This study complies with the WMA Declaration of Helsinki. The use of human cells from biopsy samples of healthy subjects (cardiomyopathies ruled out) was approved by IEO-CCM IRCCS Ethic Committee (project CCM1072). Written informed consent was obtained from all participants.

Cells were obtained from endomyocardial specimens and characterized as previously described (Pilato et al., 2018) and cultured with Iscove’s Modified Dulbecco’s Medium (Thermo Fisher Scientific, MA, United States) supplemented with 20% Fetal Bovine Serum (FBS), 10 ng/ml basic fibroblast growth factor, 10,000 U/ml Penicillin, 10,000 μg/ml Streptomycin, and 0.02 M L-Glutamine.

Physiological salt solution (PSS) had the following composition (in mM): 150 NaCl, 6 KCl, 1.5 CaCl₂, 1 MgCl₂, 10 Glucose, 10 Hepes. In Ca²⁺-free solution (0Ca²⁺), Ca²⁺ was substituted with 2 mM NaCl, and 0.5 mM EGTA was added. Solutions were titrated to pH 7.4 with NaOH. The osmolality of the extracellular solution, as measured with an osmometer (Wescor 5500, Logan, UT, United States), was 300–310 mmol/kg.

C-MSCs were loaded with 2 µM fura-2 acetoxymethyl ester (fura-2/AM; 1 mM stock in dimethyl sulfoxide) in PSS for 30 min at room temperature (RT). The details of the Ca²⁺ recording set-up have been described in Moccia et al. (2021b) and are reported in the Supplementary Material. All the experiments were performed at RT. The amplitude of intracellular Ca²⁺ release in response to each agonist (NAADP or FBS) or drug [Gly-Phe β-naphthylamide (GPN), nigericin, bafilomycin A1, and cyclopiazonic acid (CPA)] was measured as the difference between the ratio at the peak of intracellular Ca²⁺ mobilization and the mean ratio of 1 min baseline before the peak. Pooled data are given as mean ± SE and statistical significance (p < 0.05) was evaluated by the Student’s t-test for unpaired observations or one-way Anova analysis followed by the post-hoc Dunnett’s test as appropriate (Negri et al., 2021a; Remigante et al., 2021). Data relative to Ca²⁺ signals are presented as mean ± SE, while the number of cells analysed is indicated in the corresponding bar histograms.

Cell cultures were lysed in RL lysis buffer (Norgen Biotek Corp., Thorold, ON, Canada). RNA was isolated from cells by using a Total RNA Purification kit (Norgen Biotek Corp., Thorold, ON, Canada). The quantification of the isolated RNA was determined by NanoDrop spectrophotometer (ND-1000, EuroClone, Milan, Italy). Reverse transcription was conducted with SuperScript III (Invitrogen, Carlsbad, CA, United States) following the manufacturer’s instructions. qRT-PCR was performed with the use of the iQTM SYBR Green Super Mix (Bio-Rad Laboratories, Hercules, CA, United States) and specific primers (reported in Table 1). All reactions were performed in a 96-well format with the 7900HT Fast Real-Time PCR System (Thermo Fisher Scientific, MA, United States). The relative quantities of specific mRNA were obtained with the use of the comparative Ct method and were normalized to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Maione et al., 2021; Zuccolini et al., 2022). The expression of each target gene was assessed in triplicate (Ferrera et al., 2021; Maione et al., 2021).

C-MSCs were lysed in cell lysis buffer (Cell Signalling Technology, Danvers, MA, United States) supplemented with protease and phosphatase inhibitor cocktails (Sigma-Aldrich, Saint Louis, MO, United States). Total protein extracts were subjected to SDS-PAGE and transferred onto a nitrocellulose membrane (Bio-Rad, CA, United States). The membranes were blocked for 1 h at room temperature in 5% non-fat dry milk in Wash Buffer (Tris Buffer Sulfate, 0.1% Tween-20) and then incubated O/N at 4°C with the appropriate primary antibodies (reported in Table 2). The membranes were incubated with peroxidase-conjugated secondary antibodies (GE Healthcare, Chicago, IL, United States) for 1 h. Signals were visualized using the LiteUP Western Blot Chemiluminescent Substrate (EuroClone, Milan, Italy). Images were acquired with the ChemiDocTM MP Imaging System (Bio-Rad, CA, United States), and densitometric analysis of membranes was performed using the ImageJ software (National Institutes of Health, Bethesda, MD, United States). C-MSC proteins were normalized according to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) signal.

For transmission electron microscopy (TEM) analysis, following trypsinization cells were centrifuged at 800 rpm for 5 min and then fixed with 2.5% glutaraldehyde in culture medium, for 2 h at RT (Carriero et al., 2021). The cell pellet was then rinsed in PBS overnight, post-fixed in 1% aqueous OsO4 for 3 h at room temperature and rinsed in H2O. Cells were pre-embedded in 2% agarose in water, dehydrated in acetone and then embedded in epoxy resin (Electron Microscopy Sciences, EM-bed812). Ultrathin sections (60–80 nm) were cut on a Reichert OM-U3 ultramicrotome, collected on nickel grids and then stained with uranyl acetate and lead citrate. The specimens were observed with a JEM 1200 EX II (JEOL, Peabody, MA, United States) electron microscope operating at 100 kV and equipped with a MegaView G2 CCD camera (Olympus OSIS, Tokyo, Japan).

C-MSCs were plated in 6-well plates (100,000 cells/well) and serum starved for 4 h. Cells were then stimulated with 20% FBS in the absence (Ctrl) or presence of 100 µM of NED-19, a selective TPC blocker (Galione, 2015; Jin et al., 2020). 24 and 48 h after stimulation with FBS, the medium was removed, cells detached from the plates, and counted.

To evaluate whether blocking TPCs with NED-19 was able to induce apoptosis in C-MSCs, Annexin V Alexa Fluor™ 488 Dye (Thermo Fisher Scientific, MA, United States) has been used, according to the manufacturer’s instructions. Briefly, cells were detached using TrypLE™ Select Enzyme (Thermo Fisher Scientific, MA, United States) and incubated with Annexin V Alexa Fluor™ 488 Dye for 15 min at RT. The fluorescence emission at 530 nm corresponding to apoptotic cells has been measured using flow cytometry (Gallios, Beckman Coulter, Brea, CA, United States).

In order to assess whether they are endowed with a NAADP-sensitive Ca²⁺ store, C-MSCs were loaded with Fura-2/AM (2 µM), a Ca²⁺ sensitive fluorophore, as shown elsewhere (Maione et al., 2020a). Human MSCs may exhibit spontaneous oscillations in [Ca²⁺]_i (Kawano et al., 2002; Kawano et al., 2003; Kawano et al., 2006). Consistently, a fraction of C-MSCs (≈56.4%) exhibited a few (1-4) Ca²⁺ spikes in the absence of extracellular stimulation (Supplementary Figure S1). These cells were, therefore, discarded from subsequent analysis as shown elsewhere (Zuccolo et al., 2017; Zuccolo et al., 2019), since the spontaneous, unpredictable Ca²⁺ activity could mask or even prevent (in case of transient depletion of endogenous target organelle) the Ca²⁺ response to NAADP. We then assessed whether NAADP-AM, a membrane-permeable analogue of NAADP (Macgregor et al., 2007; Brailoiu et al., 2010), was able to increase the [Ca²⁺]_i in C-MSCs. NAADP-AM (1 µM) evoked a short train of intracellular Ca²⁺ oscillations that declined ≈25 min after their onset in the presence of extracellular Ca²⁺ in 88 out of 164 cells (53.6%) (Figure 1A). In 58 out of 164 cells (35.4%), NAADP-AM (1 µM) induced a transient increase in [Ca²⁺]_i that lasted ≈800 s returned to the baseline in the continuous presence of the agonist (Figure 1A). Eighteen cells (11%) were not responsive to NAADP-AM (1 µM). Under 0Ca²⁺ conditions, NAADP-AM (1 µM) induced only a transient increase in [Ca²⁺]_i that was not followed by additional Ca²⁺ spikes (Figure 1B). Intriguingly, the duration of the elevation in [Ca²⁺]_i was significantly shorter, i.e., ≈280 s, while the peak amplitude was higher (Figure 1C), as compared to the Ca²⁺ transient recorded in the presence of extracellular Ca²⁺. The subsequent restitution of extracellular Ca²⁺ after the full recovery of [Ca²⁺]_i to the baseline resulted in a second Ca²⁺ signal that was due to extracellular Ca²⁺ entry (Figure 1B). NAADP-AM was removed from the perfusate 100 s before re-addition of extracellular Ca²⁺ (Figure 1B), which suggests that the Ca²⁺ entry pathway recruited downstream of NAADP-AM-induced Ca²⁺ release is provided by store-operated Ca²⁺ entry (SOCE), as more widely discussed below (Yamazaki et al., 2007; Sanchez-Hernandez et al., 2010; Negri et al., 2020). The statistical analysis of the two distinct components of the Ca²⁺ response to NAADP-AM (i.e., endogenous Ca²⁺ release and SOCE) is presented in Figure 1D. NAADP is recognized as a mobilizer of the lysosomal Ca²⁺ pool (Galione, 2015; Patel, 2015). Control experiments confirmed that adding back extracellular Ca²⁺ after 700 s exposure to 0Ca²⁺ conditions did not increase the in C-MSCs (Supplementary Figure S2). In accord, NAADP-AM-evoked intracellular Ca²⁺ release was significantly (p < 0.001) reduced by discharging the lysosomal Ca²⁺ store with the lysosomotropic compound, dipeptide glycyl-l-phenylalanine 2-naphthylamide (GPN; 200 μM, 30 min) (Kilpatrick et al., 2013; Yuan et al., 2021) (Figures 1E,F). Of note, GPN has recently been reaffirmed as a reliable pharmacological tool to mobilize lysosomal Ca²⁺ (Yuan et al., 2021). Furthermore, NAADP-AM-evoked endogenous Ca²⁺ mobilization was abolished by collapsing the lysosomal H⁺ gradient that maintains lysosomal Ca²⁺ refilling with the H⁺/K⁺ ionophore, nigericin (50 μM, 30 min) (Figures 1E,F), or with the v-ATPase inhibitor, bafilomycin A1 (1 μM, 30 min) (Figures 1E,F) (Morgan et al., 2011; Ronco et al., 2015; Faris et al., 2019; Yuan et al., 2021). Supplementary Figure S3 shows that GPN (200 µM), nigericin (50 µM), and bafilomycin A1 (1 µM) induced a remarkable reduction in Lysotracker Red fluorescence, thereby confirming that all of these drugs target lysosomal Ca²⁺ (Pandey et al., 2009; Faris et al., 2019; Yuan et al., 2021). In accord with these observations, ammonium chloride (NH₄Cl), which disrupts the lysosomal Ca²⁺ pool by inducing intraluminal alkalinization (Christensen et al., 2002), also reduced Lysotracker Red Fluorescence and impaired NAADP-AM-evoked intracellular Ca²⁺ mobilization (Supplementary Figure S4).

Overall, these findings provide the first evidence that NAADP may induce lysosomal Ca²⁺ release followed by extracellular Ca²⁺ entry in C-MSCs.

TPCs mediate NAADP-induced intracellular Ca²⁺ release throughout the phylogenetic tree (Patel, 2015; Galione, 2019; Jin et al., 2020), including the cardiovascular system (Moccia et al., 2021a; Negri et al., 2021b). In accord, qRT-PCR analysis showed that both TPC1 and TPC2 transcripts are expressed in C-MSCs, although TPC1 mRNA is slightly more abundant (Figure 2A). Negative controls were performed by omitting reverse transcriptase from the reaction (not shown) (Faris et al., 2019). Immunoblotting confirmed that TPC1 and TPC2 are also expressed at protein level. Two single bands of, respectively, 94 and 85 kDa were found for TPC1 and TPC2 proteins (Figure 2B). C-MSCs are not amenable for lipofectamine-mediated transfection of selective small interfering RNAs (Maione, Sommariva, and Pompilio, unpublished results), which is the strategy we have recently employed to downregulate TPC1 expression in different cellular models (Faris et al., 2019; Moccia et al., 2021b). Therefore, we probed the effect of NED-19, a selective TPC inhibitor (Galione, 2015; Jin et al., 2020), which has been widely employed to inhibit NAADP-dependent TPC activation throughout the cardiovascular system (Macgregor et al., 2007; Jiang et al., 2013; Hu et al., 2021; Moccia et al., 2021a; Negri et al., 2021a). As predicted, NED-19 (100 μM, 30 min) fully suppressed NAADP-AM-evoked intracellular Ca²⁺ mobilization (Figures 2C,D). Likewise, NED-K (10 μM, 30 min), a chemically modified analogue of NED-19 that has recently been shown to selectively inhibit TPC1 (Davidson et al., 2015), and tetrandrine (10 μM, 30 min), a traditional Chinese herbal remedy that block both TPC1 and TPC2 (Sakurai et al., 2015; Moccia et al., 2021a), respectively, inhibited (p < 0.001) and abrogated NAADP-AM-evoked intracellular Ca²⁺ release (Figures 2C,D). In aggregate, these data demonstrate that NAADP stimulates TPCs to mobilize lysosomal Ca²⁺ in c-MSCs.

The local release of lysosomal Ca²⁺ evoked by NAADP has long been known to be amplified into a global increase in [Ca²⁺]_i by the recruitment of juxtaposed InsP₃Rs on the ER membrane (Churchill and Galione, 2001; Kinnear et al., 2004; Davis et al., 2012; Faris et al., 2019; Moccia et al., 2021b). To assess whether the ER Ca²⁺ store is required to maintain lysosomal Ca²⁺ release, we first exploited cyclopiazonic acid (CPA), an established inhibitor of Sarco-Endoplasmic reticulum Ca²⁺-ATPase activity, as recently shown elsewhere (Kilpatrick et al., 2013; Faris et al., 2019; Moccia et al., 2021b). In the absence of extracellular Ca²⁺ (0Ca²⁺), CPA (30 µM) induced a transient elevation in [Ca²⁺]_i due to Ca²⁺ efflux into the cytosol through ER leakage channels followed by Ca²⁺ extrusion across the plasma membrane (Figure 3A). While NAADP-AM (1 µM) was able to induce robust Ca²⁺ release in not-treated cells (Figure 3B), it failed to evoke endogenous Ca²⁺ mobilization upon CPA-induced depletion of the ER Ca²⁺ store (Figure 3A). A preliminary characterization of the Ca²⁺ handling machinery revealed that C-MSCs express InsP₃Rs, but not RyRs, and that InsP₃-induced ER Ca²⁺ discharge activates SOCE (Maione et al., 2020a). To assess the contribution of InsP₃Rs to NAADP-induced intracellular Ca²⁺ mobilization, we adopted a similar strategy to that described in (Kilpatrick et al., 2013; Kilpatrick et al., 2016; Faris et al., 2019; Moccia et al., 2021b). The transient increase in [Ca²⁺]_i evoked by NAADP-AM (1 µM) was significantly (p < 0.001) reduced by blocking InsP₃Rs with 2-Aminoethoxydiphenyl borate (2-APB) (50 μM, 30 min) (Figure 3C) (Kilpatrick et al., 2013; Kilpatrick et al., 2016) and was suppressed by inhibiting the basal production of InsP₃ with U73122 (10 μM, 10 min) (Figure 3C), which selectively interferes with phospholipase C (PLC) activity (Moccia et al., 2006; Negri et al., 2021a). The statistical analysis of these data has been presented in Figure 3D. The lack of full inhibition of NAADP-AM-evoked intracellular Ca²⁺ mobilization could be due to the incomplete inhibition of InsP₃Rs, as also reported in ECFCs (Moccia et al., 2021b), rat gastric smooth muscle cells (Pereira et al., 2014), and MDA-MB-231 breast cancer cells (Vismara et al., 2021). Therefore, InsP₃Rs provide a robust source of Ca²⁺ during lysosomal Ca²⁺ mobilization and, based upon previous observations (Davis et al., 2012; Kilpatrick et al., 2013; Ronco et al., 2015; Kilpatrick et al., 2016; Faris et al., 2019; Moccia et al., 2021b), it can be concluded that they can be recruited by CICR upon NAADP-induced lysosomal Ca²⁺ release. TEM was then exploited to assess whether MCSs between lysosomal vesicles and ER cisternae can also be detected and thereby sustain the Ca²⁺-dependent cross-talk between the two organelles also in C-MSCs (Kilpatrick et al., 2013). For this purpose, after glutaraldehyde fixation cells have been post-fixed in aqueous OsO₄ in order to darkly stain lipids and membranes, as described in Section 2. TEM micrographs indicated extensive ER-lysosome MCSs (<20 nm, 14.3 ± 1.13, n = 27 from five cells) with ultrastructural resolution (Figure 4). As also reported in human fibroblasts, in the regions of close appositions (e.g., Figure 4A), fibres that appear to tether lysosomes and ER membranes were clearly discernible. In addition, we detected regions where the apposing membranes appeared to be physically coupled with no visible space between them (e.g., Figures 4B,C). Quantification in random sections showed that 60.5% of lysosomes established contact sites with the ER. As also discussed in Kilpatrick et al. (2013), this is likely to be an underestimate as lysosomal diameter spans between 200 and 500 nm and is, therefore, predicted to extend over several sections above and below the selected plane, where additional contact sites might have been established. Of note, lysosomes could establish extensive contact sites both with the smooth (Figure 4D) and the rough (Figure 4E) ER. We also found that ER cisternae could come in direct contacts with more than one lysosome (Figure 4E). Overall, these findings provide the ultrastructural evidence that the architecture of lysosomes and ER MCSs is fully consistent with the recruitment of ER-embedded InsP₃Rs by NAADP-induced lysosomal Ca²⁺ release through TPCs.

Figure 1B clearly shows that NAADP-AM-induced mobilization of intercellularly stored Ca²⁺ resulted in extracellular Ca²⁺ entry even after the agonist washout from the perfusate. This feature clearly hints at SOCE as the Ca²⁺ entry pathway sustaining the long-lasting increase in [Ca²⁺]_i evoked by NAADP in the presence of extracellular Ca²⁺. Indeed, InsP₃-dependent ER Ca²⁺ mobilization results in SOCE activation virtually in all mammalian cells (Prakriya and Lewis, 2015; Emrich et al., 2021), including C-MSCs (Maione et al., 2020a). In order to assess whether NAADP-AM-induced lysosomal Ca²⁺ release can lead to SOCE via intermediate ER Ca²⁺ depletion, we repeated the “Ca²⁺ add-back” protocol described in Figure 1 in the absence and presence of BTP-2 or Pyr6, two selective blockers of SOCE (Schleifer et al., 2012; Moccia et al., 2016). This strategy has long been exploited to selectively evaluate the blocking effect of SOCE-targeting drugs on agonist-evoked extracellular Ca²⁺ entry rather than on the previous phase of endogenous Ca²⁺ mobilization (Sanchez-Hernandez et al., 2010; Jairaman et al., 2015; Rahman and Rahman, 2017; Scarpellino et al., 2019; Negri et al., 2020; Schach et al., 2020). The influx of Ca²⁺ secondary to Ca²⁺ restitution to the perfusate after removal of NAADP-AM (Figure 5A) from the perfusate was significantly (p < 0.001) attenuated by BTP-2 (20 μM, 20 min) and abrogated by Pyr6 (10 μM, 10 min) (Figure 5A). The statistical analysis of these data has been presented in Figure 5B. These observations demonstrate that NAADP-induced lysosomal Ca²⁺ mobilization in C-MSCs is functionally coupled to SOCE via InsP₃-dependent ER Ca²⁺ release. Therefore, lysosomal Ca²⁺ release must induce depletion of ER Ca²⁺ via InsP₃Rs, thereby leading to SOCE recruitment on the plasma membrane. To further support this conclusion, Supplementary Figure S5A shows that also the pharmacological depletion of the lysosomal Ca²⁺ store with nigericin (50 µM) induced both intracellular Ca²⁺ release and extracellular Ca²⁺ entry. Furthermore, the intracellular Ca²⁺ response to nigericin (50 µM) was significantly (p < 0.001) reduced by blocking InsP₃Rs with 2-APB (50 μM, 30 min) (Supplementary Figures S5B,C) and by interfering with basal InsP₃ production with U73122 (10 μM, 10 min) (Supplementary Figures S5B,C), as recently shown in primary cultures of colorectal cancer cells (Faris et al., 2019) and in circulating ECFCs (Moccia et al., 2021b). Finally, nigericin-evoked extracellular Ca²⁺ entry was significantly (p < 0.001) attenuated by blocking SOCE with either BTP-2 (20 μM, 20 min) or Pyr6 (10 μM, 10 min) (Figures 5C,D). This result is, therefore, consistent with the evidence reported above that NAADP-induced Ca²⁺ release through TPCs is able to induce ER Ca²⁺ depletion followed by SOCE activation.

FBS has been shown to induce intracellular Ca²⁺ signals to stimulate proliferation in primary MSCs harvested from rat bone marrow (Foreman et al., 2006). 20% FBS induced intracellular Ca²⁺ oscillations also in ≈26% of C-MSCs, whereas it promoted a transient increase elevation in [Ca²⁺]_i in the remaining 74% cells (Figure 6A). Intracellular Ca²⁺ oscillations lasted for at least 30 min, while the transient Ca²⁺ signal took approximately 13 min to decline to pre-stimulation levels (Figure 6A). In the absence of extracellular Ca²⁺ (0Ca²⁺), 20% FBS induced a rapid (≈3 min) increase in [Ca²⁺]_i that reflected endogenous Ca²⁺ mobilization. The subsequent re-addition of extracellular Ca²⁺, 100 s after FBS removal from the bath, resulted in a second bump in [Ca²⁺]_i, which was due to extracellular Ca²⁺ entry and was likely to be mediated by SOCE (Figure 6B). FBS-induced intracellular Ca²⁺ signals are known to be triggered by InsP₃-induced ER Ca²⁺ mobilization and maintained over time by SOCE (Foreman et al., 2006; Hu et al., 2009; Zuccolo et al., 2018b). Preliminary experiments confirmed that 20% FBS-induced intracellular Ca²⁺ release was abrogated by depleting the ER Ca²⁺ store with CPA (30 μM, 30 min) (Figures 6C,D), inhibiting InsP₃Rs with 2-APB (50 μM, 30 min) (Figures 6C,D), and blocking PLC with U73122 (10 μM, 10 min) (Figures 6C,D). Furthermore, 20% FBS-induced extracellular Ca²⁺ entry was significantly (p < 0.001) reduced by inhibiting SOCE with BTP-2 (20 μM, 20 min) or Pyr6 (10 μM, 10 min) (Figures 6E,F).

The evidence reported above clearly showed that NAADP-induced lysosomal Ca²⁺ release via TPCs was able to promote InsP₃-induced Ca²⁺ release from the ER, thereby resulting in SOCE activation on the plasma membrane. Therefore, we sought to assess the role of NAADP-induced lysosomal Ca²⁺ release in the Ca²⁺ response to 20% FBS (Figure 7A). The depletion of the lysosomal Ca²⁺ store with either GPN (200 μM, 30 min) (Figures 7A,B) or nigericin (50 μM, 30 min) (Figures 7A,B) abrogated FBS-induced intracellular Ca²⁺ mobilization. The same effect was achieved upon pharmacological blockade of TPCs with NED-19 (100 μM, 30 min) (Figures 7C,D), NED-K (10 μM, 30 min) (Figures 7C,D), and tetrandrine (Figures 7C,D). Therefore, NAADP plays a crucial role in igniting the Ca²⁺ response to 20% FBS in C-MSCs.

In order to assess the physiological role of NAADP-induced lysosomal Ca²⁺ release through TPCs, 20% FBS-induced C-MSC proliferation was evaluated in the absence (Ctrl) and presence of NED-19 (100 μM, 30 min). Figure 8A shows that the pharmacological blockade of TPCs significantly (p < 0.05) reduced the total cell number at 24 and 48 h, thereby showing the crucial role of TPCs in supporting C-MSC proliferation. Flow cytometric analysis of Annexin V fluorescence confirmed that pre-treating C-MSCs with NED-19 did not induce apoptosis (Supplementary Figure S6). In order to determine whether TPCs recruit mitogen-associated protein kinases (MAPKs), we evaluated the phosphorylated levels of the Ca²⁺-dependent extracellular signal-regulated protein kinases 1 and 2 (ERK1/2) and of the survival kinase, Akt (Zuccolo et al., 2018a; Faris et al., 2019; Negri et al., 2021b). Figures 8B,C illustrate that 20% FBS-induced ERK1/2, but not Akt, phosphorylation was significantly (p < 0.05) inhibited by blocking TPCs with NED-19 (100 μM, 30 min). Overall, these findings demonstrate that NAADP-induced lysosomal Ca²⁺ release through TPCs stimulates C-MSC proliferation by engaging ERK1/2.

Intracellular Ca²⁺ signals tightly control a plethora of crucial functions in human MSCs from multiple sources, as reviewed in Moccia et al. (2015), Forostyak et al. (2016), and Jiang et al. (2017). According to the canonical model, the Ca²⁺ response evoked by chemical stimulation in human MSCs is triggered by InsP₃-dependent ER Ca²⁺ and may be maintained over time by SOCE (Kawano et al., 2002; Peng et al., 2016; Kotova et al., 2018). Herein, we expanded our knowledge of the molecular mechanisms shaping intracellular Ca²⁺ signalling in these cells by focusing on the role played by NAADP in C-MSCs. The intracellular delivery of NAADP mobilizes acidic Ca²⁺ stores throughout the cardiovascular system (Moccia et al., 2021a), e.g., in guinea pig ventricular (Macgregor et al., 2007) and atrial (Collins et al., 2011) cardiomyocytes, rat pulmonary artery VSMCs (Kinnear et al., 2004), human aortic endothelial cells (Brailoiu et al., 2010), mouse brain endothelial cells (Zuccolo et al., 2019), and circulating ECFCs (Di Nezza et al., 2017; Moccia et al., 2021b). Likewise, NAADP-AM, a membrane-permeable analogue of NAADP, could induce either a transient elevation in [Ca²⁺]_i or a burst of intracellular Ca²⁺ oscillations. This latter observation is in accord with the evidence that: 1) intracellular delivery of NAADP may induce oscillatory Ca²⁺ signals in human Jurkat T-lymphocytes (Berg et al., 2000), cytotoxic T lymphocytes (Davis et al., 2012), and human pancreatic β-cells (Johnson and Misler, 2002); 2) NAADP contributes to agonist-induced repetitive Ca²⁺ spikes in several types of endothelial cells (Zuccolo et al., 2019; Berra-Romani et al., 2020; Balducci et al., 2021), and that 3) NAADP induces intracellular Ca²⁺ oscillations in mouse cardiomyocytes during reperfusion injury (Davidson et al., 2015). Early work conducted on echinoderms first suggested that NAADP was able to elicit repetitive Ca²⁺ oscillations by promoting a Ca²⁺-dependent crosstalk between two different Ca²⁺ pools (Churchill and Galione, 2001), which were later shown to be located in acidic vesicles and ER (Churchill et al., 2002; Moccia et al., 2006).

The Ca²⁺ response to NAADP in C-MSCs comprised an early phase of intracellular Ca²⁺ mobilization followed by a later phase of extracellular Ca²⁺ entry, which required the previous depletion of the endogenous Ca²⁺ pool but not the NAADP-AM presence in the perfusate. First, we found that GPN, nigericin, and bafilomycin A1, which provide three established pharmacological tools to mobilize acidic Ca²⁺ stores (Kilpatrick et al., 2013; Ronco et al., 2015; Morgan and Galione, 2021; Yuan et al., 2021), prevent NAADP-induced intracellular Ca²⁺ mobilization. In agreement with the hypothesis that the lysosomal compartment represents the primary source of this increase in [Ca²⁺]_i, all of these drugs, as well as NH₄Cl, induced a rapid reduction in Lysotracker Red fluorescence. Although a recent investigation questioned the documented GPN ability to release lysosomal Ca²⁺ (Atakpa et al., 2019), Patel’s group provided the clear-cut evidence that this compound mobilizes Ca²⁺ from acidic organelles and may, therefore, be safely exploited to probe the endogenous store primarily targeted by NAADP (Morgan et al., 2020; Yuan et al., 2021). We further showed that C-MSCs express both TPC1 and TPC2 transcripts and proteins, and that the Ca²⁺ response to NAADP was inhibited by blocking TPCs with two selective antagonists, such as NED-19 (Macgregor et al., 2007; Di Nezza et al., 2017; Jin et al., 2020; Moccia et al., 2021a) and NED-K (Davidson et al., 2015), and the traditional Chinese herbal remedy, tetrandrine, which can target both TPC1 and TPC2 (Sakurai et al., 2015; Moccia et al., 2021a). As recently reviewed in Moccia et al. (2021a) and Negri et al. (2021b), TPC1 and TPC2 are both present in mouse ventricular cardiomyocytes, but this is the first time that they were reported in any other cellular component of the human heart. As reviewed in Pitt et al. (2016), TPC1 presents a limited Ca²⁺ permeability, while TPC2 is predicted to release more Ca²⁺ upon activation. Nevertheless, it has been shown that even a small Ca²⁺ flux through TPC1 can generate a global increase in [Ca²⁺]_i when lysosomal vesicles are juxtaposed to ER cisternae (Galione, 2019). For instance, TPC1 alone supports NAADP-induced intracellular Ca²⁺ oscillations in circulating ECFCs (Di Nezza et al., 2017; Moccia et al., 2021b) and in mouse cardiac myocytes undergoing the ischemia-reperfusion injury (Davidson et al., 2015). Furthermore, TPC1 was sufficient to maintain the intracellular Ca²⁺ response to nutrients or incretins in mouse pancreatic β cells deficient for TPC2 (Cane et al., 2016). Three pieces of evidence suggest that InsP₃Rs in ER cisternae contribute to amplify NAADP-induced lysosomal Ca²⁺ release. First, depletion of the ER Ca²⁺ pool with CPA suppressed or attenuated the intracellular Ca²⁺ release evoked by both NAADP and the H⁺/K⁺ antiporter, nigericin. Second, NAADP-induced endogenous Ca²⁺ mobilization was impaired by inhibiting InsP₃Rs with 2-APB and by blocking basal InsP₃ production with U73122. Conversely, functional RyRs are absent in C-MSCs (Maione et al., 2020a). The requirement for InsP₃Rs to sustain the increase in [Ca²⁺]_i resulting from NAADP-AM-evoked Ca²⁺ release from lysosomal vesicles is in full agreement with previous work carried out on human fibroblasts (Kilpatrick et al., 2013), human ECFCs (Moccia et al., 2021b), COS-7 cells (Morgan and Galione, 2021), HeLa cells (Ronco et al., 2015), and human metastatic colorectal cancer cells (Faris et al., 2019). Third, TEM revealed clearly discernible ER-lysosomes MCSs, which closely resemble those previously described in human fibroblasts (Kilpatrick et al., 2013) and could provide the sub-cellular framework to enable InsP₃R recruitment by local Ca²⁺ release through TPCs (Penny et al., 2014). Likewise, the MCSs between lysosomal vesicles and ER cisternae in C-MSCs are similar to the cytoplasmic nanojunctions between lysosomes and sarcoplasmic reticulum (SR) recently reported in rat aortic VSMCs (Fameli et al., 2014).

The different extent of coupling between lysosomal TPCs and ER-embedded InsP₃Rs (due to changes in either their distribution or density at MCSs) could explain the onset of a long-lasting elevation in [Ca²⁺]_i that replaces the intracellular Ca²⁺ oscillations in a fraction of C-MSCs. For instance, computational modelling indicated that TPC clustering within the microdomain could accelerate the frequency of InsP₃Rs-driven Ca²⁺ oscillations (Penny et al., 2014), which could ultimately lead to the fusion of the Ca²⁺ spikes and the occurrence of a single, broader increase in [Ca²⁺]_i (Bartlett et al., 2020).

Removal of extracellular Ca²⁺ shortened the duration of the Ca²⁺ response to NAADP-AM. Therefore, NAADP is predicted to gate a Ca²⁺-permeable pathway on the plasma membrane. This observation is supported by the evidence that restitution of extracellular Ca²⁺ following exposure to NAADP (or nigericin) under 0Ca²⁺ conditions, results in a second bump in [Ca²⁺]_i that reflects extracellular Ca²⁺ entry. This influx of Ca²⁺ occurs after washout of the agonist from the bath and, therefore, it is exclusively coupled to the previous depletion of endogenous Ca²⁺ stores. As discussed elsewhere (Yamazaki et al., 2007; Sanchez-Hernandez et al., 2010; Negri et al., 2020), this feature hints at SOCE as being responsible for NAADP-induced extracellular Ca²⁺ entry. In agreement with this hypothesis, NAADP-evoked Ca²⁺ influx was remarkably reduced in the presence of either BTP-2 or Pyr6, two different inhibitors of Orai1 channels, which provide the pore-forming subunit of store-operated Ca²⁺ channels in non-excitable cells (Prakriya and Lewis, 2015; Emrich et al., 2021) and MSCs (Lee et al., 2016; Peng et al., 2016). SOCE activation ultimately results from the reduction of ER Ca²⁺ concentration (Emrich et al., 2021). As discussed elsewhere (Davis et al., 2012; Brailoiu and Brailoiu, 2016), the engagement of SOCE by NAADP (and nigericin) hints at the depletion of the ER Ca²⁺ content as the intermediate step between lysosomal Ca²⁺ release and extracellular Ca²⁺ entry. However, extracellular Ca²⁺ entry directly evoked by NAADP delivery was not always engaged during acidic Ca²⁺ signalling in the cell types where this functional interplay has been investigated (Faris et al., 2019; Moccia et al., 2021b). Therefore, it is conceivable that lysosomal Ca²⁺ release recruits ER sub-domains that are functionally coupled to the SOCE machinery in C-MSCs, but not in other cell types, as widely discussed in Parekh and Putney (2005). These observations hint at NAADP as a Ca²⁺-releasing second messenger that can trigger a functional crosstalk among multiple Ca²⁺ sources (lysosomes, ER, and plasma membrane) in C-MSCs. In these cells, NAADP may serve as a provider of the “trigger” Ca²⁺ response to extracellular stimulation that is subsequently amplified by InsP₃Rs on the ER and maintained over time by SOCE activation on the plasma membrane, as previously reported in many mammalian cells, including human fibroblasts (Kilpatrick et al., 2013), human ECFCs (Moccia et al., 2021b) and brain microvascular endothelial cells (Zuccolo et al., 2019), human metastatic colorectal cancer cells (Faris et al., 2019), human primary CTL cells (Davis et al., 2012), and rat pulmonary artery VSMCs (Kinnear et al., 2004).

It has long been known that FBS stimulates proliferation through an increase in [Ca²⁺]_i that can adopt either a biphasic (Faris et al., 2019) or an oscillatory pattern (Tao et al., 2011) in a variety of cell types, including rat bone marrow MSCs (Foreman et al., 2006). FBS-induced intracellular Ca²⁺ signals are known to impinge on the interplay between InsP₃-induced Ca²⁺ release from the ER and SOCE (Foreman et al., 2006; Hu et al., 2009). Intriguingly, a recent investigation reported the first evidence that NAADP-evoked lysosomal Ca²⁺ release via TPC1 interacts with InsP₃-dependent ER Ca²⁺ mobilization and SOCE to promote FBS-induced proliferation in human metastatic colorectal cancer cells (Faris et al., 2019). Unveiling the molecular mechanisms that drive C-MSC proliferation is crucial to improve the therapeutic outcome of regenerative strategies aiming at utilizing these cells to promote cardiac repair (Bagno et al., 2018; Braunwald, 2018). Preliminary analysis showed that FBS evoked a complex increase in [Ca²⁺]_i also in C-MSCs, which displayed either an oscillatory or a biphasic Ca²⁺ signal. Pharmacological manipulation confirmed that the Ca²⁺ response to FBS comprised InsP₃-induced ER Ca²⁺ mobilization followed by SOCE activation. Indeed, FBS-induced intracellular Ca²⁺ release was suppressed by inhibiting InsP₃Rs with 2-APB, by blocking basal InsP₃ production with U73122 and by depleting the ER Ca²⁺ store with CPA, whereas FBS-induced extracellular Ca²⁺ entry was remarkably attenuated by blocking SOCE with BTP-2 and Pyr6. Next, we provided the evidence that the NAADP-sensitive acidic Ca²⁺ store is crucial to FBS-induced intracellular Ca²⁺ signals and proliferation in C-MSCs. Indeed, FBS-induced intracellular Ca²⁺ release was abrogated by depleting the lysosomal Ca²⁺ store with either GPN or nigericin, as previously shown in human metastatic colorectal cancer cells (Faris et al., 2019). In agreement with these observations, the selective blockade of TPCs with NED-19, NED-K or tetrandrine also abolished the intracellular Ca²⁺ response to FBS. Therefore, NAADP-induced lysosomal Ca²⁺ release is indispensable to trigger the cytosolic Ca²⁺ response to FBS and this requires the functional recruitment of InsP₃Rs on the ER via CICR at lysosomal-ER MCSs. That the ER is depleted via InsP₃Rs-mediated ER Ca²⁺ release following NAADP-induced lysosomal Ca²⁺ mobilization in response to FBS is also suggested by FBS-induced SOCE activation, which requires a reduction in ER Ca²⁺ concentration (Brailoiu et al., 2009; Davis et al., 2012). The mechanism whereby FBS stimulation results to intracellular NAADP generation in C-MSCs, as well as in human metastatic cancer cells (Faris et al., 2019), remains to be elucidated. Nevertheless, FBS is likely to engage the multifunctional enzyme CD38, which catalyses the “base exchange” of the nicotinamide moiety of NADP with nicotinic acid, thereby resulting in NAADP production in most cell types (Galione, 2015), including cardiomyocytes (Negri et al., 2021b). A recent paper suggested that the dual NADPH oxidases, DUOX1 and DUOX2, contribute to NAADP biosynthesis in murine T lymphocytes (Gu et al., 2021), but their role in NAADP-dependent Ca²⁺ response to FBS is yet to be investigated.

The physiological role of NAADP-induced intracellular Ca²⁺ signals were further assessed by evaluating the effect of NED-19 on C-MSC proliferation. The pharmacological blockade of TPCs with NED-19 strongly reduced FBS-induced C-MSC proliferation at 24 and 48 h. Preliminary experiments indicated that the massive release of Ca²⁺ induced by nigericin per se resulted in C-MSC cell death already at 24 h from exposure to this lysosomotropic compound. While this observation is in accord with the reported effects of nigericin on various cell models (Murakami et al., 2012), it prevented us from probing its ability to interfere with FBS-induced proliferation. Previous work showed that NAADP-induced Ca²⁺ release may stimulate proliferation by recruiting the Ca²⁺-dependent ERK1/2 and Akt signalling pathways (Faris et al., 2019; Negri et al., 2021b). Consistently, FBS-induced ERK1/2 phosphorylation was impaired by the pharmacological blockade of TPCs with NED-19, whereas Akt engagement was unaffected. Interestingly, ERK1/2, but not Akt, was harnessed by intracellular Ca²⁺ oscillations to drive FBS-induced proliferation also in human bone marrow MSCs (Tao et al., 2011). Additionally, NAADP-induced intracellular Ca²⁺ oscillations could underpin another crucial function of C-MSCs, i.e., the regulation of extracellular matrix (ECM) composition (Maione et al., 2020b). For instance, bone marrow-derived human MSCs exhibited repetitive Ca²⁺ spikes during aligned collagen matrix formation (Gilchrist et al., 2019), whereas extracellular Ca²⁺ entry in human airway epithelial cells drives the expression and secretion of matrix-degrading enzymes, such as matrix metalloprotease 1 (Li et al., 2011). Interestingly, an increase in [Ca²⁺]_i in cardiac fibroblasts may also regulate collagen remodelling in mouse hearts (Adapala et al., 2020). Therefore, future studies will have to assess the role of NAADP-induced Ca²⁺ signalling in the modulation of ECM composition by C-MSCs.

In conclusion, this study demonstrated that NAADP induces intracellular Ca²⁺ signals in C-MSCs by promoting lysosomal Ca²⁺ release via TPCs that is in turn amplified by ER-embedded InsP₃Rs at lysosomal-ER MCSs. The following depletion of the ER Ca²⁺ pool activates SOCE, which prolongs the Ca²⁺ response to NAADP. FBS impinges on the NAADP-induced Ca²⁺-dependent crosstalk between lysosomes and ER to stimulate proliferation through the Ca²⁺-dependent ERK1/2 signalling pathway. These findings pave the way for future studies assessing whether NAADP signalling in C-MSCs could be targeted to favour cardiac repair upon an ischemic insult or to other pathologies associated to maladaptive cardiac remodelling, such as ACM, heart failure and cardiac fibrosis.