Ipr1 Regulation by Cyclic GMP-AMP Synthase/Interferon Regulatory Factor 3 and Modulation of Irgm1 Expression via p53

Intracellular pathogen resistance 1 (Ipr1) has been found to be a mediator to integrate cyclic GMP-AMP synthase (cGAS)–interferon regulatory factor 3 (IRF3), activated by intracellular pathogens, with the p53 pathway. Previous studies have shown the process of Ipr1 induction by various immune reactions, including intracellular bacterial and viral infections. The present study demonstrated that Ipr1 is regulated by the cGAS-IRF3 pathway during pathogenic infection.

ABSTRACT

Intracellular pathogen resistance 1 (Ipr1) has been found to be a mediator to integrate cyclic GMP-AMP synthase (cGAS)–interferon regulatory factor 3 (IRF3), activated by intracellular pathogens, with the p53 pathway. Previous studies have shown the process of Ipr1 induction by various immune reactions, including intracellular bacterial and viral infections. The present study demonstrated that Ipr1 is regulated by the cGAS-IRF3 pathway during pathogenic infection. IRF3 was found to regulate Ipr1 expression by directly binding the interferon-stimulated response element motif of the Ipr1 promoter. Knockdown of Ipr1 decreased the expression of immunity-related GTPase family M member 1 (Irgm1), which plays critical roles in autophagy initiation. Irgm1 promoter characterization revealed a p53 motif in front of the transcription start site. P53 was found to participate in regulation of Irgm1 expression and IPR1-related effects on P53 stability by affecting interactions between ribosomal protein L11 (RPL11) and transformed mouse 3T3 cell double minute 2 (MDM2). Our results indicate that Ipr1 integrates cGAS-IRF3 with p53-modulated Irgm1 expression.

INTRODUCTION

Viral and bacterial infection of cells begins with the transfer of pathogenic DNA into the cytosol of the cell, which initiates various intracellular defenses. Cyclic GMP-AMP (cGAMP) synthetase (cGAS) acts as an initial intracellular pattern recognition receptor that senses the presence of pathogenic DNA in the cytosol and then transfers the signal to stimulator of interferon genes (STING), TANK binding kinase 1 (TBK1), and finally interferon regulatory factor 3 (IRF3). IRF3 is the key transcriptional regulator of the cGAS-IRF3 axis and plays a critical role in the innate immune response against viral DNA and RNA infection (1, 2). Clinical studies have shown that Listeria monocytogenes and Mycobacterium tuberculosis infections activate cGAS-dependent pathways upon injection of pathogenic DNA into the cell. Furthermore, the cGAS-IRF3 axis can be activated by interferon (IFN)-stimulatory DNA (ISD) (45-bp non-CpG oligomers from L. monocytogenes), cytosolic RNA, DNA hybrids, and minor DNA damage (3, 4). The cGAS-IRF3 axis functions in defense against microbial infection, tumor initiation, growth, and metastasis via autophagy regulation (5). Moreover, several novel IRF3-dependent genes that regulate cellular autophagy pathways have also been identified (6, 7). These studies indicated that the intracellular-DNA-activated cGAS-IRF3 signaling pathway may be involved in regulation of the expression of autophagy-related genes, such as the IRG family M protein 1 gene (Irgm1).

The Intracellular pathogen resistance 1 gene (Ipr1) is involved in mycobacterium-related immunity and can be activated by type II IFNs (8); the Speckle protein 110b gene (SP110b) is the Ipr1 homologue in humans (9). Pan and colleagues determined that Ipr1 expression in the extremely susceptible mouse strain C3HeB/FeJ significantly limited M. tuberculosis bacterial proliferation in the lung, suggesting that Ipr1 effectively controls the multiplication of intracellular pathogens (10). Moreover, human SP110 may function as a transcriptional coactivator of nuclear hormone receptors (11). Ipr1/SP110b consists of an N-terminal Sp100 domain and a C-terminal SAND domain conserved in Sp100, autoimmune regulator 1, nuclear phosphoprotein 41/75, and deformed epidermal autoregulatory factor 1. The Sp100 domain is critical for Ipr1 protein translocation into promyelocytic leukemia (PML) nuclear bodies (NBs) by forming a homodimer, whereas the SAND domain is conserved in a range of proteins, and the SAND domain in deformed epidermal autoregulatory factor 1 has been shown to bind the TTCG motif and acts as a transcriptional regulator (9, 12). Though the domains of Ipr1 have been verified, the function of Ipr1 in regulating gene expression is still unknown.

Like Ipr1, immunity-related GTPases (IRGs) belong to the family of IFN-inducible (IFI) GTPases and function as a cell-autonomous resistance system for detection of pathogens, such as M. tuberculosis (13). Irgm1, also known as IFI1, is essential for host macrophage resistance to phagosomal pathogens and stem cell development (14–16). Irgm1 has been proven to promote phagosome maturation and induction of autophagy, especially in mycobacterial immunity (14, 17). Knockdown of Irgm1 in RAW264.7 cells could significantly reduce LC3 spots induced by IFN-γ (14). Furthermore, modulation of IRGM in humans affected the efficacy of autophagy, and decreased IRGM was involved in Crohn’s disease (14, 18). Though Ipr1 and Irgm1 play principal roles in host resistance to microbial infection, the exact transcriptional regulation of these genes has not yet been revealed.

Recent studies have indicated that p53, a well-known tumor suppressor, is another direct transcriptional target of type I IFNs upon viral infection (19). In turn, p53 influences the type I IFN pathway through direct transcriptional upregulation of several genes, including IRF9, IRF5, and the Toll-like receptor 3 gene (20). Under normal conditions, p53 is maintained at low levels by ubiquitin-mediated proteasomal degradation, which is regulated by Mdm2, encoding one of the ubiquitin E3 ligases. Dactinomycin (Act D) is a chemotherapeutic used to treat a number of cancers, because low-dose Act D can activate p53. Furthermore, recent studies have shown that the PML NB potentiates p53 function and increases its half-life by interacting with Mdm2 (21). Though Ipr1 is a component of PML NBs, it is unclear whether Ipr1 participates in the regulation of p53 or p53-induced gene expression.

Therefore, the present study determined that Ipr1 was regulated by the cGAS-IRF3 pathway and participated in P53-mediated Irgm1 transcriptional regulation, which revealed an unknown function of Ipr1 and gave new insight into cross-links between different signaling pathways.

RESULTS

ISD induces Ipr1 expression by regulating its promoter.

Previous animal-based and clinical research identified Ipr1 as a genetic determinant that confers host immunity to M. tuberculosis infection, while its human ortholog, SP110b, was found to contribute to hepatitis B virus (HBV) persistence (Fig. 1A) (9, 22). In the present study, the core promoter regions of Ipr1 and SP110b were found to share highly similar cis-regulatory element distributions (8). The NCBI Gene Expression Omnibus (GEO) profiles (https://www.ncbi.nlm.nih.gov/geoprofiles/) were used to find activators of Ipr1/SP110; viruses such as Sendai virus, influenza virus H1N1, and hepatitis B virus, as well as ordinary (e.g., H37Rv) strains of M. tuberculosis, etoposide, and cytarabine, were found to significantly enhance Ipr1/SP110 expression (Fig. 1B to F and G). One of the most common occurrences in viral or bacterial infection is the transfer of microbial DNA fragments into host cells (23). Normally, the microbial nucleic acids are sensed by cGAS, leading to activation of innate immune responses (24), and the ISD sequence was first identified in L. monocytogenes as an activator of the cGAS-IRF3 axis (24). To test the effect of exogenous DNA on Ipr1 expression, RAW264.7, J774A.1, and NIH 3T3 cells and bone marrow-derived macrophages (BMDMs) were transfected with ISD (2 μg/ml), which resulted in the specific increase of Ipr1 transcription and subsequent protein levels in RAW264.7 cells and BMDMs (Fig. 2A and B). The ISD sequence was transiently transfected into RAW264.7 and J774A.1 cells with Ipr1-PRO-594 (which contains the Ipr1 promoter in its core region) and pGL4.10, respectively, which caused remarkable induction of Ipr1 promoter transcriptional activity (Fig. 2C). Like type II IFNs or interleukins (ILs), ISD can enhance immune-related pathways, such as NF-κB (NF-κB–luc) and IRF (ISRE-luc) pathways, which are highly activated in RAW264.7 cells (Fig. 2D). To identify the cis-regulatory elements of the promoter, the Ipr1 promoter sequences of various species were downloaded from the University of California—Santa Cruz (UCSC) database to determine their similarity to the mouse sequence. The homology of various promoter regions was analyzed using DNAMAN (version 6.0.3.99; Lynnon Corp., Quebec, Canada), and the evolutionary tree suggested the evolution of mouse and rat Ipr1 promoters diverged widely from that of the human promoter (Fig. 2E). cis-regulatory elements were predicted by the JASPAR database (http://jaspardev.genereg.net/) using the Ipr1 core promoter region of various species followed by clustering of predicted transcription factors (Fig. 2F), and a total of 10 transcription factors were found to be common to various species (listed in Table 1), from which we found that the IFN-stimulated response element gene (ISRE) motifs (identified by IRFs) were common between mouse and human. Subsequently, we determined that ISD treatment upregulated the expression of SP110 in THP-1 cells (Fig. 2G).

FIG 1.

Upregulation of SP110 upon microbe infection or molecular treatment. (A) The expression of SP110 in an HBV-associated acute liver failure group was higher than in a normal-liver group. The values of HBV-associated acute liver failure groups were normalized to the mean of the normal-liver group (+SD). The GEO profile number is GDS4387 (GPL570 platform no. 209761_s_at). The data set title is “Hepatitis B virus (HBV)-associated acute liver failure (ALF) patients: liver explant” (58). **, P < 0.01. (B) Sendai virus infection effect on SP110 induction in a monocytic cell line. The values of Sendai virus-infected groups were normalized to the mean of the uninfected (UI) group (+SD). The GEO profile number is GDS6082 (GPL570 platform no. ILMN_2415144). The data set title is “Sendai virus infection effect on monocytic cell line: dose response” (59). (C) Pandemic and seasonal influenza virus H1N1 infection effect on SP110 induction in bronchial epithelial cells. The values of various H1N1 strain-infected groups were normalized to the mean of the UI group (+SD). The GEO profile number is GDS4855 (GPL570 platform no. 208012_x_at). The data set title is “Pandemic and seasonal H1N1 influenza virus infections of bronchial epithelial cells in vitro” (60). (D) Effect of H37Rv infection on SP110 induction in THP-1 cells. The values of the H37Rv-treated group were normalized to the mean of the UI group (+SD). The GEO profile number is GDS4781 (GPL570 platform no. 8059650). The data set headline is “Mycobacterium tuberculosis H37Rv-infected macrophage response to vitamin D” (61). (E) Effect of Mycobacterium bovis BCG infection on SP110 induction in THP-1 cells. The values of the BCG-treated group were normalized to the mean of the UI group (+SD). The GEO profile number is GDS2180 (GPL570 platform no. AGhsA070521). The data set title is “M-CSF and GM-CSF differentiated macrophage response to bacillus Calmette-Guerin (subset A)” (62). (F) The expression of SP110 in the etoposide-treated group was higher than that in the negative-control (NC) group. The values of the etoposide-treated group were normalized to the mean of the NC group (+SD). The GEO profile number is GDS5809 (GPL570 platform no. 8059650). The data set heading is “Anti-leukemia drug etoposide effect on ecotropic viral integration site 1-overexpressing myeloid cells” (63). (G) Cytarabine treatment leading to SP110 induction. The values of the DMSO-, doxorubicin-, puromycin-, and cytarabine-treated group were normalized to the mean of the DMSO (24-h) group (+SD). The GEO profile number is GDS2733 (GPL570 platform no. 208012_x_at). The data set headline is “Cytosine arabinoside effect on Ewing’s sarcoma cell line: time course and dose response” (64).

FIG 2.

ISD induces Ipr1 expression through its promoter. (A) ISD activates Ipr1 expression in murine macrophages (RAW264.7 and J774A.1 cells), BMDMs, and fibroblasts (NIH 3T3 cells). The cells were transfected with ISD (2 μg/ml). After 36 h, the expression of Ipr1 was determined by qRT-PCR analysis. The values of ISD⁺ groups were normalized to the mean of ISD⁻ groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (*, P < 0.05; **, P < 0.01). (B) ISD induced Ipr1 expression in RAW264.7 cells and BMDMs by Western blotting in a time gradient. RAW264.7 cells and BMDMs were transfected with ISD for the indicated time intervals, and then protein levels were determined using anti-SP110 and anti-GAPDH by Western blotting. (C) ISD upregulates transcriptional activity of Ipr1 promoter reporter constructs in RAW264.7 and J774A.1 cells. Cells were transfected with promoter reporter constructs and the pGL4.73 vector for 24 h, and then ISD was transfected and the cells were incubated for 8 h. The transcriptional activity of reporter constructs was determined by luciferase assays. Values were normalized to the mean of the pGL4.10 (ISD⁻) group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (D) ISD activates luciferase activities of pathway reporter constructs ISRE-luc and NF-κB-luc. RAW264.7 cells were transfected with reporter plasmids pISRE-luc, pNF-κB-luc, pSRE-luc, pNFAT-luc, and pGRE-luc. After 24 h of transfection, the cells were transfected with 2 μg/ml ISD for 8 h, and then, the activity (+SD) of reporter constructs was determined by luciferase assays. The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (E) Homology analysis of the Ipr1 promoter region in different species. The different species’ Ipr1 promoter regions were downloaded from the UCSC database, and the sequence alignment was done with DNAMAN. (F) Classified analysis of transcription factors in different species’ Ipr1 promoters. The transcription factors of various species’ Ipr1 promoters were predicted using the JASPAR database with a relative profile score threshold of 80%. The clustering analysis used the online tool Venny 2.1.0 (bioinfogp.cnb.csic.es/tools/venny). (G) ISD activates SP110 expression in THP-1 cells. THP-1 cells were transfected with ISD for 24 h. The expression of Ipr1 was determined by qRT-PCR analysis (t test; **, P < 0.01).

TABLE 1.

Ten shared genes in various species and predicted site sequences in the mouse Ipr1 promoter

Gene name	Starta	Enda	Strand	Predicted site sequence
SP1	−75	−65	−1	CCTCCACCCCT
ZNF354C	−71	−66	−1	CTCCAC
KLF5	−74	−65	−1	CCTCCACCCC
ELF5	−274	−266	1	CACTTCCTA
HLTF	−179	−170	1	ATCCTTTGCC
SPIB	−93	−87	1	TGGGGAA
MEIS1	−164	−150	1	GCCTGTCACTCCACT
NFIC	−375	−370	−1	CTGGCA
RHOXF1	−182	−176	1	TAATCC
IRF3	−13	+7	1	GATCACTTTCATTTTTCTTTT

^aDistance from TSS (nucleotides).

IRF3 binds the ISRE of the Ipr1 promoter to regulate Ipr1 expression.

In our previous studies, signal transducer and activator of transcription binding elements gene (STATx), Rel, and ISRE motifs were found in the mouse Ipr1 promoter; the Rel motif was mouse specific, while the ISRE motif was common to the species discussed above (Fig. 2F and Table 1). In this study, different stimulators that activate macrophages through STATx, Rel, or ISRE motifs were used to stimulate Ipr1 expression. Nfkbia, Tgf-β, and Irf1 were selected as positive controls for tumor necrosis factor α (TNF-α), interleukin 13 (IL-13), and IFN-β, respectively (25–27). IFN-β, but not TNF-α and IL-13, was found to induce Ipr1 expression in RAW264.7 and J774A.1 cells (Fig. 3A and B). Next, the STATx (TTCNNNGAA), Rel (GGGRNTTT), and ISRE (GAAANNGAAANN) motifs were mutated according to nucleotide conservation collected from the JASPAR database (Fig. 3C). By performing a dual-luciferase reporter (DLR) assay, we found that mutation of ISRE decreased the transcriptional activity of the murine Ipr1 promoter (Fig. 4A), and the promoter containing a mutated ISRE motif failed to respond to ISD treatment (Fig. 4B). In previous studies, STAT1 was reported to be a key component of the IFN-β pathway. RAW264.7 cells were treated with ISD or not after transient transfection of pCMV-HA, pCMV-HA-STAT1, and Ipr1 promoter reporter plasmids with wild-type or mutated STATx motifs. This suggested that the presence of STAT1 affected Ipr1 promoter activity and that there should be some other mechanisms of ISD-induced Ipr1 promoter regulation (Fig. 4C). Considering that STAT1 was capable of recognizing the ISRE motif, pCMV-HA or pCMV-HA-STAT1 was transiently transfected into RAW264.7 cells with pGL4.10, Ipr1-PRO-594, or Ipr1-P594-ISREmut. The results showed that STAT1 could increase luciferase activity of Ipr1-PRO-594 and Ipr1-P594-ISREmut (Fig. 4D). To determine whether IRF3 could recognize the ISRE motif in the Ipr1 promoter, the pCMV-IRF3 or pCMV-Myc plasmid was cotransfected with the DLR plasmids into RAW264.7 cells. The results showed a large increase in luciferase activity in the wild-type group but almost no change in the mutated group (Fig. 4E). In addition, overexpression of myc-IRF3 in RAW264.7 cells significantly enhanced Ipr1 transcription (Fig. 4F). Furthermore, chromatin immunoprecipitation (ChIP) and electrophoretic mobility shift assay (EMSA) showed IRF3 specifically bound to the DNA sequence in the Ipr1 promoter (Fig. 4G and H).

FIG 3.

IFN-β induced Ipr1 in RAW264.7 and J774A.1 cells. (A) IFN-β induced Ipr1 transcription in RAW264.7 cells. RAW264.7 cells were treated with TNF-α, IFN-β, and IL-13 for 8 h. The expression of Ipr1 was measured by qRT-PCR analysis (NC, sterile PBS containing 0.1% bovine serum albumin). The values of the treated groups were normalized to the mean of the NC groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (B) IFN-β induced Ipr1 transcription in J774A.1 cells. J774A.1 cells were treated with TNF-α, IFN-β, and IL-13 for 8 h, and the expression of Ipr1 was measured by qRT-PCR analysis. The values of the treated groups were normalized to the mean of the NC groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (C) Locus-specific mutation of STATx, Rel, and ISRE motifs in the Ipr1 promoter. The motifs were downloaded from JASPAR, and nucleotide mutation relied on conservation of specific sites.

FIG 4.

The ISRE motif plays a critical role in ISD-induced Ipr1 expression. (A) ISRE is the core motif in the Ipr1 promoter. RAW264.7 cells were transfected with wild-type and mutated Ipr1 promoter reporter plasmids for 24 h, and transcriptional activity was examined by dual-luciferase assay. The values of wild-type and mutated Ipr1 promoter groups were normalized to the mean of the pGL4.10 group (+SD). The statistical significance of differences in relative luciferase activities was assessed by one-way ANOVA (**, P < 0.01). (B) Mutation of the ISRE motif impairs ISD-induced Ipr1 expression. RAW264.7 cells were transfected with wild-type and mutated Ipr1 promoter reporter plasmids for 24 h, followed by 8 h of transfection with ISD. Transcriptional activity was examined by dual-luciferase assay. The values of wild-type and mutated Ipr1 promoter groups were normalized to the mean of the pGL4.10 group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (C) Effect of STAT1 and/or ISD on the Ipr1 promoter. pCMV-HA or pCMV-HA-STAT1 was cotransfected with Ipr1-PRO-594 or Ipr1-P594-STATmut, respectively, into RAW264.7 cells for 36 h, and then the cells were treated with ISD for 8 h and the transcriptional activity was determined by luciferase assay. Values were normalized to the mean of the ISD⁻ pCMV-HA⁺ group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (D) Effect of STAT1 on Ipr1-P594-ISREmut. pCMV-HA or pCMV-HA-STAT1 was cotransfected with Ipr1-PRO-594 or Ipr1-P594-ISREmut, respectively, into RAW264.7 cells for 36 h, and the transcriptional activity was determined by luciferase assay. Values were normalized to the mean of the pGL4.10 pCMV-HA⁺ group (+SD). The statistical significance of differences in relative luciferase activity was assessed by two-way ANOVA (*, P < 0.05; **, P < 0.01). (E) Effect of IRF3 on the Ipr1 promoter. pCMV-myc or pCMV-IRF3 was cotransfected with Ipr1-PRO-594 or Ipr1-P594-ISREmut, respectively, into RAW264.7 cells for 36 h, and then transcriptional activity was determined by luciferase assay. The values of various treated groups were normalized to the mean of the Ipr1-PRO-594 group (+SD). The statistical significance of differences in relative luciferase activities was assessed by one-way ANOVA (*, P < 0.05; **, P < 0.01). (F) IRF3 increased Ipr1 transcription. RAW264.7 cells were transfected with pCMV-Myc and pCMV-IRF3 for 24 h, and then Ipr1 expression was measured by qRT-PCR (t test; **, P < 0.01). (G) ChIP analysis of IRF3 on the Ipr1 promoter region. RAW264.7 cells were transfected with pCMV-Flag-IRF3, and ChIP was performed using anti-Flag antibody, anti-H3 as a positive control antibody, and anti-IgG as a control antibody to detect enriched fragments. The values of ChIP groups were normalized to the mean of the IgG groups (+SD). The statistical significance of differences in relative enrichment was assessed by two-way ANOVA (**, P < 0.01). (H) the ISRE motif is critical for protein-promoter interaction. The sequence tgctcagtgtgctgagatcactttCATTTTTCTTTTCttgaagcctgact (lowercase letters are the flanking sequences of the ISRE motif) with biotinylation at the 5′ end was used as a nucleic acid probe (Ipr1-Biotin) and without modification was used as Ipr1-control, and the core sequence (CATTTTTCTTTTC) was mutated to CACCCTCACCTTTC and used as Ipr1-mut-Biotin. As shown, there was a shift band in the lane loaded with lysates and Ipr1-Biotin compared to Ipr1-mut-Biotin, as well as a significant decrease in the protein-probe complex when Ipr1-control was added.

ISD induces Irgm1 expression by regulating its promoter.

Irgm1 plays a role in the innate immune response via regulating autophagy formation, and its expression is associated with susceptibility to Crohn’s disease and tuberculosis (28). The expression of Irgm1 was synergistic with that of Ipr1. Similar to Ipr1, the Irgm1 mRNA and protein increased over time when RAW264.7 cells were treated with ISD (Fig. 5A and B). To illuminate the mechanism of ISD-induced Irgm1 expression in RAW264.7 cells, the promoter of Irgm1 was cloned and confirmed using DLR assay (Fig. 5C). H3K4me3 and H3K27ac are the markers of activated promoter regions. In murine bone marrow macrophages, the Irgm1 promoter showed much higher H3K4me3 and H3K27ac peaks than the control group of input signals (Fig. 5D), and the promoter region we cloned showed high levels of H3K4me3 and H3K27ac peaks in spleen and thymus (Fig. 5E). The DLR assay showed that the promoter region ranging from positions −470 upstream to +462 downstream of the Irgm1 transcription start site (TSS) had the highest transcriptional activity and the highest induction of ISD treatment (Fig. 5F).

FIG 5.

ISD induces Irgm1 expression through its promoter. (A and B) ISD induced Irgm1 expression in a time course in RAW264.7 cells. RAW264.7 cells were transfected with ISD (2 μg/ml). The mRNA was harvested after 0 h, 8 h, 16 h, and 24 h of treatment, and the total protein was harvested at 0 h, 24 h, 48 h, and 72 h. The expression of Irgm1 was determined using qRT-PCR and Western blotting. The values of the ISD-treated groups were normalized to the mean of the ISD 0-h group (+SD). The statistical significance of differences in relative mRNA expression was assessed by one-way ANOVA (**, P < 0.01). The proteins were determined with anti-LRG47 and anti-GAPDH. (C) Analysis of Irgm1 promoter activities. Irgm1 promoters with different lengths were transfected into RAW264.7 cells for 36 h. The relative luciferase activities of Irgm1 promoter groups were normalized to the mean of the pGL4.10 group (+SD). The statistical significance of differences in relative luciferase activities was assessed by one-way ANOVA (**, P < 0.01). (D and E) The Irgm1 promoter region has higher H3K4me3 and H3K27ac peaks in bone marrow macrophages, spleen, and thymus. The peaks of H3K4me3 and H3K27ac were obtained from the WashU EpiGenome Browser, and the peak scales were fixed from a minimum of 0 to a maximum of 30. (F) Effect of ISD on the Irgm1 promoter. Irgm1 promoter reporter plasmids were transfected into RAW264.7 cells for 24 h. Then, ISD (2 μg/ml) was transfected into RAW264.7 cells for 8 h. The transcriptional activity was examined by luciferase assay. Values were normalized to the mean of the Irgm1-PRO-500 (ISD⁻) group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (*, P < 0.05; **, P < 0.01).

Knockdown of Ipr1 decreases Irgm1 transcription.

Previously, we demonstrated that Ipr1 overexpression in RAW264.7 cells resulted in upregulation of miR155 and downregulation of miR125a (29). During M. tuberculosis infection, miR155 has been shown to promote autophagy, while miR125a inhibits autophagy activation (30, 31). Irgm1 is an autophagy initiation gene and has been identified as a critical anti-M. tuberculosis gene because of its effect on edophilin 1-dependent autophagy (32). To investigate whether Ipr1 affects Irgm1, Ipr1 was knocked down in RAW264.7 cells, which resulted in downregulation of Irgm1 mRNA and protein whether the cells were treated with ISD or not (Fig. 6A and B). As Ipr1/SP110b expression decreases transcription of Tnf-α (9, 29), the Tnf-α promoter region (−500 to +900 from the TSS) was cloned into the pGL4.10 vector as a positive control. Silencing of Ipr1 resulted in increasing Tnfa-PRO-1400 luciferase activity and decreasing Irgm1-PRO-900 luciferase activity (Fig. 6C). Cotransfection of pEGFP-Ipr1 and Irgm1 promoter reporter plasmids showed that Ipr1 could increase the promoter transcriptional activity of Irgm1 in RAW264.7 and J774A.1 cells (Fig. 6D). The luciferase activity of the Tnf-α promoter was significantly decreased when cells were transfected with pEGFP-Ipr1, which was reported in human macrophages, as well (9).

FIG 6.

Knockdown of Ipr1 decreases Irgm1 transcription. (A) Knockdown of Ipr1 downregulates Irgm1. siIpr1-1, siIpr1-2, and siNC were transfected into RAW264.7 cells with or without ISD treatment. The mRNAs were harvested after 36 h, and expression of Ipr1 and Irgm1 was examined by qRT-PCR. Values were normalized to the mean of the siNC ISD⁻ group (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (*, P < 0.05; **, P < 0.01). (B) Knockdown of Ipr1 downregulates Irgm1 at the protein level. siIpr1-1, siIpr1-2, and siNC were transfected into RAW264.7 cells with or without ISD, and the total protein was harvested after 48 h and detected with anti-SP110, anti-LRG47, and anti-GAPDH by Western blotting. (C) Knockdown of Ipr1 decreases transcriptional activity of the Irgm1 promoter. Irgm1 promoter reporter plasmids were cotransfected with siIpr1-1 or siIpr1-2 for 48 h, and the transcriptional activities were examined by luciferase assays. Values were normalized to the mean of the siNC-treated groups (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (D) Ipr1 enhances the transcriptional activity of the Irgm1 promoter. Irgm1 promoter reporter plasmids were cotransfected with pEGFP-C1 or pEGFP-Ipr1 into RAW264.7 and J774A.1 cells for 36 h, and then, the transcriptional activities were examined by luciferase assays. The Tnf-α promoter reporter plasmid was used as a positive control. The values of the pEGFP-Ipr1 groups were normalized to the mean of the pEGFP-C1 groups (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (E) Downregulation of Irgm1 and p21 after Ipr1 knockdown. RAW264.7 cells were transfected with siIpr1-1 and siNC for 36 h, and then the cells were treated with ISD for 8 h. The expression was determined by qRT-PCR. Values were normalized to the mean of those of the siNC (ISD⁻) groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (F) Proportion (percent) of G₀/G₁ cells (t test; **, P < 0.01).

In addition, we found that ISD treatment was able to decrease cell proliferation, and a previous study showed that activation of the p53-p21 pathway blocked the cell cycle in the G₀/G₁ phase (33). Therefore, RAW264.7 cells were detected by quantitative real-time PCR (qRT-PCR) and flow cytometry, and we found that ISD could induce p21 transcription and block the RAW264.7 cell cycle at the G₀/G₁ phase (Fig. 6E and F). Knockdown of Ipr1 decreased p21 and Irgm1 expression (Fig. 6E), suggesting that p21 and Irgm1 are regulated by similar mechanisms.

P53 upregulates Irgm1 by the p53 motif in the Irgm1 promoter.

To investigate the regulation of Irgm1 expression, the Irgm1 promoter region from positions −470 to +462 was analyzed using the JASPAR database, and p53 was predicted to interact with it. Knockdown of Ipr1 decreased the luciferase activity of p53-luc, which could be rescued by overexpression of Ipr1 (Fig. 7A). P53 is a transcription activator; once it is activated by stimulators, such as Act D and adriamycin (ADR), its target gene, p21, is activated. Both Act D and ADR treatments could significantly enhance Irgm1 transcription. Furthermore, knockdown of p53 by small interfering RNA (siRNA) or pifithrin-α (PFT-α) decreased Irgm1 (Fig. 7B and C). Overexpression of p53 in RAW264.7 cells increased Irgm1 transcription and its promoter activity (Fig. 7D and E). Mutation of the p53 motif in the Irgm1 promoter (Irgm1-P900-p53mut) caused a decrease of transcriptional activity compared to the wild type, regardless of Act D or ISD treatment (Fig. 7F and 8A). Transfection of the myc-p53 expression plasmid with Irgm1-P900-p53mut into RAW264.7 cells revealed that activity of p53 promoting transcription activity in Irgm1-P900-p53mut was lower than that in Irgm1-PRO-900 (Fig. 8B). RAW264.7 cells were transfected with siNC plus pCMV-p53, siNC plus pCMV-myc, siIpr1-1 plus pCMV-p53, and siIpr1-1 plus pCMV-myc and confirmed by Western blotting (Fig. 8C), which suggested that overexpression of p53 could rescue the decreased Irgm1 activity caused by Ipr1 knockdown (Fig. 8D).

FIG 7.

P53 plays a critical role in Ipr1-regulated Irgm1 expression. (A) Knockdown of Ipr1 impairs the luciferase activity of signal pathway reporter construct p53-luc. RAW264.7 cells were cotransfected with siIpr1-1, siIpr1-2, and siNC and the reporter plasmid p53-luc, and then, the cells were transfected with pcDNA3.1 or pcDNA3.1-Ipr1. After 36 h, the luciferase activity was determined by DLR. The values were normalized to the mean of the pcDNA3.1 siNC group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (B) Knockdown of p53 led to downregulation of Irgm1 and p21 transcription in RAW264.7 cells. RAW264.7 cells were transfected with sip53-1 or siNC for 48 h. Then, Act D (5 nM) was added for 8 h. Transcription of Irgm1 and p21 was determined by qRT-PCR. The values were normalized to the mean of those in the DMSO siNC groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA. (C) PFT-α inhibited the expression of p21 and Irgm1. The cells were treated with ADR (200 nM) for 24 h. Then, PFT-α (30 μM) was added for 8 h. Transcription of Irgm1 and p21 was determined by qRT-PCR. The values were normalized to the mean of those in the PBS DMSO groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (D) P53 increased Irgm1 promoter activity. RAW264.7 cells were cotransfected with pCMV-p53, pCMV-myc, and Irgm1-PRO-500, Irgm1-PRO-900, and Irgm1-PRO-1400 for 36 h. Promoter transcriptional activities were examined by luciferase assay. Values were normalized to the mean of the Irgm1-PRO-500 pCMV-myc group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (E) P53 induced Irgm1 and p21 expression. RAW264.7 cells were transfected with pCMV-myc and pCMV-p53 for 36 h. Gene expression was examined by qRT-PCR. The values of pCMV-p53 groups were normalized to the mean of those in the pCMV-myc groups (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (F) Mutation of the p53 element in the Irgm1 promoter region. The p53 motif was predicted in the Irgm1 promoter using the JASPAR database. Nucleotide mutation relied on the conservation of specific sites in the p53 motif.

FIG 8.

The p53 motif plays a critical role in Irgm1 transcriptional activation. (A) Mutation of the p53 motif impairs Irgm1 promoter transcription activity. RAW264.7 cells were transfected with the luciferase reporter plasmid Irgm1-PRO-900 and the mutated plasmid Irgm1-P900-p53mut. After 36 h of transfection, Act D and ISD were added to the culture medium for 8 h. Then, the transcriptional activity was determined by luciferase assay. Values were normalized to the mean of those in the Irgm1-PRO-900-DMSO group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (*, P < 0.05; **, P < 0.01). (B) The p53 motif plays an important role in the Irgm1 promoter. The gene expression plasmid pCMV-p53 was cotransfected with the reporter plasmids Irgm1-PRO-900 and Irgm1-P900-p53mut for 36 h. Promoter transcriptional activity was determined by luciferase assay. Values were normalized to the mean of those in the Irgm1-PRO-900-pCMV-myc group (+SD). The statistical significance of differences in relative luciferase activities was assessed by two-way ANOVA (**, P < 0.01). (C) pCMV-p53 or pCMV-myc was cotransfected with siNC or siIpr1-1 into RAW264.7 cells. After 48 h, cells were harvested and analyzed by Western blotting with anti-SP110 for IPR1 and anti-myc for myc-P53. Ipr1 siRNA, but not siNC, reduced the expression of Ipr1 successfully. (D) p53 rescued Irgm1 expression that was decreased by siIpr1-1. siNC, siIpr1-1, pCMV-myc, and pCMV-p53 were co-transfected into RAW264.7 cells for 36 h. The expression of Irgm1 was determined by qRT-PCR. Values were normalized to the mean of the siNC⁺ pCMV-myc group (+SD). The statistical significance of differences in relative mRNA expression was assessed by one-way ANOVA (**, P < 0.01).

Ipr1 enhances the interaction between Rpl11 and Mdm2 to release p53.

P53 is a transcriptional activator in p21 and Irgm1 expression. When P53 was knocked down by siRNA, IRGM1 protein was decreased whether the cells were treated by ISD or not (Fig. 9A). P53 protein was decreased when Ipr1 was knocked down in RAW264.7 cells (Fig. 9B). Using the Biological General Repository for Interaction Data Sets (BioGRID) database to combine high-throughput data sets with individual focused studies, 116 proteins that interacted with P53 and 87 proteins that interacted with MDM2 in mouse were characterized. Previously, 259 proteins were identified using coimmunoprecipitation (co-IP) and mass spectrometry targeting IPR1 (29). In this article, interactions among MDM2, P53 (TRP53), and IPR1 networks were drawn using the STRING database, and major ribosomal proteins were found to connect with MDM2, P53, and IPR1, including RPS10, RPL23, RPL11, and nucleophosmin 1 (NPM1) (Fig. 9C). Interactome clustering showed that RPL11 and NPM1 were involved in IPR1, P53, and MDM2 interactions (Fig. 9D). Cotransfection of pEGFP-Ipr1 with Flag-Rpl11, Flag-Npm1, or HA-Mdm2 into HEK293T cells was carried out to confirm the interaction between IPR1 and RPL11, IPR1 and NPM1, or IPR1 and MDM2 by Co-IP (Fig. 9E). As a nucleosome protein, H2B was used as an indicator of nuclear localization. In our research, mRuby-H2b was cotransfected with pEGFP-Ipr1 in order to confirm the distribution of IPR1. Fluorescence imaging in living cells of IPR1 sublocalization in RAW264.7 cells with RPL11, NPM1, and MDM2 revealed that interactions of IPR1 with MDM2, RPL11, and NPM1 occurred in the nucleus (Fig. 9F).

FIG 9.

IPR1 interacted with RPL11, MDM2, and NPM1. (A) Knockdown of p53 downregulates Irgm1 in protein. sip53-1, sip53-2, and siNC were transfected into RAW264.7 cells with or without ISD. Cells were harvested after 72 h and detected by anti-P53, anti-LRG47, and anti-GAPDH via Western blotting. (B) Knockdown of Ipr1 downregulates p53 in protein. siIpr1-1, siIpr1-2, and siNC were transfected into RAW264.7 cells with or without ISD. Cells were harvested after 72 h and detected by anti-P53, anti-SP110, and anti-GAPDH via Western blotting. (C) Protein-protein interaction network among IPR1, P53, and MDM2. P53 and MDM2 interactomes were obtained from BioGRID; the IPR1 interactome was determined in previous research. The interaction network was established with STRING. (D) RPL11 and NPM1 were common proteins in the P53 (Trp53), MDM2, and IPR1 interactome. The interactomes were clustered using Venny 2.1.0. (E) IPR1 interacted with RPL11, NPM1, and MDM2. EGFP-Ipr1 was coexpressed with Flag-Npm1, Flag-Rpl11, or HA-Mdm2 in HEK293T cells for 48 h. Anti-Flag or anti-HA was used for immunoprecipitation and anti-GFP for Western blotting (IB). (F) Live imaging of RAW264.7 cells expressing pEGFP-Ipr1 with mRuby, mRuby-H2b, mRuby-Mdm2, mRuby-Rpl11, or mRuby-Npm1. Scale bar, 10 μm.

As nucleolar components, RPL11 and NPM1 are necessary intermediary elements between IPR1 and P53; therefore, the relationship of NPM1 and RPL11 with P53 and IPR1, as well as MDM2, was investigated. Both RPL11 and NPM1 have been reported to regulate P53 activity; however, knockdown of RPL11, but not NPM1, has been shown to downregulate P53 at the protein level (34). To investigate whether IPR1 influences the interaction between RPL11 and MDM2, Flag-Rpl11 and HA-Mdm2 were cotransfected into HEK293T cells with pEGFP-C1 (Fig. 10A, lanes i) or pEGFP-Ipr1 (Fig. 10A, lanes ii) respectively. Co-IP experiments showed that the enrichment of MDM2 by RPL11 was significantly increased by IPR1 incorporation (Fig. 10A). Moreover, to confirm the function of IPR1 in mediating interaction among RPL11, MDM2, and P53, Flag-p53 and HA-Mdm2 were cotransfected into HEK293T cells with pEGFP-C1 (Fig. 10B, lanes i) or pEGFP-Ipr1 (Fig. 10B, lanes ii), respectively. By adding enhanced green fluorescent protein (EGFP)-IPR1, the interaction of P53 and MDM2 was reduced slightly (Fig. 10B). However, when Flag-Rpl11 was added, P53 release from MDM2 was significantly enhanced, mediated by EGFP-IPR1 (Fig. 10C). Knockdown of RPL11 decreased the enrichment of P53 with or without ISD treatment (Fig. 10D). To further investigate the effect of ISD on MDM2/IPR1/RPL11, RAW264.7 cells were treated with ISD (Fig. 10E, lanes ii) or not (Fig. 10E, lanes i). Co-IP experiments showed that the interaction between MDM2 and RPL11/IPR1 was enhanced by ISD (Fig. 10E). MDM2 is an E3 ubiquitin ligase that results in the ubiquitination of P53. Flag-p53, Flag-Rpl11, and HA-Mdm2 were cotransfected into HEK293T cells with pEGFP-C1 or pEGFP-Ipr1, and the cells were treated with 50 μM MG132 for 4 h before harvest. The ubiquitination of P53 was significantly reduced and the protein level of P53 was increased by adding EGFP-IPR1 (Fig. 10F).

FIG 10.

IPR1 promotes P53 release by increasing interaction between RPL11 and MDM2, and the model of Ipr1 functions in signal transduction from ISD to Irgm1. (A) IPR1 enhanced interaction between RPL11 and MDM2. Flag-Rpl11 and HA-Mdm2 were transiently coexpressed in HEK293T cells with pEGFP-Ipr1 (lanes ii) or pEGFP-C1 (lanes i) for 48 h, and interactions were detected using anti-Flag for co-IP and anti-HA and anti-GFP for Western blotting. (B) IPR1 slightly decreased interaction between P53 and MDM2. Flag-p53 and HA-Mdm2 were transiently coexpressed in HEK293T cells with pEGFP-Ipr1 (lanes ii) or pEGFP-C1 (lanes i) for 48 h, and interactions were detected using anti-HA for co-IP and anti-Flag and anti-GFP for Western blotting. (C) IPR1 increased interaction between RPL11 and MDM2 and decreased interaction between P53 and MDM2 significantly. Flag-p53, Flag-Rpl11, and HA-Mdm2 were transiently coexpressed in HEK293T cells with pEGFP-Ipr1 (lanes ii) or pEGFP-C1 (lanes i) for 48 h, and interactions were detected using anti-HA for co-IP and anti-Flag and anti-GFP for Western blotting. (D) Knockdown of Rpl11 downregulates p53. siRpl11-1, siRpl11-2, and siNC were transfected into RAW264.7 cells with or without ISD treatment. Cells were harvested after 72 h and detected by anti-P53, anti-Rpl11, and anti-GAPDH by Western blotting. (E) ISD enhanced interaction between RPL11/IPR1 and MDM2. RAW264.7 cells were treated with ISD (lanes ii) or not (lanes i) for 60 h. Anti-MDM2 antibody was used for immunoprecipitation, and anti-Rpl11 and anti-SP110 were used for Western blotting. (F) Ubiquitination status of Flag-P53. Flag-p53, Flag-Rpl11, and HA-Mdm2 were cotransfected into HEK293T cells with pEGFP-Ipr1 or pEGFP-C1 for 60 h; 4 h prior to harvesting, cells were treated or not with MG132 (50 μM). Lysates were subjected to IP using anti-Flag. Bound proteins were analyzed by Western blotting using anti-Flag, anti-HA, anti-GFP, or anti-GAPDH. (G) Effects of ISD and Act D on THP-1 cells. THP-1 cells were treated with ISD or Act D for 16 h. Transcription of SP110, IRGM, and p21 was determined by qRT-PCR. The values were normalized to the mean of those in the ISD⁻ or DMSO group (+SD). The statistical significance of differences in relative mRNA expression was assessed by two-way ANOVA (**, P < 0.01). (H) Effect of ISD on THP-1 cells. THP-1 cells were treated with ISD for 24 h. The proteins were determined by Western blotting using anti-SP110, anti-LRG47, and anti-GAPDH. (I) Model of Ipr1 functions in signal transduction from ISD to Irgm1. First, ISD is transfected into cells and arrested by cGAS. Then, the signals are transferred through cGAS-STING-TBK1 to IRF3, resulting in Ipr1 transcriptional activation. Afterward, IPR1 increases the release of P53 by promoting the interaction of RPL11 and MDM2. The released P53 regulates Irgm1 expression through the p53 motif in the Irgm1 promoter.

DISCUSSION

Ipr1 is located in the sst1 locus and participates in host M. tuberculosis resistance. Pan et al. found that expression of the Ipr1 gene in M. tuberculosis-susceptible C3Heb/FeJ mice showed significant limits to bacterial replication in lungs (10). Wu et al. cloned cows, constantly expressing mouse Ipr1, that showed much higher M. tuberculosis resistance in their lungs, as well (35). These results suggest that the abundance of Ipr1 may have effects on the innate immunity of the host. However, the details of Ipr1’s functions and its precise transcriptional regulation in innate immunity have not been discovered. During intracellular mycobacterial infection, IFN-γ, a stimulator of Ipr1, activates macrophages to promote innate immunity. Previous research demonstrated that IFN-γ activates Ipr1 through its promoter and identified the core region of the Ipr1 promoter. In addition to IFN-γ, M. tuberculosis, HBV, influenza virus H1N1, and Sendai virus can induce Ipr1 expression, as well (Fig. 1). At the time of initiation of infection by virus and intracellular bacteria (such as M. tuberculosis), pathogenic DNA is released into the cytoplasm, in which it is arrested by intracellular receptors, resulting in various intracellular defenses. Furthermore, Ipr1 could be upregulated by small molecules, such as etoposide and cytarabine, which could activate the cGAS-IRF3 pathway (Fig. 1F and G) (36).

The cGAS-STING-IRF3 pathway is a component of the innate immune system that responds to both exogenous and endogenous double-stranded DNA (dsDNA) by initiating autophagy (36–38). When it detects dsDNA, cGAS forms homodimers, which catalyze the production of 2′,3′-cGAMP from ATP and GTP. cGAMP then binds to STING, acting as a second messenger triggering activation of IRF3, which leads to transcription initiation of IFN-stimulated genes. We found that Ipr1 can be induced by ISD, an activator of the cGAS-STING-IRF3 axis, in BMDMs and NIH 3T3, J774A.1, and RAW264.7 cell lines, and Ipr1 mRNA increased the most in macrophages (RAW264.7 and J774A.1 cells) (Fig. 2A). Previously, we confirmed the core region of the Ipr1 promoter (chr1:85,598,673 to 85,599,270) and found it was highly homologous between mice and humans (Fig. 2E), which suggests that similar regulatory factors may be shared in human and mouse. Cluster analysis of transcription factors of the Ipr1 promoter in various species showed a shared ISRE motif in the promoters, which suggests that IRFs play critical and conserved roles in regulating the expression of Ipr1 in various species. It was confirmed that Ipr1 is regulated by the cGAS-IRF3 axis via ISD, because IRF3 could bind the ISRE motif in the Ipr1 promoter. Ipr1-PRO-594 is highly activated in RAW264.7 cells because it contains a core element (the ISRE motif) of the Ipr1 promoter and is able to respond to the endogenous transcription activators, while mutated ISRE of Ipr1-P594-ISREmut failed to be recognized and performed its function of transcriptional activation. In addition, the cGAS-IRF3 axis is required for IFN-β expression, and IFN-β was found to upregulate Ipr1 transcription, as well. IFN-β signaling is transmitted through the JAK/STAT pathway to stimulate gene expression (39). In a previous study, we determined that STAT1 participates in the activation of Ipr1 expression (8). These results indicate that during viral or bacterial infection, Ipr1 is directly and rapidly activated by the cGAS-IRF3 pathway and upregulated by IFN-β (via Stat1) to potentiate the immune reaction.

We and others have previously demonstrated the critical functions of Ipr1 in initiating apoptotic pathways in M. tuberculosis-infected macrophages (8, 10). In addition to apoptosis, autophagy serves as another self-protection mechanism against viral or bacterial infection (40), and the cGAS-IRF3 pathway is critical for autophagy modulation (6). After recognition of cytoplasmic DNA, such as ISD, cGAS was activated, and on one hand directly interacted with Beclin-1 autophagy protein to mediate autophagy and on the other hand promoted signal transduction from STING to IRF3, a transcription factor. As a target gene of the cGAS-IRF3 pathway, the current studies determined that Ipr1 participates in transcriptional regulation of Irgm1, an autophagy-related gene. To identify whether IPR1 regulates Irgm1 transcription, the promoter region of Irgm1 was cloned, and DLR assay results demonstrated that Ipr1 can excite luciferase activity of Irgm1-PRO-900. Additionally, analysis of the Irgm1 promoter revealed the presence of a p53 motif. P53 is a regulatory transcription factor that responds to different cellular stresses and manages multiple cell processes, including cell cycle arrest and senescence. Recently, accumulating evidence has strongly indicated that p53 plays broader roles in either innate or adaptive immune responses, such as apoptosis and autophagy (41–43). P53 directly activates expression of immune-responsive genes, including CC-chemokine ligand 2 (CCL-2), IRF5, RNA-activated protein kinase, and Toll-like receptor 3, some of which are indispensable to the initiation of autophagy (44–46). Irgm1 encodes a member of a family of immunity-related GTPases that functions as a direct effector inducing autophagy in cell-autonomous resistance against intracellular pathogens (47). Act D is a chemotherapeutic known to activate p53 signaling (48, 49). Here, Act D was found to induce Irgm1 expression through the p53 motif in the Irgm1 promoter. Because both Ipr1 and p53 are able to activate Irgm1 transcription, it is plausible that they cooperate to do so. Knockdown of Ipr1 led to a decrease in P53, suggesting that Ipr1 may regulate gene expression by relying on P53. Under normal conditions, P53 is rapidly degraded by the MDM2-modulated ubiquitin proteasome pathway (50, 51). When cells are subjected to stress that inhibits the function of RNA polymerase I, many ribosomal proteins (RPL11, RPL5, RPL23, RPS7, and RPS3) relocate from the nucleolus to the nucleoplasm, triggering interaction with MDM2 and resulting in P53 stabilization (52). To investigate the relationship between IPR1 and P53, we focused on the IPR1, P53, and MDM2 interactome and found NPM1 and RPL11 to be mutual proteins (i.e., both found in the IPR1, P53, and MDM2 interactomes). RPL11 and NPM1 are mainly located in the nucleolus organizer region of the nucleus, while IPR1 is mainly located in the nucleoplasm. However, the interaction between IPR1 and NPM1, and RPL11 and MDM2, was validated by co-IP, suggesting IPR1 can bind such partners in the nucleus and modulate their functions. These results are in step with Rps relocation when cells are under stress. Both Npm1 and Rpl11 have been reported to regulate the function of p53 (53, 54). However, knockdown of RPL11, but not NPM1, resulted in downregulation of P53 (34). Upon ribosomal stress, RPL11 forms a stable complex with MDM2 through direct contact with its zinc finger and inhibits MDM2-mediated P53 ubiquitination (55, 56). IPR1 can enhance the formation of the RPL11-MDM2 complex and reduce the interaction between MDM2 and P53 (Fig. 10C), resulting in P53 release. Therefore, the present results indicate that IPR1 influences P53 stabilization by interacting with RPL11 and MDM2, thereby leading to regulation of Irgm1 expression in mouse macrophages. Irgm1 is critical to the formation of large vacuolar compartments, and increased LC3B-II could be detected when IRGM1 was increased. Previously, researchers found that Irgm1 is required for the autophagic pathway regulated by IFN-γ (14). Like IFN-γ, ISD treatment could induce cell autophagy, as well as Irgm1 expression, which suggests that Irgm1 takes part in cGAS-mediated autophagy. In this study, we found that Ipr1 participated in inducing Irgm1 expression when RAW264.7 cells were treated with ISD. In particular, knockdown of Ipr1 decreased Irgm1 transcription, whether ISD was added or not. These results suggest that Ipr1 may participate in autophagy regulation.

Considering the high similarity of promoter regions between Ipr1 and SP110b, we speculated that ISD can induce SP110 expression and can have effects on induction of IRGM, as well. However, ISD failed to induce IRGM expression, while SP110 was upregulated in THP-1 cells (Fig. 10G and H), which suggests different roles of Ipr1 and SP110 in humans and mice. This phenomenon may be caused by several factors. First, low homology of Irgm1 and IRGM in the promoter or gene body sequence may cause diversity in DNA sequence-specific transcriptional regulation (57). Besides this, previous studies have defined several kinds of isoforms of SP110 proteins, which could possess different functions (9).

In summary, this study provides novel evidence regarding the role of Ipr1 in regulation of cellular signal transmission (Fig. 10I). We identified an unrevealed function of Ipr1, a target gene of the cGAS-IRF3 axis, working as an intermediary between ISD-cGAS-IRF3 and Irgm1 transcriptional regulation. Ipr1/SP110 is upregulated by M. tuberculosis or viral infection (Fig. 1) and translates stress signals into Irgm1 expression. By interacting with RPL11 and MDM2, IPR1 promotes P53 release and participates in ISD-induced Irgm1 expression. The current findings define a previously undocumented function of IPR1 in immunity-related gene regulation. As Ipr1 is modulated by cGAS-IRF3, it not only serves as an M. tuberculosis resistance gene but also may play important roles in virus infection. The present study provides novel mechanistic insights into cellular immune responses and reveals potential new molecular targets for developing host-directed therapy for M. tuberculosis or viral infection.

MATERIALS AND METHODS

Cell culture and transfection.

RAW264.7, NIH 3T3, and HEK293T cells obtained from the American Type Culture Collection were maintained in Dulbecco’s modified Eagle’s medium (Gibco, USA) containing 10% fetal bovine serum (BI, Israel). All the cells were maintained at 37°C and 5% CO₂ in a humidified incubator (Thermo Fisher). Lipofectamine 2000 (Life, USA) was used for cellular transfection, using a plasmid-to-transfection-reagent ratio of 1:4 and an ISD-to-transfection-reagent ratio of 1:2. An equal volume of sterile water was added to the transfection system, which was used as the negative control (ISD⁻). BMDMs were isolated from Mus musculus (C57BL/6; 2 male, 2 female; 6 months old) according to a cell isolation protocol from Bing Ren laboratory's ENCODE Project (https://www.encodeproject.org/experiments/ENCSR000CFG/). Prior approval was obtained from the Institutional Animal Care and Use Committee (IACUC). The sequences of siRNAs were as follows: siIpr1-1, 5′-GAAGGGCGUUAUACGUUGUTT-3′; siIpr1-2, 5′-CCAAAGUGGUGCACAAUAUTT-3′ (GenePharma, Shanghai, China). siRpl11-1 (product no. siB1912020137181627), SiRpl11-2 (product no. siB1912020127563204), sip53-1 (product no. siB111118152519), and sip53-2 (product no. siB111118152510) were from Ribobio, Guangzhou, China. The sequence of siNC is 5′-UUCUCCGAACGUGUCACGUTT-3′ (GenePharma, Shanghai, China). IL-13, TNF-α, and IFN-β were reconstituted in sterile phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin. Act D was reconstituted in dimethyl sulfoxide (DMSO).

Reverse transcription and qRT-PCR.

Total RNA was isolated using TransZol reagent (TransGen, China) and transcribed into cDNA using EasyScript One-Step RT-PCR SuperMix (TransGen). qRT-PCR amplification of Ipr1, Irgm1, and p21 was performed using TransStart Green qPCR SuperMix (TransGen) on a Step One Plus II system (ABI, USA) with β-actin as a normalization control. The cycling conditions were 5 min at 94°C for initial denaturation, with amplification performed with 40 cycles of 94°C for 5 s and 60°C for 30 s. The melting curve was 65 to 95°C at a rate of 0.5°C/cycle. Three independent experiments were performed for all genes. Details of the primer sequences used are shown in Tables 2 and 3.

TABLE 2.

qRT-PCR primers of mouse used in this study

Primer name	Sequence (5′–3′)
qGapdh-F	GTGTTCCTACCCCCAATGTGT
qGapdh-R	ATTGTCATACCAGGAAATGAGCTT
qIpr1-F	GAAAAACAGACCCCCACTGA
qIpr1-R	TGAGGACAGATGACTTGGCA
qp21-F	GCAGATCCACAGCGATATCC
qp21-R	CAACTGCTCACTGTCCACGG
qIFN-β-F	CAGCTCCAAGAAAGGACGAAC
qIFN-β-R	GGCAGTGTAACTCTTCTGCAT
qIrgm1-F	CGGCTTCCTCAGAGACCCTAA
qIrgm1-R	GCTCTTTCGGTGCTCCTACT
qTGF-β-F	CTCCCGTGGCTTCTAGTGC
qTGF-β-R	GCCTTAGTTTGGACAGGATCTG
qNfkbia-F	AATGGTGAAGGAGCTGCGG
qNfkbia-F	TGATCACAGCCAAGTGGAGT

TABLE 3.

qRT-PCR primers of human used in this study

Primer name	Sequence (5′–3′)
Sp110-F	CCTGCACTCATCCAGGAAGG
Sp110-R	GGGGATCAGGTTGTCACTGG
IRGM-F	GGAAGCCATGAATGTTGA
IRGM-R	TAGGAGGCACATCTTTGG
P21-F	CGATGGAACTTCGACTTTGTCA
P21-R	GCACAAGGGTACAAGACAGTG
Gapdh-F	AATGAAGGGGTCATTGATGG
Gapdh-R	AAGGTGAAGGTCGGAGTCAA

DLR assay.

RAW264.7 and J774A.1 cells were cotransfected with promoter reporter plasmids and pGL4.73 (10:1; Promega) using Lipofectamine 2000 (Life, USA). Cells were incubated with lysis buffer for 10 min, with a shaker rate of 180 rpm to obtain protein lysates, which were analyzed using a DLR assay system (TransGen, China) according to the manufacturer’s instructions. Firefly luciferase activity in lysates was assessed on a Victor X5 multilabel plate reader (PerkinElmer, USA) and normalized to the expression of Renilla luciferase. The mouse Ipr1 core promoter region (cloned into the Ipr1-PRO-594 reporter vector) that we have determined is located in chr1:85,598,673 to 85,599,270 (mm10). The homologous regions of the mouse Ipr1 promoter in other species were predicted using Lift genome annotations of the UCSC database.

Western blotting.

Following treatment, cells were harvested using trypsin digestion and lysed in ice-cold immunoprecipitation (IP) assay buffer (50 mM Tris [pH 7.4], 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate) with a protease inhibitor cocktail (Roche, Switzerland). Equal amounts of proteins were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (0.22 μm; Millipore, Germany) using a Mini-Protean Tetra Cell system (Bio-Rad, USA). The membranes were incubated with one or more of the following primary antibodies: monoclonal rabbit anti-myc antibody (1:500; Beyotime; AM933), mouse monoclonal anti-Flag M2 (1:1,000; F1804; Sigma, USA), mouse monoclonal anti-LRG47 (1:100; sc-517338; Santa Cruz, USA), mouse monoclonal anti-green fluorescent protein (anti-GFP) (1:1,000; 66002-1-Ig; Proteintech, USA), rabbit polyclonal anti-Sp110 (M-190) (1:200; sc-98365; Santa Cruz, USA), and mouse polyclonal anti-GAPDH (anti-glyceraldehyde-3-phosphate dehydrogenase) (1:1,000; A01622; TransGen, China). Corresponding horseradish peroxidase-linked secondary-antibody detection and visualization were completed using an ECL detection kit (P0018A; Beyotime) according to the manufacturer’s instructions.

Co-IP.

HEK293T cells were transfected with appropriate expression plasmids for at least 36 h and then harvested using trypsin digestion. Whole-cell protein extracts were obtained by incubating cells in Pierce IP lysis buffer (25 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% NP-40, 1 mM EDTA, 5% glycerol, complete protease inhibitor cocktail; Merck, Germany) for 30 min on ice and homogenizing them by pipette every 5 min. The protein extracts were then incubated for 16 h at 4°C with mouse monoclonal anti-Flag M2 (Sigma) or control anti-IgG (Beyotime) antibody, and the precipitated proteins were captured on Pierce protein A/G-agarose beads (Life, USA) for 2h at 4°C. Beads bound to antibodies interacting with the appropriate proteins were rotated and washed three times with IP wash buffer. Then, SDS loading buffer (250 mM Tris-HCl, 10% SDS, 0.5% bromophenol blue, 50% glycerol, and 5% β-mercaptoethanol; Sigma) was added to the precipitation mixture and boiled for 10 min before being separated by SDS-PAGE and identified by Western blotting as described above.

Oligonucleotide annealing.

ISD and EMSA probes (Sangon, China) were prepared from equimolar amounts of sense and antisense DNA oligonucleotides (100 μM). The oligonucleotides were first heated at 95°C for 5 min and cooled to melting temperature by 1°C/min. Then, the oligonucleotides were incubated at melting temperature for 30 min before cooling to room temperature. The sense strand sequences are shown in Table 4. All of the oligonucleotides were synthesized by Sangon Biotech (China). The oligonucleotide annealing was performed on a T100 thermal cycler.

TABLE 4.

DNA sequences used for cell treatment or EMSA

Name	Orientation	Sequence (5′–3′)
ISD	Sense	TACAGATCTACTAGTGATCTATGACTGATCTGTACATGATCTACA
	Antisense	TGTAGATCATGTACAGATCAGTCATAGATCACTAGTAGATCTGTA
Ipr1-biotin	Sense	Biotin-TGCTCAGTGTGCTGAGATCACTTTCATTTTTCTTTTCTTGAAGCCTGACT
	Antisense	Biotin-AGTCAGGCTTCAAGAAAAGAAAAATGAAAGTGATCTCAGCACACTGAGCA
Ipr1-mut-biotin	Sense	Biotin-TGCTCAGTGTGCTGAGATCACCCTCACCTTTCTTTTCTTGAAGCCTGACT
	Antisense	Biotin-AGTCAGGCTTCAAGAAAAGAAAGGTGAGGGTGATCTCAGCACACTGAGCA
Ipr1-control	Sense	TGCTCAGTGTGCTGAGATCACTTTCATTTTTCTTTTCTTGAAGCCTGACT
	Antisense	AGTCAGGCTTCAAGAAAAGAAAAATGAAAGTGATCTCAGCACACTGAGCA

EMSA.

EMSAs were performed using a Chemiluminescent EMSA kit (Beyotime) according to the manufacturer’s instructions. Oligonucleotides were biotinylated on the 5′ side during chemical synthesis. pCMV-IRF3 expression plasmids were transfected into HEK293T cells over 48 h, and then, cells were harvested by pipette and centrifuged at 2,000 rpm for 5 min. The cells were lysed with Pierce IP lysis buffer on ice for 30 min. DNA probes (100 fmol) and cell lysates were incubated with binding buffer at room temperature for 20 min before adding loading buffer without bromophenol blue (GS009-2; Beyotime, China). The interaction between the probe and transcription factors was examined with PAGE, followed by Western blotting as described above. Shifted DNA fragments on nylon membranes (0.45 μm; Millipore, USA) were detected using horseradish peroxidase-linked streptomycin.

ChIP.

ChIP assays were performed with an enzymatic chromatin IP kit (no. 9003; CST, USA) according to the manufacturer’s instructions. Briefly, RAW264.7 cells transfected with Flag-IRF3 over 36 h were cross-linked with 1% formaldehyde for 10 min; glycine (final concentration, 0.125 M) was added to terminate the cross-linking. Nuclei were isolated, subjected to micrococcal nuclease (MNase) digestion as described above, and then sonicated to disrupt nuclear membranes and isolate DNA; the DNA fragments were approximately 500 bp in length. Immune complexes were formed by adding 10 μg of mouse anti-Flag M2 antibody, diluting in ChIP buffer, and incubating overnight at 4°C. The next day, samples were immunoprecipitated with protein A/G magnetic particles. Unbound protein was removed by washing with ChIP wash buffer. Protein-DNA cross-links were reversed for 2 h with proteinase K and 5 M NaCl at 65°C. The precipitated DNA was purified utilizing a PCR product purification kit (28106; Qiagen, USA) according to the manufacturer’s instructions. The precipitated DNA fragments were amplified by qRT-PCR (qRT-PCR ChIP primers: F, TGCTAACCACACTGGGGAAC; R, CGAGGTTATGCTGCCTTGGA; chr1:85598741 to 85598899).

In silico prediction and statistical analysis.

Transcription factors were predicted using the JASPAR database and according to the promoter sequences. Protein interaction networks were constructed using the BioGRID and STRING databases. The data for H3K4me3 and H3K27ac enrichment were obtained from the WashU EpiGenome Browser (https://epigenomegateway.wustl.edu/). The BMDMs used for ChIP sequencing (ChIP-seq) were isolated from 8-week-old C57BL/6 mice. The standard input signal (accession no. ENCFF001JYW [see https://www.encodeproject.org/]) was used as a control for most experiments (accession no. ENCSR000CFD, target, H3K27ac; accession no. ENCSR000CFF, target, H3K4me3). The GEO data sets used for the control, H3K27ac, and H3K4me3 are GSM918737, GSM1000074, and GSM1000065. The spleens and thymuses from 8-week-old C57BL/6 mice were used for ChIP-seq targets H3K27ac and H3K4me3 (accession no. ENCSR000CCH, target, thymus H3K27ac; accession no. ENCSR000CDJ, target, spleen H3K27ac; accession no. ENCSR000CCJ, target, thymus H3K4me3; accession no. ENCSR000CBD, target, spleen H3K4me3).

The results were from at least three independent experiments. The data are reported as means and standard deviations (SD) and were analyzed using analysis of variance (ANOVA) unless otherwise indicated; a P value of <0.05 or <0.01 was considered statistically significant.