Non-NF-kappaB elements are required for full induction of the rat type II nitric oxide synthase in vascular smooth muscle cells

Abstract

We have investigated the role of the NF-kappaB binding sites and other promoter elements beyond NF-kappaB in iNOS induction in rat vascular smooth muscle cells (SMC). Rat aortic SMC transfected with iNOS promoter constructs with either mutation or deletion of the downstream NF-kappaB site exhibited about 50% reduction in promoter activity in response to a cytokine mixture, whereas either mutation or deletion of the upstream NF-kappaB site reduced promoter activity by 90%, suggesting that the latter site is the most important, and that co-existence of two NF-kappaB sites is necessary for iNOS induction. Nuclear NF-kappaB activity was robustly induced by TNF-alpha. However, TNF-alpha alone did not induce iNOS promoter activity, protein expression, or nitrite production, indicating that NF-kappaB activation alone is not sufficient for iNOS induction. The construct up to -890 bp, containing the downstream NF-kappaB site, exhibited little response to cytokines. The construct up to -1.0 kb, containing the two NF-kappaB sites exhibited only 22% of full promoter activity. The regions -1001 to -1368 bp and -2 to -2.5 kb contributed an additional 43 and 22% promoter activity, respectively. Internal deletion or reversal of the orientation of -1001 to -1368 bp in the full promoter resulted in 40% reduction in promoter activity. These data suggest that the co-existence of two NF-kappaB sites is essential for core promoter activity, but that full induction of the rat SMC iNOS gene requires other elements located between -1.0 to -1.37 and -2.0 to -2.5 kb of the promoter.

Figure 1

Induction of rat iNOS promoter activity, protein expression, nitric accumulation and cytotoxicity by cytokines in rat aortic SMC. (a) Confluent RASMC were incubated with individual cytokines or combinations for 24 h and nitrite production was determined by the Griess reaction. Data are presented as mean±s.e.mean, n=5–7 wells per bar, from two separate experiments and two separate RASMC harvests. (b) The 3.2 kb rat iNOS promoter was transfected into RASMC, stimulated with individual cytokine or combinations for 6 h and then luciferase activity was analysed. Data are presented as mean±s.e.mean, n=3–4 wells per bar from one experiment and one RASMC harvest. (c) Confluent RASMC were incubated with individual cytokine or combinations for 24 h and the iNOS protein expression was analysed by Western blot. Data represents three separate experiments with cells obtained from two different harvests. (d) Confluent RASMC were treated as in (a), and cells were incubated with MTT. The colour development was by addition of DMSO. The ODs from controls are expressed as 100%. Data are presented as mean±s.e.mean, n=6–8 wells per bar from three separate experiments and two separate RASMC harvests. No significant difference among Control, TNF-α, IFN-γ and TNF-α+IFN-γ in luciferase activity and nitrite production; ^*P<0.05, compared to control; #P<0.05, compared to IL-1β; and $P<0.05, compared to IL-1β+TNF-α or IL-1β+IFN-γ by ANOVA.

Figure 2

The effects of mutating or deleting the down-stream (D-κB-Mut. or Del.) and the up-stream NF-κB sites (U-κB-Mut. or Del.) on the induction of iNOS promoter in response to CM in RASMC. In the 3.2 kb iNOS promoter, downstream NF-κB or upstream NF-κB site was mutated (GGG to CTC) or deleted, and the construct DNA was transfected into RASMC, stimulated with CM for 6 h and then luciferase activity was analysed. Data are presented as mean±s.e.mean, n=7–33 wells per bar from 3–5 separate experiments with RASMC obtained from four harvests. ^*P<0.05 compared to wild-type; and #P<0.05 compared to D-κB-Mut or D-κB-Del by ANOVA.

Figure 3

Cytokine induction of nuclear NF-κB DNA binding activity in RASMC by EMSA. RASMC were stimulated with individual cytokine or combinations for 30 min, and nuclear extracts were prepared. EMSA was performed by incubating an equal amount of the nuclear protein with ³²P-labelled NF-κB oligonucleotides. Bound and free oligonucleotides were separated by electrophoresis on a native gel. (a) Autoradiograph of nuclear NF-κB activity from RASMC in response to the stimulation of various cytokines. The figure represents two separate experiments with cells obtained from two harvests. (b) Quantification of nuclear NF-κB activity by phosphorimage. Data are expressed as percentage of maximal NF-κB activity by CM and presented as mean±s.e.mean, n=2–4 wells per bar from four separate experiments with cells from three harvests. No significant difference between IL-1β and TNF-α; ^*P<0.05, compared to control; and #P<0.05, compared to TNF-α or IL-1β alone by ANOVA.

Figure 4

The maximal induction of rat iNOS promoter requires more than two NF-κB binding sites in RASMC. Deletion constructs with different lengths of the 5′-flanking regions of iNOS gene were generated and transfected into RASMC. Luciferase activity was analysed after stimulation with CM for 6 h. Data are expressed as per cent of full promoter activity (3.2 kb wild type) induced by CM and presented as means±s.e.mean with number of wells per data point shown in parentheses, derived from 12 experiments with RASMC from seven separate harvests. ^*P<0.05, compared to −893 bp; #P<0.05, compared to −1000 bp; and $P<0.05, compared to 1368 bp by ANOVA.

Figure 5

The effect of internal deletion of the fragment −1001 to −1368 bp on iNOS promoter induction in RASMC. (a) Schematic representation of internal deletion (−1001 to −1368 bp) constructs of the rat 3.2 kb iNOS promoter. (b) Wild-type or deletion constructs were transfected into RASMC, and luciferase activity was analysed after stimulation with CM for 6 h. Data are expressed as per cent of the promoter activity (3.2 kb wild-type) induced by CM and presented as means±s.e.mean, n=9 wells per bar, obtained from three separate experiments with cells from three harvests. ^*P<0.05, compared to wild type by ANOVA.

Figure 6

The effect of reversed orientation of the −1001 to −1368 bp fragment on induction of iNOS promoter activity in RASMC. (a) Schematic representation of the reversed orientation of −1001 to −1368 bp fragment of the rat iNOS promoter. (b) Activities of the wild-type and iNOS promoter with the reversed orientation of the 368 bp fragment in response to CM. Data are expressed as per cent of full promoter activity (3.2 kb wild-type) induced by CM and presented as means±s.e.mean, n=3–12 wells per bar from two separate experiments with cells obtained from two harvests. ^*P<0.05, compared to wild-type; #P<0.05, compared to −1368 bp or original orientation by ANOVA.

Figure 7

Deletion analysis of the 368 bp fragment on the induction of iNOs promoter activity. Deletion constructs with different lengths at −1001 to −1368 bp were generated and transfected into RASMC. Luciferase activity was analysed after stimulation with CM for 6 h. Data are expressed as per cent of the −1368 bp promoter activity induced by CM and presented as means±s.e.mean, n=6–21 wells per bar from three separate experiments with RASMC from two harvests. ^*P<0.05, compared to the −1368 bp; #P<0.05, compared to the −1191 or −1268 bp; and $P<0.05, compared to the −1100 bp by ANOVA.