Clinical Significance of Phd Finger Protein 2 As a Novel Ikaros Target in Acute Lymphoblastic Leukemia

Background: PHD finger protein 2 (PHF2) mediates demethylation of dimethylated 'Lys-9' of histone H3 (H3K9me²), followed by transcription activation of target genes. PHF2 is a positive epigenetic regulator and is associated with tumor suppression in several types of cancer. Until now, there have been no reports about the PHF2 expression in leukemia patients and its roles in acute lymphoblastic leukemia (ALL).Genetic alterations of IKZF1 encoding the lymphoid transcription factor IKAROS are a hallmark of high risk B-progenitor ALL and are associated with poor outcome. However, it is still undetermined whether and how IKAROS regulates PHF2 expression. Methods: The 164 patients with newly-diagnosed ALL (107 B-ALL and 57 T-ALL; 12-77 years old) were recruited between June 2008 and June 2016 at the First Affiliated Hospital of Nanjing Medical University and Zhongda Hospital Southeast University. The study was approved by the Ethics Committee of the two hospitals. Patients were allocated into high or low PHF2 expression cohorts (4^th quartile vs . 1^st-3^rd quartiles) with a cut-off value determined by SPSS 20.0. The genomic amplification combined direct sequencing was used for detection of IKZF1 deletion. PHF2 expression was determined by qPCR in cells . The enrichment of IKAROS binding and H3K4me3 in the promoter of PHF2 was determined by the chromatin-immunoprecipitation assay combined qPCR (qChIP). IKZF1 knockdown was achieved in cells by transfection of human IKZF1 shRNA constructs in the GFP vector using the Neon Transfection System and with scrambled shRNA as a control. Median differences between the cohorts were evaluated using a Mann-Whitney U-test. Frequency differences were analyzed using univariate and multivariate Cox models. Relapse-free survival and overall survival were estimated by the Kaplan-Meier method and compared by log-rank test. Statistical analyses used SPSS version 20.0. Results: PHF2 mRNA levels in bone marrow samples from adults with ALL, especially those with B-cell ALL (B-ALL), were significantly lower than those in normals. The B-ALL cohort with low PHF2 mRNA levels was associated with a higher median bone marrow blast percentage (90.4% vs. 82.4%; P=0.001), a higher percentage of stem cell marker CD34+ (82.3% vs. 60.0%; P=0.021), myeloid marker CD33+ (48.5% vs. 25.0%; P=0.046), splenomegaly (46.8% vs. 23.1%, P=0.034) and delayed remission more than 4 weeks (48.7% vs. 19.0%; P=0.015), also a higher frequency of Ik6 (+) (46.3% vs. 8.0%, P=0.001) and a lower median PLT count in comparison to the high PHF2 expression cohort. Moreover, ChIP-seq data identified IKAROS binding peaks in the PHF2 promoter region in Nalm6 B-ALL and primary B-ALL cellsand the bindings were confirmed by the qChIP assay . IKAROS overexpression promotes PHF2 mRNA levels, conversely, IKZF1 knockdown decreased PHF2 expression in Nalm6 and CEM cells. These data indicate a direct effect of IKAROS on PHF2 transcription. The patients with IKZF1 deletion (Ik 6) have significantly lower PHF2 expression (0.03201 ± 0.000527 vs. 0.11331± 0.05773; P=0.033), which suggests that the IKZF1 deletion may contribute to low PHF2 expression in ALL patients. Treatment of Nalm6 and CEM cells with CK2 inhibitor CX-4945, which functions as IKAROS activator, promoted expression of PHF2 mRNA in ALL cell lines and primary cells and IKZF1 knockdown blocks this effect. CX-4945 treatment also increases the binding of IKAROS and H3K4me3 enrichment to the PHF2 promoter region in Nalm6 and CEM ALL cells as well as in primary ALL cells. Conclusion: PHF2 low expression was significantly associated with leukemia proliferation and several poor prognostic indicators in adult ALL patients. IKAROS directly regulates PHF2 transcription, and that CK2 inhibitor promotes PHF2 transcription through histone modification in an IKAROS dependent manner. Our data demonstrates the oncogenic effects of IKAROS dysfunction induced decreases in PHF2 expression. This is the first report to demonstrate that low PHF2 expression is significantly associated with Acute Lymphoblastic Leukemia. We suggest that PHF2 expression levelsmay be integrated into future adult ALL risk-stratification models.