These authors contributed equally to this work and share first authorship
This article was submitted to Vascular Physiology, a section of the journal Frontiers in Physiology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Vascular calcification (VC) is associated with a number of cardiovascular diseases, as well as chronic kidney disease. The role of smooth muscle cells (SMC) has already been widely explored in VC, as has the role of intracellular Ca2+ in regulating SMC function. Increased intracellular calcium concentration ([Ca2+]i) in vascular SMC has been proposed to stimulate VC. However, the contribution of the non-selective Piezo1 mechanosensitive cation channels to the elevation of [Ca2+]i, and consequently to the process of VC has never been examined. In this work the essential contribution of Piezo1 channels to arterial medial calcification is demonstrated. The presence of Piezo1 was proved on human aortic smooth muscle samples using immunohistochemistry. Quantitative PCR and Western blot analysis confirmed the expression of the channel on the human aortic smooth muscle cell line (HAoSMC). Functional measurements were done on HAoSMC under control and calcifying condition. Calcification was induced by supplementing the growth medium with inorganic phosphate (1.5 mmol/L, pH 7.4) and calcium (CaCl2, 0.6 mmol/L) for 7 days. Measurement of [Ca2+]i using fluorescent Fura-2 dye upon stimulation of Piezo1 channels (either by hypoosmolarity, or Yoda1) demonstrated significantly higher calcium transients in calcified as compared to control HAoSMCs. The expression of mechanosensitive Piezo1 channel is augmented in calcified arterial SMCs leading to a higher calcium influx upon stimulation. Activation of the channel by Yoda1 (10 μmol/L) enhanced calcification of HAoSMCs, while Dooku1, which antagonizes the effect of Yoda1, reduced this amplification. Application of Dooku1 alone inhibited the calcification. Knockdown of Piezo1 by siRNA suppressed the calcification evoked by Yoda1 under calcifying conditions. Our results demonstrate the pivotal role of Piezo1 channels in arterial medial calcification.
Vascular calcification (VC) is defined as a deposition of calcium-phosphate complexes in large and medium sized blood vessels. The process is highly correlated with several cardiovascular diseases, depending on the organs affected and the degree of progress.
Mineral deposition can affect both the intimal and medial layers of blood vessels. Calcification occurring in the intimal layer of arteries (AIC), which is normally composed of endothelial cells and subendothelial connective tissue, is associated with atherosclerotic lesions (
The calcification of the medial layer (AMC (arterial medial calcification), or Monckeberg’s sclerosis), which consists of SMCs and the elastin-rich extracellular matrix, is not necessarily associated with lipid accumulation and inflammatory cell invasion, but other factors mentioned above in AIC are known drivers of AMC as well. Furthermore, cell senescence is also suggested to initiate AMC. AMC leads to vascular stiffness, systolic hypertension and increases the prevalence of diastolic dysfunction, increased afterload, and heart failure (
Whichever case is examined, partially overlapping mechanisms that were proposed to induce calcification all implicate the involvement of SMCs of the medial layer. It is now generally accepted that VC, although previously thought to be a passive, is an active process whose regulation is highly dependent on the SMCs. While the contractile phenotype contributes significantly to the physiological function of the cardiovascular system, and the maintenance and remodelling of the extracellular matrix, upon pathological signals the phenotypic plasticity of SMCs makes them capable of transforming into other phenotypes including adipocytic, senescent, foam cell, synthetic and osteochondrogenic (
Since the identification of Piezo channels over the past decade (
Here we propose that the mechanosensitive Piezo1 channels expressed on the SMCs of large systemic arteries have pivotal role in VC. They are overexpressed on calcifying SMCs, and, furthermore, their activation augments, while their inhibition or the downregulation of their expression hinders AMC.
Human aortic SMCs (HAoSMC; 354-05; Cell Applications Inc., San Diego, CA, United States) were maintained in growth medium (GM) that was prepared by supplementing Dulbecco’s Modified Eagle Medium (DMEM, D6171, Sigma) with 10% FBS (10270-106, Gibco, Grand Island, NY, United States), antibiotic antimycotic solution (A5955, Sigma), sodium pyruvate (S8636, Sigma) and L-glutamine (G7513, Sigma). Cells were maintained at 37°C in a humidified atmosphere containing 5% CO2. Cells were grown to confluence and used from passages 5 to 8.
To induce calcification we cultured VSMCs in osteogenic medium (OM) that was obtained by supplementing GM with inorganic phosphate (Pi) (NaH2PO4-Na2HPO4, 1.5 mmol/L, pH 7.4) and calcium (CaCl2, 0.6 mmol/L). A stock solution of Yoda1 (50 mmol/L) was prepared in DMSO. In calcification experiments Yoda1 was used at a concentration of 10 μmol/L and DMSO was used as a vehicle control. Yoda1 and Dooku1 were co-incubated for 5 min before changing the media every 2 days (
Human internal carotid artery, and superficial femoral artery samples were obtained from vascular surgical interventions. The samples were collected, stored and handled based on and in accordance with the permit number DE RKEB/IKEB 4916-2017 issued by Regional and Institutional Research Ethics Committee of the Clinical Center of the University of Debrecen.
Vascular smooth muscle cells were obtained from adult (24-week-old) C57BL6 mice. Mice were euthanized by isoflurane then placed in supine position under a dissecting microscope. The skin was removed from the thorax and abdomen. Thorax was opened and the aorta was perfused with sterile PBS. Abdominal aorta was removed from the heart to the iliac bifurcation. Adventitia was removed from aorta. The smooth muscle layer of the aorta was cut into approximately 1 × 1 mm pieces. The smooth muscle tissue was treated with collagenase type II (C-6885, Sigma, Burlington, MA, United States) solution at 37 C, 5% CO2 for 5 hours. The cells were centrifuged for 5 min, 300 ×
Freshly isolated mouse aortic smooth muscle cells and HAoSMCs were seeded onto glass coverslips and kept at 37°C for 1–2 h while attached to the surface of the coverslips. After fixed in 4% PFA and repeated washing in glicine-PBS solution (100 mM, 15 min, 22°C), cells were permeabilized with Triton-X-100 (0.5 v/v%, 10 min, 22°C). After PBS washing (3x) non-specific biding sites were blocked with Carbo-Free Blocking Solution (SP-5040, VECTOR Laboratories, Burlingame, CA, United States). Immunostaining was performed using anti-Piezo1 (Thermo Fisher Scientific, Rockford, IL, United States, MA5-32876, mouse-IgG, monoclonal) primary antibody with Cy3 anti-mouse secondary antibody labelling (A10521, Life technologies, Eugene, OR, United States). Images were taken using Zeiss laser scanning confocal microscope (Zeiss LSM880 Airyscan; Zeiss, Oberkochen, Germany) using 405, 488 and 543 nm excitation wavelengths and 10x or 63x oil immersion objective. On HAoSMCs anti-Piezo1 and anti-alfa smooth muscle specific actin (αSMA) (Thermo Fisher Scientific, Rockford, IL, United States, MA5-32876, mouse-IgG, monoclonal; PA516697, rabbit-IgG, polyclonal, respectively) primary antibodies were applied.
Paraffin embedded human aorta samples were subjected to deparaffinization process, and the protocol described above was applied as staining procedure.
Q-PCR was performed on Stratagene Mx3005P QPCR System from Agilent Technologies (Santa Clara, California, United States) using the 5′ nuclease assay. Total RNA was isolated using TRIzol from LifeTechnologies (Carlsbad, California, United States), DNase 5 treatment was performed according to the manufacturer’s protocol, and then 1 μg of total RNA were transcribed into cDNA using 15 IU of AMV reverse transcriptase (4368814, Thermo Fisher Scientific, Rockford, IL, United States). PCR amplification was performed using the TaqMan primers and probes and the TaqMan universal PCR master mix protocol (4369016, Thermo Fisher Scientific, Rockford, IL, United States). As internal control transcripts of Peptidylprolyl isomerase—A (PPIA) (Hs99999904_m1, Thermo Fisher Scientific, Rockford, IL, United States) was applied each cases. Piezo1 gene mRNA expression was measured using PIEZO1 primers (Hs00207230_m1, Thermo Fisher Scientific, Rockford, IL, United States). The amount of the transcripts was normalized at first, to the relevant housekeeping gene using the ΔCT method. The final results were then normalized to the expression of the control or scrambled samples (ΔΔCT method).
RNA from human aortic VSMCs was isolated using TRIzol (CS502, RNA-STAT60, Tel-Test Inc.; Friendswood, TX, United States) followed by cDNA synthesis using High-Capacity cDNA Reverse Transcription Kit (4368813, Applied Biosystems, Waltman, MA, United States). The qPCR was performed in triplicate using LightCycler 480 SYBR GREEN I Master Mix (04887352001, Roche Diagnostics GmbH, Mannheim, Germany). Reactions were performed in a Real Time PCR System (Bio-Rad Laboratories, Hercules, CA, United States). Primers used are as follows: forward (5′-TGCTGGGCCATTCATTTTG-3′) and reverse (5′-TCTTCCGAGTCCAGGTACAC-3′). Relative mRNA expressions were calculated with the ∆∆Ct method, using HPRT as internal control.
For Western blot experiments total cell lysates were isolated from control, calcified and gene silenced HAoSMC cell lines. The samples were subjugated to sonication. Protein concentration were measured. Samples were subjected to SDS-PAGE (10% gels loaded with 20 μg protein per lane) and transferred to nitrocellulose membranes (Bio-Rad). The protein-binding nitrocellulose membranes were then blocked with 5% dry milk-PBS solution. Proteins were probed with anti-Piezo1 (Thermo Fisher Scientific, Rockford, IL, United States, MA5-32876, mouse-IgG, monoclonal) or with anti-Runx2 (20700-1-AP; Proteintech, Rosemont, IL) primary antibodies followed by HRP-conjugated secondary antibody labeling. The primary-secondary antibody complexes were detected using an enhanced chemiluminescence Western blotting Pico or Femto kit (Thermo Scientific, Rockford, IL, United States) in a Fujifilm Labs-3000 dark box. According to the manufacturers’ descriptions, the band corresponding to the Piezo1 protein in cell lysates can be detected above 250 kDa, at ∼280–300 kDa, while that of the Runx2 protein is 57–60 kDa.
Cells were loaded with 10 mM Fura-2-AM (F1221, Thermo Fisher Scientific, Rockford, IL, United States) for 50 min at 37°C. After loading the cells were kept in Tyrode’s solution. Fura-2 was excited with a CoolLED pE-340fura light source (CoolLED LTD., Andover, England) mounted on a ZEISS Axiovert 200 m inverted microscope (Zeiss, Oberkochen, Germany). The excitation wavelengths were alternating between 340 and 380 nm wavelength, the emission was detected with a band-pass filter 505–570 nm. The image acquisition and post processing was made with the AxioVision (rel. 4.8) software (Zeiss, Oberkochen, Germany). The [Ca2+]i was calculated from the ratio of the images taken at 340 and 380 nm after background correction with constants determined during
After washing with DPBS (Dulbecco’s Phosphate Buffered Saline, 14190144, Gibco, Waltham, MA, United States) the cells were fixed in 4% paraformaldehyde (16005, Sigma, Burlington, MA, United States) and rinsed with deionized water thoroughly. Cells were stained with Alizarin Red S (A5533, Sigma, Burlington, MA, United States) solution (100 mmol/L) to the wells and measured optical density (OD) at 560 nm using hexadecyl-pyridinium chloride solution as blank.
Cells grown on 96-well plates were washed twice with DPBS, and decalcified with HCl (30721, Sigma, 0.6 mol/L) for 30 min at room temperature. Calcium content of the HCl supernatants was determined by QuantiChrome Calcium Assay Kit (DICA-500, Gentaur, Kampenhout, Belgium). Following decalcification, cells were washed twice with DPBS, and solubilized with a solution of NaOH (S8045, Sigma, Burlington, MA, United States, 0.1 mol/L) and sodium dodecyl sulfate (11667289001, Sigma, Burlington, MA, United States, 0.1%), and protein content of samples were measured with BCA protein assay kit (23225, Pierce Biotechnology, Rockford, IL, United States). Calcium content of the cells was normalized to protein content and expressed as μg/mg protein.
For osteocalcin (OCN) detection, the ECM of the cells grown on 6-well plates was dissolved in 100 μl of EDTA (E6758, Sigma, 0.5 mol/L, pH 6.9). OCN content of the EDTA-solubilized ECM samples was quantified by an enzyme-linked immunosorbent assay (ELISA) (DY1419-05, DuoSet ELISA, R&D, Minneapolis, MN, United States) according to the manufacturer’s protocol.
C57BL/6 mice (8–12 weeks old male, n = 18) were exterminated by CO2 inhalation and perfused with 5 ml of sterile DPBS. The entire aorta was harvested and cleaned under aseptic conditions and cut into pieces. Aorta rings were maintained in control, OM and OM plus Yoda1 (OMY). After 7 days the aorta pieces were washed in DPBS, opened longitudinally and decalcified in 25 µl of 0.6 mmol/L HCl overnight. Calcium content was determined by QuantiChrom Ca-assay kit as described previously.
Aortic rings from the
During Masson’s trichrome staining after deparaffinization procedure and antigen retrieval Weigert´s Iron hematoxylin solution (solution A: 5 mg Hematoxylin dissolved in 500 ml 95% ethanol; Solution B: ferric chloride, distilled water, hydrochlorid acid; mixed in 1:1 ratio) was applied for 10 min then samples were rinsed in tap water (5 min) and distilled water. In the next step Biebrich Scarlet stain (Biebrich scarlet 2.7 mg, acid fuchsin 0.3 mg, distilled water 300 ml, glacial acetic acid 3 ml) was subjected (5 min) followed by rinse in distilled water. Next phosphotungstic/phosphomolybdic acid (in 1:1 ratio, dissolved in distilled water) for 10 min was used, then slides were transferred into Aniline blue (dissolved in distilled water and supplemeted with 1 V/V% Glacial acetic acid) for 5 min 1% acetic acid solution (1 min), final washing with distilled water, dehydration, clearing of the slides and application of coverslip completed the staining protocol.
To knock-down Piezo1 gene expression we used
Pooled data were expressed as mean ± standard error of the mean (SEM) or standard deviation (SD). The differences between statistical groups were assessed using Student’s t-test with Prism (GraphPad Software, San Diego, CA, United States) and a
Initial investigations tested the relevance of Piezo1 channels in human vascular SMCs. Human internal carotid artery, as well as superficial femoral artery samples have been examined. Immunohistochemistry of these arterial rings revealed that the Piezo1 channels are expressed in the smooth muscle layers of adult human arteries. Masson’s trichrome stain has been used to localize the smooth muscle layer in the sections, and DAB staining has been applied for the visualisation of Piezo1 channels (
Verification of the expression of Piezo1 channels.
Due to the limited availability of human samples, functional measurements were carried out on a Human Aortic Smooth Muscle Cell line (HAoSMC). In order to rule out that the observed effects are attributable to the chosen HAoSMC cell line primary SMCs freshly isolated from mouse abdominal artery were also tested in a portion of the experiments. Immunostaining of Piezo1 confirmed the presence of the channel on both cell types (
After having established that SMCs express Piezo1, alteration of its expression upon calcification was investigated. To induce calcification, HAoSMC cells were cultured in osteogenic medium (OM) (growth medium (GM) supplemented with 1.5 mmol/L phosphate and 0.6 mmol/L calcium) for 4 days, as in previous studies (
After confirming the expression of Piezo1 and changes in its expression level upon calcification, activity of the channels was examined in both cell types under different conditions. First, Piezo1 channels on SMCs isolated from mouse abdominal artery were tested. Cells were loaded with the ratiometric calcium-sensitive fluorescent dye Fura-2 and alterations in the intracellular calcium concentrations were measured using the CoolLed pE-340fura illuminator mounted on a Zeiss Axioimager fluorescent microscope. Two different mechanisms were used to activate the channels: application of hypoosmotic extracellular fluid to induce mechanical stretch, and administration of Yoda1, the specific activator of Piezo1 channels. The hypoosmotic shock, as well as Yoda1 application were able to evoke calcium transients on primary SMCs isolated from mouse abdominal artery (
Alteration of Ca2+ transients evoked by pharmacological and mechanical activation (due to cell volume regulation) of Piezo1 channels on control and calcified smooth muscle cells. Representative transients and mean of Fura-2 fluorescence intensity changes measured on individual cells are shown.
The influence of Piezo1 activation, and the consequently elevated intracellular calcium concentration on calcification of HAoSMCs were then addressed. Calcification of HAoSMCs was induced using an osteogenic medium (OM) in the absence or presence of Yoda1 (10 μmol/L). As revealed by Alizarin red (AR) staining and calcium measurement of the extracellular matrix (ECM) after 6 days of treatment Yoda1 largely amplified OM-induced calcification of HAoSMCs (
Activation of Piezo1 by Yoda1 increases Pi-mediated osteogenic differentiation and extracellular matrix (ECM) calcification of HAoSMCs and aorta rings.
For further confirmation, an
To address whether osteogenic differentiation of HAoSMCs occurred upon stimulation with Yoda1, OM or OMY, the expression of Runx2 protein, the master transcription factor of osteogenesis was determined. Runx2 was found to be upregulated in HAoSMCs stimulated with OM and OMY as compared to control cells (
Upregulation of Runx2 and Transglutaminase 2 (Tg2) in calcifying HAoSMCs
Next, Dooku1, an antagonist of Yoda1-evoked Piezo1 channel activity was tested to confirm the role of Piezo1 in Yoda1-induced calcification of HAoSMCs. As revealed by AR staining and calcium measurement of the ECM Dooku1 at a concentration of 5 μmol/L inhibited OMY-induced calcification of HAoSMCs (
Dooku1 inhibits Yoda1-induced increase in HAoSMCs calcification.
To further confirm the role of Piezo1 in OMY-induced calcification, Piezo1 expression has been downregulated by siRNA in HAoSMC cells (KD cells). Quantitative PCR analysis indicated significantly decreased expression of Piezo1 at mRNA level (
Piezo1 knockdown inhibits Yoda1-induced increase in HAoSMCs calcification.
Knockdown of Piezo1 by siRNA attenuated OMY-induced VSMCs calcification as detected by AR staining, as well as by the calcium content of the ECM (
Finally, to address the pathophysiological role of Piezo1 in calcification, calcification experiments in the absence of Yoda1 were performed. First, the effect of Dooku1 was tested on OM-induced calcification. As revealed by AR staining and ECM Ca measurement Dooku1 attenuated the OM-induced calcification of HAoSMCs without additional stimulation of Piezo1 channels (
Piezo1 is critical in OM-induced HAoSMCs calcification.
Taken together, these findings demonstrate the fundamental role of Piezo1 in calcification. Without sufficient expression or function of the channel, administration of the osteogenic medium has no or limited impact on calcification as indicated by AR staining, and ECM calcium content.
Our idea of VC has been transformed fundamentally in the last decades from a picture of a passive degenerative disease to an active, tightly regulated process. The role of SMCs in VC is by now undisputed. The importance of altered intracellular calcium homeostasis both in SMC dysfunction and calcification is also unquestionable. However, the causal relationship between these two phenomena, and the response to how intracellular calcium levels of the SMCs may increase leading to VC, are only partially understood. Our observations provide one of the missing links, revealing a potential source of [Ca2+]i rise in HAoSMCs leading to calcification. Furthermore, they also demonstrate that this effect significantly accelerates calcification.
In our study evidence that the Piezo1 channel is involved in arterial medial calcification has been provided. First, Piezo1 expression was confirmed on human aorta rings, and on primary abdominal aortic smooth muscle cells isolated from mice, as well as on HAoSMC cell line. These observations alone were not surprising, since Piezo1 expression has already been detected in several smooth muscle cells (
Next, both membrane stretch associated with cell swelling induced by hypotonic extracellular solution and pharmacological (Yoda1) stimulations were demonstrated to activate Piezo1 resulting in [Ca2+]i rises. However, hypotonicity has been reported to trigger calcium influx through other mechanosensitive channels. Such widely studied plasmalemmal proteins are the members of TRP channels family. The contribution of TRP channels in arterial smooth muscle cells to physiological and pathological processes has not been fully elucidated and is sometimes controversial. Especially, their contribution to AMC has not been studied to date, although the potential expression of about 13 different TRP channels in vascular smooth muscle cells has been suggested in previous studies (
L-type calcium channels expressed in SMCs may contribute to an increased intracellular calcium level, although being voltage sensitive, significant calcium entry cannot be assumed in the absence of depolarization. Work of Retailleau et al. did not reveal any remarkable effect of nifedipine on cytosolic calcium concentration on isolated caudal arteries (
Retailleau et al. also described the mechanosensitive properties of the voltage-gated CaV1.2 channels on arterial muscle cells in association with smooth muscle filamin A expression (
On the other hand, Yoda1 a small-molecule activator of Piezo1 has no effect either on TRP channels, or on other stretch-activated channels. The increased calcium influx measured using Yoda1 can thus be attributed to the Piezo1 channels. Besides HAoSMC, SMCs isolated from mouse abdominal artery were also examined, and they were also responsive to stimulations of Piezo1, ruling out that the effect seen was the peculiarity of the HAoSMCs. The use of subcultured human aortic smooth muscle cells with a high passage number may pose a risk due to the changed morphological and physiological properties of the aging cells (
The Yoda1 analogue, Dooku1, known to reversibly antagonize Yoda1 stimulation of Piezo1, significantly reduced the amplitude of calcium transients on HAoSMCs following Yoda1 treatment (
Furthermore, we have shown that phosphate-induced calcification of HAoSMCs is associated with an increase in Piezo1 expression level, and accordingly, stimulation of Piezo1 channels resulted in a greater increase in intracellular calcium levels as compared to control cells. These observations clearly indicate that the Piezo1 expression, the intracellular Ca2+ level, and the calcification are interrelated. We also pointed out that Yoda1-induced activation of Piezo1 channels significantly enhances calcification on both HAoSMCs and
The central role of both extracellular and intracellular calcium concentrations in calcification have been extensively documented (
These observations led to the conclusion that calcium is released from internal stores. However, this conclusion needs to be reviewed and supplemented because calcium can also enter the cell from the extracellular space through the Piezo1 channels, as shown in the work described here. This mechanism is also consistent with the finding that extracellular calcium levels play an important role in the mineralization of SMCs (
The process of VC is analogous to bone formation in many aspects. Both processes involve changes in gene transcription, overproduction of mineralization-regulating proteins, and concomitant release of mineralization-competent matrix vesicles (
In mesenchymal stem cells the expression level of Piezo1 has been proven to correlate with the degree of osteoblast differentiation and the suppression of adipocyte differentiation (
Moreover, Piezo1 involvement in hypertensive arterial remodelling has already been characterized (
It is arguable, of course, to what extent the activation of the Piezo1 channel by Yoda1 can be considered a physiological/pathophysiological stimulus. Nevertheless, the use of a specific agonist provides a clear environment for studying the consequences of channel activation. In addition, the fact that the use of the Piezo1 inhibitor Dooku1 alone, without activating the Piezo1 channel, reduces the rate of calcification confirms our hypothesis that the Piezo1 channel is an active participant in calcium homeostasis under physiological conditions and is a major contributor to AMC formation. Although Dooku1 was originally documented as an inhibitor of Yoda1-induced Piezo1 channel activity, our experiments also suggest that constitutively active Piezo1 channels are expressed on HAoSMCs, and the inhibitor has an effect on constitutively active Piezo1 channels as well, as demonstrated in recent studies on perivascular adipose tissue and odontoblasts (
Our findings that the Piezo1 channel has remarkable impact on the regulation of cytosolic calcium levels in VSMCs and thus it is a determinant in the formation and progression of calcification may explain the hitherto unclear question of what leads to a significant increase in intracellular calcium concentration in VSMCs, which might be able to stimulate the further steps of mineralization studied extensively earlier. Based on the role of the Piezo1 channel revealed here, they may be a potential therapeutic target that could prevent or slow the progression of AMC.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by Regional and Institutional Research Ethics Committee. The patients/participants provided their written informed consent to participate in this study.
LC, BD, and VJ designed the experiments; wrote the manuscript, contributed to discussion and reviewed/edited the manuscript; LS, NB and AA participated in cell culturing, immunohisochemstry, and calcium measurements; AT performed experiments of calcification. MG, DC carried out PCR and Western blot analysis, CT, IB provided the human samples and assisted the treatment of samples. All authors contributed to discussion, laboratory support, statistical analysis and reviewed/edited the manuscript.
This work was supported by grants from the Hungarian National Research, Development and Innovation Office EFOP3.6.2-16-2017-00006, project no. TKP2020-NKA-04 (under the 2020-4.1.1-TKP2020 funding scheme), project no. TKP2021-EGA-18 (under the TKP2021-EGA funding scheme), project no. GINOP-2.3.3- 15-2016-00020, and grant K131535 (to VJ). This research was funded by the Hungarian Academy of Sciences, MTA-DE Lendület Vascular Pathophysiology Research Group, grant number 96050 (to VJ). This work was also supported by the Eötvös Lorand Research Network (ELKH).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.