Departamento de Bioquímica y Biología Molecular, Facultad de Veterinaria, Universidad Complutense de Madrid, Madrid, Spain
Address correspondence and reprint requests to Dr Javier Gualix, Departamento de Bioquímica y Biología Molecular IV, Facultad de Veterinaria, UCM, Av. Puerta de Hierro s/n, 28040-Madrid, Spain. E-mail: firstname.lastname@example.org
Neuro-2a (N2a) neuroblastoma cells display an ectoenzymatic hydrolytic activity capable of degrading diadenosine polyphosphates. The ApnA-cleaving activity has been analysed with the use of the fluorogenic compound BODIPY® FL guanosine 5′-O-(3-thiotriphosphate) thioester. Hydrolysis of this dinucleotide analogue showed a hyperbolic kinetic with a Km value of 4.9 ± 1.3 μM. Diadenosine pentaphosphate, diadenosine tetraphosphate, diadenosine triphosphate, and the nucleoside monophosphate AMP behaved as an inhibitor of BODIPY® FL guanosine 5′-O-(3-thiotriphosphate) thioester extracellular degradation. Ectoenzymatic activity shared the typical characteristics of the ectonucleotide pyrophosphatase/phosphodiesterase family, as hydrolysis reached maximal activity at alkaline pH and was dependent on the presence of divalent cations, being strongly inhibited by EDTA and activated by Zn2+ ions. Both NPP1 and NPP3 isozymes are expressed in N2a cells, their expression levels substantially changing when cells differentiate into a neuronal-like phenotype. In this sense, it is relevant to point the expression pattern of the NPP3 protein, whose levels were drastically reduced in the differentiated cells, being almost completely absent after 24 h of differentiation. Enzymatic activity assays carried out with differentiated N2a cells showed that NPP1 is the main isozyme involved in the extracellular degradation of dinucleotides in these cells, this enzyme reducing its activity and changing its subcellular location following neuronal differentiation.
We described the presence of an ectoenzymatic activity able to hydrolyse diadenosine polyphosphates (ApnA) in N2a cells. This activity displays biochemical features that are typical of the ectonucleotide pyrophosphatase/phosphodiesterase (E-NPP) family members, as demonstrated by the use of the fluorogenic compound BODIPY-FL-GTPγS. Both NPP1 and NPP3 ectoenzymes are expressed in N2a cells, their levels dramatically changing when cells differentiate into a neuronal-like phenotype. Activity assays in differentiated cells showed that the ApnA-hydrolytic activity largely depends on the NPP1 isozyme.
Diadenosine polyphosphates comprise a group of compounds formed by two adenosine nucleosides joined by a phosphate chain of variable length. They are commonly abbreviated as ApnA, where n represents the number of phosphate residues in the polyphosphate chain. Members of this family of dinucleotides have been identified in the releasable content of storage granules in specialized neurosecretory cells and platelets and are thought to function as extracellular signalling molecules in the vascular and nervous systems (Flores et al. 1999; Miras-Portugal et al. 1999). In the nervous system, ApnA (n = 2–6) and other dinucleoside polyphosphates have been found to be costored with mononucleotides, mainly ATP and ADP, in the catecholaminergic and cholinergic vesicles from adrenomedullary chromaffin cells and neurons (Rodriguez del Castillo et al. 1988; Pintor et al. 1992a,b,c). All these secretory systems respond to depolarizing agents or secretagogues by releasing their vesicular content to the extracellular medium (Pintor et al. 1991, 1992a,c). In this regard, push-pull cannula experiments performed in living rats showed that after amphetamine stimulation, rat neostriatum releases diadenosine tetraphosphate, Ap4A, and diadenosine pentaphosphate, Ap5A, which can be detected in the perfusion samples at concentrations in the nanomolar range (Pintor et al. 1995). More recently, nanomolar concentrations of diadenosine polyphosphates (including diadenosine triphosphate, Ap3A, together with Ap4A and Ap5A) have been also measured in microdialysis samples from the cerebellum of conscious rats under basal conditions, i.e. in the absence of any exogenously added stimulating substance (Gualix et al. 2014). Exocytotically released ApnA interact with specific dinucleotide receptors but the ionotropic P2X, metabotropic P2Y and P1 adenosine receptors may also function as ApnA targets (Pintor et al. 2000; Hoyle et al. 2001). Results obtained using very different experimental approaches support the idea that extracellular ApnA, acting on P1, P2 or their specific dinucleotide receptors, can effectively modulate neural functions (Pereira et al. 2000; Oaknin et al. 2001; Jimenez et al. 2002; Delicado et al. 2006). Moreover, neuroprotective effects against injuries induced by ischemia or 6-hydroxydopamine injection in rat brain have been described for Ap4A (Wang et al. 2003).
The inactivation of ApnA by ectoenzymes at the cell surface provides a necessary mechanism to regulate the receptor-mediated actions of these dinucleotides (Zimmermann et al. 2012). Ectoenzymatic hydrolysis of ApnA has been analysed in adrenomedullary chromaffin cells (Rodriguez-Pascual et al. 1992; Ramos et al. 1995) and synaptic membranes of the Torpedo electric organ (Mateo et al. 1997). Available data indicate that the membrane-bound enzymes catalyse the hydrolytic cleavage at the polyphosphate chain of the dinucleotides to produce the mononucleotidic moieties AMP + adenosine 5′(n − 1) phosphate. The subsequently generated nucleotides are further degraded, yielding adenosine as the final product, which may be recovered into the cells through transport systems. Most of these ApnA-hydrolysing enzymes display biochemical characteristics typical of members of the ectonucleotide pyrophosphatase/phosphodiesterase (E-NPP) family. This family contains seven members, which have been numbered NPP1–NPP7 according to their order of cloning. NPP1–NPP3 have been previously described as the only members of the family able to hydrolyse diadenosine polyphosphates. Each of the three enzymes cleaves Ap3A, Ap4A and Ap5A at comparable rates with Michaelis constants (Km) in the low micromolar range (Vollmayer et al. 2003). However, it has been recently demonstrated that NPP4 can also hydrolyse Ap3A and Ap4A, although the Km of NPP4 binding to these dinucleotides is much weaker than previously reported for NPP1–NPP3 (Albright et al. 2012, 2014). NPP6 and NPP7 are only known to hydrolyse phosphodiester bounds in lysophospholipids or other choline phosphodiesters, whereas substrates for NPP5 have not yet been identified. Except for NPP2, which only exists as a secreted protein, all E-NPPs are single-span transmembrane proteins (Stefan et al. 2005, 2006; Zimmermann et al. 2012). Recent data on the ApnA-hydrolytic activity in plasma and synaptic membranes isolated from rat brain point to NPP1 as the main ectoenzyme involved in the cleavage of ApnA by glial cells and neurons (Asensio et al. 2007). NPP1 activity is widely distributed through the rat forebrain, with the highest activity in hypothalamus (Oaknin et al. 2008). These results add new support for a signalling function of diadenosine polyphosphates in brain.
Neuro-2a (N2a) is a mouse neural crest-derived tumour cell line that has been extensively used to study neuronal differentiation, axonal growth, and signalling pathways. The presence of functional P2X7 nucleotide receptors that are involved in the regulation of neuronal differentiation and neurite formation has been described in these cells (Gomez-Villafuertes et al. 2009; Wu et al. 2009). These results together with those obtained in human neuroblastoma models, showing that P2X7 receptor stimulation resulted into an enhancement of cell proliferation (Raffaghello et al. 2006), support a role for the purinergic signalling in the mechanisms that allow a fine control of the balance between proliferation and differentiation in neuroblastoma cells. N2a cells also constitutively express P2Y2 receptors which are sensitive to dinucleoside tetraphosphates (such as Ap4A) and whose activation enhances α-secretase activity and the non-amyloidogenic processing of the amyloid precursor protein (Leon-Otegui et al. 2011), thus precluding the formation of toxic β-amyloid peptides.
In this work, we analysed the presence of ectoenzymatic activities that are capable of hydrolysing extracellular diadenosine polyphosphates in N2a cells. The ApnA-cleaving activity in these cells displays biochemical features that are typical of the members of the E-NPP family as has been demonstrated by the use of the fluorogenic dinucleotide analogue BODIPY® FL guanosine 5′-O-(3-thiotriphosphate) thioester (BODIPY-FL-GTPγS) (Invitrogen, San Francisco, CA, USA). Both NPP1 and NPP3 ectoenzymes are expressed in N2a cells and their expression levels dramatically change when cells differentiate into a neuronal-like phenotype. Activity assays carried out with differentiated cells showed that the dinucleotide hydrolytic activity largely depends on the NPP1 isozyme.
Materials and methods
Ap5A, Ap4A, Ap3A, ATP, ADP, AMP, dibutyryl-cAMP (DiBucAMP) and monoclonal mouse anti-α-tubulin and mouse anti-MAP-2 antibodies were purchased from Sigma-Aldrich (St Louis, MO, USA). Micro-Bradford assay, acrylamide electrophoresis reagent and molecular weight protein standards were from Bio-Rad (Munich, Germany). Acetonitrile, HPLC grade, was purchased from Scharlau (Barcelona, Spain). The commercial antibodies used in this study were raised against NPP1 (also known as PC-1) or NPP3 (also known as CD203c) of human origin but are also suitable for detection of the corresponding mouse proteins and were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Fluorescence immunodetection of NPP1 was carried out using a different antibody raised against a synthetic peptide corresponding to C-terminal amino acids 915-925 of human NPP1 and was purchased from Abcam (Cambridge, UK). All other reagents not specified were routinely supplied by Sigma-Aldrich.
N2a cells were cultured in Dulbecco's modified Eagle's medium (Sigma-Aldrich) supplemented with glutamax (Invitrogen, San Francisco, CA, USA), penicillin⁄streptomycin (Invitrogen) and 10% heat-inactivated foetal bovine serum (FBSi, Euro-Clone, Padova, Italy). Cells were grown at 37°C in a humidified atmosphere containing 5% CO2.
In differentiation studies, cells were incubated with a lower percentage of FBSi (0.5%) and treated with 1 mM DiBucAMP, a well-established inductor of N2a differentiation into a neuronal-like phenotype (Fischer et al. 1986).
HPLC studies on extracellular degradation of ApnA
N2a cells were plated at 750 000 cells/well in six-well plates. After 24 h, culture medium was removed and replaced by Locke's solution (composition in mM: NaCl, 140; KCl, 4.5; CaCl2, 2.5; KH2PO4, 1.2; MgSO4, 1.2; glucose, 5.5; HEPES, 10; pH 7.4). 10-μM Ap5A was added to the wells and cells were maintained at 37°C in a humidified atmosphere containing 5% CO2. To analyse Ap5A cleavage, samples of the incubation medium were taken at different times (t = 0, 30 min, 1 h, 2 h) and the amount of Ap5A was measured by HPLC detection. As a control, Ap5A was incubated in parallel but in the absence of the cells.
The chromatographic system consisted of a Waters (Milford, MA, USA) 1515 isocratic HPLC pump, a 2487 dual absorbance detector and a Reodyne injector, all managed by the Breeze software from Waters. Analysis was performed under ion-pair chromatography conditions by equilibrating the chromatographic system with the following mobile phase: 10 mM KH2PO4, 2 mM tetrabutylammonium hydrogen sulphate and 20% acetonitrile, pH 7.5. The column was a Nova-Pak C-18 (15-cm length, 0.4-cm diameter) from Waters. Detection was monitored at 260-nm wavelength. Ap5A peaks were identified by their retention times. Integrated peak areas of the samples were normalized with respect to those obtained at t = 0, which were set as a 100%.
Fluorimetric enzyme assays
For the ectonucleotidase activity assays, cells were plated at 50 000 cells/well in black with clear bottom 96-well plates and allowed to settle for 2 h. After that, culture medium was removed and replaced by 50 μL of Locke′s solution. Reaction was started by the addition of the fluorogenic substrate BODIPY-FL-GTPγS dissolved in 50 μL of Locke′s solution for a final reaction volume of 100 μL. Increase in fluorescence emission owing to BODIPY-FL-GTPγS cleavage was monitored using a FLUOstar Optima plate reader (BMG Labtech, Ortenberg, Germany). Appropriate wavelengths were selected with the use of 485-nm excitation and 520-nm emission filters. Basal fluorescence was recorded by assaying the fluorescent substrate in the presence of 50 mM EDTA (a divalent cation chelator that blocks ectoenzymatic dinucleotide-hydrolysing activity in N2a cells) and was subtracted in all the cases. To establish equivalence between fluorescence units and substrate moles, fluorescence was read again after total substrate hydrolysis, which was ensured by the addition of Crotalus adamanteus phosphodiesterase (0.02 U, Sigma-Aldrich). Enzymatic activity values were calculated by determining the amount of substrate that is hydrolysed after the initial 30 seconds of reaction in which a linear time-dependent fluorescence increase was obtained and substrate consumption was below 10% at any of the substrate concentrations assayed. For pH and cation dependence studies, reaction mixture was accordingly modified.
Reverse transcription PCR (RT-PCR) and quantitative real-time PCR (Q-PCR)
Total RNA from N2a cells was extracted using RNeasy® plus mini kit (Qiagen, Hilden, Germany), following the manufacturer's instructions. After digestion with TURBO DNase (Ambion, Austin, TX, USA), total RNA was quantified and reverse transcribed using M-MLV reverse transcriptase, 6 μg of random primers and 350 μM dNTPs (Invitrogen, San Francisco, CA, USA). PCR reactions were carried out using AmpliTools Master Mix (Biotools), 5 μL of the RT product and specific oligonucleotide primers for mouse ectonucleotide pyrophosphatases/phosphodiesterases NPP1 and NPP3 (Sigma-Aldrich). For NPP1 (NM_008813.3), the primers used were: forward 5′-CGGACGCTATGATTCCTTAGA-3′ and reverse 5′-AGCACAATGAAGAAGTGAGTCG-3′. For NPP3 (NM_134005.2), the primers used were: forward 5′-GATGCACAGGACGAGGAGAC-3′ and reverse 5′-TGCACGTCCATATTTGAGTTG-3′. Amplicon lengths were 93 bp and 75 bp for NPP1 and NPP3 respectively. PCR reactions involved an initial denaturation step at 94°C for 5 min, followed by 40 cycles of amplification (94°C for 30 s; 60°C for 30 s and 72°C for 30s) conducted with a 2720 Thermal Cycler (Applied Biosystems). Control reactions were carried out in the absence of RT product (template) to avoid cross contamination. Amplified PCR products were electrophoresed on a 2.5% agarose gel and visualized by SYBR® Safe DNA gel stain (Invitrogen). GeneRuler 1 kb Plus DNA Ladder (Thermo Scientific, Rockford, IL, USA) was used as DNA size marker.
Q-PCR reactions were performed using the previous gene-specific primers in combination with specific Taqman 3′-minor groove binder (MGB)-DNA probes (Roche, Basel, Switzerland). The probes designed were FAM-5′-GCCAGGAA-3′-MGB and FAM-5′-CTGCTGGG-3′-MGB for NPP1 and NPP3 respectively. Fast thermal cycling was performed using a StepOnePlus® Real-Time System (Applied Biosystems, Foster City, CA, USA) as follows: denaturation, one cycle of 95°C for 20 s, followed by 40 cycles each of 95°C for 1 s and 60°C for 20 s. The results were normalized as indicated by parallel amplification of glyceraldehyde-3-phosphate dehydrogenase housekeeping gene. For glyceraldehyde-3-phosphate dehydrogenase, commercial primers and TaqMan MGB probe were supplied by Applied Biosystems.
Western blot detection of E-NPP isozymes
N2a cells were lysed and homogenized for 1 h at 4°C in lysis buffer containing 50 mm Tris⁄HCl, 150 mm NaCl, 1% Nonidet P-40 and Complete® Protease Inhibitor Cocktail Tablets (Roche), pH 7.4. Protein content in the cell extracts was determined by the Bradford method and then samples were mixed with loading buffer 5X (50% glycerol, 250 mM Tris, 10% sodium dodecyl sulphate, 0.5% bromophenol blue, 7.7 % dithiothreitol, pH = 6.8) and heated at 99°C for 5 min. Aliquots of 30-μg protein were subjected to sodium dodecyl sulphate gel electrophoresis using 8% acrylamide gels and transferred to nitrocellulose membranes (Amersham GE Healthcare, Madrid, Spain). Membranes were blocked for 1 h at 25°C with 5% non-fat dried milk or 10% donkey serum in phosphate-buffered saline containing 0.1% Tween 20 [phosphate-buffered saline (PBS)-Tween] and incubated overnight at 4°C with the following antisera at the dilutions specified in parentheses: mouse anti-NPP1 (1 : 100), goat anti-NPP3 (1 : 200) and mouse anti-α-tubulin (1 : 10 000). Blots were then washed in PBS-Tween and incubated for 1 h at 25°C with goat anti-mouse or rabbit anti-goat IgGs coupled to horseradish peroxidase (Dako Cytomation, Glostrup, Denmark), both used at 1 : 1000 dilution. Protein bands were detected by enhanced chemiluminescence detection (Amersham GE Healthcare). Chemiluminescence images were acquired using the ImageQuant LAS 500 imager (GE Healthcare Bio-Sciences, Uppsala, Sweden) and quantified by means of the ImageQuant TL 8.1 software (GE Healthcare Bio-Sciences).
N2a cells cultured on coverslips placed in 35-mm dishes (350 000 cells/well) were washed with PBS and fixed with 4% paraformaldehyde for 15 min. After washing with PBS, cells were permeabilized with 0.1% Triton X-100 and blocked with 5% donkey serum and 3% bovine serum albumin in PBS for 1 h at 25°C. After washing with 3% bovine serum albumin in PBS, the cells were incubated for 1 h with primary antibodies: goat anti-NPP1 (1 : 250), goat anti-NPP3 (1 : 50) and mouse anti-MAP-2 (1 : 100). Positive immunostaining was revealed using Alexa Fluor 647 donkey anti-goat and Alexa Fluor 488 goat anti-mouse IgGs (Invitrogen). Nuclei were counter-stained with 4′,6-diamidino-2-phenylindole (Invitrogen). Coverslips were mounted on glass slides using ProLong® Gold Antifade Reagent (Life Technologies, Carlsbad, CA, USA). Images were acquired using a Leica CTR 6500 confocal microscope (Leica Microsystems, Wetzlar, Germany) with an ACS APO 40× immersion oil objective (NA 1.15).
Data were analysed using one-way anova followed by Dunnett's or Bonferroni's post hoc comparisons test (Graph Pad Prism 5, Graph Pad Software Inc., San Diego, CA, USA). Data are expressed as the mean ± SEM. Differences were considered significant at p ≤ 0.05.
Presence of an ectoezymatic ApnA-cleaving activity in N2a cells
To analyse the presence of ectoenzymatic activities able to hydrolyse diadenosine polyphosphates, N2a cells were incubated with 10 μM Ap5A and aliquots of the incubation medium were taken at different times (t = 0, 30 min, 1 h and 2 h) to follow the extracellular dinucleotide degradation by HPLC detection. As it is shown in Fig. 1a and b, extracellular levels of Ap5A were significantly reduced after 30 min and 1 h of incubation with the N2a cells and this dinucleotide could not be detected at measurable quantities in the extracellular medium after a 2-h incubation period. The presence of mononucleotide moieties that are generated as intermediate products of Ap5A hydrolysis could not be accurately analysed as these compounds do not accumulate in the extracellular medium. Rather, they are rapidly degraded by nucleotidases and converted to adenosine which appeared at the chromatogram front (Fig 1a). No significant hydrolysis of Ap5A was observed in the absence of cells (Fig. 1b).
BODIPY-FL-GTPγS: a fluorogenic substrate for the characterization of the ApnA-cleaving activity in N2a cells
ApnA-hydrolysing enzymatic activity was analysed with the use of the fluorogenic compound BODIPY-FL-GTPγS, formed by the BODIPY FL fluorophore linked through the γ-thiol of guanosine 5′-O-(3-thiotriphosphate) (GTPγS). BODIPY-FL-GTPγS cleavage is accompanied by a 5-fold increase in fluorescence emission because BODIPY fluorescence is quenched by stacking with the guanosine base in the intact molecule (Draganescu et al. 2000). Time-dependent increase in the fluorescence emission because of BODIPY-FL-GTPγS cleavage can be used to calculate the hydrolysis rate.
Extracellular BODIPY-FL-GTPγS was degraded by intact N2a cells. Cleavage of the fluorogenic dinucleotide analogue was concentration dependent, showing a typical hyperbolic kinetic with a Km value of 4.9 ± 1.3 μM (Fig 2a). Similar Km values in the low micromolar range have been observed for the extracellular hydrolysis of dinucleotides by chromaffin cells and synaptic membranes from Torpedo electric organ or rat brain (Rodriguez-Pascual et al. 1992; Ramos et al. 1995; Mateo et al. 1997; Asensio et al. 2007). Moreover, heterologously expressed NPP1, NPP2 and NPP3 also hydrolyse the diadenosine polyphosphates with Km values in the low micromolar range (Vollmayer et al. 2003). The extracellular hydrolysis of BODIPY-FL-GTPγS was inhibited by Ap5A in a dose-dependent manner, this dinucleotide reducing the cleavage of the fluorogenic substrate by a 78% at 200 μM concentration (Fig. 2b). A similar behaviour was observed when Ap4A was assayed as an inhibitor of BODIPY-FL-GTPγS degradation, this compound being able to reduce the hydrolysis of the fluorogenic substrate by a 61% at 100-μM concentration (Fig. 2c). Ap3A, at 100 μM concentration, also acted as an inhibitor by reducing extracellular BODIPY-FL-GTPγS cleavage by a 65% (Fig. 2c). These results indicate that a common enzymatic activity showing broad substrate specificity is the responsible for both the extracellular hydrolysis of BODIPY-FL-GTPγS and diadenosine polyphosphates, ApnA then acting as competitors on BODIPY-FL-GTPγS degradation. Similar broad substrate specificity has been described for the ectoenzymatic activity able to hydrolyse diadenosine polyphosphates in chromaffin cells and Torpedo electric organ or rat brain membranes (Rodriguez-Pascual et al. 1992; Ramos et al. 1995; Mateo et al. 1997; Asensio et al. 2007). Also, heterologously expressed NPP1, NPP2 and NPP3 have been shown to be able to hydrolyse several different ApnA along with the diguanosine polyphosphate, diguanosine tetraphosphate (Gp4G) (Vollmayer et al. 2003). Thus, BODIPY-FL-GTPγS can be used as a useful substrate with which to analyse and characterize the ectoenzymatic activity responsible for the extracellular degradation of dinucleotides in N2a cells. A similar strategy, using etheno-derivatives of ApnA as fluorogenic substrates, has been previously used to characterize the dinucleotide-hydrolysing activities in chromaffin cells and Torpedo electric organ and brain membranes (Ramos et al. 1995; Mateo et al. 1997; Asensio et al. 2007). Regarding mononucleotides, only AMP was effective by reducing the extracellular cleavage of BODIPY-FL-GTPγS. This nucleotide inhibited BODIPY-FL-GTPγS-hydrolysing activity by a 55% at 100-μM concentration (Fig. 2c). It is relevant to note that E-NPP enzymes asymmetrically cleave the polyphosphate chain of ApnA to produce the mononucleotidic moieties AMP + adenosine 5′(n − 1) phosphate as reaction products (Bollen et al. 2000; Vollmayer et al. 2003; Stefan et al. 2005; Zimmermann et al. 2012). AMP is not further hydrolysed by the E-NPP, as expected from the absence of phosphodiester or pyrophosphate bonds, but exerts competitive product inhibition on the ectoenzyme activity (Bollen et al. 2000; Vollmayer et al. 2003; Zimmermann et al. 2012). In this sense, strong inhibition by the reaction product AMP has been reported for the NPP1 activity present in brain membranes (Asensio et al. 2007).
Biochemical features of the ApnA-cleaving activity in N2a cells
The effect of divalent cations and pH on the ectoenzymatic ApnA-hydrolysing activity was analysed by the use of fluorogenic substrate BODIPY-FL-GTPγS. Dinucleotide-cleaving activity in the N2a cells was dependent on the presence of divalent cations in the extracellular medium, as demonstrated by the effect of EDTA which reduced the hydrolytic activity in a dose-dependent manner, BODIPY-FL-GTPγS cleavage being almost completely abolished in the presence of 50 mM of the cation chelator (Fig 3a). When the enzymatic activity was assayed in a Mg2+-free Locke's solution (containing only 2.5 mM Ca2+ as divalent cation), activity was 63.4 ± 6.5 % of that in the standard medium (containing both 1.2 mM Mg2+ and 2.5 mM Ca2+). Similarly, hydrolysing activity was reduced to a 33.9 ± 1.6 % in a virtually Ca2+-free medium (obtained by the addition of 6 mM of the selective Ca2+ chelator EGTA to the Locke's solution). This indicated an additive effect of both Mg2+ and Ca2+ stimulating ectoenzymatic dinucleotide-cleaving activity. When assays were carried out in the absence of these divalent cations (in a Mg2+-free Locke's solution supplemented with 6 mM EGTA), BODIPY-FL-GTPγS-cleaving activity was reduced to a 19.5 ± 4.0 %, this residual activity being increased to a 36.2 ± 2.5 % when 200 μM Zn2+ was added to the medium (Fig. 3b), thus demonstrating the activatory effect exerted by Zn2+ ions.
Regarding the influence of pH, dinucleotide hydrolase activity was stimulated in alkaline conditions. In the pH range 6.5–8.5, maximal activity was observed at pH 8.5, as it is shown in Fig. 3c. There were no changes in the fluorescence emission of BODIPY-FL-GTPγS within this interval.
N2a cells express NPP1 and NPP3 isozymes
General characteristics of ectoenzymatic dinucleotide-cleaving activity in N2a cells, such as alkaline optimum pH, activation by Ca2+, Mg2+ and Zn2+ ions and inhibition by EDTA fit well with those of the E-NPP family members that hydrolyse ApnA (NPP1-NPP4). NPP4 has shown a lower affinity for the diadenosine polyphosphates than those reported here for the dinucleotide-hydrolysing activity in N2a cells. NPP4 present on the surface of vascular endothelium hydrolyses Ap3A and Ap4A with Km values of 843 μM and 210 μM respectively (Albright et al. 2012), whereas a soluble form of the enzyme, obtained via C-terminal truncation to eliminate the cytoplasmic and transmembrane domains of human NPP4, cleaved Ap3A with a Km value of 685 μM (Albright et al. 2014). However, the affinity values observed here for the dinucleotide hydrolysis by N2a cells, which are in the low μM range, fit well with those described for NPP1–NPP3 (Bollen et al. 2000; Vollmayer et al. 2003; Zimmermann et al. 2012). From these, NPP2 only exists as a secreted protein (Stefan et al. 2005, 2006; Zimmermann et al. 2012). As culture medium was replaced by Locke's solution immediately before the enzyme activity measurements, it seems unlikely that a secreted protein could be the responsible of the dinucleotide-cleaving activity observed in our assays. Thus, we analysed the expression of NPP1 and NPP3 to discriminate the isozyme involved in the extracellular degradation of dinucleotides by N2a cells.
RT-PCR analysis demonstrated that both NPP1 and NPP3·are simultaneously expressed in N2a cells (Fig. 4a). To characterize the relative amount of each NPP transcript, Q-PCR assays were performed. As it is shown in Fig. 4b comparable levels were obtained for both NPP1 and NPP3 mRNAs. Western blot analyses were performed on total protein extracts obtained from N2a cells using commercial isozyme-specific antibodies. As expected, the expression of NPP proteins correlated well with the results obtained by quantitative PCR. Thus, bands corresponding to NPP1 and NPP3 proteins were immunodetected (Fig. 4c). These proteins contain several consensus sequences for N-linked glycosylation. The occurrence of N-glycosylation in vivo affects the electrophoretic mobility of the proteins (Bollen et al. 2000; Zimmermann et al. 2012) that can vary depending on the cell type and species. The bands obtained in this study were consistent with the range of molecular sizes reported for NPP1 and NPP3 in the literature (Dong et al. 2005; Jankowski et al. 2011).
Changes in E-NPP expression in differentiated N2a cells
It is well known that purinergic signalling may regulate neuroblastoma cells growth (Raffaghello et al. 2006) and changes in the expression and function of nucleotide receptors occurs when N2a cells differentiate into neuronal-like cells (Gomez-Villafuertes et al. 2009; Wu et al. 2009). Ectonucleotidases play a pivotal role in the regulation of the extracellular levels of the nucleotidic effectors and hence variations in their activity can have a deep influence in the cell proliferation and differentiation processes (Stefan et al. 2005, 2006; Zimmermann et al. 2012). Here, we analyse the expression of the E-NPP proteins when N2a cells differentiate. Variations in the expression of these enzymes, both at the transcriptional and protein level, compared to non-differentiated cells, were measured.
Q-PCR studies revealed that the endogenous levels of the NPP1 transcript were very significantly reduced after 24 h of differentiation treatment, NPP1 mRNA levels being recovered at longer times, 48 h and 72 h (Fig. 5a). Curiously, western blot analysis showed that NPP1 protein levels only experimented a substantial decrease after 72 h and 96 h of differentiation (Fig. 5b and c). Immunofluorescence studies confirmed that NPP1 protein is present in non-differentiated cells (Fig. 5d–f), whereas their levels are significantly reduced after 96 h of differentiation treatment (Fig. 5g–i). It is relevant to note that the remaining NPP1 protein seems to change its intracellular distribution from a somatic to an axon-like location in the differentiated cells (Fig. 5g and i).
A similar pattern to that obtained for NPP1 was observed when the levels of the NPP3 transcript were analysed. NPP3 mRNA was substantially reduced in the differentiated cells after 24 h of treatment, and transcript levels were recovered at longer times (Fig. 6a). However, a completely different behaviour was observed when the amounts of protein were analysed by western blot. NPP3 protein was dramatically reduced in the differentiated N2a cells, being almost completely absent after 24 and 48 h of treatment and showing only a very slight recovery at longer times of treatment (Fig. 6b and c). Fluorescence immunodetection experiments confirmed that NPP3 protein is present in non-differentiated cells (Fig. 6d–f), whereas their levels are undetectable in neurite-bearing N2a cells that express the neuronal marker MAP-2 after 96 h of differentiation (Fig. 6g–i).
NPP1 is the main isozyme involved in the extracellular cleavage of ApnA by N2a cells
To determine the NPP isozyme that is involved in the extracellular degradation of dinucleotides in N2a cells, we analysed the changes in the ectoenzymatic hydrolytic activity in differentiated cells after 24 h or 96 h of treatment (in which one or both of the E-NPP proteins significantly reduced their expression when compared to untreated cells). As it is shown in Fig. 7, hydrolytic activity was only partially reduced in differentiated cells after 24 h of treatment, in which NPP3 protein was almost absent and NPP1 only partially reduced it expression. Moreover, in differentiated cells after 96 h of treatment, in which NPP3 still had a very low expression rate and expression of NPP1 was substantially reduced, ectoenzymatic activity was also significantly decreased (when compared to both untreated cells or differentiated cells after 24 h of treatment). These results seem to indicate that ectoenzymatic activity able to hydrolyse ApnA in N2a cells largely resides in the NPP1 isozyme, as changes in ectoenzymatic hydrolytic activity appeared to match changes in NPP1 protein expression. Reduction in NPP3 expression, on the contrary, seems to have little or no effect on ectoenzymatic dinucleotide-cleaving activity.
Diadenosine polyphosphates (ApnA) have emerged as a new family of extracellular signalling molecules that can interact with several membrane protein targets (Miras-Portugal et al. 1999; Pintor et al. 2000; Hoyle et al. 2001), being involved in neurotransmitter/neuromodulatory functions in the nervous system (Pereira et al. 2000; Oaknin et al. 2001; Jimenez et al. 2002; Delicado et al. 2006). Ectonucleotidases readily hydrolysing ApnA into AMP and adenosine 5′(n−1) phosphate could provide a well-suited mechanism to enzymatically regulate signalling by these dinucleotides (Rodriguez-Pascual et al. 1992; Ramos et al. 1995; Mateo et al. 1997; Asensio et al. 2007).
HPLC analysis has shown that murine N2a neuroblastoma cells possess an ectoenzymatic activity that can hydrolyse ApnA, as demonstrated by the rapid and complete degradation of extracellularly added Ap5A. This activity has been characterized with the help of the fluorogenic substrate BODIPY-FL-GTPγS, formed by the BODIPY FL fluorescent dye linked through a thioester bound to the terminal phosphate of GTPγS. This compound can be considered as a dinucleotide analogue containing a fluorophore in place of one of the nucleosides. Because of the interaction of the BODIPY FL dye with the guanine base of GTP, BODIPY-FL-GTPγS shows significant fluorescence quenching that is relieved when the phosphate chain of the molecule becomes cleaved (Draganescu et al. 2000). This fluorogenic compound has been used to analyse the diadenosine triphosphate hydrolase activity of Fhit, one of the most frequently inactivated proteins in lung cancer that functions as a tumour suppressor (Draganescu et al. 2000). BODIPY-FL-GTPγS is cleaved by the N2a cells in a dose-dependent manner with a Km value in the low micromolar range. Inhibition studies showed that BODIPY-FL-GTPγS cleavage becomes reduced by Ap5A, Ap4A and Ap3A, thus indicating that a common activity with a broad substrate specificity is responsible for the extracellular degradation of both ApnA and the fluorogenic substrate. BODIPY-FL-GTPγS cleavage was also inhibited by the nucleoside monophosphate AMP. These results point to a member of the E-NPP family as being responsible for the extracellular dinucleotide degradation in the N2a cells. NPP1-NPP3 enzymes have been found to be able to hydrolyse pyrophosphate and phosphodiester bonds in a wide range of nucleotidic substrates, including ApnA which exhibited affinity values in the low micromolar range (Bollen et al. 2000; Vollmayer et al. 2003; Asensio et al. 2007; Zimmermann et al. 2012). Nucleoside monophosphates such as AMP are not hydrolysed but exert competitive product inhibition on NPP reaction (Bollen et al. 2000; Vollmayer et al. 2003; Zimmermann et al. 2012). In addition, the ectoenzymatic dinucleotide cleaving activity in N2a cells exhibited biochemical features that are characteristic of the members of the E-NPP family (Bollen et al. 2000; Vollmayer et al. 2003; Zimmermann et al. 2012): (i) catalytic activity was dependent on divalent cations, being strongly inhibited by EDTA, (ii) activity was drastically reduced in the absence of both Ca2+ and Mg2+ ions, but enzyme reactivation was observed by the addition on Zn2+, (iii) finally, pH dependence studies revealed an alkaline optimum pH for the enzyme activity.
Fluorimetric enzyme assays were performed by replacing the incubation medium by fresh Locke′s solution immediately before the start of the experiments, which avoids the participation of a secreted enzyme such as NPP2 (Stefan et al. 2005, 2006; Zimmermann et al. 2012). Regarding NPP1 and NPP3, both could be detected in the N2a cells at the transcriptional and protein levels. Q-PCR analysis showed that these two isozymes are expressed to a similar extent in the neuroblastoma cells.
E-NPP expression levels vary when N2a differentiate into neuronal-like cells, which was achieved by incubating them in a medium with a low (0.5%) percentage of FBSi, supplemented with DiBucAMP. Both NPP1 and NPP3 mRNA become significantly reduced at the early stages of differentiation (after 24 h of treatment with DiBucAMP). However, there are differences in the time required for these changes to be translated into changes in the protein levels. NPP3 protein was dramatically reduced in the differentiated N2a cells being almost completely absent after 24 h of treatment with DiBucAMP. However, NPP1 protein levels have only been partially reduced in the differentiated cells after 24 or 48 h of treatment with DiBucAMP, requiring longer treatment times to be substantially decreased. In addition, the remaining NPP1 protein in the differentiated cells seems to change its intracellular distribution from a somatic to a ‘neurite-like’ location. These differences may reflect differences in the turnover of both NPP1 and NPP3 proteins. Rapid fall in the levels of NPP3 protein when N2a cells differentiate could make this protein a suitable surface marker for undifferentiated N2a cells. NPP1 (also known as PC-1) is expressed at limited stages of antibody-producing B-cell differentiation (Anderson et al. 1984), whereas NPP3 seems to have a role in the differentiation and invasive properties of glial cells (Deissler et al. 1999). Regarding tumour cells, NPP3 has been suggested to be an early marker of cholangiocarcinoma, an adenocarcinoma derived from biliary cells (Meerson et al. 1998). Moreover, NPP1 has been shown to be expressed in astrocytic brain tumours, and a correlation was found between the up-regulated expression of this ectoenzyme and the histological grade of the astrocytoma (Aerts et al. 2011b). Besides, cyclic AMP-dependent induction of differentiation inhibits NPP1 expression in the rat C6 glioma cells (Aerts et al. 2011a). However, the physiological implications of changes in the expression of E-NPP in the differentiated N2a cells remain to be elucidated.
Changes in ectoenzymatic dinucleotide-hydrolysing activity seem to reflect variations in the levels of NPP1 protein, as the activity is only partially reduced in differentiated cells after 24 h of treatment with DiBucAMP, being substantially decreased after 96 h: a pattern that quite exactly matches changes in NPP1 protein expression. However, N2a cells still retain a significant capacity to hydrolyse dinucleotides in conditions, in which NPP3 protein is almost completely loss, i.e. after 24 h of treatment with DiBucAMP. These results indicate that the dinucleotide cleaving activity of N2a cells largely resides in NPP1, NPP3 contributing in a minor, if any, extent to the extracellular degradation of dinucleotides in these cells. However, it is necessary to take into account that both NPP1 and NPP3 can hydrolyse several mononucleotides in addition to dinucleotides. Moreover, their spectrum of substrates is not restricted to nucleotidic compounds, these enzymes being able to hydrolyse pyrophosphate and phosphodiester bonds in a wide variety of substrates (Bollen et al. 2000; Vollmayer et al. 2003; Zimmermann et al. 2012). Thus, actions of NPP3 on extracellular compounds, other than dinucleotides, cannot be discarded. On the other hand, it is necessary to bear in mind that although the major functional role of ectonucleotidases is the extracellular hydrolysis of mono- and dinucleotides and the production of extracellular nucleosides, thus regulating the ligand availability at nucleotide and adenosine receptors, some ectonucleotidases are multifunctional proteins that can interact with the extracellular matrix or even signal into the cell (Zimmermann et al. 2012).
Acknowledgements and conflict of interest disclosure
This work has been supported by research grants from Ministerio de Ciencia e Innovación (BFU2011-24743), the Spanish Ion Channel Initiative (CSD2008-00005) and Fundación Marcelino Botín.
All experiments were conducted in compliance with the ARRIVE guidelines. The authors have no conflict of interest to declare.