Interferon-inducible factor 16 is a novel modulator of glucocorticoid action

Abstract

Previously, we used cDNA expression profiling to identify genes associated with glucocorticoid (Gc) sensitivity. We now identify which of these directly influence Gc action. Interferon-inducible protein 16 (IFI16), bone morphogenetic protein receptor type II (BMPRII), and regulator of G-protein signaling 14 (RGS14) increased Gc transactivation, whereas sialyltransferase 4B (SIAT4B) had a negative effect. Amyloid beta (A4) precursor-protein binding, family B, member 1 (APBB1/Fe65) and neural cell expressed developmentally down-regulated 9 (NEDD9) were without effect. Only IFI16 potentiated Gc repression of NF-kappaB. In addition, IFI16 affected basal expression, and Gc induction of endogenous target genes. IFI16 did not affect glucocorticoid receptor (GR) expression, ligand-dependent repression of GR expression, or the ligand-dependent induction of GR phosphorylation on Ser-211 or Ser-203. Coimmunoprecipitation revealed an interaction, suggesting that IFI16 modulation of GR function is mediated by protein crosstalk. Transfection analysis with GR mutants showed that the ligand-binding domain of GR binds IFI16 and is the target domain for IFI16 regulation. Analysis of human lung sections identified colocalization of GR and IFI16, suggesting a physiologically relevant interaction. We demonstrate that IFI16 is a novel modulator of GR function and show the importance of analyzing variation in Gc sensitivity in humans, using appropriate technology, to drive discovery.

Figure 1.

Bioinformatic analysis of 20 candidate Gc sensitivity modulating genes. Network analysis describing interactions between genes/gene products from the steroid-sensitive gene list with each other and with other well-characterized genes defined by canonical Gc metabolic and GR signaling pathways. Blue represents genes/gene products from the steroid-sensitive gene list and red denotes genes represented on canonical Gc pathways. Connecting lines represent interactions.

Figure 2.

Expression of BMPRII, IFI16, and RGS14 in HeLa cells enhances Gc-induced expression of the TAT3-Luc reporter gene. A) HeLa cells were transiently cotransfected with 2 μg of TAT3-Luc reporter, 0.5 μg of Renilla reporter, and either 0.6 or 1.2 μg of SRC2, IFI16, BMPRII, APBB1, RGS14, SIAT4B, NEDD9, or empty pcDNA3 construct. Cells were incubated for 24 h with Dex (0.1–10 nM) and assayed for luciferase. B) HeLa cells were transiently cotransfected with 2 μg of either C/EBP-Luc or TAT3ΔGRE-Luc and 0.5 μg of Renilla reporter control reporter gene constructs along with 1.2 μg of SRC2, IFI16, BMPRII, or empty pcDNA3 expression vector. Cells were incubated for 24 h with Dex (0.1–10 nM) and assayed for luciferase. C) HeLa cells stably expressing TAT3-Luc were cotransfected with 0.5 μg of Renilla construct and SRC2, IFI16, BMPRII, or empty pcDNA3/CMV SPORT6 construct. Twenty-four hours after transfection, cells were incubated with Dex (0.1–10 nM) for 16 h before harvest and luciferase analysis. Graphs depict means ± sd of triplicate wells, representative of 3 independent experiments. *P = <0.05, **P = <0.01 vs. empty vector control. RLU, relative light units.

Figure 3.

Overexpression of IFI16 enhances Gc-induced repression of the NRE-Luc reporter gene. HeLa cells were transiently cotransfected with 2 μg of NRE-Luc, 0.5 μg of Renilla reporter, and 0.6 μg of SRC2, IFI16, BMPRII, APBB1, RGS14, SIAT4B, NEDD9, or empty pcDNA3 construct. Twenty-four hours after transfection, cells were incubated with 10 nM Dex and 0.5 or 5 ng/ml TNF for 16 h before luciferase analysis. Graphs depict means ± sd of triplicate wells, representative of 3 independent experiments. *P < 0.05, **P < 0.01 vs. empty vector control.

Figure 4.

IFI16 modulates GR regulation of endogenous genes. A) HeLa cells transfected with 10 nM IFI16 or lamin siRNA for 48 h were lysed and immunoblotted for IFI16, GR, or tubulin. Representative images are shown. B, C) HeLa cells were transfected with 10 nM IFI16 or lamin siRNA, treated with vehicle or 100 nM Dex for 4 h, and then lysed and RNA-processed. Effect on basal (B) and Gc-regulated (C) expression of 7 GC-regulated genes (ASPH, FKBP5, MT1X, IL6ST, MYO1B, GLUL, and GILZ) was analyzed by qRT-PCR. Graphs depict means ± sd of triplicate wells, representative of 3 independent experiments. *P < 0.05, **P < 0.01 vs. lamin siRNA control.

Figure 5.

Expression and activation of GR is not affected by overexpression of BMPRII, IFI16, RGS14, or SIAT4B. A, B) HeLa cells were transiently transfected to overexpress IFI16, BMPRII, RGS14, SIAT4B, or empty pcDNA3 construct. Cells were then incubated with 100 nM Dex for either 24 h to examine GR expression (A) or 30 min to examine GR phosphorylation by immunoblot analysis (B). C) Tubulin was also included as a loading control. Representative images are shown. Phosphorylated GR was analyzed using ImageJ, and densitometry from 3 independent experiments was expressed as a ratio of phospho (P)-GR/tubulin. D) HeLa cells transfected to overexpress IFI16 or empty pcDNA3 construct were incubated with 100 nM Dex for up to 24 h and immunoblotted for GR, P-Ser-203-, P-Ser-211-, and P-Ser-226-GR. Immunoblots for IFI16 and tubulin are included as controls (CTRL). Representative images are shown.

Figure 6.

IFI16 and GR subcellular localization. A, B) HeLa cells expressing endogenous IFI16 (A) or transfected with 1.2 μg of IFI16 expression vector (B) were fixed and double-labeled with antibodies specific to GR (green) and IFI16 (red), and nuclei were counterstained with DAPI (blue). C) Arrowheads identify cells in which IFI16 and GR are localized to the nucleus. HeLa cells expressing endogenous IFI16 were incubated with vehicle or 100 nM Dex for 2 h. Cells were fixed and then were labeled with antibodies specific to GR (green) and IFI16 (red), and nuclei were counterstained with DAPI (blue). Arrowhead indicates Dex-treated cells in which GR and IFI16 localize to the nucleus. D) HeLa cells transfected with IFI16 expression vector (IFI16) were treated with vehicle or Dex for 1 h and then lysed, and cell extracts were immunoprecipitated (IP) with two different antibodies (rabbit and mouse) for either IFI16 or GR. Precipitates were immunoblotted (IB) for IFI16 and GR. Samples immunoprecipitated with mouse or rabbit IgG were included as species-specific immunoglobulin controls. E) Schematic diagram depicting the exons encoding the three IFI16 isoforms.

Figure 7.

IFI16 interacts with the GR in an LBD-dependent manner. A, B) HEK293 cells were transiently transfected with 2 μg of TAT3-Luc, 0.5 μg of Renilla, 1.2 μg of IFI16 expression vector, and 1 μg of either full-length GRα or GR ΔAF-1, which lacks the AF-1 domain (A), or GR N500, which lacks the LBD (B). After transfection cells were incubated with increasing concentrations of Dex (0.001–1 nM) for 16 h before analysis. Graphs depict means ± sd of triplicate wells, representative of 3 independent experiments. *P < 0.05, **P < 0.01. RLU, relative light units. C) HEK293 cells were transiently transfected with 2 μg of full-length GRα, GR ΔAF-1, or GR N500 myc fusion proteins and then lysed, and cell extracts were immunoprecipitated (IP) with an antibody raised against myc. Precipitates were immunoblotted (IB) for myc, IFI16, and GR. Samples immunoprecipitated with mouse IgG were included as controls. wt, wild-type. D) Schematic representation of the putative domain structure of IFI16 protein. The N-terminal domain of IFI16 (solid black) contains several regions important for IFI16 activity, including the DNA-binding domain (DBD), nuclear localization signal (NLS), pyrin/dapin/PAAD domain (pyrin domain), and site of phosphorylation by casein kinase 2 (CK2). Checkered regions indicate duplicate HIN 200 domains A and B. These are separated by a serine-threonine-proline (S/T/P)-rich spacer region (hinge region). IFI16 isoforms arise due to alternative RNA splicing in exons encoding the S/T/P domain. Numbers identify amino acid residues. LxxLL-like motifs, serine phosphorylation, and nuclear localization signal are also shown.

Figure 8.

Expression of IFI16 in human lung. Representative photomicrographs of IFI16 expression in areas of T- and B-lymphocyte aggregates within human lung tissue. Detection of IFI16 (green) within immune cells of a parenchymal tertiary lymphoid follicle from a patient with COPD. IFI16 was expressed in CD3⁺ T cells (A) and CD20⁺ B cells (B). Left panels are low-magnification views; right panels are expanded views.

Figure 9.

Colocalization of IFI16 and GR in human lung. Representative photomicrographs of IFI16 and GRα. Lung sections from a patient with COPD (A) and a nonsmoking individual with normal lung function (B) were subject to immunofluorescence with GR and IFI16 antibodies. All sections were counterstained with DAPI (blue). Arrows indicate inflammatory cells coexpressing GR and IFI16. Asterisk indicates single expression of GR or IFI16. Top rows are low-magnification views; bottom rows are expanded views.