Selective induction of endothelial P2Y6 nucleotide receptor promotes vascular inflammation

Abstract

During a systemic inflammatory response endothelial-expressed surface molecules have been strongly implicated in orchestrating immune responses. Previous studies have shown enhanced extracellular nucleotide release during acute inflammatory conditions. Therefore, we hypothesized that endothelial nucleotide receptors could play a role in vascular inflammation. To address this hypothesis, we performed screening experiments and exposed human microvascular endothelia to inflammatory stimuli, followed by measurements of P2Y or P2X transcriptional responses. These studies showed a selective induction of the P2Y(6) receptor (> 4-fold at 24 hours). Moreover, studies that used real-time reverse transcription-polymerase chain reaction, Western blot analysis, or immunofluorescence confirmed time- and dose-dependent induction of P2Y(6) with tumor necrosis factor α or Lipopolysaccharide (LPS) stimulation in vitro and in vivo. Studies that used MRS 2578 as P2Y(6) receptor antagonist showed attenuated nuclear factor κB reporter activity and proinflammatory gene expression in human microvascular endothelial cells in vitro. Moreover, pharmacologic or genetic in vivo studies showed attenuated inflammatory responses in P2Y(6)(-/-) mice or after P2Y(6) antagonist treatment during LPS-induced vascular inflammation. These studies show an important contribution of P2Y(6) signaling in enhancing vascular inflammation during systemic LPS challenge and implicate the P2Y(6) receptor as a therapeutic target during systemic inflammatory responses.

Figure 1

Endothelial P2 receptor expression after inflammatory stimulation. HMEC-1 cells were exposed to 10 ng/mL TNF-α for 24 hours. Receptor expression levels of the 15 known P2 receptors were determined in controls and TNF-α–stimulated cells by real-time RT-PCR. Data were calculated relative to the internal housekeeping gene β-actin and are expressed as mean ± SD fold change compared with the control (without TNF-α) (n = 3-4). Primers and PCR conditions are summarized in the supplemental data.

Figure 2

Endothelial P2Y₆ receptor up-regulation after TNF-α stimulation. HCAECs (A-B) or HMEC-1 cells (C-D) were exposed to TNF-α for the indicated time periods or with indicated doses. P2Y₆ receptor transcript was determined by real-time RT-PCR. Data were calculated relative to the internal housekeeping gene (β-actin) and are expressed as mean ± SD fold change compared with controls (without TNF-α) (n = 3-4). (E) HMEC-1 cells were exposed to LPS, IL-1α, and TNF-α for 24 hours. (F) HCAECs were grown to confluence on cover glasses and exposed to 10 ng/mL TNF-α for 24 hours. Cell layers were stained with antibodies specific for human P2Y₆ receptor and Alexa Fluor 488–coupled secondary antibody (green) or isotype controls and Alexa Fluor 488–coupled secondary antibody or Alexa Fluor 488–coupled secondary antibody only. DAPI (4′,6-diamidino-2-phenylindole) was used as nuclear counterstain (blue). Slides were kept on ice before image acquisition. Probes were analyzed by confocal microscopy with the use of Zeiss Laser Scanning Microscope LSM 710 and the Plan-Apochromat 63×/1.40 Oil Dic M27 objective lens with oil immersion. Zen software was used for acquisition and image processing (γ adjustment) applied equally to all images. One representative image of 3 is displayed. (G) HMEC-1 cells were exposed to TNF-α for the indicated time periods, and P2Y₆ receptor protein was determined by Western blotting. The same blot was stripped and reprobed for human β-actin as a control for protein loading. One of 3 representative Western blots is displayed.

Figure 3

Effect of P2Y₆ receptor antagonist MRS 2578 on endothelial NF-κB activity. To measure the NF-κB activity, vascular endothelia (HMEC-1) were transfected with 0.25 μg of either NF-κB promoter reporter (Clontech) or control pGL3 vector. Cells were exposed to the P2Y₆ receptor antagonist MRS 2578 either with or without TNF-α for the indicated time periods. As readout for NF-κB activity cells were lysed, and luciferase activity was determined relative to the total protein concentration. (A-B) Unstimulated NF-κB activity after exposure to indicated times or concentrations of MRS 2578. (C) NF-κB activity after stimulation with MRS 2578 (30 minutes) and after stimulation with TNF-α (10 ng/mL) for 2 hours. Results are displayed as mean ± SD (n = 3).

Figure 4

Effect of P2Y₆ receptor antagonist MRS 2578 on TNF-α–induced gene expression in vascular endothelia. Vascular endothelia (HMEC-1) were pretreated with vehicle (DMSO) or MRS 2578 (10μM) and after 30 minutes were exposed to TNF-α (10 ng/mL). Transcript levels of IL-8 (A), VCAM (B), or intercellular adhesion molecule (C) were determined by real-time RT-PCR relative to housekeeping gene β-actin 2 hours after TNF-α exposure (n = 4).

Figure 5

P2Y₆ receptor expression after intravenous LPS treatment and attenuated inflammatory responses in P2Y6^−/− mice after intravenous LPS exposure. (A-B) C57BL/6 mice received an intravenous injection of LPS (300 μg; E coli O26:B6) or vehicle control (phosphate-buffered saline [PBS]). Induction of P2Y₆ mRNA was seen in kidney (A) and heart (B). (C) C57BL/6 mice received an intravenous injection of LPS (300 μg; E coli O26:B6) or vehicle control (PBS). Abdominal aorta was stained with antibodies specific for murine P2Y₆ receptor and Alexa Fluor 488–coupled secondary antibody (green) or isotype controls and Alexa Fluor 488–coupled secondary antibody or Alexa Fluor 488–coupled secondary antibody only. DAPI (4′,6-diamidino-2-phenylindole) was used as nuclear counterstain (blue). Slides were kept on ice before image acquisition. Probes were analyzed by confocal microscopy with the use of Zeiss Laser Scanning Microscope LSM 710 and the Plan-Apochromat 20×/0.8 M27 objective lens with oil immersion. Zen software was used for acquisition and image processing (γ adjustment) applied equally to all images. White arrows indicate luminal side. One representative image of 3 is displayed. (D) Determination of plasma KC levels by enzyme-linked immunoabsorbent assay (ELISA) in C57Bl/6 wild-type mice treated with MRS 2578 and DMSO (solvent of MRS 2578) 2 hours after intravenous injection of LPS (300 μg; E coli O26:B6) or vehicle control (PBS) (wild-type control, n = 8; wild-type LPS, n = 8; P2Y₆^−/− control, n = 6; P2Y₆^−/− LPS, n = 6). (E) Determination of plasma KC levels by ELISA in C57Bl/6 wild-type versus P2Y₆^−/− mice 2 hours after intravenous injection of LPS (300 μg; E coli O26:B6) or vehicle control (PBS) (wild-type control, n = 10; wild-type LPS, n = 10; P2Y₆^−/− control, n = 8; P2Y₆^−/− LPS, n = 14). (F-G) Two hours after the LPS injection, mice were killed, and vascular organs (heart, kidney) were harvested. Transcript levels of VCAM were determined by real-time RT-PCR relative to the housekeeping gene β-actin in vascular organs (F) kidney (control, n = 3; LPS, n = 6) or (G) heart (control, n = 4; LPS, n = 6). All results are displayed as mean ± SD. (H-I) Determination of Evans blue concentration in kidney and heart tissue of C57Bl/6 wild-type versus P2Y₆^−/− mice 2 hours after intravenous challenge with 300 μg of LPS. (H) Kidney (wild-type control, n = 4; wild-type LPS, n = 10; P2Y₆^−/− control, n = 6; P2Y₆^−/− LPS, n = 12). (I) Heart (wild-type control, n = 4; wild-type LPS, n = 10; P2Y₆^−/− control, n = 6; P2Y₆^−/− LPS, n = 11).