WNT5A is epigenetically silenced in hematologic malignancies and inhibits leukemia cell growth as a tumor suppressor

To the editor:

Aberrant activation of the Wnt/β-catenin signaling pathway is a hallmark of cancers.^1,2  A recent Blood report showed the down-regulation of multiple Wnt-signaling inhibitors including SFRPs, WIF1, DKK3, and DACT1 in acute lymphoblastic leukemia (ALL) cell lines and tumors due to promoter methylation, leading to the activation of Wnt signaling.³ 

WNT5A is another noncanonical and nontransforming Wnt protein.⁴  Although there are conflicting data regarding its expression and functions in tumorigenesis, WNT5A signals through the noncanonical Wnt/Ca⁺⁺ pathway to block the Wnt-signaling cascade and inhibits B-cell proliferation.⁵  Its expression was down-regulated in most acute leukemia patients.⁵ 

We identified WNT5A as an up-regulated gene in nasal NK/T-cell lymphoma (NL) after 5-aza-2′-deoxycitidine (Aza) global demethylation and microarray expression analysis (Figure 1A), indicating that WNT5A is a methylation-silenced gene. Semiquantitative reverse transcriptase–polymerase chain reaction (RT-PCR) showed its down-regulation or silencing in multiple hematologic tumor cell lines including NL (4/4), leukemia (4/4), and Burkitt lymphoma (BL; 6/6) but not in normal lymph node (LN), peripheral blood mononuclear cell (PBMC), and lymphoblastoid cell lines (LCLs; Figure 1C). The WNT5A promoter is a typical CpG island (Figure 1B), thus subject to methylation silencing.^6,7  We analyzed its methylation by methylation-specific PCR (MSP) and detected methylation in all silenced cell lines (Figure 1C). Detailed methylation analysis with bisulfite genomic sequencing (BGS) confirmed the MSP analysis and revealed a high density of methylated CpG sites in silenced cell lines (Figure 1D). After exposure to Aza, WNT5A expression was dramatically increased in silenced cell lines (Figure 1C), further elucidating that methylation directly contributes to WNT5A silencing in tumor cell lines. MSP analysis also showed that WNT5A was methylated in 50% (5/10) of BL, 73% (22/30) of NL, and 31% (11/36) of other types of non-Hodgkin lymphoma but not in 6 normal lymph nodes, 9 PBMC samples, and normal NK cells. Down-regulation of WNT5A was detected in methylated BL samples (Figure 1E). These results demonstrated that WNT5A methylation is common and tumor specific in lymphomas.

Further functional studies showed that ectopic expression of WNT5A resulted in inhibition of tumor cell growth and clonogenicity (Figure 1F-H), consistent with previous studies that WNT5A inhibited B-cell proliferation and WNT5A heterozygous mice developed myeloid leukemias and B-cell lymphomas,⁵  suggesting that WNT5A could serve as a tumor suppressor.

Thus, we showed that WNT5A is frequently silenced by methylation in hematologic malignancies in a tumor-specific manner. Mutations of APC and epigenetic silencing of other Wnt pathway components including APC, SFRPs, DKKs, and WIF1 lead to the activation of canonical Wnt/β-catenin signaling. The noncanonical Wnt proteins (such as WNT5A), by activating alternative signaling pathways (such as the orphan tyrosine kinase ROR2), inhibit β-catenin stabilization^8,9  or induce Ca²⁺ flux to block canonical signaling downstream through inhibiting TCF-mediated transcription,^5,10  also preventing inappropriate signaling of the canonical Wnt pathway. WNT5A silencing might act as an additional mechanism to activate the canonical Wnt-signaling pathway. As targeting the Wnt/β-catenin signaling pathway is an attractive therapeutic strategy, our findings also indicate a possibility that WNT5A might be used as a therapeutic target.