Expression Profiling Of Selected Polycomb Complex Genes In Childhood Acute Myeloid Leukemia Revealed An Overexpression Of EZH2 In Acute Promyelocytic Leukemia

Acute promyelocytic leukemia (APL) in children and adolescents accounts for about 20% of cases of acute myeloid leukemia (AML) in Brazil. The reasons for the relatively high incidence of APL among children and adolescents in Brazil and other Latin countries remain elusive. Epigenetic constitutional and/or environmental factors might be implicated in the mechanism of APL. The Polycomb group (PcG) genes are critical for differentiation and cell-cycle regulation and maintenance of epigenetic memory of living organisms. Aberrant expression of PcG genes has been observed in human tumors, including AML. In this study, we sought to determine the expression levels of 4 genes from the PcG repressive complexes EZH2, YY1, BMI1 and SUZ12 in acohort of 52 children with AML or APL (male, 32; female, 20; median age 7.8 years, range 4 months-18 years).

Cells from healthy children (male, 2; female, 2; median age, 10.7 years, range 6-15 years) were used as the control group. Quantitative determination of mRNA levels was performed using Power SYBR Green PCR Master Mix® (Applied Biosystems, Foster City, CA, USA) in a Rotor Gene® thermocycler (QIAGEN). Expression levels were estimated in triplicate, and ß-actin was used as an internal control. All statistical analyses were performed using the GraphPad Prim 5.0 System. Multiple pairwise comparisons were made using a one-way analysis of variance (ANOVA) test; P<0.05 was considered statistically significant.

Despite showing broad variation among patients and controls, the expression levels of YY1 (controls, 1.16 ±SE 0.73; M1/M2, 0.63±1.07; M4/M5 0.217±0.71; APL, 0.62±3.24; p=0.19); SUZ12 (control, 1.11 ±0.93; M1/M2, 0.12±1.65; APL 2.4±3.42; M4/M5, 0.34 ± 10.30; p=0.18), and BMI1 (control, 0.041 ± 0.77; M1/M2: 0.043 ±0.10; M4/M5, 0.016± 0.23; APL, 0.12 ± 0.59; p=0.37)did not differsignificantly between controls and patients grouped according to the major AML subtypes (M1/M2, 15 cases, median age 10.8 years, range 4-18 years; M4/M5, 21 cases,median 4.5 years, range 5 months-18 years; APL 16 cases, median 9.8 years, range, 1-17 years). However, the EZH2 expression levels (control, 0.0008818±0.03675; M1/M2, 0.001495±0.03296; M4/M5 0.008215±0.09313; APL, 0.07180± 0.3402; p=0.0092) were significantly higher in APL. Among patients with APL, the expression levels of EZH2 did not differ significantly according to age (EZH2 expression, 0.23±0.44, age 0-8 years; 0.05±0.46, age 9-18 years; p=0.112), white blood cell count (EZH2 expression 0.07±0.65, WBC < 10 x 10⁹/L; 0.10±0.38, WBC ≥ 10 x 10⁹/L; p=0.148) or platelet count (EZH2 expression 0.20±0.49, platelet count < 40 x10⁹/L; 0.06 ±0.04, platelet count ≥ 40 x 10⁹/L; p=0.13).

Several studies have shown that EZH2 is deregulated in several human cancers. Here we extend this data by informing that EZH2 is highly overexpressed in pediatric APL. APL in children less than 8 years of age tends to be associated with higher EZH2 gene expression levels, suggesting that constitutional epigenetic factors may play a driver role in pediatric APL leukemogenesis.