Background: AT-rich interactive domain-containing protein 5B (ARID5B) forms a complex with PHD finger protein 2 (PHF2) and the ARID5B-PHF2 complex then induces the demethylation of di-methylated 'Lys-9' of histone H3 (H3K9me2) leading to transcription activation target genes. Mutations and single nucleotide polymorphisms of ARID5B are associated with the development of acute lymphoblastic leukemia (ALL) and treatment outcome. So far there have been no reports about ARID5B expression in ALL and the clinical significance of ARID5BlowPHF2low expression in ALL patients. Ikaros, the product of the IKZF1 gene, plays a critical role in lymphocyte development and functions as a tumor suppressor of leukemia. ChIP-seq data show the obvious Ikaros binding peaks in promoter of ARID5B, which suggested the regulatory role of Ikaros on ARID5B expression in ALL cells. Methods: 164 Subjects with newly-diagnosed ALL (107 B-cell and 57 T-cell ALL; 12-77 years old) between June 2008 and June 2016 were studied at the First Affiliated Hospital of Nanjing Medical University and Zhongda Hospital Southeast University. The study was approved by the Ethics Committee of the institutes. Subjects were allocated in a high or low ARID5B expression cohort (4th quartile vs . 1st-3rd quartiles) with a cut-off value determined by SPSS 20.0. The qChIP assays were performed by incubating chromatin with antibodies against Ikaros, H3K4me3 or normal rabbit IgG as a control. The Nalm6 and CEM cells were transfected with human IKZF1 shRNA or Scrambled shRNA constructs in the GFP vector (pGFP-v-RS) by Neon system. Median differences between the cohorts were evaluated using a Mann-Whitney U-test. Frequency differences were analyzed using uni- and multivariate Cox model. Relapse-free survival (RFS) and overall survival (OS) were estimated by the Kaplan-Meier method and compared by log-rank test. Results: We observed the ARID5B mRNA level is significantly down-regulated in subset of ALL patients. Low ARID5B mRNA levels were associated with leukemic cell proliferation and poor prognostic indicators. Particularly ARID5B expression is positively correlated with PHF2 expression in ALL. The ARID5BlowPHF2low was associated with a higher percentage of bone marrow blasts (91.2% vs. 82.4%, P =0.000), stem cell marker CD34+ (88.2% vs. 55.6%, P =0.000), myeloid marker CD33+ (50.9% vs. 28.6%, P =0.036), splenomegaly (50.0% vs. 22.9%, P =0.008), a complete remission(CR) time ≥4 weeks (53% vs. 21%, P =0.003) and higher frequency of Ik 6(+) (49.3% vs. 15.8%, P =0.001) and lower median PLT count (109/L) (32.0 vs. 58.5, P =0.020) compared to patients with non- ARID5BlowPHF2low cohort. The multivariable analysis confirmed the association with bone marrow blasts, Ik 6(+) and a CR time ≥4 weeks. ChIP-seq data shows Ikaros binding peaks in the ARID5B promoter region in Nalm6 B-ALL and primary B-cell ALL cells and the bindings wereconfirmed by individual qChIP assay. Moreover, expression of Ikaros promotes ARID5B expression, while efficient Ikaros knockdown decreased ARID5B expression in Nalm6 and CEM cells. We also found ARID5B expression was significantly lower in B-ALL with Ik6(+) (0.3153 ± 0.0938 vs. 1.2052 ± 0.58441, P =0.02439). Treatment of Nalm6 and CEM cells with CX-4945, which functions as IKAROS activator, promotes expression of ARID5B mRNA in a dose-dependent manner in ALL cell lines and patients' samples, and Ikaros knockdown significantly blocked this effect. Moreover,CX-4945 treatment significantly increases binding of Ikaros and enrichment of H3K4me3 to the ARID5B promoter region in Nalm6 and CEM ALL cells, as well as in patients' samples. Conclusion: We identified that ARID5Blow, particularly ARID5BlowPHF2low expression is associated with leukemia proliferation and poor prognostic markers. Our results also revealed the oncogenic effect of ARID5BlowPHF2low expression in ALL, and identified a high-risk subgroup of ALL characterized by Ikaros dysfunction and ARID5BlowPHF2low expression.

Disclosures

Dovat: ELF Zone Inc: Membership on an entity's Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution