The two-pore channel TPCN2 mediates NAADP-dependent Ca2+-release from lysosomal stores
- 1.1k Downloads
Second messenger-induced Ca2+-release from intracellular stores plays a key role in a multitude of physiological processes. In addition to 1,4,5-inositol trisphosphate (IP3), Ca2+, and cyclic ADP ribose (cADPR) that trigger Ca2+-release from the endoplasmatic reticulum (ER), nicotinic acid adenine dinucleotide phosphate (NAADP) has been identified as a cellular metabolite that mediates Ca2+-release from lysosomal stores. While NAADP-induced Ca2+-release has been found in many tissues and cell types, the molecular identity of the channel(s) conferring this release remained elusive so far. Here, we show that TPCN2, a novel member of the two-pore cation channel family, displays the basic properties of native NAADP-dependent Ca2+-release channels. TPCN2 transcripts are widely expressed in the body and encode a lysosomal protein forming homomers. TPCN2 mediates intracellular Ca2+-release after activation with low-nanomolar concentrations of NAADP while it is desensitized by micromolar concentrations of this second messenger and is insensitive to the NAADP analog nicotinamide adenine dinucleotide phosphate (NADP). Furthermore, TPCN2-mediated Ca2+-release is almost completely abolished when the capacity of lysosomes for storing Ca2+ is pharmacologically blocked. By contrast, TPCN2-specific Ca2+-release is unaffected by emptying ER-based Ca2+ stores. In conclusion, these findings indicate that TPCN2 is a major component of the long-sought lysosomal NAADP-dependent Ca2+-release channel.
KeywordsTPCN2 Two-pore channels NAADP Ca2+-release Lysosome Acidic stores
Ca2+-release from intracellular stores is a fundamental cellular process that is involved in the control of numerous physiological functions including muscle contraction, secretion, cell motility, and control of immune response [3, 11]. Given the diversity and complexity of these processes, it does not come as a surprise that Ca2+-release is controlled by a complex array of signaling pathways and second messengers [19, 23, 29, 32]. NAADP is the latest entry in the list of Ca2+-releasing molecules [19, 20]. This molecule is synthesized in the cell by ADP cyclases by exchanging the nicotinamide moiety of NADP with nicotinic acid . Notably, among all known Ca2+-releasing agents, NAADP is the most potent one acting in the low-nanomolar concentration range . The Ca2+-releasing activity of NAADP was first discovered in the mid-1990s in acidic lysosome-related organelles of sea urchin eggs  and has since been demonstrated in many mammalian tissues and cell types [13, 17]. NAADP-mediated Ca2+-release is characterized by a number of characteristic features. First of all, NAADP-dependent Ca2+-release displays a characteristic bell-shaped dose-response curve with a peak in the low nanomolar concentration range and total loss of activity at high micromolar concentrations [8, 22]. Moreover, the native NAADP-dependent Ca2+-release channel is insensitive to NADP which differs from NAADP only by the replacement of a carboxyl by an amid group . Finally, in contrast to IP3- and Ca2+/cADPR-sensitive stores, NAADP mainly seems to induce Ca2+-release from acidic lysosomal stores [10, 13], although in some cell types, NAADP-mediated Ca2+-release was also found in ER-derived stores, including the nuclear envelope 
The molecular identity of the cellular NAADP receptor has been a matter of controversy [13, 17]. It was suggested that there may be no unique receptor for NAADP but that NAADP, like Ca2+ and cADPR, is an agonist of the ryanodine receptor (RyR) complex present in ER-derived calcium stores . Other studies suggested that members of the transient receptor potential (TRP) channel family including TRPM2  and TRPML1  may be the cellular target protein of NAADP. However, the micromolar concentrations of NAADP required to fully activate these channel are one to two orders higher than those found for NAADP-mediated Ca2+-release in a variety of cells. Since none of the so-far analyzed proteins matches the hallmarks of native NAADP-sensitive release channels in a satisfying fashion, we hypothesized that the major receptor of NAADP may be formed by another not yet characterized protein.
Materials and methods
Cloning of TPCN1 and TPCN2
For cloning of TPCN1, two cDNA fragments were amplified from mouse brain RNA by RT-PCR using the SuperScript III-One-Step Kit (Invitrogen; primers for fragment 1: TTA GAA ATG CCT CTG ATG GAA and ATG CTT TGA GGT GAA ATT CCC A; primers for fragment 2: CAG ACA GCA TTG GTT TGA TGA G and GTG GGG TTT CTT CCT AGT GGT CTC GAG CCT CAT ACC AGA CAC AGA CAC). Following restriction digests, the two fragments were assembled in pcDNA3 using the Hind III site in the 5′ untranslated region, the central EcoR I site, and the Xho I site generated by PCR in the 3′ untranslated region (fragment 1 = 1,413 bp, fragment 2 = 1,548 bp). For cloning of TPCN2, the complete open-reading frame was amplified from mouse brain RNA. The forward primer (GGC TGT GGT ACC GCC ACC ATG GCG GCA GAA GAG CAG C) binds to the 5′ end and contains a Kozak consensus sequence and a Kpn I restriction site for cloning; the reverse Primer (GCC GCT CGA GAG GCC AGC ATC TCT GTC CTG) binds to the 3′ untranslated region and contains a Xho I restriction site for cloning. After restriction digest with KpnI and XhoI, the 2,255-bp fragment was cloned into pcDNA3 vector (Invitrogen). All cloned sequences were verified by sequencing.
A mouse multiple tissue Northern blot (Applied Biosystems/Ambion, Austin, TX) containing 2 µg poly(A) RNA from heart, brain, liver, spleen, kidney, lung, thymus, testis, ovary, and embryo was hybridized with probes against TPCN1 and TPCN2. Probes covered the entire coding regions and were generated by random-primed labeling with a α32P-dCTP (Roche, Basel, Switzerland). The membrane was hybridized overnight in stringent conditions. After two washing steps, the hybridized probe was visualized with a phosphorimager after an exposure time of 24 h or using a film (Hyperfilm MP, GE Healthcare, Piscataway, NJ) after a radiographic exposure time of 14 days at –80°C. The same blot was probed subsequently with TPCN1 and TPCN2 probes. The blot was stripped between each hybridization with SDS-containing buffer.
Production of TPCN2-specific antibodies
Polyclonal rabbit antisera were raised against a peptide (CDILE EPKEE ELMEK LHKHP) corresponding to aa 707 to 725 of TPCN2 plus a N-terminal cystein. For affinity purification of antibodies, the peptide was coupled to SulfoLink Coupling Gel (Thermo Fisher Scientific, Walthan, USA) in a column. Serum was loaded onto column and washed. Bound antibodies were eluted with 100 mM Glycin-HCl pH 2.8 in 800 µl aliquots and immediately neutralized with Tris-base.
The glycosylation deficient TPCN2 mutant (N594Q/N601Q) was cloned using the QuickChange Kit (Stratagene, La Jolla, CA). The deglycosylation assay was performed as described previously . Briefly, HEK293 cells were transfected with TPCN2, TPCN2N594Q/N601Q, or EGFP-TPCN1 using the calcium phosphate method and lysed; 10 µg of whole-cell lysates were incubated in the presence or absence of PNGase F or endoglycosidase H (both New England Biolabs, Ipswich, MA), according to the manufacturer’s instruction. The protein denaturation was performed for 10 min at 45°C, the enzyme incubation for 1 h at 37°C. Cell lysates were then analyzed. Subsequently, proteins were subjected to Western blot analysis.
Coimmunoprecipitation in HEK293 cells
For coimmunoprecipitation experiments, TPCN2 and TPCN1 fusion proteins were generated with either myc tag fused at the N-terminus or with EGFP sequence fused at the N-terminus. For expression of recombinant proteins, HEK293 cells were transfected using the calcium phosphate method. Three days after transfection, HEK293 cells were lysed as described previously . Equal amounts of protein (measured by Bradford assay) were incubated overnight at 4°C with 40 µl of protein A-Sepharose (GE Healthcare, Chalfont St. Giles, UK) and a specific antibody (Cell Signaling Technology, Beverly, MA) directed against the myc tag of examined proteins and 500 µl of buffer containing 50 mM Tris HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, and 50 mM Triton X-100, supplemented with PI. Beads were pelleted by centrifugation and washed three times with cold buffer containing 20 mM Tris HCl (pH 7.4), 5 mM MgCl2, 0.5 mM DTT, 20% (v/v) Glycerol supplemented with PI. Interacting proteins were visualized after boiling for 5 min in Laemmli sample buffer by SDS/PAGE, Western blot analysis, and ECL (GE Healthcare). The antibodies used were as follows: anti-GFP for detection of EGFP-TPCN2 fusion protein. An antibody against GST (GE Healthcare) was used as control.
For immunodetection, COS-7 cells grown on glass coverslips in a 24-well plate were transiently transfected with TPCN2 cDNA using Fugene 6 (Roche Diagnostics). Cells were cultured in DMEM supplemented with 10% fetal calf serum and kept at 37°C, 10% CO2 and fixed 1 or 2 days after transfection with 4% paraformaldehyde in PBS. Cells were permeabilized and blocked with 0.5% Triton X-100 and 5% Chemiblock (Millipore, Billerica, USA) in PBS. Cells were incubated with anti-TPCN2 (1:1000), anti-Myc (Cell Signaling, Danvers, MA; 1:1,000), anti-GFP (Clontech, Mountain View, USA; 1:500), anti-HA (Cell Signaling; 1:1000), anti-lamp1 (Abcam, Cambridge, USA; 1:250), and anti-Calnexin (Santa Cruz Biotechnology, Santa Cruz, USA; 1:100) primary antibodies in 2% Chemiblock (Millipore) and 0.3% Triton X-100 for 1 h. Then, cells were washed four times with PBS for 5 min each and incubated with AMCA (1:100), Cy2 (1:200), Cy3 (1:400), and Cy5 (1:400) secondary antibodies (all Jackson ImmunoResearch, West Grove, PA) in 2% Chemiblock (Millipore) for 1 h. Unbound antibodies were removed by three washing steps with PBS. For the staining of membranes and organelles, TRITC-WGA (membranes; 1:100; Sigma), Mitotracker Red CMX Ros (Mitochondria; 1:1,500; Invitrogen), and Hoechst 33342 (Nuclei; 0.4 µg/ml; Invitrogen) were used according to manufacturers’ instructions. For visualization of lysosomes, in some experiments, a lamp1-EGFP fusion protein was co-expressed with TPCN2 (kind gift of Esteban C. Dell'Angelica, Dept Human Genetics, UCLA, USA). The glass coverslips were then mounted onto glass slides with Permafluor (Sigma, St. Louis, MO). Cells were visualized under a Zeiss LSM 510 laser confocal microscope as previously described .
Cell surface expression
To determine surface expression of TPCN2 in COS-7 cells, the HA epitope was inserted into the loop between the putative first and second transmembrane domains of TPCN2. Two mutagenic primers (GTC GGG AAC ATC GTA TGG GTA AGT CTT TGT GAA GGA AGA TGG G and TAC GAT GTT CCC GAC TAC GCA GAT GTG CGC TAC CGT TC) have been used to insert the hemagglutinin sequence (YPYDVPDYA) in frame between amino acid residues 92 and 93 of TPCN2 using the QuikChange Kit (Stratagene) according to the manufacturer’s instruction. The sequence has been verified by sequencing. After transfection, cells were analyzed by immunocytochemistry as described above except for omitting the permeabilization step with Triton X-100 (nonpermeabilized cells). Control cells were treated equally but permeabilized.
Two days after transfection with TPCN2-EGFP or with the empty EGFP vector using Fugene 6 (Roche, Basel, Switzerland), HEK293 cells were loaded with Fura-2AM (5 µM) before measuring [Ca2+]i and subsequently placed in extracellular recording solution containing (in mM) 140 NaCl, 5 KCl, 3 MgCl2, 10 glucose, and 10 HEPES, pH 7.4. The fluorescence intensity of individual patch-clamped cells was monitored with a dual excitation fluorometric system using a Carl Zeiss Axiovert 135 fluorescence microscope equipped with a ×40 Plan-Apochromat 1.3 oil objective. The monochromatic light source (Polychrom II, TILL-Photonics) was tuned to excite Fura-2 fluorescence at 340 nm (F340) and 380 nm (F380) with a sampling frequency of 0.5 Hz for 3 ms each. Emission was detected at 450–550 nm using a PCO SensiCam VGA camera (Kelheim, Germany) and recorded and analyzed using Tillvision (Till Photonics, Martinsried, Germany). For internal perfusion with NAADP, NADP, or IP3, Fura-2AM-loaded HEK293 cells were patched in standard whole-cell configuration. After establishing base line [Ca2+]i, the membrane was perforated, and the recording was continued in current clamp mode (I = 0) . Patch pipette solution contained in mM—140 KCl, 10 HEPES, 1 MgCl2, and 0.05 BAPTA (pH 7.4) . Fura-2 (pentapotassium salt, Invitrogen, Carlsbad, USA) was added to the standard internal solution at 7 µM in addition to preloading. Changes in Fura-2 fluorescence are reported as the fluorescence ratios (F340/F380) and are translated into apparent intracellular calcium concentration according to .
All values are given as mean ± SEM; n is the number of experiments. An unpaired t test was performed for the comparison between two groups. Significance was tested by ANOVA followed by Dunett test if multiple comparisons were made. Values of P < 0.05 were considered significant.
We next analyzed the subcellular expression of heterologously expressed TPCN2 in COS-7 cells. Using wild-type TPCN2 (Fig. 2a) and TPCN2 channels carrying a HA tag in the linker region between S1 and S2 (Fig. 2b, c), we confirmed that the channel is expressed purely in intracellular compartments. Within the cell, TPCN2 was present in the ER (Fig. 2d–f) and colocalized with the lysosomal-associated membrane protein 1 (lamp1) that is a specific marker for acidic lysosomes (Fig. 2g–i, Supplementary Fig. S5). By contrast, there was no significant overlap of the TPCN2 signal with a mitochondrial marker (MitoTracker, Supplementary Fig. S6).
Here, we show that TPCN2, a novel member of the two-pore cation channel family, displays the basic functional properties of the native NAADP-dependent Ca2+-release channel. TPCN2 transcripts are widely expressed in the body and have been found in all tissues and organs investigated. TPCN2 encodes a glycosylated 75 kD protein forming homomers. Given the principal transmembrane topology of pore-loop cation channels, it is very likely that two TPCN channel subunits assemble with each other to form a complex with pseudo-tetrameric symmetry. This finding is in principal agreement with the reported molecular mass of the NAADP receptor of approximately 120 kDa [2, 9] that roughly corresponds to a TPCN2 dimer. The TPCN2 protein is specifically localized in intracellular compartments and is totally absent from the plasma membrane. Within the cells, the protein was present in acidic lysosomes and in the ER. Since our functional studies strongly indicate that only the lysosomal fraction of TPCN2 is sensitive to NAADP, the presence of TPCN2 in the ER probably reflects an overexpression phenomenon, as reported for other proteins, including ion channels .
Several lines of evidence support the notion that TPCN2 is a key component of long-sought lysosomal NAADP receptor. First of all, TPCN2 is the first channel identified that responds to NAADP in the correct physiological concentration range (low nM) . Secondly, as described for native channels, TPCN2 is inactivated by micromolar NAADP concentrations [8, 22]. Thirdly, the channel is insensitive to the NAADP homolog NADP . Also, the homologous TPCN1 channel is totally insensitive to NAADP. Fourthly, the increase in the intracellular Ca2+ concentration in cells expressing TPCN2 was in the same range as described for native NAADP-release channels [24, 28]. Fifthly, TPCN2 is localized in the acidic lysosomes, the subcelluar organelles where NAADP-dependent Ca2+-release was originally described . Importantly, TPCN2 activity is abolished by maneuvers that deplete the lysosomes [10, 13] indicating that the protein is fully operative in these organelles. By contrast, TPCN2 present in the ER was functionally silent. While we cannot exclude that TPCN2 requires assembly with another cellular protein to form a functional Ca2+-release channel, the simplest interpretation of our work is that TPCN2 itself forms the lysosomal NAADP-sensitive Ca2+-release channel. Further in vivo studies will be required to dissect the signaling cascades that control activation and modulation of TPCN2.
Lamp1-EGFP was a kind gift of Esteban C. Dell'Angelica (Dept Human Genetics, UCLA; USA). This work was supported by the Deutsche Forschungsgemeinschaft (DFG).
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
- 7.Calcraft PJ, Ruas M, Pan Z, Cheng X, Arredouani A, Hao X, Tang J, Rietdorf K, Teboul L, Chuang KT, Lin P, Xiao R, Wang C, Zhu Y, Lin Y, Wyatt CN, Parrington J, Ma J, Evans AM, Galione A, Zhu MX (2009) NAADP mobilizes calcium from acidic organelles through two-pore channels. Nature 459:596–600PubMedCrossRefGoogle Scholar
- 34.Zhang F, Jin S, Yi F, Li PL (2008) TRP-ML1 functions as a lysosomal NAADP-sensitive Ca(2+) release channel in coronary arterial myocytes. J Cell Mol Med. doi:10.1111/j.1582-4934.2008.00486.x