Suppression of β‐catenin signaling in colon carcinoma cells by a bacterial protein

Colorectal cancer is one of the leading causes of cancer‐related death worldwide. The adenomatous polyposis coli (APC) gene is mutated in hereditary colorectal tumors and in more than 80% of sporadic colorectal tumors. APC mutations impair β‐catenin degradation, leading to its permanent stabilization and increased transcription of cancer‐driving target genes. In colon cancer, impairment of β‐catenin degradation leads to its cytoplasmic accumulation, nuclear translocation, and subsequent activation of tumor cell proliferation. Suppressing β‐catenin signaling in cancer cells therefore appears to be a promising strategy for new anticancer strategies. Recently, we discovered a novel Vibrio cholerae cytotoxin, motility‐associated killing factor A (MakA), that affects both invertebrate and vertebrate hosts. It promotes bacterial survival and proliferation in invertebrate predators but has unknown biological role(s) in mammalian hosts. Here, we report that MakA can cause lethality of tumor cells via induction of apoptosis. Interestingly, MakA exhibited potent cytotoxic activity, in particular against several tested cancer cell lines, while appearing less toxic toward nontransformed cells. MakA bound to the tumor cell surface became internalized into the endolysosomal compartment and induced leakage of endolysosomal membranes, causing cytosolic release of cathepsins and activation of proapoptotic proteins. In addition, MakA altered β‐catenin integrity in colon cancer cells, partly through a caspase‐ and proteasome‐dependent mechanism. Importantly, MakA inhibited β‐catenin‐mediated tumor cell proliferation. Remarkably, intratumor injection of MakA significantly reduced tumor development in a colon cancer murine solid tumor model. These data identify MakA as a novel candidate to be considered in new strategies for development of therapeutic agents against colon cancer.


| INTRODUCTION
Cancer is not a single disease but a set of complex events with distinct genetic and histological features. However, all cancers are characterized by several biological traits, among which uncontrolled cell proliferation is the most fundamental. 1,2 The observation that spontaneous tumor regression sometimes may occur in cases associated with microbial infection has inspired the idea that bacterial cells and/or products may have potential for development of new anticancer therapies, for example, as immunotherapeutic agents in cancer therapy. In 1890, the bone surgeon Dr William Coley, New York City, invented the "Coley's toxin" treatment. Coley noticed that cancer patients who came down with infections after surgical treatment seemed to do better than people who did not get infections. He suggested that infections could stimulate the immune system to also fight the cancer. 3 However, there is still lack of evidence to support the use of Coley's toxin for treatment or prevention of cancer. Nevertheless, several bacterial species are being explored as possible agents against cancer.
Initially developed as a vaccine against tuberculosis, the Bacillus Calmette-Guerin (BCG) is also a bacterial agent approved for treatment of bladder cancer since the late 1970s. 4 Among different types of cancer, colon cancer is considered the third most common type of cancer and the second leading cause of cancer-related death in industrialized countries. 5 Colon cancer originates from neoplastic transformation of epithelial cells of the colon.
Both environmental and genetic factors play important roles in the development of colon cancer. In particular, the Wnt signaling pathway has emerged as a key pathway for colon cancer tumorigenesis. 6 Wnt/β-catenin signaling is an evolutionarily conserved intracellular signaling pathway in several organisms ranging from worms to mammals and is responsible for regulating multiple steps during embryonic development, cell proliferation and differentiation. [7][8][9] Wnt signaling regulates cytoplasmic and nuclear levels of β-catenin. 10 The degradation of β-catenin occurs via a "destruction complex" that consists of different protein components, including the scaffold protein Axin, the tumor suppressor protein adenomatous polyposis coli (APC), glycogen synthase kinase (GSK3β) and casein kinase 1 (CK1). 11 Mutations in Wnt pathways trigger the development of multiple types of cancer including melanoma and colon cancer. 11 The familial forms of colon cancer are caused by a loss of APC function, often owing to truncating mutations in the gene. A mutation in the APC gene leads to dysfunction of the destruction complex, which is important for β-catenin degradation. This in turn leads to a reduction in β-catenin proteolysis, thus enhancing its nuclear translocation. In the nucleus, β-catenin binds to T-cell factor/lymphoid enhancer factor and contributes to activation of target genes, including VEGF, cyclin D1 and COX-2, thereby increasing tumor cell migration and proliferation. [12][13][14][15] Oncogenes such as β-catenin represent attractive molecular targets for the development of cancer therapy. Unlike, for example, inactivating mutations of p53, activation of β-catenin is a gain-of-function mutation. As such, it is feasible to use well-established drug development strategies to develop β-catenin signaling inhibitors. Extracellular Wnt inhibitors, including the secreted frizzled-related proteins, which act at the cell surface to inhibit Wnt signaling through its receptors, have been pursued as potential therapeutics. 16 Clinical treatment of established colon cancer remains a major challenge. Although there are several strategies available to combat colon cancer, including surgery, chemotherapy and radiation therapy, unfortunately, all come with major side effects. 17 Thus, there is an urgent need to find novel candidate anticancer molecules that can be developed as therapeutics to combat colon cancer. Among current approaches, studies of bacterial proteins and peptides have revealed promising bioactive molecules with anticancer treatment potential.
We recently identified the protein, motility-associated killing factor A (MakA) from Vibrio cholerae as a virulence factor against Caenorhabditis elegans and zebrafish, that is, acting in both invertebrate and vertebrate hosts. 18 In the present study, we investigated the effect of MakA on mammalian cells. Our data demonstrate that MakA readily targeted cancer cells and suppressed β-catenin signaling in colon carcinoma cells. Moreover, we demonstrated its potential as a therapeutic against cancerous tissue growth in a murine solid colon tumor model.

What's new?
While immune responses induced by bacterial infection may be leveraged to fight cancer, few bacterial proteins have been investigated for anticancer potential. Here, the bacterial protein and virulence factor MakA, from Vibrio cholerae, was examined for possible anticancer effects. In vitro, MakA was found to be lethal to tumor cells, exhibiting potent cytotoxic activity in different cancer cell lines, with less toxicity in non-transformed cells. In colon cancer cells specifically, MakA inhibited β-catenin-mediated tumor cell proliferation.
Moreover, in a murine colon cancer model, MakA intratumoral injection reduced tumor growth, suggesting that MakA is a viable candidate for novel anticancer strategies.

| Isolation of human and murine bone marrow-derived monocytes
For isolation of human monocytes, venous blood was collected from donors using the cell preparation tubes (CPT) and the peripheral blood mononuclear cells (PBMCs) were obtained according to the manufacturer's protocol (Becton Dickinson, NJ). For monocyte isolation, 1 × 10 8 to 2 × 10 8 PBMCs were plated in T-75 cell culture flasks and allowed to adhere in CO 2 incubator at 37 for 2 hours. Nonadherent cells were removed by washing twice with RPMI-1640. Adherent cells were harvested using a cell scraper, washed in phosphate-buffered saline (PBS) with 1% fetal calf serum (FCS) and resuspended in a complete media. For isolation of mouse monocytes, bone marrow cells were harvested from both femurs and tibias of C57BL/6 mice. Monocytes were isolated using an EasySep Mouse Monocytes Isolation Kit (STEMCELL, Vancouver, BC, Canada). Viability of the cells was determined prior to subsequent functional assays.

| MakA purification and labeling
MakA was purified as previously described. 18 For cellular uptake, MakA was labeled with Alexa Fluor 568 using an Alexa Fluor 568 protein labeling kit (Thermo Fisher) according to the manufacturer's instructions.

| Hemolytic assay
The hemolytic activity of MakA was determined by measuring the release of hemoglobin from lysed red blood cells (RBCs). A 1% (w/v) erythrocyte suspension was prepared from human blood by washing with PBS repeatedly by centrifugation at 1200g for 5 minutes and resuspending cells in PBS, pH 7.4. Erythrocyte suspension was treated with MakA (2 μM) in a 96-well plate and incubated at 37 C for 24 hours.
The Triton X-100 (0.1%) was used as the positive control and PBS was the negative control. Images were captured at the end of the treatment.

| Clonogenic and tumor cell migration assay
To assess the effect of MakA on tumor cell proliferation, a clonogenic assay was performed. HCT8 cells were seeded in a 24-well plate (500 cells/well) overnight and treated with increasing concentrations of MakA (125-500 nM) for 10 days, fixed (4% PFA, 30 minutes) and stained with crystal violet. The number of colonies (>50 cells) was assessed in triplicate wells. A dose-dependent decrease in tumor cell migration was quantified by transwell migration/invasion assay as previously described. 20 The number of migrated cells was counted after staining with Hoechst 33342. Fluorescence and bright-field images were captured with a fluorescence microscope (Nikon, Eclipse Ti).
Images were processed using the NIS-Elements (Nikon) and ImageJ software.
F I G U R E 1 Legend on next page. Nuclei were counterstained with DAPI (10 minutes, RT) and slides were mounted with Fluoromount aqueous mounting medium (Sigma), examined using an EZC1 Eclipse laser scanning confocal microscope (Nikon), using a 63×/1.4 plan Apo λs lens. Images were captured and processed using the NIS-Elements (Nikon) and ImageJ software.

| Statistical analysis
Results are presented as mean ± SD. Statistical analysis was performed using Student's t-test, one-way or two-way analysis of variance (ANOVA) or the Mann-Whitney test at different statistical levels of significance: *P < .05 and **P < .01.

| MakA induces cell death in cancer cells
MakA has recently been identified as a putative virulence factor in V.
cholerae with a role in defense against predators as manifested in studies with C. elegans as model host organism. 18

| MakA induces apoptosis in colon carcinoma cells
The process of apoptosis, also referred to as programmed cell death, is characterized by highly complex and sophisticated molecular mechanisms, involving an energy-dependent cascade of molecular events. 21 Flow cytometry analysis following Annexin V/PI staining was employed to monitor the occurrence of apoptotic cells. The results F I G U R E 1 MakA induces apoptosis in tumor cells. A, MakA kills colon carcinoma cells (HCT8 and DLD1) in a concentration-dependent manner. Loss of cell viability was measured by decrease in MTS absorbance. Mean ± SD of three independent experiments; two-way analysis of variance (ANOVA) with Tukey's multiple comparison test (*P ≤ .05, **P ≤ .01, ns = no significant difference). B,C, Vehicle-or MakA-treated HCT8 cells were labeled with Annexin-FITC or propidium iodide followed by flow cytometric analysis. Statistical significance was calculated by twotailed t-test (*P ≤ .05, **P ≤ .01). C, Quantification of flow cytometric data from (b). Data are expressed as means ± SD of three independent experiments. D, Western blot analysis of HCT8 cells treated with increasing concentration of MakA. Actin was used as a loading control. The figure shows results of one representative experiment. E, Histograms indicating quantification of Western blots (D) for Bax and active caspase 3 normalized against actin. The Bax expression is presented as relative to control (Ct) vehicle. Data are from three (Bax expression) or two (active caspase 3 expression) independent experiments; bar graphs show mean ± SD. Significance was determined using a one-way ANOVA with Dunnett's post-test against vehicle control (Ct). *P ≤ .05 or ns = not significant. F, Accumulation of Bax (green) at the mitochondria of HCT8 cell undergoing apoptosis in response to MakA. DAPI was used as a nuclear marker. Scale bar = 10 μm. G, MakA-induced HCT8 cell apoptosis was confirmed by an increase in number of active caspase-3-positive cells as detected by FAM-FLICA Caspase-3 assay. Scale bar = 20 μm. H, zVAD, a pan-caspase inhibitor blocked MakA-induced cell death as quantified by Presto Blue cell viability assay. Mean ± SD of four independent experiments; two-way ANOVA with Sidak's multiple comparison test (*P ≤ .05, **P ≤ .01). I, Flow cytometry-based apoptosis assay showed inhibition of apoptotic population in response to MakA in the presence of zVAD, a pan-caspase inhibitor. Statistical significance was calculated by two-tailed t-test (**P ≤ .01). J, Quantification of flow cytometry data from (I)  kinase 3β (GSK3β). 24 The β-catenin protein is a key molecule of the canonical Wnt/β-catenin signaling pathway, which has oncogenic capacity leading to development of human colon cancers. 6  TCF/LEF reporter activity was quantified using the TOP-flash luciferase assay as previously described. 32 MakA caused a dose-dependent decrease in the activity of β-catenin ( Figure 5H). Taken together, these results suggest that MakA colocalizes with β-catenin and inhibits its transcriptional activity in colon cancer cells.

| MakA is therapeutically active against colon cancer
β-catenin plays an important role in tumor cell proliferation, a predominant feature of tumor cells. 6 The findings that MakA caused β-catenin fragmentation and a decrease in its total expression prompted us to investigate further the effect of MakA on tumor cell proliferation.
MakA caused a dose-dependent decrease in tumor cell proliferation, as observed by a decrease in the number of tumor cell colonies ( Figure 5I). To assess if this decrease is due to cell death or a change in the cell proliferation-related proteome, we treated colon carcinoma cells with MakA and stained for the cell proliferation marker Ki67. A significant (P ≤ .01) increase in the number of Ki67 negative cells was observed in colon carcinoma (HCT8 and DLD1) cells treated with MakA ( Figure 5J and Figure S6C,D). Importantly, MakA also caused reduction in the expression of VEGF ( Figure 5C) involved in colon cell migration and invasion. 33 Therefore, we aimed to investigate if MakA reduces tumor cell migration. Indeed, we observed a MakA-mediated, dose-dependent decrease in tumor cell migration of HCT8 cells ( Figure 5K). We also tested the effect of MakA on primary cells, such as human red blood cells and monocytes and murine monocytes ( Figure 5L and Figure S6A).
MakA caused no detectable cytolytic effect on these primary cells ( Figure S7A). Furthermore, the nontransformed colon cells  Recently, it was reported that FadA selectively stimulates the growth of colorectal cancerous cells through activating Annexin A1 (ANXA1), a member of the Annexin family of Ca 2+ -dependent phospholipidbinding proteins. 36 The AvrA protein, a type III secretion system effector protein from S Typhimurium, also activates β-catenin signals and enhances colonic tumorigenesis. 37 Earlier studies also showed that H pylori infection can promote cancer stem cell characteristics in gastric cancer cells by activating Wnt/β-catenin signaling in a process dependent on the cytotoxin CagA protein. 38  We propose that this colocalization may directly regulate the exposure of β-catenin to the proteasome, which leads to proteasomemediated β-catenin degradation.
Increased knowledge of the genetic and epigenetic alterations of Wnt/β-catenin signaling involved in progression of human cancer has led to several approaches targeting Wnt/β-catenin signaling as a means to develop cancer therapies. 40 Use of bacteria as anticancer agents through enhancing human immunity has been described in several studies. [41][42][43][44] Prodigiosin, a natural red pigment produced by Serratia marcescens, has exhibited promising anticancer activity through an unknown mechanism. 45 Recently, it was demonstrated that Obatoclax, a synthetic prodigiosin analog, inhibits Wnt/β-catenin signaling and reduces cyclin D1 levels, suggesting potential as a therapeutic in advanced breast cancers. 46 Furthermore, prodigiosin produced by Serratia marcescens subsp. lawsoniana, also known as Chamaecyparis lawsoniana, was reported to have anticancer activities when used in human choriocarcinoma (JEG3) and prostate cancer cell lines (PC3) in vitro, as well as in JEG3 and PC3 tumor-bearing nude mice in vivo. 47 Several pharmacologically active small molecular antagonists of the protein-protein interaction between the transcription factor β-catenin and TCF have shown promising results in preclinical settings. 48 In addition to these approaches, several studies have linked the protective effect of nonsteroidal anti-inflammatory drugs on colorectal cancer with molecular pathways linked to nuclear inhibition of β-catenin. 49,50 Recently, it has been shown that a small molecule, MSAB, binds to β-catenin and stimulates its proteasomal degradation, therefore suppressing oncogenic Wnt/β-catenin signaling. 51 In addition, nonsteroidal anti-inflammatory drugs such as indomethacin disrupt lysosomal function, which may contribute to reduced β-catenin/ TCF signaling activity in colon carcinoma cells. 52 Figure 6H). To the best of our knowledge, MakA is the first bacterial protein reported to lower β-catenin levels. Taken together, our results suggest that the MakA protein may be considered a novel candidate for development of new therapeutic strategies against colon cancer.