Colorectal cancer exhibits aberrant activation of Wnt/β-catenin signaling. Many inhibitors of the Wnt/β-catenin pathway have been tested for Wnt-dependent cancers including colorectal cancer, but are unsuccessful due to severe adverse reactions. FL3 is a synthetic derivative of natural products called flavaglines, which exhibit anti-inflammatory and cytoprotective properties in intestinal epithelial cells, but has not been previously tested in cell or preclinical models of intestinal tumorigenesis. In vitro studies suggest that flavaglines target prohibitin 1 (PHB1) as a ligand, but this has not been established in the intestine. PHB1 is a highly conserved protein with diverse functions that depend on its posttranslational modifications and subcellular localization. Here, we demonstrate that FL3 combats intestinal tumorigenesis in the azoxymethane-dextran sodium sulfate and ApcMin/+ mouse models and in human colorectal cancer tumor organoids (tumoroids) by inhibiting Wnt/β-catenin signaling via induction of Axin1 expression. FL3 exhibited no change in cell viability in normal intestinal epithelial cells or human matched-normal colonoids. FL3 response was diminished in colorectal cancer cell lines and human colorectal cancer tumoroids harboring a mutation at S45 of β-catenin. PHB1 deficiency in mice or in human colorectal cancer tumoroids abolished FL3-induced expression of Axin1 and drove tumoroid death. In colorectal cancer cells, FL3 treatment blocked phosphorylation of PHB1 at Thr258, resulting in its nuclear translocation and binding to the Axin1 promoter. These results suggest that FL3 inhibits Wnt/β-catenin signaling via PHB1-dependent activation of Axin1. FL3, therefore, represents a novel compound that combats Wnt pathway–dependent cancers, such as colorectal cancer.

Significance:

Targeting of PHB1 by FL3 provides a novel mechanism to combat Wnt-driven cancers, with limited intestinal toxicity.

Patients with chronic inflammatory bowel disease (IBD) of the colon have increased risk for colorectal cancer, accounting for nearly 10%–15% of deaths in patients with IBD (1). In these cases, colorectal cancer develops as a consequence of complications brought on by chronic colonic inflammation and is referred to as colitis-associated colorectal cancer. Because of the well-established association of chronic inflammation with colorectal cancer development and malignant progression, use of anti-inflammatory agents to impede carcinogenesis is considered to be a highly effective strategy for colorectal cancer chemoprevention (2).

Flavaglines are a relatively new class of molecules derived from medicinal Aglaia (family Meliaceae) plants (3). Flavaglines have shown anticancer, anti-inflammatory, and cytoprotective activities in vitro (4). FL3 is a synthetic analogue of the flavagline rocaglaol, but lacking the structural components linked to causing MDR and suboptimal pharmacokinetics, making it an ideal candidate for further investigation (3, 5). Previous studies demonstrated that FL3 at nanomolar concentrations exhibits cytoprotective effects in neurons (6), cardiomyocytes (7), and intestinal epithelial cells (8) and presents little toxicity to healthy cells (3, 9, 10). Our previous data suggest that FL3 combats inflammation using a mouse model of colitis, although we did not elucidate the mechanism of FL3 action (8). In vitro studies suggest that flavaglines target prohibitins as ligands, but this has not been established in the intestine (3). Prohibitin 1 (PHB1) is a highly conserved protein with diverse functions, including regulation of cell-cycle progression, apoptosis, and transcription, depending on its posttranslational modifications and subcellular localization (11). In intestinal epithelial cells, PHB1 is predominantly localized in the mitochondria, where it has been shown to be required for optimal activity of the electron transport chain (12–15).

Accumulation of select DNA mutations in tumor suppressors and oncogenes in colonic epithelial cells drives colorectal cancer, including colitis-associated colorectal cancer, initiation, and progression (16). The majority of colorectal cancers, including sporadic (up to 80%) and inflammation induced (∼50%), carry a genetic mutation in either Adenomatous polyposis coli (APC) or CTNNB1 (encoding β-catenin), resulting in aberrant Wnt activation (17, 18). Although Wnt/β-catenin pathway mutations tend to occur late in colitis-associated colorectal cancer development, studies suggest that activation of the Wnt/β-catenin pathway by numerous inflammatory signaling networks contributes to the onset and progression of colitis-associated colorectal cancer (19). APC suppresses tumors due to its scaffolding role in the “β-catenin destruction complex,” which is a key node in regulating Wnt signaling (20). AXIN1 and AXIN2 are pivotal scaffold proteins that coordinate the assembly of APC, kinases CK1α and GSK3β, and their substrate β-catenin to facilitate the phosphorylation and subsequent proteasomal degradation of β-catenin under basal conditions. During Wnt stimulation, the destruction complex is inhibited, resulting in accumulation and nuclear translocation of β-catenin and transcriptional activation by binding to TCF/LEF sites on Wnt target genes that control proliferation, survival, and migration (21). Alterations in the APC gene found in colorectal cancer generate truncated mutants lacking all binding sites for AXIN and abolish the formation of the β-catenin destruction complex. Current therapeutics cannot directly target APC. Colorectal cancer mutations in CTNNB1 render β-catenin resistant to phosphorylation and proteasomal degradation, leading to overexpression of β-catenin and Wnt target genes (22).

Many inhibitors of the Wnt/β-catenin pathway have been characterized, with some reaching early clinical trials (23). However, current Wnt-targeting compounds exhibit adverse reactions because Wnt signaling regulates homeostasis of many adult tissues, with the intestine being extremely vulnerable. For this reason, advancement of these compounds in clinical trials is impeded (24). Here, using human colorectal cancer cell lines with or without mutations in the CTNNB1 gene, two mouse models of intestinal tumorigenesis, and mouse and human colorectal cancer tumoroids, we demonstrate that the targeting of PHB1 by FL3 provides a novel mechanism to inhibit aberrant Wnt signaling without intestinal toxicity.

Description of histologic dysplasia/adenoma scoring, SDS-PAGE, Western immunoblot analysis, cell viability assay, MTT proliferation assay, cell migration assay, RNA isolation, qRT-PCR analysis, TCF/LEF luciferase reporter, terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) staining, immunofluorescence and IHC, chromatin immunoprecipitation (ChIP), and RNA-sequencing (RNA-seq) analysis is provided in the Supplementary Materials and Methods.

FL3 synthesis

FL3 was synthesized in our laboratory as described previously (25).

Animal models

All mice were grouped-housed in standard cages under a controlled temperature (25°C) and photoperiod (12-hour light/dark cycle) and were allowed standard chow and tap water ad libitum. All experiments were approved by the Baylor Research Institute (Dallas, TX) and the University of Colorado School of Medicine (Aurora, CO) Institutional Animal Care and Use Committees. Experiments were performed with age- and gender-matched littermate mice.

AOM-DSS model: age-matched male and female wild-type (WT) C57Bl/6 mice (The Jackson Laboratory), Phb1fl/fl mice (C57Bl/6 genetic background), and Phb1iΔIEC mice (C57Bl/6 background and described previously; ref. 26) at 8 weeks old were intraperitoneally injected with 7.6 mg/kg azoxymethane (AOM; Sigma-Aldrich), followed by two cycles of 3.0% (w/v) dextran sodium sulfate (DSS; molecular weight, 50,000; MP Biomedicals) in drinking water for 7 days followed by 14 or 21 days of recovery (regular drinking water). One day before and each day of dextran sodium sulfate administration, mice were intraperitoneally injected daily with 0.1 mg/kg FL3 or vehicle (100 μL volume). Upon sacrifice, colon was excised from the ileocecal junction to anus, cut open longitudinally, neoplasia number and size were quantitated using a dissecting microscope, and prepared for histologic evaluation or biochemical analyses.

ApcMin/+ model: age-matched male and female ApcMin/+ mice (The Jackson Laboratory) and WT C57BL/6 littermates were intraperitoneally injected with 0.1 mg/kg FL3 or vehicle (100 μL volume) twice per week starting at 15 weeks to 20 weeks. Upon sacrifice, the entire small intestine and colon were excised, cut open longitudinally, neoplasia number and size were quantitated using a dissecting microscope, and prepared for histologic evaluation or biochemical analyzes.

Human colorectal cancer and matched-normal specimens

Fresh, deidentified specimens of colorectal cancer, stage II and III, and matched-normal colon tissue were obtained from Department of Pathology, Baylor University Medical Center (Dallas, TX), which obtained written informed consent from patients undergoing resection surgery. The studies were conducted in accordance with the Belmont Report ethical principles and under institutional review board approval. Tissue specimens were stored in Advanced DMEM/F12 supplemented with penicillin/streptomycin until culturing, which averaged 2–8 hours later.

Human or mouse colorectal cancer tumoroid and colonoid culture

Colorectal cancer tumoroids were derived from tumors digested in 125 μg/mL dispase type II, 75 U/mL collagenase type IX, and 0.2% Primocen. Colonoids were derived from isolated colonic crypts from mice or human specimens and cultured in Matrigel Basement Membrane Matrix (Corning) as described previously (27). Colonoids were grown in WENR Media (Advanced DMEM/F12 supplemented with penicillin/streptomycin, 10 mmol/L HEPES, 1 × l-glutamine, 1 × N2, 1 × B27, all from Life Technologies), containing 100 ng/mL Wnt3α, 50 ng/mL recombinant EGF, 500 ng/mL R-spondin1, and 100 ng/mL Noggin (all from R&D Systems). Tumoroids were grown in WNR Medium (containing Wnt3α, Noggin, and R-spondin1) because colorectal cancer cells require no addition of growth factors such as EGF (27). Tumoroids and colonoids were grown for 7 days and treated with 10 or 50 nmol/L FL3. For collection of protein lysates for Western blot analysis, tumoroids were collected using Cultrex Organoid Harvesting Solution (Thermo Fisher Scientific) to remove Matrigel.

To determine whether human tumoroids exhibited deletion at S45 of β-catenin, genomic DNA was isolated from cultured tumoroids using the DNeasy Kit (Qiagen), 100 ng genomic DNA was amplified by PCR using the following primers: sense: 5′-CAATGGGTCATATCACAGATTCTT-3′; antisense: 5′-TCTCTTTTCTTCACCACAACATTT-3′, and the PCR products were gel-purified and sequenced.

For transfection of sgPHB1, human tumoroids grown in Matrigel were trypsinized into single cells and Nucleofected with human prohibitin 1 CRISPR/Cas9 KO Plasmid (sc-416271, Santa Cruz Biotechnology) and human Prohibitin 1 HDR Plasmid (sc-416271-HDR, Santa Cruz Biotechnology) using Hepatocyte Nucleofector Kit (VPL-1004, Lonza) for 72 hours.

Cell culture

RKO (CRL-2577), HCT116 (CCL-247), SW48, and IEC6 cells were acquired from the ATCC in 2017. Cells were cultured in 1× DMEM supplemented with 10% FBS, 40 mg/L penicillin, and 90 mg/L streptomycin. Cells were maintained in an incubator with 5% CO2 at 37°C. All experiments were performed on RKO cells between passages 5 and 19, HCT116 between passages 7 and 12, SW48 between passages 5 and 9, and IEC6 between passages 19 and 26. All cell lines were verified to be Mycoplasma free using Genlantis MycoScope PCR Detection Kit (Thermo Fisher Scientific) as recently as March 15, 2020 (RKO, HCT116, and IEC6) or April 7, 2020 (SW48).

To generate RKO cells stably expressing β-catenin harboring deletion of S45, RKO cells were transfected with Pcl-neo-β-catenin-Δ45 (Addgene) using Nucleofector T Kit (Lonza). Forty-eight hours following transfection, transfected cells were selected with 1,000 μg/mL of G418 (Geneticin; Gibco) for 3–5 weeks. Single clones were then isolated, maintained under 200 μg/mL G418 selection, and DNA was sequenced for validation of exogenous β-catenin-Δ45 expression using the following primers: sense: 5′-ATGGCCATGGAACCAGACAG-3′; antisense: 5′-CTGAGAAAATCCCTGTTCCCAC-3′. Because these cell lines still express WT endogenous β-catenin, sequencing confirmed heterozygous S45 deletion. To knockdown AXIN1 expression, RKO cells were nucleofected with 20 μmol/L of three pooled unique 27mer siRNA duplexes against AXIN1 (SR305433; OriGene) or 20 μmol/L Stealth RNAi nonspecific Negative Control Med GC (siNC; Invitrogen).

Statistical analysis

Data are presented as individual data points ± SEM. An unpaired two-tailed Student t test was used for single comparisons, one-way ANOVA with Bonferroni post hoc test for multiple comparisons, and two-way ANOVA with Bonferroni post hoc test for assessing the combination of AXIN1 knockdown and FL3 treatment (GraphPad Prism 8.0). For RNA-seq analysis, genes were filtered for counts per million ≥ 1 and differential gene expression was determined using R package DESeq2 (Bioconductor), P < 0.05, FDR ≤ 0.05, and log2 fold change >0.32. Ingenuity Pathway Analysis (Qiagen) identified significantly altered pathways.

FL3 decreases AOM-DSS–induced colonic tumorigenesis

To induce colitis-associated colorectal cancer in mice, WT C57BL/6 mice were intraperitoneally injected with azoxymethane, followed by two cycles of dextran sodium sulfate, consisting of 3.0% (w/v) dextran sodium sulfate given in their drinking water for 7 days and 14 or 21 days recovery (regular drinking water). One day prior to and during each cycle of dextran sodium sulfate, mice were intraperitoneally injected daily with 0.1 mg/kg FL3 or vehicle. FL3 decreased the number and size of AOM-DSS–induced colonic neoplasia (Fig. 1A and B) and decreased the histologic occurrence of adenocarcinoma (Fig. 1C and D), suggesting that FL3 prevents the progression of colitis to cancer. Importantly, the mice displayed no overt signs of toxicity or abnormal behavior from FL3 treatment. Colon tumoroids cultured from 3 independent AOM-DSS–treated mice and treated with FL3 at a low nanomolar concentration (10 nmol/L) exhibited decreased viability and increased apoptosis (Fig. 1E and F; Supplementary Fig. S1A). RNA-seq analysis of AOM-DSS colon tumoroids treated with FL3 followed by ingenuity pathway analysis identified Wnt/β-catenin signaling as the top canonical pathway altered by FL3 (P = 4.3E-04; Fig. 1G). Of the genes identified by RNA-seq, the expression of Axin1 was most significantly altered by FL3 (P = 1.6E-08), followed by Smek1 (P = 4.6E-07), Tle4 (P = 2.8E-07), and Klf4 (P = 2.3E-05). Western blotting confirmed increased AXIN1 protein expression by FL3 as compared with negligible increases in TLE4, SMEK1, or KLf4 (Supplementary Fig. S1B). Significant increases in Jun (P = 8.9E-07) and ep300 (P = 4.6E-07) were also demonstrated by RNA-seq, but this was not evident at the protein level (Supplementary Fig. S1B). These results identify a potential role of Wnt/β-catenin inhibition in mediating antitumorigenic properties of FL3.

Figure 1.

FL3 decreases AOM-DSS–induced colonic tumorigenesis. One day prior to and during each cycle of dextran sodium sulfate, mice were intraperitoneally injected daily with 0.1 mg/kg FL3 or vehicle (veh). A, Average neoplasia counts per mouse. B, Average neoplasia size per mouse. C, Hematoxylin and eosin staining of colon histology from AOM-DSS–treated mice. Scale bars, 250 μm. D, Percentage of mice displaying highest extent of dysplasia. n = 7 per group. E–G, Colonic tumoroids were cultured from AOM-DSS–induced tumors from three independent mice and grown in Matrigel for 7 days. Tumoroids were then treated with 10 nmol/L FL3 or vehicle. E, Morphologic changes in dead tumoroids treated with FL3 for 16 hours (arrows). Scale bars, 100 μm. F, Cell viability measured by lactate dehydrogenase release after 16 hours of FL3 treatment. n = 8 per group. G, Heatmap of genes implicated in regulating Wnt/β-catenin signaling identified by RNA-seq analysis using total RNA from tumoroids treated with FL3 for 2 hours. *, P < 0.05; **, P < 0.01 by unpaired, two-tailed Student t test.

Figure 1.

FL3 decreases AOM-DSS–induced colonic tumorigenesis. One day prior to and during each cycle of dextran sodium sulfate, mice were intraperitoneally injected daily with 0.1 mg/kg FL3 or vehicle (veh). A, Average neoplasia counts per mouse. B, Average neoplasia size per mouse. C, Hematoxylin and eosin staining of colon histology from AOM-DSS–treated mice. Scale bars, 250 μm. D, Percentage of mice displaying highest extent of dysplasia. n = 7 per group. E–G, Colonic tumoroids were cultured from AOM-DSS–induced tumors from three independent mice and grown in Matrigel for 7 days. Tumoroids were then treated with 10 nmol/L FL3 or vehicle. E, Morphologic changes in dead tumoroids treated with FL3 for 16 hours (arrows). Scale bars, 100 μm. F, Cell viability measured by lactate dehydrogenase release after 16 hours of FL3 treatment. n = 8 per group. G, Heatmap of genes implicated in regulating Wnt/β-catenin signaling identified by RNA-seq analysis using total RNA from tumoroids treated with FL3 for 2 hours. *, P < 0.05; **, P < 0.01 by unpaired, two-tailed Student t test.

Close modal

FL3 decreases transcriptional activation by β-catenin and protein expression in RKO and SW48 colorectal cancer cells

To determine whether FL3 inhibits the Wnt/β-catenin pathway, we utilized various human colorectal cancer cell lines with known APC and CTNNB1 gene mutation status. FL3 (10 nmol/L) significantly decreased viability of RKO and SW48 cells, but was not effective against HCT116 cells or the nontransformed intestinal epithelial cell line IEC6 (Supplementary Fig. S2A). HCT116 cells harbor a heterozygous deletion of S45 of β-catenin coupled with LOH, which is the site of phosphorylation by CK1α and the priming step necessary for phosphorylation of β-catenin at other sites that drive it to proteasomal degradation (22). Phosphorylation at S33 of β-catenin (mutant in SW48 cells) is dispensable if S45, S37, and T41 are phosphorylated (22). FL3 induced cell death as measured by TUNEL staining and cleaved caspase-3 expression in RKO and SW48 cells, but not in HCT116 or IEC6 cells (Supplementary Fig. S2B–S2D). FL3 did not alter cell proliferation in any cell line tested and decreased migration of RKO and SW48 cells, but not HCT116 cells (Supplementary Fig. S3). These results suggest that FL3′s therapeutic response to decrease cell viability and migration is dependent on β-catenin mutation status.

We next measured β-catenin activation and expression during FL3 treatment. Western blotting and immunofluorescence staining confirmed that 10 nmol/L FL3 decreased β-catenin protein expression in RKO and SW48 cells, but not in HCT116 or IEC6 cells (Fig. 2A and B). Human colorectal cancer cell lines were transfected with a TCF/LEF luciferase reporter plasmid to measure transcriptional activation by β-catenin and treated with FL3. FL3 (10 nmol/L) abolished relative luciferase expression to levels similar to negative control in RKO and SW48 cells, but in HCT116 cells relative luciferase expression remained 6-fold higher than negative control (Fig. 2C). Indeed, FL3 decreased expression of Wnt target genes in RKO cells (Fig. 2D) and increased phosphorylation of β-catenin at S45 that inversely correlated with β-catenin degradation (Supplementary Fig. S4A). Because HCT116 cells harbor heterozygous G13D mutation of K-Ras in addition to heterozygous deletion of S45 of β-catenin, we stably transfected RKO cells with a plasmid carrying the same HCT116 S45-β-catenin deletion (RKO mut). RKO mut cells were no longer susceptible to FL3-induced β-catenin degradation (Supplementary Fig. S4B and S4C) and cell death (Supplementary Fig. S4D–S4F), suggesting that S45 of β-catenin is crucial for FL3 action. A cell viability comparison of FL3 with common colorectal cancer chemotherapeutic drugs 5-fluorouracil (5-FU), irinotecan, and oxaliplatin in a normal intestinal epithelial cell line (IEC6), a FL3-responsive colorectal cancer cell line (RKO), and a FL3-nonreponsive colorectal cancer cell line (HCT116) demonstrates that FL3 did not decrease viability in IEC6 cells compared with 5-FU, irinotecan, or oxaliplatin (Supplementary Fig. S5). 5-FU or oxaliplatin decreased cell viability in IEC6 cells similarly to that in RKO or HCT116 cells, suggesting equal toxicity by these drugs in normal intestinal epithelial cells and colorectal cancer cells (Supplementary Fig. S5). Irinotecan did not alter viability compared with vehicle in any cell line tested.

Figure 2.

FL3 decreases transcriptional activation by β-catenin and protein expression in RKO and SW48 cells. IEC6 (normal intestinal epithelial cells), RKO, HCT116, and SW48 (colorectal cancer cells) were treated with 10 nmol/L FL3 for 16 hours. A, Total β-catenin by Western blotting. B, Immunofluorescence staining for β-catenin. Scale bars, 100 μm; boxed pullouts, 50 μm. C, Relative luciferase expression of transcriptional activation by β-catenin. Negative control (NC) cells were transfected with a noninducible firefly reporter construct. n = 6 for IEC6; n = 3 for all other cell lines. D, Relative mRNA expression of Wnt target genes in RKO cells. *, P < 0.05 versus vehicle (veh) by one-way ANOVA, followed by Bonferroni test. n = 4–5 after outlier removed by Grubs test. *, P < 0.05; **, P < 0.01; ***, P < 0.005 versus vehicle by one-way ANOVA, followed by Bonferroni test.

Figure 2.

FL3 decreases transcriptional activation by β-catenin and protein expression in RKO and SW48 cells. IEC6 (normal intestinal epithelial cells), RKO, HCT116, and SW48 (colorectal cancer cells) were treated with 10 nmol/L FL3 for 16 hours. A, Total β-catenin by Western blotting. B, Immunofluorescence staining for β-catenin. Scale bars, 100 μm; boxed pullouts, 50 μm. C, Relative luciferase expression of transcriptional activation by β-catenin. Negative control (NC) cells were transfected with a noninducible firefly reporter construct. n = 6 for IEC6; n = 3 for all other cell lines. D, Relative mRNA expression of Wnt target genes in RKO cells. *, P < 0.05 versus vehicle (veh) by one-way ANOVA, followed by Bonferroni test. n = 4–5 after outlier removed by Grubs test. *, P < 0.05; **, P < 0.01; ***, P < 0.005 versus vehicle by one-way ANOVA, followed by Bonferroni test.

Close modal

FL3 inhibition of Wnt/β-catenin and induction of cell death in RKO cells is dependent on AXIN1

AXIN1 and AXIN2 are pivotal scaffold proteins crucial for the formation of the β-catenin destruction complex. AXIN is the least abundant component of the destruction complex due to proteasomal degradation by tankyrase and even small increases in AXIN protein alter β-catenin (20). Unlike AXIN2, which is a transcriptional target of β-catenin–dependent Wnt signaling, AXIN1 is ubiquitously expressed and is classically thought of as a tumor suppressor protein (28). RKO cells with knockdown of AXIN1 by three pooled siRNAs were resistant to FL3-induced β-catenin degradation (Fig. 3A and B), β-catenin transcriptional activation (Fig. 3C), and cell death (Fig. 3D–F). These results suggest that FL3 anticancer mechanism in colorectal cancer is dependent on AXIN1 degradation of β-catenin.

Figure 3.

FL3 inhibition of Wnt/β-catenin and induction of cell death in RKO cells is dependent on AXIN1. RKO cells were transfected with three pooled unique siRNA duplexes against AXIN1 (siAxin1) or RNAi nonspecific negative control (siNC) for 48 hours and then treated with 10 nmol/L FL3 for 16 hours. A, Representative Western blots to validate efficiency of AXIN1 knockdown and effect on β-catenin protein expression. B, Immunofluorescence staining for β- catenin. Scale bars, 100 μm. C, Relative luciferase expression of β-catenin transcriptional activation. Negative control (NC) cells were transfected with a noninducible firefly reporter construct. n = 3 in two independent experiments. D, Cell viability measured by lactate dehydrogenase release. n = 8 per group. E, TUNEL immunofluorescence staining. Scale bars, 100 μm. F, Percentage of TUNEL+ cells per field; 10 fields quantitated and averaged across 10 wells. ***, P < 0.005 versus vehicle (veh) siNC by two-way ANOVA, followed by Bonferroni test.

Figure 3.

FL3 inhibition of Wnt/β-catenin and induction of cell death in RKO cells is dependent on AXIN1. RKO cells were transfected with three pooled unique siRNA duplexes against AXIN1 (siAxin1) or RNAi nonspecific negative control (siNC) for 48 hours and then treated with 10 nmol/L FL3 for 16 hours. A, Representative Western blots to validate efficiency of AXIN1 knockdown and effect on β-catenin protein expression. B, Immunofluorescence staining for β- catenin. Scale bars, 100 μm. C, Relative luciferase expression of β-catenin transcriptional activation. Negative control (NC) cells were transfected with a noninducible firefly reporter construct. n = 3 in two independent experiments. D, Cell viability measured by lactate dehydrogenase release. n = 8 per group. E, TUNEL immunofluorescence staining. Scale bars, 100 μm. F, Percentage of TUNEL+ cells per field; 10 fields quantitated and averaged across 10 wells. ***, P < 0.005 versus vehicle (veh) siNC by two-way ANOVA, followed by Bonferroni test.

Close modal

FL3 decreases intestinal tumorigenesis in the ApcMin/+ mice

To determine whether FL3 is able to ameliorate colon tumorigenesis in a model of robust Wnt/β-catenin overexpression, ApcMin/+ mice were administered 0.1 mg/kg FL3 or vehicle twice weekly from 15 to 20 weeks of age. FL3 significantly decreased the number and size of adenomas and severity of dysplasia in ApcMin/+ mice (Fig. 4A–C). This was associated with decreased β-catenin expression (Fig. 4D) and increased cell death in adenomas (Fig. 4E–G), but no change in cell proliferation (Supplementary Fig. S6). Quantitation of TUNEL+ cells in nontumor regions demonstrated that FL3 did not alter the number of apoptotic cells in normal crypts or villi in vivo (Fig. 4F). These studies demonstrate that FL3 is sufficient to impact tumor growth in the absence of normal APC activity. WT mice injected with FL3 alongside ApcMin/+ mice demonstrated normal gastrointestinal histology including small intestinal villus length and small intestinal and colonic crypt length (Supplementary Fig. S7).

Figure 4.

FL3 decreases intestinal tumorigenesis in ApcMin/+ mice. Beginning at 15 weeks of age, ApcMin/+ mice were intraperitoneally injected with 0.1 mg/kg FL3 or vehicle (veh) twice weekly for 5 weeks. A, Average adenoma counts per mouse. B, Average adenoma size per mouse. C, Percentage of mice displaying highest extent of dysplasia. LGD, low-grade dysplasia; HGD, high-grade dysplasia; Intramuc. AC, intramucosal adenocarcinoma. D, IHC staining of β-catenin in adenomas. E, TUNEL immunofluorescence staining in adenomas. Scale bars, 250 μm; boxed pullouts, 50 μm. F, Quantification of TUNEL+ cells per adenoma, crypt, or villus. G, Western blot analysis of apoptosis marker cleaved caspase-3 from total protein of isolated adenomas. n = 10 mice per group. *, P < 0.05; **, P < 0.01; ***, P < 0.005 by unpaired, two-tailed Student t test.

Figure 4.

FL3 decreases intestinal tumorigenesis in ApcMin/+ mice. Beginning at 15 weeks of age, ApcMin/+ mice were intraperitoneally injected with 0.1 mg/kg FL3 or vehicle (veh) twice weekly for 5 weeks. A, Average adenoma counts per mouse. B, Average adenoma size per mouse. C, Percentage of mice displaying highest extent of dysplasia. LGD, low-grade dysplasia; HGD, high-grade dysplasia; Intramuc. AC, intramucosal adenocarcinoma. D, IHC staining of β-catenin in adenomas. E, TUNEL immunofluorescence staining in adenomas. Scale bars, 250 μm; boxed pullouts, 50 μm. F, Quantification of TUNEL+ cells per adenoma, crypt, or villus. G, Western blot analysis of apoptosis marker cleaved caspase-3 from total protein of isolated adenomas. n = 10 mice per group. *, P < 0.05; **, P < 0.01; ***, P < 0.005 by unpaired, two-tailed Student t test.

Close modal

FL3 decreases viability of human colorectal cancer tumoroids but not matched-normal colonoids

Historically, preclinical gastrointestinal translational research has relied on cell cultures that are of limited relevance to human physiology (27). These models often fail to predict human toxicity and/or efficacy (29). The establishment of 3D Matrigel-based intestinal enteroid cultures in recent years provides a rapid and simple means to test toxicity and efficacy of pharmaceutical compounds in human specimens (27). To assess translational utility of FL3, tumoroids were cultured from human colorectal cancer specimens and colonoids from matched-normal tissue and treated with FL3. FL3 decreased viability of colorectal cancer tumoroids derived from 5 of 6 patients, but did not affect cell viability in normal colonoids (Fig. 5A and B). Sequencing analysis confirmed that tumoroids derived from patient 3, which were resistant to FL3 treatment, harbored heterozygous S45 deletion of β-catenin, corroborating our results in HCT116 cells. All other patients were WT at S45 of β-catenin.

Figure 5.

FL3 decreases viability of human colorectal cancer (CRC) tumoroids, but not matched-normal colonoids. A, Light microscopy of tumoroids derived from stage II N1a colorectal cancer demonstrates that FL3 disrupted morphology, indicative of dead cells. Scale bars, 100 μm. B, Cell viability was measured by lactate dehydrogenase release from colorectal cancer or matched-normal colonoids from 6 patients treated with 50 nmol/L FL3 for 16 hours. *, P < 0.05; **, P < 0.01; ***, P < 0.005 versus tumoroid vehicle (veh) by one-way ANOVA, followed by Bonferroni test. n = 6 wells per treatment. C and D, Colorectal cancer tumoroids derived from patient 1 were dissociated into single cells, nucleofected with CRISPR-Cas9 sgRNAs targeting PHB1, and 72 hours later treated with 50 nmol/L FL3 for 16 hours. C, Efficiency of PHB1 knockdown and AXIN1 protein expression measured by Western blotting. D, Lactate dehydrogenase release. E, Western blots of glycocalyx markers in human colonoids and tumoroids from patient 1. **, P < 0.01 versus control vehicle by one-way ANOVA followed by Bonferroni test. n = 10 per treatment.

Figure 5.

FL3 decreases viability of human colorectal cancer (CRC) tumoroids, but not matched-normal colonoids. A, Light microscopy of tumoroids derived from stage II N1a colorectal cancer demonstrates that FL3 disrupted morphology, indicative of dead cells. Scale bars, 100 μm. B, Cell viability was measured by lactate dehydrogenase release from colorectal cancer or matched-normal colonoids from 6 patients treated with 50 nmol/L FL3 for 16 hours. *, P < 0.05; **, P < 0.01; ***, P < 0.005 versus tumoroid vehicle (veh) by one-way ANOVA, followed by Bonferroni test. n = 6 wells per treatment. C and D, Colorectal cancer tumoroids derived from patient 1 were dissociated into single cells, nucleofected with CRISPR-Cas9 sgRNAs targeting PHB1, and 72 hours later treated with 50 nmol/L FL3 for 16 hours. C, Efficiency of PHB1 knockdown and AXIN1 protein expression measured by Western blotting. D, Lactate dehydrogenase release. E, Western blots of glycocalyx markers in human colonoids and tumoroids from patient 1. **, P < 0.01 versus control vehicle by one-way ANOVA followed by Bonferroni test. n = 10 per treatment.

Close modal

PHB1 expression is required for FL3 antitumorigenic action in the colon

Mice with Villin-CreERT2 tamoxifen-inducible deletion of Phb1 from the intestinal epithelium (Phb1iΔIEC) were recently described previously (26). We utilized this model to determine whether PHB1 is necessary for FL3 to combat colonic tumorigenesis. Because we did not observe spontaneous colonic neoplasia in Phb1iΔIEC mice up to 12 weeks following deletion of Phb1, we induced colitis-associated colorectal cancer in Phb1iΔIEC and Phb1fl/fl control mice using AOM-DSS and cultured tumoroids. FL3 induced death of tumoroids derived from Phb1fl/fl mice, whereas FL3 had no effect on tumoroids derived from Phb1iΔIEC mice (Supplementary Fig. S8A and S8B). In addition, FL3 induction of AXIN1 was absent in Phb1-deficient tumoroids (Supplementary Fig. S8C). We next selected human colorectal cancer tumoroids from a patient responsive to FL3 (patient 1; Fig. 5B) and targeted PHB1 using CRISPR-Cas9 single-guide RNAs (sgRNA; Fig. 5C). FL3 was ineffective to increase AXIN1 expression (Fig. 5C) and to decrease viability in PHB1-targeted human colorectal cancer tumoroids (Fig. 5D). These data suggest that PHB1 is required for FL3 antitumor action in the colon. To demonstrate aberrant glycosylation in our tumoroid model compared with colonoids, which is an important feature of cancer cells (30), colorectal cancer tumoroids exhibited increased sialylated-Tn (a shortened glycan formed from incomplete synthesis of O-glycans common in cancer) and increased N-acetylglucosamine transferase (GnT-V) that is commonly upregulated in cancer (Fig. 5E). These findings are important for the translational relevance of FL3 using colorectal cancer tumoroid models and are in agreement with previous reports demonstrating glycocalyx expression in intestinal and colonic organoids (31, 32).

The role of PHB1 in cancer depends upon the tumor type, with expression of PHB1 being reported as increased in colorectal cancer (33, 34). Dynamic partitioning of PHB1 between nuclear and cytoplasmic compartments of tumor cells has been shown to be signal dependent and necessary for the induction of apoptosis (35). Compared with the nontransformed cell line IEC6, colorectal cancer cells exhibited increased phosphorylation of PHB1 at Thr258 (Fig. 6A), a posttranslational modification that induces mitochondrial localization of PHB1 in other types of cancer and correlates with enhanced proliferation, invasion, and metastasis (36, 37). FL3 inhibited phosphorylation of PHB1 at Thr258 in colorectal cancer cells (Fig. 6A) with a concomitant nuclear translocation of PHB1 in colorectal cancer cells, but not in normal IECs (Fig. 6B). In the nucleus, PHB1 has been shown to act as a coregulator of transcription by interacting with various transcription factors such as p53, Rb, and E2F (3). A previous study identified PHB1 as a novel factor able to bind to the (TGYCC)n motif in the promoter of p53-inducible gene 3 (PIG3; ref. 38). This motif is similar to the underlined portion of the p53 binding site (RRRCWWGYYY; R = A or G, W = A or T, Y = C or T), which repeats this motif with a 0–21 bp spacer in-between. Using UCSC Genome Browser, the promoter regions of AXIN1, SMEK1, TLE4, and KLF4 genes (the genes most significantly increased by FL3 in RNA-seq analysis shown in Fig. 1G) were searched for WGYYY or RGYYY motifs with or without a spacer. None were found in SMEK1, TLE4, or KLF4 promoters. However, the AXIN1 promoter contains the sequence 5′-GGCCTGGGCTTCGGCGCTCTGGCTCGGGCTCTGGCTC-3′ located −5668 to −5631 from the transcriptional start site in the 5′ untranslated region (UTR), which contains five RGYYY motifs (underlined), and we will refer to these as putative PHB1-binding motif herein. ChIP assays demonstrated that PHB1 associates with this sequence in the AXIN1 promoter and that PHB1 binding is enriched by FL3 treatment in RKO cells (Fig. 6C and D). Collectively, these results suggest a model for the in vivo anticancer mechanism of FL3 involving PHB1-induced AXIN1 expression and β-catenin degradation.

Figure 6.

In colorectal cancer cells, FL3 blocks phosphorylation of T258-PHB1 and induces nuclear translocation of PHB1, where it binds to the promoter of AXIN1. A, Western blot analysis of cells treated with 10 nmol/L FL3 for 2 hours. B, Colorectal cancer RKO cells and normal IEC6 epithelial cells were treated with 10 nmol/L FL3 for 2 hours. Immunofluorescence staining of PHB1. C and D, Chromatin was isolated from RKO cells treated with vehicle (veh) or 10 nmol/L FL3 for 1 or 2 hours. ChIP assays were performed using PHB1 or IgG control antibodies (Ab) and primers spanning the putative PHB1-binding site in the AXIN1 promoter. Immunoprecipitates were analyzed by PCR (C) and qPCR (D) amplification to show the relative enrichment of PHB1 binding as percentage of total input DNA.

Figure 6.

In colorectal cancer cells, FL3 blocks phosphorylation of T258-PHB1 and induces nuclear translocation of PHB1, where it binds to the promoter of AXIN1. A, Western blot analysis of cells treated with 10 nmol/L FL3 for 2 hours. B, Colorectal cancer RKO cells and normal IEC6 epithelial cells were treated with 10 nmol/L FL3 for 2 hours. Immunofluorescence staining of PHB1. C and D, Chromatin was isolated from RKO cells treated with vehicle (veh) or 10 nmol/L FL3 for 1 or 2 hours. ChIP assays were performed using PHB1 or IgG control antibodies (Ab) and primers spanning the putative PHB1-binding site in the AXIN1 promoter. Immunoprecipitates were analyzed by PCR (C) and qPCR (D) amplification to show the relative enrichment of PHB1 binding as percentage of total input DNA.

Close modal

FL3 is a novel pharmacologic agent, not previously tested in cell or preclinical models of intestinal tumorigenesis. Here, we provide proof of principle for fundamental mechanisms of FL3 to combat intestinal tumorigenesis without significant toxicity. Using an unbiased approach (RNA-seq analysis), we identify inhibition of Wnt/β-catenin as the mechanism of FL3 anticancer activity dependent on PHB1 signaling. Importantly, we demonstrate translational utility of FL3 using human colorectal cancer tumoroids and matched-normal colonoids. FL3 provides a novel small molecule to combat cancers dependent on the Wnt pathway such colorectal cancer.

It is well established that genetic mutations in various genes are associated with different prognosis and response to colorectal cancer therapy. For instance, activating mutations in K-Ras render EGFR inhibition ineffective (39) and TP53 mutation diminishes response to 5-FU (40). Response to FL3 was diminished in colorectal cancer cell lines harboring deletion of S45 of CTNNB1 including HCT116. HCT116 cells harbor heterozygous G13D mutation of K-Ras in addition to heterozygous deletion of S45 of β-catenin. To determine which mutation diminishes FL3 response, stably transfected RKO cells with a plasmid carrying the same HCT116 S45-β-catenin deletion (RKO mut cells) were no longer susceptible to FL3–induced β-catenin degradation and cell death, suggesting that S45 of β-catenin is crucial for FL3 action. DNA sequencing revealed that the human colorectal cancer tumoroids unresponsive to FL3 (patient 3) carried mutation at S45 of β-catenin. Collectively, these results suggest that FL3 anticancer action is dependent on degradation of β-catenin, requiring phosphorylation at S45.

FL3 decreased tumor number, size, and severity of dysplasia, while increasing tumor cell death in two mouse models of intestinal tumorigenesis, one inflammation-induced model (AOM-DSS) and one sporadic model (ApcMin/+ mice). FL3 efficacy in ApcMin/+ mice demonstrates that FL3 is sufficient to combat tumorigenicity in a model driven by aberrant Wnt/β-catenin overexpression and in the absence of normal APC activity. This is especially important for the translational potential of FL3 because mutation in the APC gene is present in 49.5% of all patients with colorectal cancer, a much higher percentage than mutation in the CTNNB1 gene (4.99% of patients with colorectal cancer) or AXIN1 gene (2.72% of patients with colorectal cancer; ref. 41), suggesting that the majority of patients with colorectal cancer will likely respond to FL3 treatment based on APC and CTNNB1 genetic mutation status. Heterogeneity of colitis-associated colorectal cancer and sporadic colorectal cancer includes involvement of chromosomal instability (CIN), microsatellite instability (MSI), and promoter CpG island methylator phenotype (CIMP; refs. 42–44). Future studies will further analyze these classifications in regards to FL3 response. On the basis of the colorectal cancer cell lines used in this study, HCT116, RKO, and SW48 are CIN/MSI-high, and therefore, does not alter response to FL3. CIMP, which is the status of HCT116 cells, could contribute to diminished FL3 responsiveness, whereas CIMP+, which is the status of RKO and SW48, could be an important indicator of FL3 effectiveness. Future studies using other CIN+/MSS/CIMP cells such as Caco2 and HT29 will further elucidate FL3 response with these classifications.

The role of PHB1 in cancer depends upon the tumor type, with expression of PHB1 being reported as increased in colorectal cancer (33, 34). Dynamic partitioning of PHB1 between nuclear and cytoplasmic compartments of tumor cells has been shown to be signal dependent and necessary for the induction of apoptosis in tumor cells (35). Although human PHB1 does not possess a mitochondrial targeting sequence, the N-terminal transmembrane domain of PHB1 (amino acid 2–24) targets it to the mitochondria where it is anchored to the inner mitochondrial membrane (45). At the C-terminus of PHB1, a leucine/isoleucine nuclear export sequence between amino acid 257–270 facilitates its export from the nucleus via CRM-1 export receptor to the cytoplasm (35). This specific structure of PHB1 allows active shuttling between the organelles. Previous studies have shown that phosphorylation of PHB1 at Thr258 enhances its mitochondrial localization in bladder and cervical cancer cells and is crucial to increased proliferation, invasion, and metastasis in these cancer cells (10, 36, 37). Here, we demonstrate that RKO, HCT116, and SW48 colorectal cancer cells exhibit increased phosphorylation of PHB1 at Thr258, which is inhibited by FL3. Nuclear localization of PHB1 is induced by FL3 treatment in RKO colorectal cancer cells, but not in nontransformed IEC6 cells. We show by ChIP that PHB1 binding to the 5′UTR of the AXIN1 gene is increased by FL3 in RKO colorectal cancer cells and that PHB1 deficiency in mouse and human colorectal cancer tumoroids abolishes FL3-induced AXIN1 expression and cell death. These results suggest that, FL3 inhibits Wnt/β-catenin via blockade of PHB1 phosphorylation at Thr258 and subsequent PHB1 nuclear translocation where it induces transcription of AXIN1, the core molecule and limiting component of the β-catenin degradation complex. PHB1 was shown to inhibit Wnt signaling in murine liver and human hepatocellular carcinoma cells via regulation of the transcription factor E2F1 (46). Interestingly, similar to FL3, other AXIN1 stabilizers that act via tankyrase inhibition have been shown to impact tumorigenesis in the absence of normal APC activity (47). The tight control AXIN1 expression exerts on the β-catenin destruction complex is a property currently being exploited in the development of novel therapeutics for Wnt-driven cancers (20) and our results suggest that FL3 at low nanomolar concentrations upregulates AXIN1 expression. We admit that PHB1 regulates multiple signaling pathways dependent on cell type studied, posttranslational modifications of PHB1, and subcellular localization of PHB1. As examples, PHB1 can interact with C-RAF, AKT, IKK, and STAT3 (48). Given our unbiased RNA-seq data implicating FL3 inhibition of Wnt/β-catenin signaling in colorectal cancer, we currently focused on this pathway. Future studies will elucidate the role of other known PHB1 signaling pathways in FL3 antitumorigenic properties in colorectal cancer.

Polymorphisms of the PHB1 gene have been linked to gastric, breast, ovarian, and skin cancers, but it remains unknown whether these genetic alterations affect PHB1 posttranslational modifications or function (49). Our results, elucidating FL3 mechanism via PHB1 action may provide an ability to stratify patients with colorectal cancer by those who are most likely to respond to FL3 treatment (those with high p-Thr258-PHB1 expression and absence of S45 β-catenin mutation). FL3 treatment of nontransformed cells alongside colorectal cancer cells and colorectal cancer tumoroids, as well as no change by FL3 treatment in the number of TUNEL+ cells in normal crypts or villi in vivo, suggests that FL3 lacks toxicity in normal intestinal epithelial cells. We propose that the targeting of PHB1 by FL3 provides a novel mechanism to inhibit Wnt signaling without intestinal toxicity because PHB1 nuclear shuttling response to FL3 is specific to colorectal cancer cells. Although many inhibitors of the Wnt/β-catenin pathway have been characterized with some reaching early clinical trials (23), these compounds exhibit significant adverse reactions given the role of Wnt signaling in maintaining homeostasis of many adult tissues, particularly the intestine. As examples, Wnt pathway inhibitors, XAV939 and LGK974, result in severe intestinal toxicity in mice and OMP18R5 (vantictumab) induces abdominal pain and diarrhea in patients preventing advancement in clinical trials (23). FL3 provides a novel small-molecule Wnt/β-catenin inhibitor presenting little toxicity to healthy intestinal epithelial cells, as shown previously in other healthy cell types (3, 9, 10).

Our study defines a link between FL3, PHB1, and Wnt/β-catenin inhibition, which is a key pathway mediating intestinal carcinogenesis. These results provide important insight in the nuclear trafficking of PHB1 and its contribution to tumor death dependent on cellular localization and posttranscriptional modifications. We provide the first identification of in vivo molecular targets of FL3 and specifically identify a novel role of PHB1 in regulating AXIN1 levels. FL3 provides a novel small-molecule inhibitor of β-catenin via upregulation of AXIN1 expression as a potential therapeutic for Wnt pathway–driven cancers such as colorectal cancer. Future studies are poised to build the preclinical biosafety profile of FL3 and test the role of FL3 as a sensitizing agent in chemoresistant colorectal cancer cell lines.

No potential conflicts of interest were disclosed.

D.N. Jackson: Data curation, software, formal analysis, writing-original draft. K.M. Alula: Data curation, software, formal analysis, validation. Y. Delgado-Deida: Data curation, software, formal analysis, validation. R. Tabti: Data curation. K. Turner: Data curation, writing-review and editing. X. Wang: Data curation, software, formal analysis. K. Venuprasad: Formal analysis, writing-review and editing. R.F. Souza: Conceptualization, writing-review and editing. L. Désaubry: Conceptualization, resources, writing-review and editing. A.L. Theiss: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, project administration, writing-review and editing.

We thank Dr. Xiaofang Huo and Dr. Qiuyang Zhang for providing key antibodies, Jie Han and Masha Sorouri for technical assistance, Beth Cook for histology processing, Cynthia Smitherman for RNA-seq analysis, and Jinghua Gu for biostatistical support (Baylor Scott & White Research Institute). This work was supported by NIH grants (R01-DK117001 to A.L. Theiss) and Litwin IBD Pioneers Crohn's Colitis Foundation 301869 (to A.L. Theiss).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Mattar
MC
,
Lough
D
,
Pishvaian
MJ
,
Charabaty
A
. 
Current management of inflammatory bowel disease and colorectal cancer
.
Gastrointest Cancer Res
2011
;
4
:
53
61
.
2.
Fajardo
AM
,
Piazza
GA
. 
Chemoprevention in gastrointestinal physiology and disease. Anti-inflammatory approaches for colorectal cancer chemoprevention
.
Am J Physiol Gastrointest Liver Physiol
2015
;
309
:
G59
70
.
3.
Thuaud
F
,
Ribeiro
N
,
Nebigil
CG
,
Desaubry
L
. 
Prohibitin ligands in cell death and survival: mode of action and therapeutic potential
.
Chem Biol
2013
;
20
:
316
31
.
4.
Li-Weber
M.
Molecular mechanisms and anti-cancer aspects of the medicinal phytochemicals rocaglamides (=flavaglines)
.
Int J Cancer
2015
;
137
:
1791
9
.
5.
Liu
T
,
Nair
SJ
,
Lescarbeau
A
,
Belani
J
,
Peluso
S
,
Conley
J
, et al
Synthetic silvestrol analogues as potent and selective protein synthesis inhibitors
.
J Med Chem
2012
;
55
:
8859
78
.
6.
Fahrig
T
,
Gerlach
I
,
Horvath
E
. 
A synthetic derivative of the natural product rocaglaol is a potent inhibitor of cytokine-mediated signaling and shows neuroprotective activity in vitro and in animal models of Parkinson's disease and traumatic brain injury
.
Mol Pharmacol
2005
;
67
:
1544
55
.
7.
Bernard
Y
,
Ribeiro
N
,
Thuaud
F
,
Turkeri
G
,
Dirr
R
,
Boulberdaa
M
, et al
Flavaglines alleviate doxorubicin cardiotoxicity: implication of Hsp27
.
PLoS One
2011
;
6
:
e25302
.
8.
Han
J
,
Zhao
Q
,
Basmadjian
C
,
Desaubry
L
,
Theiss
AL
. 
Flavaglines ameliorate experimental colitis and protect against intestinal epithelial cell apoptosis and mitochondrial dysfunction
.
Inflamm Bowel Dis
2016
;
22
:
55
67
.
9.
Qureshi
R
,
Yildirim
O
,
Gasser
A
,
Basmadjian
C
,
Zhao
Q
,
Wilmet
JP
, et al
FL3, a synthetic flavagline and ligand of prohibitins, protects cardiomyocytes via STAT3 from doxorubicin toxicity
.
PLoS One
2015
;
10
:
e0141826
.
10.
Yuan
G
,
Chen
X
,
Liu
Z
,
Wei
W
,
Shu
Q
,
Abou-Hamdan
H
, et al
Flavagline analog FL3 induces cell cycle arrest in urothelial carcinoma cell of the bladder by inhibiting the Akt/PHB interaction to activate the GADD45alpha pathway
.
J Exp Clin Cancer Res
2018
;
37
:
21
.
11.
Basmadjian
C
,
Thuaud
F
,
Ribeiro
N
,
Desaubry
L
. 
Flavaglines: potent anticancer drugs that target prohibitins and the helicase eIF4A
.
Future Med Chem
2013
;
5
:
2185
97
.
12.
Bourges
I
,
Ramus
C
,
Mousson de Camaret
B
,
Beugnot
R
,
Remacle
C
,
Cardol
P
, et al
Structural organization of mitochondrial human complex I: role of the ND4 and ND5 mitochondria-encoded subunits and interaction with prohibitin
.
Biochem J
2004
;
383
:
491
9
.
13.
Hsieh
SY
,
Shih
TC
,
Yeh
CY
,
Lin
CJ
,
Chou
YY
,
Lee
YS
. 
Comparative proteomic studies on the pathogenesis of human ulcerative colitis
.
Proteomics
2006
;
6
:
5322
31
.
14.
Theiss
AL
,
Idell
RD
,
Srinivasan
S
,
Klapproth
JM
,
Jones
DP
,
Merlin
D
, et al
Prohibitin protects against oxidative stress in intestinal epithelial cells
.
FASEB J
2007
;
21
:
197
206
.
15.
Tsutsumi
T
,
Matsuda
M
,
Aizaki
H
,
Moriya
K
,
Miyoshi
H
,
Fujie
H
, et al
Proteomics analysis of mitochondrial proteins reveals overexpression of a mitochondrial protein chaperon, prohibitin, in cells expressing hepatitis C virus core protein
.
Hepatology
2009
;
50
:
378
86
.
16.
Van Der Kraak
L
,
Gros
P
,
Beauchemin
N
. 
Colitis-associated colon cancer: is it in your genes?
World J Gastroenterol
2015
;
21
:
11688
99
.
17.
Conlin
A
,
Smith
G
,
Carey
FA
,
Wolf
CR
,
Steele
RJ
. 
The prognostic significance of K-ras, p53, and APC mutations in colorectal carcinoma
.
Gut
2005
;
54
:
1283
6
.
18.
Yamauchi
M
,
Morikawa
T
,
Kuchiba
A
,
Imamura
Y
,
Qian
ZR
,
Nishihara
R
, et al
Assessment of colorectal cancer molecular features along bowel subsites challenges the conception of distinct dichotomy of proximal versus distal colorectum
.
Gut
2012
;
61
:
847
54
.
19.
Koch
S
. 
Extrinsic control of Wnt signaling in the intestine
.
Differentiation
2017
;
97
:
1
8
.
20.
Wang
Z
,
Tacchelly-Benites
O
,
Yang
E
,
Thorne
CA
,
Nojima
H
,
Lee
E
, et al
Wnt/Wingless pathway activation is promoted by a critical threshold of Axin maintained by the tumor suppressor APC and the ADP-ribose polymerase tankyrase
.
Genetics
2016
;
203
:
269
81
.
21.
Daugherty
RL
,
Gottardi
CJ
. 
Phospho-regulation of β-catenin adhesion and signaling functions
.
Physiology
2007
;
22
:
303
9
.
22.
Tortelote
GG
,
Reis
RR
,
de Almeida Mendes
F
,
Abreu
JG
. 
Complexity of the Wnt/β-catenin pathway: searching for an activation model
.
Cell Signal
2017
;
40
:
30
43
.
23.
Krishnamurthy
N
,
Kurzrock
R
. 
Targeting the Wnt/beta-catenin pathway in cancer: update on effectors and inhibitors
.
Cancer Treat Rev
2018
;
62
:
50
60
.
24.
Shang
S
,
Hua
F
,
Hu
ZW
. 
The regulation of β-catenin activity and function in cancer: therapeutic opportunities
.
Oncotarget
2017
;
8
:
33972
89
.
25.
Thuaud
F
,
Bernard
Y
,
Turkeri
G
,
Dirr
R
,
Aubert
G
,
Cresteil
T
, et al
Synthetic analogue of rocaglaol displays a potent and selective cytotoxicity in cancer cells: involvement of apoptosis inducing factor and caspase-12
.
J Med Chem
2009
;
52
:
5176
87
.
26.
Jackson
DN
,
Panopoulos
M
,
Neumann
WL
,
Turner
K
,
Cantarel
BL
,
Thompson-Snipes
L
, et al
Mitochondrial dysfunction during loss of prohibitin 1 triggers Paneth cell defects and ileitis
.
Gut
2020
Feb 28 [Epub ahead of print].
27.
Sato
T
,
Stange
DE
,
Ferrante
M
,
Vries
RG
,
Van Es
JH
,
Van den Brink
S
, et al
Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett's epithelium
.
Gastroenterology
2011
;
141
:
1762
72
.
28.
Mazzoni
SM
,
Fearon
ER
. 
AXIN1 and AXIN2 variants in gastrointestinal cancers
.
Cancer Lett
2014
;
355
:
1
8
.
29.
Howell
JC
,
Wells
JM
. 
Generating intestinal tissue from stem cells: potential for research and therapy
.
Regen Med
2011
;
6
:
743
55
.
30.
Kang
H
,
Wu
Q
,
Sun
A
,
Liu
X
,
Fan
Y
,
Deng
X
. 
Cancer cell glycocalyx and its significance in cancer progression
.
Int J Mol Sci
2018
;
19
:
2484
.
31.
Fujii
M
,
Clevers
H
,
Sato
T
. 
Modeling human digestive diseases with CRISPR-Cas9-modified organoids
.
Gastroenterology
2019
;
156
:
562
76
.
32.
Rouhanifard
SH
,
Lopez Aguilar
A
,
Meng
L
,
Moremen
KW
,
Wu
P
. 
Engineered glycocalyx regulates stem cell proliferation in murine crypt organoids
.
Cell Chem Biol
2018
;
25
:
439
46
.
33.
Chen
YW
,
Chou
HC
,
Lyu
PC
,
Yin
HS
,
Huang
FL
,
Chang
WS
, et al
Mitochondrial proteomics analysis of tumorigenic and metastatic breast cancer markers
.
Funct Integr Genomics
2011
;
11
:
225
39
.
34.
Hammoudi
A
,
Song
F
,
Reed
KR
,
Jenkins
RE
,
Meniel
VS
,
Watson
AJ
, et al
Proteomic profiling of a mouse model of acute intestinal Apc deletion leads to identification of potential novel biomarkers of human colorectal cancer (CRC)
.
Biochem Biophys Res Commun
2013
;
440
:
364
70
.
35.
Rastogi
S
,
Joshi
B
,
Fusaro
G
,
Chellappan
S
. 
Camptothecin induces nuclear export of prohibitin preferentially in transformed cells through a CRM-1-dependent mechanism
.
J Biol Chem
2006
;
281
:
2951
9
.
36.
Chiu
CF
,
Ho
MY
,
Peng
JM
,
Hung
SW
,
Lee
WH
,
Liang
CM
, et al
Raf activation by Ras and promotion of cellular metastasis require phosphorylation of prohibitin in the raft domain of the plasma membrane
.
Oncogene
2013
;
32
:
777
87
.
37.
Jiang
L
,
Dong
P
,
Zhang
Z
,
Li
C
,
Li
Y
,
Liao
Y
, et al
Akt phosphorylates prohibitin 1 to mediate its mitochondrial localization and promote proliferation of bladder cancer cells
.
Cell Death Dis
2015
;
6
:
e1660
.
38.
Guan
X
,
Liu
Z
,
Wang
L
,
Johnson
DG
,
Wei
Q
. 
Identification of prohibitin and prohibiton as novel factors binding to the p53 induced gene 3 (PIG3) promoter (TGYCC)(15) motif
.
Biochem Biophys Res Commun
2014
;
443
:
1239
44
.
39.
Shi
C
,
Washington
K
. 
Molecular testing in colorectal cancer: diagnosis of Lynch syndrome and personalized cancer medicine
.
Am J Clin Pathol
2012
;
137
:
847
59
.
40.
Iacopetta
B
. 
TP53 mutation in colorectal cancer
.
Hum Mutat
2003
;
21
:
271
6
.
41.
Consortium
APG
. 
AACR project GENIE: powering precision medicine through an international consortium
.
Cancer Discov
2017
;
7
:
818
31
.
42.
Habermann
JK
,
Upender
MB
,
Roblick
UJ
,
Kruger
S
,
Freitag
S
,
Blegen
H
, et al
Pronounced chromosomal instability and multiple gene amplifications characterize ulcerative colitis-associated colorectal carcinomas
.
Cancer Genet Cytogenet
2003
;
147
:
9
17
.
43.
Scarpa
M
,
Scarpa
M
,
Castagliuolo
I
,
Erroi
F
,
Kotsafti
A
,
Basato
S
, et al
Aberrant gene methylation in non-neoplastic mucosa as a predictive marker of ulcerative colitis-associated CRC
.
Oncotarget
2016
;
7
:
10322
31
.
44.
Shivakumar
BM
,
Rotti
H
,
Vasudevan
TG
,
Balakrishnan
A
,
Chakrabarty
S
,
Bhat
G
, et al
Copy number variations are progressively associated with the pathogenesis of colorectal cancer in ulcerative colitis
.
World J Gastroenterol
2015
;
21
:
616
22
.
45.
Peng
YT
,
Chen
P
,
Ouyang
RY
,
Song
L
. 
Multifaceted role of prohibitin in cell survival and apoptosis
.
Apoptosis
2015
;
20
:
1135
49
.
46.
Mavila
N
,
Tang
Y
,
Berlind
J
,
Ramani
K
,
Wang
J
,
Mato
JM
, et al
Prohibitin 1 acts as a negative regulator of Wingless/Integrated-Beta-catenin signaling in murine liver and human liver cancer cells
.
Hepatol Commun
2018
;
2
:
1583
600
.
47.
Lau
T
,
Chan
E
,
Callow
M
,
Waaler
J
,
Boggs
J
,
Blake
RA
, et al
A novel tankyrase small-molecule inhibitor suppresses APC mutation-driven colorectal tumor growth
.
Cancer Res
2013
;
73
:
3132
44
.
48.
Wang
D
,
Tabti
R
,
Elderwish
S
,
Abou-Hamdan
H
,
Djehal
A
,
Yu
P
, et al
Prohibitin ligands: a growing armamentarium to tackle cancers, osteoporosis, inflammatory, cardiac and neurological diseases
.
Cell Mol Life Sci
2020
Feb 15 [Epub ahead of print].
49.
Chowdhury
D
,
Kumar
D
,
Sarma
P
,
Tangutur
AD
,
Bhadra
MP
. 
PHB in cardiovascular and other diseases: present knowledge and implications
.
Curr Drug Targets
2017
;
18
:
1836
51
.