P2Y1 and P2Y13 purinergic receptors mediate Ca²⁺ signaling and proliferative responses in pulmonary artery vasa vasorum endothelial cells

Abstract

Extracellular ATP and ADP have been shown to exhibit potent angiogenic effects on pulmonary artery adventitial vasa vasorum endothelial cells (VVEC). However, the molecular signaling mechanisms of extracellular nucleotide-mediated angiogenesis remain not fully elucidated. Since elevation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) is required for cell proliferation and occurs in response to extracellular nucleotides, this study was undertaken to delineate the purinergic receptor subtypes involved in Ca²⁺ signaling and extracellular nucleotide-mediated mitogenic responses in VVEC. Our data indicate that stimulation of VVEC with extracellular ATP resulted in the elevation of [Ca²⁺]_i via Ca²⁺ influx through plasma membrane channels as well as Ca²⁺ mobilization from intracellular stores. Moreover, extracellular ATP induced simultaneous Ca²⁺ responses in both cytosolic and nuclear compartments. An increase in [Ca²⁺]_i was observed in response to a wide range of purinergic receptor agonists, including ATP, ADP, ATPγS, ADPβS, UTP, UDP, 2-methylthio-ATP (MeSATP), 2-methylthio-ADP (MeSADP), and BzATP, but not adenosine, AMP, diadenosine tetraphosphate, αβMeATP, and βγMeATP. Using RT-PCR, we identified mRNA for the P2Y1, P2Y2, P2Y4, P2Y13, P2Y14, P2X2, P2X5, P2X7, A1, A2b, and A3 purinergic receptors in VVEC. Preincubation of VVEC with the P2Y1 selective antagonist MRS2179 and the P2Y13 selective antagonist MRS2211, as well as with pertussis toxin, attenuated at varying degrees agonist-induced intracellular Ca²⁺ responses and activation of ERK1/2, Akt, and S6 ribosomal protein, indicating that P2Y1 and P2Y13 receptors play a major role in VVEC growth responses. Considering the broad physiological implications of purinergic signaling in the regulation of angiogenesis and vascular homeostasis, our findings suggest that P2Y1 and P2Y13 receptors may represent novel and specific targets for treatment of pathological vascular remodeling involving vasa vasorum expansion.

purine and pyrimidine nucleotides are released from both resident vascular and circulating blood cells as a result of fluid shear stress, changes in osmolarity, hypoxia, cell activation in response to various stimuli, and neurotransmission, and they play an important role in regulating vascular functions (5–7, 13, 21, 30, 45, 46). The effects of extracellular nucleotides and nucleosides are mediated by P2 and P1 purinergic receptors, respectively. Ubiquitous expression of purinergic receptors by the cells present in the vascular wall indicates that extracellular nucleotides are important signaling molecules involved in modulating vascular responses under physiological and pathological conditions. It has been demonstrated that endothelial cells from various vascular beds express multiple sets of purinergic receptor subtypes, including P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2X4, P2X5, and P2X7 depending on the vascular bed they originate from (7, 8, 37, 43, 50). Activation of P2 receptors in vascular endothelial cells induces Ca²⁺-dependent NO production and release of endothelium-dependent hyperpolarizing factor, prostacyclin, and tissue plasminogen activator (8, 30, 50). A number of studies have demonstrated a contribution of P2 receptors to cytokine release (5, 18), platelet aggregation (53), and immune cell activation and proliferation (5). In addition, involvement of P2 receptors in chemotactic, morphogenetic, and proliferative responses of endothelial cells also has been reported, generating considerable interest in the field of purinergic signaling and its relevance to angiogenic regulation (1, 9, 21, 29, 45, 52). However, the data on the role of individual P2Y receptor subtypes in angiogenic responses in endothelial cells remain very limited.

Ca²⁺ signaling plays an important role in regulation of physiological responses in endothelial cells (3, 4, 20, 38, 39). Angiogenic activation of endothelial cells in response to growth factors such as VEGF, basic FGF, and IGF2 promotes an elevation of intracellular Ca²⁺ followed by the activation of Ca²⁺-dependent signaling events (38). Ca²⁺-mediated signal transduction has been implicated in the endothelial and epithelial cell responses, notably the mechanically stimulated intercellular and intracellular calcium waves (20, 30), vesicular exocytosis, and cytoskeleton alterations (27, 39). In addition, a role of Ca²⁺ signaling has been demonstrated in airway ciliary activity (56), oxidative stress regulation (2, 27), and P2Y-induced (48), Zn²⁺-dependent (24), transient receptor potential (TRP) channel-regulated (31), and other types of cellular responses (4, 38). It also has been demonstrated that in epithelial cells, an ATP-induced increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) may be linked to a store-operated Ca²⁺ release (36), may be regulated by Na⁺-K⁺-ATPase in an inositol 1,4,5-trisphosphate (IP₃)-dependent fashion (12), and may originate from distinct sustained and transient Ca²⁺ influx via activated P2X receptors (57). Thus it can be concluded that the variety of mechanisms of Ca²⁺ responses and their role in endothelial cell physiology strongly suggest a fundamental importance of intracellular Ca²⁺ for regulation of angiogenesis (38).

A number of studies have addressed a possibility that the changes in [Ca²⁺]_i can be observed in both cytosolic and nuclear compartments and that increases in nucleoplasmic Ca²⁺ are required for cell proliferation (4, 44). It also was reported that a mitogenic effect of hepatocyte growth factor observed in SkHep1 cells involves c-Met translocation to the nucleus and elevation of nucleoplasmic [Ca²⁺]_i (22, 44). Moreover, Ca²⁺ signaling plays a role in VEGF- and Wnt5a-induced endothelial cell proliferation (11, 34) as well as in EGF signaling in embryonic stem cells (25). Accordingly, inhibition of VEGF- and FGF-induced angiogenesis by endostatin and angiostatin was associated with a diminished Ca²⁺ response (28). Thus it has been suggested that Ca²⁺ signaling plays an important role in endothelial cell physiology and may be involved in regulation of angiogenesis (38). Because the Ca²⁺ responses in the endothelium in the context of purinergic receptors have not been extensively characterized, we aimed to investigate a role of extracellular nucleotide-induced Ca²⁺ signaling in endothelial cells and its possible role in angiogenesis.

We recently reported that, unlike endothelial cells of large systemic and pulmonary vessels, vasa vasorum endothelial cells (VVEC) isolated from pulmonary arteries of chronically hypoxic calves exhibit a distinct endothelial phenotype characterized by increased mitogenic responses to extracellular ATP. Moreover, we found that even more dramatic VVEC mitogenic responses were observed to extracellular ADP, but not to extracellular UTP and UDP (21, 52). These observations suggested that P2Y1 and/or P2Y13 receptors responding to ADP may play a unique role in regulation of VVEC angiogenesis and in adventitial vasa vasorum expansion. To further delineate the role of purinergic receptors in VVEC angiogenesis, this study has been focused on extracellular nucleotide-induced intracellular Ca²⁺ signaling and proliferative responses, including DNA synthesis and phosphorylation of ERK1/2, Akt, and S6 ribosomal protein. Our data demonstrate that in VVEC, extracellular ATP induced Ca²⁺ influx through both the plasma membrane channels and Ca²⁺ mobilization from the intracellular stores. Ca²⁺ responses were simultaneously generated in the cytosolic and the nuclear compartments, consistent with the role of Ca²⁺ signaling in mediating VVEC proliferation. Furthermore, establishing a purinergic receptor expression profile by RT-PCR and pharmacological analysis, with the use of P2 receptor agonists and antagonists, revealed that mitogenic and Ca²⁺ responses in VVEC were mediated predominantly by P2Y1 and P2Y13 receptors. Together, our findings present new evidence for the role of P2Y1 and P2Y13 receptors in VVEC and suggest that these receptors may be potential therapeutic targets for modulating physiological and pathological angiogenesis.

MATERIALS AND METHODS

Culturing of VVEC from pulmonary artery adventitia.

Pulmonary arteries were obtained from male Holstein calves that had been exposed to hypobaric hypoxia for 2 wk (barometric pressure = 430 mmHg). All animal work was undertaken using standard veterinary care in the Department of Physiology, School of Veterinary Medicine, Colorado State University. The protocol was approved by the Institutional Animal Care and Use Committee at Colorado State University (no. 08-090A-02; animal welfare assurance no. A3572-01). Adventitia was dissected from the media and extensively washed in phosphate-buffered saline solution (PBS), and VVEC were isolated as previously described (21). Cells were cultured in DMEM (Cellgro Mediatech, Manassas, VA) supplemented with 20 mM l-glutamine (Cellgro), nonessential amino acids (1:100 vol/vol; Sigma, St. Louis, MO), 100 U/ml penicillin and 100 μg/ml streptomycin (Sigma), 10% fetal bovine serum (FBS; Gemini Bio-Products, Woodland, CA), and 15 μg/ml endothelial cell growth supplement (Upstate Biotechnology, Charlottesville, VA). VVEC were characterized by the expression of endothelial markers, including von Willebrand factor, endothelial nitric oxide synthase, and platelet endothelial cell adhesion molecule-1; binding of the lectin Licopercsicon esculentum; and incorporation of acetylated low-density lipoproteins labeled with 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (14). All studies were performed on cells between passages 2 and 7.

Intracellular Ca²⁺ measurements.

VVEC were cultured on glass-bottom dishes (MatTek, Ashland, MA) to 80% confluence and growth-arrested in DMEM without serum for 72 h. For intracellular Ca²⁺ measurements, cells were loaded with 1 μM fura-2 AM (Invitrogen, Carlsbad, CA) for 30 min at room temperature, washed with medium (or Ca²⁺-free Ringer's solution where indicated), and left to recover for 10 min to reduce spontaneous intrinsic Ca²⁺ activity. Time-lapse λ340/380-nm image acquisition was started, and cells were perfused with a solution containing a stimulus as indicated. Experiments were performed at room temperature. The imaging system was constructed by Intelligent Imaging Innovations (Denver, CO) utilizing a Nikon TE2000 microscope with a motorized Z-axis focusing mechanism, a Sutter Arc lamp with fiber optical light guide, and a Cooke SensiCam charge-coupled device camera. SlideBook software was used for image acquisition and analysis, and a ratiometric technique was used for monitoring intracellular Ca²⁺ over time in a single cell (32). Single-cell Ca²⁺ traces (typically from 5–10 cells) were averaged.

Analysis of purinergic receptor expression by RT-PCR.

Total cellular RNA was isolated from VVEC extracts using an RNA isolation kit (Qiagen, Santa Clarita, CA) and was treated with RQ1 DNase 1 (Promega, Madison, WI) to remove residual contaminating DNA. Two micrograms of RNA were reverse-transcribed into first-strand cDNA using the SuperScript III First Strand and Platinum PCR SuperMix kits (Invitrogen) and quantified by spectrophotometric analysis. PCR amplification was carried out on equal amounts of cDNA for 35 cycles using the following parameters: initial denaturation step, 5 min at 94°C; amplification steps, 30 s at 55°C and 2 min at 72°C; and final elongation step, 10 min at 72°C. Amplification products were resolved on 2% agarose gel and stained with ethidium bromide, and the fragment sizes were determined by comparison to known DNA standards. Receptor-specific oligonucleotide primers were chosen based on the published sequences for purinergic receptors and the GenBank database of the National Center for Biotechnology Information.

DNA synthesis measurement.

VVEC were seeded on 24-well plates at a density of 1.2 × 10⁴ cells per well in DMEM supplemented with 10% FBS. The next day, cells were rinsed with PBS and growth-arrested by culturing in DMEM without serum for 72 h. Cells were stimulated with 100 μM ATP, ADP, ATPγS, ADPβS, 2-methylthio-ATP (MeSATP), or 2-methylthio-ADP (MeSADP) (Sigma) in the presence of 0.125 μCi of [methyl-³H]thymidine (NEN Life Science Products, Boston, MA) for 24 h. To evaluate a contribution of P2Y1 and P2Y13 receptors, cells were preincubated with the P2Y1 receptor antagonist MRS2179 (10 μM, 30 min) or the P2Y13 receptor antagonist MRS2211 (10 μM, 30 min) before stimulation with extracellular nucleotides. P2Y13 receptors were shown to be coupled to Gα_i protein; therefore, in some experiments cells were preincubated with its inhibitor, pertussis toxin (PTx; 100 ng/ml, 18 h). In the experiments designed to evaluate a role of Ca²⁺ in DNA synthesis, cells were preincubated with cell-permeable BAPTA-AM (10 and 30 μM; Invitrogen) for 30 min. After pretreatment, cells were stimulated with ATP (100 μM) in the presence of 0.125 μCi of [methyl-³H]thymidine for 24 h. Incorporated radioactivity in total cell lysates was determined using a Beckman LS6500 counter.

Preparation of cell extracts and Western blot analysis.

VVEC were cultured to near confluence, growth-arrested in DMEM for 72 h, and stimulated with ATP, ADP, ATPγS, ADPβS, MeSATP, or MeSADP (100 μM) in serum-free medium for 10 min. To evaluate a role of P2Y1 and P2Y13 receptors in nucleotide-induced signaling, cells were preincubated with the P2Y1 receptor antagonist MRS2179 (10 μM, 30 min), the P2Y13 receptor antagonist MRS2211 (10 μM, 30 min), or PTx (100 ng/ml, 18 h). Total cell lysates were prepared as described previously (21). Equivalent amounts of total cell protein (20 μg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Proteins were transferred to polyvinylidene difluoride (PVDF) membranes and probed with rabbit polyclonal antibodies against phospho-ERK1/2 (Tyr²⁰²/Thr²⁰⁴), phospho-Akt (Ser⁴⁷³), phospho-S6 (Ser²³⁶), following conditions recommended by the manufacturer (Cell Signaling Technology, Danvers, MA). After being washed with Tris-buffered saline-Tween buffer, membranes were incubated with donkey anti-rabbit peroxidase-conjugated IgG at 1:20,000 dilution (Amersham Biosciences, Piscataway, NJ) for 1 h at room temperature. Immunoreactive bands were detected using the Renaissance ECL kit (NEN Life Science Product) followed by exposure to Hyperfilm. In all experiments, equal sample loading and transfer were verified by staining PVDF membranes with Ponseau and by immunoprobing with anti-GAPDH antibody (Cell Signaling Technology).

Statistical analysis.

For the analysis of variances between groups of data, one-way ANOVA was performed, followed by the Dunnett or Bonferroni test, using GraphPad Prism 3.0 (GraphPad Software). Density of protein bands from Western blot images was quantified using NIH ImageJ software.

RESULTS

Purinergic receptor agonists increase [Ca²⁺]_i in VVEC.

First, we screened selected nucleotides and their analogs for the ability to initiate Ca²⁺ signaling. VVEC exhibited robust Ca²⁺ flux within seconds after stimulation with MeSATP, ADP, ADPβS, ATPγS, 2′(3′)-O-(4-benzoylbenzoyl)adenosine 5′-triphosphate (BzATP), or UTP (Fig. 1A; concentrations as indicated). Similar Ca²⁺ responses were observed for ATP and MeSADP (as shown below), and smaller Ca²⁺ influx was induced by UDP. VVEC were unresponsive to αβ-methylene-ATP (αβMeATP), βγ-methylene-ATP (βγMeATP), diadenosine tetraphosphate (Ap₄A), AMP, and adenosine (Ado), indicating that several P2 receptor subtypes, but not P1 (adenosine) receptors, are coupled to intracellular Ca²⁺ signaling in VVEC.

Fig. 1.

Extracellular nucleotides induce Ca²⁺ responses in vasa vasorum endothelial cells (VVEC). A: growth-arrested VVEC (72 h in serum-free DMEM) exhibited robust Ca²⁺ influx in response to stimulation with extracellular nucleotides (concentrations as indicated). Image acquisition began at time 0; after 100 s, cells were stimulated with nucleotides (arrows), and data acquisition continued in the presence of stimuli for an additional 1,100 s. [Ca²⁺]_i, intracellular Ca²⁺ concentration; MeSATP, 2-methylthio-ATP; BzATP, 2′(3′)-O-(4-benzoylbenzoyl)adenosine 5′-triphosphate; αβATP, αβ-methylene-ATP; βγATP, βγ-methylene-ATP. B: representative images of Ca²⁺ responses in VVEC stimulated with ATP followed by ionomycin (images taken 10 s after each stimulation). Note that nuclear Ca²⁺ levels in response to both stimuli were higher than those in the cytoplasm.

Purinergic stimulation of VVEC induces cytosolic and nucleoplasmic [Ca²⁺]_i increase via both influx of extracellular Ca²⁺ and Ca²⁺ mobilization from intracellular stores.

To further characterize Ca²⁺ responses in VVEC, we investigated localization and mechanisms of intracellular Ca²⁺ transients. The experiment shown in Fig. 1B demonstrates characteristic subcellular patterns of Ca²⁺ increases in VVEC 10 s after the addition of ATP. The data obtained from several (n = 3–5) distinct VVEC populations revealed the following nucleotide rank order of potency to induce Ca²⁺ response: MeSATP ≥ MeSADP ≥ ADP = ATP > ADPβS ≥ ATPγS = BzATP > UTP > UDP. The observed pharmacological profile suggests an involvement of several purinergic receptor subtypes in an increase in [Ca²⁺]_i, with a predominant role of P2Y1 and P2Y13 receptors and, to a lesser extent, a role of P2Y2 and P2X7 receptors. Our data also showed that although subcellular Ca²⁺ levels were relatively homogeneous within resting cells, after the stimulation with extracellular ATP (100 μM), [Ca²⁺]_i appeared to be higher in the nucleus than in the cytoplasm (Fig. 2A). These differences were further magnified by the addition of ionomycin (Fig. 2A). Temporal differences between increases in nuclear and cytoplasmic Ca²⁺ levels were undetectable (data not shown).

Fig. 2.

Purinergic stimulation of VVEC evokes an increase in cytosolic and nucleoplasmic Ca²⁺ levels via influx of extracellular Ca²⁺ and mobilization of Ca²⁺ from intracellular stores. A: comparison of averaged peak Ca²⁺ levels in response to ATP in the nucleus and cytoplasm. B: VVEC Ca²⁺ response to thapsigargin in Ca²⁺-free medium (dashed line) and to ATP in normal medium (solid line). The presence of a relatively small transient increase in cytoplasmic Ca²⁺ level is indicative of its release from intracellular stores and thus the existence of a store-derived component of Ca²⁺ signaling in VVEC. C: Ca²⁺ response to ATP in Ca²⁺-free medium followed by the addition of Ca²⁺. Each of the superimposed Ca²⁺ traces represents the recording from a single cell. D: multiphased Ca²⁺ fluctuations exhibited by an individual VVEC in response to ATP. Red traces represent individual cells; green trace is an average for a whole cell population. Results are representative of 3 experiments. Image acquisition began at time 0; after 100 s, cells were stimulated with ATP (arrows), and data acquisition continued in the presence of stimulus for an additional 1,100 s.

Although the store-operated Ca²⁺ entry (SOC) of ATP-mediated Ca²⁺ influx has been reported in epithelial cells (36), cardiac fibroblasts (17), and mesenteric microvessels (42), the presence of this mechanism in VVEC has not been established. To address this, we stimulated cells with 1 μM thapsigargin (TG) in Ca²⁺-free medium. TG is a specific inhibitor of sarco(endo)plasmic reticulum Ca²⁺-ATPases, known to induce depletion of intracellular Ca²⁺ stores. We observed a small transient increase of Ca²⁺ in response to TG that was severalfold lower than that induced by ATP in Ca²⁺-containing medium (Fig. 2B). This type of increase in Ca²⁺ concentration is indicative of the presence of a store-derived component of the Ca²⁺ pool in VVEC. Data presented in Fig. 2C provide additional evidence in support of this finding. Stimulation of VVEC with ATP in Ca²⁺-free medium (200 s after image acquisition was started, as indicated by arrow) resulted in the release of Ca²⁺ from intracellular stores, whereas cell perfusion with Ca²⁺-containing medium at 400 s resulted in a rapid influx of newly added Ca²⁺ through Ca²⁺ channels on the plasma membrane. Thus ATP-induced changes in [Ca²⁺]_i include initial release of Ca²⁺ from intracellular stores followed by influx through membrane Ca²⁺ channels.

Moreover, we observed variations in multiphased Ca²⁺ fluctuations exhibited by some individual VVEC in response to ATP stimulation (Fig. 2D). Such Ca²⁺ fluctuation patterns or even stable oscillatory responses were observed in other types of endothelial cells and are believed to play a role in intercellular signaling that synchronizes regulatory mechanisms at the tissue level (16, 17, 54).

Overall, data presented in Figs. 1 and 2 demonstrate that the primary response of VVEC to nucleotides is Ca²⁺ release. The presence of a store-operated component in VVEC Ca²⁺ signaling is consistent with the G protein-coupled characteristics of metabotropic P2Y receptors.

VVEC express multiple purinergic receptor subtypes.

Expression of heterogenous purinergic receptors in endothelial cells has been previously reported (37, 50). The efficacy of various extracellular nucleotides to induce Ca²⁺ responses suggested that multiple purinergic receptors are expressed in VVEC. Our RT-PCR analysis revealed the expression of metabotropic P2Y1, P2Y2, P2Y4, P2Y13, and P2Y14 receptors and ionotropic P2X1, P2X2, P2X5, and P2X7 receptors in VVEC (Fig. 3 and Table 1). In addition, VVEC express A1, A2b, and A3 adenosine receptors. Considering potent mitogenic effects of extracellular ATP and ADP, but not UTP and UDP, in VVEC (21), these data strongly suggest that P2Y1 and P2Y13 receptors may play a predominant role in nucleotide-mediated angiogenic signaling in these cells.

Fig. 3.

PCR analysis of purinergic receptor expression in VVEC. PCR was performed as described in materials and methods using primers specific for P2 and P1 receptors (see Table 1). Amplification products were identified using a 2% agarose gel containing ethidium bromide and visualized under ultraviolet light. Sizes of the PCR products were verified by comparison to DNA standards.

Table 1.

Primers used for PCR analysis of purinergic receptor expression in VVEC

Gene/Accession No.	Oligonucleotide	Sequence	Product Size, bp
P2Y1	Sense primer	5′-GCT GCA GAG GTT CAT CTT CC-3′	204
NM 174410	Antisense primer	5′-CCG GTG CCA GAG TAG AAG AG-3′
P2Y2	Sense primer	5′-GGT AAG GAC TGA GCC AGC AG-3′	240
XM 612432	Antisense primer	5′-GCC GTA ACC TCA TCC TCA AA-3′
P2Y4	Sense primer	5′-GCA AGT TCA TCC GCT TTC TC-3′	206
XM 592537	Antisense primer	5′-GGT GGT GAC GAA GAA CAG GT-3′
P2Y6	Sense primer	5′-CCA ACA CAC TTG GGG CTA CT-3′	171
XM 588870	Antisense primer	5′-CCT GGA CTC CAT CTT CCA AA-3′
P2Y11*	Sense primer	5′-CGC TCA GGT GAG AGA AAC TG-3′	162
	Antisense primer	5′-TCT TCA TCA CCT GCA TCA GC-3′
P2Y12	Sense primer	5′-ACG GAA CAC GTT CTC CAA TC-3′	224
NM 001001174	Antisense primer	5′-GGC TCG TGT CTC TCT CTG CT-3′
P2Y13*	Sense primer	5′-TCA AGT CAG CCA CCA AAG TG-3′	163
	Antisense primer	5′-CCC AGA GAC CTG AAG ACA GC-3′
P2Y14*	Sense primer	5′-TTC CAG GCA AGG ATA GGA TG-3′	193
	Antisense primer	5′-CTC CTG CTG CTT CCC TAA TG-3′
P2X1*	Sense primer	5′-ACA GGT GGA TCC AGA GGT CA-3′	163
	Antisense primer	5′-CCG CCC CCA TCT ATC TTT-3′
P2X2*	Sense primer	5′-AGA CAA GGG CCA GGG TCA C-3′	153
	Antisense primer	5′-ATC TGT CTT GCT GGC CTC AG-3′
P2X3*	Sense primer	5′-GGA CCA TTG GGA TCA TCA AC-3′	188
	Antisense primer	5′-TGG GGT CAC GTA GTC AGA CA-3′
P2X4	Sense primer	5′-TGA CAC AGA AGT GCC AAA GC-3′	158
NM 001034049	Antisense primer	5′-TCG CAT CAT AAA TGC ACG TT-3′
P2X5	Sense primer	5′-AGC AGG AAT CTG AGC ACG TT-3′	203
XM 594012	Antisense primer	5′-GGT ACT CCA AAG GCA TGG AA-3′
P2X6*	Sense primer	5′-CGG GAG CTC TTC TCT TAC CC-3′	170
	Antisense primer	5′-GGG GGC TTC TGG ATT ACA A-3′
P2X7	Sense primer	5′-TGC TTT GGT GAG TGA CAA GC-3′	165
XM 591410	Antisense primer	5′-TGC AAG GGG AAG GTG TAA TC-3′
ADORA1	Sense primer	5′-GTT GTG TGT GTG CTG GGA AC-3′	185
NM 174497	Antisense primer	5′-CCC CAC TTG GCA CAT CTA GT-3′
ADORA2a	Sense primer	5′-TCA ACA GCA ACC TGC AGA AC-3′	154
	Antisense primer	5′-ACA AAG CAG GCG AAG AAG AG-3′
A2b	Sense primer	5′-CCA GTC CAC ACC ATC AAC TG-3′	190
NM 001075925	Antisense primer	5′-GGC AGA GGA TGT ACC TGG AA-3′
ADORA3	Sense primer	5′-TTT GTG CTG GAC ACA GCT TC-3′	242
NM 001079645	Antisense primer	5′-TCA TCT CTG GCA AAG TCA CG-3′

Sense and antisense primers used for PCR analysis of purinergic receptor expression in vasa vasorum endothelial cells (VVEC) are shown with their product size in base pairs (bp). Primers were designed based on 100% homology sequence between human and mice cDNA.

^*Bos Taurus gene accession numbers are not available.

P2Y1 and P2Y13 receptors mediate elevation of intracellular Ca²⁺ in VVEC.

To determine whether P2Y1 and P2Y13 receptors mediate Ca²⁺ responses in VVEC, we used a pharmacological approach that involves P2Y1 and P2Y13 agonists and antagonists. The agonists were chosen based on the published pharmacological profile of purinergic receptors (7, 49). Stimulation of VVEC with extracellular nucleotides ADP, MeSADP, ADPβS, ATP, MeSATP, and ATPγS (100 μM each) resulted in significant elevation of intracellular Ca²⁺ (Fig. 4A). The strongest responses were observed when VVEC were stimulated with MeSATP. Preincubation with the P2Y1 receptor-specific antagonist MRS2179 (10 μM, 30 min) suppressed Ca²⁺ responses to all these nucleotides, with the exception of ADPβS. ADPβS is a less potent P2Y1/P2Y2/P2Y13 receptor agonist; therefore, it may act on P2Y2 and/or P2Y13 receptors. Preincubation with the P2Y13 receptor-specific antagonist MRS2211 resulted in significant inhibition of Ca²⁺ responses to ATP, MeSATP, ATPγS, ADP, and ADPβS. The response to MeSADP was less sensitive to MRS2211, suggesting the possibility that in bovine VVEC, MeSADP may act preferentially on P2Y1 receptors. In human and rat tissues and some cell lines, P2Y13 was characterized as a Gα_i protein-coupled receptor (7, 49, 55). Therefore, to further distinguish a contribution of P2Y1 and P2Y13 to Ca²⁺ responses, we examined the effects of nucleotides on PTx-treated cells (Fig. 4B). Preincubation of VVEC with PTx (100 ng/ml, 18 h) decreased the responses to ADPβS (by 38%) and ATPγS (by 43%). A small but not significant inhibitory effect of PTx was observed in cells stimulated with MeSADP (by 4%) and MeSATP (by 13%). The responses to ADP and ATP remained unaffected. Given that ADPβS and ATPγS can activate Gα_i/α_q-coupled P2Y2 receptors, our results indicate the involvement of both Gα_i-dependent (P2Y13 or P2Y2) and Gα_i-independent (P2Y1, P2Y2 and P2Y13) receptors in nucleotide-induced Ca²⁺ signaling in VVEC.

Fig. 4.

P2Y1 and P2Y13 receptors are involved in nucleotide-induced Ca²⁺ responses in VVEC. Fura-2-loaded growth-arrested VVEC were pretreated for 30 min with 10 μM MRS2179, 10 μM MRS2211, or a vehicle control (A) or were pretreated for 18 h with 100 ng/ml PTx before fura-2 loading (B). Cells were stimulated with ATP, ADP, MeSATP, MeSADP, ADPβS, and ATPγS (100 μM). Ca²⁺ responses from 5–10 single cells in the field were monitored as described in materials and methods. *P < 0.05; **P < 0.01 vs. nucleotide-stimulated cells. cont, Control.

Intracellular Ca²⁺ is required for VVEC proliferation.

Because elevated nucleoplasmic Ca²⁺ might be required for mitogenic responses in VVEC, we next examined whether nucleotide-induced Ca²⁺ responses are functionally important to VVEC proliferation. The cell-permeable Ca²⁺ chelator BAPTA-AM (10 and 30 μM) completely inhibited extracellular ATP-induced DNA synthesis in VVEC (Fig. 5). A similar inhibitory effect was observed when cells were stimulated with extracellular ADP and selected nonhydrolyzable nucleotide analogs (data not shown).

Fig. 5.

Ca²⁺ is required for nucleotide-induced DNA synthesis in VVEC. Growth-arrested VVEC (72 h in serum-free DMEM) were preincubated with or without BAPTA-AM (10 and 30 μM) for 30 min and then stimulated with ATP (100 μM) in the presence of 0.125 μCi of [³H]thymidine for 24 h. Incorporated radioactivity was measured in total cell lysates using a beta counter. Data are means ± SE from 4 experiments conducted on 2 distinct VVEC populations. *P < 0.001 vs. nonstimulated control. #P < 0.001 vs. ATP-stimulated cells.

P2Y1 and P2Y13 receptors are involved in mitogenic responses in VVEC.

To further explore the involvement of P2Y1 and P2Y13 receptors in VVEC mitogenesis, we examined the role of these receptors in DNA synthesis. We found that pretreatment with the P2Y1 receptor antagonist MRS2179 resulted in a significant >50% inhibition of DNA synthesis in response to ATP, MeSATP, and MeSADP. A lesser inhibitory effect was observed in ATPγS- and ADP-stimulated cells and was not observed in ADPβS-stimulated cells (Fig. 6A). Pretreatment of VVEC with MRS2211 had an even larger effect on nucleotide-mediated DNA synthesis. The most pronounced inhibitory effects were observed in VVEC stimulated with MeSATP, ATPγS, ADP, and ADPβS (>50%). MRS2211 also suppressed the effects of ATP and MeSADP. Furthermore, PTx treatment had an inhibitory effect on DNA synthesis (by 32–68%) in responses to all nucleotides and on activation of the intracellular kinases ERK1/2 (by 28–57%), Akt (by 22–73%), and S6 (by 10–38%), indicating a substantial contribution of Gα_i-coupled P2Y13 receptors in the integrated growth response in VVEC (Fig. 6B; see Supplemental Fig. 2). (Supplemental data for this article is available online at the American Journal of Physiology-Cell Physiology website.)

Fig. 6.

P2Y1 and P2Y13 receptor agonists induce DNA synthesis in VVEC in a Gα_i-dependent and -independent manner. Growth-arrested VVEC (72 h in serum-free DMEM) were preincubated with MRS2179 (10 μM, 30 min; A), MRS2211 (10 μM, 30 min; A), or PTx (100 ng/ml, 18 h; B) or remained untreated and were then stimulated with the indicated nucleotides (100 μM) in the presence of 0.125 μCi of [³H]thymidine for 24 h. Incorporated radioactivity was determined in total cell lysates as described in materials and methods. Data are means ± SE from 4–6 independent experiments conducted on 3 distinct VVEC populations. *P < 0.05; **P < 0.01 vs. nucleotide-stimulated cells.

Finally, a role of P2Y1 and P2Y13 receptors in mitogenic activation of VVEC was examined by agonist-dependent activation of intracellular mitogenic signaling pathways in the presence and absence of MRS2179 and MRS2211. Western blot analysis of phospho-ERK1/2, phospho-Akt, and phospho-S6 ribosomal protein (Fig. 7 and Table 2) showed that P2Y1 and P2Y13 purinergic receptor antagonists MRS2179 and MRS2211 exhibited an overlapping but slightly different inhibitory effect on nucleotide-mediated responses in VVEC. MRS2179 (10 μM) had the most dramatic inhibitory effect on MeSATP-induced ERK1/2 phosphorylation; but also had an effect on ATP-, MeSATP-, and MeSADP-induced Akt phosphorylation and MeSATP-, ATPγS-, MeSADP-induced S6 ribosomal protein phosphorylation. Less significant inhibition of ERK1/2 and Akt phosphorylation was observed in VVEC stimulated with ADPβS. MRS2211 (10 μM) effectively inhibited ERK1/2 and Akt phosphorylation in VVEC stimulated with MeSADP and ADPβS, and S6 protein phosphorylation in cells stimulated with MeSATP and MeSADP. Because MRS2179 and MRS2211 act as competitive antagonists at P2Y1 and P2Y13 receptors, we examined the effects of these compounds under conditions when cells were stimulated at the lower (1 μM) concentration of some agonists (see Table 2, values in parentheses). The data showed that MRS2179 had a more potent inhibitory effect on ATP- and MeSATP-induced ERK1/2 and Akt phosphorylation and a lesser effect on ATP- and MeSATP-induced S6 phosphorylation. MeSADP-induced responses remained without significant changes. MRS2211 exhibited a more potent antagonistic effect on ATP- and MeSATP-induced ERK1/2 phosphorylation and MeSATP-induced Akt and S6 phosphorylation. The responses to MeSADP also remained without significant changes. Collectively, the inhibitory effects of MRS2179 and MRS2211 indicate that both P2Y1 and P2Y13 receptors are involved in VVEC mitogenesis to some extent, through the activation of ERK1/2 and Akt/mTOR/S6 signaling pathways.

Fig. 7.

P2Y1 and P2Y13 receptors are involved in activation of mitogenic signaling pathways in VVEC. Growth-arrested VVEC were preincubated with MRS2179 (10 μM, 30 min) or MRS2211 (10 μM, 30 min) or remained untreated and were then stimulated with selected agonists (100 μM, 30 min) for 10 min. Phoshoproteins in total cell lysates (20 μg) were determined with antibodies to phospho-ERK1/2 (Tyr²⁰²/Thr²⁰⁴), phospho-Akt (Ser⁴⁷³), and phospho-S6 (Ser²³⁶). Data are representative of at least 3 independent experiments.

Table 2.

Contribution of P2Y1 and P2Y13 receptors to activation of mitogenic pathways in VVEC

Agonist	−MRS2179	+MRS2179	−MRS2211	+MRS2211
p-ERK1/2
Control	1 (1)	0.8 (1.4)	1 (1)	1 (1.3)
ATP	35 (8.6)	40 (2.8)	14.7 (8.6)	15 (3.6)
MeSATP	33.4 (10.2)	21 (2.5)	18.7 (10.2)	22.4 (2.4)
ATPγS	48	57	23.7	22
ADP	56.2	56.2	28.7	32.1
MeSADP	65 (11.9)	60.2 (10.3)	11.1 (11.9)	5 (7)
ADPβS	46	39	24.1	8
p-Akt
Control	1 (1)	0.5 (1.3)	1 (1)	0.4 (1.2)
ATP	7 (1.9)	4.2 (1.4)	2.3 (1.9)	2.6 (1.65)
MeSATP	4 (2.4)	2.4 (1.7)	2.2 (2.4)	3.3 (1.1)
ATPγS	6.2	6.6	3	3.1
ADP	6.2	7.1	3.3	4.1
MeSADP	4.1 (1.8)	3.1 (1.7)	4.5 (1.8)	2.3 (1.3)
ADPβS	11	8.5	1.6	0.7
p-S6
Control	1 (1)	0.73 (1.3)	1 (1)	0.2 (1.2)
ATP	2 (1.9)	1.7(1.7)	4.1 (1.9)	3.5 (1.8)
MeSATP	1.6 (2.4)	0.2 (1.7)	3.5	1.8 (1.1)
ATPγS	2.4	1.2	3.7	3.5
ADP	1.8	1.5	4.7	3.8
MeSADP	2.4 (1.8)	1.4 (1.7)	3.5	0.8 (1.3)
ADPβS	2.1	2.1	4.4	3.7

Values indicate the relative increase in ERK1/2, Akt, and S6 phosphorylation in response to selected agonists with (+) and without (−) the P2Y1 and P2Y13 specific antagonists MRS2179 and MRS2211, respectively. Cells were treated with P2Y1 and P2Y13 receptor antagonists and agonists as described in materials and methods. The activation of ERK1/2, Akt, and S6 was determined by Western blot analysis. Values without parentheses represent a 100 μM nucleotide concentration; values given in parentheses represent a 1 μM nucleotide concentration. The calculated values represent the degree of phospho-ERK1/2, phospho-Akt, and phospho-S6 increase in nucleotide-stimulated cells compared with that in nonstimulated cells. The densities of protein bands were normalized to GAPDH; normalized basal levels (control) were taken as having a value of 1.

DISCUSSION

Our previous studies demonstrated potent angiogenic effects of extracellular ATP in pulmonary artery VVEC (21, 52). In this study, we further investigated the role of purinergic signaling and intracellular Ca²⁺ in VVEC mitogenesis. Data obtained in different endothelial cell models demonstrate that, as a second messenger, Ca²⁺ is involved in the regulation of various physiological and pathological cell responses. Purinergic receptor-mediated increase in [Ca²⁺]_i has been recognized as the intracellular response leading to NO production in the vascular endothelium and regulation of vasorelaxation and vascular permeability (15, 26, 30, 43). However, the cellular mechanisms and the role of Ca²⁺ signaling in angiogenic activation of endothelial cells in response to extracellular nucleotides remain largely unexplored.

Ca²⁺-mediated activation of several signal transduction pathways has been described in endothelial and epithelial cells (2, 4, 17, 20, 27, 48, 52). The importance of Ca²⁺ signaling in angiogenesis is supported by evidence that inhibitors of Ca²⁺ entry and inhibitors of TRP channels suppress cell proliferation and angiogenesis in vitro and in vivo (reviewed in Ref. 38). It also has been demonstrated that ATP-induced Ca²⁺ influx involves a store-operated Ca²⁺ entry, or capacitative Ca²⁺ entry (36). This mechanism is a dominant signaling pathway in nonexcitable cells (47) and involves the opening of Ca²⁺-permeable channels in the plasma membrane in response to the depletion of intracellular Ca²⁺ stores (reviewed in Ref. 41). The results of our study demonstrate that nucleotide-induced elevation in [Ca²⁺]_i in VVEC includes both the release from intracellular stores and Ca²⁺ influx from the extracellular milieu. Most likely, this process involves IP₃ receptor-coupled Ca²⁺ channels in the endoplasmic reticulum, as described in other nonexcitable cells featuring store-operated Ca²⁺ entry (12, 33). However, this is probably not the only mechanism of ATP-mediated Ca²⁺ signaling in VVEC, since an increase in nuclear [Ca²⁺]_i observed in our experiments (Figs. 1B and 2A) indicates the presence of additional Ca²⁺ pathways. These possibilities are currently under study in our laboratory.

Our study revealed multiphased Ca²⁺ fluctuations exhibited by individual VVEC following ATP stimulation. Such Ca²⁺ fluctuation patterns or even more stable oscillatory responses at a single-cell level have been shown in several epithelial and endothelial cell types. These responses are believed to play a role in the generation of intercellular Ca²⁺ signaling “waves” providing an intercellular communication to synchronize regulatory mechanisms at the tissue level (16, 17). Importantly, a recent study demonstrated that extracellular ATP-induced Ca²⁺ cytosolic oscillations in lung epithelial and endothelial cells mediate paracrine effects, including NO production and lung expansion in the model of lung mechanical ventilation (30). However, the fact that in our experimental setting only a part of VVEC responded in this manner may be attributed to different differentiation stages or individual cell phenotypes, as well as to variations in confluence of the cell culture, which affects the number of gap junctions responsible for intercellular signaling. Nevertheless, this interesting phenomenon in VVEC certainly deserves additional investigation in specifically optimized cell culture conditions.

Real-time measurements of intracellular Ca²⁺ in VVEC performed in our study demonstrated increases in Ca²⁺ levels in both cytoplasm and nucleus on ATP stimulation. Moreover, our results indicate that Ca²⁺ signaling is required for nucleotide-mediated VVEC proliferation. These findings are in agreement with the reports showing that elevation of nucleoplasmic Ca²⁺ is involved in growth factor-induced proliferative responses (4, 44), necessary for activation of the nuclear PLC/PKC pathway and transcription factors, and in centrosome separation during early prophase (22, 23, 40, 44). The importance of nucleoplasmic Ca²⁺ in cell proliferation is also supported by reports showing bradykinin B₂ and metabotropic glutamate mGluR5 receptor expression in hepatocyte nuclei and in nuclei of primary neurons, respectively (reviewed in Ref. 4). Our future studies need to explore a possibility of nuclear localization of purinergic receptors and their role in mediating VVEC nucleoplasmic Ca²⁺ responses and proliferation.

The sensitivity of VVEC to stimulation with extracellular nucleotides suggests that multiple purinergic receptors contribute to the angiogenic phenotype of these cells. Indeed, RT-PCR analysis revealed mRNA for several metabotropic P2Y receptors (PY1, P2Y2, P2Y4, P2Y13, and P2Y14), ionotropic P2X receptors (P2X2, P2X5, and P2X7), and P1 adenosine receptors (A1, A2b, and A3). In line with this observation, expression of P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2X4, P2X5, and P2X7 has been previously reported in vascular endothelium (8, 37). The rank order of potency (MeSATP ≥ MeSADP ≥ ADP = ATP > ADPβS ≥ ATPγS = BzATP > UTP > UDP) of extracellular nucleotides to induce Ca²⁺ responses suggests a contribution of more than one receptor with the overlapping pharmacological profile. The application of MRS2179, a selective antagonist of P2Y1 receptor, and MRS2211, a selective antagonist of P2Y13 receptor, strongly blocked the agonist-induced Ca²⁺ responses in VVEC. In addition, treatment with PTx demonstrated an involvement of Gα_i-coupled P2 receptors in mediating Ca²⁺ responses. Together, these studies revealed a predominant role of P2Y1 and P2Y13 receptors and, to a lesser degree, contribution of P2Y2 and P2X7 receptors in observed changes in [Ca²⁺]_i. Interestingly, we observed some differences in relative efficacy of MeSATP, MeSADP, ATP, and ADP on an increase of [Ca²⁺]_i in distinct VVEC populations, pointing to possible variations in a relative expression, conformation state, and/or functional involvement of P2Y1, P2Y2, and P2Y13 receptors. In contrast to nucleotide di- and triphosphates, AMP and Ado did not elevate [Ca²⁺]_i and were much less effective in activating selected kinases and proteins belonging to mitogenic pathways (Supplemental Fig. 1).

Although our results demonstrate that UTP, a ligand for P2Y2 receptors, is capable of raising intracellular Ca²⁺ levels in VVEC, it did not stimulate VVEC proliferation (21). Importantly, in contrast to extracellular ATP and ADP, extracellular UTP and UDP are less hydrolysable by ectonucleotidases/NTPDases, implying their possible role in longer term effects in endothelial cells. The involvement of ATP and UTP, and hence P2Y2 and, possibly, P2Y4 receptors, in endothelial cell migration has been previously reported (29). The role of extracellular UTP and UDP in VVEC angiogenic responses such as migration, invasion, and tube formation needs to be further investigated. In addition, our study demonstrated that bradykinin, but not endothelin, 5-HT, angiotensin II, or norepinephrine, elevated [Ca²⁺]_i in VVEC (data not shown). These observations suggest that VVEC exhibit a distinct endothelial phenotype with selective reliance on extracellular nucleotides as angiogenic stimuli. Complementary to our study, ATP-mediated Ca²⁺ responses were detected in tumor endothelial cells and were associated with an in vitro tubulogenic effect (19).

Potent and selective purinergic receptor agonists and antagonists represent an advanced pharmacological approach to establish a functionality of individual P2 receptor subtypes. Our study demonstrates the ability of P2Y1 and P2Y13 agonists to activate ERK1/2, Akt, and S6 ribosomal protein signaling pathways in VVEC. Furthermore, using newly developed selective P2Y1 and P2Y13 antagonists, MRS2179 and MRS2211, respectively, we confirmed a role of these receptors in nucleotide-induced signaling in VVEC. Although the association of some P2Y receptor subtypes with ERK, phosphatidylinositol 3-kinase/PDK/PKCζ, and Ca²⁺/PLC pathways and cross talk between P2Y1 and VEGF receptor 2 signaling in endothelial cells have been reported (1, 9, 29, 35, 45), specific signaling functions of P2Y13 receptors have not been investigated. The expression of the P2Y13 receptors was found to be the highest in spleen, brain, and lymph nodes, as well as in hematopoietic, neuronal, and red blood cells. Nucleotides such as ADP and MeSADP are the most potent ligands for human and rat P2Y13, whereas MeSATP is the most potent ligand for the bovine P2Y13 receptors (7, 49). In red blood cells, ADP-induced activation of P2Y13 receptors decreases ATP release and diminishes the vasodilatory response to hypoxia (51). In mouse and human tissues, P2Y13 receptor was identified as a Gα_i-coupled orphan receptor, SP174, pointing out a possible role for this receptor in cell proliferation (55). The expression and function of P2Y13 receptors in endothelial cells, as well as in bovine tissue, have not been documented yet. The results of our study present new evidence for the expression of the P2Y13 receptors in bovine pulmonary artery VVEC and their functional role in Ca²⁺ and mitogenic signaling responses. As outlined above, PTx was used to distinguish between a contribution of Gα_i-coupled receptors vs. Gα_q-coupled receptors to nucleotide-induced Ca²⁺ increase, DNA synthesis, and protein phosphorylation. Our results suggest that Gα_q-coupled P2Y1 receptors are predominantly involved in acute Ca²⁺ responses, whereas both Gα_q-coupled P2Y1 (and possibly P2Y2) receptors and Gα_i-coupled P2Y13 receptors are involved in VVEC growth responses. Keeping in mind that isolated VVEC exhibit highly proliferative potential and that these cells may be enriched with endothelial progenitor cells, which may contribute to a neovascularization process, future studies are needed to validate the expression patterns of P2Y13 receptors within VVEC populations, identify G protein α-subunits coupled to P2Y13 receptors, and unveil signal transduction pathways leading to angiogenic activation of VVEC. Given that elevated levels of extracellular ATP and ADP are observed under various pathological conditions associated with hypoxia and inflammation, our findings indicate that P2Y1 and P2Y13 receptors may be targeted by these nucleotides and therefore may play a role in regulating angiogenic expansion of the vasa vasorum and, potentially, other microvascular endothelial cell types.

GRANTS

This work was funded by American Heart Association Grant 0665464Z (to E. V. Gerasimovskaya) and National Heart, Lung, and Blood Grants R01 HL086783 (to E. V. Gerasimovskaya), PPG HL014985 (to K. R. Stenmark), and SCCOR HL084923 (to K. R. Stenmark).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).