Mutant glucocorticoid receptor binding elements on Interleukin-6 promoter regulate dexamethasone effects

Glucocorticoid has been widely used as an important modulator for clinical infectious and inflammatory disease. Glucocorticoid receptor (GR) is a transcription factor belonging to the family of nuclear receptors, regulated anti-inflammatory process and the release of pro-inflammatory cytokines. Five putative GR and other transcription factor binding sites on interleukin (IL)-6 promoter were identified and dexamethasone could reduce LPS-induced IL-6 release. Among them, the mutant transcriptional factors NF-κB, AP-1, and Sp1-2 site decreased the basal and effects of lipopolysaccharide (LPS)-induced IL-6 promoter activities in different responses. GR2/3 seemed to be an important role in both basal and inducible promoter activities in LPS-induced inflammation. We concluded that the selective GR2/3 modulators may have agonistic and antagonistic combined effects and activate important signaling pathway during LPS-stimulated inflammatory process.


Introduction
Severe sepsis has been assumed to relate with immune dis-equilibrium. The immune dysregulation in the early phase of sepsis is result from inadequate endogenous glucocorticoid mediated regulation of nuclear factor-κВ (NF-κВ) activation, which leads to its over-expression and release massive pro-inflammatory cytokines (the socalled "cytokine storm") [1] . "Glucocorticoid sensitivity" in critically ill patient with septic shock is also associated with disease severity and outcome [2]. Administration of low low-dose steroid in the early septic phase, although could not reduce 28-day mortality, it could reduce duration of shock and mechanical ventilator-dependent [3][4][5]. The effect of low-dose steroid in short-and longer mortality of septic patients remain controversial. Glucocorticoids (GCs) regulate many biological processes through their intracellular glucocorticoid receptors [GRs]. The GCs diffuse through the cell membrane bind to its GRs and the activated GRs complex translocated into the nucleus and expressed inflammatory mediators [6].
GCs are considered as immunosuppressive and anti-inflammatory agents and adjuvant therapy for patients with severe sepsis and septic shock. The activated GR is proposed to directly and indirectly interact with promoters to perform the negative regulation of cytokine gene expression [7,8]. Selective knockout of GR gene results in 5 sensitive to LPS treatment, whereas, endogenous GCs decrease LPS-induced inflammation via inhibition of cytokine gene expression, such as IL-12 [9]. GCs also negatively regulate IL-6 gene expression through downregulation of its promoter activity in various tissues [10,11]. The regulation of several cytokine gene expressions by activated GRs, which are interact with the GCs, may be through binding to genes or indirective interaction with other transcription factors, in particular NF-κB and activator protein-1 (AP-1) complexes [7,[12][13][14]. Transcription factors, such as NF-κB and AP-1, are known to mediate gene expression during inflammatory reaction extensively, especially in inflammatory cytokines, MMPs and cyclo-oxygenase-2 [12,15,16]. The synthetic glucocorticoid, dexamethasone (DEX), has been shown to inhibit inflammatory response via downregulation of transcription factor AP-1 in human lung epithelial cells [16].
We have previously demonstrated that the increase of macrophage migration inhibitory factor (MIF) was noted in severe sepsis such as Vibrio vulnificus-infected model [17]. In V. vulnificus-infected mice, MIF could regulate interleukin (IL)-6 in a time-dependent manner. Serum MIF regulates the NF-κB to modulate the release of IL-6 in transcriptional level. We further to investigate the mechanisms of transcription factors involved in regulating IL-6 promoter. We attempt to elucidate the roles of putative transcription factor binding sites, such as NF-κB, Sp1 and AP1, play in LPS-7

Cell Culture
Two cell lines were used in our studies. One was the RAW264. 7 8 Total RNA was isolated from the cultured cells using TRIzol reagent (Invitrogen).

RNA isolation and semi-quantitative RT-PCR analysis
RT-PCR was performed as described previously (Chang and Huang, 2005). Briefly, total RNA (2 μg) was reverse-transcribed into cDNA in 20 μl of 1X first strand buffer PCR products were analyzed on a 1.2% agarose gel.

Plasmid constructs
pGL4.10-Basic and pRL-TK luciferase reporter vectors (Promega, Madison, WI) were used for the promoter reporter assays. The promoter construct of the IL-6 gene was generated by PCR amplification of a 349 bp fragment that was then inserted the KpnΙ/HindⅢ fragments of the human IL-6 gene promoter into pGL4.10-Basic vector.
The resulting plasmid was named pGL4.10-IL-6 construct. Promoter constructs containing nucleotide substitutions in the sequence motifs of AP1, NF-κB, Sp1 and 5 GR sites were individually generated by PCR amplification with primer pairs spanning 9 the mutant nucleotides according to the protocol of site-directed mutagenesis by overlap extension [18]. The GR cDNA fragment containing NheI and XbaI sites was cloned by overlap extension using RT-PCR and inserted into pCMS-EGFP, a mammalian expression vector. The sequences of primers used in cloning IL-6 promoter and GR cDNA fragments and site-directed mutagenesis are shown in supplemental information.

Transient transfection
Exponentially growing RAW264.7 cells were seeded at a density of 2.

Dual-luciferase assay
After LPS (1μg/ml) and/or DEX (10μM/ml) treatment for 24 h, the cells were washed with PBS and then lysates were prepared by scraping the cells from plates in the presence of 1× Passive lysis buffer (Promega). Luciferase assays were performed according to Dual-Luciferase Assay System (Promega) and then detected by a Sirius luminometer (Berthold Detection System, Pforzheim, Germany).

Nuclear protein extraction of human IMR-32 cells
IMR-32 cells were collected from culture dish and centrifuged for 5 minutes at 500 x g. The cell pellets were mixed with 10-fold volume of buffer A (10 mM HEPES (pH 7.9), 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM DTT, 1 tablet of protease and phosphatase inhibitor (Thermo Scientific, USA)) and incubated on ice for 8 minutes followed by shortly centrifuged for 10 seconds at 12,000 x g at 4 °C. The buffer A-treated cell pellets were then mixed with 2-fold volume of buffer C (20 mM HEPES (pH 7.9), 25% glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM DTT, 1 tablet of protease and phosphatase inhibitor) and incubated on ice for 16 minutes. After 12,000 x g centrifugation at 4°C, the supernatant containing the nuclear protein of IMR-32 cells were collected and stored at -80°C.

Gel electrophoretic mobility shift assays (EMSA)
In EMSA, The DNA probes with putative GR-binding sites were prepared by annealing the two complementary biotin-labeled single-strand oligonucleotides as Chromatin Immunoprecipitation (ChIP) Assay [19] ChIP assays were performed with an EZ ChIP kit purchased from Upstate

Statistical analyses
All values are expressed as means ± SD. Groups were compared using Student's twotailed unpaired t-test. A value of P < 0.05 was considered to be statistically significant.

Results
The LPS-induced morphology changes of RAW264.7 and its

IL-6 gene expression reversed by co-treated dexamethasone
To identify the LPS-induced effects of glucocorticoids on IL-6 gene expression, we treated LPS and DEX in RAW264.7 cells and measured the promoter activities and RNA levels of IL-6 gene. We found that the morphology of RAW264.7 cells changed under LPS treatment (Fig 1A). To study the regulation of IL-6 gene, we constructed a 349-bp promoter fragment of IL-6 gene into the pGL4.10-Basic vector, which contains a reporter luciferase gene for measuring promoter activity. We also used RT-PCR to detect IL-6 gene transcript. We found that the promoter activities and mRNA levels were induced by LPS and reversed by DEX (Figs 1B and C). However, treatment of DEX alone did not change the cellular morphology, promoter activities and mRNA levels of IL-6 gene. These results indicate that glucocorticoids modulate LPS-induced morphological change and IL-6 gene expression.

Different changes of promoter activities in mutant putative sites on IL-6 promoter by using site-directed mutagenesis method
To predict which proteins bind to IL-6 promoter to mediate the gene expression, we searched the transcription factor binding sites on IL-6 promoter region using 14 bioinformatics tools and found several putative sites, such as AP-1, NF-κB, Sp1 and 5 GR sites. The sequences of these sites show high similarity among different species (Supplemental material, Fig S1). To clarify which sites are important for regulating IL-6 gene expression, we used the site-directed mutagenesis to construct the mutant IL-6 promoter-reporter vectors and transfected theses constructs to RAW264.7 cells.
Mutation of AP-1 site decreased the basal and effects of LPS-induced and addition of DEX in IL-6 promoter activities (Fig 2A). Importantly, mutation of NF-B site dramatically decreased the promoter activities ( Fig 2B). There are two Sp1 sites on IL-6 promoter, denoted Sp1-1 and Sp1-2. Mutation of Sp1-2 but not Sp1-1 sites decreased IL-6 promoter activity, suggesting these two sites exert differential function in IL-6 expression in this cell line (Fig 2C). These results suggest that AP-1 and NFB sites are important for IL-6 express in RAW264.7 cells but only one Sp1 site is partially involved in this process.

The basal and inducible changes of IL-6 promoter activities in mutant five GR binding sites and the reversed effect of dexamethasone treatment
There are five putative GR binding sites on IL-6 promoter. We transfected the mutant promoter constructs of GR sites to detect which sites are important for the IL-6 promoter activities and the results showed that mutation the GR2 and 3 sites decreased the basal and LPS-induced promoter activities. Furthermore, the DEXreversed effects were also altered. However, mutation of the GR1, 4 and 5 sites did not affect the effects of LPS and DEX treatments (Fig 3). These results suggest that GR2 and GR3 are important for IL-6 promoter activities.

The changes of IL-6 promoter activities and gene expression under over-expression of GR
To understand the role of GR in regulating the expression of IL-6, we over-expressed GR in RAW264.7cells by transfecting various doses of GR cDNA constructs and found the IL-6 promoter activities were increased in a dose dependent manner ( Fig   4A). To identify if GR binds to IL-6 promoter, we used ChIP and EMSA to investigate the protein-DNA binding in vitro and in vivo, respectively. In the ChIP assay, GR can bind to IL-6 promoter indeed (Fig 4B). In the EMSA assay, the sequence analysis shows that GR2 and 3 overlap partially, so we used one probe containing these two sites for EMSA. The EMSA showed that the probe of GR 2/3 site but not others presents shifted bands. Furthermore, we used GR binding motif of elastin promoter as cold probes for competition assay and found the shifted bands were eliminated (Fig 4C), suggesting that GR binds to GR2/3 sites. However, there are no sifted bands in EMSA using the probes containing GR1, 4 and 5 sites. 16 According to these results, we confirmed that GR binds to GR2/3 sites but not others on the IL-6 promoter region.

LPS-stimulated IL-6 expression via GRs
We proposed the schematic model for DEX negative regulates LPS-stimulated IL-6 expression in RAW 264.7 cells (Fig 5). The LPS stimulates nuclear IL-6 promoter activity through NF-κB pathway and the counter-regulation of NF-κB by the GR was proposed on IL-6 promoter.  [20].

Discussion
The molecular mechanism of GCs through interaction with their receptor GRs to regulate the inflammation responses has been well-studied. GCs bind to GRs and are transported into the nucleus subsequently to form a dimer and also combinations of transcriptional complexes, such as AP-1 and NF-κB, either in transrepression or in transactivation, which are responsible for anti-inflammatory action of GCs [21,22].
The results from previous studies suggest that the treatment of naïve monocytes with fluticasone or dexamethasone, did not cause a global suppression of the activated monocytic functions instead of induction of cellular differentiation with an antiinflammatory phenotype [23,24]. Recently, growing evidences show that GCs could efficiently inhibit these processes by down-regulating pro-inflammatory mediators from macrophages and monocytes and their migration toward inflammatory stimuli. 18 GCs remove endo-and exogenous danger signals by an increased phagocytic capacity, and limit T-cell activation [25].
LPS also resulted in the increases of IL-6 and its promoter activity. The increase of LPS-stimulated IL-6 promoter activities could be suppressed by DEX, esp. in NF-κB and AP-1, Sp1-2, but not in Sp1-1. These results are consistent with previous wellknown studies [29,30]. It has been shown in the previous study that two Sp1 sites on IL-6 promoter and Sp1 binds to this region indeed in the human monocytes [31].
However, there is no report to elucidate which Sp1 site are important for IL-6 expression. We first discovered that only the Sp1-2 site is involved in IL-6 expression. Taken together, the transcription factors NF-κB, AP-1, Sp1-2 bind to IL-6 promoter and play important roles in basal and in inducible expression of the murine IL-6 gene.
In our study, we have found five putative GRs binding sites on IL-6 promoter and DEX could decrease the promoter activities of IL-6 GR binding sites. The promoter activity of IL-6 after LPS stimulation and the construct sequence, showed a significant increase of LPS-inducible promoter activity. Among them, GR2/3 seemed to be an important role in both basal and inducible promoter activities in LPS-induced inflammation. Comparing to GR1, GR4, GR5, they did not show the similar changes of LPS-inducible or basal promoter activities. Although previous study shows that recombinant GR binds to IL-6 promoter regions respect to GR1-5 elements in vitro [10], there were no significant shifted bands using the probes of GR1, GR4, GR5 for EMSA and neither changes of the promoter activities being found after mutation of these three sites (data not shown). It has been proposed that GR interacts with transcription factors, such as AP-1and NF-κB, to mediate the downstream proinflammatory genes, including IL-6 [32]. The exact targeted regulation of GR2/3 on IL-6 and the possible mechanism related to transcriptional synergism with GRE remain unknown. Whether GR2/3 plays a novel negative GRE [33], directly binding by GRα, the pivotal subunit of GR, which could involve the interactions between GRα and AP-1and NF-κB, needs further studies to identify.
Our results provide the possible mechanism of controversial steroids should be administrated in hemodynamic unstable septic patient whether in low-or high doses. The GR2/3 binding site might provide the important selection of candidate therapeutics in vulnerable septic patients in the future.