The Jumonji JmjC domain containing KDM family

The Jumonji (JmjC) domain-containing KDM family members can be clustered into seven subfamilies (KDM2 – KDM8). The KDM2 cluster catalyzes H3K36me2/me1 demethylation; in addition, KDM2B/JHDM1B/FBXL10 has been suggested to catalyze demethylation of H3K4me3 [3]. Cell culture studies support that KDM2 proteins might have context-dependent pro- and anti-oncogenic functions. FBXL11 can suppress NF-κB-dependent growth of colon cancer 16,16-Dimethyl Prostaglandin E2 [4], while FBXL10 has been reported to inhibit proliferation of tumor cells correlating with repression of rRNA genes [3]. Several reports show that FBXL10 plays a critical role in cell proliferation, apoptosis, differentiation, senescence, and tumorigenesis by exerting its H3K36 demethylase activity at different genomic loci. Murine Fbxl10 is a protein which is expressed in two isoforms. Isoform 1 representing full-length Fbxl10 protein contains the N-terminal catalytic JmjC domain, followed by a CxxC zinc finger domain, a plant homeodomain finger (PHD), a Fbox, and 8 leucine-rich repeats. A shorter isoform lacks the JmjC domain. Both isoforms are expressed in vivo as shown by Fukuda et al. [5]. Loss of Fbxl10 has been associated with genomic instability as found in a Blm- (Bloom syndrome) deficient mouse strain, suggesting that Fbxl10 can act as a tumor suppressor [6]. In contrast, expression profiling of human cancers showed FBXL10 overexpression in pancreatic ductal carcinomas, basal-like triple negative breast cancers and various leukemias, indicating that FBXL10 possibly acts as an oncogene [7], [8], [9], [10]. In addition, mutation affecting FBXL10 are commonly found in human large B cell lymphoma [11]. In mouse embryonic fibroblasts, expression of Fbxl10 blocked cellular senescence, with Fbxl10 promoting immortalization by silencing the tumor suppressor miRNAs let-7b and miR-101, which target EZH2 [12]. However, the physiological role of Fbxl10 in vivo remains unclear. Deletion of the full length isoform of Fbxl10 resulted in abnormal increase in both cell proliferation and apoptosis accompanied by neural tube defects and exencephaly, suggesting that Fbxl10 is required for regulation of proliferation and apoptosis during early neurogenesis [5]. In a current study, Boulard et al. further established mice which lack both isoforms of Fbxl10; these mutants did not progress beyond mid-gestation and arrested with multiple developmental defects, suggesting substantial deregulation of developmental gene expression [13]. The more severe phenotype of the gene-trap mutant without any Fbxl10 expression indicates that Fbxl10 has a predominant function during embryogenesis. In the present study, we established a gain of function mouse model to analyze the role of Fbxl10 in vivo. Fbxl10 overexpressing mouse embryonic fibroblasts (MEFs) have the capability to migrate faster than wild-type MEFs. We attempted to investigate the molecular mechanism by which Fbxl10 participates in migration induction employing transcriptome and genome-wide chromatin binding analyses.
Materials and methods
Results
Discussion In the present study, we show that overexpression of Fbxl10 in MEFs by dox-induction for two days leads to an increase in migration, whereas proliferation was not altered. Contradictory reports on Fbxl10 promoting or blocking proliferation were published [3], [22], [25], [26], [27]. Several studies highlight the fact that most, if not all histone demethylases including Jmj-type chromatin enzymes function as parts of transcriptional or chromatin protein complexes, thereby modulating specificity and mechanism of action [8], [28], [29]. Depending on its association, Fbxl10 maybe promote or block proliferation by regulating gene expression of target genes. In our cellular context, Fbxl10 overexpressing MEFs have the capability to migrate faster than wild-type MEFs. Due to the low endogenous Fbxl10 level in MEFs, the migratory phenotype could be only partially reverted by RNAi-mediated knockdown of Fbxl10.