Virus-Induced Differential Expression of Nuclear Receptors and Coregulators in Dendritic Cells: Implication in Interferon Production

Abstract

We investigated mRNA expression of 49 nuclear hormone receptors (NRs) and 35 transcriptional coregulators in mouse bone marrow-derived dendritic cells (DCs) upon infection with Newcastle Disease virus or murine cytomegalovirus. These viruses regulated mRNA expression of some NRs among which NOR1 and LXRα were highly induced at mRNA and protein levels. Exogenous expression of the latter NRs repressed IRF3- or IRF7-induced transactivation of the interferon β promoter and NDV infection further potentiated their repressive effect. The viral infection also significantly regulated mRNA expression of some coregulators, including HDAC1. Toll-like receptor ligands regulated NR and coregulator mRNA expression similar to the viruses. Thus, NRs and coregulators are integral components of DC-organizing anti-viral response wherein NOR1 and LXRα participate in regulating interferon production.

1. Introduction

Nuclear hormone receptors (NRs) are a family of DNA-binding transcription factors, which regulate a broad spectrum of physiological processes including cell cycle, cellular metabolism, organ homeostasis and embryonic development [1]. Ligand-dependent NRs associated with their lipophilic ligands, such as steroid hormones and various metabolic molecules, bind specific DNA sequences in the promoter region of their responsive genes, recruit coregulators and other cofactor molecules, and finally stimulate the transcription of downstream coding sequences [2]. Currently, 285 coregulators are reported and their activities are associated with various physiologic processes and pathologic conditions [2]. Some NR members, on the other hand, do not have ligands, thus they are called as orphan nuclear receptors [3].

Dendritic cells (DCs) play a pivotal role in the anti-viral immune response by activating innate and acquired immunity [4]. Virus-infected DCs stimulate naive T-cells in lymphoid organs and produce type I interferons (IFNα and β), which subsequently stimulate the expression of hundreds of interferon-responsive genes at various local effector cells [5–6]. Among the DC family, the conventional (cDCs) and the plasmacytoid (pDCs) DCs have been described [7]. cDCs are more potent than pDCs in antigen presentation and secretion of the T-helper (Th) 1 cytokine interleukin (IL)-12, while pDCs are specialized to produce tremendous amounts of type I IFNs [7]. DCs employ pattern recognition receptors, the toll-like receptors (TLRs), to detect pathogen components [4]. Binding of TLR ligands to their respective receptors activate interferon regulatory factors (IRF3 and IRF7), which in turn promote expression and secretion of type I IFNs and inflammatory cytokines [4].

Previous studies indicate that NRs and their coregulators play important roles in DC-mediated immune response [1]. For example, PPARγ inhibits maturation of DCs, and suppresses their cytokine expression and potency to activate T-cells [8]. The liver X receptor (LXR) α and the vitamin D (VDR) receptor reduce capacity of DCs to present antigens to T-cells [9–10]. In addition, acetylation of chromatin-associated histones by histone acetylases modulate the balance between the T-helper (Th) 1 and Th2 lymphocytes [11]. Therefore, we characterize the expression profiles of NRs and coregulators in murine DCs upon viral infection, and examined the effect of selected NRs on the IFNβ promoter activity.

2. Material and methods

2.1 Mice

Male and female C57BL/6 mice at 6–8 week old were used in accordance with guidelines of the NIH animal care and use committee under the approved animal study proposal 09–008.

2.2 DC culture, virus infection and TLR stimulation

Bone marrow-derived DCs were generated by culturing bone marrow hematopoietic cells in the presence of the FMS-like tyrosine kinase 3 ligand for 8 days [12]. For purification of pDCs and cDCs, pDCs were labeled with the anti-B220 microbeads (Miltenyi Biotech, Bergisch Gladbach, Germany) at 4° C for 20 min, and these DC subtypes were separated following the manufacturer’s instruction (purity > 95%, by flow cytometry). Total DCs, pDCs or cDCs (1 × 10⁶ cells) were infected with NDV (Hertz strain, at multiplicity of infection (MOI=10) or MCMV (Smith strain, at MOI=10) at 37° C for indicated periods. TLR ligands, including the polyinosine-polycytidylic acid (TLR 3 ligand: Poly IC, 20 µg/ml; Invitrogen, Carlsbad, CA), resiquimod (TLR7 ligand: R848, 100 nM; Alexis, San Diego, CA) and the unmethylated CpG oligodeoxynucleotides (TLR9 ligand: CpG, 1 µg/ml; Operon, Huntsville, AL) were used to stimulate DCs.

2.3 Real-time quantitative PCR (qPCR) and PCR array

Total RNA (0.5 µg) was extracted using the RNeasy mini kit (Qiagen, Valencia, CA) and were converted to cDNA using the TaqMan reverse transcription reagents (Applied Biosystems, Foster City, CA). qPCR was performed by using 20 ng of cDNA, 5 µM of primers and the SYBR green PCR Master Mix (Applied Biosystems). Primer sequences were listed in Supplementary Table 1. RT² Profile Custom PCR Arrays (SA Bioscience, Frederick, MD) were used to examine simultaneously the mRNA levels of 89 NRs and coregulators and 5 housekeeping genes (Gusb, Hprt1, Hspcb, Gapdh and Actb) in 96-well plates. Each reaction included 380 ng of cDNA. Obtained Ct (threshold cycle) values were normalized by those of the 5 housekeeping genes (for PCR array) or Gapdh (for qPCR), and fold changes were calculated by using the comparative C_t method (2^−ΔΔCt where ΔΔC_t = ΔC_t, sample − ΔC_t, reference).

2.4 Immunoblot and histone deacetylase (HDAC) 1 activity assay

DCs (1 × 10⁷ cells) were infected/stimulated with NDV (MOI=10) or CpG (1 µg/ml) for indicated periods. Whole cell extracts or nuclear extracts were then prepared by using the nuclear protein extraction kit (Active Motif, Carlsbad, CA). Whole cell extracts (5 µg) were run on 8–12% NuPAGE Bis-Tris gels (Invitrogen), transferred to nitrocellulose membranes and immunoblotted with the rabbit anti-NOR1, anti-RXRα or anti-β-actin antibody (Santa Cruz Biotechnologies Inc., Santa Cruz, CA). The HDAC1 activities of nuclear extracts were measured by using the HDAC activity kit (Active Motif). HDAC activities obtained in the absence of infection/stimulation at each time-point were defined as 100%.

2.5 Luciferase reporter assay

HCT116 cells were cultured in McCoy’s 5A medium supplemented with 10% fetal calf serum and antibiotics. Using Lipofectamine 2000 (Invitrogen), cells were transfected with 0.2 µg/ml of the NOR1-expressing plasmid (a gift from Dr. Naganari Ohkura, National Cancer Center Research Institute, Tokyo, Japan) or the LXRα- and RXRα-expressing plasmids (gifts from Dr. David Mangelsdorf, University of Texas Southwestern Medical Center, Dallas, TX), 0.2 µg/ml of the IRF-expressing plasmid, 0.5 µg/ml of the pGL4 vector-based reporter plasmid carrying the IFNβ promoter and 0.5 µg/ml of the pGL4.73[hRluc/SV40] renilla control plasmid (Promega, Madison, MI) for 24 h. Cells were then infected with NDV (MOI=10) or treated with GW3965 (1 µM) for an additional 24 h. Lysates were analyzed for the firefly and renilla luciferase activities using the dual-luciferase assay kit (Promega).

2.6 Statistical analysis

Student’s t test with two-tailed p values was employed for evaluation of statistical significance.

3. Results

3.1 Basal expression of NRs and coregulators in DCs

We first examined basal mRNA expression of 49 NRs and 35 coregulators in DCs. DCs expressed most of the steroid and metabolic ligand receptors but only half of the orphan receptors (Supplementary Table 2A & 2B). Besides, DCs expressed 34 out of 35 coregulators examined (Supplementary Table 2C & 2D).

3.2 Modulation of NR and coregulator expression in virus-infected DCs

We next explored the effect of viral infection on mRNA expression of NRs and coregulators in DCs. After infecting DCs with NDV or MCMV, we monitored NR and coregulator mRNA expression at 3 time-points (Table 1). Both viruses induced pronounced DC activation but did not affect the percentages of cDCs and pDCs (Supplementary Figure 1 & 2). Although the 2 viruses induced unique patterns of NR and coregulator mRNA expression, they shared 6 NRs (Figure 1A & 1B) and 6 coregulators (Figure 1C & 1D) as commonly regulated molecules.

Table 1.

Viral infection modulates NR and coregulator mRNA expression in mouse DCs

A. NDV infection
Nuclear Receptors
Name	1h		7h		13h
	fold change	p-value	fold change	p-value	fold change	p-value
NOR1	−1.04	0.513	23.09	0.0002	3.24	0.118
AR	−1.34	0.458	4.64	0.030	3.06	0.181
LXRα	−1.31	0.039	4.18	0.016	7.77	0.034
GR	−1.13	0.239	3.23	0.017	2.19	0.162
TRα	−1.08	0.210	−2.41	0.0003	−5.63	0.038
NURR1	1.70	0.326	−3.16	0.014	−2.83	0.047
EAR2	−1.15	0.056	−3.18	0.001	−2.37	0.075
RXRα	−1.01	0.691	−4.60	0.001	−5.30	0.016
NUR77	1.69	0.232	−6.19	0.003	−15.30	0.009
PPARγ	−1.25	0.192	−6.28	0.002	−16.36	0.010
Coregulators
Name	1h		7h		13h
	fold change	p-value	fold change	p-value	fold change	p-value
i. General Coregulators
NCOA2	1.06	0.680	2.08	0.013	4.93	0.010
NCOR2	−1.11	0.169	−3.89	0.003	−2.87	0.034
ii. Histone Modifiers
HDAC1	−1.16	0.091	3.70	0.001	2.41	0.228
HDAC4	−1.37	0.072	−4.88	0.0003	−8.65	0.015
HDAC6	1.07	0.775	−3.57	0.0001	−2.93	0.023
HDAC10	1.03	0.609	−7.99	0.0001	−6.23	0.005
iii. Specific NR Coregulators
AKAP13	−1.03	0.355	3.02	0.002	1.31	0.778
CRSP2	1.05	0.795	−1.93	0.005	−3.04	0.044
iv. Cell Cycle/Growth Coregulators
Set-TAF-Iβ	−1.02	0.363	−3.30	0.015	−5.54	0.098
SMC4	−1.22	0.147	−1.91	0.001	−3.77	0.076

B. MCMV infection
Nuclear Receptors
Name	1h		7h		13h
	fold change	p-value	fold change	p-value	fold change	p-value
NOR1	1.26	0.314	20.72	0.0003	5.95	0.003
LXRα	−1.56	0.059	2.07	0.242	3.35	0.035
NURR1	1.61	0.972	−3.66	0.013	−2.46	0.031
VDR	1.07	0.150	−3.69	0.001	−1.15	0.403
RXRα	−1.28	0.055	−4.17	0.0005	−3.06	0.002
PPARγ	−1.16	0.188	−5.33	0.003	−5.37	0.001
NUR77	1.88	0.625	−6.37	0.002	−11.71	0.001
Coregulators
Name	1h		7h		13h
	fold change	p-value	fold change	p-value	fold change	p-value
i. General Coregulators
NCOA2	−1.06	0.059	2.04	0.011	5.79	0.011
NCOR2	−1.07	0.150	−3.79	0.002	−2.32	0.009
ii. Histone Modifiers
HDAC1	−1.45	0.034	3.33	0.003	3.07	0.022
HDAC4	−1.36	0.078	−4.51	0.003	−5.19	0.001
HDAC10	1.04	0.076	−3.67	0.004	−2.39	0.030
HDAC11	−2.07	0.013	−1.25	0.071	3.10	0.003
iii. Specific NR Coregulators
NRIP1	−1.42	0.082	2.26	0.057	3.30	0.030
iv. Cell Cycle/Growth Coregulators
Set-TAF-Iβ	−1.36	0.247	−3.80	0.002	−3.56	0.0002

Strongly induced genes (fold change>5)

Weakly induced genes (3>fold change>5)

Weakly suppressed genes (−5<fold change<−3)

Strongly suppressed genes (−5<fold change)

Figure 1.

Venn diagrams show common or specific NRs and coregulators up-regulated or down-regulated by viral infection in DCs.

Regarding the time course of mRNA expression after NDV or MCMV infection, some NRs/coregulators (e.g. NOR1 and HDAC10) demonstrated the highest expression at 7 h time-point after infection, while others (e.g. NUR77 and HDAC4) developed it at 13 h Table 1). Interestingly, 2 groups of NRs showed significant and characteristic changes after viral infection: Three NR4A family members (NOR1, NURR1 and NUR77) and 3 metabolite receptors (LXRα, PPARγ and RXRα). While NOR1 was significantly up-regulated (over 20-fold), NUR77 and NURR1 were down-regulated (by almost 80%). LXRα and PPARγ, which are master regulators of the cholesterol and carbohydrate metabolism respectively, were significantly modulated after viral infection (LXRα: ~3–6 fold up-regulation; PPARγ and RXRα ~60–80% down-regulation). RXRα is essential for LXRα and PPARγ to form functional heterodimers [13].

To investigate whether the above-mentioned transcriptional regulation occurs differently in DC subtypes, we purified cDCs and pDCs from total DCs, infected them with viruses separately, and monitored mRNA expression of selected NRs and coregulators (12 overlapping genes shown in Figure 1) by qPCR. We found that cDCs and pDCs exhibited similar induction/reduction profiles on most of the genes examined, whereas some genes showed stronger response in particular DC subtypes (Figure 2 & Supplementary Figure 3). For example, cDCs showed stronger induction of NOR1 and HDAC1 mRNA than pDCs upon viral infection. pDCs, on the other hand, exhibited stronger repression on NURR1 and HDAC4 transcription than cDCs (Figure 2). Both DC subtypes highly expressed NOR1 and LXRα at baseline and after viral infection (Supplementary Table 3).

Figure 2. Viral infection alters NR and coregulator mRNA expression in pDCs and cDCs.

pDCs and cDCs were purified from DCs, and were infected with NDV (MOI=10) or MCMV (MOI=10). Total RNA was isolated at 1 h, 7 h or 13 h of post-infection. The mRNA levels of indicated NRs and coregulators were measured with qPCR. Data were normalized with Gapdh mRNA abundance and the mean values +/− SEM of 3 independent experiments are shown.

3.3 TLR ligands modulated NR and coregulator mRNA expression in DCs

The above results indicate that NDV and MCMV infection modulated mRNA expression of a similar set of NRs and coregulators in DCs. DCs express TLRs to detect viral genetic materials including viral DNA and RNA [14]. We therefore examined the effect of specific synthetic TLR ligands to verify if TLRs participate in the transcriptional regulation of NRs and coregulators upon NDV (single strand RNA virus) or MCMV (double strand DNA virus) infection. All of the tested TLR ligands strongly activated DCs (Supplementary Figure 4), and induced similar effects as viral infection on the mRNA expression of selected NRs and coregulators, although with differential potencies (Figure 3 and Table 1).

Figure 3. TLR ligands alter NR and coregulator mRNA expression in DCs.

DCs were treated with CpG (1 µg/ml), Poly IC (20 µg/ml) or R848 (100nM) for 7 h. The mRNA levels of NRs and coregulators (A) up-regulated (B) down-regulated by treatment are shown. Data were normalized with Gapdh mRNA abundance and compared with control, and the mean values +/− SEM of 3 independent experiments are shown.

3.4 NDV and CpG changed NR protein expression in DCs

To verify whether the changes found at mRNA levels are also observed at protein levels, we examined protein expression of the selected NRs in DCs after NDV infection or CpG stimulation. Both NDV and CpG up-regulated NOR1 and LXRα expression at 16 h time-point after the infection/stimulation and sustained to 24 h, whereas RXRα was highly down-regulated continuously (Figure 4A).

Figure 4. Viral infection modulates NRs protein expression and HDAC activity in DCs.

(A) Nuclear extracts were prepared from DCs infected with NDV (MOI=10) or stimulated with CpG (1 µg/ml) for 16 h or 24 h. Equal amounts of protein (5 µg) were loaded in each lane, and Western blots evaluating expression of NOR1, LXRα, RXRα and control β-actin were performed. Western blots for these molecules in HCT116 cells transfected with the respective NR-expressing plasmid are served as positive controls (right gel images). Representative images of 3 independent experiments are shown.

(B) Nuclear extracts were prepared from DCs infected with NDV (MOI=10) or stimulated with CpG (1 µg/ml) for 13 h and 19 h. HDAC activity of each time-point is shown as % of the control obtained in the absence of viral infection at the same time-point. The mean values +/− SEM of 3 independent experiments are shown.

3.5 NDV and/or CpG changed mRNA expression of the NR-responsive genes in DCs

To examine biological impact of the altered NR expression caused by viral infection, we infected DCs with NDV and stimulated them with the LXRα agonist GW9635 or the PPARγ agonist ciglitazone. NDV enhanced GW3965-dependent mRNA expression of LXRα-responsive genes (ATP-binding cassette sub-family G member 1 (ABCG1) and sterol regulatory element-binding protein-1c (SREBP-1c) [15–16]), while it potentiated ciglitazone-induced suppression of the PPARγ-responsive gene mRNA expression (adipocyte fatty acid-binding protein (aP2) and ABCG2 [17–18]) (Supplementary Table 4). NDV infection and CpG treatment strongly stimulated mRNA expression of the NOR1-responsive genes (small-inducible cytokine B10 (IP10) and myristoylated alanine-rich C-kinase substrate (MARCKS) [19]) (Supplementary Table 4). These results indicate that NDV infection changes the actions of these NRs on their responsive genes through alteration of their expression levels in DCs.

3.6 NDV and CpG increased the HDAC1 activity in DCs

Among the HDACs found to be modulated upon viral infection (Figure 1C & 1D), HDAC1 belongs to class I HDACs, which constitutively localize in the nucleus and mainly deacetylate chromatin-associated histones [20]. In contrast, HDAC4 and HDAC10 are class II HDACs, which localize in the cytoplasm and deacetylate various cytoplasmic proteins [20]. Nuclear extracts obtained from DCs infected/stimulated with NDV or CpG showed increased HDAC activity (up-regulated by 50–75%) after 13 h of treatment (Figure 4B), which was consistent with the up-regulation of HDAC1 mRNA expression after NDV infection (Table 1). The HDAC activity returned to baseline after 19 h of the treatments, indicating that the effect of infection/treatment was transient (Figure 4B).

3.7 NOR1 and LXRα repressed IRF-induced transcriptional activity of the IFNβ promoter

Pathogen-stimulated TLRs activate several transcription factors including IRF3 (by TLR3) and IRF7 (by TLR7 and TLR9) for induction of the type I IFNs and inflammatory cytokines in DCs [21]. To address immunological consequences of viral infection-induced alteration of NR expression in DCs, we examined the influence of NOR1 or LXRα to the transcriptional activity of IRF3 or IRF7 on the IFNβ promoter in the absence and presence of NDV infection. Without infection, overexpression of either NOR1 or LXRα/RXRα strongly repressed IRF3- or IRF7-induced transcriptional activity of the IFNβ promoter (Figure 5A: NOR1: 80% and 50% respectively; Figure 5B: LXRα/RXRα: 30% and 50%, respectively). Addition of the LXRα ligand GW3965 further enhanced the repressive effect (Figure 5C). These receptors, however, did not affect IRF8-induced transcriptional activity on the same promoter (Figure 5A & B). Thus, NOR1 and LXRα specifically suppress IRF3- and IRF7-induced induction of IFNs. Further, NDV infection significantly reduced IRF3- or IRF7-induced transcriptional activity of the IFNβ promoter in the absence of NR overexpression, and potentiated the repressive effect of exogenously expressed NOR1 and LXRα/RXRα on the transcriptional activity of IRF3 and IRF7 (Figure 5A & 5B, right half panels). These results suggest that NDV stimulated expression of endogenous NOR1 and LXRα/RXRα, which suppressed the transcriptional activity of IRF3 and IRF7 in cooperation with exogenously expressed NRs.

Figure 5. NOR1 or LXRα represses IRF3- and IRF7-induced transcriptional activity on the IFNβ promoter upon NDV infection in HCT116 cells.

(A & B) HCT116 cells were transfected with the indicated protein-expressing plasmids together with the pGL4 reporter carrying the IFNβ promoter and the pGL4.73[hRluc/SV40] control plasmid. Cells were subsequently infected with NDV (MOI=10) for 24 h. (C) The transfected cells were treated with GW3965 (1 µM) for 24 h. Firefly luciferase activities were normalized by renilla luciferase activities to account for transfection efficiency. The mean values +/− SEM of 3 independent experiments are shown. *: p<0.05, **: p<0.01, n.s.: not significant, compared to the conditions indicated. RLU: relative light unit.

4. Discussion

Among all the NRs expressed in mice, we found that NR4A family members showed distinct expression profiles in response to viral infection. We also found that mRNA expression of LXRα was up-regulated by ~3–6 fold, while PPARγ was down-regulated by ~60–80%. This viral infection-mediated alteration of NR expression further modulated mRNA expression of the genes responsive to these NRs. In reporter assays employing the IFNβ promoter construct, NOR1 and LXRα/RXRα strongly repressed IRF3- and IRF7-induced transcriptional activities. The effect may be mediated by competition at coactivator levels, as exogenous expression of p300/CBP alleviated the negative effect of LXRα/RXRα on IRF3-induced transcriptional activity in RAW264.7 cells [22]. These NRs may act as counter regulatory factors for TLR activation, preventing the latter’s prolonged stimulation of the type I IFN production, which may exert detrimental effects on local tissues. In contrast, these NRs did not influence the transcriptional activity of IRF8 on the IFNβ promoter. IRF8 promotes the second phase of the type I IFN production in DCs upon pathogen infection but not responsible for its first phase mediated by IRF3 and IRF7 [12]. Thus, these results indicate the functional specificity of these NRs among the IRF subtypes, selectively affecting the first phase production of the type I IFNs.

All three members of the NR4A family are expressed in macrophages [23–24], and are strongly induced by LPS and IFNγ treatment possibly through activation of the NF-κB pathway [23–24]. Since both DCs and macrophages are professional antigen-presenting cells, it is possible that similar regulation on the expression of the NR4A family members may occur in DCs. NOR1 was reported to regulate apoptosis in DCs and T-cells, and simultaneous inactivation of NOR1 and NUR77 develops acute leukemia in mice [25–27]. Thus, NOR1 and other NR4A family members might play essential roles in the regulation of various cellular activities in DCs, such as differentiation and cytokine production, in addition to the regulation of the IRF activity on the type I IFN production.

In regard to LXRα and PPARγ, these NRs have strong and diverse modulatory effects on immunity and inflammation in vitro and in vivo [28–29]. For example, stimulation of LXRs in macrophages alleviates inflammation and relieves plaque formation in atherosclerotic vasculatures [28], while LXR-null macrophages are defective in response to intracellular pathogens [30]. Further, PPARγ redirects DCs toward a less stimulatory condition, and modulates migration of Langerhan cells from infection sites to drain lymph nodes for T-cell activation [31–32]. Moreover, leukotriene, a prostanoid inflammatory mediator, acts as a ligand for PPARγ and modulates expression of IL-10 and IL-12 through activation of PPARγ in DCs [33]. These previous studies indicate the importance of LXRα and PPARγ in many aspects of the DC biology, and further suggest the possibility that viral infection influences DC activity by modulating expression of these NRs.

Most of the coregulators examined were expressed at baseline and their mRNA levels after viral infection were relatively stable compared to those of NRs, suggesting that viral infection mainly alters DC activity by regulating the expression of DNA-binding factors including NRs. Nevertheless, DCs significantly altered mRNA levels of Set-TAF-Iβ, NCOR2, NCOA2, and HDACs upon viral infection, which are either chromatin-modifying enzymes or essential cofactors for assembling basal transcriptional machineries [2]. HDACs modulate DC-mediated immune activity in part by inducing differentiation of precursor DCs into appropriate subtypes and by stimulating expression of costimulatory molecules for antigen presentation [11]. Among the HDACs significantly regulated by viral infection, HDAC1 is essential for induction of the IFN-responsive genes by H4 deacetylation [34]. Given that viral infection strongly stimulated HDAC1 mRNA expression/activity in DCs, it is likely that virus alters IFN-mediated anti-viral response in part through regulation of the HDAC1 expression/activity.