Volume 592, Issue 2 p. 244-255
Research Letter
Free Access

Selective regulation of type II interferon‐inducible genes by NPM1/nucleophosmin

Mayumi Abe

Faculty of Medicine, University of Tsukuba, Japan

PhD Program of Human Biology, School of Integrative and Global Majors, University of Tsukuba, Japan

Search for more papers by this author
Jianhuang Lin

Faculty of Medicine, University of Tsukuba, Japan

PhD Program of Human Biology, School of Integrative and Global Majors, University of Tsukuba, Japan

Search for more papers by this author
Kyosuke Nagata

Faculty of Medicine, University of Tsukuba, Japan

Search for more papers by this author
Mitsuru Okuwaki

Corresponding Author

Faculty of Medicine, University of Tsukuba, Japan

PhD Program of Human Biology, School of Integrative and Global Majors, University of Tsukuba, Japan

Correspondence

M. Okuwaki, Faculty of Medicine, University of Tsukuba, 1‐1‐1 Tennodai, Tsukuba, Japan

Fax: +81 29 853 3942

Tel: +81 29 853 7950

E‐mail: mokuwaki@md.tsukuba.ac.jp

Search for more papers by this author
First published: 18 December 2017
Citations: 4
Edited by Wilfried Ellmeier

Abstract

Nucleophosmin (NPM1) is a multifunctional nucleolar protein. Here, we analyze the role of NPM1 in gene expression using our previous microarray data and find a relationship between NPM1 and interferon (IFN)‐γ‐inducible genes. We show that NPM1 selectively regulates the expression of a subset of IFN‐γ‐inducible genes and directly binds to two important transcription factors in the type II IFN pathway: signal transducer and activator of transcription 1 and interferon regulatory factor 1 (IRF1). Furthermore, NPM1 is found to regulate the IFN‐γ‐inducible promoter activity of major histocompatibility complex class II transactivator (CIITA), and mutation of the IRF1‐binding site on the CIITA promoter abolishes the effect of NPM1. Our results suggest a novel mechanism for IFN‐γ‐mediated gene expression by NPM1.

Abbreviations

CIITA, class II transactivator

GAS, gamma‐associated site

IFN, interferon

IFNAR, IFN‐α receptor

IFNGR, IFN‐γ receptors

IRF1, interferon regulatory factor 1

IRF9, IFN‐regulatory factor 9

IRF‐E, IRF‐binding element

ISGF3, IFN‐stimulated gene factor 3

ISRE, IFN‐stimulated response elements

MHC, major histocompatibility complex

STAT, signal transducer and activator of transcription

Nucleophosmin (NPM1) is a phosphoprotein that is mainly localized in the nucleolus, although it constantly shuttles between the nucleolus, the nucleus, and the cytoplasm [1]. Importantly, NPM1 is highly expressed in human solid malignancies and has been implicated in tumorigenesis; genetic mutations of its gene are frequently found in acute myeloid leukemia [2, 3]. Therefore, it is important to understand the functions of NPM1 in both normal and malignant cells. NPM1 is a multifunctional protein, which is involved in the regulation of ribosome biogenesis, DNA replication, apoptosis, centrosome duplication, and cell proliferation [3, 4]. We previously reported that NPM1 shows histone chaperone activity in vitro and participates in the regulation of chromatin structure [5]. It has also been shown that NPM1 interacts with transcription factors, including c‐Myc, NFκB, YY1, AP‐2γ, and interferon regulatory factor 1 (IRF1) and is required for the regulation of their target genes [6-10]. Consistent with these observations, our recent microarray analysis demonstrated that NPM1 is involved in the regulation of various genes [7]; however, the molecular mechanism by which NPM1 regulates that the expression of those genes is not well understood.

Interferons (IFNs) are cytokines that play important roles in antiviral and antiproliferative responses [11]. IFNs are classified into type I, II, and III based on receptor specificity and sequence homology. The main signaling pathway activated by IFNs is the Janus‐activated kinase (JAK) signal transducer and activator of transcription (STAT) pathway [11, 12]. The binding of type I IFNs to the IFN‐α receptor (IFNAR) results in the autophosphorylation and activation of the receptor‐associated JAK1 and tyrosine kinase 2 (TYK2) pathways, which in turn regulate the tyrosine phosphorylation of STAT1 and STAT2. Tyrosine‐phosphorylated STAT1 and STAT2 heterodimers translocate to the nucleus, where they assemble with IFN‐regulatory factor 9 (IRF9) to form a complex called IFN‐stimulated gene factor 3 (ISGF3). This complex binds to specific elements, termed IFN‐stimulated response elements (ISREs), which are present in the promoters of IFN‐stimulated genes to initiate transcription. The type III IFNs bind to a receptor complex composed of interferon lambda receptor 1 (IFNLR1) and interleukin‐10 receptor B (IL10RB) and use the JAK‐STAT signal transduction pathway similar to type I IFNs [13]. In contrast, the only type II IFN, IFN‐γ, binds to the IFN‐γ receptor (IFNGR), followed by JAK1‐ and JAK2‐mediated phosphorylation of STAT1. Phosphorylated STAT1 homodimers translocate to the nucleus and bind to the DNA sequence termed the IFN‐γ activation site (GAS) to initiate transcription.

Interferon regulatory factor 1 is induced by both IFN‐α/β and IFN‐γ and binds to the ISRE/IRF‐binding element (IRF‐E) on the target genes’ promoters. The major histocompatibility complex (MHC) I and II genes, which are required for antigen presentation, are induced by IRF1 on stimulation with IFN‐γ [14]. The NOD‐like receptor family CARD domain containing 5 (NLRC5), and MHC class II transactivator (CIITA) genes are also required for the expression of MHC I and MHC II genes, respectively [15-17].

From the previous microarray data, we found that the expression of IFN‐γ‐inducible genes is decreased by NPM1 knockdown. Interestingly, we demonstrated that NPM1 binds directly to both STAT1 and IRF1 and participated in the transcriptional regulation of a subset of IFN‐γ‐inducible genes. We propose a novel mechanism for the type II IFN signaling pathway by NPM1.

Materials and methods

Plasmid construction

Plasmids pGEX2T‐NPM1, pET14b‐NPM1, pET14b‐B23.2, pET14b‐B23.3, pET14b‐NPM1ΔA, pET14b‐NPM1ΔC, pET14‐NPM1ΔN, pET14b‐NPM1CR, and pET14b‐NPM1CR1.5 were described previously [18]. The STAT1 and IRF1 were amplified by PCR using primer sets 5′‐aaaggatccatgtctcagtggtacgaact‐3′ and 5′‐aaaggatccctatactgtgttcatcatac‐3′ and 5′‐agctggatccatgcccatcactcggatgcg‐3′ and 5′‐agcgaattctacggtgcacagggaatggcc‐3′ with cDNA prepared from HeLa cells as a template. The amplified cDNAs were subcloned into BamH I and EcoR I sites of pcDNA3.1‐Flag vector. To construct pCAGGS‐Flag‐IRF1, the IRF1 cDNA was cut out from pcDNA3.1‐Flag‐IRF1 by Hind III and EcoR I, blunted by Klenow Fragment (Toyobo, Osaka, Japan), and subcloned into pCAGGS treated with Xho I and Klenow fragment. To construct pGEX6P‐1‐IRF1, the IRF1 cDNA was cloned into BamH I and EcoR I sites of pGEX6P‐1 vector. The promoter IV sequence of the human CIITA gene (CIITA‐237) was amplified by PCR using a primer set 5′‐AAAAGATCTGGGGCCTGGGACTCTCCCCG‐3′ and 5′‐AAAAAGCTTCCCGACCTTAGGGGTTACAG‐3′ with genomic DNA extracted from HeLa cells as a template. To construct a series of 5′ deletion mutants, forward primers 5′‐AAAAGATCTTTGGGATGCCACTTCTGATA‐3′ for CIITA‐154, 5′‐AAAAGATCTCAGCGCTGCAGAAAGAAAGT‐3′ for CIITA‐82, or 5′‐AAAAGATCTGAAAAAGAACTGCGGGGAGG‐3′ for CIITA‐54, and the reverse primer described above were used. The amplified DNA was subcloned into Bgl II and Hind III sites of pGV‐B vector. Site‐directed mutations at the IRF1 recognition sequence and the GAS in pGV‐B‐CIITA‐237 were introduced by primer sets 5′‐CTTTTTCTCGAGCACTGTCTTTCTGCAGCGCTGAGCTCG‐3′ and 5′‐GCAGAAAGACAGTGCTCGAGAAAAAGAACTGCGGGGAGG‐3′, and 5′‐CACGTGCTTTAGAATTCGTGGCATCCCAACTGCCTGG‐3′ and 5′‐ATGCCACGAATTCTAAAGCACGTGGTGGCCACAGTAG‐3′, respectively.

Cell culture, transfection, and reagents

HeLa and 293T cells were maintained in Dulbecco's modified Eagle's medium (Nacalai Tesque, Kyoto, Japan) supplemented with 10% heat‐inactivated fetal bovine serum at 37 °C with 5% CO2. The stable HeLa cell line expressing EGFP‐Flag‐NPM1 was established previously [19] and maintained as described above. Transient transfections of plasmid DNA and siRNAs were performed using GeneJuice (Novagen, Darmstadt, Germany) and Lipofectamine RNA iMAX (Life Technologies, Waltham, MA, USA), respectively, according to the manufacturer's instructions. Stealth RNAs for negative controls and human NPM1 were described previously [19]. Antibodies used were NPM1 (Life Technologies), Flag‐tag (M2; Sigma Aldrich, St. Louis, MO, USA), STAT1 (sc‐346; Santa Cruz, Dallas, TX, USA), p‐STAT1 (Y701) (D4A7, CST), IRF1 (ab26109; Abcam, Cambridge, UK), and β‐actin (sc‐47778; Santa Cruz). Recombinant human IFN‐β and IFN‐γ (PEPROTECH, Rocky Hill, CT, USA) were commercially available.

Purification of recombinant proteins

For expression and purification of GST‐tagged proteins, BL21 (DE3) and BL21 (RIL) were transformed with pGEX2T‐NPM1 and pGEX6P‐1‐IRF1, respectively. The transformed E.coli was grown at 37 °C until OD600 reached 0.4. Expression of the recombinant proteins was induced by the addition of isopropyl β‐d‐thiogalactopyranoside at 16 °C for 16 h. Bacterial cell lysates expressing GST‐tagged proteins were sonicated in buffer A [50 mm Tris‐HCl (pH 7.9), 0.1% Triton X‐100, and 1 mm PMSF] containing 100 mm NaCl. For purification of Flag‐tagged STAT1, 293T cells transfected with pcDNA3.1‐Flag‐STAT1 were suspended in buffer B (0.2% Triton X‐100, 20 mm Tris‐HCl pH 7.9, 10 mm KCl, 1.5 mm MgCl2, 0.5 mm PMSF) containing 400 mm NaCl on ice for 10 min and rotate at 4 °C for 30 min followed by centrifuge at 21 500 g, 4 °C for 15 min. The supernatants were recovered and diluted with twice volumes of buffer B without NaCl. The cell extracts were incubated with anti‐Flag M2 affinity gels (Sigma Aldrich) for 2 h at 4 °C and then washed by buffer A containing 300 mm NaCl. The proteins bound with the resin were eluted with buffer A containing 150 mm NaCl and Flag peptide (Sigma Aldrich), and the eluted proteins were dialyzed against buffer H (20 mm Hepes‐NaOH pH7.9, 50 mm NaCl, 0.1 mm EDTA, 1 mm DTT, 0.5 mm PMSF, and 10% glycerol). Purification of His‐tagged proteins was described previously [18].

Immunoprecipitation and GST pull‐down assays

Flag‐IRF1 was transiently expressed in 293T cells. The cells were treated with or without IFN‐γ for 6 h, collected, and sonicated in buffer A containing 100 mm NaCl. The cell lysates were incubated with anti‐Flag M2 affinity gels (Sigma Aldrich) in buffer A containing 100 mm NaCl. The resins were washed extensively with the same buffer. The proteins bound with the resin were eluted with buffer A containing 100 mm NaCl and Flag peptide (Sigma Aldrich), separated by SDS/PAGE and analyzed by western blotting. For immunoprecipitation of STAT1, HeLa cells were treated without or with IFN‐γ for 1 h and the cell lysates were prepared. The extracts were subjected to immunoprecipitation with control IgG or anti‐STAT1 antibody, and immunoprecipitated proteins were separated by SDS/PAGE and detected by western blotting. For GST pull‐down assays, GST, GST‐NPM1, or GST‐IRF1 was immobilized on glutathione sepharose beads. The beads were mixed with Flag‐STAT1, His‐NPM1, or His‐NPM1 deletion mutants, and washed with buffer A containing 100 mm NaCl. Proteins were eluted from the beads by an SDS sample buffer, separated by SDS/PAGE, and visualized by CBB staining or western blotting.

Reporter assay

HeLa cells (4 × 104 per well) transfected with control or NPM1 siRNA were seeded in 24‐well plates and transfected with 125 ng of pGV‐B‐CIITA, pGAS‐TA‐Luc (Clontech, Mountain View, CA, USA), or pISRE‐TA‐Luc (Clontech; Firefly luciferase) and 125 ng of pTA‐RL (Renilla luciferase) 24 h after siRNA transfection. Twenty‐four hours after plasmid DNA transfection, cells were treated with IFN‐γ (20 ng·mL−1) for 24 h. For pISRE‐TA‐Luc reporter, cells were treated with INF‐β (1000 IU·mL−1) for 3 h. Luciferase assay was performed using Renilla Luciferase Assay System kit (Promega Corporation, Madison, WI, USA) according to the manufacturer's instructions.

RT‐qPCR

HeLa cells were transfected with siRNA for NPM1 or negative control for 48 h and IFN‐γ (20 ng·mL−1) was added and further incubated for 24 h. Total RNA was extracted using RNeasy Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. cDNA was prepared from purified RNA (1 μg) using ReverTraAce (Toyobo) with oligo dT primer. Real‐time PCR was performed in triplicate with SYBR Green Real‐time PCR Master Mix‐Plus (Toyobo) in the Thermal Cycler Dice Real‐time PCR system (TaKaRa, Shiga, Japan). Primer sets for RT‐PCR are listed in Table 1.

Table 1. Primers used for RT‐PCR
Primers Sequences
HLAB‐F GCGGCTACTACAACCAGAGC
HLAB‐R GATGTAATCCTTGCCGTCGT
HLADR‐F GAGTTTGATGCTCCAAGCCCTCTCCCA
HLADR‐R CAGAGGCCCCCTGCGTTCTGCTGCATT
HLADQ‐F GGGCTGACTGAAACTATGGC
HLADQ‐R AGGGTGGGAACACAAGGAAG
STAT1‐F CCATCCTTTGGTACAACATGC
STAT1‐R TGCACATGGTGGAGTCAGG
IRF1‐F GAACTCCCTGCCAGATATCGAG
IRF1‐R TGCTCTTAGCATCTCGGCTGGA
CIITA‐F CTGAAGGATGTGGAAGACCTGGGAAAGC
CIITA‐R GTCCCCGATCTTGTTCTCACTC
NLRC5‐F CTGGCCAGTCTCACCGCACAA
NLRC5‐R CCAGGGGACAGCCATCAAAATC
JAK1‐F AAATCGCACCATCACCGTTG
JAK1‐R ATTGTCGTTGGTTCCATGCC
JAK2‐F AGTGGCGGCATGATTTTGTG
JAK2‐R TCTAACACTGCCATCCCAAGAC
IFNGR1‐F TTTCTCCTACCCCTTGTCATGC
IFNGR1‐R TTAGTTGGTGTAGGCACTGAGG
IFNGR2‐F AAGATTCGCCTGTACAACGC
IFNGR2‐R GCCGTGAACCATTTACTGTCG
GAPDH‐F CCACATCGCTCAGACACCAT
GAPDH‐R GCGCCCAATACGACCAAA

Immunofluorescence

The cells on cover slips were fixed with 3% paraformaldehyde in PBS, permeabilized in a buffer (300 mm sucrose and 3 mm MgCl2 in PBS) containing 0.5% Triton X‐100 and incubated in PBS containing 0.5% nonfat dry milk and 0.1% Triton X‐100. The fixed and permeabilized cells were incubated with anti‐STAT1 or IRF1 antibodies diluted with PBS containing 0.5% nonfat dry milk. The cells on coverslips were washed with PBS containing 0.1% Triton X‐100 (PBST), incubated with secondary antibodies conjugating with AlexaFluor dyes (Molecular Probes, Eugene, OR, USA), washed extensively with PBST, and incubated with TO‐PRO‐3 (molecular probes). All fluorescence images were captured by a confocal microscope (LSM 5 Exciter; Carl Zeiss, Oberkochen, Germany).

Results

NPM1 regulates the transcription of IFN‐γ‐induced genes

Previously, we performed a comprehensive microarray analysis of the effect of NPM1 knockdown on gene expression in HeLa cells [7], where 539 genes were found to be downregulated (< 0.669‐fold). Gene ontology analysis of these genes showed functional enrichment in immune responses including antigen processing and presentation via MHC class I (Fig. 1A). We also noticed that the immune response genes decreased by NPM1 knockdown are induced by IFN‐γ; therefore, we questioned whether NPM1 is involved in the type II IFN signaling pathway. We first focused on the genes encoding the class I and II antigen presentation machinery. To confirm the microarray results, RT‐qPCR was performed using HeLa cells treated with control or NPM1 siRNA and IFN‐γ (Fig. 1B,C). NPM1 was efficiently reduced by NPM1 siRNA treatment (Fig. 1B). The expression of the human MHC class I gene, HLAB, was detected at a low level and that of the MHC class II genes, HLADR and HLADQ, was not detected under nonstimulated conditions. The expression of both MHC class I and II genes was greatly increased upon IFN‐γ treatment. Interestingly, we demonstrated that the expression of these MHC genes decreased by NPM1 knockdown, suggesting that NPM1 is involved in the type II IFN signaling pathway. To gain insight into the function of NPM1 in the type II IFN signaling pathway, we next focused on the transcription regulators of IFN‐γ‐induced transcription (Fig. 1D). It is well‐established that STAT1 is a master regulator of the type II IFN signaling pathway and activated STAT1 induces downstream genes such as IRF1, CIITA, NLRC5, and STAT1 itself by binding to the consensus sequence (GAS) in their promoters [12, 14, 20]. These transcription factors induced by STAT1 are required for IFN‐γ‐induced expression of the MHC genes. Under nonstimulated conditions, the expression of CIITA was not detected, indicating that CIITA is required for the expression of the MHC II genes in HeLa cells. The expression levels of STAT1, IRF1, CIITA, and NLRC5 were increased by IFN‐γ treatment and those of STAT1, IRF1, and CIITA, but not NLRC5, were significantly reduced by NPM1 knockdown. These results raised the possibility that NPM1 is selectively involved in the regulation of a subset of STAT1 target genes. We also examined whether NPM1 knockdown decreases the levels of STAT1 and IRF1 by quantitative western blotting (Fig. 1E) using the level of actin as a loading control. Consistent with the RT‐qPCR results, the expression of the STAT1 protein in NPM1 knockdown cells was lower than that in control cells 6–24 h after IFN‐γ treatment. However, the levels of IRF1 in control and NPM1 siRNA‐treated cells were similarly increased after IFN‐γ treatment.

image
NPM1 regulates the transcription of IFN‐γ‐induced genes. (A) Gene ontology analysis of the gene set decreased by NPM1 knockdown. The gene ontology analysis was performed using the 539 downregulated genes and the list of the enriched functions was shown. The previous microarray data (NCBI Gene Expression Omnibus (GEO) under accession number GSE81785) was used. (B) Knockdown of NPM1 by siRNA. Expression of NPM1 in HeLa cells treated with control or NPM1 siRNA was examined by western blotting using anti‐NPM1 antibody. Actin was used as a loading control. (C,D) RT‐qPCR analyses. RNA was extracted from control or NPM1 knockdown HeLa cells treated without or with IFN‐γ (20 ng·mL−1) for 24 h as indicated at the bottom of the graphs and RT‐qPCR was performed using gene‐specific primers. White and black bars indicate the results from control and NPM1 siRNA, respectively. Relative mRNA levels were normalized by the expression level of GAPDH. Three independent experiments were performed and error bars indicate ± SD. The results were statistically analyzed by t‐test and ***, **, and * represent P < 0.001, 0.01, and 0.05, respectively. (E) Effect of NPM1 knockdown on the expression level of STAT1 and IRF1. HeLa cells were treated with control or NPM1 siRNA for 72 h and IFN‐γ (20 ng·mL−1) was added and further incubated for 6, 12, and 24 h. The expression levels of STAT1, IRF1, NPM1, and β‐actin were examined by western blotting. The band intensities of STAT1, IRF1, and β‐actin were measured and those of STAT1 and IRF1 were normalized by the intensity of β‐actin. The normalized intensities of STAT1 and IRF1 in control cells at 6 h after IFN‐γ addition were set as 1.0 and the relative intensities were graphically shown at the bottom graphs. Three independent experiments were performed and error bars indicate ± SD. The results were statistically analyzed by t‐test and ns and * represents not significant and P < 0.05, respectively.

To examine the function of NPM1 in the type II IFN signaling pathway, we next examined the localization of NPM1 in cells treated with IFN‐γ using HeLa cells stably expressing EGFP‐tagged NPM1. NPM1 mainly localizes to the nucleoli in control cells and shuttles between the nucleoplasm and the nucleoli. On IFN‐γ treatment, NPM1 localization was not clearly changed, while STAT1 and IRF1 were clearly accumulated in the nuclei (Fig. 2A). We also examined the expression of the genes involved in the IFN‐γ signaling pathway: IFNGR1 and IFNGR2 and kinases (Jak1 and Jak2) that phosphorylate STAT1 (Fig. S1). The expression of these genes was not induced by IFN‐γ and NPM1 knockdown did not significantly affect their expression. We next examined STAT1 phosphorylation after IFN‐γ treatment in control and NPM1 knockdown cells (Fig. 2B,C). STAT1 phosphorylation at tyrosine 701 (Y701) was clearly detected 15 min after IFN‐γ addition in both control and NPM1 knockdown cells. Quantitative analysis by western blotting revealed that the level of STAT1 protein and its phosphorylation was not significantly affected by NPM1 knockdown during or 180 min after IFN‐γ treatment, although the level of STAT1 protein in NPM1 knockdown cells was slightly lower than in control cells (Fig. 2C). In addition, STAT1 was similarly accumulated in the nuclei in both control and NPM1 knockdown cells 1 h after IFN‐γ treatment (Fig. 2D). In parallel, we showed that accumulation of IRF1 in control and NPM1 knockdown cells was not significantly different. These results suggest that NPM1 regulates the type II IFN signaling pathway after IRF1 and STAT1 are translocated to and accumulate in the nucleus.

image
NPM1 does not affect the IFN‐γ‐induced nuclear accumulation of STAT1 and IRF1. (A) Localization of NPM1 after IFN‐γ treatment. HeLa cells stably expressing EGFP‐NPM1 were treated without or with IFN‐γ (20 ng·mL−1) for 6 h, followed by immunofluorescence analysis with anti‐STAT1 (left panels) or anti‐IRF1 (right panels) antibody. Localization of the proteins was observed by a confocal microscope. (B,C) Effect of NPM1 knockdown on the level of STAT1 Y701 phosphorylation. The cell extracts prepared from HeLa cells treated with control or NPM1 siRNA after IFN‐γ (20 ng·mL−1) treatment were separated by SDS/PAGE and analyzed by western blotting with anti‐STAT1 phosphorylated at tyrosine 701 (p‐STAT1), ‐STAT1, ‐NPM1, and ‐β‐actin antibodies. Time (min) after IFN‐γ treatment was shown at the top of the panel. The band intensities of STAT1 and STAT1 (p‐Y701) were normalized by that of β‐actin and relative intensities were graphically shown in (C). The normalized intensities of STAT1 or STAT1 (P‐Y701) in control cells at 15 min after IFN‐γ (20 ng·mL−1) addition were set as 1.0 and relative intensity was calculated. Three independent experiments were performed and error bars indicate ± SD. (D) Localization of STAT1 and IRF1 in NPM1 knockdown cells. HeLa cells were treated with control or NPM1 siRNA and stimulated by IFN‐γ for the indicated time periods. STAT1 or IRF1 was visualized by immunofluorescence staining using anti‐STAT1 and ‐IRF1 antibodies, respectively. DNA was stained with TO‐PRO‐3. Localizations were observed by a confocal microscopy.

NPM1 affects STAT1‐mediated transcription

Because NPM1 depletion decreased the IFN‐γ induced expression of STAT1 (see Fig. 1E), it is likely that NPM1 affects both the type I and II IFN signaling pathways. To address this point, we performed reporter assays using pGAS‐TA‐luc and pISRE‐TA‐luc reporter plasmids (Fig. 3A,B), which contain binding sites for STAT1 homodimer (GAS) and ISGF3 (ISRE), respectively. We observed that IFN‐γ‐induced expression of the reporter gene, but not IFN‐β‐induced expression, was significantly reduced by NPM1 knockdown. These results suggest that NPM1 is involved in the regulation of the type II IFN signaling pathway, and that the decreased type II IFN‐induced gene expression by NPM1 is not simply explained by decreased STAT1 expression.

image
NPM1 regulates the IFN‐γ‐induced, but not IFN‐β‐induced transcription through direct interaction with STAT1. (A,B) Luciferase assays with GAS‐Luc and ISRE‐Luc reporters. HeLa cells treated with control or NPM1 siRNA were transfected with pGAS‐TA‐Luc (A) or pISRE‐TA‐Luc (B) with pTA‐Renilla Luc vectors. Twenty‐four hours post‐transfection, the cells were stimulated without or with IFN‐γ (20 ng·mL−1) (A) for 24 h or IFN‐β (1000 IU·mL−1) (B) for 3 h and subjected to luciferase reporter assay. Luciferase activity of each sample was normalized to Renilla luciferase activity to calculate relative luciferase activity. Three independent experiments were performed and error bars indicate ± SD. The results were statistically analyzed by t‐test, and ** and * represent P < 0.01 and 0.05, respectively. (C) Immunoprecipitation analysis of STAT1. Cell extracts were prepared from HeLa cells treated without or with IFN‐γ (20 ng·mL−1) for 1 h, and the interaction between NPM1 and STAT1 was analyzed by immunoprecipitation with control IgG or anti‐STAT1 antibody. The input (lanes 1 and 2) and immunoprecipitated (lanes 3–6) proteins were separated by SDS/PAGE and analyzed by western blotting with anti‐NPM1, ‐STAT1, and ‐p‐STAT1 antibodies. (D) Purified recombinant proteins. Recombinant GST, GST‐tagged NPM1, and Flag‐tagged STAT1 proteins were separated by SDS/PAGE and visualized with CBB staining. Lane M is a molecular size marker. (E) GST pull‐down assay. GST or GST‐tagged NPM1 (lanes 2 and 3, 1 μg) was mixed and incubated with purified Flag‐STAT1. The protein bound to GST proteins were examined by western blotting with anti‐Flag antibody, and the GST proteins were visualized by CBB staining.

To clarify the mechanism by which NPM1 regulates STAT1‐mediated transcription, we examined the endogenous interaction between NPM1 and STAT1 by immunoprecipitation with anti‐STAT1 antibody in HeLa cells treated with or without IFN‐γ (Fig. 3C). Endogenous NPM1 was coimmunoprecipitated with STAT1 independent of IFN‐γ treatment. To test whether NPM1 directly interacts with STAT1, we prepared recombinant proteins of GST, GST‐tagged NPM1, and Flag‐tagged STAT1 (Fig. 3D), and GST pull‐down assays were performed (Fig. 3E). Flag‐tagged STAT1 precipitated with GST‐tagged NPM1, but not with GST, indicating that NPM1 directly associates with STAT1. These results suggest that NPM1 is involved in the type II IFN pathway by direct interaction with STAT1.

NPM1 regulates the CIITA gene expression

Next, we examined whether NPM1 regulates the promoter activity of STAT1 target genes. To this end, we chose the promoter activity of the CIITA gene because its expression is absolutely dependent on IFN‐γ and is significantly decreased by NPM1 knockdown in HeLa cells (see Fig. 1). The expression of the CIITA gene is controlled by four different promoters: pI, pII, pIII, and pIV [21]. CIITA pI and pIII are active in cells of myeloid and lymphoid origins, respectively, while the significance of CIITA pII remains unknown [22]. CIITA pIV is induced by IFN‐γ in most cell types; therefore, we focused on this promoter element. The proximal promoter region of pIV contains multiple cis‐acting elements recognized by transcription factors such as NF‐κB, NF‐GMa, STAT1 (GAS), USF1 (E box), and IRF1 [23]. To examine the effect of NPM1 on the promoter activity of CIITA pIV upon IFN‐γ treatment, we prepared the proximal promoter of human CIITA pIV with a series of 5′ deletion mutants and performed reporter assays (Fig. 4A). Consistent with the decreased expression of endogenous CIITA in NPM1 knockdown cells, the reporter activity of the pGV‐B‐CIITA‐237 construct, which contains 237 base pairs (bp) upstream and 115 bp downstream of the transcription start site (+1) of the CIITA gene, was significantly decreased by NPM1 knockdown. The reporter activity of the NFκB‐binding element deletion construct (pGV‐B‐CIITA‐154) was similar to that of the full‐length construct and was decreased by NPM1 depletion, suggesting that NFκB is not involved in the regulation of the CIITA gene under the assay conditions employed. Further deletion of three elements, the NF‐GMa‐binding site, GAS, and the E box (pGV‐B‐ CIITA‐82), partially reduced IFN‐γ‐induced reporter gene expression, and its reporter activity induced by IFN‐γ was decreased by NPM1 knockdown. Conversely, the deletion construct pGV‐B‐CIITA‐54 abolished IFN‐γ‐induced expression of the reporter gene, and the reporter activity of this construct was not affected by NPM1 knockdown.

image
NPM1 regulates the transcription of CIITA gene through the IRF1‐binding element. (A) Luciferase assay with CIITA pIV. The proximal cis‐acting elements [CIITA‐237, 237 bp upstream and 115 bp downstream of transcription start site (+1)] of the CIITA pIV was cloned and used for reporter assay. 5′‐deletion mutants, CIITA‐154, ‐82, and ‐54 were also constructed and examined the luciferase activity. HeLa cells treated with control or NPM1 siRNA were transfected with pGV‐B‐CIITA plasmids with pTA‐RL vectors. Twenty‐four hours post‐transfection, the cells were stimulated without or with IFN‐γ (20 ng·mL−1) for 24 h and subjected to luciferase reporter assay. Luciferase activity of each sample was normalized by Renilla luciferase activity and the activity of HeLa cells treated with control siRNA and without IFN‐γ was set as 1.0, and the relative reporter activity was calculated. Three independent experiments were performed and error bars indicate ± SD. The results were statistically analyzed by t‐test, and ** and * represent P < 0.01 and 0.05, respectively. (B,C) Luciferase assay with the reporter plasmids containing mutations at GAS (B) and IRF1‐binding site (C) in the CIITA pIV. Experiments and data calculation were performed as in (A).

To further confirm the result of the CIITA promoter analysis by NPM1 knockdown, we generated constructs with site‐specific mutations either in the GAS or IRF1‐binding sequences. In accordance with the results of the GAS element deletion construct (pGVB CIITA‐82), the mutation of GAS slightly decreased IFN‐γ induction and IFN‐γ‐induced expression of this construct was decreased by NPM1 knockdown (Fig. 4B). When the CIITA promoter contained mutations at the IRF1‐binding site, the IFN‐γ‐induced reporter activity was not only abolished but also not affected by NPM1 knockdown (Fig. 4C). These results suggest that NPM1 regulates the IFN‐γ‐induced stimulation of CIITA pIV via IRF1, although we could not completely exclude the possibility that NPM1 regulates STAT1 binding to the CIITA pIV.

NPM1 binds to IRF1 through the oligomerization domain

Previous study has shown that NPM1 interacts with IRF1 through its multifunctional domain 2 in vitro [24]. To confirm this interaction, Flag‐tagged IRF1 was expressed in 293T cells and an immunoprecipitation assay performed (Fig. 5A). 293T cells were used here to obtain sufficient amounts of Flag‐IRF1 for immunoprecipitation. We found that Flag‐tagged IRF1 binds to endogenous NPM1 in the absence or presence of INF‐γ treatment. To determine the IRF1‐binding region of NPM1, GST pull‐down assays with a series of NPM1 deletion mutant proteins were performed (Fig. 5B). The two splicing variants of NPM1/B23.1, namely B23.2 and B23.3, which lack the C‐terminal RNA‐binding domain and the basic region, respectively, interacted with IRF1 (Fig. 5C). This indicates that the C‐terminal domain and the basic region are dispensable for the interaction. The two highly acidic regions are known requirements for efficient histone binding and nucleosome assembly [25]. The deletion of these acidic regions did not affect the interaction with IRF1 (Fig. 5D, lanes 7–8). Further analyses showed that the C‐terminal half of the protein (amino acid 121–294) was dispensable for IRF1 binding, and the N‐terminal oligomerization domain (amino acid 1–120) was sufficient to interact with IRF1.

image
IRF1 interacts with the oligomerization domain of NPM1. (A) Immunoprecipitation assay. Cell extracts were prepared from 293T cells expressing Flag‐tagged IRF1 treated without or with IFN‐γ for 6 h, and immunoprecipitation was performed with anti‐Flag M2 beads. Input (lanes 1–4) and immunoprecipitated proteins (lanes 5–8) were separated by SDS/PAGE and subjected to western blotting with anti‐Flag and ‐NPM1 antibodies. (B) Diagram of the splicing variants and truncated mutants of NPM1. Black, light gray, and dark gray boxes indicate oligomerization domain, acidic regions, and the C‐terminal globular domain, respectively. (C,D) GST pull‐down assay. GST or GST‐tagged IRF1 (1 μg per sample) immobilized on glutathione sepharose beads were incubated with His‐tagged NPM1/B23 proteins (1 μg per sample). The beads were extensively washed and the bound proteins were separated by SDS/PAGE and visualized by CBB staining. The positions of the His‐tagged proteins co‐precipitated with GST‐tagged IRF1 are indicated at the left side of each lane.

Discussion

In this study, we demonstrated that NPM1 regulates a subset of IFN‐γ‐inducible genes such as the MHC class I and II genes (Fig. 1C). The effect of NPM1 knockdown is likely due to the decreased expression of the transcription regulators, STAT1, IRF1, and CIITA (Fig. 1D), all of which are induced by IFN‐γ. The STAT1 protein level induced by IFN‐γ treatment was also decreased by NPM1 knockdown (Fig. 1E), suggesting that the regulation of the STAT1 expression is a primary function of NPM1 in the type II IFN signaling pathway. Given that NPM1 did not affect the phosphorylation of STAT1 at tyrosine 701 or the nuclear accumulation of STAT1 and IRF1 (Fig. 2B–D), it is suggested that NPM1 regulates the type II IFN signaling pathway after the nuclear accumulation of STAT1 and IRF1. Although we demonstrated that NPM1 directly binds to STAT1 and regulates the expression of a reporter gene containing GAS (Fig. 3A,C,E), NPM1 failed to regulate the STAT1 target gene NLRC5. These results suggest that NPM1 confers preferential binding sequence specificity on STAT1. Sequence variation in GAS or the sequences adjacent to GAS may affect the sequence preference of the STAT1‐NPM1 complex.

In HeLa cells, the expression of the MHC class II genes and their regulator CIITA was not detected by RT‐PCR. This supports the previous finding that CIITA is an essential transcription regulator of the MHC class II genes, but not the MHC class I genes. Our results imply that NPM1 regulates the expression of the MHC class II genes via decreased expression of the CIITA gene. Although the CIITA pIV promoter contains STAT1‐binding sites, NPM1 knockdown decreased the activity of the CIITA pIV even in the absence of STAT1‐binding sites (Fig. 4A,B). This suggests that NPM1 regulates the expression of CIITA independent of STAT1. It was previously demonstrated that the binding of STAT1 to CIITA pIV depends on the transcription factor USF1, which binds to the E box on CIITA pIV [23]. This local environment of GAS on CIITA pIV may be why NPM1 does not affect STAT1 binding. Thus, it is likely that NPM1 regulates the function of IRF1 in IFN‐γ‐induced expression of CIITA. However, STAT1 did not stimulate CIITA pIV activity when the IRF1‐binding site was mutated; therefore, we could not exclude the possibility that NPM1 regulates CIITA pIV via interaction with both STAT1 and IRF1.

Consistent with previous studies [10, 24], we found that NPM1 shows potential association with IRF1 (Fig. 5). Although a previous study reported that NPM1 inhibits the DNA binding of IRF1 [10], our results suggest that NPM1 positively regulates IRF1 function. Further study is required to address this discrepancy and to clarify the molecular mechanism by which NPM1 regulates the IRF1 function.

Although the effect of NPM1 knockdown on the expression of the reporter gene containing ISRE induced by IFN‐β was not clearly observed (Fig. 3B), we cannot exclude the possibility that NPM1 regulates the type I IFN‐inducible genes that contain different ISRE sequences or ISRE adjacent to cis‐regulatory elements. It is possible that NPM1 associates with and regulates the function of ISGF3 through its STAT1‐binding activity.

Here, we demonstrated that NPM1, an oncogenic nucleolar protein, is involved in the regulation of the type II IFN signaling pathway. INF‐γ is a well‐established proinflammatory cytokine that plays critical roles in both the acquired and innate immune systems, host defense, and in tumor surveillance [12]. It also plays a role in enhancing the inflammatory responses in damaged sites and tumor microenvironments. We previously demonstrated that NPM1 regulates the TNF‐α inflammatory response by enhancing the DNA‐binding activity of NF‐κB [7]. Our results suggest a key regulatory role of NPM1 in inflammation and various diseases including cancer caused and/or enhanced by inflammation.

Acknowledgements

This work was supported by Grants‐in‐aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan to KN (24115001) and MO (26440021 and 17K07300).

    Author contributions

    MA and MO designed the research, KN and MO supervised the research, MA and JL performed experiments, all authors analyzed the data, and MA and MO wrote the manuscript. All authors read and approved the final manuscript.