Constitutive activation of the Wnt canonical pathway in mantle cell lymphoma

Abstract

Aberrations of the Wnt canonical pathway (WCP) are known to contribute to the pathogenesis of various types of cancer. We hypothesize that these defects may exist in mantle cell lymphoma (MCL). Both the upstream and downstream aspects of WCP were examined in MCL cell lines and tumors. Using WCP-specific oligonucleotide arrays, we found that MCL highly and consistently expressed Wnt3 and Wnt10. beta-catenin, a transcriptional factor that is a downstream target of WCP, is localized to the nucleus and transcriptionally active in all 3 MCL cell lines examined. By immunohistochemistry, 33 (52%) of 64 MCL tumors showed nuclear localization of beta-catenin, which significantly correlated with the expression of the phosphorylated/inactive form of GSK3beta (p-GSK3beta; P = .011, Fisher). GSK3beta inactivation is directly linked to WCP stimulation, since addition of recombinant sFRP proteins (a naturally occurring decoy for the Wnt receptors) resulted in a significant decrease in p-GSK3beta. Down-regulation of DvL-2 (an upstream signaling protein in WCP) by siRNA or selective inhibition of beta-catenin using quercetin significantly decreased cell growth in MCL cell lines. To conclude, WCP is constitutively activated in a subset of MCL and it appears to promote tumorigenesis in MCL.

Figure 1

Localization and in vitro activity of β-catenin in MCL cell lines. (A) Subcellular fractionation using the cell lysates of 3 MCL cell lines revealed that β-catenin was localized to the nucleus (N). The expression of α-tubulin in the cytoplasm (C) served as a control for the efficiency of subcellular fractionation. (B) Confocal microscopy revealed the nuclear accentuation of the β-catenin staining in MCL cells (bottom panel). The use of secondary antibody served only as negative controls (top panel). (C) The use of the TOP/FOP system confirmed that β-catenin is transcriptionally active in SP53 and Mino cells. Luciferase activity is expressed as arbitrary units (AU). Error bars indicate SD. Experiments were performed in triplicates and the differences are statistically significant (P < .05, Student t test).

Figure 2

Expression of β-catenin in MCL tumors. (A) Immunohistochemistry using paraffin-embedded MCL tumors showed nuclear staining of β-catenin in a positive case (left panel). Only weak cytoplasmic staining was found in the tumor cells in a negative case, although endothelial cells showed intense cytoplasmic staining (right panel). (B) Immunohistochemistry using paraffin-embedded MCL tumors showed intense cytoplasmic staining of p-GSK3β in a positive case (left panel), but no appreciable staining in a negative case (right panel).

Figure 3

Activation status of the WNT pathway in MCL. (A) Western blot studies revealed the strong expression of p-GSK3β in all 3 MCL cell lines. In addition, DvL-2 was highly expressed. Importantly, the slowly migrating/phosphorylated forms of DvL-2 were detected in all 3 cell lines. Mouse stem cell lysates were used as positive controls for the phosphorylated forms of DvL proteins. P indicates the phosphorylated form of DvL-2 and UnP indicates the unphosphorylated form of DvL-2. (B) Western blot studies revealed the expression pattern of p-GSK3β in 5 MCL primary tumors. P1, P3, and P5 had a relatively high level of p-GSK3β, which was associated with relatively a high level of DvL-2 and the presence of its slowly migrating forms. In contrast, P2 and P3 had weak or undetectable p-GSK3β, which correlated with a weak DvL-2 expression.

Figure 4

Effect of WNT pathway inhibition on GSF3β phosphorylation status. SP53 and Jeko-1 cells were treated with different concentrations of natural Wnt inhibitor: sFRP1 (A) or sFRP4 (B). Treatments with either of these 2 sFRP proteins led to a significant decrease in p-GSK3β. The bottom graph on each figure represents densitometry measurement of the p-GSK3β expression levels. Error bars indicate SD.

Figure 5

Biologic effect of WNT pathway inhibition by DvL-2 targeting.(A) Treatment of Jeko-1 cells with siRNA for DvL-2 for 24 hours showed a dramatic decrease in the protein expression of DvL-2. (B) Wnt pathway inhibition by DvL-2 induced a significant decrease in Jeko-1 cell growth as measured by MTS assay. Error bars indicate SD. (C) Treatments of Jeko-1 cells for 48 hours with DvL-2 siRNA demonstrate a dose-dependent cleavage of caspase-3.

Figure 6

Quercetin induce apoptosis in MCL. (A) MTS assay (left panel) and trypan blue exclusion test (right panel) were performed to assess the biologic effects of quercetin on MCL cell lines. All 3 cell lines showed a dose-dependent decrease in cell growth with quercetin treatment. Negative controls in all experiments were treated with the highest volume of DMSO used in the treated group. Results from the treated group are normalized to those of the negative controls. (B) Cell-cycle analysis by flow cytometry was performed using Mino cells with or without quercetin treatment. M1 represents the G_o/G₁ phase; M2, the S phase; M3/4, the G₂/M phase; and M5, the subG₀ apoptotic cell population. Compared with cells with quercetin treatment, treated cells showed a decrease in the proportion of cells in the S phase as well as the G₂/M phase. In addition, there was a dramatic increase in the size of the subG₀ cell population, in keeping with the occurrence of apoptosis. (C) All 3 MCL cell lines showed dose-dependent down-regulation of cyclin D1 after quercetin treatment at 24 hours. Negative controls were treated with DMSO in the same volumes used in the treatment group. (D) All 3 MCL cell lines showed dose-dependent down-regulation of Bcl-2 and Bcl-XL after quercetin treatment at 24 hours. Negative controls were treated with DMSO in the same volumes used in the treatment group. (E) All 3 MCL cell lines showed expression of cleaved caspases-3, -7, and -9, as well as PARP, after quercetin treatment at 24 hours. Negative controls were treated with DMSO in the same volumes used in the treatment group. Error bars indicate SD.

Figure 7

Treatment of Jeko-1 cells with quercetin had no detectable effect on the GSK3β phosphorylation status. Jeko-1 cells were treated with different concentrations of quercetin. Treatment with the diluent served only as the negative control.