Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Na+/Ca2+ exchanger 1 on nuclear envelope controls PTEN/Akt pathway via nucleoplasmic Ca2+ regulation during neuronal differentiation

Na+/Ca2+ exchanger 1 on nuclear envelope controls PTEN/Akt pathway via nucleoplasmic Ca2+... 2+ 2+ Nuclear envelope (NE) is a Ca -storing organelle controlling neuronal differentiation through nuclear Ca 2+ 2+ concentrations ([Ca ] ). However, how [Ca ] regulates this important function remains unknown. Here, we n n + 2+ investigated the role of the nuclear form of the Na /Ca exchanger 1(nuNCX1) during the different stages of neuronal differentiation and the involvement of PTEN/PI3’K/Akt pathway. In neuronal cells, nuNCX1 was detected on the inner membrane of the NE where protein expression and activity of the exchanger increased during NGF-induced differentiation. nuNCX1 activation by Na -free perfusion induced a time-dependent activation of nuclear-resident PI3K/Akt pathway in isolated nuclei. To discriminate the contribution of nuNCX1 from those of plasma membrane NCX, we generated a chimeric protein composed of the fluorophore EYFP, the exchanger inhibitory peptide, and the nuclear localization signal, named XIP-NLS. Fura-2 measurements on single nuclei and patch-clamp experiments in whole-cell configuration showed that XIP-NLS selectively inhibited nuNCX1. Once it reached the nuclear 2+ compartment, XIP-NLS increased the nucleoplasmic Ca peak elicited by ATP and reduced Akt phosphorylation, GAP- 43 and MAP-2 expression through nuclear-resident PTEN induction. Furthermore, in accordance with the prevention of the neuronal phenotype, XIP-NLS significantly reduced TTX-sensitive Na currents and membrane potential during neuronal differentiation. The selective inhibition of nuNCX1 by XIP-NLS increased the percentage of β III tubulin- positive immature neurons in mature cultures of MAP-2-positive cortical neurons, thus unraveling a new function for 2+ nuNCX1 in regulating neuronal differentiation through [Ca ] -dependent PTEN/PI3K/Akt pathway. Introduction dysregulation may affect cell survival, as it occurs under 2+ 2+ Nuclear Ca concentrations ([Ca ] ) regulate many hypoxic conditions where it triggers the expression of pro- 1 5,6 2+ 2+ cellular functions by modulating specific proteins .For apoptotic genes .[Ca ] is influenced by cytosolic Ca 2+ instance, the nuclear-targeted Ca buffer protein parval- levels and is regulated by the activation of specific receptors, bumin, but not the cytosolic isoform, reduces hepatocyte pumps, and exchangers localized on the nuclear envelope 2 7,8 proliferation . Moreover, the activation of the transcription (NE) . For instance, the stimulation of nuclear inositol 3 4 factors Elk1 , NFAT, DREAM , and CREB depends on (1,4,5)-triphosphate receptors (InsP R) leads to nucleo- 2+ 2+ 2+ nuclear Ca levels. On the other hand, intranuclear Ca plasmic Ca increase in isolated nuclei from Xenopus 9 10 11 oocytes ,Aplysia neurons , and pancreatic acinar cells . Furthermore, the activation of G-protein coupled receptors Correspondence: Lucio Annunziato (lannunzi@unina.it) 2+ localized on the NE modulates—via [Ca ] —gene IRCCS SDN, School of Medicine, “Federico II” University of Naples, Naples, Italy IRCCS SDN, Naples, Italy Agnese Secondo and Alba Esposito contributed equally to this work. Edited by N. Barlev © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890 Secondo et al. Cell Death Discovery (2018) 4:12 Page 2 of 14 transcription, import and export pathways of proteins, and Western blot only on the inner membrane attached to the cell cycle . In addition, outer and inner membranes of NE nuclei and in total lysate of both whole cells and total + − 2+ 13–15 are provided with K ,Cl , and R-type Ca channels , nuclei (Fig. 1D). Efficient isolation of INM was confirmed + + 16 2+ 17 Na /H exchanger 1 ,Ca -ATPase pump ,and by the expression of Lamin B1/B2 and CRM1, a member 18,19 NCX1 . Despite the existent knowledge on the role of of nuclear transport factor family (Fig. 1D). Furthermore, plasma membrane NCX in the regulation of intracellular only the preparation of outer NE membrane expressed 2+ 2+ 20–22 calreticulin, an ER/SR protein that is contiguous with the Ca concentrations and Ca -dependent pathways , nuNCX1 function remains unexplored. Interestingly, global outer NE (Fig. 1D). silencing of NCX1 prevents neurite outgrowth through the inhibition of Akt pathway thus reducing NeuN-positive Expression and function of NCX isoforms in NE neurons . On the other hand, it has been reported that In an attempt to study the involvement of nuNCX1 in the miR-29a targeting the phosphatase and tensin homologs on different stages of neuronal differentiation, immunolocali- chromosome ten (PTEN), increase phosphorylation of Akt, zation of this isoform was performed also in PC12 cells thus promoting neurite outgrowth . Indeed, deregulation exposed to NGF for 3 days before complete differentiation. of PTEN affects neurogenesis, neurite outgrowth, synapto- At 3 days, DAB immunostaining of NCX1 was almost totally 25,26 genesis, and synaptic plasticity . Therefore, the purpose detected in nuclear compartment (Fig. 2A, black arrow) and of the present study has been to investigate the contribution slightly localized in some part of plasma membrane (Fig. 2A, of nuNCX1 in the modulation of nuclear-resident PTEN/ gray arrow). qRT-PCR experiments showed that, during 2+ PI3K/Akt pathway via [Ca ] during neuronal differentiation with NGF, NCX1 transcripts increased in differentiation. PC12 cells with a significant peak at 3 days of exposure (Fig. 2B). On the other hand, the transcripts of the Results third isoform of the exchanger, NCX3, significantly Characterization of nuNCX1 expression in neuronal cells reduced at 3 and 7 days of exposure (in A.U. ± SEM: control Immunolocalization of the NCX1 isoform in NGF- = 1.02 ± 0.03; 3 days NGF= 0.52 ± 0.05 *p < 0.05 vs. differentiated PC12 cells revealed diffused DAB immune- control; 7 days NGF= 0.55 ± 0.08, *p < 0.05 vs. control). staining of this isoform. Together with its plasmamembrane Moreover, Western blot analysis revealed that nuNCX1 expression, NCX1 was also present at the nuclear level expression in PC12 cells peaked after 3 days exposure to (Fig. 1Aa,b,c). Furthermore, in specificNE areas, NCX1co- NGF in nuclear preparations (Fig. 2C), thus suggesting a localized with Lamin B1/B2 (Fig. 1Ba), a component of specific role played by nuclear NCX1 in triggering nuclear lamina connected with the nucleoplasmic face of the differentiation at that time. By contrast, NCX3 protein inner membrane (INM). However, when NGF-differentiated expression dropped dramatically in the same preparation PC12 cells were transfected with siRNA against NCX1 (data not shown). (siNCX1), Lamin B1/B2-NCX1 co-localization disappeared Then, NCX1 activity was studied by exposing to a (Fig. 1Ba, c). Moreover, in whole neuronal cells, NCX1 Na -free solution single nuclei loaded with Fura-2AM, a immunosignal was also found outside the nucleus (Fig. 1Ba) fluorescent dye exclusively concentrated in NE lumen . and disappeared in cells treated with siNCX1 (Fig. 1Bb). Na -free perfusion forced the exchanger to operate in the 2+ Confocal analysis showed a significant immunosignal of forward mode, moving Ca from the nucleoplasm to 2+ NCX1 on the NE of isolated nuclei derived from the NE lumen. This elicited an increase of Ca level in the 2+ differentiated PC12 cells (Fig. 1Ca, b). Nissl staining showed NE lumen. In particular, this [Ca ] increase induced by the lack of ER contamination and the integrity of isolated Na -free exposure was much higher in nuclei obtained nuclei while ethidium bromide staining confirmed the high from PC12 cells exposed to NGF for 3 days than those purity of these preparations and the lack of dispersed DNA exposed for 7 days or undifferentiated controls (Fig. 2D). In (data not shown). In addition, any possible contamination by the same nuclear preparation, a pharmacological char- specific markers of the plasmamembrane, mitochondria, acterization of NCX1 activity in the forward mode was Golgi complex, and ER such as the plasma membrane performed by the Na ionophore monensin, or blocking calcium ATPase, Na channels, dynamin-related protein 1, NCX by the amiloride analog 3-amino-6-chloro-5-[(4- GM130, and stromal interaction molecule 1 were detected chloro-benzyl)amino]-n-[[(2,4-dimethylbenzyl)-amino]imi- in the preparation of isolated nuclei, except for the two nomethyl]-pyrazinecarboxamide (CB-DMB) .Whereas nuclear markers H3 and Akt (data not shown). monensin was able to enhance the Na -dependent activity To further confirm the specific localization of NCX1 at of the exchanger, possibly by increasing intra-nuclear Na 2+ INM level, we performed a separation between outer and levels, CB-DMB prevented the increase in NE Ca level inner membranes of NE with cold Na-citrate in isolated associated with NCX activation (Fig. 2E). Furthermore, NE 2+ nuclei obtained from NGF-differentiated PC12 cells. Once Ca increase was completely prevented in isolated nuclei the outer membrane was removed, NCX1 was detected by from PC12 cells previously treated with siNCX1(Fig. 2E). Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 3 of 14 + 2+ Fig. 1 Localization of neuronal Na /Ca exchanger 1 at the inner membrane of the NE. a Immunolocalization of NCX1 in whole cell obtained by 3,3′-diaminobenzidine in differentiated PC12 cells treated with NGF for 7 days. Scale bars: 50 µm (a), 20 µm (b, c). b Co-localization of NCX1 (red) with Lamin B1/B2 (green), the inner membrane of the NE marker, in differentiated PC12 cells detected by confocal microscopy under control conditions treated with siControl (a) and in NCX1-silenced PC12 cells (b). Scale bars: 5 µm. c Confocal images representing immunofluorescent signal of NCX1 (green) and Hoechst (blue) in isolated nuclei from differentiated PC12 cells under control conditions (a, b). d Western blot of NCX1, Lamin B1/B2, CRM1 and calretinin in total and nuclear lysates, inner and outer NE membranes of NGF-differentiated PC12 cells. All the experiments were repeated at least three times on different preparations nuNCX1 silencing prevents nuclear-resident Akt effect was prevented by pre-treating nuclei for 10 min phosphorylation induced by Na -free perfusion with the PI3′-K inhibitor LY294002 (25 µM) before To verify whether nuNCX1 activation could modulate Na -free exposure (Fig. 3b). Interestingly, in nuclear 2+ the phosphorylation of nuclear-resident Akt, a Ca preparations obtained from neuronal cells previously -dependent kinase mainly involved in neurite outgrowth, exposed to siNCX1, Na -free perfusion (5 min) failed to we exposed isolated nuclei to a Na -free solution under induce Akt phosphorylation compared to siControl controlled conditions of temperature. Isolated nuclei were cells (Fig. 3c). obtained from PC12 cells treated with NGF for 3 days, a time of the highest expression of nuNCX1 (see Fig. 2). Selective inhibition of nuNCX1 by the chimeric protein + 2+ Na -free induced phosphorylation of nuclear-resident XIP-NLS increases nucleoplasmic Ca level thus limiting Akt in isolated nuclei from PC12 cells (Fig. 3a). Moreover, neuronal differentiation Na -free-induced Akt phosphorylation was evident Forty-eight hours after the transfection of XIP-NLS already after 5 min of exposure (Fig. 3a). The same acti- (Fig. 4a), confocal microscopy and immunoblot analysis vation of Akt was induced by exposing intact nuclei revealed that XIP-NLS was selectively localized in the to bradykinin (BK) (Fig. 3a), whose Gq-coupled nuclei of NGF-differentiated PC12 cells (Figs. 4b, c). receptors are located on the NE inner membrane . This Indeed, EYFP signal co-localized with nuclear DNA Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 4 of 14 Fig. 2 nuNCX1 expression and activity in nuclei from PC12 cells during differentiation with NGF. a Immunolocalization of NCX1 in whole cell obtained by 3,3′-diaminobenzidine in PC12 cells treated with NGF for 3 days. Scale bars: 50 µm (a), 20 µm (b). b Representative qRT-PCR of ncx1 transcript expression in control cells and in PC12 cells after 3 and 7 days exposure to NGF. c Representative Western blot and quantification of nuNCX1 protein expression in isolated nuclei from PC12 cells in control conditions and after 3 and 7 days exposure to NGF. For b and c: all the experiments were repeated at least three times on different nuclear preparations; *p < 0.05 vs. undifferentiated control, **p < 0.05 vs. undifferentiated 2+ + control and 3 days NGF. d Top: Representative traces of the effect on [Ca ] of Na -free perfusion in Fura-2-loaded nuclei isolated from undifferentiated PC12 cells (control), and PC12 cells exposed to 3 days or 7 days to NGF. Bottom: quantification of the effect of d.*p < 0.001 vs. 2+ + undifferentiated control, **p < 0.05 vs. undifferentiated control and 3 days NGF. e Quantification of the effect on [Ca ] of Na -free in isolated nuclei obtained from NGF-differentiated PC12 cells pre-treated with: monensin (10 min), CB-DMB (20 min), and siControl or siNCX1 (at DIV 5). *p < 0.001 vs. undifferentiated control, **p < 0.001 vs. all groups. Inset: Representative Western blot depicts the effect of siControl and siNCX1 on nuNCX1 protein expression in isolated nuclei from NGF-differentiated PC12 cells staining Hoechst 33258 in nuclei efficiently transfected detected by Fura-2 microfluorimetry (Figs. 5a, b). By with XIP-NLS (Fig. 4b). Furthermore, the specificity of the contrast, as expected by the nuclear selectivity of the nuclear localization of XIP-NLS was revealed by the chimeric protein, XIP-NLS did not affect NCX currents co-detection, after 48 h, with the nuclear markers CREB measured by patch-clamp at plasma membrane level in and pCREB in the same fraction. Further evidence for the NGF-differentiated PC12 cells (Figs. 5c, d). nuclear localization of XIP-NLS and for the purity of our In NGF-treated PC12 cells transfected with XIP-NLS, cellular fractions was the absence of the cytosolic protein this peptide significantly increased Fura-dextran-detected 2+ SOD1 in the nuclear fraction (Fig. 4c). nucleoplasmic Ca peak in response to the Gq-protein Furthermore, XIP-NLS inhibited Na -dependent coupled receptor agonist ATP (100 µM) compared to nuNCX1 activity elicited by Na -free perfusion in isolated mock control (Fig. 6a). Once transfected, XIP-NLS nuclei obtained from NGF-differentiated PC12 cells, as significantly reduced GAP-43 expression and Akt Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 5 of 14 Fig. 3 Nuclear-resident Akt activation by Na -free exposure in nuclear preparations from neuronal cells. a Representative Western blot and quantification of phosphorylated Akt in isolated nuclei from NGF-differentiated PC12 cells (3 days) after perfusion with Na -free or bradykinin (BK). b Representative Western blot and quantification of phosphorylated Akt in isolated nuclei from NGF-differentiated PC12 cells previously treated with the PI3’K inhibitor LY294002 c Representative Western blot and quantification of the effect of Na -free (5 min) in isolated nuclei obtained from NGF- differentiated PC12 cells (3 days) previously treated with siControl or siNCX1. For a, b, c: all the experiments were repeated at least three times on + + different nuclear preparations. *p < 0.05 vs. untreated control or siControl, both not exposed to Na -free. **p < 0.05 vs. control or siControl + 5′ Na - free phosphorylation in PC12 cells exposed to NGF for expressed karyopherin β2 (Fig. 7a). Interestingly, once 3 days compared to mock control (Fig. 6b). Interestingly, transfected at higher concentrations in cortical neurons, under the same conditions, XIP-NLS significantly XIP-NLS carrying EYFP co-immunoprecipitated with increase the expression of the negative regulator nuNCX1 as detected with anti-NCX1 in isolated nuclei of PI3K/Akt pathway, phosphatase and tensin homolog (Fig. 7b). As occurred in NGF-treated PC12 cells, (PTEN) (Fig. 6b), mainly involved in axonal growth XIP-NLS significantly increased Fura-dextran-detected 24 2+ modulation . Then, in accordance with the prevention of nucleoplasmic Ca peak in response to ATP (100 µM) the neuronal phenotype, XIP-NLS significantly reduced in cortical neurons (Fig. 7c). This suggests that XIP-NLS + 2+ TTX-sensitive Na currents in PC12 cells exposed to was able to reduce Ca buffering properties of neuronal NGF for 3 days (Figs. 6, d). Furthermore, XIP-NLS nuclei. Moreover, also in mature neurons (7 DIV), significantly decreased membrane potential in PC12 cells the chimeric protein reduced terminal differentiation exposed to NGF for 3 days compared to the respective markers, such as GAP-43 and MAP-2 (Figs. 7d, e), controls (Fig. 6e). significantly increased PTEN expression and decreased A separation between outer and inner NE membranes Akt phosphorylation not only in total lysates (Fig. 7d, e) of isolated nuclei obtained from primary cortical neurons but also in isolated nuclei (Fig. 8a) from cortical neurons. showed a specific localization of NCX1 on the inner In order to strength the evidence demonstrating the role membrane attached to the nuclei (Fig. 7a). Furthermore, of nuclear NCX1 isoform in the differentiation of primary efficient isolation of INM in cortical neurons was con- cortical neurons, we analyzed the effect of the nuclear firmed by the expression of Lamin B1/B2 (Fig. 7a). On the inhibitor of NCX1, XIP-NLS, on the markers of both other hand, the preparation of outer NE membrane immature and mature rat cortical neurons. To this aim we Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 6 of 14 Fig. 4 Subcellular localization of the chimeric protein XIP-NLS. a Scheme of XIP-NLS structure. b Immunolocalization of XIP-NLS (green, 400 ng/ ml) detected 48 h after transfection. Hoechst dye (blue) was used to mark nuclei. Under these conditions, EYFP-positive nuclei were at least 25 ± 2% of total Hoechst-positive nuclei. c Western blot of EYFP, CREB, pCREB, and SOD1 in cytosolic and nuclear fractions obtained from NGF-differentiated PC12 cells transfected with XIP-NLS (400 ng/ml). Fractions were prepared 48 h after transfection performed confocal double immunofluorescence experi- localization was paralleled by its specific activation. In ments with the anti-βIII-tubulin, a marker of immature particular, working in the forward mode of operation, 2+ neurons, and microtubule-associated protein-2 (MAP-2) nuNCX1 moved Ca from the nucleoplasm to the NE that is expressed throughout all stages of neuronal dif- lumen of neuronal cells. Indeed, NCX1 silencing, as well as ferentiation. We found that the selective inhibition of the pharmacological inhibition of its activity by the 2+ nuclear NCX1 by transfecting XIP-NLS increased the amiloride derivative CB-DMB, abrogated Ca rise in the percentage of βIII-tubulin-positive immature neurons lumen of NE due to the forward mode of the nuclear after 7 DIV compared with control cultures but did not exchanger, thus demonstrating that nuNCX1 has the same change the percentage of MAP-2-positive neurons pharmacological sensitivity of its plasmamembrane form. (Fig. 8b). Therefore, nuNCX1 contributes to the Na -dependent 2+ removal of [Ca ] from nucleoplasm of neuronal cells. In 2+ Discussion this regard, we speculated that Ca extrusion from the The present study demonstrates that the nuclear Na / total nuclear compartment of a whole cell might entail two 2+ Ca exchanger isoform 1 (nuNCX1), located at the level consecutive steps: the first step, operated by NCX1 located 2+ of the inner membrane of NE (INM) in neurons, plays a on the inner membrane of the NE, might mediate Ca relevant role in neuronal differentiation through the transport from the nucleoplasm to the lumen of the NE , 2+ modulation of nuclear Ca homeostasis. However, we and the second step, whose transporter mechanism is still 2+ found NCX1 immunosignal also outside the INM, possibly unknown, might promote Ca movement from the at the level of other intracellular organelles, whereas no lumen of the nuclear envelope into the ER. Such new detectable traces of the other two NCX isoforms, NCX2 function might be supported by the presence of a Na and NCX3 were detected on INM. Compelling evidence gradient existing between the lumen of the nuclear for NCX1 specific localization to the INM is that, in envelope and the nucleoplasm. Consistently, previous nuclear preparations obtained from cortical neurons in evidence indicates the presence of a Na concentration which outer membrane was removed, nuNCX1 was co- gradient between the lumen (80 mM) and the nucleoplasm expressed with specific markers of the INM, namely (5–20 mM), that is maintained by the nuclear form of the + + 31 laminB1/B2 and CRM1. Furthermore, this specific Na /K ATPase . Indeed, this pump is present in the Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 7 of 14 2+ + Fig. 5 Functional characterization of the chimeric protein XIP-NLS. a Representative traces of Fura-2-detected [Ca ] after Na -free perfusion in isolated nuclei prepared from NGF-differentiated PC12 cells previously transfected with mock or XIP-NLS. b Quantification of a.*p < 0.001 vs. all. Control (n = 15 cells); mock (n = 20 cells); XIP-NLS 40 (n = 15 cells); XIP-NLS 400 (n = 15 cells). c Superimposed traces of I currents recorded by patch-clamp in NCX NGF-differentiated PC12 cells previously transfected with mock or XIP-NLS. d Quantification of c.mock(n = 20 cells); XIP-NLS (n = 20 cells) inner membrane of the nuclear envelope and is oriented plasma membrane NCX level . Interestingly, XIP-NLS 27,31 2+ with the ATP hydrolysis site in the nucleoplasm . inhibited nuNCX1 Ca -extruding function without Indeed, the present finding is in line with our previous affecting the activity of plasma membrane NCX1, as results demonstrating that NCX1 participates in NGF- demonstrated by patch-clamp experiments in whole induced neuronal differentiation via nuclear Akt . neuronal cells. These data suggest that nuclear NCX1 2+ However, the present data seems to suggest that, among plays a crucial role in regulating nuclear Ca home- 2+ the different exchangers, nuNCX1 plays a major role in ostasis upon [Ca ] increase and works to remove 2+ inducing neuronal differentiation working in the early nucleoplasmic Ca accumulation in whole cell under phase of the process. Consistently, nuNCX1 expression physiological conditions. and activity progressively increased during differentiation, Furthermore, the activity of nuNCX1 was involved in the thus peaking at 3 days of NGF exposure. At this time, regulation of PTEN/PI3K/Akt pathway, a transduction nuNCX1 mainly controls nuclear-resident PTEN/Akt signal regulating neuronal differentiation. Moreover, 2+ pathway via [Ca ] modulation. Indeed, when we during neuronal differentiation, the expression of PTEN, a inhibited nuNCX1 with the chimeric fluorescent protein phosphatase that converts phosphatidylinositol 3,4,5-tri- 2+ XIP-NLS, nuclear Ca clearance decreased, whereas the sphosphate to phosphatidylinositol 3,4-bisphosphate, 2+ [Ca ] peak, elicited by ATP, increased. This effect on was found to be inversely correlated to that of activated 2+ nucleoplasmic Ca was due to the selective localization Akt. Specifically, the inhibition of nuNCX1 by XIP-NLS of XIP-NLS in the nucleus where it binded to the f-loop of leaded to an increase in nuclear-resident PTEN expression nuNCX1, as also demonstrated by the coimmunopreci- and a decrease in Akt phosphorylation. Since nuNCX1, 2+ pitation between the peptide and nuNCX1 in isolated working in the forward mode, moves Ca from nucleo- nuclei. Accordingly, the same interaction occurs at plasm to NE lumen, it is conceivable that the blockade of Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 8 of 14 2+ Fig. 6 Effect of the chimeric protein XIP-NLS on [Ca ] and neuronal differentiation in NGF-treated PC12 cells. a Representative traces of 2+ Fura-dextran-detected nucleoplasmic Ca after ATP (100 µM) perfusion in NGF-treated PC12 cells (3 days) previously transfected with mock or XIP- 2+ NLS. Bar graph depicts quantification of ATP-induced nucleoplasmic Ca increase. *p < 0.05 vs. mock. mock (n = 20 cells); XIP-NLS (n = 15 cells). b Representative Western blots and quantifications of GAP-43, PTEN protein expression, and Akt phosphorylation in NGF-treated PC12 cells (3 days) previously transfected with mock or XIP-NLS. The experiments were repeated at least three times on different preparations; *p < 0.05 vs. undifferentiated cells; **p < 0.05 vs. mock. c and d Representative traces and quantification of voltage-gated sodium currents (I ) recorded from NaV PC12 cells under control conditions (undifferentiated, n = 6), after exposure to NGF for 3 days and mock transfection (n = 10) and after exposure to NGF for 3 days and NLS-XIP transfection (n = 20). The normalization for membrane capacitances produced similar results. e Quantification of membrane potential in all the conditions aforementioned. *p < 0.001 vs. undifferentiated cells; **p < 0.001 vs. NGF-treated PC12 cells + Mock this flux may promote, at nuclear level, a nucleoplasmic positive cortical neurons, thus suggesting that the selective 2+ Ca -dependent activation of nuclear-resident PTEN. inhibition of nuNCX1 hampered neuronal differentiation Indeed, a part the well-established role in antagonizing of cortical neurons. This is in line with a very recent paper PI3’K signaling at the plasma membrane level, PTEN has showing that the chronic treatment with NCX inhibitors been localized in the nucleus where it modulates tran- produce marked reduction of the Purkinje cell dendritic 32 35 scriptional activity . In addition, PTEN appears during arbor . 33,34 elongation of growing neurites , and increases pro- Collectively, this study unravels a new function for gressively at the early stage of differentiation, possibly to nuclear NCX1 within the complex regulatory machinery prevent aberrant neurite extension. In our study, once of nuclear calcium homeostasis, suggesting that this upregulated, PTEN was able to reduce Akt phosphoryla- exchanger form may control neuronal differentiation 2+ tion, MAP-2 and GAP-43 expression, thus avoiding through the fine regulation of Ca -dependent neuronal differentiation in cortical neurons. Another transduction mechanisms. interesting finding of this study is that XIP-NLS increased the percentage of βIII-tubulin-positive immature neurons -after 7 DIV- in mature cultures of MAP-2- Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 9 of 14 2+ Fig. 7 Effect of the chimeric protein XIP-NLS on [Ca ] and neuronal differentiation in primary cortical neurons. a Western blot of NCX1, Lamin B1/B2, and karyopherin in inner and outer NE membranes of cortical neurons at 7 DIV. All the experiments were repeated at least three times on different preparations. b Nuclear lysates from neurons trasfected with XIP-NLS (48 h) and immunoprecipitated with anti-EYFP (top) or anti-NCX1 (bottom). The presence of NCX1 (top) or EYFP (bottom) was analyzed by immunoblotting performed with specific antibodies. The input consists on 2+ nuclear lysates from untreated neurons at 7 DIV. c Representative traces of Fura-dextran-detected nucleoplasmic Ca after ATP (100 µM) perfusion in 2+ cortical neurons previously transfected with mock or XIP-NLS. Bar graph depicts quantification of ATP-induced nucleoplasmic Ca increase. *p < 0.05 vs. mock. mock (n = 15 cells); XIP-NLS (n = 10 cells). d, e Representative Western blots and quantifications of GAP-43, MAP-2, PTEN protein expression, and Akt phosphorylation in cortical neurons (7DIV) previously transfected with mock or XIP-NLS. All the experiments were repeated at least three times on different preparations; *p < 0.05 vs. mock Materials and methods anesthetized and then decapitated to minimize pain and Cell cultures distress. Dissection and dissociation were performed in 2+ 2+ Clonal cells Ca /Mg -free PBS containing glucose (30 mM). Tissues PC12 cells were grown on plastic dishes in RPMI were incubated with papain for 10 min at 37 °C and dis- medium composed of 10% horse serum (HS), 5% fetal sociated by trituration in Earle’s Balanced Salt Solution bovine serum (FBS), 100 UI/ml penicillin, and 100 μg/ml containing DNase, bovine serum albumin (BSA), and streptomycin. Neuronal differentiation was induced by ovomucoid. Cells were plated at 15 × 10 in 100-mm exposing PC12 cells to NGF (50 ng/ml) were cultured in a plastic Petri dishes pre-coated with poly-D-lysine (20 μg/ humidified 5% CO atmosphere. Culture medium was ml) in minimum Eagle’s medium/F12 (Life Technologies, changed every 2 days. For functional studies, cells were Milan, Italy) containing glucose, 5% deactivated FBS, 5% seeded on glass coverslips (Fisher, Springfield, NJ, USA), deactivated HS (Life Technologies, Milan, Italy), gluta- coated with poly-L-lysine (5 μg/ml) (Sigma, St. Louis, mine, and antibiotics. Cytosine β-D-arabinofuranoside Missouri, USA), and used at least 12 h after seeding. hydrochloride (Ara-C; 10 μM) was added within 48 h of plating to prevent non-neuronal cell growth. Neurons were cultured at 37 °C in a humidified 5% CO atmo- Embryonic neurons 2 sphere and used after 7 days of culture. All experiments cortical pure neurons were prepared from brains of 16- on primary cortical neurons were performed according to day-old Wistar rat embryos. Briefly, the rats were first Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 10 of 14 Fig. 8 Effect of nuclear NCX1 inhibition on nuclear-resident PTEN/Akt pathway and on βIII-tubulin and MAP-2 immunostaining in mature cortical neurons. a Representative Western blots and quantifications of PTEN expression and Akt phosphorylation in nuclear preparations from mature cortical neurons (7DIV) transfected at 5 DIV with mock or XIP-NLS. All the experiments were repeated at least three times on different preparations. *p < 0.001 vs. mock. b Top: βIII-tubulin and MAP-2 staining in cortical neurons at 7 DIV under control conditions and in neurons transfected with XIP-NLS (bar = 20 μm; nuclei Hoechst blue). Bottom: Quantification of MAP-2 and βIII-tubulin fluorescence and % of their colocalization in cortical neurons at 7 DIV the procedures described in experimental protocols preparations were stained with ethidium bromide and/or approved by the ethical committee of the “Federico II” Nissl solution to study the purity, integrity, and possible University of Naples, Italy. ER contamination of nuclear preparations. Isolation of intact nuclei Inner and outer nuclear membranes separation Highly purified nuclei were obtained from cell cultures Freshly isolated nuclei of PC12 cells and cortical by the combined use of hypotonic shock and mechanical neurons were employed for isolation of NE as described disruption. Briefly, when they reached confluence, cells previously . Nuclear preparations were treated were washed twice in saline phosphate buffer and lysed in with 2% Na-citrate (trisodium, dihydrate) in a Nuclear hypotonic buffer (10 mM Tris-HCl, pH 7.8, 10 mM β- Resuspension Buffer containing (in mM): 10 HEPES pH mercaptoethanol, 0.5 mM phenylmethylsulfonyl fluoride 7.9, 10 KCl, 1.5 MgCl , 0.1 EGTA, 0.5 DTT, 1 (PMSF), 1 μg/ml aprotinin, leupeptin, and pepstatin). After sodium orthovanadate (Na VO ), 0.2 PMSF, 350 sucrose, 3 4 20 minutes on ice, the swollen samples were mechanically protease inhibitors. Nuclei were pelleted by brief disrupted. The nuclear pellet was obtained by centrifuga- centrifugation at 5000×g: the pellet containing inner tion at 500 g for 6 min at 4 °C and was then washed in a nuclear membranes was resuspended in nuclear solution containing 10 mM Tris-HCl, 2 mM MgCl2 plus resuspension buffer, while the supernatant containing protease inhibitors, pH 7.2. Finally, the nuclear pellet was outer nuclear membranes was precipitated with 10% resuspended in a standard solution (KCl 125 mM, HEPES trichloroacetic acid . 50 mM, K2HPO4 2 mM, EGTA 0.1 mM, CaCl2 0.3 mM, 11,34 NaCl 3 mM, containing 1 mM ATP, pH 7.4) . Nuclear Nuclear/cytosolic fractionation and western blot analysis preparations were used or stored at −80 °C until use. To Cytosolic and nuclear fractions were obtained from check nuclear integrity, isolated nuclei were plated on glass NGF-differentiated PC12 cells. In brief, cells were washed coverslips (Fisher, Springfield, NJ, USA) previously coated twice and collected in saline phosphate buffer by with poly-L-lysine (30 µg/ml), and, then, fixed with 4% centrifugation at 1500 rpm for 3 min. Then, the pellet PAF (w/v) for 20 min at RT. After fixing, nuclear was dissolved in a buffer containing (in mM) 10 Hepes Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 11 of 14 pH 7.9, 10 KCl, 1.5 MgCl , 0.1 EGTA, 0.5 DTT, 1 sodium pCREB (rabbit polyclonal anti-body, Millipore Corpora- orthovanadate (Na VO ), 0.2 PMSF, and protease inhi- tion, MA, USA). 3 4 bitors; the cytosolic fraction was then obtained by two Membranes containing inner and outer nuclear frac- consecutive centrifugations at 3000 rpm for 5 min at 4 °C. tions were incubated overnight at 4 °C in the blocked Nuclear fractions were obtained dissolving the pellet in a buffer containing the following antibodies: 1:1000 anti- solution containing (in mM) 20 Hepes pH 7.9, 400 NaCl, NCX1 (rabbit polyclonal antibody, Swant, Bellinzona, Swiss); 1:200 anti-Lamin B1/B2 (mouse monoclonal 1.5 MgCl , 0.1 EGTA, 25% glycerol, 1 Na VO ,0.2 PMSF 2 3 4 and protease inhibitors; throughout the procedure, the antibody, Abcam, Cambridge, UK), 1:1000 calretinin samples were kept on an agitation system for 30 min at (rabbit polyclonal antibody, GeneTex, Inc., GA, USA), and 4 °C. Finally, samples were centrifuged at 13,000 rpm for 1:1000 Karyopherin β2 (mouse monoclonal antibody, 15 min. Santa Cruz Biotechnologies, Inc., CA, USA). Then, For Western blot analysis, cell cultures, inner and outer membranes were washed with 0.1% Tween 20 and incu- nuclear membranes, and nuclear preparations were lysed bated with the secondary antibodies (1:2000; GE Health- with a buffer containing 20 mM tris–HCl (pH 7.5), 10 mM care, Milan, Italy) for 1 h. Immunoreactive bands were NaF, 1 mM phenylmethylsulfonyl fluoride, 1% NONIDET detected using ECL reagent kits (GE Healthcare, Milan, P-40, 1 mM Na VO , 0.1% aprotinin, 0.7 mg/ml pepstatin, Italy). The optical density of the bands was determined by 3 4 and 1 μg/ml leupeptin. Samples were cleared by cen- Chemi-Doc Imaging System (Bio-Rad Laboratories, trifugation and supernatants were used for Western blot Milan, Italy). analysis. Protein concentration in supernatants was determined by the Bradford method . Protein samples qRT-PCR analysis (50 μg) were analyzed on 8% sodium dodecyl sulfate qRT-PCR was performed in a 7500-fast real-time PCR polyacrylamide gel with 5% sodium dodecyl sulfate system (Applied Biosystems) by Fast SYBR Green Master stacking gel (SDS-PAGE) and electrotransferred onto Mix (Applied Biosystems). Samples were amplified simul- Hybond ECL nitrocellulose papers (GE Healthcare, Little taneously in triplicate as follows: heating 2 min at 50 °C, Chalfont, UK). Membranes were blocked with 5% non-fat denaturation 10 min at 95 °C, amplification and quantifica- dry milk in 0.1% Tween 20 (TBS-T; 2 mmol/l Tris–HCl, tion 35 cycles of 15 s at 95 °C; 1 min at 60 °C with a single 50 mmol/l NaCl, pH 7.5) for 2 h at RT. Subsequently, they fluorescence measurement. The data were normalized by were incubated overnight at 4 °C in the blocked buffer using hypoxanthine phosphor-ribosyl-transferase (HPRT) as containing the following antibodies: 1:1000 anti-NCX1 an internal control. Differences in mRNA content between (rabbit polyclonal antibody, Swant, Bellinzona, Swiss), groups were expressed by using 2−ΔΔCt.formula. The oli- 1:1000 antibody for β-actin (mouse monoclonal antibody, gonucleotide sequences were for NCX1, forward: AGATTC Santacruz Biotechnology, Inc. CA), 1:500 anti-pAkt CGTGACTGCAGTTGTG and reverse: ATACTGGTCCT (mouse monoclonal antibody, Cell Signaling, Inc., MA, GGGTAGCTGCTA; for NCX3, forward: CGGTCACAG USA), 1:1000 anti-Akt (rabbit polyclonal antibody, Santa CTGTTGTTTTTGT and reverse: CAGGGCAGCAGCTT Cruz Biotechnology, Inc., CA, USA), 1:1000 anti-CRM1 TGCT;for HPRT,forward:TGGAAAGAACGTCTTG (goat polyclonal antibody, Santa Cruz Biotechnology, Inc, ATTGTTGA and reverse: GCTGTACTGCTTGACC CA, USA), 1:1000 anti-calretinin (rabbit polyclonal anti- AAGGAA. body; GeneTex, CA, USA), 1:500 anti-GAP-43 (mouse monoclonal antibody, Millipore Corporation, MA, USA), Small interfering RNA 1:1000 anti-MAP-2 (mouse monoclonal antibody, Sigma- NCX1 knock down was achieved with siRNA duplexes Aldrich, Milan, Italy), 1:1000 anti-PTEN (mouse mono- against NCX1 and its non-targeting control (Invitrogen, clonal antibody, Santa Cruz Biotechnology, Inc., CA, San Giuliano Milanese, Milan, IT). The following gene- USA) and 1:2000 anti-α-tubulin (mouse monoclonal specific sequences were used successfully: antibody, Sigma-Aldrich, Milan, Italy). Membranes were (RNA)-CCAAAUGGAGAGACCACCAAGACU and (RNA)- blocked with 5% non-fat dry milk in 0.1% Tween 20 (TBS- UAGUCUUGGUGGUCUCUCCAUUUGG or (RNA)-AC T; 2 mmol/l tris–HCl, 50 mmol/l NaCl, pH 7.5) for 2 h at AUGUUCCUCGGAGUUUCUAUUAU and (RNA)-AUA RT. Membranes containing nuclear fractions from neu- AUAGAAACUCCGAGGAACAUGU. After 5 days in ronal cells treated with XIP-NLS were incubated over- culture with NGF, PC12 cells were transfected for 5 h with night at 4 °C in the blocked buffer containing the each duplex at a final concentration of 10 nM using following antibodies: 1:1000 anti-SOD1 (rabbit polyclonal HyPerFect transfection reagent (Qiagen, Milan, IT). anti-body, Santa Cruz Biotechnology, Inc., CA, USA), 1:1000 anti-GFP (mouse monoclonal anti-body, Abcam, Generation and expression of XIP-NLS Cambridge, UK), 1:1000 anti-CREB (mouse monoclonal The expressing vector encoding for XIP-NLS protein anti-body, Cell Signaling Technology, Inc. MA, USA) and was obtained by sub-cloning in frame 2 copies of each XIP Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 12 of 14 and NLS cDNAs in EYFP-C3 plasmid (Clontech Labora- and Cy3 anti-rabbit IgG (1:200, Jackson Immuno Research tories, Mountain View, CA), as shown in Fig. 4 panel A. In Laboratories, Inc. PA, USA). The immunosignal of particular, 2 copies of the cDNA (from rat NCX1.1 NCX1 shown in Fig. 1 panel A and Fig. 2 panel A was NM_019268.3) encoding for XIP domain (DRRLLFYKY- detected by using a biotinylated secondary antibody. After VYKRYRAGKQR) were obtained by PCR on a single this latter incubation, the peroxidase reaction was devel- plasmid and digested by specific restriction enzymes, oped using 3,3′-diaminobenzidine/4-HCl as a chromogen. Then, after the final wash, coverslips were mounted with whereas the two cDNAs encoding for NLS (from SV40 Large T-anti-gen) and a translational stop codon Vectashield (Vector Labs, Burlingame, CA). Fluorescence (PKKKRKV*) were obtained by in vitro annealing of two intensity of MAP2 and βIII-tubunin in cortical neurons synthetic oligodeoxy nucleotide strains (Euro- was quantified in terms of pixel intensity value by using the finsgenomics, Ebersberg, Germany). Both XIP and NLS NIH image software. Briefly, digital images were taken cDNAs were sequentially ligated in frame in the multi- with 63x objective and identical laser power settings and cloning site of the EYFP-C3 plasmid. Two or more neutral exposure times were applied to all the photographs from aminoacids separated EYFP, XIP, and NLS domains. each experimental set. The quantification of colocalization Successful construction of the plasmid encoding for YFP- between MAP2 and βIII-tubun was assessed by using the XIP-NLS was verified by sequencing both strands ‘co-localization highlighter’ plug-in for ImageJ Software (Microgem, Naples, Italy). This plasmid named XIP-NLS (NIH, Bethesda, MA, USA). Before colocalization analysis, was transiently transfected in PC12 cells with Lipofecta- threshold settings for each image were determined, and mine 2000 (Invitrogen, Carlsbad, CA), according to the quantification was achieved by counting the number manufacturer’s protocol. The efficiency of transfection colocalized points per microscope field. Results were was calculated by counting the number of EYFP-positive expressed as a percentage of colocalization. All images nuclei compared to total Hoecth-positive nuclei. were observed using a Zeiss LSM510 META/laser-scan- ning confocal microscope. Immunoprecipitation analyses 2+ 2+ Cortical neurons were transfected with an excess of [Ca ] and nuclear Ca measurements 2+ XIP-NLS (400 ng/ml). After transfection, nuclei from [Ca ] was measured by single-cell computer-assisted each population were isolated as previously described and video-imaging in isolated nuclei from NGF-differentiated 38,39 homogenized in lysis buffer containing: 50 mM HEPES, PC12 loaded with Fura-2/AM (Calbiochem) . Under 100 mM NaCl, 1.5 mM MgCl2, 1 mM PMSF, 0.2% Non- these conditions, Fura-2 is trapped within the NE lumen. 2+ idet P-40, 5 μg/ml aprotinin, 10 μg/ml leupeptin and 2 μg/ However, nucleoplasmic Ca was measured in whole ml pepstatin. The nuclear lysates were cleared by cen- cells by loading NGF-differentiated PC12 with Fura- trifugation at 12,000 rpm for 10 min. At the end, 1 mg of 1dextran (10 μM/1 h at 37 °C in the presence of 0.05% nuclear lysate was immunoprecipitated with the mouse Pluronic), a conjugated form of the fluorescent probe that monoclonal anti-EGFP antibody (1:200) or the rabbit is trapped in the nucleoplasmic space. 2+ polyclonal anti-NCX1 antibody (1:200). Then, the To perform functional analysis of Ca level in the immunoprecipitates were resolved by Western Blotting lumen of NE on isolated nuclei, we modified the methods 11 18 and immunoblot analysis was performed using anti- used by Gerasimenko et al. and Xie et al. . Fresh NCX1 or anti-GFP antibodies, respectively. nuclear preparations from NGF-differentiated PC12 cells were placed on glass coverslips previously treated with 30 Immunocytochemistry µg/ml poly-L-lysine. After 1 h, the preparation was loaded Isoform-specific antibodies for NCX1 and Lamin B1/B2, with 10 µM Fura-2/AM at 4 °C in a standard solution alone or in combination, were used for immunocy- containing 0.04% Pluronic. At the end of the Fura-2/AM tochemistry. The rabbit polyclonal anti-NCX1 antibody loading period, isolated nuclei were allowed to settle for was purchased from Swant (Bellinzona, Switzerland). Cells 15 min in a TM solution composed of: 20 mM Tris-HCl, or isolated nuclei were rinsed twice in cold 0.01 M saline 1 mM MgCl2, 150 mM NaCl with a pH of 7.4 and then phosphate buffer (PBS) at pH 7.4 and fixed in 4% (w/v) subjected to an uptake buffer named Na -free consisting paraformaldehyde (Sigma, Milan, Italy) for 20 min at RT. of: 250 mM sucrose, 50 mM Tris–HCl, 4 mM K2HPO4, 4 Following three washes in PBS, cells were blocked with 3% mM MgCl2, 2 mM EGTA, 2 mM EDTA, and 147 mM + 18 (w/v) BSA and 0.05% Triton-X (Biorad, Milan, Italy) for 1 NMDG together with 2.5 mM CaCl2, pH 7.4 .To 2+ h at RT. The coverslips were then incubated overnight measure Ca levels, both in whole cells and isolated with the primary antibody anti-NCX1 (1:1000 dilution), or nuclei, coverslips were placed into a perfusion chamber anti-Lamin B1/B2 (1:50, Abcam, UK). After three washes (Medical System, Co. Greenvale, NY, USA) mounted onto in PBS, the coverslips were incubated in the dark for 1 h at a Zeiss Axiovert 200 microscope (Carl Zeiss, Germany) RT with two secondary antibodies: Cy2 anti-mouse IgG equipped with a FLUAR 40X oil objective lens. The Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 13 of 14 experiments were carried out with a digital imaging sys- 7.3). TTX-sensitive Na currents were recorded as pre- tem composed of MicroMax 512BFT cooled CCD camera viously described by Secondo et al . The mean resting (Princeton Instruments, Trenton, NJ, USA), LAMBDA membrane potential (holding potential at 0 current) was 10-2 filter wheel (Sutter Instruments, Novato, CA, USA), measured using the above-mentioned bath solution in and Meta-Morph/MetaFluor Imaging System software absence of TTX and TEA. Capacitive currents were esti- (Universal Imaging, West Chester, PA, USA). After mated from the decay of the capacitive transient induced loading, nuclei were alternatively illuminated at wave- by 5 mV depolarizing pulses from a holding potential of lengths of 340 and 380 nm by a Xenon lamp. The emitted −80 mV and acquired at a sampling rate of 50 kHz. The light was passed through a 512 nm barrier filter. Fura-2 capacitance of the membrane was calculated according to fluorescence intensity was measured every 3 s and the following equation: Cm- = τc •Io/ΔEm (1−I∞/Io), reported as F340/F380 ratios. where Cm is membrane capacitance, τc is the time con- For the experiments of Figs. 6 and 7, NGF-differentiated stant of the membrane capacitance, Io is the maximum PC12 cells and cortical neurons at 7 DIV were illuminated capacitance current value, ΔEm is the amplitude of the at wavelengths of 480 nm to identify EYFP-positive cells voltage step, and I∞ is the amplitude of the steady state and then at 340 and 380 nm to measure, in the EYFP- current. 2+ positive cells, Fura-dextran-detected nucleoplasmic Ca 2+ levels. Ca levels have been reported as 340/380 ratio. Statistical analysis Data are expressed as mean ± S.E.M. Statistical Electrophysiological recording of NCX and voltage-gated comparisons between controls and treated experimental sodium channel activity by patch clamp groups were performed using the one-way ANOVA, The membrane voltage and current were acquired using followed by Newman Keul’s test. p < 0.05 was considered pClamp10 software. statistically significant. Plasma membrane NCX currents (I )inNGF- NCX differentiated PC12 cells were recorded by the patch- Acknowledgements clamp technique in whole-cell configuration using the This work was supported by PON01_01602 and PON03PE_00146_1 by MIUR to commercially available amplifier Axopatch200B and Digi- L.A., Progetto Giovani Ricercatori GR-2010-2318138 from Ministero della Salute and PRIN 2015 from MIUR to A.S., Futuro in Ricerca 2013 (RBFR13M6FN) from data1322A interface (Molecular Devices), as previously 28,40 MIUR to P.M., and Fondazione Italiana Sclerosi Multipla FISM 2015/R/6 to F.B. described .I were recorded starting from a holding NCX We thank Dr. Paola Merolla for the editorial revision. potential of −60 mV up to a short-step depolarization at +60 mV (60 ms). A descending voltage ramp from +60 Author contributions AS designed the study, performed video-imaging, interpreted results and mV to −120 mV was applied. I recorded in the des- NCX wrote the first draft of the manuscript. TP performed Western blot cending portion of the ramp (from +60 mV to −120 mV) experiments, NE isolation and NE membranes separation. FB assisted with were used to plot the current–voltage (I–V) relation curve. immunohistochemistry and interpreted the results. AE and VT performed biochemical experiments while NG performed qRT-PCR. PM designed and The I magnitude was measured at the end of +60 mV NCX generated XIP-NLS expressing vector; RC and AP performed (reverse mode) and at the end of −120 mV (forward electrophysiological measurements. LA and GF assisted with designing 2+ mode), respectively. The Ni -insensitive component was experiments and, interpreting results, edited the manuscript. All authors read 21,41 and approved the final manuscript. All experiments were performed at subtracted from total currents to isolate I .Neu- NCX “Federico II” University of Naples, Italy. ronal cells were perfused with external Ringer’s solution containing the following (in mM): 126 NaCl, 1.2 NaHPO4, Competing interests 2.4 KCl, 2.4 CaCl2, 1.2 MgCl2, 10 glucose, and 18 The authors declare that they have no competing interests. NaHCO3, pH 7.4. Twenty millimolar tetraethylammonium Publisher's note (TEA), 50 nM TTX, and 10 μM nimodipine were added to Springer Nature remains neutral with regard to jurisdictional claims in Ringer’s solution to abolish potassium, sodium, and cal- published maps and institutional affiliations. cium currents. The dialyzing pipette solution contained Received: 23 October 2017 Accepted: 25 November 2017 the following (in mM): 100 K-gluconate, 10 TEA, 20 NaCl, Published online: 30 April 2018 1 Mg-ATP, 0.1 CaCl2, 2 MgCl2, 0.75 EGTA, and 10 HEPES, adjusted to pH 7.2 with CsOH. For tetrodotoxin (TTX)-sensitive Na channel record- References ings, NGF –differentiated PC12 cells and untreated PC12 1. Bootman, M. D., Fearnley, C., Smyrnias, I., MacDonald, F. & Roderick, cells were perfused with an extracellular Ringer’s solution H. L. An update on nuclear calcium signalling. J. Cell. Sci. 122,2337–2350 (2009). containing 20 mM tetraethylammonium (TEA) and 5 μM 2. Rodrigues, M. A. et al. Nucleoplasmic calcium is required for cell proliferation. J. nimodipine. The pipettes were filled with 110 mM CsCl, Biol. Chem. 282,17061–17068 (2007). 10 mM TEA, 2 mM MgCl2, 10 mM EGTA, 8 mM glucose, 2 mM Mg-ATP, 0.25 mM cAMP, and 10 mM HEPES (pH Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 14 of 14 + 2+ 3. Pusl,T.etal. Epidermalgrowth factor-mediated activation of the ETS domain 23. Secondo,A.etal. Involvement ofthe Na /Ca exchanger isoform 1 (NCX1) in transcription factor Elk-1requires nuclear calcium. J. Biol. Chem. 277, neuronal growth factor (NGF)-induced neuronal differentiation through Ca 27517–27527 (2002). -dependent Akt phosphorylation. J. Biol. Chem. 290,1319–1331 (2015). 2+ 4. Ledo, F., Kremer, L., Mellstrom, B. & Naranjo, J. R. Ca -dependent 24. Zou, H. et al. MicroRNA-29A/PTEN pathway modulates neurite outgrowth in block of CREB-CBP transcription by repressor DREAM. Embo. J. 21,4583–4592 PC12 cells. Neurosci 29,289–300 (2015). (2002). 25. van Diepen, M.T.&Eickholt,B. J. FunctionofPTEN during theformation and 5. Delivoria-Papadopoulos, M., Akhter, W. & Mishra, O. P. Hypoxia-induced maintenance of neuronal circuits in the brain. Dev. Neurosci. 30,59–64 (2008). 2+ Ca -influx in cerebral cortical neuronal nuclei of newborn piglets. Neurosci. 26. Zhou, J. & Parada, L. F. PTEN signaling in autism spectrum disorders. Curr. Opin. Lett. 342,119–123 (2003). Neurobiol. 22,873–879 (2012). +/ + 6. Delivoria-Papadopoulos, M.,Ashraf,Q.M.& Mishra,O.P.Effect of hypoxiaon 27. Galva, C., Artigas, P. & Gatto, C. Nuclear Na K -ATPase plays an active 2+ expression of apoptotic proteins in nuclear, mitochondrial and cytosolic role in nucleoplasmic Ca homeostasis. J. Cell. Sci. 125,6137–6147 fractions of the cerebral cortex of newborn piglets: the role of nuclear Ca (2012). -influx. Neurochem. Res. 33,1196–1204 (2008). 28. Secondo, A. et al. Molecular pharmacology of the amiloride analog 3-amino-6- 7. Leite, M. F. et al. Nuclear and cytosolic calcium are regulated independently. chloro-5-[(4-chloro-benzyl)amino]-n-[[(2,4-dimethylbenzyl)-amino]imino- + 2+ Proc. Natl. Acad. Sci. 100, 2975–2980 (2003). methyl]-pyrazinecarboxamide (CB-DMB)asapaninhibitor of theNa -Ca 8. Nicotera, P., Orrenius, S., Nilsson, T. & Berggren, P. O. An inositol 1,4,5-tri- exchanger isoforms NCX1, NCX2, and NCX3 in stably transfected cells. J. 2+ sphosphate-sensitive Ca pool in liver nuclei. Proc. Natl. Acad. Sci. 87, Pharm. Exp. Ther. 331,212–221 (2009). 6858–6862 (1990). 29. Savard, M. et al. Expression of endogenous nuclear bradykinin B2 receptors 9. Stehno-Bittel, L.,Luckhoff, A. &Clapham,D.E.Calcium releasefromthe mediating signaling in immediate early gene activation. J. Cell. Physiol. 216, nucleus by InsP3 receptor channels. Neuron 14,163–167 (1995). 234–244 (2008). 10. Bezin,S., Fossie,P.&Cancela, J. M. Nucleoplasmic reticulum is not essential in 30. Wu, G.,Xie, X., Lu,Z. H.&Ledeen,R.W. Sodium-calciumexchanger complexed nuclear calcium signalling mediated by cyclic ADPribose in primary neurons. with GM1 ganglioside in nuclear membrane transfers calcium from nucleo- Pflug. Arch. 456,581–586 (2008). plasm to endoplasmic reticulum. Proc. Natl. Acad. Sci. 106,10829–10834 11. Gerasimenko,O.V., Gerasimenko, J. V.,Tepikin,A.V. & Petersen,O.H.ATP- (2009). + + dependent accumulation and inositol trisphosphate- or cyclic ADP-ribose- 31. Garner, M. H. Na,K-ATPase in the nuclear envelope regulates Na: K gradients 2+ mediated release of Ca from the nuclear envelope. Cell 80,439–444 (1995). in hepatocyte nuclei. J. Membr. Biol. 187,97–115 (2002). 12. Campden, R.,Audet, N.&Hébert,T.E.Nuclear Gprotein signaling: newtricks 32. Freeman,D.J.etal. PTEN tumorsuppressor regulates p53 protein levels and for old dogs. J. Cardiovasc Pharm. 65,110–122 (2015). activity through phosphatase-dependent and -independent mechanisms. 13. Mazzanti, M., Bustamante, J. O. & Oberleithner, H. Electrical dimension of the Cancer Cell. 3,117–130 (2003). nuclear envelope. Physiol. Rev. 81,1–19 (2001). 33. Musatov, S. et al. Inhibition of neuronal phenotype by PTEN in PC12 cells. Proc. 14. Tabares, L., Mazzanti, M. & Clapham, D. E. Chloride channels in the nuclear Natl. Acad. Sci. 101,3627–3631 (2004). membrane. J. Membr. Biol. 123,49–54 (1991). 34. Gerasimenko, J., Maruyama, Y., Tepikin, A., Petersen, O. H. & Gerasimenko, O. 15. Bkaily, G. et al. The use of confocal microscopy in the investigation of cell Calcium signalling in and around the nuclear envelope. Biochem. Soc. Trans. structure and function in the heart, vascular endothelium and smooth muscle 31,76–78 (2003). cells. Mol. Cell. Biochem. 172,171–194 (1997). 35. Sherkhane,P.&Kapfhammer,J.P.Chronic pharmacologicalblockadeofthe 16. Bkaily, G. et al. G-protein-coupled receptors, channels, and Na + -H + Na + /Ca2 + exchanger modulates the growth and development of the Pur- exchanger in nuclear membranes of heart, hepatic, vascular endothelial, and kinje cell dendritic arbor in mouse cerebellar slice cultures. Eur. J. Neurosci. 46, smooth muscle cells. Can. J. Physiol. Pharm. 84,431–441 (2006). 2108–2120 (2017). 2+ 17. Abrenica, B. & Gilchrist, J. S. Nucleoplasmic Ca loading is regulated by 36. Gilchrist,J. S.& Pierce, G.N.Identification and purification of a calcium- 2+ mobilization of perinuclear Ca . Cell. Calcium 28,127–136 (2000). binding protein in hepatic nuclear membranes. J. Biol. Chem. 268, 18. Xie,X., Wu,G., Lu,Z.H.& Ledeen, R.W.Potentiation ofasodium-calcium 4291–4429 (1993). exchanger in the nuclear envelope by nuclear GM1 ganglioside. J. Neurochem. 37. Bradford, M. M. A rapid and sensitive method for the quantification of 81,1185–1195 (2002). microgram quantities of protein utilizing the principle of protein-dye binding. 19. Ledeen, R. & Wu, G. GM1 in the nuclear envelope regulates nuclear calcium Ann. Biochem 72,248–254 (1976). 2+ through association with a nuclear sodium-calcium exchanger. J. Neurochem. 38. Grynkiewicz, G., Poenie, M. & Tsien, R. Y. A new generation of Ca indicators 103,126–134 (2007). with greatly improved fluorescence properties. J. Biol. Chem. 260,3440–3450 + 2+ 20. Annunziato, L.,Pignataro,G.& Di Renzo, G. F. Pharmacology of brainNa –Ca (1985). exchanger: from molecular biology to therapeutic perspectives. Pharm. Rev. 39. Secondo, A. et al. BHK cells transfected with NCX3 are more resistant to 56,633–654 (2004). hypoxia followed by reoxygenation than those transfected with NCX1 and + 2+ 21. Molinaro, P. et al. Na -Ca exchanger (NCX3) knock-out mice display an NCX2: Possible relationship with mitochondrial membrane potential. Cell. impairment in hippocampal long-term potentiation and spatial learning and Calcium 42,521–535 (2007). memory. J. Neurosci. 31, 7312–7321 (2011). 40. Pannaccione, A. et al. A new concept: Aβ1-42 generates a hyperfunctional 22. Molinaro, P. et al. Neuronal NCX1 overexpression induces stroke resistance proteolytic NCX3 fragment that delays caspase-12 activation and neuronal while knockout induces vulnerability via Akt. J. Cereb. Blood. Flow. Metab. 36, death. J. Neurosci. 32, 10609–10617 (2012). + 2+ 1790–1803 (2016). 41. Molinaro, P. et al. Neurounina-1, a novel compound that increases Na /Ca exchanger activity, effectively protects against stroke damage. Mol. Pharm. 83, 142–156 (2013). Official journal of the Cell Death Differentiation Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death Discovery Springer Journals

Na+/Ca2+ exchanger 1 on nuclear envelope controls PTEN/Akt pathway via nucleoplasmic Ca2+ regulation during neuronal differentiation

Loading next page...
 
/lp/springer-journals/na-ca2-exchanger-1-on-nuclear-envelope-controls-pten-akt-pathway-via-1Y0DiXUerR

References (46)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Stem Cells; Apoptosis; Cell Cycle Analysis
eISSN
2058-7716
DOI
10.1038/s41420-017-0018-1
Publisher site
See Article on Publisher Site

Abstract

2+ 2+ Nuclear envelope (NE) is a Ca -storing organelle controlling neuronal differentiation through nuclear Ca 2+ 2+ concentrations ([Ca ] ). However, how [Ca ] regulates this important function remains unknown. Here, we n n + 2+ investigated the role of the nuclear form of the Na /Ca exchanger 1(nuNCX1) during the different stages of neuronal differentiation and the involvement of PTEN/PI3’K/Akt pathway. In neuronal cells, nuNCX1 was detected on the inner membrane of the NE where protein expression and activity of the exchanger increased during NGF-induced differentiation. nuNCX1 activation by Na -free perfusion induced a time-dependent activation of nuclear-resident PI3K/Akt pathway in isolated nuclei. To discriminate the contribution of nuNCX1 from those of plasma membrane NCX, we generated a chimeric protein composed of the fluorophore EYFP, the exchanger inhibitory peptide, and the nuclear localization signal, named XIP-NLS. Fura-2 measurements on single nuclei and patch-clamp experiments in whole-cell configuration showed that XIP-NLS selectively inhibited nuNCX1. Once it reached the nuclear 2+ compartment, XIP-NLS increased the nucleoplasmic Ca peak elicited by ATP and reduced Akt phosphorylation, GAP- 43 and MAP-2 expression through nuclear-resident PTEN induction. Furthermore, in accordance with the prevention of the neuronal phenotype, XIP-NLS significantly reduced TTX-sensitive Na currents and membrane potential during neuronal differentiation. The selective inhibition of nuNCX1 by XIP-NLS increased the percentage of β III tubulin- positive immature neurons in mature cultures of MAP-2-positive cortical neurons, thus unraveling a new function for 2+ nuNCX1 in regulating neuronal differentiation through [Ca ] -dependent PTEN/PI3K/Akt pathway. Introduction dysregulation may affect cell survival, as it occurs under 2+ 2+ Nuclear Ca concentrations ([Ca ] ) regulate many hypoxic conditions where it triggers the expression of pro- 1 5,6 2+ 2+ cellular functions by modulating specific proteins .For apoptotic genes .[Ca ] is influenced by cytosolic Ca 2+ instance, the nuclear-targeted Ca buffer protein parval- levels and is regulated by the activation of specific receptors, bumin, but not the cytosolic isoform, reduces hepatocyte pumps, and exchangers localized on the nuclear envelope 2 7,8 proliferation . Moreover, the activation of the transcription (NE) . For instance, the stimulation of nuclear inositol 3 4 factors Elk1 , NFAT, DREAM , and CREB depends on (1,4,5)-triphosphate receptors (InsP R) leads to nucleo- 2+ 2+ 2+ nuclear Ca levels. On the other hand, intranuclear Ca plasmic Ca increase in isolated nuclei from Xenopus 9 10 11 oocytes ,Aplysia neurons , and pancreatic acinar cells . Furthermore, the activation of G-protein coupled receptors Correspondence: Lucio Annunziato (lannunzi@unina.it) 2+ localized on the NE modulates—via [Ca ] —gene IRCCS SDN, School of Medicine, “Federico II” University of Naples, Naples, Italy IRCCS SDN, Naples, Italy Agnese Secondo and Alba Esposito contributed equally to this work. Edited by N. Barlev © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890 Secondo et al. Cell Death Discovery (2018) 4:12 Page 2 of 14 transcription, import and export pathways of proteins, and Western blot only on the inner membrane attached to the cell cycle . In addition, outer and inner membranes of NE nuclei and in total lysate of both whole cells and total + − 2+ 13–15 are provided with K ,Cl , and R-type Ca channels , nuclei (Fig. 1D). Efficient isolation of INM was confirmed + + 16 2+ 17 Na /H exchanger 1 ,Ca -ATPase pump ,and by the expression of Lamin B1/B2 and CRM1, a member 18,19 NCX1 . Despite the existent knowledge on the role of of nuclear transport factor family (Fig. 1D). Furthermore, plasma membrane NCX in the regulation of intracellular only the preparation of outer NE membrane expressed 2+ 2+ 20–22 calreticulin, an ER/SR protein that is contiguous with the Ca concentrations and Ca -dependent pathways , nuNCX1 function remains unexplored. Interestingly, global outer NE (Fig. 1D). silencing of NCX1 prevents neurite outgrowth through the inhibition of Akt pathway thus reducing NeuN-positive Expression and function of NCX isoforms in NE neurons . On the other hand, it has been reported that In an attempt to study the involvement of nuNCX1 in the miR-29a targeting the phosphatase and tensin homologs on different stages of neuronal differentiation, immunolocali- chromosome ten (PTEN), increase phosphorylation of Akt, zation of this isoform was performed also in PC12 cells thus promoting neurite outgrowth . Indeed, deregulation exposed to NGF for 3 days before complete differentiation. of PTEN affects neurogenesis, neurite outgrowth, synapto- At 3 days, DAB immunostaining of NCX1 was almost totally 25,26 genesis, and synaptic plasticity . Therefore, the purpose detected in nuclear compartment (Fig. 2A, black arrow) and of the present study has been to investigate the contribution slightly localized in some part of plasma membrane (Fig. 2A, of nuNCX1 in the modulation of nuclear-resident PTEN/ gray arrow). qRT-PCR experiments showed that, during 2+ PI3K/Akt pathway via [Ca ] during neuronal differentiation with NGF, NCX1 transcripts increased in differentiation. PC12 cells with a significant peak at 3 days of exposure (Fig. 2B). On the other hand, the transcripts of the Results third isoform of the exchanger, NCX3, significantly Characterization of nuNCX1 expression in neuronal cells reduced at 3 and 7 days of exposure (in A.U. ± SEM: control Immunolocalization of the NCX1 isoform in NGF- = 1.02 ± 0.03; 3 days NGF= 0.52 ± 0.05 *p < 0.05 vs. differentiated PC12 cells revealed diffused DAB immune- control; 7 days NGF= 0.55 ± 0.08, *p < 0.05 vs. control). staining of this isoform. Together with its plasmamembrane Moreover, Western blot analysis revealed that nuNCX1 expression, NCX1 was also present at the nuclear level expression in PC12 cells peaked after 3 days exposure to (Fig. 1Aa,b,c). Furthermore, in specificNE areas, NCX1co- NGF in nuclear preparations (Fig. 2C), thus suggesting a localized with Lamin B1/B2 (Fig. 1Ba), a component of specific role played by nuclear NCX1 in triggering nuclear lamina connected with the nucleoplasmic face of the differentiation at that time. By contrast, NCX3 protein inner membrane (INM). However, when NGF-differentiated expression dropped dramatically in the same preparation PC12 cells were transfected with siRNA against NCX1 (data not shown). (siNCX1), Lamin B1/B2-NCX1 co-localization disappeared Then, NCX1 activity was studied by exposing to a (Fig. 1Ba, c). Moreover, in whole neuronal cells, NCX1 Na -free solution single nuclei loaded with Fura-2AM, a immunosignal was also found outside the nucleus (Fig. 1Ba) fluorescent dye exclusively concentrated in NE lumen . and disappeared in cells treated with siNCX1 (Fig. 1Bb). Na -free perfusion forced the exchanger to operate in the 2+ Confocal analysis showed a significant immunosignal of forward mode, moving Ca from the nucleoplasm to 2+ NCX1 on the NE of isolated nuclei derived from the NE lumen. This elicited an increase of Ca level in the 2+ differentiated PC12 cells (Fig. 1Ca, b). Nissl staining showed NE lumen. In particular, this [Ca ] increase induced by the lack of ER contamination and the integrity of isolated Na -free exposure was much higher in nuclei obtained nuclei while ethidium bromide staining confirmed the high from PC12 cells exposed to NGF for 3 days than those purity of these preparations and the lack of dispersed DNA exposed for 7 days or undifferentiated controls (Fig. 2D). In (data not shown). In addition, any possible contamination by the same nuclear preparation, a pharmacological char- specific markers of the plasmamembrane, mitochondria, acterization of NCX1 activity in the forward mode was Golgi complex, and ER such as the plasma membrane performed by the Na ionophore monensin, or blocking calcium ATPase, Na channels, dynamin-related protein 1, NCX by the amiloride analog 3-amino-6-chloro-5-[(4- GM130, and stromal interaction molecule 1 were detected chloro-benzyl)amino]-n-[[(2,4-dimethylbenzyl)-amino]imi- in the preparation of isolated nuclei, except for the two nomethyl]-pyrazinecarboxamide (CB-DMB) .Whereas nuclear markers H3 and Akt (data not shown). monensin was able to enhance the Na -dependent activity To further confirm the specific localization of NCX1 at of the exchanger, possibly by increasing intra-nuclear Na 2+ INM level, we performed a separation between outer and levels, CB-DMB prevented the increase in NE Ca level inner membranes of NE with cold Na-citrate in isolated associated with NCX activation (Fig. 2E). Furthermore, NE 2+ nuclei obtained from NGF-differentiated PC12 cells. Once Ca increase was completely prevented in isolated nuclei the outer membrane was removed, NCX1 was detected by from PC12 cells previously treated with siNCX1(Fig. 2E). Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 3 of 14 + 2+ Fig. 1 Localization of neuronal Na /Ca exchanger 1 at the inner membrane of the NE. a Immunolocalization of NCX1 in whole cell obtained by 3,3′-diaminobenzidine in differentiated PC12 cells treated with NGF for 7 days. Scale bars: 50 µm (a), 20 µm (b, c). b Co-localization of NCX1 (red) with Lamin B1/B2 (green), the inner membrane of the NE marker, in differentiated PC12 cells detected by confocal microscopy under control conditions treated with siControl (a) and in NCX1-silenced PC12 cells (b). Scale bars: 5 µm. c Confocal images representing immunofluorescent signal of NCX1 (green) and Hoechst (blue) in isolated nuclei from differentiated PC12 cells under control conditions (a, b). d Western blot of NCX1, Lamin B1/B2, CRM1 and calretinin in total and nuclear lysates, inner and outer NE membranes of NGF-differentiated PC12 cells. All the experiments were repeated at least three times on different preparations nuNCX1 silencing prevents nuclear-resident Akt effect was prevented by pre-treating nuclei for 10 min phosphorylation induced by Na -free perfusion with the PI3′-K inhibitor LY294002 (25 µM) before To verify whether nuNCX1 activation could modulate Na -free exposure (Fig. 3b). Interestingly, in nuclear 2+ the phosphorylation of nuclear-resident Akt, a Ca preparations obtained from neuronal cells previously -dependent kinase mainly involved in neurite outgrowth, exposed to siNCX1, Na -free perfusion (5 min) failed to we exposed isolated nuclei to a Na -free solution under induce Akt phosphorylation compared to siControl controlled conditions of temperature. Isolated nuclei were cells (Fig. 3c). obtained from PC12 cells treated with NGF for 3 days, a time of the highest expression of nuNCX1 (see Fig. 2). Selective inhibition of nuNCX1 by the chimeric protein + 2+ Na -free induced phosphorylation of nuclear-resident XIP-NLS increases nucleoplasmic Ca level thus limiting Akt in isolated nuclei from PC12 cells (Fig. 3a). Moreover, neuronal differentiation Na -free-induced Akt phosphorylation was evident Forty-eight hours after the transfection of XIP-NLS already after 5 min of exposure (Fig. 3a). The same acti- (Fig. 4a), confocal microscopy and immunoblot analysis vation of Akt was induced by exposing intact nuclei revealed that XIP-NLS was selectively localized in the to bradykinin (BK) (Fig. 3a), whose Gq-coupled nuclei of NGF-differentiated PC12 cells (Figs. 4b, c). receptors are located on the NE inner membrane . This Indeed, EYFP signal co-localized with nuclear DNA Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 4 of 14 Fig. 2 nuNCX1 expression and activity in nuclei from PC12 cells during differentiation with NGF. a Immunolocalization of NCX1 in whole cell obtained by 3,3′-diaminobenzidine in PC12 cells treated with NGF for 3 days. Scale bars: 50 µm (a), 20 µm (b). b Representative qRT-PCR of ncx1 transcript expression in control cells and in PC12 cells after 3 and 7 days exposure to NGF. c Representative Western blot and quantification of nuNCX1 protein expression in isolated nuclei from PC12 cells in control conditions and after 3 and 7 days exposure to NGF. For b and c: all the experiments were repeated at least three times on different nuclear preparations; *p < 0.05 vs. undifferentiated control, **p < 0.05 vs. undifferentiated 2+ + control and 3 days NGF. d Top: Representative traces of the effect on [Ca ] of Na -free perfusion in Fura-2-loaded nuclei isolated from undifferentiated PC12 cells (control), and PC12 cells exposed to 3 days or 7 days to NGF. Bottom: quantification of the effect of d.*p < 0.001 vs. 2+ + undifferentiated control, **p < 0.05 vs. undifferentiated control and 3 days NGF. e Quantification of the effect on [Ca ] of Na -free in isolated nuclei obtained from NGF-differentiated PC12 cells pre-treated with: monensin (10 min), CB-DMB (20 min), and siControl or siNCX1 (at DIV 5). *p < 0.001 vs. undifferentiated control, **p < 0.001 vs. all groups. Inset: Representative Western blot depicts the effect of siControl and siNCX1 on nuNCX1 protein expression in isolated nuclei from NGF-differentiated PC12 cells staining Hoechst 33258 in nuclei efficiently transfected detected by Fura-2 microfluorimetry (Figs. 5a, b). By with XIP-NLS (Fig. 4b). Furthermore, the specificity of the contrast, as expected by the nuclear selectivity of the nuclear localization of XIP-NLS was revealed by the chimeric protein, XIP-NLS did not affect NCX currents co-detection, after 48 h, with the nuclear markers CREB measured by patch-clamp at plasma membrane level in and pCREB in the same fraction. Further evidence for the NGF-differentiated PC12 cells (Figs. 5c, d). nuclear localization of XIP-NLS and for the purity of our In NGF-treated PC12 cells transfected with XIP-NLS, cellular fractions was the absence of the cytosolic protein this peptide significantly increased Fura-dextran-detected 2+ SOD1 in the nuclear fraction (Fig. 4c). nucleoplasmic Ca peak in response to the Gq-protein Furthermore, XIP-NLS inhibited Na -dependent coupled receptor agonist ATP (100 µM) compared to nuNCX1 activity elicited by Na -free perfusion in isolated mock control (Fig. 6a). Once transfected, XIP-NLS nuclei obtained from NGF-differentiated PC12 cells, as significantly reduced GAP-43 expression and Akt Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 5 of 14 Fig. 3 Nuclear-resident Akt activation by Na -free exposure in nuclear preparations from neuronal cells. a Representative Western blot and quantification of phosphorylated Akt in isolated nuclei from NGF-differentiated PC12 cells (3 days) after perfusion with Na -free or bradykinin (BK). b Representative Western blot and quantification of phosphorylated Akt in isolated nuclei from NGF-differentiated PC12 cells previously treated with the PI3’K inhibitor LY294002 c Representative Western blot and quantification of the effect of Na -free (5 min) in isolated nuclei obtained from NGF- differentiated PC12 cells (3 days) previously treated with siControl or siNCX1. For a, b, c: all the experiments were repeated at least three times on + + different nuclear preparations. *p < 0.05 vs. untreated control or siControl, both not exposed to Na -free. **p < 0.05 vs. control or siControl + 5′ Na - free phosphorylation in PC12 cells exposed to NGF for expressed karyopherin β2 (Fig. 7a). Interestingly, once 3 days compared to mock control (Fig. 6b). Interestingly, transfected at higher concentrations in cortical neurons, under the same conditions, XIP-NLS significantly XIP-NLS carrying EYFP co-immunoprecipitated with increase the expression of the negative regulator nuNCX1 as detected with anti-NCX1 in isolated nuclei of PI3K/Akt pathway, phosphatase and tensin homolog (Fig. 7b). As occurred in NGF-treated PC12 cells, (PTEN) (Fig. 6b), mainly involved in axonal growth XIP-NLS significantly increased Fura-dextran-detected 24 2+ modulation . Then, in accordance with the prevention of nucleoplasmic Ca peak in response to ATP (100 µM) the neuronal phenotype, XIP-NLS significantly reduced in cortical neurons (Fig. 7c). This suggests that XIP-NLS + 2+ TTX-sensitive Na currents in PC12 cells exposed to was able to reduce Ca buffering properties of neuronal NGF for 3 days (Figs. 6, d). Furthermore, XIP-NLS nuclei. Moreover, also in mature neurons (7 DIV), significantly decreased membrane potential in PC12 cells the chimeric protein reduced terminal differentiation exposed to NGF for 3 days compared to the respective markers, such as GAP-43 and MAP-2 (Figs. 7d, e), controls (Fig. 6e). significantly increased PTEN expression and decreased A separation between outer and inner NE membranes Akt phosphorylation not only in total lysates (Fig. 7d, e) of isolated nuclei obtained from primary cortical neurons but also in isolated nuclei (Fig. 8a) from cortical neurons. showed a specific localization of NCX1 on the inner In order to strength the evidence demonstrating the role membrane attached to the nuclei (Fig. 7a). Furthermore, of nuclear NCX1 isoform in the differentiation of primary efficient isolation of INM in cortical neurons was con- cortical neurons, we analyzed the effect of the nuclear firmed by the expression of Lamin B1/B2 (Fig. 7a). On the inhibitor of NCX1, XIP-NLS, on the markers of both other hand, the preparation of outer NE membrane immature and mature rat cortical neurons. To this aim we Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 6 of 14 Fig. 4 Subcellular localization of the chimeric protein XIP-NLS. a Scheme of XIP-NLS structure. b Immunolocalization of XIP-NLS (green, 400 ng/ ml) detected 48 h after transfection. Hoechst dye (blue) was used to mark nuclei. Under these conditions, EYFP-positive nuclei were at least 25 ± 2% of total Hoechst-positive nuclei. c Western blot of EYFP, CREB, pCREB, and SOD1 in cytosolic and nuclear fractions obtained from NGF-differentiated PC12 cells transfected with XIP-NLS (400 ng/ml). Fractions were prepared 48 h after transfection performed confocal double immunofluorescence experi- localization was paralleled by its specific activation. In ments with the anti-βIII-tubulin, a marker of immature particular, working in the forward mode of operation, 2+ neurons, and microtubule-associated protein-2 (MAP-2) nuNCX1 moved Ca from the nucleoplasm to the NE that is expressed throughout all stages of neuronal dif- lumen of neuronal cells. Indeed, NCX1 silencing, as well as ferentiation. We found that the selective inhibition of the pharmacological inhibition of its activity by the 2+ nuclear NCX1 by transfecting XIP-NLS increased the amiloride derivative CB-DMB, abrogated Ca rise in the percentage of βIII-tubulin-positive immature neurons lumen of NE due to the forward mode of the nuclear after 7 DIV compared with control cultures but did not exchanger, thus demonstrating that nuNCX1 has the same change the percentage of MAP-2-positive neurons pharmacological sensitivity of its plasmamembrane form. (Fig. 8b). Therefore, nuNCX1 contributes to the Na -dependent 2+ removal of [Ca ] from nucleoplasm of neuronal cells. In 2+ Discussion this regard, we speculated that Ca extrusion from the The present study demonstrates that the nuclear Na / total nuclear compartment of a whole cell might entail two 2+ Ca exchanger isoform 1 (nuNCX1), located at the level consecutive steps: the first step, operated by NCX1 located 2+ of the inner membrane of NE (INM) in neurons, plays a on the inner membrane of the NE, might mediate Ca relevant role in neuronal differentiation through the transport from the nucleoplasm to the lumen of the NE , 2+ modulation of nuclear Ca homeostasis. However, we and the second step, whose transporter mechanism is still 2+ found NCX1 immunosignal also outside the INM, possibly unknown, might promote Ca movement from the at the level of other intracellular organelles, whereas no lumen of the nuclear envelope into the ER. Such new detectable traces of the other two NCX isoforms, NCX2 function might be supported by the presence of a Na and NCX3 were detected on INM. Compelling evidence gradient existing between the lumen of the nuclear for NCX1 specific localization to the INM is that, in envelope and the nucleoplasm. Consistently, previous nuclear preparations obtained from cortical neurons in evidence indicates the presence of a Na concentration which outer membrane was removed, nuNCX1 was co- gradient between the lumen (80 mM) and the nucleoplasm expressed with specific markers of the INM, namely (5–20 mM), that is maintained by the nuclear form of the + + 31 laminB1/B2 and CRM1. Furthermore, this specific Na /K ATPase . Indeed, this pump is present in the Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 7 of 14 2+ + Fig. 5 Functional characterization of the chimeric protein XIP-NLS. a Representative traces of Fura-2-detected [Ca ] after Na -free perfusion in isolated nuclei prepared from NGF-differentiated PC12 cells previously transfected with mock or XIP-NLS. b Quantification of a.*p < 0.001 vs. all. Control (n = 15 cells); mock (n = 20 cells); XIP-NLS 40 (n = 15 cells); XIP-NLS 400 (n = 15 cells). c Superimposed traces of I currents recorded by patch-clamp in NCX NGF-differentiated PC12 cells previously transfected with mock or XIP-NLS. d Quantification of c.mock(n = 20 cells); XIP-NLS (n = 20 cells) inner membrane of the nuclear envelope and is oriented plasma membrane NCX level . Interestingly, XIP-NLS 27,31 2+ with the ATP hydrolysis site in the nucleoplasm . inhibited nuNCX1 Ca -extruding function without Indeed, the present finding is in line with our previous affecting the activity of plasma membrane NCX1, as results demonstrating that NCX1 participates in NGF- demonstrated by patch-clamp experiments in whole induced neuronal differentiation via nuclear Akt . neuronal cells. These data suggest that nuclear NCX1 2+ However, the present data seems to suggest that, among plays a crucial role in regulating nuclear Ca home- 2+ the different exchangers, nuNCX1 plays a major role in ostasis upon [Ca ] increase and works to remove 2+ inducing neuronal differentiation working in the early nucleoplasmic Ca accumulation in whole cell under phase of the process. Consistently, nuNCX1 expression physiological conditions. and activity progressively increased during differentiation, Furthermore, the activity of nuNCX1 was involved in the thus peaking at 3 days of NGF exposure. At this time, regulation of PTEN/PI3K/Akt pathway, a transduction nuNCX1 mainly controls nuclear-resident PTEN/Akt signal regulating neuronal differentiation. Moreover, 2+ pathway via [Ca ] modulation. Indeed, when we during neuronal differentiation, the expression of PTEN, a inhibited nuNCX1 with the chimeric fluorescent protein phosphatase that converts phosphatidylinositol 3,4,5-tri- 2+ XIP-NLS, nuclear Ca clearance decreased, whereas the sphosphate to phosphatidylinositol 3,4-bisphosphate, 2+ [Ca ] peak, elicited by ATP, increased. This effect on was found to be inversely correlated to that of activated 2+ nucleoplasmic Ca was due to the selective localization Akt. Specifically, the inhibition of nuNCX1 by XIP-NLS of XIP-NLS in the nucleus where it binded to the f-loop of leaded to an increase in nuclear-resident PTEN expression nuNCX1, as also demonstrated by the coimmunopreci- and a decrease in Akt phosphorylation. Since nuNCX1, 2+ pitation between the peptide and nuNCX1 in isolated working in the forward mode, moves Ca from nucleo- nuclei. Accordingly, the same interaction occurs at plasm to NE lumen, it is conceivable that the blockade of Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 8 of 14 2+ Fig. 6 Effect of the chimeric protein XIP-NLS on [Ca ] and neuronal differentiation in NGF-treated PC12 cells. a Representative traces of 2+ Fura-dextran-detected nucleoplasmic Ca after ATP (100 µM) perfusion in NGF-treated PC12 cells (3 days) previously transfected with mock or XIP- 2+ NLS. Bar graph depicts quantification of ATP-induced nucleoplasmic Ca increase. *p < 0.05 vs. mock. mock (n = 20 cells); XIP-NLS (n = 15 cells). b Representative Western blots and quantifications of GAP-43, PTEN protein expression, and Akt phosphorylation in NGF-treated PC12 cells (3 days) previously transfected with mock or XIP-NLS. The experiments were repeated at least three times on different preparations; *p < 0.05 vs. undifferentiated cells; **p < 0.05 vs. mock. c and d Representative traces and quantification of voltage-gated sodium currents (I ) recorded from NaV PC12 cells under control conditions (undifferentiated, n = 6), after exposure to NGF for 3 days and mock transfection (n = 10) and after exposure to NGF for 3 days and NLS-XIP transfection (n = 20). The normalization for membrane capacitances produced similar results. e Quantification of membrane potential in all the conditions aforementioned. *p < 0.001 vs. undifferentiated cells; **p < 0.001 vs. NGF-treated PC12 cells + Mock this flux may promote, at nuclear level, a nucleoplasmic positive cortical neurons, thus suggesting that the selective 2+ Ca -dependent activation of nuclear-resident PTEN. inhibition of nuNCX1 hampered neuronal differentiation Indeed, a part the well-established role in antagonizing of cortical neurons. This is in line with a very recent paper PI3’K signaling at the plasma membrane level, PTEN has showing that the chronic treatment with NCX inhibitors been localized in the nucleus where it modulates tran- produce marked reduction of the Purkinje cell dendritic 32 35 scriptional activity . In addition, PTEN appears during arbor . 33,34 elongation of growing neurites , and increases pro- Collectively, this study unravels a new function for gressively at the early stage of differentiation, possibly to nuclear NCX1 within the complex regulatory machinery prevent aberrant neurite extension. In our study, once of nuclear calcium homeostasis, suggesting that this upregulated, PTEN was able to reduce Akt phosphoryla- exchanger form may control neuronal differentiation 2+ tion, MAP-2 and GAP-43 expression, thus avoiding through the fine regulation of Ca -dependent neuronal differentiation in cortical neurons. Another transduction mechanisms. interesting finding of this study is that XIP-NLS increased the percentage of βIII-tubulin-positive immature neurons -after 7 DIV- in mature cultures of MAP-2- Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 9 of 14 2+ Fig. 7 Effect of the chimeric protein XIP-NLS on [Ca ] and neuronal differentiation in primary cortical neurons. a Western blot of NCX1, Lamin B1/B2, and karyopherin in inner and outer NE membranes of cortical neurons at 7 DIV. All the experiments were repeated at least three times on different preparations. b Nuclear lysates from neurons trasfected with XIP-NLS (48 h) and immunoprecipitated with anti-EYFP (top) or anti-NCX1 (bottom). The presence of NCX1 (top) or EYFP (bottom) was analyzed by immunoblotting performed with specific antibodies. The input consists on 2+ nuclear lysates from untreated neurons at 7 DIV. c Representative traces of Fura-dextran-detected nucleoplasmic Ca after ATP (100 µM) perfusion in 2+ cortical neurons previously transfected with mock or XIP-NLS. Bar graph depicts quantification of ATP-induced nucleoplasmic Ca increase. *p < 0.05 vs. mock. mock (n = 15 cells); XIP-NLS (n = 10 cells). d, e Representative Western blots and quantifications of GAP-43, MAP-2, PTEN protein expression, and Akt phosphorylation in cortical neurons (7DIV) previously transfected with mock or XIP-NLS. All the experiments were repeated at least three times on different preparations; *p < 0.05 vs. mock Materials and methods anesthetized and then decapitated to minimize pain and Cell cultures distress. Dissection and dissociation were performed in 2+ 2+ Clonal cells Ca /Mg -free PBS containing glucose (30 mM). Tissues PC12 cells were grown on plastic dishes in RPMI were incubated with papain for 10 min at 37 °C and dis- medium composed of 10% horse serum (HS), 5% fetal sociated by trituration in Earle’s Balanced Salt Solution bovine serum (FBS), 100 UI/ml penicillin, and 100 μg/ml containing DNase, bovine serum albumin (BSA), and streptomycin. Neuronal differentiation was induced by ovomucoid. Cells were plated at 15 × 10 in 100-mm exposing PC12 cells to NGF (50 ng/ml) were cultured in a plastic Petri dishes pre-coated with poly-D-lysine (20 μg/ humidified 5% CO atmosphere. Culture medium was ml) in minimum Eagle’s medium/F12 (Life Technologies, changed every 2 days. For functional studies, cells were Milan, Italy) containing glucose, 5% deactivated FBS, 5% seeded on glass coverslips (Fisher, Springfield, NJ, USA), deactivated HS (Life Technologies, Milan, Italy), gluta- coated with poly-L-lysine (5 μg/ml) (Sigma, St. Louis, mine, and antibiotics. Cytosine β-D-arabinofuranoside Missouri, USA), and used at least 12 h after seeding. hydrochloride (Ara-C; 10 μM) was added within 48 h of plating to prevent non-neuronal cell growth. Neurons were cultured at 37 °C in a humidified 5% CO atmo- Embryonic neurons 2 sphere and used after 7 days of culture. All experiments cortical pure neurons were prepared from brains of 16- on primary cortical neurons were performed according to day-old Wistar rat embryos. Briefly, the rats were first Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 10 of 14 Fig. 8 Effect of nuclear NCX1 inhibition on nuclear-resident PTEN/Akt pathway and on βIII-tubulin and MAP-2 immunostaining in mature cortical neurons. a Representative Western blots and quantifications of PTEN expression and Akt phosphorylation in nuclear preparations from mature cortical neurons (7DIV) transfected at 5 DIV with mock or XIP-NLS. All the experiments were repeated at least three times on different preparations. *p < 0.001 vs. mock. b Top: βIII-tubulin and MAP-2 staining in cortical neurons at 7 DIV under control conditions and in neurons transfected with XIP-NLS (bar = 20 μm; nuclei Hoechst blue). Bottom: Quantification of MAP-2 and βIII-tubulin fluorescence and % of their colocalization in cortical neurons at 7 DIV the procedures described in experimental protocols preparations were stained with ethidium bromide and/or approved by the ethical committee of the “Federico II” Nissl solution to study the purity, integrity, and possible University of Naples, Italy. ER contamination of nuclear preparations. Isolation of intact nuclei Inner and outer nuclear membranes separation Highly purified nuclei were obtained from cell cultures Freshly isolated nuclei of PC12 cells and cortical by the combined use of hypotonic shock and mechanical neurons were employed for isolation of NE as described disruption. Briefly, when they reached confluence, cells previously . Nuclear preparations were treated were washed twice in saline phosphate buffer and lysed in with 2% Na-citrate (trisodium, dihydrate) in a Nuclear hypotonic buffer (10 mM Tris-HCl, pH 7.8, 10 mM β- Resuspension Buffer containing (in mM): 10 HEPES pH mercaptoethanol, 0.5 mM phenylmethylsulfonyl fluoride 7.9, 10 KCl, 1.5 MgCl , 0.1 EGTA, 0.5 DTT, 1 (PMSF), 1 μg/ml aprotinin, leupeptin, and pepstatin). After sodium orthovanadate (Na VO ), 0.2 PMSF, 350 sucrose, 3 4 20 minutes on ice, the swollen samples were mechanically protease inhibitors. Nuclei were pelleted by brief disrupted. The nuclear pellet was obtained by centrifuga- centrifugation at 5000×g: the pellet containing inner tion at 500 g for 6 min at 4 °C and was then washed in a nuclear membranes was resuspended in nuclear solution containing 10 mM Tris-HCl, 2 mM MgCl2 plus resuspension buffer, while the supernatant containing protease inhibitors, pH 7.2. Finally, the nuclear pellet was outer nuclear membranes was precipitated with 10% resuspended in a standard solution (KCl 125 mM, HEPES trichloroacetic acid . 50 mM, K2HPO4 2 mM, EGTA 0.1 mM, CaCl2 0.3 mM, 11,34 NaCl 3 mM, containing 1 mM ATP, pH 7.4) . Nuclear Nuclear/cytosolic fractionation and western blot analysis preparations were used or stored at −80 °C until use. To Cytosolic and nuclear fractions were obtained from check nuclear integrity, isolated nuclei were plated on glass NGF-differentiated PC12 cells. In brief, cells were washed coverslips (Fisher, Springfield, NJ, USA) previously coated twice and collected in saline phosphate buffer by with poly-L-lysine (30 µg/ml), and, then, fixed with 4% centrifugation at 1500 rpm for 3 min. Then, the pellet PAF (w/v) for 20 min at RT. After fixing, nuclear was dissolved in a buffer containing (in mM) 10 Hepes Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 11 of 14 pH 7.9, 10 KCl, 1.5 MgCl , 0.1 EGTA, 0.5 DTT, 1 sodium pCREB (rabbit polyclonal anti-body, Millipore Corpora- orthovanadate (Na VO ), 0.2 PMSF, and protease inhi- tion, MA, USA). 3 4 bitors; the cytosolic fraction was then obtained by two Membranes containing inner and outer nuclear frac- consecutive centrifugations at 3000 rpm for 5 min at 4 °C. tions were incubated overnight at 4 °C in the blocked Nuclear fractions were obtained dissolving the pellet in a buffer containing the following antibodies: 1:1000 anti- solution containing (in mM) 20 Hepes pH 7.9, 400 NaCl, NCX1 (rabbit polyclonal antibody, Swant, Bellinzona, Swiss); 1:200 anti-Lamin B1/B2 (mouse monoclonal 1.5 MgCl , 0.1 EGTA, 25% glycerol, 1 Na VO ,0.2 PMSF 2 3 4 and protease inhibitors; throughout the procedure, the antibody, Abcam, Cambridge, UK), 1:1000 calretinin samples were kept on an agitation system for 30 min at (rabbit polyclonal antibody, GeneTex, Inc., GA, USA), and 4 °C. Finally, samples were centrifuged at 13,000 rpm for 1:1000 Karyopherin β2 (mouse monoclonal antibody, 15 min. Santa Cruz Biotechnologies, Inc., CA, USA). Then, For Western blot analysis, cell cultures, inner and outer membranes were washed with 0.1% Tween 20 and incu- nuclear membranes, and nuclear preparations were lysed bated with the secondary antibodies (1:2000; GE Health- with a buffer containing 20 mM tris–HCl (pH 7.5), 10 mM care, Milan, Italy) for 1 h. Immunoreactive bands were NaF, 1 mM phenylmethylsulfonyl fluoride, 1% NONIDET detected using ECL reagent kits (GE Healthcare, Milan, P-40, 1 mM Na VO , 0.1% aprotinin, 0.7 mg/ml pepstatin, Italy). The optical density of the bands was determined by 3 4 and 1 μg/ml leupeptin. Samples were cleared by cen- Chemi-Doc Imaging System (Bio-Rad Laboratories, trifugation and supernatants were used for Western blot Milan, Italy). analysis. Protein concentration in supernatants was determined by the Bradford method . Protein samples qRT-PCR analysis (50 μg) were analyzed on 8% sodium dodecyl sulfate qRT-PCR was performed in a 7500-fast real-time PCR polyacrylamide gel with 5% sodium dodecyl sulfate system (Applied Biosystems) by Fast SYBR Green Master stacking gel (SDS-PAGE) and electrotransferred onto Mix (Applied Biosystems). Samples were amplified simul- Hybond ECL nitrocellulose papers (GE Healthcare, Little taneously in triplicate as follows: heating 2 min at 50 °C, Chalfont, UK). Membranes were blocked with 5% non-fat denaturation 10 min at 95 °C, amplification and quantifica- dry milk in 0.1% Tween 20 (TBS-T; 2 mmol/l Tris–HCl, tion 35 cycles of 15 s at 95 °C; 1 min at 60 °C with a single 50 mmol/l NaCl, pH 7.5) for 2 h at RT. Subsequently, they fluorescence measurement. The data were normalized by were incubated overnight at 4 °C in the blocked buffer using hypoxanthine phosphor-ribosyl-transferase (HPRT) as containing the following antibodies: 1:1000 anti-NCX1 an internal control. Differences in mRNA content between (rabbit polyclonal antibody, Swant, Bellinzona, Swiss), groups were expressed by using 2−ΔΔCt.formula. The oli- 1:1000 antibody for β-actin (mouse monoclonal antibody, gonucleotide sequences were for NCX1, forward: AGATTC Santacruz Biotechnology, Inc. CA), 1:500 anti-pAkt CGTGACTGCAGTTGTG and reverse: ATACTGGTCCT (mouse monoclonal antibody, Cell Signaling, Inc., MA, GGGTAGCTGCTA; for NCX3, forward: CGGTCACAG USA), 1:1000 anti-Akt (rabbit polyclonal antibody, Santa CTGTTGTTTTTGT and reverse: CAGGGCAGCAGCTT Cruz Biotechnology, Inc., CA, USA), 1:1000 anti-CRM1 TGCT;for HPRT,forward:TGGAAAGAACGTCTTG (goat polyclonal antibody, Santa Cruz Biotechnology, Inc, ATTGTTGA and reverse: GCTGTACTGCTTGACC CA, USA), 1:1000 anti-calretinin (rabbit polyclonal anti- AAGGAA. body; GeneTex, CA, USA), 1:500 anti-GAP-43 (mouse monoclonal antibody, Millipore Corporation, MA, USA), Small interfering RNA 1:1000 anti-MAP-2 (mouse monoclonal antibody, Sigma- NCX1 knock down was achieved with siRNA duplexes Aldrich, Milan, Italy), 1:1000 anti-PTEN (mouse mono- against NCX1 and its non-targeting control (Invitrogen, clonal antibody, Santa Cruz Biotechnology, Inc., CA, San Giuliano Milanese, Milan, IT). The following gene- USA) and 1:2000 anti-α-tubulin (mouse monoclonal specific sequences were used successfully: antibody, Sigma-Aldrich, Milan, Italy). Membranes were (RNA)-CCAAAUGGAGAGACCACCAAGACU and (RNA)- blocked with 5% non-fat dry milk in 0.1% Tween 20 (TBS- UAGUCUUGGUGGUCUCUCCAUUUGG or (RNA)-AC T; 2 mmol/l tris–HCl, 50 mmol/l NaCl, pH 7.5) for 2 h at AUGUUCCUCGGAGUUUCUAUUAU and (RNA)-AUA RT. Membranes containing nuclear fractions from neu- AUAGAAACUCCGAGGAACAUGU. After 5 days in ronal cells treated with XIP-NLS were incubated over- culture with NGF, PC12 cells were transfected for 5 h with night at 4 °C in the blocked buffer containing the each duplex at a final concentration of 10 nM using following antibodies: 1:1000 anti-SOD1 (rabbit polyclonal HyPerFect transfection reagent (Qiagen, Milan, IT). anti-body, Santa Cruz Biotechnology, Inc., CA, USA), 1:1000 anti-GFP (mouse monoclonal anti-body, Abcam, Generation and expression of XIP-NLS Cambridge, UK), 1:1000 anti-CREB (mouse monoclonal The expressing vector encoding for XIP-NLS protein anti-body, Cell Signaling Technology, Inc. MA, USA) and was obtained by sub-cloning in frame 2 copies of each XIP Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 12 of 14 and NLS cDNAs in EYFP-C3 plasmid (Clontech Labora- and Cy3 anti-rabbit IgG (1:200, Jackson Immuno Research tories, Mountain View, CA), as shown in Fig. 4 panel A. In Laboratories, Inc. PA, USA). The immunosignal of particular, 2 copies of the cDNA (from rat NCX1.1 NCX1 shown in Fig. 1 panel A and Fig. 2 panel A was NM_019268.3) encoding for XIP domain (DRRLLFYKY- detected by using a biotinylated secondary antibody. After VYKRYRAGKQR) were obtained by PCR on a single this latter incubation, the peroxidase reaction was devel- plasmid and digested by specific restriction enzymes, oped using 3,3′-diaminobenzidine/4-HCl as a chromogen. Then, after the final wash, coverslips were mounted with whereas the two cDNAs encoding for NLS (from SV40 Large T-anti-gen) and a translational stop codon Vectashield (Vector Labs, Burlingame, CA). Fluorescence (PKKKRKV*) were obtained by in vitro annealing of two intensity of MAP2 and βIII-tubunin in cortical neurons synthetic oligodeoxy nucleotide strains (Euro- was quantified in terms of pixel intensity value by using the finsgenomics, Ebersberg, Germany). Both XIP and NLS NIH image software. Briefly, digital images were taken cDNAs were sequentially ligated in frame in the multi- with 63x objective and identical laser power settings and cloning site of the EYFP-C3 plasmid. Two or more neutral exposure times were applied to all the photographs from aminoacids separated EYFP, XIP, and NLS domains. each experimental set. The quantification of colocalization Successful construction of the plasmid encoding for YFP- between MAP2 and βIII-tubun was assessed by using the XIP-NLS was verified by sequencing both strands ‘co-localization highlighter’ plug-in for ImageJ Software (Microgem, Naples, Italy). This plasmid named XIP-NLS (NIH, Bethesda, MA, USA). Before colocalization analysis, was transiently transfected in PC12 cells with Lipofecta- threshold settings for each image were determined, and mine 2000 (Invitrogen, Carlsbad, CA), according to the quantification was achieved by counting the number manufacturer’s protocol. The efficiency of transfection colocalized points per microscope field. Results were was calculated by counting the number of EYFP-positive expressed as a percentage of colocalization. All images nuclei compared to total Hoecth-positive nuclei. were observed using a Zeiss LSM510 META/laser-scan- ning confocal microscope. Immunoprecipitation analyses 2+ 2+ Cortical neurons were transfected with an excess of [Ca ] and nuclear Ca measurements 2+ XIP-NLS (400 ng/ml). After transfection, nuclei from [Ca ] was measured by single-cell computer-assisted each population were isolated as previously described and video-imaging in isolated nuclei from NGF-differentiated 38,39 homogenized in lysis buffer containing: 50 mM HEPES, PC12 loaded with Fura-2/AM (Calbiochem) . Under 100 mM NaCl, 1.5 mM MgCl2, 1 mM PMSF, 0.2% Non- these conditions, Fura-2 is trapped within the NE lumen. 2+ idet P-40, 5 μg/ml aprotinin, 10 μg/ml leupeptin and 2 μg/ However, nucleoplasmic Ca was measured in whole ml pepstatin. The nuclear lysates were cleared by cen- cells by loading NGF-differentiated PC12 with Fura- trifugation at 12,000 rpm for 10 min. At the end, 1 mg of 1dextran (10 μM/1 h at 37 °C in the presence of 0.05% nuclear lysate was immunoprecipitated with the mouse Pluronic), a conjugated form of the fluorescent probe that monoclonal anti-EGFP antibody (1:200) or the rabbit is trapped in the nucleoplasmic space. 2+ polyclonal anti-NCX1 antibody (1:200). Then, the To perform functional analysis of Ca level in the immunoprecipitates were resolved by Western Blotting lumen of NE on isolated nuclei, we modified the methods 11 18 and immunoblot analysis was performed using anti- used by Gerasimenko et al. and Xie et al. . Fresh NCX1 or anti-GFP antibodies, respectively. nuclear preparations from NGF-differentiated PC12 cells were placed on glass coverslips previously treated with 30 Immunocytochemistry µg/ml poly-L-lysine. After 1 h, the preparation was loaded Isoform-specific antibodies for NCX1 and Lamin B1/B2, with 10 µM Fura-2/AM at 4 °C in a standard solution alone or in combination, were used for immunocy- containing 0.04% Pluronic. At the end of the Fura-2/AM tochemistry. The rabbit polyclonal anti-NCX1 antibody loading period, isolated nuclei were allowed to settle for was purchased from Swant (Bellinzona, Switzerland). Cells 15 min in a TM solution composed of: 20 mM Tris-HCl, or isolated nuclei were rinsed twice in cold 0.01 M saline 1 mM MgCl2, 150 mM NaCl with a pH of 7.4 and then phosphate buffer (PBS) at pH 7.4 and fixed in 4% (w/v) subjected to an uptake buffer named Na -free consisting paraformaldehyde (Sigma, Milan, Italy) for 20 min at RT. of: 250 mM sucrose, 50 mM Tris–HCl, 4 mM K2HPO4, 4 Following three washes in PBS, cells were blocked with 3% mM MgCl2, 2 mM EGTA, 2 mM EDTA, and 147 mM + 18 (w/v) BSA and 0.05% Triton-X (Biorad, Milan, Italy) for 1 NMDG together with 2.5 mM CaCl2, pH 7.4 .To 2+ h at RT. The coverslips were then incubated overnight measure Ca levels, both in whole cells and isolated with the primary antibody anti-NCX1 (1:1000 dilution), or nuclei, coverslips were placed into a perfusion chamber anti-Lamin B1/B2 (1:50, Abcam, UK). After three washes (Medical System, Co. Greenvale, NY, USA) mounted onto in PBS, the coverslips were incubated in the dark for 1 h at a Zeiss Axiovert 200 microscope (Carl Zeiss, Germany) RT with two secondary antibodies: Cy2 anti-mouse IgG equipped with a FLUAR 40X oil objective lens. The Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 13 of 14 experiments were carried out with a digital imaging sys- 7.3). TTX-sensitive Na currents were recorded as pre- tem composed of MicroMax 512BFT cooled CCD camera viously described by Secondo et al . The mean resting (Princeton Instruments, Trenton, NJ, USA), LAMBDA membrane potential (holding potential at 0 current) was 10-2 filter wheel (Sutter Instruments, Novato, CA, USA), measured using the above-mentioned bath solution in and Meta-Morph/MetaFluor Imaging System software absence of TTX and TEA. Capacitive currents were esti- (Universal Imaging, West Chester, PA, USA). After mated from the decay of the capacitive transient induced loading, nuclei were alternatively illuminated at wave- by 5 mV depolarizing pulses from a holding potential of lengths of 340 and 380 nm by a Xenon lamp. The emitted −80 mV and acquired at a sampling rate of 50 kHz. The light was passed through a 512 nm barrier filter. Fura-2 capacitance of the membrane was calculated according to fluorescence intensity was measured every 3 s and the following equation: Cm- = τc •Io/ΔEm (1−I∞/Io), reported as F340/F380 ratios. where Cm is membrane capacitance, τc is the time con- For the experiments of Figs. 6 and 7, NGF-differentiated stant of the membrane capacitance, Io is the maximum PC12 cells and cortical neurons at 7 DIV were illuminated capacitance current value, ΔEm is the amplitude of the at wavelengths of 480 nm to identify EYFP-positive cells voltage step, and I∞ is the amplitude of the steady state and then at 340 and 380 nm to measure, in the EYFP- current. 2+ positive cells, Fura-dextran-detected nucleoplasmic Ca 2+ levels. Ca levels have been reported as 340/380 ratio. Statistical analysis Data are expressed as mean ± S.E.M. Statistical Electrophysiological recording of NCX and voltage-gated comparisons between controls and treated experimental sodium channel activity by patch clamp groups were performed using the one-way ANOVA, The membrane voltage and current were acquired using followed by Newman Keul’s test. p < 0.05 was considered pClamp10 software. statistically significant. Plasma membrane NCX currents (I )inNGF- NCX differentiated PC12 cells were recorded by the patch- Acknowledgements clamp technique in whole-cell configuration using the This work was supported by PON01_01602 and PON03PE_00146_1 by MIUR to commercially available amplifier Axopatch200B and Digi- L.A., Progetto Giovani Ricercatori GR-2010-2318138 from Ministero della Salute and PRIN 2015 from MIUR to A.S., Futuro in Ricerca 2013 (RBFR13M6FN) from data1322A interface (Molecular Devices), as previously 28,40 MIUR to P.M., and Fondazione Italiana Sclerosi Multipla FISM 2015/R/6 to F.B. described .I were recorded starting from a holding NCX We thank Dr. Paola Merolla for the editorial revision. potential of −60 mV up to a short-step depolarization at +60 mV (60 ms). A descending voltage ramp from +60 Author contributions AS designed the study, performed video-imaging, interpreted results and mV to −120 mV was applied. I recorded in the des- NCX wrote the first draft of the manuscript. TP performed Western blot cending portion of the ramp (from +60 mV to −120 mV) experiments, NE isolation and NE membranes separation. FB assisted with were used to plot the current–voltage (I–V) relation curve. immunohistochemistry and interpreted the results. AE and VT performed biochemical experiments while NG performed qRT-PCR. PM designed and The I magnitude was measured at the end of +60 mV NCX generated XIP-NLS expressing vector; RC and AP performed (reverse mode) and at the end of −120 mV (forward electrophysiological measurements. LA and GF assisted with designing 2+ mode), respectively. The Ni -insensitive component was experiments and, interpreting results, edited the manuscript. All authors read 21,41 and approved the final manuscript. All experiments were performed at subtracted from total currents to isolate I .Neu- NCX “Federico II” University of Naples, Italy. ronal cells were perfused with external Ringer’s solution containing the following (in mM): 126 NaCl, 1.2 NaHPO4, Competing interests 2.4 KCl, 2.4 CaCl2, 1.2 MgCl2, 10 glucose, and 18 The authors declare that they have no competing interests. NaHCO3, pH 7.4. Twenty millimolar tetraethylammonium Publisher's note (TEA), 50 nM TTX, and 10 μM nimodipine were added to Springer Nature remains neutral with regard to jurisdictional claims in Ringer’s solution to abolish potassium, sodium, and cal- published maps and institutional affiliations. cium currents. The dialyzing pipette solution contained Received: 23 October 2017 Accepted: 25 November 2017 the following (in mM): 100 K-gluconate, 10 TEA, 20 NaCl, Published online: 30 April 2018 1 Mg-ATP, 0.1 CaCl2, 2 MgCl2, 0.75 EGTA, and 10 HEPES, adjusted to pH 7.2 with CsOH. For tetrodotoxin (TTX)-sensitive Na channel record- References ings, NGF –differentiated PC12 cells and untreated PC12 1. Bootman, M. D., Fearnley, C., Smyrnias, I., MacDonald, F. & Roderick, cells were perfused with an extracellular Ringer’s solution H. L. An update on nuclear calcium signalling. J. Cell. Sci. 122,2337–2350 (2009). containing 20 mM tetraethylammonium (TEA) and 5 μM 2. Rodrigues, M. A. et al. Nucleoplasmic calcium is required for cell proliferation. J. nimodipine. The pipettes were filled with 110 mM CsCl, Biol. Chem. 282,17061–17068 (2007). 10 mM TEA, 2 mM MgCl2, 10 mM EGTA, 8 mM glucose, 2 mM Mg-ATP, 0.25 mM cAMP, and 10 mM HEPES (pH Official journal of the Cell Death Differentiation Association Secondo et al. Cell Death Discovery (2018) 4:12 Page 14 of 14 + 2+ 3. Pusl,T.etal. Epidermalgrowth factor-mediated activation of the ETS domain 23. Secondo,A.etal. Involvement ofthe Na /Ca exchanger isoform 1 (NCX1) in transcription factor Elk-1requires nuclear calcium. J. Biol. Chem. 277, neuronal growth factor (NGF)-induced neuronal differentiation through Ca 27517–27527 (2002). -dependent Akt phosphorylation. J. Biol. Chem. 290,1319–1331 (2015). 2+ 4. Ledo, F., Kremer, L., Mellstrom, B. & Naranjo, J. R. Ca -dependent 24. Zou, H. et al. MicroRNA-29A/PTEN pathway modulates neurite outgrowth in block of CREB-CBP transcription by repressor DREAM. Embo. J. 21,4583–4592 PC12 cells. Neurosci 29,289–300 (2015). (2002). 25. van Diepen, M.T.&Eickholt,B. J. FunctionofPTEN during theformation and 5. Delivoria-Papadopoulos, M., Akhter, W. & Mishra, O. P. Hypoxia-induced maintenance of neuronal circuits in the brain. Dev. Neurosci. 30,59–64 (2008). 2+ Ca -influx in cerebral cortical neuronal nuclei of newborn piglets. Neurosci. 26. Zhou, J. & Parada, L. F. PTEN signaling in autism spectrum disorders. Curr. Opin. Lett. 342,119–123 (2003). Neurobiol. 22,873–879 (2012). +/ + 6. Delivoria-Papadopoulos, M.,Ashraf,Q.M.& Mishra,O.P.Effect of hypoxiaon 27. Galva, C., Artigas, P. & Gatto, C. Nuclear Na K -ATPase plays an active 2+ expression of apoptotic proteins in nuclear, mitochondrial and cytosolic role in nucleoplasmic Ca homeostasis. J. Cell. Sci. 125,6137–6147 fractions of the cerebral cortex of newborn piglets: the role of nuclear Ca (2012). -influx. Neurochem. Res. 33,1196–1204 (2008). 28. Secondo, A. et al. Molecular pharmacology of the amiloride analog 3-amino-6- 7. Leite, M. F. et al. Nuclear and cytosolic calcium are regulated independently. chloro-5-[(4-chloro-benzyl)amino]-n-[[(2,4-dimethylbenzyl)-amino]imino- + 2+ Proc. Natl. Acad. Sci. 100, 2975–2980 (2003). methyl]-pyrazinecarboxamide (CB-DMB)asapaninhibitor of theNa -Ca 8. Nicotera, P., Orrenius, S., Nilsson, T. & Berggren, P. O. An inositol 1,4,5-tri- exchanger isoforms NCX1, NCX2, and NCX3 in stably transfected cells. J. 2+ sphosphate-sensitive Ca pool in liver nuclei. Proc. Natl. Acad. Sci. 87, Pharm. Exp. Ther. 331,212–221 (2009). 6858–6862 (1990). 29. Savard, M. et al. Expression of endogenous nuclear bradykinin B2 receptors 9. Stehno-Bittel, L.,Luckhoff, A. &Clapham,D.E.Calcium releasefromthe mediating signaling in immediate early gene activation. J. Cell. Physiol. 216, nucleus by InsP3 receptor channels. Neuron 14,163–167 (1995). 234–244 (2008). 10. Bezin,S., Fossie,P.&Cancela, J. M. Nucleoplasmic reticulum is not essential in 30. Wu, G.,Xie, X., Lu,Z. H.&Ledeen,R.W. Sodium-calciumexchanger complexed nuclear calcium signalling mediated by cyclic ADPribose in primary neurons. with GM1 ganglioside in nuclear membrane transfers calcium from nucleo- Pflug. Arch. 456,581–586 (2008). plasm to endoplasmic reticulum. Proc. Natl. Acad. Sci. 106,10829–10834 11. Gerasimenko,O.V., Gerasimenko, J. V.,Tepikin,A.V. & Petersen,O.H.ATP- (2009). + + dependent accumulation and inositol trisphosphate- or cyclic ADP-ribose- 31. Garner, M. H. Na,K-ATPase in the nuclear envelope regulates Na: K gradients 2+ mediated release of Ca from the nuclear envelope. Cell 80,439–444 (1995). in hepatocyte nuclei. J. Membr. Biol. 187,97–115 (2002). 12. Campden, R.,Audet, N.&Hébert,T.E.Nuclear Gprotein signaling: newtricks 32. Freeman,D.J.etal. PTEN tumorsuppressor regulates p53 protein levels and for old dogs. J. Cardiovasc Pharm. 65,110–122 (2015). activity through phosphatase-dependent and -independent mechanisms. 13. Mazzanti, M., Bustamante, J. O. & Oberleithner, H. Electrical dimension of the Cancer Cell. 3,117–130 (2003). nuclear envelope. Physiol. Rev. 81,1–19 (2001). 33. Musatov, S. et al. Inhibition of neuronal phenotype by PTEN in PC12 cells. Proc. 14. Tabares, L., Mazzanti, M. & Clapham, D. E. Chloride channels in the nuclear Natl. Acad. Sci. 101,3627–3631 (2004). membrane. J. Membr. Biol. 123,49–54 (1991). 34. Gerasimenko, J., Maruyama, Y., Tepikin, A., Petersen, O. H. & Gerasimenko, O. 15. Bkaily, G. et al. The use of confocal microscopy in the investigation of cell Calcium signalling in and around the nuclear envelope. Biochem. Soc. Trans. structure and function in the heart, vascular endothelium and smooth muscle 31,76–78 (2003). cells. Mol. Cell. Biochem. 172,171–194 (1997). 35. Sherkhane,P.&Kapfhammer,J.P.Chronic pharmacologicalblockadeofthe 16. Bkaily, G. et al. G-protein-coupled receptors, channels, and Na + -H + Na + /Ca2 + exchanger modulates the growth and development of the Pur- exchanger in nuclear membranes of heart, hepatic, vascular endothelial, and kinje cell dendritic arbor in mouse cerebellar slice cultures. Eur. J. Neurosci. 46, smooth muscle cells. Can. J. Physiol. Pharm. 84,431–441 (2006). 2108–2120 (2017). 2+ 17. Abrenica, B. & Gilchrist, J. S. Nucleoplasmic Ca loading is regulated by 36. Gilchrist,J. S.& Pierce, G.N.Identification and purification of a calcium- 2+ mobilization of perinuclear Ca . Cell. Calcium 28,127–136 (2000). binding protein in hepatic nuclear membranes. J. Biol. Chem. 268, 18. Xie,X., Wu,G., Lu,Z.H.& Ledeen, R.W.Potentiation ofasodium-calcium 4291–4429 (1993). exchanger in the nuclear envelope by nuclear GM1 ganglioside. J. Neurochem. 37. Bradford, M. M. A rapid and sensitive method for the quantification of 81,1185–1195 (2002). microgram quantities of protein utilizing the principle of protein-dye binding. 19. Ledeen, R. & Wu, G. GM1 in the nuclear envelope regulates nuclear calcium Ann. Biochem 72,248–254 (1976). 2+ through association with a nuclear sodium-calcium exchanger. J. Neurochem. 38. Grynkiewicz, G., Poenie, M. & Tsien, R. Y. A new generation of Ca indicators 103,126–134 (2007). with greatly improved fluorescence properties. J. Biol. Chem. 260,3440–3450 + 2+ 20. Annunziato, L.,Pignataro,G.& Di Renzo, G. F. Pharmacology of brainNa –Ca (1985). exchanger: from molecular biology to therapeutic perspectives. Pharm. Rev. 39. Secondo, A. et al. BHK cells transfected with NCX3 are more resistant to 56,633–654 (2004). hypoxia followed by reoxygenation than those transfected with NCX1 and + 2+ 21. Molinaro, P. et al. Na -Ca exchanger (NCX3) knock-out mice display an NCX2: Possible relationship with mitochondrial membrane potential. Cell. impairment in hippocampal long-term potentiation and spatial learning and Calcium 42,521–535 (2007). memory. J. Neurosci. 31, 7312–7321 (2011). 40. Pannaccione, A. et al. A new concept: Aβ1-42 generates a hyperfunctional 22. Molinaro, P. et al. Neuronal NCX1 overexpression induces stroke resistance proteolytic NCX3 fragment that delays caspase-12 activation and neuronal while knockout induces vulnerability via Akt. J. Cereb. Blood. Flow. Metab. 36, death. J. Neurosci. 32, 10609–10617 (2012). + 2+ 1790–1803 (2016). 41. Molinaro, P. et al. Neurounina-1, a novel compound that increases Na /Ca exchanger activity, effectively protects against stroke damage. Mol. Pharm. 83, 142–156 (2013). Official journal of the Cell Death Differentiation Association

Journal

Cell Death DiscoverySpringer Journals

Published: Apr 30, 2018

There are no references for this article.