.01; ***p value ≤ 0.001; ****p value ≤ 0.0001. Mann–Whitney U test

Collectively, our findings suggest that TSC astrocytes exhibit a reduced ability to respond to stimuli, as evidenced by attenuated [Ca2+]c responses, underscoring their reduced capacity to react to environmental changes.

Furthermore, TSC astrocytes exhibited a trend toward lower [Ca2+]c compared to control astrocytes, although the difference did not reach statistical significance. These observations underscore the need for further investigations to elucidate the potential dysregulation of Ca2+ signaling pathways in TSC astrocytes and its implications in TSC pathophysiology.

TSC astrocytes displayed dysregulation of ER Ca2+ release and SOCE impairment

SOCE is a fundamental cellular process that regulates Ca2+ influx from the extracellular space into the cytosol and subsequently into the ER upon [Ca2+]ER depletion. During cellular homeostasis, the ER undergoes a continuous leakage of Ca2+ ions through numerous leak channels [53], a phenomenon consistently replenished by the activity of the SERCA protein [38, 66, 101]. In addition, the mTOR hyperactivation and the increased-ROS production in TSC impacts on the [Ca2+]ER, therefore in this study we explored ER Ca2+ handling and the ability to perform SOCE. To induce complete [Ca2+]ER depletion, thus enabling assessment of its ability to empty, a specific SERCA inhibitor TBHQ was employed. This treatment led to concentration-dependent Ca2+ accumulation within the cytosol and [Ca2+]ER stores depletion, and the consequent activation of STIM-ORAI protein interaction [87]. Subsequently, for quantifying ER replenishment capacity, Ca2+ stimulation was induced allowing the measurement of ER responsiveness to refilling demands.

In this study, Fura-2 dye was used to measure [Ca2+]c, in zero-Ca2+ KRB solution (within the first 30 s), ER Ca2+ release ability (upon TBHQ stimulation, between 30 and 300 s) and assess SOCE (upon Ca2+ stimulation at 300 s) (Fig. 5a).

Fig. 5

Ca2+ imaging in endoplasmic reticulum (ER) of control and TSC astrocytes. a. Control and TSC astrocytes, previously loaded with Fura-2/AM, were stimulated with TBHQ (50 s) and Ca2+ (300 s) in KRB solution. Representative curves of cytosolic Ca2+ responses upon the applied stimulus of three independent experiments. The blue curve showed the Ca2+ response of control astrocytes, while the dark orange curve showed the Ca2+ response of TSC astrocytes (n = 160 control cells and n = 136 TSC cells, from 14 independent coverslips). b. Bar plot visualizing significant reduction of Ca2+ concentration in the cytoplasm of TSC astrocytes compared to control astrocytes after blocking the SERCA protein with TBHQ stimuli (between 30 and 300 s). c. Bar plot visualizing basal cytosolic Ca2+ concentration in TSC astrocytes after TBHQ stimuli compared to controls. d. Bar plot visualizing significant reduction in the absolute concentration of Ca2+ in the ER of TSC astrocytes compared to control astrocytes after Ca2+ administration (at 300 s). Data are represented as mean ± SEM; p value: *p value ≤ 0.05; **p value ≤ 0.01; ***p value ≤ 0.001; ****p value ≤ 0.0001. Mann–Whitney U tests were used to evaluate significance

Upon treatment with 50 µM TBHQ, TSC astrocytes exhibited a significant reduction of Ca2+ release from the ER compared to control astrocytes (p value: 0.029 for the amplitude of [Ca2+]c, p value < 0,0001 for the baseline level after TBHQ treatment) (Fig. 5b, c), this suggests a reduction of [Ca2+]ER due to reduced-Ca2+ transport into ER and/or an increase of [Ca2+]ER leakage. Subsequently, cells were stimulated with Ca2+ to assess SOCE and TSC astrocytes displayed a decreased [Ca2+]c influx (p value < 0.0001) (Fig. 5d), indicating an impairment of SOCE. Overall, the reduction in basal [Ca2+]c levels, after SERCA blockade, and further Ca2+ stimulation suggest a dysregulation of release from the ER and SOCE in TSC astrocytes and might lead to a disruption in Ca2+ signaling and downstream Ca2+-dependent cellular mechanisms.

Reduced response to stimuli and altered Ca2+ dynamics in TSC mitochondria

The alterations detected from the previous experiments in Ca2+ storage in the ER, suggest a potential impairment in the ability to store Ca2+ and its dynamics in the different organelles in TSC astrocytes. Control and TSC astrocytes were transfected with 4mtD3cpv Ca2+ indicator, enabling targeted Ca2+ imaging in the mitochondria matrix in single cell [75].

A reduction in the basal [Ca2+]m was observed in TSC astrocytes when compared to control astrocytes (p value < 0.0001) (Fig. 6a, b and c). Subsequently, stimulation with 200 μM ATP was used to assess the mitochondrial Ca2+ uptake. It was observed that mitochondria in TSC astrocytes exhibited a significant reduction in Ca2+ influx compared to control astrocytes (Fig. 6d). Because the reduced ER Ca2+ content (Fig. 6) might affect Ca2+ dynamics in mitochondria, we used a Ca2+ ionophore, ionomycin, to induce an artificial cytosolic Ca2+ overload to expose control and TSC mitochondria to the same Ca2+ concentrations, independent of the ER Ca2+ releasing capacity. Upon treatment with 5 µM ionomycin, we found a significantly reduced (p value < 0.0001) Ca2+ influx in the mitochondrial matrix of TSC compared with control astrocytes, suggesting an intrinsic impairment of mitochondrial Ca2+ transport system (Fig. 6e).

Fig. 6

Mitochondrial Ca2+ imaging of D3-positively transfected mitochondria in control and TSC astrocytes. a. Representative curve of seven independent coverslips. Control and TSC astrocytes, previously transfected with D3-plasmid and stimulated with ATP and ionomycin in KRB solution. The blue curve showed the basal Ca2+ response of control astrocytes before and after ATP (30 s) and ionomycin (150 s) stimuli, while the dark orange curve showed the basal Ca2+ response of TSC astrocytes. b. Representative curves of seven independent coverslips. The basal Ca2+ levels were normalized. The blue curve showed the basal Ca2+ response of control astrocytes before and after ATP (30 s) and ionomycin (150 s) stimuli, while the dark orange curve showed the basal Ca2+ response of TSC astrocytes. c. The barplot showed the significant reduction concentration of mitochondrial Ca2+ in TSC astrocytes at homeostasis (at 50 s) compared to controls after ATP stimulation (n = 100 ROIs for control; n = 102 ROIs for TSC). d. The barplot showed the significant reduction of mitochondrial Ca2+ concentration in TSC astrocytes after ATP stimulation compared to control astrocytes (n = 100 ROIs for control; n = 102 ROIs for TSC). e. The barplot showed the significant reduction of mitochondrial Ca2+ concentration in TSC astrocytes after Ionomycin stimulation compared to control astrocytes (n = 70 ROIs for control n = 88 ROIs for TSC). Data are expressed as mean ± SEM; p value: *p value ≤ 0.05; **p value ≤ 0.01; ***p value ≤ 0.001; ****p value ≤ 0.0001. Mann–Whitney U tests were used to evaluate significance

These findings are supported by the transcriptional profile of genes modulating mitochondrial Ca2+ homeostasis. The substantial reduction of Ca2+ influx in mitochondrial matrix may be attributed to the compromised activity of the mitochondrial Ca2+ uniporter, potentially influenced by the elevated expression of MCUB, dominant Negative Subunit of MCU complex. Furthermore, the diminished [Ca2+]m may also stem from heightened Ca2+ efflux driven by the upregulation of NCLX localized on the IMM.

Significant depolarization of mitochondria membrane potential (ΔΨm) in TSC astrocytes

Ca2+ accumulation into mitochondria regulates its metabolism and the depolarization events of mitochondrial membrane potential. TSC astrocytes showed impairment in Ca2+ dynamics in mitochondria suggesting damages in mitochondrial membrane potential in TSC might occur. JC-1 staining was conducted to examine the ΔΨm in the experimental samples. Both control and TCS astrocytes exhibited 100% double positive staining for JC-1. The analysis of the experiments unveiled distinct subpopulations indicating different states of ΔΨm within each sample group based on the fluorescence intensity of the JC-1 dye: ‘PE low FITC low’, ‘PE high FITC high’ and ‘PE high FITC low’ (Fig. 7a). The ‘PE high FITC low’ population, constituting 97.2% and 90.1% of cells (control and TSC, respectively) were considered as cells with normally polarized mitochondria. Respectively, the ‘PE low FITC low’ and the ‘PE high FITC high’ population, which had a decreased PE/FITC (JC-1 aggregate/JC-1 monomer) ratio, were considered as cells with depolarized mitochondria. Notably, there was a significantly lower number of TSC cells in ‘PE high FITC low’ group compared to control cells (97.2%, p value: 0.0006) (Fig. 7d), while the number of TSC cells in the ‘PE high FITC high’ group was 8.8-fold higher compared to control cells (5.72%, p value: 0.0003) (Fig. 7b). No significant differences were found in ‘PE low FITC low’ group (Fig. 7c).

Fig. 7

Flow cytometry of JC-1 staining. a. Representative figure of 10,000 separate events. Flow cytometry dot-plots of JC-1 fluorescence to measure ΔΨm in control astrocytes (blue dots) and TSC astrocytes (orange dots). Three subpopulations of cells were identified in both control and TSC astrocytes: ‘PE high FITC low’, ‘PE low FITC low’, and ‘PE high FITC high’ fluorescence. b–d. Barplots showing the percentage of control and TSC astrocytes with a positive JC-1 fluorescence intensity signal. No differences were identified between control and TSC astrocytes in ‘PE low FITC low’ fluorescence signal. TSC astrocytes showed a significant increase in PE high FITC high fluorescence intensity, while they showed a significant reduction in PE high FITC low fluorescence signal. Data are expressed as mean ± SEM; p value: *p value ≤ 0.05; **p value ≤ 0.01; ***p value ≤ 0.001; ****p value ≤ 0.0001. Mann–Whitney U test were used to evaluate significance

These results indicate that mitochondria in TSC astrocytes are significantly depolarized compared to the controls potentially contributing to mitochondrial dysfunction.

Collectively, these findings provide compelling evidence of an alteration in ΔΨm, characterized by depolarization of the IMM, which suggests mitochondrial dysfunction in TSC astrocytes. The observed depolarization may indicate the presence of underlying cellular stress potentially damaging the components of the electron transport chain (ETC), impairing ATP synthesis or increased apoptosis leading to loss of mitochondria function. Therefore, we further explored the mitochondria respiratory capacity in TSC astrocytes.

Reduced oxygen consumption rate and reserve respiratory capacity in TSC astrocytes

The application of OROBOROS respirometry provided a comprehensive analysis of the respiration profiles of control and TSC astrocytes, enabling a deeper understanding of their mitochondrial function. This analysis performed in intact cells showed significant differences between control and TSC astrocytes in one of the three distinct phases of the mitochondrial respiration: ETC (p value: 0.016) (Fig. 8a and b). TSC astrocytes exhibited a reduction in basal oxygen consumption (R phase, Fig. 8b), indicating overall a lower energy demand compared to controls. The maximum capacity of the ETC was found to be significantly decreased (p value: 0.016) in TSC astrocytes when compared to control astrocytes (Fig. 8b). This reduction suggests a direct defect in the activity of the protein complexes involved in the ECT that led to an impaired ability of TSC astrocytes to generate ATP efficiently. While no difference was observed in ATP-linked respiration (Fig. 8c), TSC astrocytes showed a diminished reserve respiratory capacity (p value: 0.0286) compared to control astrocytes (Fig. 8d) further supporting the previous findings.

Fig. 8

Oxygen consumption rate of control and TSC astrocytes normalized to protein content. a. Representative tracing of high-resolution respirometry to quantify intact cell respiration; blue line: oxygen concentration, red line: oxygen flux. b. Oxygen flux in the routine state (R); in the leakage state (L) after addition of oligomycin, an inhibitor of ATP synthetase; after the addition of FCCP, an uncoupler of oxidative phosphorylation to induce maximum respiratory capacity (E). All data are expressed as specific flux, i.e., oxygen consumption normalized to the sample protein content and after non-mitochondrial oxygen flux subtraction (ROX). c. Oxygen consumption linked to ATP production, i.e., oligomycin-sensitive respiration obtained by the subtraction of L from R. d. Reserve respiratory capacity obtained by the subtraction of R from E. Data are expressed as mean ± SEM of folds changes above the control; p value: *p value ≤ 0.05; **p value ≤ 0.01; ***p value ≤ 0.001; ****p value ≤ 0.0001. Mann–Whitney U test

In conclusion, our findings reveal significant metabolic alterations in TSC astrocytes when compared to control astrocytes. The observed reductions in basal oxygen consumption, decreased ETC capacity, and diminished reserve respiratory capacity collectively suggest a compromised ability of TSC astrocytes to respond to heightened energy demands or cellular stress.

EM-revealed alterations in mitochondrial ultrastructure in TSC astrocytes

The observed alteration of mitochondrial calcium dynamics along with impairment in mitochondrial membrane depolarization and mitochondrial respiration might impact on their morphology and their ultrastructure in TSC astrocytes. EM was employed to examine the high-resolution morphology of mitochondria, enabling a detailed assessment of their structure, organization, and ultrastructural characteristics (Fig. 9a and b). Tissue samples obtained from individuals with TLE, and non-epileptogenic regions with histologically healthy-appearing neurons and astrocytes were selected as control specimens for comparative analysis with TSC tissues (Online Resource 1). Here, healthy-appearing astrocytes and neurons from individuals with TLE will be referred to as controls to evaluate mitochondrial ultrastructure. Several quantitative parameters of the mitochondria, including area, perimeter, aspect ratio (AR), roundness, circularity, Feret’s diameter, width, height, and integrated density, were measured to investigate potential alterations in mitochondrial ultrastructure in TSC astrocytes and neurons compared to the TLE control group. Mitochondria of TSC astrocytes showed significant alterations in their shape and elongation compared to control astrocytes. The observed significant reduction in perimeter and area (p value < 0.0001; p value < 0.0001, respectively), along with a significant increase in AR of mitochondria in TSC astrocytes (p value < 0.0001) (Fig. 9c, d and e), may indicate higher mitochondrial fragmentation associated with cellular dysfunction, increased fission/fusion, or changes in morphology due to cellular stress and/or changes in metabolic states [21]. Furthermore, TSC astrocytes exhibited a significant reduction in mitochondria circularity (p value < 0.0001), roundness (p value < 0.0001), and Feret’s diameter (p value < 0.0001) (Fig. 9f; Online Resource 16a and 16b) further corroborating the possibility of increased fragmentation, altered fission/fusion events, changes in the mitochondrial network, cristae structure, or remodeling of mitochondrial membranes. Lastly, the width (p value < 0.0001), the height (p value < 0.0001), and integrated density (p value < 0.0001) of mitochondria in TSC astrocytes were reduced (Fig. 9g and h; Online Resource 16c) suggesting a decrease in the size, impaired mitochondrial biogenesis, increased degradation, or dysfunction of the mitochondrial respiratory chain. These findings indicate potential alterations in the structural organization of mitochondria due to changes in mitochondrial dynamics, fusion, fission, or remodeling processes. A decrease in integrated density may also suggest mitochondrial depletion within the analyzed region.

Fig.