Therapeutic targeting of P2X4 receptor and mitochondrial metabolism in clear cell renal carcinoma models

Background Clear cell renal cell carcinoma (ccRCC) is the most common subtype of renal cancer. Large-scale metabolomic data have associated metabolic alterations with the pathogenesis and progression of renal carcinoma and have correlated mitochondrial activity with poor survival in a subset of patients. The aim of this study was to determine whether targeting mitochondria-lysosome interaction could be a novel therapeutic approach using patient-derived organoids as avatar for drug response. Methods RNAseq data analysis and immunohistochemistry were used to show overexpression of Purinergic receptor 4 (P2XR4) in clear cell carcinomas. Seahorse experiments, immunofluorescence and fluorescence cell sorting were used to demonstrate that P2XR4 regulates mitochondrial activity and the balance of radical oxygen species. Pharmacological inhibitors and genetic silencing promoted lysosomal damage, calcium overload in mitochondria and cell death via both necrosis and apoptosis. Finally, we established patient-derived organoids and murine xenograft models to investigate the antitumor effect of P2XR4 inhibition using imaging drug screening, viability assay and immunohistochemistry. Results Our data suggest that oxo-phosphorylation is the main source of tumor-derived ATP in a subset of ccRCC cells expressing P2XR4, which exerts a critical impact on tumor energy metabolism and mitochondrial activity. Prolonged mitochondrial failure induced by pharmacological inhibition or P2XR4 silencing was associated with increased oxygen radical species, changes in mitochondrial permeability (i.e., opening of the transition pore complex, dissipation of membrane potential, and calcium overload). Interestingly, higher mitochondrial activity in patient derived organoids was associated with greater sensitivity to P2XR4 inhibition and tumor reduction in a xenograft model. Conclusion Overall, our results suggest that the perturbed balance between lysosomal integrity and mitochondrial activity induced by P2XR4 inhibition may represent a new therapeutic strategy for a subset of patients with renal carcinoma and that individualized organoids may be help to predict drug efficacy. Supplementary Information The online version contains supplementary material available at 10.1186/s13046-023-02713-1.


Page 2 of 19
Background Deregulation of metabolism is a hallmark of cancer, including clear cell renal cell carcinoma (ccRCC), which is the most common subtype [1][2][3][4]. The development of combination therapies with immune checkpoint inhibitors and receptor tyrosine kinase inhibitors (RTKI) has revolutionized the treatment of kidney cancer. However, tumor heterogeneity remains a major challenge [5][6][7]. Increasing evidence suggests that aberrant lipid accumulation, altered metabolism, and autophagy confer a high-energy state and resistance to ccRCC treatment [3,8,9]. Integration of transcriptomic and metabolomic data selected a kidney cancer subgroup independently from histology, characterized by strengthened mitochondrial and weakened angiogenesis-related gene signatures [10]. Metabolomic analysis has identified several metabolites related to the biosynthesis of glutathione, an antioxidant that buffers reactive oxygen species produced by mitochondria, in a particularly aggressive ccRCC subclass [11]. However, the flux and activity of metabolic reactions are fundamental for evaluating the correlation between metabolism and clinical aggressiveness of ccRCC, and only limited data are available [12]. Therefore, targeting mitochondria may represent a novel therapeutic approach to inhibit tumor growth and/or enhance their sensitivity to therapy [13,14]. The mechanisms leading to the selective disruption of tumor mitochondria are largely unknown. Compared to other intracellular organelles, the dense hydrophobic double-membrane system and negative membrane potential of mitochondria complicate the identification of tumor-selective drug targets [15], similar to the acidic pH of lysosomes. Lysosomes are intracellular Ca 2+ hubs that are essential for membrane trafficking and signaling [16,17]. We have previously shown that lysosomal purinergic receptor 4 (P2XR4), an ATP/Ca 2+ pump [18], plays a key role in energy flux and drives motility and proliferation of endothelial cells [19]. Conversely, blockage of P2XR4 by 5-BDBD inhibits the transport of lysosomal P2XR4 to, and activity on, the cell membrane and reduces tumor angiogenesis [19]. Here, we investigated the role of P2XR4 in metabolic and energy dynamics in ccRCC cell lines. We used patient-derived organoids as personalized drug screening, system and xenograft mouse models to investigate the sensitivity of tumors to P2XR4 inhibition. We reported that P2XR4 inhibition affects tumor cell survival and mitochondrial dynamics in a specific subset of ccRCC characterized by high mitochondrial content and activity, two parameters recently reported to be associated with poor prognosis [11].

Organoids and samples collection
Patient-derived organoids were generated as previously reported [21] using a protocol approved by the Research Ethics Committee of the University of Campania (2017SZ0064) and was conducted in accordance with the Helsinki Declaration. All patients provided written informed consent prior to acquisition of tissue. Tumor histology was assessed by pathologists. Tissues from kidney carcinomas were minced and subjected to enzymatic digestion in 5 ml of collagenase IV (5 mg/ml) for 1 h at 37 °C. Cell pellets were resuspended in cold organoid culture medium (see Supplementary Information), and 5 × 10 4 cells in 40 μl droplets with 2% Matrigel were deposited into U 96-well plates. Organoids were passaged every 2-3 weeks at a split ratio of 1:2-1:3 with TripLE (Gibco) and cultured with DMEM/F12, 200 mM GlutaMAx, 1 M Hepes, 1 × B27,100 ng/ml FGF-10 50 ng/ ml EGF (PeproTech).

Image-based drug screening in organoids
For drug screening, 5 × 10 4 cells in 40 μl droplets with 2% Matrigel were deposited into U 96-well plates and cultured in 100µl of medium. When they reached a size of 200-300 μm after three days of culturing, they were transferred into wells with basket. Test compounds were added to apical and basolateral compartments and equilibrated [22]. Culture medium supplemented with different drug concentrations was replaced every day until harvest time. 20 PDO replicates of individual biopsies were tested with each drug dose. Organoids were exposed to control (0.1% DMSO) or drugs (5-BDBD, everolimus and EZD8055) at different doses for additional 96 h. Images were recorded over time by a Zeiss microscope and sizes of organoids analyzed by ImageJ and CELIGO System (version 1.51). To determine viability, organoids were seeded in 384-well culture plates at 2.5 × 10 4 cells/15μL/ well in organoid medium containing 2% Matrigel. When organoids reached 200µm, 10-fold drug dilution series from 0 to 100 or 200 μM were added to 20 organoid biological replicates. After 96 h, cell viability was quantified using a CellTiter-Glo 3D (ATP) luminescent assay according to the manufacturer's instructions (Promega), or MTT assay kit and TECAN Infinity 2000 Elisa reader system. The absorbance value of untreated organoid controls (n = 20) was considered equal to 1 then the value of organoids treated (n = 20 each dose) was reported as ratio of controls. Dose-response curves were fitted using nonlinear squares regression with variable slope (four parameters). Half-maximal inhibitory concentration (IC50) and the fraction of dead cells were determined by the area under the curve (AUC), using GraphPad Prism 9. Statistical evaluation was by two-way ANOVA, with Dunnett's multiple comparison test comparing each treatment to insensitive PDOs.

Immunohistochemistry analysis
Consecutive sections of formalin-fixed, paraffin-embeddedtumors were subjected to IHC analysis for the P2XR4 receptor using a DAB substrate kit (Maixin, Fuzhou, China), following the manufacturer's instructions. The slides were deparaffinized in xylene and rehydrated using graded ethanol to water ratios before staining. All sections were treated with EDTA (pH 8.0) for antigen retrieval and with 3% H 2 O 2 to inactivate endogenous peroxidases. Sections were incubated with anti-P2XR4 (Invitrogen, PA5-83,466) and CD31 (Dako Omnis; Clone MIB-1) anti-mitochondrial antibodies (SPM198; Abcam) at 1:200 dilution overnight at 4 °C. After washing, sections were stained with a secondary antibody for 30 min at room temperature. Phosphate-buffered saline was substituted for each primary antibody as a negative control. Five random fields were examined under a light microscope. The IHC staining results were independently evaluated by two pathologists who were blinded to the clinical data. The staining intensity was scored as 0 (negative), 1 (low), 2 (medium), and 3 (high). The staining extent was scored as 0 (0% stained), 1 (1%-25% stained), 2 (26%-50% stained), and 3 (51%-100% stained). The final score was determined by multiplying the intensity by the staining extent, and ranged from 0 to 9. Final scores ≤ 4 were considered low staining and > 4 high staining.

Cell proliferation assay
Cell proliferation was determined by the MTT assay (Sigma, St. Louis, MO, USA), according to the manufacturer's protocol. Cells (2 × 10 3 ) were seeded in 96-well plates and exposed to various concentrations of 5-BDBD for 24 h. Assays were read at 570 nm using a TECAN Infinity 2000 microplate reader (Molecular Devices, Sunyvale, CA). Cell proliferation was measured on different days and reported as absorbance at 570 nm or vitality as ratio of controls.

Seahorse assay
Mitochondrial and glycolytic rates were measured using the Mitochondrial Stress Test (MST) and Glycolysis Stress Test (GST) Agilent assay kits using a Seahorse XF96 extracellular flux analyzer (Seahorse Biosciences, Billerica, MA, USA). Cells (1.8 × 10 4 per well) were plated in quadruplicate in XF96 extracellular flux assay plates with 200 μL of XF base medium equilibrated to a pH of 7.4. For organoids assay we follow the Star protocol [22]. Three PDOs with size of 50µm were plated in individual wells of 96 well plates. After adherence for 2 h at 37 °C, media was replaced with GST or MST buffer. For MST, were added the following compounds: oligomycin (1 μM), FCCP (1 μM), and rotenone/antimycin A (0.5 μM). For GST were added: glucose (10 mM), oligomycin (1 μM), and 2-DG (50 mM). Values for each measurement were averaged across quadruplicate wells and reported as the oxygen consumption rate (OCR, pmol O 2 /min) for MST and extracellular acidification rate (ECAR, mpH/min) for GST. Different concentrations of 5-BDBD or 0.1% DMSO were added 5 min before starting MST and GST assays. Data were reported per microgram of protein extracted at the end of the experiment.

Western blots and immunoprecipitation
Briefly, 30 μg of protein was lysed in ice-cold RIPA buffer with a protease inhibitor cocktail and phosphatase inhibitors (Sigma, MO, USA), loaded on a 12% Tris-glycine SDS polyacrylamide gel, and then transferred to a nitrocellulose membrane. Membranes were blocked with 5% milk in Tris-buffered saline with 0.1% Tween-20 for 20 min prior to the application of 5-BDBD (5 μM, TOCRIS). The lysosomal contribution to cytosolic Ca 2+ was tested by pretreatment of cells with an L-L-MA working solution of 10 μM (125,130,250, Thermo Fisher).

Determination of intracellular oxidative stress
Intracellular oxidative stress after 5-BDBD treatment was determined using a 2′,7′-dichlorofluorescein-diacetate (DCFDA/ H2DCFDA-Cellular ROS Assay Kit (ab113851, Abcam). The cells were seeded in 96-well plates at (1 × 10 3 ) cells/well and cultured overnight. Cells were then preincubated with 50 μM DCFDA for 15 min and treated for an additional 15 min with 0.5 μM 5-BDBD, or 0.1% DMSO a negative control, or 100 μM H 2 O 2 , or 25 mM CaCl 2 , or carbonyl cyanide m-chlorophenyl hydrazone 10 μM CCCP as positive controls [23]. Fluorescence images were captured by Zeiss confocal microscope and analyzed the intensity with ZEN 2.0 program (Germany). The mean of fluorescence intensity at 488 nm was determined in triplicate well experiments using TECAN Infinity 2000 Elisa reader. Glutathione determination was performed with glutathione colorimetric detection kit (Invitrogen) following kit guide accordingly with to manufacture's instruction.

Mitochondrial Ca 2+ measurement
Mitochondrial Ca 2+ levels were measured using the Rhod-2 AM dye (Abcam). For flow cytometry, 10 5 plated cells were culture in 0.1% DMSO or 0.5 μM 5-BDBD (treated cells) with Rhod-2 AM dye (5 μM) for 30 min at 37 °C. Cells were then washed twice, suspended in 500 μL of PBS, and analyzed by flow cytometry. For live microscopy image, cells were grown in microplates Ibidi. Then 5 μM of Rhod-2 AM dye and 1 μM and MitoTracker green were added to the media (Cell Signaling Technologies) for 10 min, followed by 0.1% DMSO or 5 μM 5-BDBD for an additional 20 min. The images were captured using a confocal Zeiss microscope and analyzed by ZEN 2.0 program (Germany).

Mitochondrial membrane potential
Control and cells treated with 5-BDBD (0.5 μM for 15 min) were resuspended in 1 mL medium containing 5 μM JC-1 mitochondrial membrane potential probe (e-CK-A301 Elabscience). After incubation at 37° C for 30 min, the cells were washed twice, resuspended in 500 μL of PBS, and analyzed by flow cytometry. The percentage of cells that exhibited red and green fluorescence was quantified. Carbonyl cyanide m-chlorophenyl hydrazone (CCCP), an inhibitor of mitochondrial oxidative phosphorylation that causes ROS-induced mitochondrial depolarization [23], was used as a positive control. The cells were treated with 10 μM CCCP for 15 min prior to flow cytometry. Fixed cells were photographed using a confocal microscope (Zeiss, Germany) and analyzed using ImageJ.

Apoptosis assay
Cells (1 × 10 5 ) were treated with 0.1% DMSO or 5 μM 5-BDBD for 24 h, dissociated, washed in PBS, and incubated in Annexin V binding buffer containing annexin V-FITC and PI (BD Biosciences) for 15 min at room temperature. The stained cells were diluted 1:5 in binding buffer and analyzed using a FACSAria III Flow Cytometer (BD Biosciences).

In vivo mouse model
A tumor xenograft model was established as described in our previous report [19]. Briefly, A-498 cells (5 × 10 6 cells per mouse) were subcutaneously injected into the right flanks of 6-week-old BALB/c female nude mice. Equal-sized groups were generated using a randomized experimental design. Mice (n = 3 per cage) were housed in a temperature-controlled (22-24 ℃) environment with a relative humidity of 60-70% and a 12 h light/12 h dark cycle. All animals received water and food ad libitum. When the tumor size reached approximately 100 mm 3 , mice were randomly assigned to treatment groups (PBS or 10 mg/kg 5-BDBD administered by intraperitoneal injection every 3 days for 55 days, n = 5 per group). Before each administration, the body weights of the mice were monitored. Simultaneously, the tumor volumes were measured and calculated according to the formula: tumor volume (mm 3 ) = 0.5 × (length of tumor) × (width of tumor). At the end of the experiment, the mice were killed by an overdose of pentobarbitone (200 mg/kg), and the tumors were dissected, fixed, and used for immunohistochemical analysis.

Statistical analysis
All data were analyzed by one-way ANOVA with Tukey's post hoc analysis using or Student's t test using Graph-Pad Prism 9 (LA USA). Dose-response curves were fitted using nonlinear squares regression with variable slope (four parameters). Half-maximal inhibitory concentration (IC50) and the fraction of dead cells were determined by the area under the curve (AUC), using GraphPad Prism 9. Statistical evaluation was by two-way ANOVA, and Dunnett's multiple comparison test was used to compare each treatment to control. results are expressed as mean ± SEM or SD. The threshold (α) of significance was set at p < 0.05.

P2XR4 overexpression in ccRCC correlates with poor prognosis.
Given that our previous study had established that P2XR4 affects endothelial mitochondrial activity [19], we now investigated its role in ccRCC metabolism. We began by evaluating P2XR4 gene and protein expression in human ccRCC. Bioinformatic analysis of the RNAseq database of human samples from The Cancer Genome Atlas (TCGA, accessed using the Xena browser), indicated that P2X4 mRNA was upregulated in a cohort of primary ccRCC (n = 945), compared to adjacent non-tumor tissues (Fig. 1A, p = 0.0032). This was also confirmed using the GEPIA browser (Fig. 1A, p = 0.0022). P2X4 mRNA was a variable independent of tumor grade (I-III) and its high expression correlated with poor overall survival (OS) (p = 0.043 and = 0.00028) (Fig. 1B). P2XR4 protein, assessed by immunohistochemistry in the Human Protein Atlas database (ATLAS) showed positive staining in clear cell carcinoma sections (Fig. 1C). The signal was moderate in 49% and strong in 19% of the cases, and predominantly localized in the cytoplasm and on membranes. Immunohistochemistry of our primary ccRCC tissues confirmed greater P2XR4 expression in tumors than in adjacent nonmalignant tissues, as shown in representative images (Fig. 1D). Additionally. microarrays featuring normal (N = 30) and malignant tissues (N = 48) confirmed a significantly enhanced expression of P2XR4 protein reported as a histoscore in tumor compared to normal tissues (Fig. 1F). In A-498 and SNC-12 cells, P2XR4 colocalized with the main lysosomal protein, LAMP1, whereas it was expressed at lower levels in the 786-0 clear carcinoma cell line and in normal epithelial renal cells (HRE) then in A-498 and SNC-12 cells (Fig. 1E, G).

Mitochondrial metabolism in ccRCC
Correlative analysis showed a significant association between P2X4 gene expression and glutathione peroxidase (GPX4, p = 6.8 −6 ), superoxide dismutase (SOD2, p = 2.3 -9 ) involved in the detoxification of radical oxygen, and exokinase 2 (HK2, p = 7 −12 ), which regulates the integrity of mitochondrial membrane (Suppl. Figure 1). We therefore investigated the role of P2XR4 in intracellular Ca +2 homeostasis and mitochondrial activity in ccRCC cells [24]. Confocal microscopy and FACS analysis using mitotracker staining showed twofold greater mitochondrial fluorescence intensity in A-498 cells than in SNC-12 cells, with both being higher than that in normal renal epithelial cells (HRE) ( Fig. 2A,B). The lowest mitochondrial fluorescence was observed in 786-0 cells ( Fig. 2A,B) consistent with the mean fluorescence intensity measured by FACS (Fig. 2B,C). Then mitochondrial activity was determined by dosing the oxygen consumption rate (OCR) over time and respiratory parameters calculated using inhibitors of mitochondrial function (oligomycin, FCCP and Rotenone + Antimycin) (Fig. 2D). Results showed that A-498 and SNC-12 cells had higher OCRs than 786-0 cells (Fig. 2D). OCRs associated with maximal respiration of SNC-12 and 786-O cells were 2.3-and fivefold smaller than those of A-498 cells, respectively, as well as ATP production (Fig. 2E). In contrast, the extracellular acidification rate (ECAR) in 786-0 cells was 1.8-and 2.0-fold greater than that in A-498 and SNC-12 cells, respectively (Fig. 2F). More importantly, in A-498 and SNC-12 cell lines, the OXOPHO rate (monitored as OCR) was greater than glycolysis activity measured as extracellular acidification rate (ECAR), indicating that in these cells, mitochondrial metabolism is the principal energy source instead of glycolysis (Fig. 2G). In normal HRE cells, OCRs and ECAR were well-balanced, whereas in 786-0 cells glycolysis (ECAR) was the principal source of energy production (Fig. 2G).

P2XR4 roles in mitochondrial calcium homeostasis
Since intracellular Ca 2+ is necessary for mitochondrial activity, we focused on the contribution of P2XR4 to cytosolic calcium [Ca 2+ ] cyt homeostasis. For this purpose, we measured [Ca 2+ ] cyt in A-498 cells, which showed the greatest mitochondrial activity among the cancer cell lines tested, using Fura-2 AM as an indicator. Cells were grown in Ca 2+ -free media to stimulate Ca 2+ release from storage and treated for 2 min with different doses of 5-BDBD, an inhibitor of P2XR4. 5-BDBD treatment reduced the cytosol calcium concentration compared to that in the controls in a dose-dependent manner (Fig. 3A).
In order to understand whether 5-BDBD treatment blockade the calcium release from lysosomes or endoplasmic reticulum (ER) we grew cells in Ca 2+ free and blocked its release from organelles with selected inhibitors (Fig. 3B). 5-BDBD treatment following the IP3R antagonist reduced [Ca 2+ ] cyt peak indicated that 5-BDBD did not modulate ER Ca 2+ storage. In contrast the addition of 5-BDBD-did not reduce [Ca 2+ ] cyt after lysosomal damage induced by L-leucyl-L-leucine methyl ester (L-L-MA), a lysosomotropic agent (Fig. 3B). These data indicated that P2XR4 regulates the flux of lysosomal Ca 2+ into the cytoplasm. In subsequent experiments, Ca 2+ uptake by cells was stimulated by adding 2 mM Ca 2+ to the medium by itself or in combination with 3 mM ATP (Fig. 3C, D). Again, treatment with 5-BDBD decreased Ca 2+ in the cytosol (Fig. 3C, D). Simultaneous administration of 5-BDBD and A084598, a P2XR7 inhibitor, strongly reduced [Ca 2+ ] cyt compared to 5-BDBD alone, when ATP was absent, indicating their synergy (Fig. 3C). Additionally , blocking mitochondrial activity by CBX enhanced intracellular Ca 2+ via P2XR7 and endoplasmic reticulum receptor; however, this was independent of P2XR4 ( Fig. 3E). Indeed, intracellular Ca 2+ remained unchanged when 5-BDBD was added together with CBX, compared to 5-BDBD alone (Fig. 3E). These data indicated that the contribution of P2XR4 to intracellular Ca 2+ homeostasis is linked to mitochondrial activity. Moreover, at an early time point (10 min), 5-BDBD treatment altered the mitochondrial shape (Fig. 3F). In contrast to the random distribution in the cytoplasm, mitochondria appeared fused in long filaments as shown by MitoSOX-Red staining (Fig. 3F) and after 24 h the mitochondrial mass was reduced (Fig. 3G) [25]. Accordingly, P2XR4 inhibition reduced basal and maximum mitochondrial oxygen consumption rates in A-498 and SNC-12 cells compared to HRE and 786-0 cells (Fig. 3H-K). The OCR reduction occurred in concentration-dependent manners  without increase of glycolysis (ECAR) (Fig. 3L, M). We conclude that P2XR4 contributes to the [Ca 2+ ] cyt homeostasis required to maintain mitochondrial cellular respiration.

P2XR4 promotes metabolic stress resistance
Reactive oxygen species (ROS) damage several cellular functions. Mitochondria play a role in the detoxification of ROS through specific enzymes, such as manganese superoxide dismutase (MnSOD/SOD2) and Catase [26]. Therefore, we investigated whether P2XR4 inhibition affects these pathways. First, we evaluated whether pharmacological inhibition of P2XR4 affected ROS production. A few minutes after 5-BDBD treatment, a significant increase in ROS was measured by DCFDA fluorescence in A-498 and SNC-12 cells (p < 0.001) (Fig. 4A, B) and lipid peroxidation (p < 0.01 (Fig. 4C). ROS levels were higher after 5-BDBD treatment than in response to oxidative stress induced by H 2 O 2 (Fig. 4B). Consistent with this observation, 5-BDBD treatment significantly increased the oxidant form of glutathione (measured as GSSG activity) (p < 0.001), an important regulator of redox homeostasis, and reduced the levels of detoxifying enzymes, such as Catalase, compared to control cells ( Fig. 4D-F) [27]. No significant effects on SOD protein expression were observed (data not shown). The silencing of P2XR4 recapitulated most of the above findings compared to scramble clone, such as the accumulation of radical oxygen species, reduction of growth, and reduction of the mitochondrial oxygen consumption rate (Fig. 4G-M).

P2XR4 inhibition induces mitochondrial pore dysfunction
To verify whether mitochondrial dysfunction induced by selective P2XR4 inhibition was associated with alterations in membrane potential, we used the potentialsensitive JC-1 dye (Fig. 5). Flow cytometry showed a significantly increased ratio between depolarization (green) and oxidation (red) in treated cells, compared to cells maintained in 2 mM Ca 2+ or treated with CCCP, a compound known to alter the membrane potential (Fig. 5A). Similarly, confocal images of cells showed a decrease in mitochondrial membrane oxidant potential (red) after 5-BDBD treatment and increased depolarization of membranes (green) (Fig. 5B). As mitochondrial potential is essential for regular Ca 2+ uptake, we hypothesized that 5-BDBD treatment could affect uncontrolled mitochondrial Ca 2+ entry. Using Rhodamine 2-AM as a sensor of mitochondrial Ca 2+ , we detected an overload of mitochondrial Ca 2+ after 15 min of 5-BDBD treatment dosed by fluorescence and FACS (Fig. 5C-E) which ultimately resulted in activation of mitochondrial Caspase 3 (Fig. 5F). Apoptotic FACS screening by annexin V-FITC and PI staining of SNC-12 and A-498 cells after 5-BDBD treatment, resulted in 44% and 52% of cell death, respectively (Fig. 5G,H). Moreover, the viability of cancer cell lines with high mitochondrial activity (such as A-498 and SNC-12) was sensitive to 5-BDBD treatment in a concentration-dependent manner, with an IC 50 of 1 µM and 5 µM after 24 h, respectively (Fig. 5I).
In contrast, normal epithelial cells and 786-0 cells with lower mitochondrial activity were less sensitive (IC 50 = 40 and 50 µM) (Fig. 5I). These results indicate that the efficacy of 5-BDBD is depending on the molecular characteristics of the cancer cells and P2XR4 protects against ROS via a mitochondrial pathway.

P2XR4 contributes to lysosomes integrity and function
Lysosomes are acidic, membrane-bound organelles containing hydrolytic enzymes and stored Ca +2 [28]. As in resting cells P2XR4 is mainly localized in lysosomes, we investigated whether mitochondrial Ca 2+ overload in renal cancer cells is also linked to lysosomal damage induced by 5-BDBD. We evaluated lysosome integrity before and after prolonged treatment with 5-BDBD using lysotracker staining, as this dye normally translocates rapidly to the inner membrane of acid organelles (Fig. 6A). The A-498 and SNC-12 cells were strongly stained by lysotracker, indicating a high number of acidic organelles. A time course experiment in living cells treated with 5-BDBD showed a peak of acid vesicles at 30 min, then a reduction in lysosome staining (Fig. 6A, B), coinciding with an extracellular accumulation of acid vesicles (see staining of medium in Fig. 6A) [29]. Lysotracker FACS analysis confirmed an increase in acid vesicles in treated cells at early time point, whereas after 6 h reduced (Fig. 6C). RT-PCRs confirmed the accumulation of stress gene mRNAs (Fig. 6D) and stable expression of autophagy proteins, such as p62 and the accumulation of both LC3BI and its active isoform, LC3BII, after several hours of 5-BDBD treatment (Fig. 6E). The observation that LCB3II proteins did not increase and p62 was stable for 8 h suggested that autophagy was not responsible for the loss of lysosomal integrity. Because the molecular mechanism to eliminate damaged organelles is mediated by an ubiquitin (Ub)-dependent pathway [29,30], we determined whether this event occurred also during treatment with 5-BDBD. For this purpose, culture media were immunoprecipitated for CD63 antigen and analyzed (Fig. 6F, G). Nanosize measurements indicated that CD63 + nanobody from 5-BDBD treated cells were bigger in size than controls whereas Western blotting indicated that they were polyubiquitinated, compared to untreated controls (Fig. 6F, G). Collectively, these results indicate that 5-BDBD affect the autophagy mechanism and releasing polyubiquitinated small membranes.

Therapeutic efficacy of P2XR4 inhibition in patient-derived organoids is related to mitochondrial activity
The therapeutic potential of specific P2XR4 inhibitor was tested in patient-derived 3D organoids (PDO) generated from surgically resected ccRCCs patients [31]. PDOs generation was successful in 50% of patients' biopsy samples (n = 10) [32]. The clinical pathological characteristics of patients are summarized in Supplementary Tables 1 and 2. Plating 4 × 10 4 cells PDOs reached a diameter of 200-300 µm after 3-7 days (Suppl. Figure 2) and showed typical clear cell carcinoma histology and renal cancer cell markers (Fig. 7A, B). To test the effect of the P2X4 inhibitor on PDOs growth, 384-well plates were used to have 20 PDO replicates for each dose and controls, and thus to improve the robustness of statistical analysis (Suppl. Figure 3A-C). In the first experiment, 5-BDBD at different doses was added for 4 days after PDOs had grown to a diameter of 200µm. The schematic representation of the experiment is reported in Suppl. Figure 3. Even though PDOsreplicates from the same patient showed some differences in their response to 5-BDBD treatment (Fig. 7 C,D), the mean data from the 20 replicates in each group indicated that 5-BDBD decreased PDOs size in a statistically significant dosedependent manner in 3 of 10 patients (PDOs #1-3 in Fig. 7 D). Moreover, the staining with calcein-AM (as a vital stain) and caspase 3 as an apoptotic marker, of PDOs drug treated indicated that 5-BDBD not only inhibited the growth, but also induced cell death, as evidenced in a representative image (Fig. 7E). Doseresponse curves in a viability experiment again showed that three PDOs were more sensitive to treatment that the rest, with a mean of IC 50 ranging from 10 to 32 μM after 4 days of treatment (Fig. 7F,G) although the IC 50 of individual biological replicates was different (Fig. 7H). Furthermore, area under the curve (AUC) of all PDOs confirmed the significant difference between PDOs #1-3 and the others by two-way ANOVA (Fig. 7H). In contrast, all PDOs displayed similar response patterns to everolimus and AZD8055 (Suppl. Figure 4). The consistency of the results in replicates of the same treatment suggests that the responses of PDOs generated from the same patient were stable and confirmed their suitability as a model for testing drug sensitivity (Fig. 7H, I). Because 5-BDBD inhibitory activity was greater in cells displaying pronounced mitochondrial activity (Fig. 6A), we wondered whether this effect would have observed in PDOs. To test this hypothesis, we evaluated the mitochondrial oxygen consumption rate and glycolysis in all PDOs as well as in control A-498 cells (Fig. 7J,K). We divided PDOs into two subgroups: the first subgroup with high mitochondrial activity (PDOs #1,2,3,4) with basal OCR > 50 mmol/min/µg corresponding to half value of control cells (A-498); the second group (PDOs #5-10) with low mitochondrial activity (basal OCR < 50 mmol/min/g) and variable glycolysis (Fig. 7J, K) and then analyzed variation of PDOs diameters during 5-BDBD treatment. Results indicated that three out of four PDOs with high mitochondrial activity showed significantly reduction in diameter after 4 days of 5-BDBD treatment, compared to PDOs with low mitochondrial activity (Fig. 7L). Together, these results suggest that P2X4 is a vulnerability receptor that mediates mitochondrial calcium uptake and activity.

P2XR4 expression is a therapeutic vulnerability in vivo.
To further examine the protective effect of P2XR4 on tumor growth, A-498 cells with the highest mitochondrial activity were implanted into nude mice. Beginning on day 14, when the tumors reached 50-70 mm 3 , mice were treated with vehicle control or 10 mg/kg 5-BDBD i.p. for 55 days. The tumors grew rather slowly, but the treatment effect of 5-BDBD became evident after 25 days. On day 55, the growth of the vehicle-treated tumors was more than five times greater, and they weighed 3.4 ± 4.5% more than 5-BDBD treated tumors (Fig. 8A,  B). Vehicle-treated tumor sections stained with hematoxylin and eosin showed pleiotropic cellular morphology, Ki67 positive cells and the presence of mitochondria, and the P2XR4 signal localized in the cytoplasm and on membranes (Fig. 8C). In contrast, tumor sections from 5-BDBD treated mice showed large necrotic/fibrotic areas and peripheral inflammation (indicated by yellow arrows). These sections were largely Ki67, mitochondria and P2XR4 negative, except for the residual tumor areas (Fig. 8C, D). Together, these results support the hypothesis that P2X4 receptor contributes to mitochondrial metabolic activity and represents a potential therapeutic target for the treatment of renal carcinoma.

Discussion
Our study investigated the potential role of P2XR4 in regulating mitochondrial activity in ccRCC cells in vitro, in a patient derived organoids model, and in vivo. Overall, results suggest that P2XR4 is a potential therapeutic target in ccRCC with high mitochondrial activity. Cancer cells depend on lysosomes and mitochondria as energy providers and regulators of the transition to a more aggressive state [33,34]. Following ATP stimulation, lysosomal P2XR4 releases intracellular calcium, which is necessary for promoting mitochondrial metabolism and resistance to oxidative stress and necrosis. Our studies indicate that P2XR4 is a part of the intricate network between organelles and represents a vulnerable target for treatment in ccRCC. P2XR4 was transcriptionally upregulated in the TCGA database of primary ccRCC, compared to adjacent tissues, and correlated with poor prognosis. Similarly, the expression of the P2XR4 receptor protein in the ATLAS database and our own tissue samples was weak or moderate in normal tissues, whereas tumors showed high expression. A significant positive correlation was observed between P2XR4 mRNA and proteins associated with mitochondrial activity and antioxidant mechanisms, such as superoxide dismutase 2 and glutathione peroxidase 4, in ccRCC samples from the TCGA cohort. Interestingly, our screening of renal carcinoma cell lines revealed higher mitochondrial activity in A-498 and SNC-12 cells overexpressing P2XR4 than that in other renal cancer cell lines or normal kidney cells.
P2XR4 is a member of the P2X receptor family of cation-permeable ligand-gated ion channels that open in response to ATP binding [18]. It is mainly localized on endolysosomes, but shuttles to and fuses with the cell membrane when stimulated [19,34,35]. It is widely expressed in the central nervous system and throughout the periphery and has many regulatory functions, including inflammatory pain, stimulation of the secretion of prostaglandin chemokines by immune cells, and motility of T cells [36][37][38][39]. Several studies have also shown a correlation between higher expression of the P2X4 receptor and lysosomal exocytosis of active hydrolases, increased cancer cell migration, survival, tumor progression [9,[40][41][42][43], and vesicular trafficking [44,45]. P2XR4 also stimulates the growth and motility of endothelial cells and promotes neoangiogenesis and in vivo tumor growth [19]. Our present study suggests that P2XR4 regulates the interaction between lysosomes and mitochondria by modulating antioxidant metabolism and oxidative phosphorylation in cell cultures and PDOs generated from tumor biopsies of individual patients. Pharmacological and genetic disruption of P2XR4 resulted in a dosedependent reduction in mitochondrial and glutathione activity, antioxidant response, increased lipid peroxidation, ROS formation, and cell death. Interestingly, a recent study in colon carcinoma indicated that P2XR4 inhibition increases ROS accumulation and prevents mTORC1 activity in cells sensitive to chemotherapy [46].
Our data provide the first evidence that P2XR4 is a key mediator of the interplay between lysosomes and mitochondria in tumor metabolism through Ca 2+ homeostasis [47]. Recent data have shown that mitochondriallysosomal contact regulates various organelle functions, including transfer among them of different metabolites [25,48]. We showed that inhibition of P2XR4 resulted in an overload of mitochondrial calcium and alteration of the pore membrane potential. Since the blockage of mitochondrial activity prevented P2XR4 contribution to cytoplasmic calcium, and the mechanism attributed to P2XR4 is a kiss-and-run-type fusion, in addition to full fusion [17], we can speculate that lysosomal P2XR4 provides the amount of Ca 2+ necessary for mitochondrial pore expansion, mediating efficient transfer of Ca 2+ into the mitochondria. Moreover, P2XR4 inhibition induces lysosomal damage, increased lipid peroxidation, and lysophagy.
Lysosomes have also been implicated in RTKI drug resistance [49]. Polymorphisms in lysosome autophagic genes are associated with poor outcomes in ccRCC [50], and mutations in SETD2, a frequent molecular feature of ccRCC, suppress autophagy regulation [51]. However, lysosomes are primarily Ca 2+ signaling hubs that govern multiple cellular processes and the extracellular microenvironment. Furthermore, our data suggest that the proor anti-tumor roles of lysosomal Ca 2+ channels can differ according to the specific genetic context, type of cancer, malignancy stage, and signals from the microenvironment [16].
Treatment of ccRCC is rapidly evolving. The current therapeutic landscape includes a combination of immune checkpoint inhibitors (ICIs) and RTKI. However, the development of drug resistance remains a major challenge. The development of therapeutic strategies targeting tumor metabolism in ccRCC is a rational approach. Ongoing experiments in our laboratories are testing different combination therapies with RTKI, mTOR inhibitors, and P2X4 inhibitors. The results of these studies will guide future development