Mitochondrial transcription factor B2 overexpression increases M2 macrophage infiltration via cytosolic mitochondrial DNA-stimulated Interleukin-6 secretion in ovarian cancer.

Mitochondrial transcription factor B2 (TFB2M) is a protein modulating both mitochondrial DNA (mtDNA) transcription and compacting. In this study, we explored the expression profile of TFB2M in ovarian cancer, its association with infiltration of tumor-associated macrophages (TAMs), and its influence on macrophage polarization. Serial sections of ovarian cancer tissue arrays were stained to detect TFB2M and CD163 expression. Epithelial ovarian cancer cell line OVISE and CAOV4 were used to assess the influence of TFB2M on IL-6 expression. THP-1 cells were utilized as an in vitro model for macrophage migration and polarization. Results showed that higher TFB2M expression is associated with poor survival in ovarian cancer patients. IHC staining confirmed a moderately positive correlation between TFB2M expression and the infiltration of CD163-positive cells in 68 primary ovarian cancer cases. TFB2M overexpression was associated with increased mtDNA outside the mitochondria and elevated IL-6 expression in ovarian cancer cells. When cytosolic mtDNA was selectively inhibited by DNase I, TFB2M-induced IL-6 upregulation was canceled. TFB2M overexpression could activate the nuclear factor kappa-B (NF-κB) signaling pathway via promoting nucleus entry of p65 and p-p65, which was abrogated by inhibiting cytosolic mtDNA, TLR9, or NF-κB signaling pathway. Conditioned medium from OIVSE cells with TFB2M overexpression could induce macrophage migration and M2 polarization. However, these inducing effects were abrogated by DNase I, TLR9 inhibitor, and anti-IL-6 R pretreatment. In conclusion, this study showed a novel role of TFB2M in the immunosuppressive tumor microenvironment. It promotes M2 macrophage infiltration via a cytosolic mtDNA/TLR9/NF-κB/IL-6 pathway in ovarian cancer.

Highlights

TFB2M expression is positively correlated with TAM infiltration in ovarian tumor.

TFB2M expression is positively correlated with cytosolic mtDNA stress.

Cytosolic mtDNA stress enhances the secretion of IL-6 via NF-κB signaling pathway.

Cytosolic mtDNA stress enhances macrophage infiltration and M2 polarization.

Introduction

Ovarian cancer is one of the most common malignancies in women with high mortality [1]. Approximately 90% of primary ovarian tumors are of epithelial origin [2]. Generally, for cases suitable for surgical resection, the current standardized treatment is optimal cytoreductive surgery following systemic chemotherapy [3]. Although debulking surgery combined with first-line chemotherapy might achieve complete remission, recurrence occurs in 70–80% of patients within 2–5 years, with inevitable chemoresistance [4]. Immune checkpoint inhibitors (ICIs) have been considered as a potential strategy to overcome therapeutic resistance. However, available clinical trial data suggest that the ICIs targeting programmed cell death protein 1 (PD-1), programmed death-ligand 1 (PD-L1) and cytotoxic T lymphocyte-associated antigen-4 (CTLA-4) could not improve survival in ovarian cancer [5], which is quite different from the situations in some other solid tumors, such as lung cancer, melanoma, and renal cell carcinoma [6].

The cross-talk between cancer cells and tumor microenvironment (TME) plays a critical role in the immunotherapeutic responses of tumor cells [7-9]. Generally, the cellular components of TME include tumor cells, stromal cells, fibroblasts, endothelial cells, and immune cells [10]. Similar to other solid tumors, the development of ovarian cancer is associated with alterations in the surrounding stroma [11]. Tumor cells can sculpt the surrounding microenvironment via secreting a series of cytokines, chemokines, and other factors [10]. These molecules influence the infiltration of immune cells and alter their biological behaviors. The infiltrated immune cells can suppress or promote tumor progression, depending on their compositions and phenotypic states [12]. Among the immune cells, monocyte-derived macrophages can polarize into inflammatory M1 or immune-suppressive M2 macrophages, depending on the inducers in the TME [13].

Mitochondria are double-membrane organelles with their own circular and high-copy-number DNA, encoding essential proteins to maintain normal mitochondrial functions. Besides the role of an energy provider, they regulate a series of fundamental cellular processes, such as calcium homeostasis, lipid metabolism, and apoptosis [14,15]. Mitochondrial DNA (mtDNA) is transcribed by mtDNA-specific transcriptional complex, which includes mitochondrial transcription factors A (TFAM), mitochondrial transcription factor B1 (TFB1M) or TFB2M, and a mitochondrial RNA polymerase (POLRMT) [16].

In addition to the involvement in transcription, TFAM and TFB2M also participate in compacting mtDNA into nucleoids and controlling mtDNA copy numbers [17-19]. Knockdown of TFAM or TFB2M can reduce mtDNA content, while enforced TFAM or TFB2M overexpression cancer increase mtDNA number [19,20]. This alteration might support rapid mitochondrial fission. However, mitochondrial fission is usually associated with the cytosolic release of mtDNA, which causes cytosolic mtDNA stress [21,22]. Cytosolic mtDNA escaped from mitochondria include multiple forms, including nucleoids, circular double-stranded molecules, linear single-stranded, and small fragments [23]. It can bind to certain sensors (such as Toll-Like Receptor 9, TLR9 and Z-DNA binding protein 1, ZBP1) in the cytoplasm, triggering the signaling pathways related to cytokines production [22,24].

In this study, we aimed to explore the expression profile of TFB2M in ovarian cancer, its association with the infiltration of tumor-associated macrophages (TAMs) and its regulatory effects on macrophage polarization.

Materials and methods

Bioinformatic analysis

Gene expression data were extracted from The Genotype-Tissue Expression (GTEx) and The Cancer Genome Atlas (TCGA) projects, using the UCSC Xena Browser [25]. Kaplan-Meier survival analysis was conducted using data from the Kaplan-Meier Plotter (http://kmplot.com/analysis/) [26], using the best performing threshold as the gene expression cutoff. Immunohistochemistry (IHC) staining of TFB2M in the fallopian tube and normal ovary was checked using data in the Human Protein Atlas (HPA) [27] (https://www.proteinatlas.org/).

IHC staining and scoring

Commercial ovarian cancer tissue arrays (serial sections) were purchased from Taibosi Bio (Xian, Shanxi, China). IHC staining was performed using a BOND-III Fully Automated IHC and ISH Staining System (Leica Biosystems, Nussloch, Germany) and commercially available antibodies (anti-TFB2M, 1: 200, AP10145b, ABCEPTA, Suzhou, China, anti-CD163, 1: 200, 16,646-1-AP, Proteintech, Wuhan, China).

IHC staining score is a combination of staining extensiveness and intensity, following the method introduced previously [28], with small modifications. The following scores were defined for staining extensiveness [28]: 0 = < 5%, 1 = 6% – 25%, 2 = 26% – 50%, 3 = 51% – 75%, and 4 = 76% – 100%. The following intensity scores were defined: 0 = absent, 1 = weak, 2 = moderate, and 3 = strong. The combined scores were grouped as the following + (0–1), ++ (2–3), +++ (3–4), ++++ (5–7).

The infiltration of TAMs was estimated following the method introduced previously [29]. In brief, CD163 positive macrophages (considered as TAMs) were counted. The ratio of TAMs/total number of visible cells with a monocyte/macrophage morphology with oval or rounded nuclei was calculated. For CD163 staining, only the staining extensiveness was assessed, following the criteria mentioned above. Two pathologists without authorship in this study examined these sections independently in a blinded manner, without prior knowledge of the tissue array data.

Cell culture and treatment

Epithelial ovarian cancer cell lines OVISE (TP53 wildtype) and CAOV4 (TP53 mutant) and a pro-monocytic cell-line THP-1 were purchased from Otwo Biotech (Shenzhen, China). OVISE cells were cultured in RPMI1640 medium with 10% fetal bovine serum (FBS), while CAOV4 cells were cultured in Dulbecco’s modified Eagles medium (DMEM) with 10% FBS at 37°C in a 5% CO2 atmosphere in a humidified incubator, as described previously. THP-1 cells were cultured as recommended previously [30]. Both culture mediums were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin.

Lentiviral TFB2M (NM_022366) expression particles were generated based on pHBLV-CMVIE-IRES-puro backbone, while lentiviral TFB2M shRNA particles were generated based on pHBLV-U6-puro, by Hanbio (Shanghai, China). The following shRNA sequences were used: #1, 5'-CCCAAAGCGTAGGGAATTATT-3'; #2, 5'-CCCAAAGCGTAGGGAATTATT-3'. Lentivirus was produced following the protocols recommended by the provider. In brief, lentiviral particles were co-transfected with psPAX2 packaging plasmid and pMD2.G envelope plasmid into HEK293T cells using Lipofectamine 3000 (Thermo Fisher Scientific, Carlsbad, CA, USA). Lentivirus containing supernatant was harvested 48 h later. Cells were infected at a multiplicity of infection (MOI) of 10, with the presence of polybrene (6 μg/ml).

Real-time quantitative reverse transcription-PCR (qRT-PCR) analysis

Total RNA was extracted from cell samples using the RNeasy Plus kit (Qiagen, Hilden, Germany) and reversely transcribed into cDNA using a high-capacity cDNA reverse transcription kit (Thermo Fisher Scientific), following the manufacturer’s instructions. qRT-PCR was conducted using SYBR Green qRT-PCR Master Mix (Thermo Fisher Scientific), on a 7500 Fast Real-Time PCR System (Thermo Fisher Scientific). Relative gene expression was calculated by estimating fold changes by the 2−ΔΔCt method using β-actin as an endogenous control [31]. The primers used are summarized in Table 1.

Table 1.

Primers used in this study

Primers for qRT-PCR
Gene	Primer: forward	Primer: reverse
TFB2M	5’-GGGAAAACCAAGTAGACCTCCAC-3’	5’-TTTCGAGCGCAACCACTTTGGC-3’
IL6	5’-AGACAGCCACTCACCTCTTCAG-3’	5’-TTCTGCCAGTGCCTCTTTGCTG-3’
CCL2	5’-AGAATCACCAGCAGCAAGTGTCC-3’	5’-TCCTGAACCCACTTCTGCTTGG-3’
CSF1	5’-TGAGACACCTCTCCAGTTGCTG-3’	5’-GCAATCAGGCTTGGTCACCACA-3’
TGFB1	5’-TACCTGAACCCGTGTTGCTCTC-3’	5’-GTTGCTGAGGTATCGCCAGGAA-3’
CCL22	5’-GTGGAAGACAGTATCTGCTGCC-3’	5’-AGGCTTGCGGCAGGATTTTGAG-3’
CD86	5’-CCATCAGCTTGTCTGTTTCATTCC-3’	5’-GCTGTAATCCAAGGAATGTGGTC-3’
CD163	5’-CCAGAAGGAACTTGTAGCCACAG-3’	5’-CAGGCACCAAGCGTTTTGAGCT-3
CD206	5’-GTTCACCTGGAGTGATGGTTCTC-3’	5’-AGGACATGCCAGGGTCACCTTT-3’
ACTB	5’-CACCATTGGCAATGAGCGGTTC-3’	5’-AGGTCTTTGCGGATGTCCACGT-3’.
Primers for cytosolic mtDNA quantification
MT-CO1	5’-GCCCCCGATATGGCGTTT-3’	5’-GTTCAACCTGTTCCTGCTCC-3’
18S rDNA	5’-TAGAGGGACAAGTGGCGTTC-3’	5’-CGCTGAGCCAGTCAGTGT-3’

Cytosolic mtDNA quantification

Cytosolic mtDNA quantification followed the strategies used in previous publications [32,33]. In brief, cytosolic cell fractions were prepared using the Cell Fractionation Kit (ab109719, Abcam, Cambridge, UK) according to the manufacturer’s instructions. In brief, cell samples were collected, washed in Buffer A, re-suspend in 1X Buffer A, mixed with an equal volume of Buffer B, centrifuged at 5,000 × g for 1 min at 4°C. Then, the supernatants were transferred to new tubes and further centrifuged at 20,000 × g for 10 min. The resulting supernatants were collected as cytosolic fractions and were used to isolate cytosolic DNA, using The FlexiGene DNA Kit (no. 51,206, Qiagen), following the manufacturer’s instructions. The copy number of cytosolic mtDNA was measured by qPCR, using the primers amplifying gene encoding mitochondrial cytochrome c oxidase 1 (MT-CO1), with the same volume of DNA solution. Normalization to the nuclear genome was performed using DNA isolated from tissue homogenate using primers targeting 18S rDNA (encoding 18S rRNA). The copy numbers of cytosolic mtDNA were normalized to that of the copy numbers of nuclear DNA, as described previously [33]. Then, comparison (fold change) was performed between groups. The primers for MT-CO1 and 18S rDNA are summarized in Table 1.

Western blot assay

Cell samples were collected, washed, and lysed in RIPA buffer in the presence of Halt Protease and Phosphates Inhibitor Cocktail (Thermo Fisher Scientific), as described previously [34]. Protein concentration was measured using the BCA kit (Thermo Fisher Scientific). 30 μg of proteins were loaded to each lane, separated by SDS-PAGE and transferred to nitrocellulose membranes. Then, the membranes were blocked in 5% nonfat dry milk in TBST for 1 h, and incubated overnight at 4°C with the primary antibodies. After washing, the membranes were incubated with HRP-conjugated anti-rabbit or anti-mouse IgG for 1 h at room temperature. ECL Plus reagents (Beyotime, Shanghai, China) were used for detection. Protein signals were detected using with Chemi-Doc MP System (Bio-Rad Laboratories). The following primary antibodies were used, including anti-TFB2M (anti-TFB2M, 1: 1000, AP10145b, ABCEPTA), anti-p65 (1: 2000, 66,535-1-Ig, Proteintech, Wuhan, China), anti-phospho-p65 (p-p65) (Ser536) (93H1) (#3033, 1:1000, Cell Signaling Technology, Danvers, MA, USA), anti-β-actin (1:2000, 20,536-1-AP, Proteintech) and anti-Histone-H3 (1:5000, 17,168-1-AP, Proteintech).

Detection of mtDNA and cytosolic DNA with staining

Fluorescent labeling of mtDNA and cytosolic DNA was conducted as described previously [33]. Cells were seeded in 24-well plates at a density of 5 × 104 cells/well. 24 h later, cells were subjected to lentiviral mediated TFB2M overexpression. 48 h later, the culture medium was replaced by FBS-free growth medium containing 200 nM MitoTracker Red CMXRos (Beyotime) and was further cultured for 30 min. Then, the medium was removed, and cells were washed. 1 ml working solution of PicoGreen dsDNA Reagent (Invitrogen, P7581) was added and further incubated for 10 min at 37°C in the dark. Then, the staining solution was replaced with fresh prewarmed medium.

To detect TLR9 and mtDNA co-localization, cells with or without TFB2M overexpression were fixed with 4% paraformaldehyde, treated with 0.1% Triton X-100, and blocked with 3% BSA. Then, the cells were then incubated with anti-TLR9 (1:200, DF2970, Affinity Bioscience, Changzhou, China) at 4°C overnight. After washing, the cells were incubated with goat anti-rabbit IgG (Dylight 549-conjugated) secondary antibodies (Thermo Fisher Scientific). Then, mtDNA was detected using 1 ml working solution of PicoGreen dsDNA Reagent (Invitrogen, P7581) for 10 min at 37°C in the dark. After that, fluorescence signaling was detected using IX83 Inverted Microscope (Olympus, Tokyo, Japan).

Protein transfection or inhibitor treatment

48 h after lentiviral infection, cells were collected and plated in 6-well plates at a density of 5×105/well. 24 h later, cells were subjected to protein transfections or inhibitor treatment. For protein infection, cells were washed with warm FBS-free culture medium and transfected for 4 h at 37°C with 2 μg/mL DNase I or lactate dehydrogenase as a nontargeted protein control, using PULSin reagent (Polyplus Transfection, Illkirch, France). Then, the transfection medium was removed and cells were incubated for 24 h in complete growth medium. Cells and culture medium was then collected for further studies, as described previously [29].

For TLR9 inhibitor treatment, cells were treated with ODN INH-18 (1 µM) for 6 h (InvivoGen, Toulouse, France). Then, the inhibitor-containing culture medium was removed, and cells were incubated for 24 h in complete growth medium. Cells and culture medium was then collected for further studies.

For nuclear factor kappa-B (NF-κB) inhibitor treatment, cells were treated with PDTC (10 µM) for 6 h (Selleck, Shanghai, China). Then, cells were collected for further studies.

The concentration of IL-6 in the condition mediums was measured using an ELISA assay kit (EK0410) purchased from Boster (Wuhan, China), following the manufacturer’s instructions. The cell samples were collected for qPCR to quantify cytosolic mtDNA copy numbers and western blotting to detect the expression of TFB2M and NF-κB p65 in the cytoplasm and nucleus.

Transwell assay of migration and macrophage polarization

THP-1 cells were treated with 10 ng/ml 12-O-tetradecanoylphorbol-l3-acetate (TPA) for 24 h to induce M0 macrophage [35]. Transwell assay of migration was performed using 6.5 mm Transwell with 5.0 µm Pore Polycarbonate Membrane Insert (Corning, New York, NY, USA). In brief, 5×104 THP-1 cells after TPA treatment were added to the top chamber of the insert for 1 h, following which the non-adherent cells were removed. For the Tocilizumab (an anti-IL-6 R mAb) (A2012, Selleck) treatment, THP-1 cells were pretreated with Tocilizumab (2 μg/ml) for 12 h before seeding. The bottom charmer was filled with a conditioned medium collected as described above with 20% FBS. The adherent cells were left to migrate overnight in 5% CO2 at 37°C. Then, cells on the top side were removed by a cotton swab. Cells migrated to the bottom side were fixed in methanol, stained with crystal violet, and counted under phase-contrast microscopy. Migrated cells were counted by five randomly selected fields.

To explore the effect of the conditioned medium on macrophage polarization, we seeded 5×105 THP-1 cells/well in six-well plates. After adherence, cells were treated with 10 ng/ml of TPA for 24 h and then treated with different conditioned mediums collected above for another 48 h. Then, cells were collected and washed three times with cold PBS containing 5% human serum and 0.1% NaN3. After that, cells were incubated with the FITC PE Anti-Human CD163 (PE-65169) or Anti-Human CD206 (FITC-65155) at 4°C in the dark for 30 min. Then, cells were washed, re-suspended in 200 µl of 1x flow cytometry staining buffer, and analyzed on a flow cytometer (BD Accuri C6 Plus, BD, Franklin Lakes, NJ, USA).

Statistical analysis

All statistical analyses were performed with GraphPad 8.10. Data are presented as means ± standard deviation (SD). Quantitative results are reported based on at least three independent experiments. Two-group comparisons were compared with the Unpaired t-test with Welch’s correction. To compare differences among three or more groups, One-way ANOVA followed by Dunnett’s post-hoc test was performed. p< 0.05 was considered statistically significant.

Results

In this study, we the expression profile of TFB2M in ovarian cancer, its association with the infiltration of TAMs and its regulatory effects on macrophage polarization. We found that TFB2M upregulation increases cytosolic mtDNA copy number, which stimulates IL-6 expression and secretion via a TLR9/NF-κB signaling pathway. IL-6 serves as a downstream effector in promoting macrophage infiltration and M2 macrophage polarization.

TFB2M expression is positively correlated with TAM infiltration in ovarian tumor tissues

To characterize the expression of TFB2M/TFB1M/TFAM in normal and ovarian tumor tissues, we extracted RNA-seq data from GTEx and TCGA. Compared to normal fallopian tubes, ovarian tumor tissues had a significantly higher expression of TFB2M and TFAM and a substantially lower expression of TFB1M (p< 0.05, Figure 1(a)). Survival data from the Kaplan-Meier Plotter indicated that for ovarian cancer patients, higher TFB2M expression is associated with substantially shorter progression-free survival (PFS) (Figure 1(b)) and overall survival (OS) (Figure 1(c)). In comparison, higher TFAM expression is associated with significantly better survival (data were not shown). Therefore, we only focused on TFB2M in this study.

Figure 1.

. Comparison of TFB2M/TFB1M/TFAM RNA expression in normal fallopian tube (GTEx-FT, n = 5) in GTEx project and primary ovarian cancer tissues in TCGA (TCGA-OV, n = 418). B-C. K-M survival analysis of PFS (b) and OS (c) in patients with primary ovarian cancer, using data from Kaplan-Meier Plotter. Patients were grouped by the auto select best cutoff of TFB2M expression. D. IHC staining of TFB2M in the normal fallopian tube and primary ovarian cancer tissues in the HPA dataset. Images were retrieved from the following links: https://www.proteinatlas.org/ENSG00000162851-TFB2M/tissue/fallopian+tube and https://www.proteinatlas.org/ENSG00000162851-TFB2M/pathology/ovarian+cancer#ihc. E-F. Representative IHC staining images of TFB2M and CD163 in serial ovarian cancer tissue arrays (e). Correlation between TFB2M expression and percentage of CD163+ cells in 68 ovarian cancer tissues (f). Patients were divided into four groups by the quartile value of TFB2M staining or percentage of CD163+ cells. Spearman’s correlation analysis was conducted. OVC: ovarian cancer. FT: fallopian tube. *p < 0.05, ***p < 0.001.

IHC staining data in the Human Protein Atlas (HPA) [27] (https://www.proteinatlas.org/) showed that epithelial glandular cells of fallopian tube had medium TFB2M expression. In comparison, medium to high TFB2M staining was observed in ovarian cancer tissues (Figure 1(d)).

Some recent studies showed that mtDNA stress is associated with TAM infiltration in cancer [22], we further checked the relationship between TFB2M expression and TAM infiltration in ovarian cancer. IHC staining was performed to detect the expression of TFB2M and CD163 (a TAM marker) in 68 ovarian tumor tissues, using commercial serial sections. By scoring TFB2M staining and CD163-positive cells, we observed a moderately positive correlation between TFB2M expression and the infiltration of CD163-positive cells in ovarian cancer tissues (Spearman’s r = 0.40, p < 0.001) (Figure 1(e,f)), implying a link between TFB2M upregulation and TAM infiltration.

TFB2M expression is positively correlated with cytosolic mtDNA copy number in ovarian cancer cells

Since TFB2M has an important regulatory role in mtDNA content, we further examined whether it regulates cytosolic mtDNA copy number. Ovarian cancer cell lines OVISE and CAOV4 cells were subjected to TFB2M overexpression or knockdown (Figure 2(a,b)). Then, fluorescent detection was performed to visualize mtDNA and mitochondria in OVISE and CAOV4 cells with TFB2M overexpression. PicoGreen was used to stain double-strand DNA (green). MitoTracker (red) was utilized to label mitochondria. Since cytoplasmic PicoGreen staining is specific for mtDNA [36], PicoGreen that colocalizes with MitoTracker (yellow) is considered mtDNA in mitochondria. In contrast, PicoGreen that does not colocalize with either MitoTracker or the nucleus is considered cytosolic mtDNA (Figure 2(c)). TFB2M overexpression was associated with increased cytosolic mtDNA. Then, mtDNA in the cytosolic fractions was quantified using qPCR assay. Results showed that the copy number of cytosolic mtDNA was significantly higher in cells with TFB2M overexpression (Figure 2(d,e)). These data imply that TFB2M overexpression might result in increased cytosolic mtDNA.

Figure 2.

TFB2M expression is positively correlated with cytosolic mtDNA stress in ovarian cancer cells A-B. QRT-PCR (top panels) and western blot (bottom panels) analysis of TFB2M expression at the mRNA and protein levels in OVISE and CAOV4 cells 48 h after lentiviral infection for TFB2M overexpression (a) or knockdown (b). C. Fluorescent detection was performed to visualize double-strand DNA (green, using PicoGreen) and mitochondria (red, using MitoTracker) in OVISE and CAOV4 cells 48 h after lentiviral infection for TFB2M overexpression. Cytosolic mtDNA was indicated by white arrows. D-E. qPCR was used to determine the copy number of mtDNA in the cytoplasm of OVISE and CAOV4 cells 48 h after lentiviral infection for TFB2M overexpression (d) or knockdown (e). Data shown are the mean ± SD. from three independent experiments. EV: empty vector; shTFB2M: TFB2M shRNA; scr.: shRNA scramble control. **p < 0.01, ***p < 0.001.

Cytosolic mtDNA stress enhanced the secretion of IL-6 by activating the nuclear factor-κB (NF-κB) signaling pathway

To understand the potential involvement of cytosolic mtDNA stress in TAM infiltration, we checked the expression and secretion of major cytokines and chemokines that influence TAM infiltration, including CCL2, CSF1, TGF-β, IL-6, and CCL22 in OVISE and CAOV4 cells with or without TFB2M overexpression. TFB2M overexpression significantly increased mRNA expression and protein secretion of IL-6 in both cell lines (Figure 3(a–d)). Cytosolic mtDNA is one major type of damage-associated molecular patterns (DAMPs), which trigger the inflammatory responses via binding with TLR9, including in epithelial tumor cells [26,37]. Since TFB2M overexpression can increase cytosolic mtDNA, we detected the alteration of IL-6 expression after mtDNA degradation. Depletion of cytosolic mtDNA using DNase I treatment significantly reduced IL-6 transcription and secretion and also weakened TFB2M overexpression induced IL-6 expression (Figure 3(c,d)). In addition, qPCR results confirmed that DNase I treatment significantly reduced basal cytosolic mtDNA and largely abrogated the effect of TFB2M overexpression on increasing cytosolic mtDNA (Figure 3(e)).

Figure 3.

Cytosolic mtDNA stress enhances the secretion of IL-6 by activating the nuclear factor-κB (NF-κB) signaling pathway A-B. qRT-PCR analysis for the mRNA expression of genes encoding TAM recruitment associated cytokines and chemokines including CCL2, CSF1, TGF-β, IL-6, and CCL22 in OVISE (a) and CAOV4 (b) cells 48 h after lentiviral mediated TFB2M overexpression. C-D. qRT-PCR (c) analysis of IL-6 transcription in cells and ELISA (d) assay of IL-6 secretion in the supernatants of cultured OVISE and CAOV4 cells with TFB2M overexpression alone, treated with DNase I alone or with combined TFB2M overexpression and DNase I treatment. E. qPCR was used to determine the copy number of mtDNA in the cytoplasm of OVISE and CAOV4 cells with TFB2M overexpression alone, treated with DNase I alone or with combined TFB2M overexpression and DNase I treatment. F. Immunofluorescent staining of TLR9 in OVISE and CAOV4 cells with or without TFB2M overexpression. The interactions between TLR9 and cytosolic mtDNA were indicated by white arrows. G-J. Western blot analyses were performed to detect the protein levels of TFB2M in whole cells and phospho-NF-κB p65 (p-p65) (Ser536) and p65 in the cytoplasm and nucleus of OVISE and CAOV4 cells with TFB2M overexpression alone or with combined ODN INH-18 (a TLR9 antagonist), DNase I or PDTC (an NF-κB inhibitor) treatment. Total TFB2M, nucleus p-p65, and p65 in each group were quantified using ImageJ software (h-j). Data shown are the mean ± SD. from three independent experiments. EV: empty vector. **p < 0.01, ***p < 0.001, n.s., not significant.

TLR9 has a verified regulatory effect on IL-6 expression via NF-κB signaling [38,39]. Therefore, we explored whether TFB2M promotes IL-6 expression via this mechanism in ovarian cancer cells. Immunofluorescent staining confirmed that TFB2M overexpression enhanced the binding between mtDNA and TLR9 in the two cell lines (Figure 3(f)). Western blot data showed that the nuclear translocation of NF-κB p65 and p-p65 was significantly increased by TFB2M overexpression. However, this effect could be canceled by treatment with TLR9 antagonist (ODN INH-18), DNase I, or NF-κB inhibitor PDTC (Figure 3(g–j)). Based on these findings, we infer that TFB2M could promote IL-6 expression and secretion via increasing cytosolic mtDNA and activating the downstream TLR9-NF-κB signaling pathway.

TFB2M overexpression induced cytosolic mtDNA stress enhances macrophage infiltration and M2 polarization by IL-6 in ovarian cancer

To explore whether elevated IL-6 secretion mediated by TFB2M overexpression and mtDNA stress contributes to macrophage infiltration, we tested how the conditioned medium from OVISE cells influenced macrophage recruitment by in vitro migration assay. Conditioned medium from TFB2M-overexpression OVISE cells significantly enhanced the migration of THP-1 cells compared to the medium from the empty vector (EV) control group (Figure 4(a,b)). However, the medium collected from OVISE cells treated with ODN INH-18 in or DNase I did not increase the migration (Figure 4(a,b)). In addition, when THP-1 cells were pretreated with an IL-6 R mAb (Tocilizumab), the migration capability was significantly suppressed (Figure 4(a,b)).

Figure 4.

. Representative images (a) and quantitation (b) of transwell migration assay of THP-1 cells by conditioned medium from OVISE cells as indicated. In the Tocilizumab group, cells were pretreated with 2 μg/ml Tocilizumab for 12 h. C-D. qRT-PCR analysis was performed to detect the expression of M2 macrophage markers (CD163 and CD206) (c) and M1 macrophage markers (CD86) (d) at the mRNA level in THP-1 macrophages treated with different conditioned mediums as in panel A. E-G. Representative images (e) and quantitation (f-g) of flow cytometric analysis to detect the proportion of CD163+ (e-f) or CD206+ (e and g) THP-1 macrophages treated with different conditioned mediums as in panel A. Data shown are the mean ± SD. from three independent experiments. **p < 0.01, ***p < 0.001.

Next, we studied whether TFB2M enhances the M2 polarization of THP-1 cells via secreting IL-6. QRT-PCR analysis showed that THP-1 cells treated with conditioned medium from TFB2M-overexpressed OVISE cells had a significantly increased mRNA expression of M2 macrophage markers CD163 and CD206 (Figure 4(c)), with unchanged CD86 (an M1 macrophage marker) expression (Figure 4(c,d)). Flow cytometry analysis confirmed these trends at the protein level (Figure 4(e–g)). As expected, ODN INH-18 or DNase I treatment abrogated the M2 macrophage-inducing effect of the conditioned medium (Figure 4(c,d)). When THP-1 cells were pretreated with Tocilizumab, the trend of M2 polarization was also canceled (Figure 4(c,d)).

Discussion

The contribution of dysregulated the mtDNA-specific transcriptional complex genes in cancer development has been gradually revealed. TFAM might be a target of Melatonin in suppressing the growth of malignant glioma cells [40]. TFB2M upregulation in hepatocellular carcinoma (HCC) can facilitate tumor cell growth and metastasis by activating ROS-Akt-NF-κB signaling [41]. It also activates aerobic glycolysis in HCC cells through the NAD+/SIRT3/HIF-1α signaling [42]. In this study, we found that TFB2M was upregulated in ovarian cancer tissues. Its expression was positively correlated with M2 macrophage infiltration. Macrophage recruitment and polarization in the TME can be affected by complex autocrine and paracrine mechanisms. Generally, M1 macrophages can be induced by molecules such as IFN-γ, LPS, GM-CSF, and IL-12, while M2 macrophages can be induced by multiple factors such as IL-4, IL-6, IL-10, IL-13, CSF1, CCL2, TGFβ, VEGF, CCL3, and CCL14 [10,11]. Ovarian cancer cells have complex genetic and epigenetic alterations, leading to dysregulated secretion of molecules regulating the immune cell phenotypes.

Since TFB2M modulates both mtDNA transcription and copy number, we examined its influence on cytosolic mtDNA copy number in ovarian cancer. Results showed that it positively modulates cytosolic mtDNA copy number. Ovarian tumor cells usually had significantly higher mtDNA copy numbers compared to normal ovarian tissues [43]. However, the contribution of this alteration to cancer development is still not well understood.

Recent studies have linked the TFB2M dysregulation, adaptive metabolic alteration, and cytosolic mtDNA alterations in solid tumors. TFB2M upregulation might be an adaptive metabolic alteration to support aerobic glycolysis and mitochondrial fission during tumor progression. In HCC, TFB2M upregulation can elevate the expressions of aerobic glycolysis-associated genes, such as GAPDH, LDHA, GLUT1, and HK2. On the other hand, TFB2M overexpression suppresses the peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1α), a critical regulator of mitochondrial respiration [42]. Increased aerobic glycolysis is quite common in solid tumors [44]. Besides dysregulated genes related to energy metabolism, aerobic glycolysis is also associated with increased mitochondrial fission in multiple cancers [45], including ovarian cancer [46]. Mitochondrial fission is usually associated with cytosolic release of mtDNA, which causes cytosolic mtDNA stress [21,22]. In the cytoplasm, mtDNA can bind to multiple sensor proteins and trigger different downstream signaling pathways. Notably, mtDNA can directly bind to TLR9 and trigger the inflammatory responses in epithelial tumor cells [26,37].

Therefore, we further explored whether TFB2M affects cytokine or chemokine expression and secretion in ovarian cancer cells. Among the molecules examined, we confirmed that TFB2M overexpression in OVISE and CAOV4 cells could increase IL-6 expression and secretion. When cytosolic mtDNA was selectively inhibited by DNase I, TFB2M-induced IL-6 expression was canceled. TFB2M overexpression can activate NF-κB signaling pathway via promoting nucleus entry of p-p65. However, when cytosolic mtDNA, TLR9 or NF-κB signaling pathways were selectively inhibited, the promoting effect was canceled. These findings confirmed that TFB2M upregulation could elevate IL-6 expression and secretion from ovarian cancer cells via cytosolic mtDNA/TLR9/NF-κB signaling pathway.

TLR9 has a verified regulatory effect on IL-6 expression via NF-κB signaling [38,39]. Interestingly, IL-6 has been confirmed as a critical cytokine enhancing macrophage infiltration and M2 macrophage polarization in TME [30,47]. In the current study, we showed that conditioned medium from OIVSE cells with TFB2M overexpression could induce macrophage migration and M2 polarization. However, these inducing effects were abrogated by DNase I, TLR9 inhibitor, and anti-IL-6 R pretreatment.

Acute injury or stimulation-induced cytosolic mtDNA can act as immunostimulatory molecules in normal cells by engaging the cytosolic cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway [48], some pattern-recognition receptors (PRRs) and triggering type I interferons and interferon-stimulated gene (ISG) expression [21,49]. Although acute activation of these signaling pathways can generate barriers to early tumorigenesis, their chronic activation exerts tumorigenic effects by establishing an immune-suppressive tumor microenvironment [48,50]. In the future, it is meaningful to explore the regulation of TFB2M during the early tumorigenesis and development of ovarian cancer. Previous studies reported that TP53 is a guardian of the mitochondrial genome by reducing its susceptibility to oxidative damage and replication errors [51,52]. Considering the different TP53 status in OVISE CAOV4 cells, we infer that this might be the reason why different levels cytosolic mtDNA were observed in these two cell lines after TFB2M overexpression (Figure 2(c)). Future studies are warranted to explore the underlying mechanisms.

Conclusion

This study showed a novel functional role of TFB2M in the immunosuppressive TME in ovarian cancer. Its upregulation directly leads to increased cytosolic mtDNA copy number, which stimulates IL-6 expression and secretion via a TLR9/NF-κB signaling pathway. IL-6 serves as a downstream effector in promoting macrophage infiltration and M2 macrophage polarization.