711–717 on-line at: www.actabp.pl

In our earlier studies of the signaling cross-talk between nucleotide receptors in an in vitro glioma model (C6 cell line) under prolonged serum deprivation conditions, a growth arrest of the cells and expression shift from P2Y(1) to P2Y(12) receptors was found. The aim of the present work was to test if siRNA silencing of P2Y(1) receptor changes P2Y(12) expression similarly as following the serum deprivation and which physiological downstream pathways it affects. Here we demonstrate for the first time the efficiency of siRNA technology in silencing P2Y nucleotide receptors in glioma C6 cell line. Moreover, P2Y(12) proved to be insensitive to the P2Y(1) receptor silencing. The effect of the P2Y(1) silencing on calcium signaling was less pronounced then the extent of the protein change itself, exactly as was the case for the serum starvation experiments. Phosphorylation of ERK and Akt kinases were studied as the downstream effect of P2Y(1)-evoked signaling and similar effects as in the case of serum deprivation were found for ERK, and even stronger ones for Akt phosphorylation.


INTRoduCTIoN
Signaling between the cell environment and the cytoplasm is fundamental for the ability of life to adapt to the changing conditions as well as for of the regulation of the behavior of individual cells by the organism.In the cytoplasm, the signal is transmitted by so called secondary messengers, among which calcium ions are one of the most ubiquitous and versatile ones (Rasmussen et al., 1976).
The rat glioma C6 cell line was used in this work as an in vitro model of primary glial tumor cells.The cell line belongs to a non-excitable cell group (Baranska et al., 1995) and is derived from chemically-induced rat brain tumor (Benda et al., 1968).Glioma C6 shares several properties of astrocytic as well as oligodendrocytic progenitors and is often used as a biochemical model for studies related to astrocytes (Brismar, 1995).When injected into rodent brain these cells become morphologically similar to glioblastoma multiforme (Auer et al., 1981) and therefore they are used in both in vivo and in vitro studies of this kind of tumor (Jacobs et al., 2011).The expression and activity of several P2Y receptors: P2Y 1 , P2Y 2 , P2Y 4 , P2Y 6 , P2Y 12 , P2Y 13 and P2Y 14 in glioma C6 has been well-documented in the literature (for review see: Wypych & Baranska, 2013).
In the past, several studies performed in Jolanta Baranska's laboratory, in which the authors of the present paper also participated, showed interesting crosstalk between G q -and G i -dependent pathways in the physiology of glioma C6 cells.This phenomenon was revealed by long-term serum starvation which affected the level of both P2Y 1 and P2Y 12 receptors protein.While the P2Y 1 protein level as well as the calcium signal evoked by its activation decreased during serum starvation, the level of P2Y 12 receptor rose (Krzeminski et al., 2007).Those changes were interpreted as a shift from proliferative to pro-survival behavior under the unfavorable conditions of long-time serum deprivation.
The calcium signal evoked by G q -coupled receptors, including P2Y nucleotide receptors, consists of two components and is preceded by activation of phospho-lipase Cβ and production of IP 3 followed by IP 3 action on its receptor (IP 3 R) on the endoplasmic reticulum (ER).This in turn leads to calcium release from ER cisternae (intracellular calcium stores) via IP 3 R.In SOCE (Store Operated Calcium Entry) the emptying of the ER cisternae is followed by opening of a special class of channels in the plasma membrane, so called store operated calcium channels (SOCs), and calcium flows from the cell environment into the cytoplasm.This influx is also known as capacitative calcium entry.Thus, there are two sources of calcium signal: intracellular calcium stores and the extracellular space.
A study of calcium signaling of serum-deprived glioma C6 cells revealed that the strength of the P2Y 1 receptor evoked calcium signal did not always correlate with the receptor protein level, but the observed phenomenon was explained by a compensating activity of P2Y 12 receptor (Suplat et al., 2007).
Up to now there are no literature data about the regulation of transcription of P2RY1 and P2RY12 genes.To test the hypothesis that a decreased P2Y 1 expression enhances the expression of P2Y 12 receptor, we decided to silence the P2Y 1 receptor by siRNA and to test the effect of that on P2Y 12 protein level and the ability of the cells for nucleotide-dependent calcium signaling.The RNAi technology has proven to be an efficient tool to suppress the expression of targeted genes and, although never used earlier to target P2Y nucleotide receptors in glioma cells, it proved to be quite efficient in knockingdown P2X 7 receptor (Deli et al., 2007).
P2Y 1 receptors silencing with siRNA.In all experiments a set of four dsRNAs selected for four regions of rat P2RY1 mRNA (RefSeq NM_012800) was used (Dharmacon, Thermo Scientific, USA).Their sequences are shown in Table 1.Concentrations of siRNA were determined at 260 nm.
Glioma C6 transfection.Automatic electroporation system Amaxa (Lonza, USA) was used for cell transfection and only passages under 30 were used.Briefly, after non-enzimatic detachement (Cell Dissociation Solution, Sigma-Aldrich, Germay) 1.5 × 10 6 cells were used for one reaction.After centrifugation (200 × g for 10 minutes) the cells were resuspended in 100 µl of electroporation buffer (Cell Line Nucleofector Solution V, Lonza, USA).Cells were electroporated using a survival-promoting program in a Nucleofector II electroporator (Lonza, USA).For each electroporation from 0.25 to 3 µg of siRNA was used.Immediately after electroporation cells were conditioned in 500 µl of fresh culture media and then cultured appropriately for further experiments.As a positive control of electroporation, 2.5 µg of GFPcoding plasmid was used (pmaxGFPTM, Lonza, USA).As negative controls, cells were either incubated in the presence of 3 µg siRNA without electroporation or electroporated in electroporation buffer with no siRNA.The next day, transfection efficiency was calculated as a ratio between the number of GFP-expressing cells and all visible cells observed in at least 5 fields.A Nikon Diaphot microscope equipped with a 10× 0.5NA dry LWD lens was used using overlay of fluorescence and DIC Nomarski and the images were registered with a Retiga 1300 cooled digital camera (QImaging, Canada) controlled by the µManager program (Edelstein et al., 2010).
To evaluate changes in the amounts of P2Y 1 and P2Y 12 receptors, membranes were first detected for P2Y 1 followed by β-actin and then for P2Y 12 with stripping between different antibodies.The same procedure was used for phospho-Erk and phospho-Akt.
Stripping buffer contained 0.2 M glycine, 0.1% SDS and 0.01% Tween 20.After two rounds of stripping (7 minutes each), membranes were washed twice for 10 minutes in PBS and twice for 5 minutes in PBS-T, followed by blocking and hybridization as above.The specificity of the anti-P2Y 1 and anti-P2Y 12 antibodies was confirmed as in (Krzeminski, et al., 2007) by incubation of the primary antibodies with antigen peptide provided by the manufacturer.A 1:1 antigen to antibody mixture was pre-incubated for 1 h (with agitation) in blocking buffer and then membrane was processed as for primary antibody.
Multiple bands visible on all P2Y 1 blots are common problem for most P2Y receptor detection experiment.As discussed earlier, it is not nonspecific staining but the result of several glycosylation forms of the same receptor present in the cell.Declycosylation of cell extract by N-glycosidase F results in one band on the western blot (Krzeminski et al., 2008).
Analysis of the immunodetected proteins.Western blots were imaged on X-Ray film (Kodak), scanned and then densitometry was performed with ImageJ program plugin (Schneider et al., 2012).Signal intensity was normalized first to β-actin than to that in control conditions and the values shown are mean of at least three experiments ±SD.In order to determine statistical significance of differences nonparametric Mann-Whitney U test was used.The significance is labeled (*) for p < 0.05, (***) for p < 0.001 and a lack of statistical significance is presented as (-).
Immunostaining of P2Y 1 receptors for wide field and TIRF microscopy.Cells were grown on TIRFcompatible, high precision Nº 1.5 coverslips (Paul Marienfeld GmbH&Co.KG, Germany).After washing twice with PBS, cells were fixed in cold (4ºC) acetone for 3 min, rehydrated in PBS for at least 10 minutes at RT and then permabilized for 10 minutes with 0.1% Triton X100 (Romil Chemicals, UK).To confirm specificity of the staining, primary anti-P2Y 1 antibody was preincubated with the peptide provided by the manufacturer at a 1:1 dilution, in the same way as in the case of western blot.After careful washing, slides were incubated in 2.5% NCS in BPS (ICC blocking solution) for 1 hour to block unspecific binding and then stained overnight (4ºC) with primary anti-P2Y 1 antibody (Alomone, Israel) diluted 1:200 in ICC blocking solution.After three 10-minute washes in PBS slides were incubated with secondary goat anti-rabbit IgG antibodies conjugated with Alexa488 (Invitrogen, Life Technologies, USA) in ICC blocking solution (1:600) for 2 hours at RT.After three 10-minute final washes in PBS, specimens were mounted with VectaShield mounting medium (Vector Laboratories, UK).For evaluation of P2Y 1 receptor staining a Carl Zeiss Laser TIRF3 system was used.It contained a Carl Zeiss Axio Observer Z.1 microscope stand (Carl Zeiss, Germany) equipped with a 488 nm/100 mW argon laser for TIRF and a mercury arc lamp for wide field fluorescence microscopy.High numerical aperture planapochromat lens 100× 1.46NA was used to provide total reflection of side introduced laser beam in the coverslip and allow the TIRF mode.Images were collected with a high efficiency back-thinned CCD camera Quantem 512SC (Photometrics, Germany) for n cells from at least 5 different areas.The microscope was controlled by Ax-ioVision 4.7.2.0 software (Carl Zeiss, Germany).Ratio between images of background-corrected TIRF and wide field fluorescence images was calculated with MetaMorph 7.5.6.0 software (Danaher Corporation, USA).In graphs, the ratio TIRF/WF in control conditions is presented as 1 and relative changes (mean ± S.E.) of the parameter are shown.The imaging in each experimental condition was performed at least three times.
Calcium imaging.The experiments were performed as described before (Suplat-Wypych, et al., 2010).Briefly, before experiments cells on coverslips were washed once with PBS and once with solution containing 137 mM NaCl, 2.7 mM KCl, 1 mM Na 2 HPO 4 , 25 mM glucose, 20 mM Hepes (pH 7.4), 1 mM MgCl 2 , 1% bovine serum albumin and 2 mM CaCl 2 (standard buffer).The cells were then incubated at 37°C for 30 min in the standard buffer with 1 μM Fura-2 AM.Thereafter, the coverslips were mounted in a chamber on a Nikon Diaphot inverted-stage microscope equipped with a fluo ×40/1.3NA oil-immersion objective lens.Fura-2 digital fluorescence microscopy was used to determine changes in the intracellular calcium levels [Ca 2+ ] i (Grynkiewicz et al., 1985).
Ludl Lep MAC 5000 filter wheel system with a Chroma Fura-2 filter set was used for illumination of specimens.Images were acquired using an Andor Luca R digital EMCCD camera (Andor, Irland).Data processing was carried out using iQ Live Cell Imaging Software (Andor, Irland) and Matlab ® software (The Mathworks Inc, USA).All data are expressed as 340/380 nm-induced fluorescence of Fura-2 ratio changes against time (Δ340/380).Each experiment was repeated at least three times.MeSADP (Sigma-Aldrich, Germany) at 10 μM was used to evoke P2Y 1 -dependent calcium signal.Calcium responses against time are expressed as means of (n) cells.Changes in the calcium response were calculated as the relative changes of integrals from 100 consecutive measuring points after adding the agonist (mean ± S.D.).

The receptor protein level during serum deprivation
Previously published serum starvation experiments (Krzeminski et al., 2007) were carried out in significantly different conditions.Change of the glioma C6 cell line passage from above 40 to below 30 was forced by electroporation protocol and lead to difference in obtained results large enough to repeat receptor level measurements as a reference for further research.
Using western blot analysis, we measured relative protein content of P2Y 1 and P2Y 12 receptors in glioma C6 cells during prolonged, 96 h serum deprivation as a starting point of the study (Fig. 1).The amount of P2Y 1 receptor protein fell to less than one third of control level (10-30%, depending on the experiment).At the same time the amount of P2Y 12 receptor protein rose to about 150% of the control.Although those changes were less pronounced than the ones reported previously (Krzeminski, et al., 2007), the trends were exactly the same: the level of P2Y 1 receptor was falling, and of the P2Y 12 receptor rising and, crucially, in both series of measurements, the increase of P2Y 12 receptor level during prolonged serum starvation was three times smaller than decrease of P2Y 1 receptor level.In other words, the rise of P2Y 12 is evident, but less intense than the fall of P2Y 1 receptor.Experiments were performed four times, data pooled and processed together.

silencing of P2Y 1 receptors by siRNA
Since we wanted to evaluate the effect of progressive P2Y 1 silencing, and not just a lack of P2Y 1 protein, the transient siRNA approach was chosen.The poor efficiency of other transfection methods (around 3% after lipofectamine treatment, not shown) made us use electroporation with the Amaxa system as an effective way to deliver small RNA molecules into cells.Preliminary experiment with GFP vector (pmaxGFPTM, Lonza, USA) proved transfection efficiency above 40-70%.The best result of silencing we observed 48 h after electropo-ration with siRNA.Figure 2 shows dose dependency of P2Y 1 receptor silencing on the amount of siRNA loaded into the cells.The siRNA dose of 0.25 µg was ineffective (1.02 ± 0.15), and the effect was growing with the dose of siRNA (0.72 ± 0.47 for 0.5 µg) up to 0.32 ± 0.07 for 1 µg.With 3 µg we observed erratic results (0.45 ± 0.42), showing that the effect may not be specific any longer.It could be due to a too high (higher than recommended by the producer) dilution of the sample by siRNA.The addition of 3 µg of siRNA without electroporation or the electroporation without the siRNA in the medium did not result in any statistically significant changes of P2Y 1 receptor protein level in the cells (1.02 ± 0.25 and 1.01, respectively).Altogether it shows that the observed results are specific consequences of the siRNA activity in glioma C6 cells, and are not the result of the experimental procedure itself.For further experiments we decided to use 1 µg of siRNA.
The maximal level of silencing was observed after 48 h of silencing with 1 µg siRNA (0.32 ± 0.07 of control) what may be caused by the slow turnover of the receptor in the idle cells.To find out if the protein was still present on the plasma membrane, we decided to use TIRF (total internal reflection fluorescence) microscopy (Fig. 3).TIRF microscopy allows one to observe a 100-nm layer of the specimen close to the coverglass.With this method we evaluated the amount of the P2Y 1 receptor protein present on the plasma membrane.In contrast, in wide field (WF) fluorescent microscopy the signal is proportional to the total amount of a protein in the cell.Thus, the of ratio of fluorescence intensity in the TIRF channel to that in the wide field channel is proportional to the ratio between membrane and total pool of receptors.We measured the TIRF/WF signal ratio of immunocytochemically stained P2Y 1 receptors and found that 48 h after siRNA administration the amount of P2Y 1 receptors on the cell surface fell faster than did the total receptor protein level (calibrated ratio:

Physiological effect of P2Y 1 receptor silencingcalcium signaling
Knowing the effects of individual siRNA doses on the total P2Y 1 protein level and its level in the cell membrane, we could test the influence of both the experimental procedure and receptor knock-down on the direct signaling effect of its activation: the calcium signal in the cytoplasm.Evoked by 10 µM MeSADP, it was studied with Fura-2 ratiometry, as described in Materials and Methods.Dashed lines in figure 4 present the transient plots in control experiments while solid ones in siRNA-treated cells.It worth mentioning, that electroporation of the cells without siRNA changes the shape of the calcium signal in comparison with control, even 48 hours after electroporation.The initial peak of the signal (calcium release form ER) reached higher values and the second, CCE sustained phase, even if initially stronger, disappeared faster.A similar signal shape was seen in all siRNA-treated cells, with no CCE phase, suggesting some long-term influence of electroporation on cell membranes.To quantify the effect of P2Y 1 siRNA administration on the calcium signal strength, we decided to calculate integrals of the transients, to compare plots with slightly different shapes (Fig. 5).Contrary to the protein level effects, both electroporation and incubation with siRNA without electroporation led to small but statistically significant diminishing of the signal (0.75 ± 0.05 for cells incubated with siRNA and 0.87 ± 0.03 for cells electroporated in absence of siRNA, both as fraction of control signal).Surprisingly, we did not observe such an effect after electroporation with low dose (0.25 µg) of siRNA (1.04 ± 0.04).The effect of the siRNA doses, which significantly changed the protein level was several times stronger.Surprisingly, we did not observe the dose-dependency as in the protein level (0.24 ± 0.03 for 0.5 µg of siRNA and 0.27 ± 0.02 for 1 µg of siRNA).

Physiological effect of P2Y 1 receptor silencingdownstream effects
We also studied the level of phosphorylation of ERK1⁄2 (p44⁄p42) and Akt, well known targets of the calcium signal in glioma C6 cells.Phosphorylation of both proteins was studied up to 120 min after induction of cells by 10 µM 2MeSADP (Fig. 6).Control cells (black bars) were compared with those electroporated with 1 µg of siRNA silencing the P2Y 1 receptor (grey bars).While Erk1/2 phosphorylation was only partially (more than two times) diminished, Akt phosphorylation was practically blocked.

Effect of P2Y 1 receptor knock-down by siRNA on P2Y 12 receptor protein level
The main goal of this work was to find out if the increase of the P2Y 12 protein level observed during serum deprivation of C6 glioma cells was a simple compensatory effect of the P2Y 1 receptor level fall.To evalu- Signals presented in Fig. 4 were integrated.Both negative controls groups (cells incubated with 3 µg siRNA without electroporation and electroporation in absence of siRNA) cause slight, but statistically significant inhibition of the calcium signal.Such an effect is not visible in cells electroporated with low (0.25 µg) dose of siRNA.Higher doses (0.5 and 1 µg) reduced signal very strongly, however the effect is not dose dependent.

dIssCusIoN
This study clearly shows that siRNA silencing of P2Y nucleotide receptors is a promising experimental approach.The siRNA knock-out achieved in the present work was efficient but not complete.It is better than that reported by Deli et al. (2007) for the P2X 7 receptor in melanoma cells (about 50% of downregulation) and similar to the one recently reported by Crooke et al. (2009) and Martin-Gil et al. (2012) both on P2Y 2 receptor.This level of efficiency is probably a result of two causes: the siRNA delivery efficiency and the relatively slow receptor protein turnover.The latter possibility is supported also by our data from TIRF imaging, showing gradual transfer of the P2Y 1 receptors from ventral plasma membrane into the cytoplasm.
The observed physiological effects of inhibition of Phosphorylation of Erk1/2 and Akt kinases after silencing of P2Y 1 receptors follows the Sak and Illes postulate that a general property of neuronal malignant cells appears to be the absence of P2Y 1 receptor and inhibition of PI3-K⁄Akt signaling (Sak & Illes, 2005).In the present study the effect is much more pronounced than in the serum deprivation paper, where Akt signaling was strongly affected after long-term experiment (Krzeminski, et al., 2007).Moreover, our previous work clearly shows that P2Y 12 receptors have a positive, modulatory influence (via cAMP-dependent pathways) on the P2Y 1dependent calcium signaling (Suplat, et al., 2007).However, the mechanism proposed by Hardy et al. (2004) for platelets does not seem to depend strongly on the P2Y 12 receptor protein level itself, even if the activation of this receptor strongly affects the cAMP level in glioma C6 cells via G i protein (Van Kolen & Slegers 2004).Moreover, the P2Y 1 -evoked calcium signal observed in the present work in glioma C6 cells was much less sensitive to the siRNA dose than the receptor protein level itself.Those discrepancies between the receptor protein level and the signaling efficiency can, however, have much deeper roots.
Several years ago it was discovered that another nucleotide receptor, P2Y 2 , can interact with integrins, changing the signaling activity after entering into complex with α v β 5 integrins (Bagchi et al., 2005) and changing its G protein α subunit specificity from G q to G 12 (Liao et al., ).This change switches the signal from the receptor and instead of inducing the calcium transient the receptor directly activates RhoA-dependent signaling pathway (Suzuki et al., 2003).This quite unique double specificity of GPRS receptors to G q and G 12/13 subunits of heterotrimeric G proteins may lead to the situation in which the level of receptor protein does not exactly correlate with the signal from this receptor.Even if the ability to bind integrins was described only for ATP-induced P2Y 2 receptors, notions of integrin-dependent ADP nucleotide signaling appeared recently (O'Brien et al., 2012) as well as the P2Y 12 -dependent integrin activity influencing microglia cell spreading under the ADP stimulation (Ohsawa et al., 2010).
Summarizing, this paper shows that nucleotide signaling is an interplay of two proteins more complicated than previously postulated and the cross-talk between receptors may be very indirect and finely modulated by the cell structure and function.

Figure 2 .
Figure2.silencing of P2Y 1 receptor by siRNA P2Y 1 protein level two days after electroporation of glioma C6 cells with specific siRNA.Densitometric analysis of the western blot, from left: control, cells incubated with specific siRNA without electroporation, cells electroporated in the control medium, without siRNA, growing dose of siRNA: 0,25 µg, 0,5 µg, 1 µg and 3 µg.The dose-dependent activity of specific siRNA can be observed from 0,25 µg to 1 µg dose.Presented blots come from two representative experiments, β-actin as housekeeping protein reference.
Figure 3. distribution of P2Y 1 receptor after prolonged siRNA silencing Glioma C6 cells were stained with anti-P2Y 1 antibody and visualized both in wide field (WF) and total internal reflection fluorescence (TIRF) microscopy.(A) (upper row) Fluorescence image of control C6 cell stained with antibody against P2Y 1 receptor.The same cell pictured in the TIRF mode (lower row).Left column: control cells, right column: cells with P2Y 1 receptors silenced by1 µg dose of siRNA.(B) Bars (mean ± S.E.) present the ratio between TIRF and WF fluorescence intensity, proportional to the ratio between internalized and membrane fraction of receptor.

Figure 1 .
Figure 1.Effect of long-term starvation on P2Y 1 and P2Y 12 receptors protein level (A) Densitometric analysis of western blot.Four independent experiments were performed.(B) Typical result of P2Y 1 receptor protein detection, here shown control, 48h, 72 h and 96h of serum-deprivation time points.Lower row: β-actin as housekeeping protein reference.(C) Typical example of P2Y 12 receptor protein detection, here shown control, 48 h, 72 h and 96 h time points of serum deprivation.Lower row: β-actin as housekeeping protein reference.

Figure 4 .Figure 5 .
Figure 4. siRNA silencing of P2Y1 receptor -effect on calcium signaling Glioma C6 cells were loaded with Fura-2 AM and signal evoked by 10 µM 2MeSADP in the presence of 2 mM CaCl 2 was studied.Thick grey plot shows control cells without any treatment (n = 267).Dashed plots present calcium signal of the transfection control experiments: medium grey -cells incubated with 1µg siRNA but not electroporated (n = 282) and light grey -cells electroporated in absence of siRNA (n = 412).Note that calcium signal of the electroporated cells has stronger ER release phase and shorter capacitative phase.All plots present the mean response from n number of cells as indicated in brackets.Solid plots present calcium signal of the cells with P2Y 1 protein knock-down by siRNA: black -0.25 µg (n = 213), medium grey -0.5 µg (n = 254), light grey -1 µg (n = 187).siRNA doses which earlier significantly lowered P2Y 1 receptor protein level also strongly inhibited calcium signal.

Figure 7 .Figure 6 .
Figure 7. siRNA silencing of P2Y 1 receptor -effect on P2Y 12 receptor protein level Glioma C6 cells content of P2Y 12 protein level after electroporation with 1 µg of specific siRNA.Densitometric analysis of western blot of three independent experiments.From left: control, cells incubated with 3 µg of siRNA without electroporation, cells electroporated in the eletroporation buffer, without siRNA, growing dose of siRNA: 0,25 µg, 0,5 µg, 1 µg and 3 µg.As, shown in Fig.1, the dose dependent activity of specific siRNA was observed on P2Y 1 protein level between 0,25 µg and 1 µg.There is no visible effect of this inhibition on the observed level of P2Y 12 protein level.Presented blots come from two representative experiments, β-actin as housekeeping protein reference.