A Possible Role for WNT5A Hypermethylation in Pediatric Acute Lymphoblastic Leukemia

Abstract

Objective: WNT5A is one of the most studied noncanonical WNT ligands and is shown to be deregulated in different tumor types. Our aim was to clarify whether hypermethylation might be the cause of low WNT5A mRNA levels and whether we could restore this downregulation by reversing the event.

Materials and methods: The expression of WNT5A mRNA was studied in a large acute lymphoblastic leukemia (ALL) patient group (n=86) by quantitative real-time PCR. The methylation status was detected by methylation-specific PCR (MSPCR) and bisulphate sequencing. In order to determine whether methylation has a direct effect on WNT5A expression, disease-representative cell lines were treated by 5'-aza-20-deoxycytidine.

Results: Here we designed a validation experiment of the WNT5A gene, which was previously examined and found to be differentially expressed by microarray study in 31 T-cell ALL patients. The expression levels were confirmed by quantitative real-time PCR and the expression levels were significantly lower in T-cell ALL patients than in control thymic subsets (p=0.007). MSPCR revealed that 86% of the patients were hypermethylated in the WNT5A promoter region. Jurkat and RPMI cell lines were treated with 5'-aza-20-deoxycytidine and WNT5A mRNA expression was restored after treatment.

Conclusion: According to our results, WNT5A hypermethylation does occur in ALL patients and it has a direct effect on mRNA expression. Our findings show that epigenetic changes of WNT signaling can play a role in ALL pathogenesis and reversing methylation might be useful as a possible treatment of leukemia.

Figure 1. The heat map diagram of WNT5A probes in T-ALL patients and controls. Gene and samples were clustered using Euclidean distance and complete linkage method for the probe sets in 31 T-ALL patients and thymocyte subsets as controls [CD4 single positive, CD8 single positive, CD4+CD8+ double positive, Thymus (total thymus tissue), DP3- (CD4+, CD8+ double positive CD3 negative, immature single positive and CD3-, CD4-, CD8-)]. Three probe sets for WNT5A are illustrated in the heat map analysis. Green color shows downregulation and red color shows upregulation of the targeted gene.

Figure 2. Relative WNT5A mRNA expression in B- and T-cell acute lymphoblastic leukemia patients. A) B-ALL patients’ samples were compared with healthy bone marrow samples and T-cell acute lymphoblastic leukemia patients’ were compared with healthy thymocytes (p=0.007). Each sample was studied in duplicate and threshold cycle numbers are relative to the geometric mean of 3 reference genes (ABL, β-actin, and CypA). B) Comparison of WNT5A mRNA levels between B-ALL and T-ALL samples (p<0.0001). Each sample was studied in duplicate and threshold cycle numbers are relative to the geometric mean of 3 reference genes (ABL, β-actin, and CypA).

Figure 3. MSPCR results in acute lymphoblastic leukemia patients. The samples were run on 4% agarose gel with ethidium bromide under UV light. PCR products were 178 bp long. Initial M: pUC Mix Marker 8, P: patient, M: methylated PCR, U: unmethylated PCR (patients 1, 2, 3, and 4 are methylated; patient 7 is slightly methylated; patients 5 and 6 are unmethylated), IVM: in vitro methylated control sample, Neg: negative control.

Figure 4. Aza-20-deoxycytidine (Aza) treatment experiments in cell lines. Cell lines Jurkat and RPMI 8402 were cultured for 4 days with Aza (Sigma-Aldrich, A3656) at 5 mM and 10 mM concentrations respectively in 24-well plates (5x105 cells/well) supplemented with 1 mL of RPMI 1640 medium with 10% FBS and 1% antibiotics (penicillin and streptomycin) at 37 °C in a humid atmosphere containing 5% CO2. A) MSPCR results after 0 mM, 5 mM, and 10 mM 5’-Aza treatment of RPMI 8402 cell line. B) MSPCR results after 0 mM, 5 mM, and 10 mM 5’-Aza treatment of Jurkat cell line. C) Relative WNT5A mRNA expression level after 5’-Aza treatment of RPMI 8402 cell line (5 mM, p=0.0043; 10 mM, p=0.0002). D) Relative WNT5A mRNA expression level after 5’-Aza treatment of Jurkat cell line (5 mM, p=0.004).

Figure 5. Kaplan-Meier estimate of probability of A) overall and B) event-free survival analysis.

Supplemental Figure 1. CpG island methylation status by bisulphate sequencing in ALL-derived cell lines. A) MSPCR results of cell lines, IVM: in vitro methylated control; B) bisulphate sequencing of Jurkat cell line, red bars show CpG islands; C) bisulphate sequencing of Fleb14-4 cell line, red bars show CpG islands.