Oct-1 cooperates with the TATA binding initiation complex to control rapid transcription of human iNOS

Abstract

Expression of the human inducible nitric oxide synthase (hiNOS) is generally undetectable in resting cells, but stimulation by a variety of signals including cytokines induces transcription in most cell types. The tight transcriptional regulation of the enzyme is a complex mechanism many aspects of which remain unknown. Here, we describe an octamer (Oct) element in hiNOS proximal promoter, located close to the TATA box. This site constitutively binds Oct-1 and its deletion abrogates cytokine-induced transcription, showing that it is indispensable though not sufficient for transcription. Increasing the distance between Oct and the TATA box by inserting inert DNA sequence inhibits transcription, and footprinting of this region shows no other protein binding in resting cells, suggesting an interaction between the two complexes. Chromatin immunoprecipitation assays detect the presence of Oct-1, RNA polymerase II and trimethyl K4 histone H3 on the proximal promoter in resting cells, confirming that the gene is primed for transcription before stimulation. RT-PCR of various fragments along the hiNOS gene shows that transcription is initiated in resting cells and this is inhibited by interference with Oct-1 binding to the proximal site of the promoter. We propose that, through interaction with the initiation complex, Oct-1 regulates hiNOS transcription by priming the gene for the rapid response required in an immune response.

Fig. 1

Constitutive presence of Oct-1 in human colon cancer cell lines. a EMSA of nuclear proteins (2 μg) from control (CT) or CM-stimulated (2 h) cells (CM), using a consensus Oct probe. Unlabelled probe (UP) was added in 50× excess as competitor and an anti-Oct-1 antibody (Ab) was used to identify the specific complex. b The Oct sites found in the promoter, described in Table 1, were used as probes in EMSA with nuclear extracts (2 μg) from CM-stimulated (+) and non-stimulated (−) cells. c Labelled proximal Oct site (P-Oct) was incubated with 2 μg nuclear extracts from control (CT) or stimulated (CM) cells. Specificity was confirmed with 50-fold excess of the un-labelled (UP) probe and the protein complex was identified using a specific anti-Oct-1 antibody (Ab). d Chromatin immunoprecipitation with specific anti-Oct-1 antibodies was performed on control and stimulated (CM) cells. Primers specific for the proximal promoter (P1) were used for the PCR which was run on 1.5% agarose gels

Fig. 2

Role of P-Oct in NOSII transcription. a The proximal Oct element (P-Oct) was deleted from the 7 kb hiNOS promoter and cells were transiently transfected with the full length (P7) or the deleted (OD) plasmids. Luciferase activity was measured in control cells (CT) or after 6 h CM stimulation (CM) and activity is reported as fold induction of non-stimulated cells for each plasmid. b Northern blot showing hiNOS mRNA accumulation in control (CT) cells and after 6 h cytokine stimulation (CM)

Fig. 3

NFκB activation by CM. a EMSA of nuclear proteins (2 μg) from control cells (CT) or cells stimulated with CM for 2 h, using a consensus NFκB probe. Components of the induced complexes are identified using specific antibodies as p50/p65 (P1) and p50/p50 (P2) by the formation of supershifts (S). NS non-specific band. b Transcriptional activity of NFκB was examined using a luciferase construct containing four tandem NFκB sites. Luciferase activity in transiently transfected control (CT) or 6 h stimulated (CM) cells is reported as fold induction. c Cells were transiently transfected with the 7-kb hiNOS reporter gene together with an empty vector (vector) or an expression vector for a non-phosphorylatable IκB (IκBM) and activity was determined in control (CT) or 6 h stimulated (CM) cells

Fig. 4

Interaction between P-Oct and the TATA box. a Schematic representation of the constructs introducing short sequences between the two sites. Transcriptional activity of the H3 hiNOS reporter gene and the modified constructs with 5 bp (H3 + 5) or 10 bp (H3 + 10) insertions, in response to CM or LPS stimulation for 6 h is shown. The activity of the full length P7 hiNOS reporter gene after 6 h CM stimulation is shown for reference. b In vivo footprinting of the hiNOS proximal promoter in non-stimulated cells. Genomic samples were obtained as described in “Materials and methods”. In vitro and in vivo samples were treated for ligation-mediated PCR

Fig. 5

Binding of Oct-1 and Pol II to hiNOS chromatin. Schematic representation of the localisation of the primer pairs on the hiNOS gene. ChIP assays were performed with specific antibodies to RNA Pol II, Oct-1 and H3K4me3 (MH3) as well as non-specific IgG on control (CT) and 2 h cytokine-stimulated (CM) cells. PCR with primers for the regions depicted was performed on the precipitated chromatin and on the input DNA (IN) and run on 1.5% agarose gels

Fig. 6

Transcription processivity of the hiNOS gene. a Schematic representation of the localization of the amplified fragments on the iNOS gene. RT-PCR of nuclear RNA for each fragment after the indicated CM stimulation was blotted and hybridised with labelled hiNOS probe. b Activation index obtained from nuclear run-on experiments and calculated for each probe after normalisation to the actin signal

Fig. 7

Oct-1 control of transcription initiation. Non-stimulated cells were transiently transfected with double-stranded oligonucleotides harbouring the P-Oct and U-Oct sites identified in Table 1. Total mRNA was extracted and RT-PCR performed with primers to the proximal region of the hiNOS gene (5′) corresponding to exon 1 and distal region (3′) corresponding to exons 11–17