Impact of histone demethylase KDM3A-dependent AP-1 transactivity on hepatotumorigenesis induced by PI3K activation

Abstract

Epigenetic gene regulation linked to oncogenic pathways is an important focus of cancer research. KDM3A, a histone H3 lysine 9 (H3K9) demethylase, is known to have a pro-tumorigenic function. Here, we showed that KDM3A contributes to liver tumor formation through the phosphatidylinositol 3-kinase (PI3K) pathway, which is often activated in hepatocellular carcinoma. Loss of Kdm3a attenuated tumor formation in Pik3ca transgenic (Tg) mouse livers. Transcriptome analysis of pre-cancerous liver tissues revealed that the expression of activator protein 1 (AP-1) target genes was induced by PI3K activation, but blunted upon Kdm3a ablation. Particularly, the expression of Cd44, a liver cancer stem marker, was regulated by AP-1 in a Kdm3a-dependent manner. We identified Cd44-positive hepatocytes with epithelial-mesenchymal transition-related expression profiles in the Pik3ca Tg liver and confirmed their in vivo tumorigenic capacity. Notably, the number and tumor-initiating capacity of Cd44-positive hepatocytes were governed by Kdm3a. As a mechanism in Kdm3a-dependent AP-1 transcription, Kdm3a recruited c-Jun to the AP-1 binding sites of Cd44, Mmp7 and Pdgfrb without affecting c-Jun expression. Moreover, Brg1, a component of the SWI/SNF chromatin remodeling complex, interacted with c-Jun in a Kdm3a-dependent manner and was bound to the AP-1 binding site of these genes. Finally, KDM3A and c-JUN were co-expressed in 33% of human premalignant lesions with PI3K activation. Our data suggest a critical role for KDM3A in the PI3K/AP-1 oncogenic axis and propose a novel strategy for inhibition of KDM3A against liver tumor development under PI3K pathway activation.

A potential common role of the Jumonji C domain‑containing 1A histone demethylase and chromatin remodeler ATRX in promoting colon cancer

“…Hence, JMJD1A may be more important for promoting cancer stem cells than for stimulating the growth rate of tumors. Consistent with this concept, JMJD1A has been reported to foster cancer stemness in a variety of different tumors ( 15 , 16 , 19 , 56 ).…”

A network of epigenomic and transcriptional cooperation encompassing an epigenomic master regulator in cancer

“…Original findings and reported cooperation events of epigenomic regulators with transcription factor are enumerated Symbol TFClass Cooperation event Ref. JUN 1.1.1 KDM3A assists in recruiting JUN to AP1 binding sites in regulating expression of CD44, MMP7, and PDGFRB in liver adenocarcinoma tumor formation 51 KDM4A promotes a positive feedback loop by facilitating the binding of the AP1 complex to the promoters JUN and FOSL1 in squamous cell carcinoma cells 52 CEBPB 1.1.8 Novel event KDM4B serves as a cofactor for CEBPB in preadipocytes and is recruited to the promoters of CEBPB regulated cell cycle genes 53 MYOD 1.2.2 Novel event KDM4B regulates the expression of MYOD and physically interacts with MYOD thereby controlling myogenic differentiation 54 KDM4C decreases MYOD degradation and increase MYOD transcriptional activity to facilitate skeletal muscle differentiation 55 HIF1A 1.2.5 KDM3A expression is stimulated by HIF1A binding to a response element in the promoter region of KDM3A 24 KDM3A is regulated by HIF1A stimulating tumor formation in renal cell carcinoma 22 KDM3A cooperates with HIF1A to induce glycolytic genes in urothelial bladder carcinoma 25 SREBF1 1.2.6 Novel event KDM1A regulates SREBF1 binding to the FASN promotor stimulating lipogenesis 26 MYC 1.2.6 KDM3A stimulates MYC expression and attenuates its ubiquitin-dependent degradation by binding to a E3 ubiquitin ligase 56 KDM3A regulates transcription of MYC and PAX3 by directly binding to their promotors and regulates their H3K9me 2 level in breast adenocarcinoma 57 …”

The potential of KDM3A as a therapeutic target in Ewing Sarcoma and other cancers

“…Recently, a number of studies have implicated KDM3A (JMJD1A/JHDM2A), a member of the KDM3 subfamily with specificity for removal of mono and di-methyl marks from H3K9, in tumor/metastasis promotion, chemoresistance and other phenotypes, in cancers of epithelial origin (including the common cancers of breast 4 , prostate 5 and colon 6 ), liver 7 , and the hematopoietic system 8 . Additionally, recent studies have implicated KDM3A in solid malignancies of childhood, including the metastasis of neuroblastoma 9 , a malignant pediatric tumor of peripheral nervous system origin, and both tumorigenesis and metastasis of Ewing Sarcoma 10, 11 , a pediatric sarcoma of bone and soft tissue.…”

“…Approaches could entail similar targeting of key intermolecular interactions or an enzymatic activity of a critical co-factor. Alternatively, one could focus on key druggable downstream mediators of KDM3A action, such as the cell surface molecules CD44 in hepatocellular carcinoma 7 and MCAM in Ewing Sarcoma 11 , which could be inhibited using blocking antibodies. Further understanding of KDM3A biology in cancer should inform optimal approaches to therapeutic targeting.…”

“…Alternatively, or in complement, empiric studies of candidate compounds, alone and in combinations with other therapeutic agents, may help inform such decisions. Indeed, the favorable activity profile of the pan-JHDM inhibitor JIB-04 in pre-clinical studies provides an example of the potential merits of a broader spectrum JHDM blockade approach 14 , while mechanistic studies in some cancers suggest possible opportunities for combinatorial targeting approaches (eg: KDM3A blockade and PI3K inhibition 7 ).…”