Specific role of N-methyl-D-aspartate (NMDA) receptor in elastin-derived VGVAPG peptide-dependent calcium homeostasis in mouse cortical astrocytes in vitro

Under physiological and pathological conditions, elastin is degraded to produce elastin-derived peptides (EDPs). EDPs are detected in the healthy human brain, and its concentration significantly increases after ischemic stroke. Both elastin and EDPs contains replications of the soluble VGVAPG hexapeptide, which has a broad range of biological activities. Effects of VGVAPG action are mainly mediated by elastin-binding protein (EBP), which is alternatively spliced, enzymatically inactive form of the GLB1 gene. This study was conducted to elucidate the activation and role of the N-methyl-D-aspartate receptor (NMDAR) in elastin-derived VGVAPG peptide-dependent calcium homeostasis in mouse cortical astrocytes in vitro. Cells were exposed to 10 nM VGVAPG peptide and co-treated with MK-801, nifedipine, verapamil, or Src kinase inhibitor I. After cell stimulation, we measured Ca2+ level, ROS production, and mRNA expression. Moreover, the Glb1 and NMDAR subunits (GluN1, GluN2A, and GluN2B) siRNA gene knockdown were applied. We found the VGVAPG peptide causes Ca2+ influx through the NMDA receptor in mouse astrocytes in vitro. Silencing of the Glb1, GluN1, GluN2A, and GluN2B gene prevented VGVAPG peptide-induced increase in Ca2+. Nifedipine does not completely reduce VGVAPG peptide-activated ROS production, whereas MK-801, verapamil, and Src inhibitor reduce VGVAPG peptide-activated Ca2+ influx and ROS production. These data suggest the role of Src kinase signal transduction from EBP to NMDAR. Moreover, the VGVAPG peptide affects the expression of NMDA receptor subunits.

ROS measurement. After 3-h exposure of mouse astrocytes to 10 nM VGVAPG peptide, we observed an increase in ROS production by 48.70% compared to the control. VVGPGA does not stimulate ROS production in astrocytes. MK-801 alone reduced ROS production by 16.36%, compared to control. Interestingly, nifedipine alone increased ROS production by 27.53%, compared to control. MK-801, verapamil, and Src kinase inhibitor I reduced ROS production stimulated by VGVAPG peptide to vehicle control levels. Nifedipine slightly but significantly reduced ROS production stimulated by the VGVAPG peptide by 19.65% as compared to the VGVAPG peptide-stimulated group ( Fig. 2A).
After 6 h of astrocyte exposure to VGVAPG peptide, we observed an increase in ROS production by 60.28% compared to control. VVGPGA does not stimulate ROS production in astrocytes. MK-801, verapamil, nifedipine, and Src kinase inhibitor I reduce ROS production stimulated by the VGVAPG peptide (Fig. 2B). ca 2+ level and ROS production in astrocytes after gene silencing procedure. Ca 2+ level. In cells transfected with scrambled siRNA, 15-min exposure to 10 nM VGVAPG peptide increased Ca 2+ influx by 30.16% compared to control. The Glb1 gene-silencing procedure protects cells by an increase in Ca 2+ influx. Likewise, silencing of NMDAR subunits (GluN1, GluN2A, and GluN2B) protects cells by an increase in Ca 2+ influx (Fig. 3A).
ROS production. In cells transfected with scrambled siRNA, 3-h exposure to 10 nM VGVAPG peptide increased ROS production by 34.65% compared to the control. The GluN1, GluN2A, and GluN2B gene-silencing procedure protects cells by increasing ROS production (Fig. 4A).
After 6-h exposure to VGVAPG peptide, we obtained results similar to those that were previously described. VGVAPG peptide increases ROS production by 40.16% compared to the control. The GluN1, GluN2A, and GluN2B gene-silencing procedure protects cells by increased ROS production in astrocytes (Fig. 4B). mRNA expression of GluN1, GluN2A, and GluN2B NMDAR subunits. In cells transfected with scrambled siRNA, 6-h exposure to 10 nM VGVAPG increased mRNA expression of the GluN2A NMDAR subunit by 21.66%. Silencing of the Glb1 gene protects cells through the VGVAPG peptide, and we did not observe changes in mRNA expression of the GluN1 and GluN2A genes. However, silencing of the Glb1 gene causes 10 nM VGVAPG to induce an increase in mRNA expression of the GluN2B gene (increase by 24.16% compared to control; Fig. 5).

Discussion
Our experiments are the first to show that the VGVAPG peptide increases Ca 2+ influx in to mouse primary astrocytes in vitro. The experiments showed that MK-801 (a noncompetitive antagonist of the NMDAR) and Src kinase inhibitor I completely prevent Ca 2+ influx after VGVAPG stimulation. Calcium-channel antagonists are a heterogeneous group of compounds. Nifedipine acts primarily on the smooth muscle cells of blood vessels, and has little effect on the heart. On the other hand, verapamil is cardioselective, and mainly affects the heart by blocking voltage-gated calcium channels 37 . Therefore, we chose nifedipine and verapamil for further research. Moreover, our study shows that nifedipine and verapamil prevent Ca 2+ influx stimulated by the VGVAPG peptide in mouse astrocytes in vitro, although not as strongly as MK-801 or Src kinase inhibitor. Moreover, after silencing of both the Glb1 gene and NMDAR subunits (GluN1, GluN2A, and GluN2B genes), the VGVAPG peptide does not activate Ca 2+ influx. The data we obtained suggest that the activation of EBP results in opening of the NMDAR and other types of calcium channels are less often blocked by nifedipine and verapamil. To date, it is well documented that EDPs and/or the VGVAPG peptide initiate Ca 2+ influx in different cell types such as HUVEC, SMC, fibroblasts, and gliomas [23][24][25][26]38 . Moreover, numerous reports have shown that the EDP-activated influx of Ca 2+ into cells requires EBP activation [23][24][25]38 . Some authors have suggested that G proteins and c-Src as well as ERK1/2 or MEK1/2 kinases are involved in signal transduction, although the molecular mechanism mediating the opening of Ca 2+ channels remains unclear 36,38,39 . Each point represents the mean ± SD of three independent experiments, each of which comprised six replicates per treatment group. *P < 0.05, **P < 0.01, and ***P < 0.001, versus the control cells. ### P < 0.001, versus the VGVAPG peptide-treated group.
The next step of our study was to determine ROS production as a consequence of the influx of Ca 2+ ions into cells. We found that stimulation of primary astrocytes by the VGVAPG peptide increased ROS production in both of the time intervals studied. MK-801, verapamil, and Src inhibitor prevent ROS production that was stimulated by the VGVAPG peptide. Similarly, after silencing of NMDAR subunits (GluN1, GluN2A, and GluN2B genes), the VGVAPG peptide did not cause ROS production. Interestingly, nifedipine does not completely prevent the ROS production stimulated by the VGVAPG peptide. Our results suggest that the binding of the VGVAPG peptide to EBP leads to increased ROS production, which mainly could be NMADR dependent. To date, reports concerning ROS production after EDP and/or VGVAPG stimulation are inconsistent. Our previous studies have shown that the VGVAPG peptide increases ROS production in an EBP-dependent manner in mouse primary astrocytes in vitro 32 . Similarly, an increase in ROS production after EDP stimulation has been reported in monocytes, human fibroblasts, and a neuroblastoma (SH-SY5Y) cell line 29,30,33 . Moreover, EDPs have been described to enhance the activities of antioxidant enzymes, such as superoxide dismutase (SOD), catalase (CAT), or glutathione peroxidase (GPx), and increase lipid peroxidation in human fibroblasts 40 . Furthermore, the VGVAPG peptide increases the expression and activity of GPx in SH-SY5Y cells 33 . On the other hand, the VGVAPG peptide reduced ROS production in neutrophils in controls and patients with stable chronic obstructive pulmonary disease 28 . www.nature.com/scientificreports www.nature.com/scientificreports/ Phosphorylation regulates many cellular processes, including protein activity, localization, mobility, and protein-protein interactions. Direct phosphorylation of ionotropic glutamate receptors is a key mechanism that regulates channel function and receptor localization at synapses 41,42 . The proto-oncogene tyrosine-protein kinase Src is an important modulator of NMDARs. It has been described that the NR2A, NR2B, and NR2D subunits can be tyrosine phosphorylated 41 . Moreover, phosphorylation of the abovementioned subunits results in increased NMDAR activity 43 . Our data are the first to show that, in astrocytes in vitro, the Src kinase inhibitor I prevents Ca 2+ influx and ROS production stimulated by the VGVAPG peptide. These data suggest a role of Src kinase in signal transduction from EBP to NMDAR and, probably, in other calcium channels.
It is well known that, in cells of the nervous system, the NMDAR is one of the most important ion channels [19][20][21] . NMDAR are responsible for Ca 2+ influx and cellular ROS production that can result in excitotoxic cell death. However, our previous studies have shown that VGVAPG is nontoxic, and does not activate the apoptotic process in mouse astrocytes in vitro 44 . Therefore, we decided to determine the involvement of NMDAR subunits in the mechanism of action of VGVAPG in astrocytes in vitro. Our experiment with silencing of the GluN1, GluN2A, and GluN2B genes confirmed the experiments with the NMDAR antagonist MK-801. Moreover, we determined that the VGVAPG peptide affects the mRNA expression of GluN2A NMDAR subunits. In cells transfected with scramble siRNA, 10 nM VGVAPG peptide increased the mRNA expression of the GluN2A NMDAR subunit. On the other hand, after the silencing of the Glb1 gene, we did not observe changes in the mRNA expression of GluN2A genes. However, silencing of the Glb1 gene ensured the 10 nM VGVAPG could increase the mRNA expression of the GluN2B gene. The proportion of the GluN2B and GluN2A subunits changes with age and indicates the maturation of the nervous system 20,45 . In fetal life, strong expression of the GluN2B subunit is observed whereas, after birth, this expression decreases in favor of the GluN2A subunit 46,47 . However, a heterotetramer containing GluN2B subunits is mainly responsible for excitotoxicity and the intensity of neuronal apoptosis, whereas the opposite effect is caused by the presence of GluN2A subunits promoting cell survival 48,49 . Our data suggest that the VGVAPG peptide changes the NMDAR expression profile to one similar to the adult profile. Increased expression of GluN2A can promote cell survival and supports our hypothesis postulated in our other papers that the VGVAPG peptide promotes regeneration and the healing process in astrocytes 32,44,50 . However, after Glb1 silencing, the NMDAR expression profile changed to similar to an embryonic one, and could promote cell apoptosis. Furthermore, these data suggest the involvement of other EDP receptors.

Conclusion
Our study shows, for the first time, that the VGVAPG peptide causes Ca 2+ influx through NMDAR in mouse astrocytes in vitro. Silencing of the Glb1, GluN1, GluN2A, and GluN2B genes prevented a VGVAPG peptide-induced increase in the Ca 2+ level. Nifedipine does not completely reduce VGVAPG peptide-activated ROS production, whereas MK-801, verapamil, and Src kinase inhibitor I reduce VGVAPG peptide-activated Ca 2+ influx and ROS production. There appears to be a potential role of Src kinase in signal transduction from EBP to NMDAR. Moreover, the VGVAPG peptide induced an increase in the expression of GluN2A NMDAR subunits, which promote cell survival in adult life. However, due to the lack of sufficient data explaining the molecular mechanism of action of the VGVAPG peptide in the nervous system, more studies on this topic are necessary.
Astrocyte-enriched cell culture. All experiments were conducted on mouse astrocyte cell cultures, which were prepared from the fetuses of pregnant female Swiss mice according to a previously described method that generates an almost pure culture of astrocytes (>98% astrocytes) 44 (supplementary data). Cells were cultured in DMEM/F12 1:1 without phenol red that was supplemented with 10% FBS, 100 U/mL penicillin, 0.10 mg/mL streptomycin, and 250 ng/mL amphotericin B, maintained at 37 °C in a humidified atmosphere containing 5% CO 2 . The astrocytes were seeded at a density of 5 × 10 5 per well in a 96-well plate, for colorimetric and fluorometric analysis, and 60 × 10 5 per well in a 12-well plate for mRNA analysis. The culture medium was changed prior to the experiment with the VGVAPG peptide and a tool compound was selected for this study.
siRNA gene-silencing procedure. Glb1, GluN1, GluN2A, and GluN2B siRNA were used to inhibit the expression of EBP and NMDAR subunits in mouse primary astrocytes. The experiment was conducted in accordance with a previously described procedure 51 . Briefly, appropriate siRNA was applied for 7 h at a final concentration of 50 nM in antibiotic-free medium containing the siRNA transfection reagent INTERFERin. After transfection, the astrocytes were cultured for 12 h prior to commencing the experiment. Vehicle controls included positive siRNA and scramble siRNA containing a scrambled sequence that did not lead to specific degradation of any known cellular mRNA. Knockdown of the Glb1 gene was estimated at 39% of the control mRNA and 60.05% of the control protein as previously described 44,50 . Knockdown of the GluN1, GluN2A, or GluN2B genes were estimated at 22.02, 25.21 and 24.32%, respectively of the control mRNA. ca 2+ determination in astrocytes. To measure the calcium level, we used the Fluo-3/AM indicator, which exhibits increased fluorescence emission intensity on binding to Ca 2+ and, therefore, is used to detect increases in the level of intracellular calcium 52 . Briefly, a final concentration of 4 μM Fluo-3/AM dye and 0.04% Pluronic F-127 in the loading medium (DMEM/F12 without FBS) was used for 60 min before compound treatment. Thereafter, the cells were washed three times with 200 μL PBS without Ca 2+ supplemented with HEPES. After 15 and 30 min of incubation of the cells with 10 nM VGVAPG or VVGPGA peptide (5% CO 2 at 37 °C), the fluorescence of Fluo-3 was measured with a microplate reader (FilterMax F5) at maximum excitation and emission spectra of 485 and 535 nm, respectively. ROS measurement. The fluorogenic dye H 2 DCFDA was used to detect intracellular ROS according to a previously described method 53 . Briefly, a final concentration of 5 μM H 2 DCFDA was used. The cells were incubated in H 2 DCFDA in a serum-free and phenol red-free DMEM/F12 medium for 45 min before compound treatment. After 3-and 6-h incubation of the cells with 10 nM VGVAPG or VVGPGA peptide (5% CO 2 at 37 °C), the culture medium was replaced with fresh medium to remove the extracellular residual DCF and the studied compound to reduce the fluorescence background. Cells treated with 0.3% hydrogen peroxide (H 2 O 2 ) were used as a positive control (results not shown). DCF fluorescence was measured with a microplate reader (FilterMax F5) at maximum excitation and with emission spectra of 485 and 535 nm, respectively.
Real-time PCR analysis of mRNAs specific to genes encoding GluN1, GluN2A, and GluN2B. The experiment was conducted through a previously described procedure 54 . For the real-time PCR assay, astrocytes were seeded onto 12-well plates and initially cultured for 24 h. After 6-h exposure to 10 nM VGVAPG peptide, the samples were collected, and total RNA was extracted from the astrocytes using a Qiagen RNeasy Mini Kit according to the manufacturer's instructions. Both the quality and quantity of the RNA were determined spectrophotometrically at 260 and 280 nm, respectively (ND/1000 UV/Vis; Thermo Fisher NanoDrop, USA). Two-step real-time reverse transcription (RT)-PCR was conducted with both the RT reaction and the quantitative PCR (qPCR) run using the CFX Real Time System (BioRad, USA). The RT reaction was carried out at a final volume of 20 μL with 180 ng RNA (as a cDNA template) using the cDNA reverse transcription kit in accordance with the manufacturer's instructions. Products from the RT reaction were amplified using the FastStart Universal Probe Master (Rox) kit with TaqMan probes as primers for specific genes encoding Actb, Glb1, GluN1, GluN2A, and GluN2B. Amplification was carried out in a total volume of 20 μL containing 1× FastStart Universal Probe Master (Rox) and 1.5 μL of the RT product, which was used as the PCR template. Standard qPCR procedures were carried out as follows: 2 min at 50 °C and 10 min at 95 °C, followed by 45 cycles of 15 s at 95 °C and 1 min at 60 °C. The threshold value (Ct) for each sample was set during the exponential phase, and the ΔΔCt method was used for data analysis. Actb was used as the reference gene. Random-peptide-sequence VVGPGA was used as control and did not affect GluN1, GluN2A, and GluN2B mRNA expression (data not shown).