Lovastatin Inhibits RhoA to Suppress Canonical Wnt/β-Catenin Signaling and Alternative Wnt-YAP/TAZ Signaling in Colon Cancer.

Statins are first-line drugs used to control patient lipid levels, but there is recent evidence that statin treatment can lower colorectal cancer (CRC) incidence by 50% and prolong CRC patient survival through mechanisms that are poorly understood. In this study, we found that the treatment of APCmin mice by the mevalonate pathway inhibitor lovastatin significantly reduced the number of colonic masses and improved hypersplenism and peripheral anemia. Furthermore, reverse transcription polymerase chain reaction (RT-PCR) analysis of colonic mass tissues showed a potent inhibitory effect in both Wnt/β-catenin signaling and YAP/TAZ signaling in the lovastatin treatment group. The results of our transcriptomic analyses in RKO indicated that lovastatin regulated several proliferation-related signaling pathways. Moreover, lovastatin suppressed important genes and proteins related to the canonical Wnt/β-catenin and alternative Wnt-YAP/TAZ signaling pathways in RKO and SW480 cells, and these effects were rescued by mevalonic acid (MVA), as confirmed through a series of Western blotting, RT-PCR, and reporter assays. Given that statins suppress oncogenic processes primarily through the inhibition of Rho GTPase in the mevalonate pathway, we speculate that lovastatin can inhibit certain Rho GTPases to suppress both canonical Wnt/β-catenin signaling and alternative Wnt-YAP/TAZ signaling. In RKO cells, lovastatin showed similar inhibitory properties as the RhoA inhibitor CCG1423, being able to inhibit β-catenin, TAZ, and p-LATS1 protein activity. Our results revealed that lovastatin inhibited RhoA activity, thereby suppressing the downstream canonical Wnt/β-catenin and alternative Wnt-YAP/TAZ pathways in colon cancer cells. These inhibitory properties suggest the promise of statins as a treatment for CRC. Altogether, the present findings support the potential clinical use of statins in non-cardiovascular contexts and highlight novel targets for anticancer treatments.

Introduction

Statins inhibit the activity of the key rate-limiting mevalonate pathway enzyme HMG-CoA reductase, which is required for the synthesis of metabolic compounds including cholesterol . As such, statins can reduce hepatic cholesterol production and increase low-density lipoprotein (LDL) receptor levels, thereby contributing to a reduction in levels of plasma cholesterol . Through the inhibition of the mevalonate pathway, statins can reduce levels of farnesyl pyrophosphate (FPP) and geranylgeranyl pyrophosphate (GGPP), which are isoprenyl groups involved in the modification of Ras, Rac, Rho, and other small G proteins important for the regulation of cellular migration, survival, and proliferation . As statins have been increasingly employed in recent years, their non-lipid-related regulatory effects have been a focus of growing research. Notably, many statins have been found to exhibit inhibitory activity in a range of tumors[4-12]. In large-scale clinical studies, statins were shown to decrease colorectal cancer (CRC) incidence by 50% and to extend the duration of CRC patient survival[13-16]. As such, the mechanistic basis for the antitumor activity of statins has been a focus of growing research interest in recent years.

Ras gene mutations correspond to a common feature of many tumor types, and preliminary studies suggest that the inhibitory activities of statins are primarily attributable to the inhibition of the mevalonate pathway downstream target Rho GTPase[17-19]. There are approximately 20 Rho GTPase superfamily members that have been described to date, including RhoA, Rac1, and Cdc42 . In addition to regulating normal cellular differentiation, apoptosis, and proliferation, these signaling molecules are closely associated with tumor development and metastatic progression[21,22]. The serine/threonine protein kinase ROCK (Rho-related coil protein kinase) is a key downstream Rho GTPase effector molecule.

Wnt signaling regulates migratory, proliferative, and fate determination activities in cells[23-25]. Wnt signaling is commonly aberrantly activated in CRC and other tumor types, and consists of both canonical and alternative signaling pathways[25-27]. β-catenin and the TCF/LEF (T-cell factor/lymphoid enhancer factor family) transcription factors mediate canonical Wnt signaling[24,25]. Aberrant Wnt pathway activation can contribute to oncogenesis and growth defects, with abnormal Wnt activation being evident in roughly 90% of intestinal tumors . When this pathway is inactive, the GSK3β, APC, and AXIN1 complex can promote β-catenin phosphorylation, resulting in its ubiquitination and consequent degradation . When the function or structural characteristics of this complex are disrupted, β-catenin is not phosphorylated, leading to its cytoplasmic accumulation and subsequent nuclear entry wherein it can bind to the transcription factor TCF4, leading to the activation of downstream gene expression and the regulation of a variety of pathological and physiological processes including cellular proliferation and apoptosis . This pathophysiological aberration manifests in the APCmin mouse, which presents a mutant allele of APC gene and possesses a constitutive activation of Wnt signaling in the intestine, resulting in intestinal tumorigenesis .

Accumulating evidence suggests that YAP (yes-associated-protein)/TAZ (tafazzin) can serve as downstream effector molecules in an alternative Wnt signaling pathway . YAP/TAZ are also the core components of the Hippo pathway, which plays a central role in tumor development. Mammals exhibit significant conservation of Hippo pathway components such as MST1/2 (macrophage stimulating factor1/2), WW45 (Sav, Sav homolog), LATS1/2 (large tumor inhibitory kinase1/2), MOB1 (mps one binder 1), YAP, and the YAP paralog TAZ. MST1/2 are central regulators of this Hippo pathway, phosphorylating LATS1/2, WW45, and MOB1, ultimately leading to the phosphorylation of YAP/TAZ downstream[30-32]. After phosphorylation, YAP/TAZ can interact with cytoskeletal proteins to regulate organ size and volume[33-37]. High YAP/TAZ levels or nuclear YAP/TAZ enrichment can be evident in a range of tumor types including liver, breast, ovarian, lung, and colon cancers[38-46].

A pathological link exists between signaling cascades such as Wnt/β-catenin and Wnt-YAP/TAZ with cancer stem cells (CSCs) . β-catenin regulates stem cell differentiation, self-renewal, and pluripotency . Thus, the abnormal activation of Wnt/β-catenin can promote the progression of CSCs, leading to metastasis. YAP regulates differentiation in stem cells during organogenesis and cancer development . Activation of the YAP/TAZ pathway promotes induction of CSCs in a large variety of human cancers. Because of the interaction between the mevalonate pathway and its downstream effectors, β-catenin and YAP, with CSCs, mevalonate pathway inhibitors, including lovastatin, have been implicated as potential antineoplastic agents when used together with chemotherapy in adjuvant cancer treatment .

Mevalonate pathway activation can promote YAP/TAZ nuclear localization and activation[51,52]. Lovastatin and other statins can block this mevalonate pathway, thereby disrupting such YAP/TAZ-mediated transcriptional activation[51,52]. Statins can thus impact the alternative Wnt signaling pathway through a mechanism potentially linked to Rho GTPase inhibition. Whether statins further impact canonical Wnt signaling pathway activation in CRC cells, however, remains to be established. Given the clear evidence that statins can inhibit colon cancer in clinical contexts and the importance of canonical signaling in CRC onset and progression, we hypothesize that statins can exert anti-CRC efficacy by inhibiting both alternative Wnt signaling (YAP/TAZ) and canonical Wnt signaling. In this study, we sought to test this hypothesis and to further explore the potential mechanistic link between such inhibition and Rho GTPase activity. Overall, this study offers valuable new insights into the potential clinical use of statins in noncardiovascular contexts and highlights novel targets for anticancer treatments.

Materials and Methods

Animals

APCmin mice were purchased from Model Animal Research Center of Nanjing University (Nanjing, China). The mice were housed under standard light/dark cycling and given ad libitum access to food and water. All the protocols were approved by the local ethics committee. To evaluate the effects of lovastatin in vivo, 5- to 7-week-old APCmin mice were fed with lovastatin (100 mg/kg/day) for 8 weeks. Then, blood was harvested through the eyeballs into anticoagulant blood collection tubes timely for further testing of blood routine and liver function. Spleens of mice were surgically removed to be weighed and photographed. In addition, small intestine segments were removed for manual counting of masses. Mass tissues were collected and stored in liquid nitrogen for further reverse transcription polymerase chain reaction (RT-PCR).

Cell Culture

RKO and SW480 cells were cultured in DMEM containing 10% fetal bovine serum and penicillin/streptomycin at 37°C in a 5% CO2 incubator. MVA (79849) and GGPP (G6025) were from Sigma Aldrich (Shanghai, China). NSC23766, ML141, CCG1423, and Y27632 2HCl were from Selleck Chemicals (Houston, TX, USA). Lovastatin was from MedChemEx-press (Shanghai, China).

Western Blotting

A standard protocol was used for Western blotting using primary antibodies specific for β-catenin (ab32572), TAZ (4883S), p-YAP (13008T), p-GSK-3β (5558), p-β-catenin (2009), p-MOB1 (D2F10), and p-LATS1 (9157) from Cell Signaling Technology (Danvers, MA, USA), as well as anti-GAPDH from Abcam (Shanghai, China). Immobilon ECL Ultra Western HRP Substrate (Millipore, Bedford, MA, USA) was used to develop blots, which were imaged with a ChemiDoc MP Imaging System (Bio-Rad, Shanghai, China).

RT-PCR

RNA was isolated using Trizol (Life Technologies, CA, USA), after which an All-in-One First-Strand cDNA Synthesis Kit (Genecopoeia, Rockville, MD, USA) was used to prepare cDNA from 500 ng of RNA per sample. An All-in-One qPCR Mix (Genecopoeia) and an Applied Biosystems 7500 Real-Time PCR detection system were then used for RT-PCR analyses, with 36B4 being used for reference. Primer sequences are listed in Supplement Table 1.

Luciferase Reporter Assay

The 8×GTIIC luciferase reporter (a gift from Stefano Piccolo, Addgene plasmid #34615) and TCF/LEF luciferase reporter (E461A; Promega, Madison, WI, USA) were used to conduct a luciferase reporter assay using the protocols provided by Promega (Promega; Part# 9PIE461), as detailed previously . RKO cells were transfected with luciferase reporter constructs, and 18 h after such transfection (0.15 μg/cm2), cells were treated with ANG II (5 μM) for 24 h, after which they were lysed with the ONE-Glo™ Luciferase Assay System detection reagent (E6110, Promega).

RNA Sequencing

RNA sequencing (RNA-seq) analyses were conducted as in prior reports . Trizol (Life Technologies, CA, USA) was used to isolate total RNA from RKO cells treated for 48 h with lovastatin (5 μM) with or without MVA (0.5 mM). RNA degradation and possible contamination were assessed via agarose gel electrophoresis, with RNA integrity and concentrations being assessed with an Agilent 2100 bioanalyzer instrument. A total of 3 µg of RNA per sample was sequenced with a Hiseq 2500 instrument.

Statistical Analysis

Data are means ± SEM and were compared via Mann-Whitney U tests or Student’s t tests with P < 0.05 as the significance threshold.

Results

Effects of Lovastatin Treatment on APCmin Mice

After 8 weeks of treatment by lovastatin (100 mg/kg/day), we found that the number of colonic masses was significantly reduced (Fig. 1B) and the anemia was significantly improved. The number of leukocytes and erythrocytes and the content of the hemoglobin all increased in lovastatin-treated APCmin mice (Fig. 1A). And triglyceride and cholesterol levels were measured to verify the effect of lovastatins. The results showed a significant decrease in triglyceride and no significant change in cholesterol (Fig. 1A). In addition, previous studies show that aged APCmin mice develop splenomegaly caused by severe anemia[53,54]. Therefore, anemia improvement by lovastatin might suppress extramedullary hematopoiesis, and then decrease the size and weight of spleen. As predicted, our results showed lovastatin treatment indeed significantly improved splenomegaly in APCmin mice (Fig. 1C, D). Considering that APCmin mice present a mutant allele of APC gene has constitutive activation of Wnt signaling. YAP/TAZ is a novel mediator of alternative Wnt signaling . RT-PCR was conducted to test the target gene expression of Wnt/β-catenin signaling and YAP/TAZ signaling. Results revealed that Wnt/β-catenin signaling target genes including AXIN2, Myc, CCND1 and BIRC5 were all downregulated in the lovastatin treatment group (Fig. 1E), as same as YAP/TAZ signaling target genes (CTGF, CYR61, and AMOLT2) (Fig. 1F).

Figure 1.

Effects of lovastatin treatment on APCmin mice. (A) Results of blood routine and liver function. The number of leukocytes and erythrocytes and the content of the hemoglobin, as well as total plasm protein all increased in lovastatin-treated (100 mg/kg/day) APCmin mice. Triglyceride and cholesterol levels decreased by lovastatin treatment. (B) Small intestine segments for manual counting of masses by three divided ranges: <1 mm, 1–2 mm and >2 mm. (C–D) The size (C) and weight (D) of spleen in the NT group and lovastatin treatment group. (E–F) The target gene expression of Wnt signaling (E) and YAP/TAZ signaling (F) in intestine masses of APCmin mice treated with or without lovastatin (*P < 0.05. **P < 0.01). There were no differences in demographic characteristics between the treatment and control group apart from the lovastatin treatment received. YAP: yes-associated-protein; TAZ: tafazzin; WBC: white blood cell; RBC: red blood cell; Hb: hemoglobin; TC: total cholesterol; TG: triglyceride; NT: no treatment.

RNA-Seq Analysis of RKO Cells Treated With Lovastatin With or Without MVA

An RNA-seq analysis performed on RKO cells treated for 48 h with lovastatin (5 μM) with or without MVA (0.5 mM) revealed that lovastatin treatment markedly altered RKO transcriptional profiles (Fig. 2A, B). As shown with volcano plots, relative to the control group, there were 1333 differentially expressed genes (DEGs) in the Lova group (814 upregulated, 519 downregulated) (Fig. 2A). MVA treatments reversed many of these Lova-induced transcriptional changes (Fig. 2B). Heatmaps further revealed significant differences in Lova-treated RKO cells relative to control cells, with CTGF and CYR61 being among the most downregulated genes following Lova treatment (Fig. 2C).

Figure 2.

RNA sequencing of RKO cells treated with lovastatin with or without MVA. (A) Volcano plot depicting the differentially expressed genes detected by RNA sequencing in RKO cells (Lova vs no treatment). (B) Heatmap depicting the differentially expressed genes detected by RNA sequencing in RKO cells. RKO cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, and total RNA was isolated for RNA sequencing. (C) A selection of the most differentially expressed genes in RKO cells treated with lovastatin or lovastatin plus MVA. CTGF and CYR61 were among the most downregulated genes in lovastatin-treated cells, and such downregulation was rescued by MVA. MVA: mevalonic acid.

The Impact of Lovastatin on Canonical Wnt Signaling

Our RNA-seq results suggested that the mevalonate pathway may regulate canonical Wnt signaling. To further confirm this possibility, we assessed Wnt target gene expression levels in RKO cells (Wnt pathway is functioning normally) and SW480 cells (the APC gene is mutated causing a constitutively active Wnt pathway) via RT-PCR. Both in RKO cells and in SW480 cells, treated for 48 h with lovastatin alone or lovastatin plus MVA, respectively, lovastatin treatment significantly (P < 0.05) decreased Wnt target gene expression levels (AXIN2, ENC1, CCND1, BIRC5, and BCL2L1), while this was reversed by MVA treatment (P < 0.05) (Fig. 3A, B). Glycogen synthase kinase-3β (GSK3β) and casein kinase Iα (CKIα) facilitate the phosphorylation of β-catenin, leading to its ubiquitination and degradation[27,56]. The catalytic activity of GSK3β can be inhibited by serine 9 phosphorylation[53,56]; thus, we focused on analyzing the active forms of GSK3b at serine 9 locus. We found that lovastatin inhibited such GSK3β phosphorylation at serine 9, thus contributing to the observed increase in β-catenin phosphorylation and subsequent degradation within RKO cells (Fig. 3C) and SW480 cells (Fig. 3D). Following β-catenin nuclear translocation, it can bind to the TCF/LEF transcription factor and thus modulate gene expression . We therefore explored the ability of lovastatin to modulate TCF/LEF luciferase reporter activity (Fig. 3E, G). Consistent with the above results, lovastatin inhibited TCF/LEF reporter activity (P < 0.05), whereas both MVA and GGPP reversed these changes (P < 0.05). Together, these data suggested that the mevalonate pathway can regulate canonical Wnt signaling.

Figure 3.

The impact of lovastatin on canonical Wnt signaling. (A–B) Lovastatin treatment suppressed the expression of Wnt target genes, while exogenous MVA reversed these changes. RKO (A) and SW480 (B) cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, and RNA was isolated for RT-PCR. *P < 0.05, **P < 0.01 (Lova vs no treatment); #P < 0.05, ##P < 0.01 (L+MVA vs Lova). (C–D) Western blotting was used to assess RKO (C) and SW480 (D) cells treated with lovastatin or lovastatin plus MVA. Note that we focused on GSK3b phosphorylation (p-GSK3b) in Fig. 3C and D. Cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, and protein was then isolated for Western blotting; (E–G) Lovastatin treatment suppressed TCF/LEF reporter activity, while MVA (F) or GGPP (G) restored such activity. The TCF/LEF luciferase reporter was transfected into RKO cells, and then the cells were treated with lovastatin (5 μM), lovastatin plus MVA (0.5 mM), or lovastatin plus GGPP (20 μM) for 48 h, and luminescence was measured. **P < 0.01, ***P < 0.001 (Lova vs no treatment); #P < 0.001 (L+Licl vs Lova); *P < 0.05, **P < 0.01 (Lova vs L+MVA or L+GGPP). MVA: mevalonic acid; GGPP: geranylgeranyl pyrophosphate; RT-PCR: reverse transcription polymerase chain reaction; RFU: relative fluorescence units; DMSO: dimethyl sulfoxide.

The Impact of Lovastatin on YAP/TAZ-Mediated Alternative Wnt Signaling

Our RNA-seq analyses revealed the marked downregulation of the YAP/TAZ target genes CTGF, CYR61, and AMOTL2 following lovastatin treatment, while MVA treatment reversed these changes . They further confirm the ability of lovastatin to regulate both canonical Wnt signaling and YAP/TAZ-mediated alternative Wnt signaling; we thus sought to verify these results in RKO and SW480 cells via RT-PCR (Fig. 4A, B). We found that lovastatin also reduced TAZ protein levels and significantly increased p-YAP and p-LATS1 levels, but p-MOB1 levels were unaffected, whereas MVA addition reversed these changes (Fig. 4C). Together, these data suggest that lovastatin can suppress YAP/TAZ activity in a manner dependent upon LATS1, but not MOB1. Luciferase reporter assay showed that lovastatin was able to potently suppress the activity of the 8×GTIIC luciferase reporter (Fig. 4D), which contains multimerized response elements for TEAD, the primary YAP/TAZ DNA-binding cofactor . The treatment of cells with MVA was sufficient to fully restore YAP/TAZ signaling activity, thus confirming that lovastatin can regulate YAP/TAZ-mediated alternative Wnt signaling.

Figure 4.

The impact of lovastatin on YAP/TAZ-mediated alternative Wnt signaling. We next tested core proteins of YAP/TAZ-mediated alternative Wnt signaling including TAZ, p-YAP, p-LATS1, and p-MOB1. (A–B) Lovastatin treatment suppressed the expression of YAP/TAZ target genes, while exogenous MVA restored them. RKO (A) and SW480 (B) cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, and RNA was isolated for RT-PCR analyses. **P < 0.01, ***P < 0.001 (Lova vs no treatment); #P < 0.05, ##P < 0.01 (L+MVA vs Lova). (C–D) Western blotting analyses of RKO (C) and SW480 (D) cells treated with lovastatin or lovastatin plus MVA. RKO cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, and then proteins were isolated for Western blotting. (E) Lovastatin treatment suppressed 8×GTIIC-Lux reporter activity, while MVA reversed these changes. The 8×GTIIC-Lux reporter was transfected into RKO cells, and then cells were treated with lovastatin (5 μM) or lovastatin plus MVA (0.5 mM) for 48 h, after which the luminescence was measured. ***P < 0.001 (Lova vs no treatment); ***P < 0.001 (Lova vs L+MVA). YAP: yes-associated-protein; TAZ: tafazzin; MOB1: mps one binder 1; MVA: mevalonic acid; RT-PCR: reverse transcription polymerase chain reaction; NT: no treatment.

The Impact of Different Rho GTPase Inhibitors on Wnt-YAP/TAZ Signaling

To gain further insight into the mechanistic basis for lovastatin-mediated inhibition of Wnt-YAP/TAZ signaling, we employed a series of Rho GTPase inhibitors including the Rac1 inhibitor NSC23766 (20 mM), the Cdc42 inhibitor ML141 (20 μM), the ROCK inhibitor Y27632 (1 µM), and the RhoA inhibitors CCG1423 and CT04 (10 mM and 0.1 μg/μl, respectively). Cells were stimulated for 16 h, after which Western blotting was performed revealing that relative to control treatment, β-catenin and TAZ were significantly reduced in the CCG1423 treatment group, whereas p-LATS1 levels were increased (Fig. 5). No other treatments significantly altered levels of any analyzed proteins in these cells. Overall, these results suggest that lovastatin is able to suppress canonical Wnt signaling and alternative YAP/TAZ signaling primarily via suppressing RhoA activation in RKO cells without impacting Cdc42, Rac1, or ROCK.

Figure 5.

The effects of different Rho GTPase inhibitors on Wnt-YAP/TAZ signaling activity. Western blotting assays of RKO cells treated with NSC23766, ML141, CCG1423, and Y-27632 2HCL. RKO cells were treated with NSC23766 (20 mM), ML141 (20 μM), CCG1423(10 mM), or Y-27632 2HCL (1 μM) for 16 h, and then protein was isolated for Western blotting. YAP: yes-associated-protein; TAZ: tafazzin.

Discussion

Lovastatin closely participates in CSC pharmacology. CSCs represent a small portion of cells that have been identified in most human cancers . This subpopulation of cells shares properties similar to normal stem cells such as sphere formation in vitro, self-renewal, and the ability to differentiate into multiple cell lineages . Unlike normal stem cells that undergo differentiation to form cell types with no proliferative potential, CSCs will give rise to progeny that have not undergone terminal differentiation but instead will proliferate uncontrollably giving rise to the bulk formation of the tumor mass . Proliferation of a small subset of drug-resistant CSCs leads to early recurrence, whereas activation of dormant CSCs into their cycling counterparts leads to a later relapse of the disease . Analogous to normal stem cells, CSCs are dependent upon certain signaling pathways that regulate their ability for self-renewal and differentiation.

Two pathways, in particular Wnt/β-catenin and Wnt-YAP/TAZ signaling cascades, may mediate CSC fate. The canonical Wnt/β-catenin cascade is responsible for the self-renewal and differentiation of stem cells . Loss of repression of β-catenin and its following localization to the nucleus leads to binding and activation of the family of transcription factors TCF/LEF. Activation of TCF/LEF leads to transcription of target genes such as CCND1, FGF20, DKK1, MYC, and WISP1 that regulate cell fate . The Wnt-YAP/TAZ pathway is involved in both stem cell survival and differentiation . YAP can also dedifferentiate mature cells into progenitor cells, which is seen in the reprogramming of cancer cells into CSCs . In the alternative Wnt pathway, YAP and TAZ are translocated to the nucleus where they activate the family of transcription factors TEADS and non-TEADS such as RUNX1/2, SMAD, and OCT4, leading to the transcription of genes involved in cell fate and survival . The Wnt/β-catenin and Wnt-YAP/TAZ signaling cascades, among others, are interconnected through the mevalonate pathway .

Due to these overlapping cell signaling cascades and pathological links, the mevalonate pathway has become a promising target for cancer therapy. Mevalonate pathway inhibitors such as lovastatin have been shown to inhibit metastasis and induce cell growth arrest and apoptosis. These factors provide the scientific rationale for probing lovastatin and other mevalonate pathway inhibitors as innovative stem cell–based treatments that warrant additional clinical studies on their use as adjuvants to chemotherapy .

Prior large-scale clinical studies have demonstrated the ability of statins to inhibit colon cancer, but whether the underlying mechanisms governing such inhibition are associated with Wnt signaling remains uncertain[13-16]. Here, we determined the therapeutic benefits of lovastatin on APCmin mice and then explored the mechanisms whereby statins regulate canonical Wnt signaling and alternative Wnt signaling (YAP/TAZ) in an effort to gain insight into these potential underlying mechanisms.

The mevalonate pathway rate-limiting enzyme HMG-CoA reductase acts a target of statins. In RKO cells, lovastatin treatment markedly altered transcriptional profiles detected in an RNA-seq assay. The YAP/TAZ pathway target genes CTGF and CYR61 were among the genes that were most significantly downregulated following lovastatin treatment, while MVA treatment reversed such downregulation. We additionally found that treatment with lovastatin inhibited the phosphorylation of GSK3β, leading to increases in β-catenin phosphorylation and degradation within these RKO cells. Moreover, lovastatin was able to significantly inhibit TCF/LEF luciferase reporter activity, while MVA and GGPP were able to rescue such suppression. RT-PCR, Western blotting, and reporter assays in RKO cells as well as RT-PCR in SW480 cells all confirmed that lovastatin was able to modulate canonical Wnt signaling. In addition, these data suggest that targets of lovastatin within this canonical pathway may lie downstream of GGPP within the mevalonate pathway.

YAP/TAZ may serve as central facilitators of alternative Wnt signaling activity[26,27]. In this study, we observed significant reductions in TAZ protein levels together with an increase in p-YAP levels following lovastatin treatment, whereas MVA addition was sufficient to reverse these changes, in line with previous findings. The upstream YAP/TAZ regulator MOB1 can bind to MTS1/2 and be phosphorylated, in addition to binding and activating the downstream target proteins LATS1/2[37,64]. Mevalonate pathway–mediated YAP/TAZ activation was LATS1/2-independent . In contrast, we found that lovastatin treatment significantly increased p-LATS1 protein levels without altering p-MOB1 levels in treated cells. These findings align with other prior studies, confirming the ability of the mevalonate pathway to regulate YAP/TAZ activity in a manner dependent on LATS1, but not MOB1[65,66].

The ability of statins to inhibit oncogenic processes primarily involves the inhibition of Rho GTPases downstream of GGPP within the mevalonate pathway . To date, studies of the associations between Rho GTPases and canonical Wnt signaling have been limited. Some reports have suggested that Rac1, which is expressed at high levels in tumor cells, can increase β-catenin protein levels, while Cdc42 deletion can drive cytoplasmic β-catenin degradation within tumor cells . Other reports indicate that the activation of RhoA is associated with tumor development, progression, and related processes including cellular invasion, proliferation, and migration . RhoA knockout in xenograft mouse model systems has been shown to decrease colon cancer cell proliferation . These data suggest that overexpressing or activating the Rho GTPases Cdc42, Rac1, and RhoA can enhance tumor cell proliferation in a manner potentially associated with canonical Wnt signaling pathway activation.

While RhoA participates in the alternative Wnt/PCP pathway in the context of embryonic development and certain cancers, in colon cancer cells, RhoA activity was not altered in colon cancer cells following stimulation with the alternative Wnt ligand Wnt5a, suggesting that this Wnt/PCP pathway is not the primary regulator of RhoA within this cancer cell type[69,70]. Rho-YAP/TAZ pathway is capable of regulating YAP/TAZ phosphorylation in the context of alternative Wnt signaling[71,72], with this effect being unrelated to canonical Wnt signaling . In the present report, we found β-catenin and TAZ protein levels to be significantly reduced following RhoA inhibitor (CCG1423) treatment, whereas p-LATS1 levels were significantly elevated. Overall, these results indicate that lovastatin is capable of suppressing both the canonical Wnt/β-catenin and alternative Wnt-YAP/TAZ signaling pathways in CRC cells, owing to its ability to inhibit RhoA activity without affecting Cdc42, Rac1, or ROCK (Fig. 6). However, the mechanisms whereby RhoA suppresses GSK3β phosphorylation and promotes LATS1 phosphorylation remain unclear.

Figure 6.

Lovastatin signaling pathway. Model depicting that lovastatin inhibits RhoA to suppress canonical Wnt/β-catenin signaling and alternative Wnt-YAP/TAZ signaling. YAP: yes-associated-protein; TAZ: tafazzin.

Conclusion

Lovastatin suppresses both canonical Wnt/β-catenin signaling and alternative Wnt-YAP/TAZ signaling in colon cancer cells by inhibiting RhoA.

Click here for additional data file.

Supplemental material, sj-pdf-1-cll-10.1177_09636897221075749 for Lovastatin Inhibits RhoA to Suppress Canonical Wnt/β-Catenin Signaling and Alternative Wnt-YAP/TAZ Signaling in Colon Cancer by Yi Xiao, Qin Liu, Nanyin Peng, Yuzhang Li, Danyang Qiu, Tianlun Yang, Richard Kang, Ahsan Usmani, Efosa Amadasu, Cesario V. Borlongan and Guolong Yu in Cell Transplantation