Nutraceutical Evaluation of Trigonelline's Therapeutic Potential by Targeting Bladder Cancer Stem Cells and Cancer-Associated Fibroblasts via Downregulation of TGFβ3/GLI2/YAP1 Signaling Hub

Abstract

Trigonelline (TGN), an alkaloid identified in medicinal plants such as coffee (Coffea spp.) and fenugreek (Trigonella foenum-graecum), has demonstrated significant anticancer properties across various malignancies, yet its efficacy in bladder cancer (BLCA) remains underappreciated. This study investigates TGN's role in modulating cancer stem cells (CSCs) and the tumor microenvironment (TME), two key contributors to BLCA progression and chemoresistance. Through comprehensive bioinformatics analyses of BLCA patient datasets, a TGY signature (TGFβ3, GLI2, YAP1) was identified as a critical signaling hub associated with poor prognosis, therapeutic resistance, and CSC generation. Computational docking studies revealed TGN's high binding affinity to the TGY signature, TGFβ3 (ΔG = -3.9 kcal/mol), GLI2 (ΔG = -4.2 kcal/mol), YAP1 (ΔG = -3.4 kcal/mol), suggesting its potential to disrupt this signaling axis. In vitro experiments demonstrated that TGN effectively inhibited BLCA cell proliferation, colony formation, and tumorspheroid growth while significantly enhancing cisplatin sensitivity in resistant cell lines. Notably, TGN reduced the transformation of fibroblasts into cancer-associated fibroblasts (CAFs) through the downregulation of α-SMA and FAP (Fibroblast activation protein) expression, indicating its capacity to normalize the TME. Real-time PCR analysis revealed that TGN treatment significantly reduced markers of epithelial-mesenchymal transition and stemness pathways. Our preclinical mouse study demonstrated that combining TGN and cisplatin significantly reduced tumorigenesis in cisplatin-resistant bladder tumoroids harboring CAFs. Importantly, this combination therapy showed no apparent systematic toxicity, suggesting a favorable safety profile. Our findings reveal novel molecular targets of TGN in bladder cancer; TGN acts as a potent disruptor of the TGY signaling axis and a normalizer of the TME by reducing CAF transformation. In sum, our findings advocate for TGN's further exploration as a candidate for combination therapy in drug-resistant BLCA, with the potential to improve patient outcomes by simultaneously targeting both CSCs and the TME, serving as a foundation for future clinical trials.

Keywords: TGFβ3/GLI2/YAP1 signaling; bladder cancer; cancer stem cells; cancer-associated fibroblasts; cisplatin resistance; trigonelline.

Figure 1

Increased TGFβ3/GLI2/YAP1 (TGY signature) signaling is predominately associated with EMT and stemness in bladder cancer patients. (A) A heatmap depicts the expression correlation between GLI1 and GLI2 and TGFβ3 genes in TCGA cancer types. Notably, TGFβ3 is expressed highest among the bladder cancer cohort's three TGFB family members (1, 2, and 3). (B) A heatmap represents the Pearson correlations between GLI2 and TGFβ3 and the signature of EMT/stemness in TCGA cohorts. (C) TGFβ3 gene enrichment analysis. The bubble plot shows the signaling and metabolic pathways curated from the KEGG database 2021. (D) Kaplan-Meier survival curves showing the expression of TGFβ3/GLI2/YAP1 in TCGA cohorts of bladder cancer. (E) Kaplan-Meier survival curves demonstrate that the higher expression of the TGY signature is associated with lower overall survival (right) and disease-free survival (left) of the TCGA bladder cancer cohorts.

Figure 2

TGFβ3 expression patterns are associated with bladder cancer progression and immune-suppressive tumor microenvironment (TME). (A) TCGA database analyses revealed that TGY signature expression was significantly correlated with various immune cells and CAFs (p < 0.01). (B) CAF was found to be the predominant infiltrated immune cells. (C). High expression of the TGY signature statistically worsens the OS rate of mUC patients. (D). TGFβ3/GLI2/YAP1 was closely associated with the CAF-mediated T cell exclusion phenotype.

Figure 3

Elevated TGFβ3, GLI2, and YAP1 expression in bladder tumorspheroids and CAF infiltration. (A) Flow cytometric analysis demonstrates a higher % CD44 cells in the T24 and 5637 bladder tumorspheroids than their parental counterparts. (B) Representative images (left panel) of the tumor spheroid-generating ability of the CD44⁺ T24 and 5637 cells when compared with their parental counterpart; the right panel shows the spheroid quantification. (C) Comparative western blots show the upregulation of TGFβ3, GLI2, and YAP1 in CD44⁺ T24 and 5637 tumorspheroids compared with their parental counterparts. (D) Microscopic images of cancer-associated fibroblasts (CAF) infiltration levels between the CD44⁺ tumorspheroids and their parental counterparts. The intensity of the red fluorescence reflects the CAF infiltration. The fluorescence quantification is displayed in the bar graphs (right panel). (E) Dot blots of IL-6 secretion comparison between the CD44⁺ CAF-infiltrated tumorspheroids and their parental counterparts. (F) Quantitative PCR analysis comparing the mRNA levels of TGFβ3, GLI2, and YAP1, α-SMA (alpha-smooth muscle actin), vimentin, and FAP (Fibroblast activation protein) between the CD44+ CAF-infiltrated tumorspheroids than their parental counterparts.

Figure 4

TGFβ3-silencing resulted in significantly reduced bladder tumorigenic properties. (A) Bar graph of mRNA expression levels confirming the TGFβ3-silencing effect of siRNA on T24 and 5637 cells. Western blotting shows the downregulation of TGFβ3, GLI2, YAP1, and β-catenin in TGFβ3-silenced T24 and 5637 cells compared with the siCtrl. Microscopic image (left panel) showing the (B) reduced tumorsphere generating ability, and (C) CAF-infiltration of TGFβ3-silenced T24 and 5637 cells when compared with the siCtrl: the left panel shows the quantification of the spheroid and CAF-infiltration respectively. (D) Bar graph showing the downregulation of the mRNA levels of α-SMA, vimentin, and FAP in the TGFβ3-silenced T24 and 5637 cells as compared with the siCtrl counterparts. The dot blot on the lower panel also shows decreased TGFβ3 secreted in the TGFβ3-silenced cells than their siCtrl counterparts. (E) A drug-response curve showing higher sensitivity of TGFβ3-silenced 5637 and T24 cells to cisplatin treatment than the siCtrl. Results are expressed as mean ± SD of assays performed 3 times in triplicate. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

In vitro characterization of TGN's anticancer functions and CAF transformation. (A) Molecular docking simulation of TGN: TGFβ3 binding complex. (i) A 3D molecular complex demonstrates TGN (red) in TGFβ3's binding pocket. (ii) A stick illustration demonstrates the amino acid residues participating in the hydrophobic interactions within the pocket. The numbers represent the molecular distance of the bond. (iii) A 2D ligand interaction map of the TGN: TGFβ3 binding complex. This illustration shows the interacting amino acids, the type of interaction, and the bond distance. (iv) The table shows the binding energy (Gibb's free energy) calculated in each complex. (B) Effect of TGN on the tumor spheroid-forming ability of T24 and 5637 cells. The right panel represents a quantitative analysis of the spheroid formation between the TGN-treated (11 and 7 µM, 72 hs, for 5637 and T24 spheres, respectively) and the control cells. (C) Flow cytometric analysis shows the decreased percentages of CD44⁺ cells in TGN-treated T24 and 5637 cell lines compared to the control (non-treated) counterparts. (D) Representative micrographs demonstrating the TGN's inhibitory effects on CAF-infiltration in T24 and 5637 tumor spheroids (denoted as TMD). The bar graphs on the right represent the quantitative analysis of CAF infiltration levels between TGN-treated and control tumorspheroids. (E) The drug-response curve showed that TGN treatment increased the sensitivity of 5637 and T24 tumoroids to cisplatin treatment. (F) Western blots comparing the expressions of TGFβ3, GLI2, YAP1, and β-catenin in TGN-treated and control T24 and 5637 TMD. (G) Comparative qPCR analysis shows TGN-treated bladder tumorspheroids (TMD) expressed significantly lower mRNA expression levels of α-SMA, vimentin, and FAP than their control counterparts. (H) Comparative dot blot analysis of the culture media from control and TGN-treated tumoroids (11 and 7 µM, 72 hs, for 5637 and T24 TMD, respectively). TGFβ3, IL-6, and VEGF levels were markedly lower in the TGN-treated group. Results are expressed as mean ± SD of assays performed 3 times in triplicate. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6

TGN treatment delayed tumoroid growth and overcame cisplatin resistance in a mouse xenograft model. (A) The tumor volume versus time graph demonstrates the effect of trigonelline (TGN) monotherapy, cisplatin (CDDP) monotherapy, and combination therapy in a CAF-containing 5637 tumoroid xenograft mouse model. Vehicle control (PBS), trigonelline monotherapy (50 mg/kg, 3 times/week, i.p), cisplatin monotherapy (1 mg/kg, 3 times/week, i.p), and combination therapy (TGN: 50 mg/kg, 3 times/week, i.p; cisplatin: 1 mg/kg, 5 times/week, i.p). N=5 for each group. The upper right panel shows the average tumor weight from each group; the lower right panels demonstrate the photos of the tumor samples harvested after experiments. (B) Mean body weight versus time graph. The body weight of the mice was recorded weekly. The average weights were then plotted against time for monitoring purposes. (C) Tumor spheroid forming assay. The secondary tumor spheroid-forming ability of the tumor cells harvested from each group was analyzed. The diameter represented the average size of the tumor spheroids. (D) Quantitative PCR analysis comparing the expression of TGFβ3/GLI2/YAP1 (the TGY signature) and CAF markers (α-SMA and vimentin) among different groups. Results are expressed as mean ± SD of assays performed 3 times in triplicate. **p < 0.01, ***p < 0.001, ****p < 0.0001.