The cytosolic role of EZH2-IMPDH2 complex in melanoma progression and metastasis via GTP regulation


 The enhancer of zeste homolog 2 (EZH2) oncogene is a histone methyltransferase that functions canonically as a catalytic subunit of the polycomb repressive complex 2 (PRC2) to tri-methylate histone H3 at Lys 27 (H3K27me3). Although targeting of EZH2 methyltransferase is a promising therapeutic strategy against cancer, methyltransferase-independent oncogenic functions of EZH2 are also described. Moreover, pharmacological EZH2 methyltransferase inhibition was only variably effective in pre-clinical and clinical studies, suggesting that targeting EZH2 methyltransferase alone may be insufficient. Here, we demonstrate a non-canonical mechanism of EZH2’s oncogenic activity through interactions with inosine monophosphate dehydrogenase 2 (IMPDH2) and downstream promotion of guanosine-5'-triphosphate (GTP) production. Liquid Chromatography-Mass Spectrometry (LC-MS) of EZH2 immunoprecipitates from melanoma cell lines and human patient-derived xenografts (PDXs) revealed EZH2-IMPDH2 interactions that were verified to occur between the N-terminal EED-binding domain of cytosolic EZH2 and the CBS domain of IMPDH2 in a PRC2- and methylation-independent manner. EZH2 silencing reduced cellular GTP, ribosome biogenesis, RhoA-mediated actomyosin contractility and melanoma cell proliferation and invasion by impeding the activity and cytosolic localization of IMPDH2. Guanosine, which replenishes GTP, reversed these effects and thereby promoted invasive and clonogenic cell states even in EZH2 silenced cells. IMPDH2 silencing antagonized the proliferative and invasive effects of EZH2, also in a guanosine-reversible manner. In human melanomas, high cytosolic EZH2 and IMPDH2 expression were associated with nucleolar enlargement, a marker for ribosome biogenesis. We also identified EZH2-IMPDH2 complexes in a range of cancers in which Sappanone A (SA), which inhibits EZH2-IMPDH2 interactions and thereby IMPDH2 tetramerization, was anti-tumorigenic, although notably non-toxic in normal human melanocytes and bone marrow derived blood progenitor cells that lacked observable EZH2-IMPDH2 interactions. These findings illuminate a previously unrecognized, non-canonical, methyltransferase-independent, but GTP-dependent mechanism by which EZH2 regulates tumorigenicity in melanoma and other cancers, opening new avenues for development of anti-EZH2 therapeutics.

.  EF1a-V5-LIC vector backbone's SrfI/NotI RE using the cloning primers listed in Table   175 S1. pCMVHA hEZH2 and V5-EZH2 vector was used to generate EZH2-H689A mutant 176 vector using the mutagenesis primers listed in Table S1 with QuikChange II site-177 directed mutagenesis kit (Agilent) following the manufacturer's instructions. Custom 178 designed siRNA oligonucleotides listed in Table S1 were purchased from Bioneer  Table S1 182 using the Lipofectamine RNAiMAX transfection reagent (Invitrogen). 72 hours after 183 siRNA transfection, cells were used for functional assays or collected for western blot 184 analysis.    Proteins were run on 4-20% Mini-PROTEAN TGX Stain-Free Protein Gels (BioRad, 243 4568096) and then transferred to PVDF membrane by wet transfer system.

278
To measure cell proliferation rates, we plated equal numbers of cells in 6-well plates.

279
Cells were trypsinized and counted on the indicated days by haemocytometer after 280 trypan blue staining.

281
For clonogenicity assay, cells were fixed with ice-cold absolute methanol for 20 min 282 and air-dried for 15 minutes. Cells were stained with 0.5% Crystal Violet for 20 min at 283 room temperature and then rinsed with tap water to remove excess dye. Five random 284 fields of stained cells were imaged using bright field microscopy at 40 × magnification 285 and average cell numbers per field were plotted as a function of time.

334
Slides were rinsed in water three times and incubated with 0.2% gold chloride solution 335 (Sigma-Aldrich) for 2 minutes. Rinsed slides were incubated with 5% sodium 336 thiosulfate for 2 minutes. After rinsing with water twice, slides were counterstained with 337 10 µg/mL DAPI solution for 5 minutes. Slides were rinsed and mounted with 338 fluorescence mounting medium (Dako).  Table S1 was   For proximity ligation assays, cells were seeded on round coverslips. After 24 h of 368 seeding, cells were fixed with 4% PFA for 15 min at RT, rinsed three times with PBS, 369 and blocked for 1 h using blocking buffer(5% normal goat serum containing 0.3%        515 We recently reported that EZH2 abundance, but not its methyltransferase 516 activity, may be the key therapeutic target to lower tumorigenic potential and 517 metastasis of low pigmented melanoma cells 35 (Fig. 1A, S1A and S1B), they had no effect on cell growth, clonogenicity, migration, 526 invasion, or pigmentation in BRAFV600E mutant cell lines A375 and IGR37, but partial 527 effect on NRASQ61K mutant C006-M1 melanoma cells (Fig. 1A-1F and S1A-S1H).

528
On the contrary, siEZH2-mediated knockdown or EZH2 degradation by DZNep

703
GTP is also essential for G-protein activity 11 . Next, we focused on Rho-   shown to be the docking site for IMPDH2 ( Figure 5A). In addition, stable EZH2 720 knockdown reduced cell proliferation, migration and invasion but these effects were  (Table S2 and Figure 6N). In parallel, SA treatment has no 797 cytotoxic effect on the blood progenitor cells in a time-dependent manner ( Figure 6O).

798
In addition, SA treatment of human melanocytes for 7 days at 2 to 5 µM did not show 799 significant cell growth attenuation ( Figure 6P). These data suggest that cytosolic EZH2/IMPDH2 interaction is a novel druggable target in these cancers.

824
Collectively, these data support a model in which nuclear EZH2 mediates 825 transcriptional silencing and concomitantly, cytosolic EZH2 contributes to the 826 activation of IMPDH2 mediated GTP synthesis that facilitates ribosome biogenesis 827 and actomyosin contractility contributing to melanoma progression and metastasis 828 ( Figure 7L). We propose that the molecular mechanism through which EZH2 activates 829 IMPDH2 mediated GTP synthesis is methyltransferase independent and involves 830 IMPDH2 cytosolic sequestration and tetramerization. Through an unbiased proteomics approach, in this study we uncovered 858 methyltransferase-independent binding partners of EZH2 in multiple melanoma cells.

859
Among the top EZH2-binding interactors is IMPDH2, the critical enzyme in de novo 860 GTP synthesis. In this study, we also showed that EZH2/IMPDH2 interaction is shuttling has remained unexplored in this study.

880
Nuclear Drosophila IMPDH has been shown to be accumulated during the G2 881 phase of the cell cycle or following replicative/ oxidative stress and to be bound to