Mammalian cell lines, culture, and molecular biology

A variant of the HEK293 cell line, 293H (Invitrogen), which has better adherence in monolayer culture, was used throughout the study. SK-BR-3 cells were obtained from ATCC. HEK293 and SK-BR-3 cells were cultured in Dulbecco’s modified Eagle’s medium and RPMI 1640 medium, respectively, supplemented with 10% fetal bovine serum, 50 units/ml penicillin, and 50 mg/ml streptomycin. For quantitative proteomic analysis of NAADP and TPC-interacting proteins, HEK293 cells were cultured in SILAC-compatible DMEM supplemented with 10% dialyzed fetal bovine serum and 100 mg/L ¹³C or ¹²C-labeled lysine and arginine (all reagents were from Thermo Fischer Scientific) for at least 6 days to achieve an incorporation efficiency of >90%.

Recombinant cDNA constructs of human TPC1 (GenBank: AY083666.1) and TPC2 (GenBank: BC063008.1) with/without FLAG and/or V5 epitopes on their C-termini were constructed with pCDNA6 vector (Invitrogen). The TPC1 and TPC2 constructs with an additional C-terminal tag of eGFP were used in the proteomic experiments. To facilitate the identification of transfected cells, an IRES-containing bicistronic vector, pCDNA6-TPC2-V5-IRES-AcGFP, was also generated and used in electrophysiological experiments. Recombinant human Lsm12 (GenBank: AAH44587.1) -expression plasmids were obtained either commercially (pCMV-Lsm12-Myc-FLAG from OriGene) or constructed with pCDNA6 (pCDNA6-Lsm12-FLAG-His). The plasmid (pGP-CMV-GCaMP6f) expressing the Ca²⁺ reporter GCaMP6f²⁶ was obtained from Addgene (Cat # 40755). Cells were transiently transfected with plasmids with transfection reagent of Lipofectamine 2000 (Invitrogen), PolyFect (QIAGEN), or FuGENE HD (Promega) and subjected to experiments within 16–48 h after transfection.

An Lsm12-KO cell line of HEK293 cells was generated with Synthego’s chemically modified sgRNA 5′-CCAGAAUGUCCCUCUUCCAG-3′ and GeneArt Platinium Cas9 nuclease (ThermoFisher Scientific) that were transfected together into cells using Lipofectamine CRISPRMAX Cas9 transfection reagent (ThermoFisher Scientific). KD of Lsm12, Lsm5, and Lsm11 expression in SK-BR-3 and HEK293 cells was achieved with predesigned dicer-substrate siRNA (DsiRNA) (Cat # hs.Ei.LSM12.13.2, hs.Ri.LSM5.13.2, and hs.Ri.LSM11.13.2 from IDT). A non-targeting DsiRNA (Cat # 51-01-14-03 from IDT) was used as a negative control. The real-time PCR assay was conducted in a MiniOpticon system (Bio-Rad Inc.) using KiCqStart (Millipore Sigma) SYBR Green ReadyMix (Cat # KCQS00) and predesigned primer pairs: TCGTGATGAAGAGTGATAAGG and CAAACTCAGTGACATCTTCC for Lsm5; AAAGCATATGAACGGGATTC and AGTGGATGTCTGTGAGTATC for Lsm11; and GCTTTAAAATGTCCCTCTTCC and GGCAAGCTTACTAACATTGAG for Lsm12. The real-time PCR reactions were performed as 95 °C for 30 s and then 34 cycles of 95 °C for 15 s, 58 °C for 30 s, and 72 °C for 30 s. The specificity and efficiency of the primers were analyzed by melting curve analysis and reactions with serially diluted template DNA, respectively.

Quantitative proteomic analyses of NAADP- or TPC-interacting proteins

Immobilized NAADP was used for affinity-precipitation of NAADP-interacting proteins from HEK293 cells transfected with constructs of TPC1-eGFP-FLAG or TPC2-eGFP-FLAG. Immobilized NAADP was prepared as previously reported²⁵. Briefly, 1 μmol of NAADP was incubated with 100 μl of 100 mM sodium periodate/100 mM sodium acetate (pH 5.0) mixture in the dark for 1 h. The unreacted sodium periodate was removed by adding 200 mM potassium chloride to the solution, incubating on ice for 5 min, and then centrifuging at 170,000xg for 10 min. The supernatant was then incubated with sodium acetate, which was pre-balanced with adipic acid dihydrazide-Agarose beads (Cat# A0802 from Sigma-Aldrich) at 4 °C in the dark for 3 h. The NAADP-conjugated beads were then washed sequentially with 1 M NaCl and phosphate-buffered saline (PBS). Anti-FLAG antibody was immobilized to protein-A agarose beads without chemical crosslinking.

For affinity-precipitation, proteins (~5 mg) were solubilized from five dishes (round, 10 cm in diameter) of the heavy- and light-labeled cells, respectively, with a lysis buffer containing 2% dodecyl maltoside in 150 mM NaCl and 20 mM Tris-HCl (pH 7.4). The solubilized proteins were collected as supernatants after centrifugation at 17,000xg for 10 min, and then incubated overnight with beads of immobilized NAADP or anti-FLAG antibody at 4 °C. A protease inhibitor cocktail (cOmplete, EDTA free; Cat # 11697498001 from Roche) was added in all buffers throughout the affinity-precipitation according to the manufacture’s instruction. After being washed with lysis buffer three times, the bound proteins were eluted with 100 μg/ml FLAG peptide for collecting interacting proteins of FLAG-tagged TPCs, or NAADP (100 µM), and then 4% SDS for collecting NAADP-interacting proteins.

For SILAC quantification by mass spectrometry, the heavy- and light-labeled eluates originated from the same amount of total protein inputs (supernatants of cell lyses) were merged. After a brief separation by a short run of SDS-PAGE, the protein-containing gels were excised into three or five different bands. Each gel band was thoroughly washed with 50% acetonitrile in 25 mM ammonium bicarbonate and then subjected to in-gel reduction with dithiothreitol, alkylation with iodoacetamide, and digestion with trypsin (Mass Spectrometry Grade; Cat # V5280 from Promega) as we previously described³⁹. Digested peptide mixtures were extracted, dried in a speed vacuum concentrator, reconstituted in 2% acetonitrile and 0.1% trifluoroacetic acid, and subjected to LC-MS/MS analysis with either a Q-Exactive Orbitrap mass spectrometer or an Orbitrap Fusion Lumos mass spectrometer (ThermoFisher Scientific). The mass spectrometer was operated in positive ion mode. MS/MS spectra were acquired with either higher-energy collisional dissociation (HCD) in Orbitrap (Q-Exactive) or collision-induced dissociation (CID) in ion trap (Fusion Lumos). The spectra labeling, database search, and quantification were carried out with Mascot Distiller (v 2.6) and Mascot (v 2.4) (MatrixScience) by searching against SwissProt (2016_04) human protein database (20,200 sequences). Database searches were performed with a peptide mass tolerance of 10 or 20 ppm and MS/MS tolerance of 20 mmu (Q-Exactive) or 0.5 Da (Fusion Lumos), allowing two missed cleavage sites and variable modifications of carbamidomethyl (C), oxidation (M), and N-terminal pyroglutamate (pyro-Glu). Quantification was obtained with Mascot Distiller with quantitation method of SILAC (K + 6, R + 6) from spectra pairs whose correlation coefficient between the observed precursor isotope distributions and the predicted ones is ≥0.6 and whose standard error after the fit to the signal intensities of each precursor pair is ≤1. For peptide sequence match (PSM), an identity threshold (E value ≤0.05) was required for all assigned spectra. Only proteins with heavy:light ratio of ≥3 and a mascot score of ≥40 were reported as differential or putative interacting proteins. The false discovery rate (FDR) was estimated to be <1% for peptides and <3% for proteins. The TPC (TPC1 and TPC2) and NAADP interactomes were identified from protein samples of five replications and single preparation, respectively.

Immunoprecipitation, immunoblotting, and immunofluorescence analyses

Immunoprecipitation, immunoblot, and immunofluorescence analyses of protein interactions and expression were performed similarly to analyses we reported³⁹. Rabbit polyclonal anti-FLAG antibody (Cat# F7425 from Sigma-Aldrich) and mouse anti-V5 (clone V5-10) agarose affinity gel (Cat# A7345 from Millipore) were used in immunoprecipitation. FLAG- or V5-peptide was used to elute the antibody-trapped proteins from the beads. Mouse monoclonal anti-FLAG M2 antibody (Cat# F3165 from Sigma-Aldrich) at 1:1000 dilution and mouse monoclonal anti-V5 antibody (Cat# R96125 from Invitrogen) at 1:10,000 dilution were used for immunoblotting. Rabbit monoclonal anti-LSM12 antibody (Cat# EPR12282 from Abcam) at a dilution factor of 1:1000 and 1:100 was used for immunoblotting and immunofluorescence, respectively.

In situ PLA

In situ PLA was performed using Duolink In Situ PLA reagents (Sigma-Aldrich) as we reported³⁹. HEK293 cells transfected with FLAG-tagged Lsm12 and/or V5-Tagged TPC2 constructs were fixed in 4% paraformaldehyde 16–24 h after transfection. After permeabilization treatment (0.05% Triton X-100), cells were

incubated with mouse monoclonal anti-V5 (1:200 dilution; Cat# SAB2702199 from Sigma-Aldrich) and rabbit anti-FLAG (1:500 dilution; Cat# F7425 from Sigma-Aldrich) in phosphate-buffed saline (PBS) at room temperature for 1 h. HEK293 cells probed with primary antibodies were first incubated with secondary anti-mouse and anti-rabbit antibodies conjugated with oligonucleotides of a PLA probe and then subjected to oligonucleotide hybridization, ligation, amplification, and detection following the manufacturer’s instructions. Finally, cells were mounted on slides using a mounting medium with DAPI and observed under a confocal microscope (FV1000; Olympus).

Expression and purification of recombinant human Lsm12 with E. coli

The bacterial expression plasmid pET-6His-TB-Lsm12, which encodes human Lsm12 protein with an N-terminally tagged cleavable thrombin (TB) cleavage motif after 6×His tag, was constructed with bacterial pET vector and transformed into E. coli strain BL21 (DE3). The cells were grown at 37 °C and collected 4 h after induction with 1 mM IPTG. Cells were harvested by centrifugation and broken by sonication in 50 mM Tris-HCl pH 8.0, 500 mM NaCl, and 5% glycerol supplemented with 1 mM PMSF and Pierce^TM Protease inhibitor tablet (Cat # A32965 from ThermoFisher). After centrifugation at 16,000xg for 40 min, the soluble fraction was collected as supernatant. The 6×His-tagged LSM12 (hLsm12-His_E.coli) was purified by passing the soluble fraction through a Ni-NTA column and then eluting the column with 300 mM imidazole in the same buffer. Imidazole and excessive salts in elute were removed by dialysis in 25 mM Tris-HCl pH 8.0 and 50 mM NaCl.

NAADP binding assays

The NAADP binding assay with immobilized NAADP was performed in a similar manner as that described above for affinity-precipitation of NAADP-interacting proteins except that the NAADP-conjugated beads were preincubated with various concentrations of label-free/unmodified NAADP or NADP before incubation with purified hLsm12-His_E.coli (5 nM) or HEK293 cell lysis. The ³²P-NAADP was synthesized from ³²P-NAD (Cat# BLU023 from PerkinElmer Co.) following two sequential enzymatic reactions²⁸. ³²P-NAD was first catalyzed into ³²P-NADP with 0.5 U/ml NAD kinase (Cat# AG-40T-0091from Adipogen Co.) in the presence of 5 mM Mg²⁺-ATP and 100 mM HEPES, pH 7.4 at 37 °C for 4 h. The synthesized ³²P-NADP was converted into ³²P-NAADP with 1 μg/ml ADP-ribosyl cyclase (Cat# A9106 from Sigma-Aldrich) and 100 mM nicotinic acid. The synthesized ³²P-NAADP was validated and purified with polyethyleneimine cellulose thin-layer chromatography developed in a solvent system of isobutyric acid−500 mM NH₄OH (5:3 v/v). For the assay involving ³²P-NAADP binding to purified hLsm12-His_E.coli, the protein at 30 nM was incubated with various concentrations of non-radiolabeled NAADP or NADP for 30 min and then further incubated with 1 nM ³²P-NAADP at room temperature for 1 h. The NAADP-protein complex was captured by Ni-NTA beads, washed three times with TBS, and eluted with 300 mM imidazole in TBS. For the assay involving ³²P-NAADP binding to cell membranes, we followed a previously reported procedure⁶ with slight modifications. Cell membranes corresponding to 1 mg protein were incubated with 1 nM ³²P-NAADP after preincubation with non-radiolabeled NAADP/NADP. Unbound ³²P-NAADP was separated from the membrane through rapid vacuum filtration through GF/B filter paper. The radioactivity in the eluates or filter papers was counted with a liquid scintillation counter.

Generation of Lsm12^Δ45-50 mutant mice and preparation of MEFs

The founder lines of Lsm12 mutant mice were generated using the CRISPR/Cas9 method via pronuclear injection of a mixture of sgRNA 5′-CCAGAAUGUCCCUCUUCCAG-3′ (Synthego, unmodified) and Cas9 into embryonic stem cells derived from C57BL/6 mice at the MD Anderson Cancer Center Genetically Engineered Mouse Facility. Homozygous Lsm12^Δ45–50 mutant mice and commercial C57BL/6 WT mice were used for the preparation of MEFs. MEFs were isolated and cultured as previously described⁴⁰. Briefly, Embryos were isolated from E12.5–E13.5 mouse embryos. After the head and most of the internal organs were removed, each embryo was minced and digested for 15 min, and then cultured at 37 °C, 5% CO₂ in freshly prepared MEF medium composed DMEM, 10% FBS, 2 mM l-glutamine, and 100 U/ml penicillin-streptomycin. For Ca²⁺-imaging analysis, AAV1 particles (Addgene, Cat # 100836-AAV1) carrying CAG-driven GCaMP6f Ca²⁺ sensor were added in the medium for 16–24 h at 37 °C. Cells were analyzed 24–48 h after a change of the medium to remove the virus. All animal experiments were carried out according to protocols and guidelines approved by the Institutional Animal Care and Use Committee of the University of Texas MD Anderson Cancer Center.

Imaging analysis of NAADP-evoked Ca²⁺ release

Transfected cells were identified by fluorescence of the Ca²⁺ reporter GCaMP6f. Fluorescence was monitored with an Axio Observer A1 microscope equipped with an AxioCam MRm digital camera and ZEN Blue 2 software containing a Physiology module (Carl Zeiss) at a sampling frequency of 2 Hz. Cell injection was performed with a FemtoJet microinjector (Eppendorf). The pipette solution contained (mM): 110 KCl, 10 NaCl, and 20 HEPES (pH 7.2) supplemented with Dextran (10,000 MW)-Texas Red (0.3 mg/ml) and NAADP (100 nM) or vehicle. The bath was Hank’s Balanced Salt Solution (HBSS) that contained (mM): 137 NaCl, 5.4 KCl, 0.25 Na₂HPO₄, 0.44 KH₂PO₄, 1 MgSO₄, 1 MgCl₂, 10 glucose, and 10 HEPES (pH 7.4). When purified hLsm12-His_E.coli was used, it was injected at 100 ng/µl (~4 μM) with NAADP in the same pipette solution. To minimize interference by contaminated Ca²⁺, the pipette solution was always treated with Chelex 100 resin (Cat# C709, Sigma-Aldrich) immediately before use. Microinjection (0.5 s at 150 hPa) was made ~30 s after pipette tip insertion into cells. Only cells that showed no response to mechanical puncture, i.e., no change in GCaMP6f fluorescence for ~30 s, were chosen for pipette solution injection. The successful injection was verified by fluorescence of the co-injected Texas Red. Elevation in intracellular Ca²⁺ concentration was reported by a change in fluorescence intensity ΔF/F₀, calculated from NAADP microinjection-induced maximal changes (ΔF at the peak) in fluorescence divided by the baseline fluorescence (F₀) immediately before microinjection.

Patch-clamp recording of TPC currents

To identify transfected cells under a fluorescence microscope for patch-clamp recording, TPC1 or TPC2 was co-expressed with GFP in HEK293 cells via transfection with either a biscistronic vector pCDNA6-TPC2-V5-IRES-AcGFP or the channel’s cDNA construct (pCDNA6-TPC1-V5-FLAG or pCDNA6-TPC2-V5-FLAG) together with pEGFP-C1. NAADP-induced TPC activation was recorded by whole-cell patch-clamp recordings on HEK293 cells with asymmetric Na⁺ (outside)/K⁺ (inside) solutions using a MultiClamp 700B amplifier and pCLAMP software (Axon Instruments) at room temperature. Bath solution contained 145 mM NaMeSO₃, 5 mM NaCl, and 10 mM HEPES (pH 7.2). Pipette electrodes (3–5 MΩ) were filled with 145 mM KMeSO₃, 5 mM KCl, and 10 mM HEPES (pH 7.2). The cells were visualized under an infrared differential interference contrast optics microscope (Zeiss). Currents were recorded by voltage ramps from −120 to +120 mV over 400 ms for every 2 s with a holding potential of 0 mV. After a whole-cell recording configuration was achieved, an injection pipette was inserted into the cell and the baseline of the whole-cell current was recorded. Microinjection of NAADP and purified Lsm12 protein was performed as above in imaging analysis of NAADP-evoked Ca²⁺ release. The NAADP-induced currents were obtained by subtraction of the baseline from NAADP injection-induced currents. PI(3,5)P₂-activated TPC2 activation was recorded by a patch-clamp recording of whole enlarged endolysosomes as reported^14,41. Cells were treated with vacuolin (1 µM) overnight to enlarge endolysosomes. Patch pipettes for recording were polished and had a resistance of 5–8 MΩ. The cytoplasmic solution contained 160 mM NaCl and 5 mM HEPES (pH was adjusted with NaOH to 7.2). The luminal solution contained 105 mM CaCl₂, 5 mM HEPES, and 5 mM MES (pH was adjusted to 4.6 with methanesulfonic acid).

Statistics and reproducibility

The data was processed and plotted with Igor Pro (v5), GraphPad Prism (v8), or OriginLab (v2015 or 2017). Unpaired Student’s t-test (two-tailed) was used to calculate p values. Unless indicated, all measurements or repeats were taken with distinct samples or cells. Independent experiments with similar results related to representative results were done three times for Fig. 1e left panel, four times for Fig. 1e right panel, two times for Fig. 1f left panel, three times for Fig. 1f right panel, three times for Fig. 2a, two times for Fig. 2b, one time for Fig. 2f, one time for Fig. 4d, two times for Fig. 4e, one time for Fig. 4f, two times for Fig. 5a, two times for Fig. 5b, two times for Fig. 5e, one time for Fig. 7d, three times for Fig. 7e, two times for Fig. 7f, one time for Supplementary Fig. 3c, three times for Supplementary Fig. 5a, one time for Supplementary Fig. 5b, and four times for Supplementary Fig. 5c.

Reporting Summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.