H3K9me3-mediated repression of KLF6: Discovering a novel tumor suppressor in liposarcoma using a systematic epigenomic approach

ABSTRACT

The pathogenesis of well-differentiated and dedifferentiated liposarcoma is poorly understood. We recently reported Kruppel-like factor 6 (KLF6) as a histone H3 lysine 9 trimethyl (H3K9me3)-regulated and differentially expressed transcription factor serving a previously unappreciated tumor suppressor role in liposarcoma. Mechanistically, KLF6 may drive adipocytic differentiation through increased expression of known regulators of adipogenesis.

Liposarcoma (LPS) is one of the most common soft tissue sarcomas. The two most common subtypes, well-differentiated (WD) and dedifferentiated (DD) liposarcoma, are characterized by markedly different clinical behavior and patient prognosis. WDLPS are low-grade tumors with a more indolent course whereas DDLPS are high-grade aggressive tumors that systemically metastasize in up to 20% of patients.^1,2 Surgical resection is the standard of care as liposarcomas are generally resistant to conventional chemotherapies; however, most patients with DDLPS of the retroperitoneum will succumb to disease within 5 years. Thus, gaining a better understanding of the pathogenesis of WDLPS and DDLPS is a critical first step toward developing effective targeted therapies for liposarcoma. The evolution of WDLPS and DDLPS is poorly understood and whether one arises from the other or they arise independently is unknown. Interestingly, they often co-occur in patients either at the time of initial presentation or at disease recurrence and both are characterized by chromosome 12q14–15 amplification, resulting in mouse double minute 2 homolog (MDM2) and cyclin-dependent kinase 4 (CDK4) amplification in the majority of cases.^3-6 The similar presence of these genetic drivers suggests that epigenetic alterations may underlie the differences between WDLPS and DDLPS, but a comprehensive understanding of epigenetic differences is lacking.^7-8

In our study,⁹ we explored the epigenomic differences between WDLPS and DDLPS in cell-based systems and human tumor samples using unbiased approaches. We first performed a global analysis of 9 common epigenetic marks across human LPS tumors and found that DDLPS tumors exhibited higher levels of histone 3 lysine residue 9 trimethylation (H3K9me3). Increased levels of H3K9me3 in DDLPS contributed to the aggressiveness and dedifferentiation phenotype as treatment with chaetocin, a methyltransferase inhibitor, significantly reduced proliferative capacity and increased differentiation potential. We also noted significant differences in DNA methylation at the level of intensity and number of positive cells; the functional significance of this change must be determined in future studies.

To understand the implications of differences in H3K9me3 we performed ChIP-sequencing analysis of this mark across 7 human LPS cell lines, which confirmed higher H3K9me3 levels in DDLPS and showed enrichment for regulatory regions of genes related to cellular differentiation and migration. These results suggest that the repressive histone mark H3K9me3 may regulate aggressiveness in LPS by blocking the maintenance of adipocytic differentiation. Intersection of H3K9me3 ChIP-seq and gene expression microarray data sets in the DDLPS and WDLPS cells lines identified the transcription factor kruppel-like factor 6 (KLF6) as a potential target. KLF6 is significantly underexpressed, and an upstream regulatory region of KLF6 showed higher levels of H3K9me3, in DDLPS cell lines compared with WDLPS. This observation was confirmed in DDLPS and WDLPS tumors, in that KLF6 was underexpressed at both the mRNA and protein level in DDLPS tumors. Additionally, we showed that pharmacologic inhibition of H3K9me3 decreased H3K9me3 levels at the KLF6 locus and resulted in increased expression of this gene. In sum, we identified that repression of KLF6 in DDLPS is a result of increased H3K9me3 at its associated regulatory region.

KLF6 acted as a driver of the DDLPS phenotype as rescuing KLF6 expression in DDLPS cell lines decreased proliferation and delayed tumor onset when injected into nude mice. KLF6 overexpressing cells also exhibited morphologic changes consistent with induction of adipocytic differentiation and cellular senescence as well as increased expression of proadipogenic factors such as peroxisome proliferator-activated receptor gamma (PPARγ), CCAAT/enhancer binding protein β (CEBPα), and CCAAT/enhancer binding protein α [ΑΘ2]).KLF6 knockdown abrogated these effects of pharmacologic H3K9me3 inhibition. Overall, our data suggest a tumor suppressive role for KLF6 in liposarcoma and provide evidence that increased H3K9me3 levels promote the aggressive dedifferentiated phenotype of DDLPS via epigenetic silencing of KLF6 (Fig. 1). Although our studies suggest KLF6 silencing as a primary target underlying H3K9me3-mediated aggressiveness of DDLPS, other targets of this mark might contribute to its function in this disease. Indeed, H3K9me3 is primarily a heterochromatin silencing mark and therefore differences at the global level between WDLPS and DDLPS may result in loss of silencing of transposable elements, thus creating genetic instability that may contribute to the aggressiveness of DDLPS.

Figure 1.

KLF6: a differentially expressed and epigenetically regulated oncosuppressive transcription factor. In dedifferentiated liposarcoma (DDLPS), compared to well-differentiated liposarcoma (WDLPS), higher histone H3 lysine 9 trimethyl (H3K9me3) levels at the Kruppel-like factor 6 (KLF6) locus suppress gene expression, resulting in decreased expression of proadipocytic regulators peroxisome proliferator-activated receptor gamma (PPARγ), CCAAT/enhancer binding protein β (CEBPβ), and CCAAT/enhancer binding protein α (CEBPα).

Our results provide new insights into liposarcomagenesis and suggest that a potential transition between WDLPS and DDLPS occurs that involves pre-existing networks utilized during normal adipocytic differentiation. It might be speculated that H3K9me3 loss opens up the chromatin at certain super-enhancer regions, thereby helping proadipogenic transcriptional reprogramming. We postulate that H3K9me3 inhibition may be of potential therapeutic benefit in liposarcoma. Future studies to elucidate the mechanisms by which the KLF6 locus is targeted for differential H3K9 trimethylation, as well as the pathways by which KLF6 mediates its proadipocytic differentiation effects, will be critical and of clinical importance in a disease that we understand poorly and for which the only current means of disease control is often repeated and morbid surgical resections. Importantly, our work identifying and validating KLF6 as an H3K9me3-regulated factor in liposarcoma also highlights the value of applying epigenomic approaches as a discovery platform to identify novel targets in liposarcomagenesis and other diseases.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.