The expression of IRF8 was originally detected in hematopoietic cells and its expression has long been thought to be restricted to hematopoietic cells
[56]. Although IRF8 is abundantly expressed in hematopoietic cells, particularly in B cells and DCs, it has since been determined that IRF8 is also expressed in epithelial cells in the intestine, colon, ocular lens, skin, lung, liver, and heart
[8][10][37][38][39][40][41][42][43][44][45]. Analysis of IRF8 expression from single cell RNA sequencing datasets revealed that IRF8 is indeed expressed in several types of non-hematopoietic cells, including theca cells, melanocytes, enterocytes and intestinal cells in humans.
5. IRF8 Expression and Function in Non-Hematopoietic Cancer Cells
While much research has been dedicated to understanding the role of IRF8 in hematopoietic cells, less research has investigated the role of IRF8 in non-hematopoietic cells. In certain types of cancers, IRF8 functions as a tumor suppressor gene
[18][19][42][43][44], and its expression is regulated by DNA methylation or micro-RNA interference
[74]. In breast and lung cancer, IRF8 is downregulated as a result of promoter hypermethylation
[43][75][76], and a higher IRF8 expression level was associated with a better prognosis
[19]. Similarly, in gastric, nasopharyngeal, esophageal, colorectal, lung, prostate, and renal cell carcinomas the expression of IRF8 is downregulated
[43][44][77][78][79][80][81]. IFN-γ induced expression of IRF8 was increased when gastric cancer cells were treated with a demethylating agent
[77]. In other cancer cell lines with a methylated IRF8 promoter, IFN-γ treatment showed decreased ability to induce IRF8 expression
[78]. Taken together, these studies provide evidence for promoter methylation as a mechanism for IRF8 silencing in certain types of human cancer cells. Analysis of human lung tumor specimens at the single cell level validate the early finding that IRF8 is silenced in the tumor cells. Except for certain tumor-infiltrating immune cells, including certain B cells, pDC and myeloid cells, IRF8 is down-regulated in most cell types and is undetectable in human lung cancer cells
[82].
Within non-hematopoietic cancer cells, IRF8 plays a protective role against the formation of a metastatic phenotype. In human colon cancer cells, IRF8 protein levels are inversely correlated with the metastatic phenotype
[46][47]. IRF8 expression sensitized the metastatic tumor cells to Fas-mediated apoptosis, indicating that loss of IRF8 in cancers could promote earlier metastasis
[46]. It has been suggested that colon cancer cells silence IRF8 expression through inhibition of pSTAT1 function at the IRF8 promoter
[47]. IRF8 may also function in non-hematopoietic cancers to negatively regulate the expression of MMP3, a matrix metalloproteinase associated with a poor prognosis and metastasis in various solid organ cancers
[83][84][85]. In a chemically-induced sarcoma model, an inverse relationship between IRF8 and MMP3 expression was demonstrated; in a mouse model of mammary cancer, loss of MMP3 reduced spontaneous lung metastasis
[86]. This suggests that IRF8 may negatively regulate MMP3 expression, protecting against MMP3-induced metastasis in solid organ tumors. While most research supports an anti-metastasis role for IRF8 in non-hematopoietic cancers, one study found that IRF8 promoted EMT-like phenomena, cell motility, and invasion in a human osteosarcoma cell line providing evidence for a possible role in the acquisition of a metastatic-like phenotype
[87].
IRF8 appears to impede the progression and formation of cancer by increasing the expression of tumor suppressor genes, such as Caspase 1, p21, p27, and PTEN
[44][81][88]. Expression of tumor suppressor genes inhibited lens cell carcinoma growth upon transfection with IRF8
[88]. Similarly, IRF8-induced expression of p27 induced lung cancer cell senescence
[44]. Mechanistically, IRF8 was shown to inhibit Akt activity thereby inducing the accumulation of p27 and promoting senescence. While IRF8 increases tumor suppressor genes, one study also showed an IRF8-induced decrease in oncogene (Yap1 and Survivin) expression in renal cell carcinoma cell lines
[81]. In this cell line, ectopic expression of IRF8 induced cell cycle G2/M arrest and apoptosis further supporting IRF8’s role in impeding cancer progression.
IRF8 functions as a promoter of Fas-mediated apoptosis. This form of extrinsic apoptosis is key to immune clearance of tumor cells and is targeted in cancer therapy
[89][90]. Modification of Fas expression can render immune checkpoint inhibitor (ICI) therapies unsuccessful. Colon cancer cells with low Fas expression exhibited decreased sensitivity to FasL-induced apoptosis (a mechanism utilized by ICIs)
[91] and lower Fas expression was correlated with decreased survival in colon cancer patients
[92]. One mechanism by which cancer cells may alter the expression of Fas, and thereby reduce sensitivity to apoptosis, is through downregulation of IRF8.
As previously mentioned, IRF8 is downregulated in many cancers. Studies revolving around a loss of IRF8 have elucidated ways in which IRF8 helps facilitate Fas-mediated apoptosis and resulting in tumor immune sensitivity. In soft tissue sarcoma cells, IRF8 was found to be a repressor of FLICE-like protein (FLIP)
[93]. Silencing of FLIP and ectopic IRF8 expression each increased susceptibility to FasL-induced apoptosis. This suggests that by repressing FLIP, IRF8 promotes Fas-mediated apoptosis susceptibility. Mechanistically, IRF8 disruption diminishes JAK1 expression and inhibits STAT1 phosphorylation to block IFN-γ induced Fas upregulation in sarcoma cells
[94]. IRF8 has also been found to regulate pro-apoptotic and anti-apoptotic factors contributing to Fas-mediated apoptosis. In tumor-induced MDSCs, levels of IRF8 are decreased. IRF8-deficient MDSCs showed an increase in anti-apoptotic Bcl-xL and a decrease in pro-apoptotic Bax, providing a mechanism by which IRF8 downregulation can impede the Fas-mediated apoptosis pathway to evade elimination by T lymphocytes
[32]. Another mechanism by which IRF8 can promote Fas-mediated apoptosis is through repression of acid ceramidase (A-CDase). A-CDase upregulation has been implicated in certain cancers such as prostate and breast
[95][96][97]. IRF8 induced repression of A-CDase resulted in C16 ceramide accumulation and increased apoptosis sensitivity
[31]. This suggests that IRF8 downregulation likely contributes to decreased Fas-mediated apoptosis sensitivity through failure to repress A-CDase. In summary, IRF8 promotes Fas-mediated apoptosis through repression of FLIP and A-CDase and by regulating apoptotic factors such as Bcl-xL and Bax
[31][32][93].
IRF8 functions as a link between cancer and the immune system. Poor tumoral homing of DCs and T cells was observed in melanoma-bearing IRF8 KO mice
[42]. Additionally, the few DCs that were able to infiltrate the tumor were immature and unable to produce T cell-mediated immune responses. Mechanistically, the decreased tumoral immune infiltration observed in IRF8 KO mice is suggested to occur through the aberrant expression of multiple chemokine receptors and ligands
[42]. When IRF8 was induced, immune infiltration was reestablished, and the chemokine expression pattern was reversed. Using a microfluidic device, in which melanoma cells and splenocytes were separated by a microchannel, IRF8 KO splenocytes exhibited a marked decrease in the ability to cross the channel to infiltrate the melanoma cells while melanoma cells exhibited marked propensity to invade the channels suggesting existence of an IRF8-regulated communicating soluble factor
[98].
OPN has recently emerged as another immune checkpoint that may compensate PD-L1 function to promote colon tumor immune evasion
[99]. IRF8 is expressed in human and mouse gastric and colon epithelial cells
[38][41] and binds to the ISRE elements at the
Spp1 promoter to repress OPN expression
[100]. In human colon and pancreatic cancers, IRF8 expression is down-regulated and OPN is up-regulated
[99][100][101][102][103]. scRNA-Seq analysis indicates that OPN is abundantly expressed in tumor cells, MDSCs, and ILCs
[99]. Furthermore, scRNA-Seq analysis revealed the tumor-promoting role of SPP1
+ cells in cancer
[104][105]. OPN protein level is highly elevated in the peripheral blood in human cancer patients and OPN expression is up-regulated in several different types of human cancers
[106][107][108][109][110][111][112][113][114]. Silenced IRF8 expression in both tumor cells and tumor-induced myeloid cells by DNA methylation and H3K9me3 deposition at the
Irf8 promoter are likely major mechanisms underlying OPN elevation in cancer patients and tumor-bearing mice
[41][46][47][78][115]. It was observed that IRF8 KO mice exhibited deficient generation of antigen-specific T cells and this deficiency was due to OPN engagement of CD44 to directly suppress T cell activation
[100]. Accordingly, OPN blockade increased activation of tumor-specific T cells and suppressed tumor growth in vivo
[99]. OPN is highly expressed in human pancreatic tumor cells and may bind to CD44 on tumor cells via autocrine and paracrine manners to promote pancreatic cancer stemness and progression
[101][103]. Therefore, in addition to functioning as a repressor of
SPP1 in immune cells such as MDSCs and ILCs, IRF8 not only functions as an intrinsic tumor suppressor
[41] but also suppresses tumorigenesis and progression through repressing OPN expression in tumor cells
[100].