Institut für Neurobiochemie, Medizinische Fakultät, Otto-von-Guericke Universität, Magdeburg, Germany
Address correspondence and reprint requests to Prof Georg Reiser, Institut für Neurobiochemie, Otto-von-Guericke-Universität Magdeburg, Medizinische Fakultät, Leipziger Straße 44, 39120 Magdeburg, Germany. E-mail: email@example.com
The P2Y11 nucleotide receptor detects high extracellular ATP concentrations. Mutations of the human P2RY11 gene can play a role in brain autoimmune responses, and the P2Y11 receptor alanine-87-threonine (A87T) polymorphism has been suggested to affect immune-system functions. We investigated receptor functionality of the P2Y11A87T mutant using HEK293 and 1321N1 astrocytoma cells. In HEK293 cells, the P2Y11 receptor agonist 3′-O-(4-benzoylbenzoyl)adenosine 5′-triphosphate (BzATP) was completely inactive in evoking intracellular calcium release while the potency of ATP was reduced. ATP was also less potent in triggering cAMP generation. However, 1321N1 astrocytoma cells, which lack any endogenous P2Y1 receptors, did not display a reduction. Only when 1321N1 cells were co-transfected with P2Y11A87T and P2Y1 receptors, the calcium responses to the P2Y11 receptor-specific agonist BzATP were reduced. It is already known that P2Y1 and P2Y11 receptors interact. We thus conclude that the physiological impact of A87T mutation of the P2Y11 receptor derives from detrimental effects on P2Y1–P2Y11 receptor interaction. We additionally investigated alanine-87-serine and alanine-87-tyrosine P2Y11 receptor mutants. Both mutations rescue the response to BzATP in HEK293 cells, thus ruling out polarity of amino acid-87 to be the molecular basis for altered receptor characteristics. We further found that the P2Y11A87T receptor shows complete loss of nucleotide-induced internalization in HEK293 cells. Thus, we demonstrate impaired signaling of the P2Y11 A87T-mutated receptors when co-operating with P2Y1 receptors.
The human P2Y11 nucleotide receptor plays key role in immune-responses in brain and other tissues. We provide evidence for significant functional disturbance of the P2Y11 receptor carrying the Alanine-87-Threonine mutation caused by natural polymorphism. This receptor defect is apparent only when co-expressed with P2Y1 receptors. We found reductions in ligand-induced calcium and cAMP responses and in nucleotide-induced receptor internalization / resensitization. Thus, prolonged nucleotide treatments are the basis for the molecular defects of the mutant receptor in diseases.
In the brain, the P2Y11 receptor has been reported to play a role in autoimmune responses leading to narcolepsy–cataplexy (Kornum et al. 2011). This disease is characterized by loss of neurons in the hypothalamus. The single-nucleotide polymorphism (SNP) rs2305795 in the 3′ untranslated region (UTR) of the human P2RY11 receptor gene causes reduced receptor expression levels in T-lymphocytes and natural killer cells and impairs P2Y11 receptor-mediated protection against ATP-induced cell death (Kornum et al. 2011). Very little is known about functions of the P2Y11 receptor in the brain. High extracellular concentrations of ATP as a result of cell lysis promote inflammatory responses via P2Y receptor activation (Di Virgilio et al. 2009; Vitiello et al. 2012). As the P2Y11 receptor detects high extracellular ATP in the μM range (Haas et al. 2013), this receptor is predestined to playing a major role in inflammatory processes in the brain.
Another SNP in the coding region of the gene (rs3745601) results in amino acid substitution of alanine to threonine at position 87 of the P2Y11 receptor. This SNP was found to be linked to an increased risk for acute myocardial infarction (AMI) and increased levels of C-reactive protein, indicating inflammatory reactions (Black et al. 2004; Amisten et al. 2007). The association was strongest for homozygous mutations and genetically related early onset and family history AMI. AMI is often caused by atherosclerosis, which is known to be an immune-system-related disease (Tabas 2010). However, the functional and physiological impact of the A87T mutation of the human P2Y11 receptor remains unclear. We address this question in this study.
In natural killer cells which express all P2Y receptor types, ATP released by endothelial cells activates the P2Y11 receptor and inhibits chemotaxis and cytotoxicity (Gorini et al. 2010). In dendritic cells, the P2Y11 receptor regulates ATP-dependent maturation (Wilkin et al. 2001), cell migration (Schnurr et al. 2003), release of interleukin-8 (Meis et al. 2010) and interleukin-12, as well as stimulation of interleukin-10 production (Wilkin et al. 2002). The activation of the P2Y11 receptor by ATP in dendritic cells either promotes inflammation or supports immune tolerance by shaping T helper cell responses (Wilkin et al. 2002). Activation of the P2Y11 receptor with extracellular ATP inhibits the constitutive apoptosis of neutrophils (Vaughan et al. 2007). Autocrine activation of P2Y11 receptors by ATP release from intracellular vesicles causes activation of macrophages with IL-6 secretion (Sakaki et al. 2013).
In humans, the mRNA for the P2Y11 receptor is abundant in the brain, spleen, and lymphocytes, but can also be detected in macrophages, platelets, neutrophils, dendritic cells, and the heart (Berchtold et al. 1999; Moore et al. 2001; Schnurr et al. 2003; Wang et al. 2004; Wihlborg et al. 2006). The expression in platelets remains unclear as Wang and coworkers (Wang et al. 2003) found no indication of the presence of the P2Y11 receptor mRNA in platelets.
The P2Y11 receptor is a 7-transmembrane domain, G protein-coupled receptor, which belongs to the family of eight human P2Y receptors (Communi et al. 1997; Abbracchio et al. 2006). The human P2Y1 receptor is the closest homolog of the human P2Y11 receptor with 33% amino acid identity. Each P2Y receptor subtype has a distinct pattern of physiologically active adenine (P2Y1, P2Y11, P2Y12, and P2Y13 receptors) and uridine nucleotide ligands (P2Y4, P2Y6, and P2Y14 receptors). The P2Y2 receptor is activated equally well by both ATP and UTP. The receptors P2Y1 to P2Y11 are coupled to Gq signaling and therefore mediate the rise of intracellular calcium. In addition, the P2Y2 and P2Y4 receptors are linked to Gi signaling (Communi et al. 1996; Murthy and Makhlouf 1998). A unique feature within the P2Y receptor family is the P2Y11 receptor coupling to both Gq and Gs signal transduction. The latter mediates adenylyl cyclase activation and thus cAMP accumulation. The P2Y12 and P2Y13 receptors inhibit the adenylyl cyclase via Gi signaling.
The physiological standard agonist for the P2Y11 receptor is ATP, although ATP also activates P2Y1 and P2Y2 receptors. Therefore, 3′-O-(4-Benzoylbenzoyl)adenosine 5′-triphosphate (BzATP) is most frequently used as a potent agonist, which is selective for the P2Y11 receptor within the P2Y receptor family.
An interaction of endogenous P2Y1 receptors with P2Y11 receptors has been shown in HEK293 cells (Ecke et al. 2008). In that study, it was demonstrated that the receptor interaction results in distinct functional and pharmacological properties. Several cell types have been reported to express the P2Y1 receptor, in addition to the P2Y11 receptor. Therefore, it is important to study the consequences of an interaction between P2Y1 receptors and P2Y11 mutant receptors.
Here, we investigated the P2Y11A87T receptor in comparison with the non-mutated wild-type P2Y11 receptor to obtain insights into alterations of receptor characteristics. As P2Y11 receptor interaction with P2Y1 receptors represents a physiologically relevant situation, we used HEK293 cells to mimic this condition. We also analyzed in 1321N1 astrocytoma cells the P2Y11 receptor and its mutant. These cells lack any endogenous P2Y receptor expression. We determined the characteristics of the ligand-induced intracellular calcium responses, ATP-induced cAMP accumulation, nucleotide-induced receptor internalization, as well as the resensitization of the calcium response after a prolonged receptor desensitization period. Our results show that an impaired function of the P2Y11 receptor carrying the A87T mutation occurs only in cells that also express the P2Y1 receptor.
Materials and methods
The pEGFP-N1 expression vectors were from Clontech Laboratories Inc. (Palo Alto, CA, USA), the pcDNA 3.1/myc-His vector from Life Technologies Corporation (Carlsbad, CA, USA). Dulbecco's modified Eagles' medium, penicillin and streptomycin were from Biochrom AG (Berlin, Germany). Fetal calf serum was from PAA Laboratories GmbH (Pasching, Austria), G418 (geneticine) from Merck Chemicals GmbH (Schwalbach/Ts., Germany). Fura 2/AM was supplied by Life Technologies Corporation (Carlsbad, CA, USA), FuGENE 6 Transfection Reagent, and DOTAP by Roche Diagnostics GmbH (Mannheim, Germany). All other chemicals used for cell culture and single-cell calcium measurements were obtained from Carl Roth GmbH (Karlsruhe, Germany).
1321N1 astrocytoma cells or HEK293 cells were used to stably express green fluorescent protein (GFP) constructs of human P2Y1, P2Y11, P2Y11A87T, P2Y11A87S, and P2Y11A87Y receptors. The receptor cDNA was cloned into a pEGFPN1 vector. 1321N1 astrocytoma cells were further used for P2Y1-myc/His receptor expression using a pcDNA/myc-His vector. Cell culture conditions and transfection were described before (Ecke et al. 2008). All transfected HEK293 and 1321N1 astrocytoma cell cultures used in the present experiments were derived from the same stock wild-type cell cultures to guarantee equal conditions for all experiments.
For site-directed mutagenesis, the QuickChange Site-Directed Mutagenesis Kit (Agilent Technologies Inc., Santa Clara, California, USA) was used. Mutagenesis primers were designed according to the manual. A Pfu polymerase was used instead of the Pfu Turbo polymerase. The PCR cycler program was performed as follows: initial denaturation at 98°C for 30 s, second denaturation at 95°C for 30 s, primer annealing at 55°C for 1 min, polymer extension at 70°C for 12 min. The steps 2–4 were repeated 16 times. After mutagenesis PCR, 1 μL of DpnI restriction enzyme was added to the PCR product to digest methylated template DNA. Competent DH5α E. coli bacteria were used for amplification of the plasmid. Plasmids from selected clones were extracted and digested with HindIII and BamHI to verify correct fragment size. Positive plasmids were then sequenced to confirm the successful introduction of the desired point mutation.
Single-cell calcium measurements
For calcium measurements, cells were grown on 22-mm coverslips as described in Ecke et al. (2008). The nucleotide-induced change in [Ca2+]i was monitored by detecting the respective 510-nm emission intensity of the ratiometric calcium-sensitive dye Fura 2 after excitations at 340 nm and 380 nm. The ratio of the emission intensities of 340 nm and 380 nm excitations represents the intracellular calcium concentration. SigmaPlot (SPSS Inc., Chicago, IL, USA) was used to derive the concentration–response curves and EC50 values from the average response amplitudes. Nucleotide concentrations relevant for EC50 value calculation were tested in at least three independent experiments with an average of 10–20 cells per experiment. Measurements for comparison of simultaneous nucleotide-induced rise of [Ca2+]i in GFP-negative and GFP-positive HEK293 cells were repeated at least 2–3 times with 5–40 cells per experiment. Experiments with 1321N1 astrocytoma cells co-expressing the P2Y1 and P2Y11 or P2Y11A87T receptors were repeated 3–9 times with 4–13 cells per experiment. The statistical analysis of the results was investigated using a one-way anova with a Tamhane-T2 post hoc test. Only cells with a clearly membrane-localized GFP signal and with the typical calcium response kinetics upon agonist pulse application were included in the data analysis of P2Y receptor-expressing cells. The GFP-tagged P2Y receptors are suitable for pharmacological and physiological studies, as previously reported (Tulapurkar et al. 2004, 2006; Ecke et al. 2006; Zylberg et al. 2007).
1321N1 astrocytoma cells or HEK293 cells expressing the P2Y11 receptor or P2Y11A87T receptor were grown in cell culture dishes (5 cm diameter) to approximately 70% confluency. The cells were incubated in Na-HBS buffer-containing 500-μM isobutyl-methyl-xanthine for 30 min at 37°C followed by a 10-min incubation at 37°C with the respective nucleotide concentrations. The cells were lysed with 0.1 M HCl, scraped from the dishes, and briefly sonicated. Cell extract (100 μL) was kept for the determination of total protein concentration. The following steps were conducted according to the manufacturer's manual. For the measurements, the direct cyclic AMP enzyme immunoassay kit from ENZO Life Sciences Inc. (Farmingdale, NY, USA) was used. The mean values given are based on at least two samples with the majority of experiments carried out at least six times.
The statistical analysis of the results was investigated using a one-way anova with a Tamhane-T2 post hoc test.
Nucleotide-induced receptor internalization
Investigation of nucleotide-induced internalization of the P2Y11 receptor and the A87T, A87S, and A87Y P2Y11 receptor mutants was done using a LSM 510 META laser scanning confocal microscope (Carl Zeiss AG, Oberkochen, Germany). HEK293 cells expressing the respective GFP-labeled receptor were grown on a cover slide to approximately 70% confluency. After incubation of the cells with the respective nucleotide for 60 min, cells were fixed with para-formaldehyde and the localization of GFP fluorescence was detected.
Nucleotide-induced intracellular calcium responses in 1321N1 astrocytoma cells and HEK293 cells, which express P2Y11 wild-type or A87T, A87S, A87Y receptor mutants
1321N1 astrocytoma cells are commonly used for the investigation of isolated pharmacological properties of P2Y receptors, as they lack endogenous P2Y receptor expression (Communi et al. 1997; Ecke et al. 2006, 2008; Meis et al. 2010). In HEK293 cells, it was shown recently that the expression of P2Y11 receptors leads to the formation of receptor hetero-oligomers with the endogenously expressed P2Y1 receptor displaying new pharmacological properties (Ecke et al. 2008). The pharmacological properties are distinct from the known properties of the P2Y1 or the P2Y11 receptors, when they are expressed in the absence of the respective interaction partners. As this interaction between P2Y11 and P2Y1 receptors is of physiological relevance, we additionally used HEK293 cells to express the P2Y11 receptor and the P2Y11 receptor mutants generated by us. A possible effect of the A87T mutation on the P2Y1/P2Y11 receptor oligomer might otherwise remain undetected. We confirmed that the expression levels for the P2Y11 receptor and the P2Y11 receptor mutants in HEK293 cells were comparable (Figure S6).
In an additional analysis (Figure S3), we verified in HEK293 and 1321N1 astrocytoma cells that no other P2Y receptors would be expressed (Ginsburg-Shmuel et al. 2010) to interfere with the responses detected in our experiments.
Concentration–response curves of nucleotide-induced Ca2+ responses were established to investigate the potency of BzATP at the P2Y11 wild-type and P2Y11A87T mutant receptors in 1321N1 astrocytoma cells (Fig. 1a). For both receptors, the nucleotide potencies were virtually identical, with EC50 values of 0.8 ± 0.1 μM for the P2Y11 receptor and 0.9 ± 0.07 μM for the P2Y11A87T receptor. There were no significant differences between both calcium response curves. Analysis of the P2Y11 receptor expressed in HEK293 cells gave a comparable EC50 value of 1.0 μM for BzATP.
Astonishingly, for the P2Y11 receptor A87T mutant, the BzATP-induced intracellular calcium response in HEK293 cells was completely abolished (Fig. 1b). The EC50 values for HEK293 cells, as summarized in Table 1, demonstrate a BzATP-induced calcium response of the P2Y11 receptor with A87Y mutation at the level of the unmutated P2Y11 receptor (EC50 = 1.6 μM), whereas the affinity of the P2Y11A87S receptor was significantly higher (EC50 = 0.8 μM; p < 0.01). HEK293 wild-type cells show no calcium response to BzATP, which is expected because of the lack of P2Y11 receptors in these cells (Figure S2).
Table 1. EC50 values of ATP, BzATP, and 2-MeS-ADP at the P2Y11, P2Y11A87T, P2Y11A87S, and P2Y11A87Y receptors expressed in HEK293 cells in comparison with HEK293 wild-type cells
HEK293 cell-type response to nucleotides – EC50 (μM ± SEM)
HEK293 wild-type cells
Data were obtained from concentration–response curves, as shown in Fig. 1. The cells were pre-incubated with 2 μM Fura-2 AM for 30 min and the change in fluorescence (Δ F340 nm/F380 nm) was detected as described in Methods. n.r. = no intracellular calcium response for nucleotide concentrations of up to 100 μM.
1.0 ± 0.05
0.8 ± 0.06
1.6 ± 0.1
0.14 ± 0.01
1.2 ± 0.3
0.5 ± 0.09
0.1 ± 0.04
0.04 ± 0.006
1.8 ± 0.08
2.8 ± 0.2
2.9 ± 0.3
3.5 ± 0.5
1.8 ± 0.02
Testing again 1321N1 astrocytoma cells, we found that 2-methylthioadenosine 5′-diphosphate (2-MeS-ADP) was a very weak agonist of both the P2Y11 wild-type and the P2Y11A87T mutant receptor with respective EC50 values of 24.2 ± 0.3 μM and 35.6 ± 0.001 μM (Fig. 1c). For validation of these results, we also analyzed the P2Y1 receptor expressed in 1321N1 cells. This receptor showed a low EC50 value of 0.33 ± 0.18 nM, demonstrating the P2Y1 receptor specificity of 2-MeS-ADP. Wild-type HEK293 cells also show a prominent intracellular calcium response to 2-MeS-ADP (Fig. 1d) as a result of the presence of the endogenous P2Y1 receptor (EC50 = 0.04 μM, Table 1). However, the potency of 2-MeS-ADP at HEK293 cells additionally expressing the P2Y11 receptor was significantly (p < 0.01) reduced (EC50 = 0.14 μM). The A87T mutation had a further detrimental effect on the potency of 2-MeS-ADP, as the EC50 value was reduced about 9-fold to only 1.2 μM. The average calcium responses to 2-MeS-ADP in HEK293 cells expressing the P2Y11A87T receptor were significantly reduced (p < 0.01; Fig. 1d) compared with the P2Y11 receptor-expressing cells. Also, the A87S mutation decreased the 2-MeS-ADP potency compared with the unmutated P2Y11 receptor. However, with EC50 = 0.5 μM, the decrease was less pronounced than with the A87T mutation (Table 1). Expression of the P2Y11A87Y receptor lead to a 2-MeS-ADP potency comparable with that of the unmutated P2Y11 receptor (EC50 = 0.1 μM, Table 1).
Next, we determined the EC50 values for the physiological ligand ATP. In 1321N1 astrocytoma cells, the EC50 values were almost identical with 2.1 ± 0.2 μM for the P2Y11 wild-type receptor and 2.4 ± 0.7 μM for the P2Y11A87T receptor-expressing cells (Fig. 1e).
In HEK293 cells with P2Y11 receptor expression (Fig. 1f), the EC50 value of ATP was identical to that of HEK293 wild-type cells (EC50 = 1.8 μM, Table 1). Both the A87T and A87S mutations of the P2Y11 receptor similarly reduced the potency of ATP (EC50 = 2.8 μM and 2.9 μM, respectively). The ATP potency was weakest at HEK293 cells expressing the P2Y11A87Y receptor (EC50 = 3.5 μM). However, at a nucleotide concentration of 3 μM, close to the EC50 value of these receptors, there was no significant difference in the calcium responses of HEK293 cells expressing either of the mutated receptors. Therefore, ATP has about equal potency at the P2Y11A87T, P2Y11A87S, and P2Y11A87Y receptors. The calcium responses of the HEK293 cells expressing the P2Y11A87T receptor were significantly lower than those of cells expressing the P2Y11 receptor or HEK293 wild-type cells (p < 0.01; Fig 1f).
ATP-induced cAMP accumulation in 1321N1 astrocytoma cells and HEK293 cells expressing the P2Y11 wild-type or P2Y11A87T receptor
The P2Y11 receptor is dually coupled to the Gq and the Gs signaling pathways. As we detected a detrimental effect of the A87T mutation of the P2Y11 receptor on Gq-mediated intracellular calcium signaling in HEK293 cells, it was important to test also the effect on Gs signaling. Therefore, we investigated whether the A87T mutation of the P2Y11 receptor would have an effect on the nucleotide-induced cAMP accumulation in 1321N1 astrocytoma cells (Fig. 2a) and HEK293 (Fig. 2b).
1321N1 astrocytoma cells showed a similar amount of cAMP accumulation for cells expressing either one or the other receptor, testing ATP concentrations of up to 50 μM. However, in HEK293 cells expressing the P2Y11A87T receptor, we found lower cAMP-level elevation for ATP (at concentrations of 20–100 μM) compared with cells expressing the P2Y11 receptor. The difference between the P2Y11 and the P2Y11A87T receptor in cAMP signaling can thus be detected only in HEK293 cells.
Intracellular calcium responses, measured simultaneously in non-transfected HEK293 cells and HEK293 cells transfected with the P2Y11 or P2Y11A87T receptors
In single-cell calcium measurements, the transfected cells can be identified in our experiments via the GFP label of the P2Y11 receptors. We next analyzed the intracellular calcium responses simultaneously in both non-transfected (GFP-negative) and transfected (GFP-positive) cells under identical experimental conditions. This is possible because after transfection usually only about 50% of the cells in the culture show expression of GFP-labeled receptors. The responses were simultaneously evoked within the same culture of HEK293 cells with BzATP, 2-MeS-ATP, and ATP. The results are given in Table 2.
Table 2. Comparison of average amplitudes of the rise of [Ca2+]i after stimulation with BzATP, 2-MeS-ADP, or ATP alone or in combination with the P2Y1 receptor antagonist MRS2179
Data were obtained from measurements with HEK293 cell cultures containing non-transfected cells (GFP− = GFP-negative cells) and cells expressing the P2Y11 or P2Y11A87T receptor (GFP+ = GFP-positive cells; compare Figure S1a). If necessary, cells were additionally pre-incubated with antagonist before the measurements. The cells were pre-incubated with 2 μM Fura-2 AM for 30 min and the ratio of the change in fluorescence (Δ F340 nm/F380 nm), representing the calcium response amplitude, was simultaneously detected for GFP-negative and GFP-positive cells, as described in Methods. For data comparison, the factor (F) of the calcium response amplitudes of GFP-positive/GFP-negative cells is given.
BzATP (3 μM)
0.2 ± 0.02
1.9 ± 0.07
0.09 ± 0.02
0.09 ± 0.02
0.07 ± 0.01
BzATP (10 μM)
0.2 ± 0.01
1.8 ± 0.1
0.2 ± 0.04
0.07 ± 0.01
0.07 ± 0.003
2-MeS-ADP (1 μM)
1.9 ± 0.06
1.1 ± 0.1
1.5 ± 0.1
0.5 ± 0.08
2.1 ± 0.06
2-MeS-ADP (3 μM)
3.0 ± 0.1
2.5 ± 0.1
2.6 ± 0.15
0.8 ± 0.03
2.5 ± 0.06
ATP (3 μM)
1.6 ± 0.07
2.0 ± 0.07
1.9 ± 0.08
1.2 ± 0.06
1.8 ± 0.05
ATP (3 μM) + MRS2179 (100 μM)
0.4 ± 0.07
1.9 ± 0.1
0.2 ± 0.04
0.1 ± 0.01
0.2 ± 0.03
ATP (10 μM)
1.8 ± 0.06
2.2 ± 0.11
2.5 ± 0.06
1.6 ± 0.05
2.2 ± 0.06
For the analysis we here introduced the factor F. F is defined as the ratio of the nucleotide-induced calcium response amplitude of GFP-positive cells divided by the response amplitude of GFP-negative cells. This factor describes the consequences of the expression of either the P2Y11 or the P2Y11A87T receptor in HEK293 cells. F > 1 indicates an enhanced response of GFP-positive cells compared with GFP-negative cells, and F < 1 indicates a reduced response of the GFP-positive cells. For HEK293 cells expressing the P2Y11 receptor, F was 9.5 upon treatment with 3 μM BzATP. This means that the response amplitude to 3 μM BzATP was 9.5 times higher than the response amplitude of GFP-negative cells. BzATP of 10 μM leads to comparable calcium response amplitudes (F = 9.0). This is consistent with our finding that 3 μM BzATP triggered an already near-maximal rise of [Ca2+]i (see Fig. 1b). In HEK P2Y11A87T cultures, GFP-positive cells showed a negligible calcium response amplitude to 3 μM and 10 μM BzATP, similar to that in GFP-negative cells. The response amplitudes of all GFP-negative cells were comparable with those of HEK293 wild-type cells evaluated in a separate culture.
As shown in Table 2, a decrease in the calcium response to 1 μM 2-MeS-ADP (F = 0.6) in HEK293 cells expressing the P2Y11 receptor could also be detected for 3 μM 2-MeS-ADP (F = 0.8). The calcium response to 3 μM 2-MeS-ADP in HEK293 cells expressing the P2Y11A87T receptor was stronger reduced when compared with GFP-negative cells. The F-value for 2-MeS-ADP in P2Y11A87T expressing cells was 0.3. Therefore, the difference in the calcium responses between GFP-positive and GFP-negative cells was again greater in HEK293 cells expressing the P2Y11A87T receptor than in cells expressing the P2Y11 wild-type receptor. HEK293 wild-type cells of a separate culture showed intracellular calcium responses to 2-MeS-ADP comparable with the GFP-negative cells in P2Y11 and P2Y11A87T receptor-transfected cultures. In all measurements, HEK293 cells expressing the P2Y11A87T receptor displayed equal or lower calcium response amplitudes than the GFP-negative cells in the same culture. We attribute this reduction in the response to the receptor interaction of P2Y11A87T with P2Y1 receptors in these cells.
We employed the P2Y1 receptor-specific antagonist MRS2179 to identify the P2Y receptor subtype responsible for the nucleotide-induced rise of [Ca2+]i in GFP-negative cells (Table 2). The calcium response amplitudes of GFP-negative cells to 3 μM ATP was strongly inhibited by the antagonist. The ATP-induced calcium response of HEK293 wild-type cells from control cultures was similarly inhibited. As Table 2 further shows, interestingly, there was no significant reduction in the response amplitude in cells expressing the P2Y11 receptor (amplitude = 1.9). However, in cells expressing the P2Y11A87T receptor, the reduction was similar to that in GFP-negative cells (amplitudes = 0.1 and 0.2, respectively).
The calcium response amplitudes of HEK293 cells expressing the P2Y11 or P2Y11A87T receptors challenged with ATP are in accordance with our data from the concentration–response experiments (Fig. 1f). We found that cells expressing the P2Y11 receptor showed a significantly increased (p < 0.05) intracellular calcium response to 3 μM ATP (amplitude = 2.0) as compared with GFP-negative cells in the same culture (amplitude = 1.6). Cells expressing the P2Y11A87T receptor, however, showed a significantly (p < 0.01) decreased intracellular calcium response with an amplitude of 1.2 compared to 1.9 for GFP-negative cells.
Table 2 further delineates that for an ATP concentration of 10 μM, HEK293 cells expressing the P2Y11 receptor showed an increased calcium response amplitude compared with GFP-negative cells (F = 1.2) similar to what we observed for 3 μM ATP (F = 1.3). Cells expressing the P2Y11A87T receptor showed a reduced response to 10 μM and similarly to 3 μM ATP (F = 0.6).
We also conducted a technical control experiment to rule out the possibility that simply the presence of the GFP tag at the receptor protein is the cause for the observed differences in the calcium response amplitudes between the GFP-positive and GFP-negative cells under the conditions given. We therefore conducted measurements with HEK293 cells expressing the GFP-tagged protease-activated receptor 2 (PAR2; data not shown). Transfected and non-transfected HEK293 cells in these cultures showed virtually identical intracellular calcium responses to 3 μM ATP (amplitudes = 1.4 ± 0.04 and 1.5 ± 0.05, respectively) and 10 μM ATP (amplitudes = 2.0 ± 0.06 and 2.0 ± 0.08, respectively). Responses were also comparable with response levels in HEK293 wild-type cells.
Co-expression of the P2Y11 wild-type or P2Y11A87T mutant receptors with the P2Y1 receptor in 1321N1 astrocytoma cells
The most striking observation in this study is the effect described in Fig. 1a and b: While the P2Y11A87T receptor and the wild-type P2Y11 receptor expressed in 1321N1 astrocytoma cells revealed no difference in BzATP-induced calcium responses (Fig. 1a), in HEK293 cells the expressed P2Y11A87T receptor resulted in a complete loss of this calcium response (Fig. 1b). We hypothesize that the A87T mutation of the P2Y11 receptor affects the functionality of the P2Y1/P2Y11 receptor oligomer in HEK293 cells. A crucial step to test this hypothesis was the following experiment: We co-expressed the P2Y1 receptor with the P2Y11 or the P2Y11A87T receptor in 1321N1 astrocytoma cells. In these cells, we compared the response amplitudes induced by 1 μM of the P2Y11 receptor-specific agonist BzATP and of the P2Y1 receptor-specific agonist 2-MeS-ADP (Fig. 3). The nucleotides were applied with a recovery period of at least 2 min after the first treatment with BzATP. The concentration of 1 μM 2-MeS-ADP was high enough to strongly activate P2Y1 receptor-mediated intracellular calcium signaling and was at the same time too low to trigger a P2Y11 receptor-mediated response (Fig. 1c).
For the results shown in Fig. 3, the type of receptor or combination of receptors expressed in the 1321N1 astrocytoma cells is given at the bottom of the graph. Cells expressing the P2Y1 receptor alone did not respond to the BzATP stimulus, but showed the expected strong response to 2-MeS-ADP. Cells expressing the P2Y11 or the P2Y11A87T receptor alone showed an intracellular calcium response to BzATP of comparable amplitude, but no response to 2-MeS-ADP.
Importantly, in 1321N1 astrocytoma cells co-expressing the P2Y11A87T receptor mutant and the P2Y1 receptor, we observed a significantly (p < 0.01) reduced response amplitude to BzATP compared with cells co-expressing the unmutated P2Y11 receptor and the P2Y1 receptor. The response amplitude was also significantly (p < 0.01) lower than in cells expressing the P2Y11 or the P2Y11A87T receptor in absence of the P2Y1 receptor. Furthermore, cells co-expressing the P2Y11 and the P2Y1 receptor showed a calcium response to BzATP that was not significantly different from cells expressing the P2Y11 receptor alone.
The reduction in the calcium response amplitude to BzATP in cells co-expressing the P2Y11A87T and the P2Y1 receptor is consistent with our findings in HEK293 cells (see Fig. 1b). The data in Fig. 3 clearly demonstrate that the A87T mutation of the P2Y11 receptor has a significant impact in cells co-expressing the P2Y1 receptor. The reduction in the 1321N1 cells was not as severe as in HEK293 cells. The smaller reduction probably is due to different cell-type-specific characteristics. Along the same line, the reduction in the 2-MeS-ADP potency seen above in HEK293 cells expressing the P2Y11 or P2Y11A87T receptor cannot be demonstrated in 1321N1 astrocytoma cells co-expressing the P2Y11A87T or P2Y11 receptor and the P2Y1 receptor. In these cells, the response amplitudes to 2-MeS-ADP were equal at a 1 μM concentration.
The co-expression of the P2Y1-myc/His receptor and the P2Y11-GFP or P2Y11A87T-GFP receptors in 1321N1 astrocytoma cells is shown in Figure S4. The fluorescence intensities of the GFP and of Alexa555-marked α-myc-antibody were equal (Figure S5) in both cultures verifying equal receptor expression levels.
Nucleotide-induced internalization of P2Y11 receptors and P2Y11(Ala-87) Thr/Ser/Tyr receptor mutants in HEK293 cells and de-/resensitization of calcium responses
The localization of the GFP fluorescence of the P2Y receptor fusion protein in HEK293 cells was investigated using a confocal laser scanning microscope. Untreated control cells show a localization of the fluorescence mostly at the plasma membrane (Fig. 4), indicating the correct localization of the P2Y receptors. The wild-type P2Y11 receptor expressed in HEK293 cells was internalized during a 60-min lasting treatment with ATP (100 μM). This is indicated by a strong intracellular accumulation of GFP fluorescence. The P2Y11 receptor-specific agonist BzATP (100 μM) caused a strong receptor internalization.
In HEK293 cells expressing the P2Y11A87T mutant receptor, no internalization could be detected after 60 min treatment with any of the nucleotides. In HEK293 cells expressing the P2Y11A87S receptor, a slight internalization could be observed after treatment with ATP (100 μM) and a clear internalization after BzATP (100 μM) application. HEK293 cells expressing the P2Y11A87Y receptor showed no receptor internalization.
The lack of nucleotide-induced receptor internalization in HEK293 cells expressing the P2Y11A87T receptor raised the question whether this would result in a higher responsiveness of the cells after a prolonged ATP stimulus. We therefore treated cells expressing the P2Y11, the P2Y11A87T, or the P2Y11A87S receptor with 100 μM ATP for 30 min, which is sufficient to induce receptor internalization in P2Y11 receptor-expressing cells. This was followed by a 60 min recovery period in nucleotide-free Na-HBS buffer. After that time, a brief (1 min) second test treatment with ATP (100 μM) was applied to determine the extent of the recovery of the calcium response. First, the absolute amplitudes (Δ F340 nm/F380 nm) of the immediate calcium response to ATP at 0 min, and second of the intracellular calcium level after the continuous 30-min incubation are given in Fig. 5. The third bar in Fig. 5, respectively, displays the response to the second short ATP test stimulus at 90 min. The response at 90 min represents the expected physiological excitability of cells co-expressing the P2Y1 and P2Y11 or P2Y11A87T receptors after long-term exposure to ATP.
In HEK293 cells expressing the P2Y11 receptor, the calcium response after 30 min continuous ATP treatment was significantly reduced (p < 0.01) to 25% of the instantaneous response at 0 min. The recovery of the calcium response from 30 to 90 min was statistically significant (p < 0.01), but the increase was only 1.4-fold, reaching 34% of the 0-min response.
In HEK293 cells expressing the P2Y11A87T receptor, the calcium response after 30 min continuous ATP treatment was reduced significantly (p < 0.01) to 15% of the response at 0 min. The level of desensitization after 30 min was significantly higher (p < 0.01) than in cells expressing the P2Y11 receptor. The response at 90 min was significantly increased, by a factor of 3.4 (p < 0.01) compared to the response at 30 min and to 51% of the response at 0 min. Therefore, the recovery of the calcium response in HEK293 cells expressing the P2Y11A87T receptor was higher than in HEK293 cells expressing the P2Y11 receptor.
In HEK293 cells expressing the P2Y11A87S receptor, the calcium response after 30 min continuous ATP treatment was significantly (p < 0.01) reduced to 12% of the response at 0 min. This was comparable with the P2Y11A87T receptor-expressing cells after 30 min. At 90 min, the response was increased significantly (p < 0.01) 4.8-fold compared with the response at 30 min and to 57% of the response at 0 min. Therefore, in cells expressing the P2Y11A87S receptor, desensitization and the recovery of the calcium response were comparable with that in cells expressing the P2Y11A87T receptor.
The immediate calcium response amplitudes upon the ATP stimulus at 0 min were comparable for the P2Y11 and P2Y11A87S receptors; however, the response amplitude of the P2Y11A87T receptor was significantly lower (p < 0.01).
We additionally made a technical control experiment with a 90-min superfusion of HEK293 wild-type cells with nucleotide-free buffer (data not shown). This treatment was not able to reduce the amplitude of the intracellular calcium response to ATP (100 μM) in comparison with the same ATP stimulus without the preceding superfusion with buffer. This confirms that possible stress caused by the experimental conditions was not responsible for the observed changes in intracellular calcium responses of the HEK293 cells.
The A87T mutation of the P2Y11 receptor reduces nucleotide-induced [Ca2+]i and cAMP responses in HEK293 cells, but not in 1321N1 astrocytoma cells
BzATP is a potent P2Y11 receptor agonist. In 1321N1 astrocytoma cells expressing either P2Y11 or P2Y11A87T receptors, we found virtually identical EC50 values for the rise of [Ca2+]i (Fig. 1a). HEK293 cells expressing the P2Y11 receptor also showed a strong calcium response (Fig. 1b). This is consistent with the notion that BzATP is able to activate the P2Y1–P2Y11 receptor oligomer (Ecke et al. 2008). However, with the A87T mutation of the P2Y11 receptor, the intracellular calcium response to BzATP in HEK293 cells was completely abolished.
The shift from a non-polar to a polar amino acid was suggested to have an effect on P2Y11 receptor function (Amisten et al. 2007). To explore this possibility, we generated P2Y11 receptors with A87S and A87Y point mutations. However, the A87S and A87Y receptor mutations rescued the calcium response, as BzATP had EC50 values in the range of that of the unmutated P2Y11 receptor (Table 1). As serine and tyrosine are both polar amino acids like threonine, a switch in polarity at position 87 of the P2Y11 receptor can be ruled out to be the reason for altered receptor functionality.
A high sensitivity to 2-MeS-ADP of HEK293 wild-type cells confirms the endogenous P2Y1 receptor expression in these cells (Fig. 1d). The 3.5-fold rise of the EC50 value in cells additionally expressing the P2Y11 receptor is likely because of the interaction of endogenous P2Y1 receptors with the P2Y11 receptors. Interestingly, the expression of P2Y11A87T receptors resulted in an additional increase in the EC50 value.
Besides the artificial nucleotides BzATP and 2-MeS-ADP, we used ATP, which is the physiological agonist not only for the P2Y11 receptor but also for the P2Y1 receptor. In 1321N1 astrocytoma cells, we again found no significant differences in nucleotide potency. However, in HEK293 cells, there was a significant albeit small difference in calcium response between the P2Y11 and P2Y11A87T receptor-expressing cells (Fig. 1f).
These findings substantiate our hypothesis that the A87T mutation of the P2Y11 receptor per se is not sufficient to have a detrimental effect on the downstream calcium response. The co-expression of the P2Y1 receptor and the possible interaction between both receptors are required for the impact of the A87T mutation on intracellular calcium signaling.
Within the human P2Y receptor family, the dual coupling to Gq and Gs signaling pathways is unique for the P2Y11 receptor. Therefore, we investigated the impact of the A87T mutation of the P2Y11 receptor also on Gs signaling, measuring the ATP-induced cAMP accumulation (Fig. 2). The impact of the A87T mutation of the P2Y11 receptor could be found only in HEK293 cells (Fig. 2b). Thus, for both signaling pathways, the A87T mutation of the P2Y11 receptor has a detrimental effect in HEK293 cells.
We used P2Y receptors labeled with GFP. The GFP tag enables us to verify receptor localization at the plasma membrane. We previously confirmed (Ecke et al. 2006) that the rank order of nucleotide potencies of the GFP-labeled P2Y11 receptors is similar to that for receptors without the GFP tag (Communi et al. 1999).
We explored under identical conditions the calcium response amplitudes to BzATP, 2-MeS-ADP, and ATP in HEK293 cell cultures containing non-transfected (GFP-negative) and transfected (GFP-positive) cells (Figure S1). We found negligible calcium response amplitudes to BzATP in GFP-positive cells of HEK P2Y11A87T cultures, like in GFP-negative cells (Table 2). This was in clear contrast to GFP-positive cells expressing the P2Y11 receptor and corroborates our results depicted in Fig. 1b showing a complete loss of BzATP potency at the P2Y11A87T receptor. GFP-positive HEK293 cells expressing the P2Y11 or P2Y11A87T receptor both showed reduced calcium response amplitudes to the P2Y1 receptor agonist 2-MeS-ADP as compared with GFP-negative cells (Table 2). This is in accordance with the increased EC50 value of the agonist at these receptors compared with HEK293 wild-type cells (Fig. 1d) and indicates an interaction between endogenous P2Y1 and the transfected P2Y11 receptors. Furthermore, the response amplitude to ATP of GFP-positive HEK293 cells expressing the P2Y11A87T receptor was lower than in GFP-negative cells (Table 2).
A very weak ATP-induced intracellular calcium response in HEK293 wild-type cells and GFP-negative cells after co-application of the P2Y1 receptor-specific antagonist MRS2179 verified the activity of the P2Y1 receptor in HEK293 cells (Table 2). Interestingly, MRS2179 inhibited the calcium response in HEK293 cells expressing the P2Y11A87T receptor but not in cells expressing the P2Y11 receptor. This further illustrates that the A87T mutation of the P2Y11 receptor has an adverse effect on the P2Y1–P2Y11 oligomer function.
To test whether P2Y11A87T receptor signaling indeed is affected only in cells co-expressing simultaneously the P2Y1 receptor, we double-transfected 1321N1 astrocytoma cells (Fig. 3). Consistent with our hypothesis, cells co-expressing the P2Y11A87T mutant receptor and the P2Y1 receptor showed much lower response amplitudes to BzATP than cells co-expressing the P2Y11 wild-type receptor and the P2Y1 receptor. The response was also significantly reduced compared with cells expressing the P2Y11 or P2Y11A87T receptor alone.
The A87T mutation of the P2Y11 receptor abolishes nucleotide-induced receptor internalization in HEK293 cells and improves receptor resensitization
Receptor internalization is an important way of modifying cellular responsiveness to ligands. The P2Y1 receptor undergoes nucleotide-induced internalization (Mundell et al. 2006). Internalization of G protein-coupled receptors involves receptor phosphorylation by G protein-coupled receptor kinases and the subsequent recruitment of β-arrestins (Kohout and Lefkowitz 2003). The P2Y1 and the P2Y11 receptor internalization has been shown to depend on dynamin and to induce specifically translocation of β-arrestin 2 to the plasma membrane upon stimulation of the receptors by nucleotides (Hoffmann et al. 2008).
As a result of the interaction with the P2Y1 receptor, the P2Y11 receptor acquires the ability to undergo nucleotide-induced internalization (Ecke et al. 2008). Thus, internalization in HEK293 cells seems to be because of the formation of receptor oligomers with the P2Y1 receptor.
In our present study, HEK293 cells expressing the P2Y11A87T or P2Y11A87Y receptor did not show any internalization after treatment with ATP or BzATP (Fig. 4). Interestingly, in P2Y11A87S receptor-expressing cells, a slight ATP-induced and strong BzATP-induced receptor internalization was observed. This finding is consistent with the single-cell calcium measurements, where the P2Y11 and P2Y11A87S receptor showed similar potencies for BzATP. This indicates that some functionality, which is lost as a result of the A87T mutation, is retained with the A87S mutation.
As P2Y11A87T receptors are not removed from the cell surface, the mutation exerts its major influence on the long-term activity of cells facing high levels of extracellular ATP. The potency of ATP for triggering intracellular calcium signals in HEK293 cells expressing the P2Y11A87T receptor compared with the P2Y11 receptor was only slightly reduced (Table 1).
We therefore investigated the levels of [Ca2+]i after 30-min sustained exposure to ATP to induce receptor internalization and/or desensitization (Fig. 5). In HEK293 cells expressing the P2Y11 receptor, we detected a significant reduction in the intracellular calcium level in the course of 30-min ATP treatment. After a 60-min recovery period, a 1-min ATP pulse was used to determine the extent of the recovery of the calcium response. The cells showed only a slightly recovered calcium response probably because of the internalization of the P2Y11 receptors. This long-term desensitization reflects the physiological situation in cells facing high extracellular ATP concentrations.
HEK293 cells expressing the P2Y11A87T receptor showed a lower intracellular calcium level after 30 min ATP treatment than P2Y11 receptor-expressing cells. Considering the impaired internalization of the mutant receptor, this reduction could be a result of a compensating desensitization. On the other hand, the recovery of the calcium response after 90 min was 2.4 times higher than in P2Y11 receptor-expressing cells. HEK293 cells expressing the P2Y11A87S receptor showed only slight ATP-induced receptor internalization. Accordingly, these cells had a sensitization pattern similar to cells expressing the P2Y11A87T receptor. Therefore, we propose that the impact of the A87T mutation of the P2Y11 receptor relies mainly on the long-term kinetics of the responses of cells/tissues challenged with high ATP concentrations.
A mutation in the 3′ UTR of the P2RY11 gene impairs immune cell functions by reduced P2Y11 receptor expression levels (Kornum et al. 2011). Therefore, the deleterious implications are possibly caused by altered long-term responsiveness of immune cells.
Finally, the detrimental effects of the A87T SNP can be seen only when the P2Y11 receptor is co-expressed with P2Y1 receptors. P2Y1–P2Y11 receptor interaction was reported before (Ecke et al. 2008). In conclusion, future investigations of immune- and autoimmune-related diseases in brain, like narcolepsy, therefore should explore P2Y1–P2Y11 receptor interactions.
Acknowledgments and conflict of interest disclosure
Studies were supported by Deutsche Forschungsgemeinschaft.
All experiments were conducted in compliance with the ARRIVE guidelines. The authors have no conflict of interest to declare.