Lysophosphatidylcholine up-regulates human endothelial nitric oxide synthase gene transactivity by c-Jun N-terminal kinase signalling pathway

Abstract

Human endothelial nitric oxide synthase (eNOS) plays a pivotal role in maintaining blood pressure homeostasis and vascular integrity. It has recently been reported that mitogen-activated protein kinases (MAPKs) are intimately implicated in expression of eNOS. However detailed mechanism mediated by them remains to be clarified. In this study, eNOS gene transactivity in human umbilical vein endothelial cells was up-regulated by stimulation of lysophosphatidylcholine (LPC). The stimulation of LPC highly activated both extracellular signal-regulated kinase 1/2 (ERK1/2) and c-Jun N-terminal kinase (JNK), with differences in the dynamic processes of activation between them. Unexpectedly, p38 MAPK could not be activated by the stimulation of LPC. The activation of JNK signalling pathway by overexpression of JNK or its upstream kinase active mutant up-regulated the transactivity of eNOS significantly, but the activation of p38 signalling pathway down-regulated it largely. The inhibition of either ERK1/2 or JNK signalling pathway by kinase-selective inhibitors could markedly block the induction of the transactivity by LPC. It was observed by electrophoretic mobility shift assay that LPC stimulated both SP1 and AP1 DNA binding activity to go up. Additionally using decoy oligonucleotides proved that SP1 was necessary for maintaining the basal or stimulated transactivity, whereas AP1 contributed mainly to the increase of the stimulated transactivity. These findings indicate that the up-regulation of the eNOS gene transactivity by LPC involves the enhancement of SP1 transcription factor by the activation of JNK and ERK1/2 signalling pathways and AP1 transcription factor by the activation of JNK signalling pathway.

Figure 1

Time-dependent effect of LPC stimulation on the activities of MAPKs. HUVEC-12 cells were treated with 40 μM/l of LPC at the different intervals of 0 min., 15 min., 30 min., 1 hr, 2 hrs, 3 hrs, 4 hrs, 5 hrs and 6 hrs, respectively (lanes 1–9). After harvest and lysis of the cells their cytoplasmic proteins were extracted, then Western blot was performed to detect unphosphorylated (upper panel) and phosphory-lated (lower panel) levels of ERK1/2 (A, B), JNK1/2 (C, D) and p38 (E, F), respectively. The phosphorylated levels of them were subjected to densitometric analysis. Each bar is means ± S.D. of 3 independent experiments. *P< 0.05, **P< 0.01, compared with LPC stimulation group for 0 min.

Figure 2

Effects of the selective inhibitors of MAPK signalling pathways on MAPK activities stimulated by LPC. HUVEC-12 cells were pretreated with selective inhibitors including PD98059 (50 μm/l), SB203580 (15 μm/l) and curcumin (30 μm/l) for 1.5 hrs, respectively, followed by stimulation with or without 40 μm/l of LPC (lanes 1–5). The cells were harvested, lysed and cytoplasmic proteins of them were extracted 2 hrs after the treatment of LPC, then Western blot was performed to detect unphosphorylated (upper panel) and phosphorylated (lower panel) levels of ERK1/2 (A, B), JNK1/2 (C, D) and p38 (E, F), respectively. The phosphorylated levels of them were subjected to densitometric analysis. Each bar is means ± S.D. of 3 independent experiments. M, molecular mass markers; Pur, purified ERK1/2 protein that bacterially expressed was phosphorylated or unphosphorylated by MEK as a positive control of ERK1/2; UV, HEK293 cells were treated with or without ultraviolet ray as a positive control of JNK1/2; Ani, C6 glioma cells were treated with or without anisomycin as a positive control of p38; control, HUVEC-12 cells were untreated with LPC as a control. *P< 0.05, **P< 0.01, compared with the control; ^#P< 0.05, ^##P< 0.01, compared with LPC group.

Figure 3

Effect of the stimulation of LPC and the activation of MAPKs on human eNOS transcriptional activity. (A) Cultured HUVEC-12 cells were co-transfected with pDseNOSRed plus pcDNA3 or ERK2 or JNK1 or p38 MAPK expression vector, respectively. (B) The cultured cells were co-transfected with pDseNOSRed plus pcDNA3 or MEK1(E) or MKK4(E) or MKK6b(E) active mutant, respectively. These cells were treated with or without stimulation of LPC (40 julMol/l) 45 hrs after the transfection. The red fluorescence-emitting cells were subjected to dynamic observation under an inverted fluorescence microscope. The fluorescence intensity of RFP expressed by the cells that represents the eNOS transactivity was quantitatively analysed using the fluorescence analysis software, Image-Pro Plus, 96 hrs after transfection. This result represents means ± S.D. of 3 independent experiments. *P< 0.05, **P< 0.01, compared with pDseNOSRed group.

Figure 4

Effects of the selective inhibitors of MAPKs on human eNOS transcriptional activity induced by LPC. (A) After transfected with pDseNOSRed, HUVEC-12 cells were treated without or with PD98059 (50 μMol/l) or SB203580 (15 μMol/l) or curcumin (30 μMol/l), respectively, for 1.5 hrs before treatment of LPC. (B) After transfected with pDseNOSRed plus pcDNA3 or JNK1 expression vector, the cells were treated without or with curcumin (30 μMol/l) at the same time. 45 hrs after the transfection, the cells were stimulated without or with LPC (40 μMol/l). The treatment of each group is shown as indicated in the figure. The red fluorescence-emitting cells were subjected to dynamic observation under an inverted fluorescence microscope. The fluorescence intensity of RFP expressed by the cells that represents the eNOS transactivity was quantitatively analysed using the fluorescence analysis software, Image-Pro Plus, 96 hrs after transfection. The result represents means ± S.D. of 3 independent experiments. **P< 0.01, compared with pDseNOSRed group; ^#P< 0.05, ^##P< 0.01, compared with pDseNOSRed plus LPC group (A) or compared with pDseNOSRed plus pcDNA3 and LPC group (B); ^AAP< 0.01, compared with pDseNOSRed plus JNK1 expression vector and LPC group.

Figure 5

SP1 and AP1 DNA binding activities induced by LPC stimulation. After treatment of LPC, HUVEC-12 cell nuclear extracts were prepared at the different intervals for detection of SP1 (A, B) and AP1 (C, D) DNA binding activity via EMSA. The extracts were incubated with a ³²P-labelled SP1 oligodeoxynucleotide probe or ³²P-labelled AP1 oligodeoxynucleotide probe and subjected to nondenaturing gel electrophoresis, followed by autoradiography. Lane 1, free probe; lane 2, the cells untreated with LPC served as a control; lane 3, 100-fold molar excess of cold probe of SP1 or AP1 was used to confirm SP1- or AP1-specific DNA binding; lane 4–10, the cells were treated with 40 μMol/l of LPC for 15 min., 30 min., 1 hr, 2 hrs, 3 hrs, 4 hrs and 6 hrs, respectively. SP1 or AP1 DNA binding is indicated by the upper arrow, and free probe is indicated by the lower arrow. The result represents means ± S.D. of 3 independent experiments. *P< 0.05, **P< 0.01, compared with the control.

Figure 6

Effects of the selective inhibitors of MAPK on DNA binding activities of SP1 and AP1 in HUVEC-12 cells stimulated with LPC. The nuclear extracts from HUVEC-12 cells treated without or with 40 μMol/l of LPC for 2 hrs were used for EMSA to detect effect of the specific inhibitors of MAPKs on SP1 and AP1 DNA binding activity by stimulation of LPC. Lane 1, free probe; lane 2, treated without LPC as a control; lanes 3∼4 (A, B), 50- and 100-fold molar excess of cold probe of SP1 was used to determine SP1-specific DNA binding; lane 3 (C, D) 100-fold molar excess of cold probe of AP1 was done; lanes 5 (A, B) and 4 (C, D), treated with LPC as a stimulation group; lanes 6∼8 (A, B) and 5∼7 (C, D), pretreated with PD98059 (50 μMol/l), curcumin (30 μMol/l) and SB203580 (15 μMol/l), respectively, for 1.5 hrs before the stimulation of LPC (40 μMol/l). The data represent means ± S.D. of 3 independent experiments. *P< 0.05, **P< 0.01, compared with the control; ^##P< 0.01, compared with the stimulation group.

Figure 7

Effects of JNK signalling pathway on the DNA binding activities of SP1 and AP1 in HUVEC-12 cells by treatment of LPC. The nuclear extracts from HUVEC-12 cells treated without or with 40 μMol/l of LPC for 2 hrs were used for EMSA to detect effect of a specific inhibitor of JNK on SP1 and AP1 DNA binding activity enhanced by overexpression of JNK1 in the presence or absence of stimulation of LPC. Lane 1, free probe; lane 2, treated without LPC as a control; lanes 3∼4 (A, B), 50- and 100-fold molar excess of cold probe of SP1 was used to determine SP1-specific DNA binding; lane 3 (C, D) 100-fold molar excess of cold probe of AP1 was done; lanes 5 (A, B) and 4 (C, D), transfected alone with a JNK1 expression vector; lanes 6 (A, B) and 5 (C, D), treated with 40 μMol/l of LPC 45 hrs after tansfected with the JNK1 expression vector; lanes 7∼9 (A, B) and 6∼8 (C, D), transfected with the JNK1 expression vector, then pretreated with PD98059 (50 μMol/l), SB203580 (15 μMol/l) and curcumin (30 μMol/l) for 1.5 hrs, respectively, followed by stimulation with LPC (40 μMol/l). The data represent means ± S.D. of 3 independent experiments. *P< 0.05, **P< 0.01, compared with the control; ^#P< 0.05, compared with the alone transfected JNK1 group; ^ΔΔP< 0.01, compared with the transfected JNK1 plus LPC group.

Figure 8

Effects of decoy ODNs of SP1, AP1 and SSRE on human eNOS transcriptional activity. The human eNOS promoter reporter gene construct, pDseNOSRed, was co-transfected with or without 5 pMol/l of wild-type ODNs of AP1, SP1 and SSRE (i.e. AP1w, SP1w and SSREw) or 5 pMol/l of mutant ODNs of them (e.g. AP1m, SP1m and SSREm), respectively, into HUVEC-12 cells to evaluate the contribution of AP1 (A), SP1 (B) and SSRE (C) in LPC-induced eNOS transcription. 45 hrs after the transfection, the cells were treated with or without 40 μMol/l of LPC. The red fluorescence-emitting cells were subjected to dynamic observation under an inverted fluorescence microscope. The fluorescence intensity of RFP expressed by the cells that represents the eNOS transactivity was quantitatively analysed using the fluorescence analysis software, Image-Pro Plus, 96 hrs after transfection. The result represents means ± S.D. of 4 independent experiments. *P< 0.05, **P< 0.01, compared with the pDseNOSRed group without the treatment of LPC; ^#P< 0.05, ^##P< 0.01, compared with the pDseNOSRed group treated by LPC.