Wnt signaling is conserved from invertebrates to vertebrates and regulates early embryonic development, as well as the homeostasis of adult tissues; as a central pathway in both physiological processes, dysregulation of Wnt signaling is associated with many human diseases, and particularly with cancer . Recently, Wnt signaling has also been implicated in hematopoiesis, both in self-renewal and in differentiation [1, 10, 18]. Based on these observations, it is hypothesized that dysregulation of the WNT pathway might contribute to the pathogenesis of lymphoproliferative diseases .
Despite the modest number of reports on the potential roles of Wnt signaling in leukemia, it is increasingly clear that Wnt signaling is dysregulated in several types of leukemia [12, 18, 27]. Some of these findings involve over- or underexpression of several Wnt ligands or Frizzled receptors [16, 19, 28, 29], hypermethylation of natural WNT inhibitors , and overexpression of β-catenin .
Despite this knowledge, there are a very limited number of publications on the expression of WNT7A and its role in the biology of blood cells. One of the first observations of the implication of WNT7A in hematological disorders was the frequent deletion of the 3p25 chromosome band observed in patients with AML, CML, and ALL . As is known, WNT7A is also localized at this chromosomal region [32, 33] and its deletion could be an important step during the neoplastic transformation.
In this paper, we report the expression of WNT7A in normal peripheral T-lymphocytes and strongly reduced WNT7A expression, not only in leukemia-derived cell lines, but also in the peripheral blood cells of patients with leukemia.
We were able to demonstrate that T-lymphocytes, but not B-lymphocytes, express WNT7A (ΔCP 11.47 ± 1.2). In agreement with this observation, Lu et al. also found expression of this ligand in peripheral blood lymphocytes (ΔCP 11.81 ± 0.99), but do not determined that this expression was afforded mainly from T-cells . In contrast, Sercan et al. found WNT7A expression in both T- and B-cells obtained from healthy volunteers . Discrepancies in these data could be due to the different method employed for quantification. The previously mentioned research group quantified WNT7A expression by comparing the densities of amplified WNT7A and β-actin PCR products visualized on agarose gels, while our group did this by performing qRT-PCR assays, which afford very precise data for quantification analysis.
We found from 38- to 500-fold lower expression in leukemia-derived cell lines than in healthy control cells (see Figure 2B). These results are in agreement as reported recently by Sercan et al., in which they did not find WNT7A expression in leukemia-derived cell lines K562, HL60, Jurkat, and Namalwa . However, the authors measured qualitatively, while we determined WNT7A expression quantitatively.
On the other hand, expression of WNT7A in hematological diseases has been only determined in patients with CLL and AML. Memarian et al. observed reduced expression of WNT7A in Iranian patients with AML compared with normal subjects ; however, the authors did not find this difference in patients with CLL . It is noteworthy that in both of these previously mentioned reports, WNT7A expression was calculated using the band densities of WNT7A and β-actin after conventional PCR. In agreement with the results of Memarian et al. we also observed reduced expression of WNT7A also in patients with AML, but statistical significance was not reached, probably due to the low number of patients with AML whom we analyzed (data not shown). Regarding expression of this ligand in patients with CLL; Lu et al. also observed lower WNT7A expression in patients with CLL (ΔCP 15.43 ± 2.94) when compared with healthy peripheral blood lymphocytes (11.81 ± 0.99). In this sense, we also observed this behavior in 4 out of 5 CLL patients (ΔCP 16.3 ± 1.5). Despite this low number of CLL patients, we found a statistical significance of p ≤0.02 when compared with healthy control cells (ΔCP 11.47 ± 1.2) (data not shown).
Interestingly, when we analyzed peripheral blood cells from 14 patients with ALL, these also expressed reduced WNT7A expression (ΔCP 15.19 ± 2.5) and we found a statistically significant difference of p ≤0.001 (GAPDH) and p ≤0.003 (RPL32) when compared with the control group (Figure 3).
Another important observation that we discerned is that WNT7A decreases acutely after PHA activation. To our knowledge, this is the first report evidencing that lymphocytes require reduction of their WNT7A levels in order to proliferate and suggests that dysregulation in the expression of this ligand needs to occur during oncogenesis to lose control of cell proliferation. Interestingly, it has been reported that T-cell activation by phytohemagglutinin results in a strong increase of phosphorylated GSK3β , which in turn targets beta-catenin for ubiquitylation and proteasomal degradation .
With respect to the reference genes used in the qRT-PCR assays, it is important to mention that there are no perfect reference genes for every treatment in every cell line. Thus, we used at least two reference genes in each assay and also evaluated some samples with a total of five reference genes (please see Additional Files 1 and 2). It has been determined that one of the reference genes that we used (RPS18) is useful as internal control for quantitative PCR in human lymphoblastoid cells, because constant levels of expression across the cell lines used were found following exposure to ionizing radiation as well as to PHA . However, it could be that some, but not all, of the changes in WNT7A expression may be caused by changes in reference-gene expression when cells were treated with PHA.
To our knowledge, no other papers relating WNT7A and leukemia have been published; however, reduced or absent expression of WNT7A has also been observed in lung cancer when compared with normal lung and mortal, short-term bronchial epithelial culture by qRT-PCR assay [36, 37].
Furthermore, it has been reported that WNT7a activates E-cadherin expression in lung cancer cells and that WNT7A loss may be important in lung cancer development or in progression due to its effects on E-cadherin, because E-cadherin in cancer has been associated with dedifferentiation, invasion, and metastasis . In addition to the role of WNT7A observed in leukemia and lung cancer, disruptions or alterations of the WNT7A gene have also been found in oral premalignant lesions  and in esophageal squamous cells .
We were also able to demonstrate, in the Jurkat leukemia-derived cell line, that restoration of WNT7A (by lentiviral overexpression or the addition of human recombinant protein) inhibits cell proliferation (Figure 5). Moreover, with the inducible-lentiviral overexpression system, we also confirmed this observation in K562, BJAB, and CEM cells after WNT7a expression; however, induction of cell death was not observed (Figure 8). Spinsanti et al. also reported an anti-proliferative action of WNT7a expression in undifferentiated PC12 cells . Additionally, recent studies have demonstrated that the combined expression of WNT7A and Frizzled 9 (Fzd9) in Non-small cell lung cancer (NSCLC) cell lines inhibits transformed growth by activating ERK5 and increasing PPARgamma activity, representing a novel tumor suppressor pathway in lung cancer [36, 42, 43]. However, the biological role of WNT7A action in cancer is controversial at present; some evidence supports its activity as an oncogene, but there is also evidence of its tumor suppressor action [36, 44, 45]. This dual role can be explained by the FZD proteins that bind WNT7a. It has been reported that the binding of WTN7a and FZD5 induces the canonical pathway, which has been related with cancer development [46, 47]. On the other hand, WTN7a can also bind FZD-10 and -9, which in turn activated the c-Jun NH2-terminal kinase pathway (JNK). Activation of JNK has been shown to antagonize the canonical pathway .
Due to the reported dual behavior of WNT7A as a cell-proliferation inducer or blocker, it is reasonable to think that Jurkat cells preferentially express anti-proliferative FZD partners of WNT7A. To address this question, we analyzed the presence of FZD mRNAs in Jurkat cells compared with T-lymphocytes from healthy controls and found overexpression of FZD-3 and -6 and downmodulation of FZD-5 and -10 in Jurkat cells (data not shown). On analyzing hematopoietic cells and leukemia-derived cells, Sercan et al. also found expression of FZD-3 and -6 in leukemia-derived T-lymphocytes . The presence of FZD-6 in lymphocytes is interesting, because it has been shown that FZD-6 can act as a negative regulator of the canonical pathway . Whether FZD-6 can interact with WTN7a in lymphocytes and what the biological consequences of this interaction would be are questions that remain open.
It has been observed that increased expression of some WNT ligands such as WNT3a, induces activation of the canonical pathway, accompanied by an increase in the proliferation and survival of leukemia cells . In addition, it has been reported that β-catenin comprises an integral part of AML cell proliferation and cell cycle progression . Because we observed downmodulation of WNT7A in leukemia-derived cells, it appears that WNT7a in Jurkat cells does not activate the WNT/β-catenin pathway. Evidence that supports this notion is the finding that mRNA levels of the putative canonical target genes AXIN2, MYC, JUN, and FRA-1 were not increased after WNT7a restoration (see Figure 6). In contrast, mRNA from JUN and FRA-1 were strongly downregulated in WNT7A-expressing cells, but again restored (FRA-1) or even upregulated (JUN) when cells were treated with LiCl (Figures 6A & 6B). Concerning this point, it has been reported that the β-catenin - T cell-factor/lymphoid-enhancer-factor complex directly interacts with the promoter region of JUN and FRA-1 . Because we observed restoration of the expression levels of JUN and FRA-1 after LiCl treatment, it is very probable that LiCl antagonize WNT7a activity in these cells. An additional observation that supports the idea that WNT7a is working in this model in a non-canonical pathway is that expression of β-catenin was not increased after WNT7a restoration (as seen in Figure 6C).