White button mushroom (Agaricus bisporus) disrupts androgen receptor signaling in human prostate cancer cells and patient-derived xenograft

Abstract

White button mushroom (WBM) (Agaricus bisporus) is a potential prostate cancer (PCa) chemo-preventative and therapeutic agent. Our clinical phase І trial of WBM powder in patients with biochemically recurrent PCa indicated that WBM intake reduced the circulating levels of prostate-specific antigen (PSA). We hypothesized that WBM exerts its effects on PCa through the androgen receptor (AR) signaling axis. Therefore, we conducted a reverse translational study with androgen-dependent PCa cell lines (LNCaP and VCaP) and patient-derived-xenografts (PDX) from a prostate tumor (TM00298). In both LNCaP and VCaP cells, western blots and qRT-PCR assays indicated that WBM extract (6-30 mg/mL) suppressed DHT-induced PSA expression and cell proliferation in a dose-dependent manner. Immunofluorescence analysis of AR revealed that WBM extract interrupted the AR nuclear-cytoplasmic distribution. PSA promotor-luciferase assay suggested that WBM extract inhibited DHT-induced luciferase activity. RNA-Seq on WBM-treated LNCaP cells confirmed that WBM treatment suppressed the androgen response pathways and cell-cycle control pathways. Our PDX showed that oral intake of WBM extract (200 mg/kg/d) suppressed tumor growth and decreased PSA levels in both tumors and serum. In the present study, we also identified a conjugated linoleic acid isomer (CLA-9Z11E) as a strong AR antagonist by performing LanthaScreen TR-FRET AR Coactivator Interaction Assays. The inhibitory effect of CLA-9Z11E (IC50: 350 nM) was nearly two times stronger than the known AR antagonist, cyproterone acetate (IC50: 672 nM). The information gained from this study improves the overall understanding of how WBM may contribute to the prevention and treatment of PCa.

Keywords: 11E)-linoleic acid; Androgen receptor; Cell cycle; Conjugated (9Z; Prostate cancer; Prostate-specific antigen; White button mushroom.

Figure 1.. WBM extract causes a dose-dependent decrease in cell proliferation, cellular AR, and PSA levels.

Cells were treated with the mushroom extract (WBM, 1 μL/mL~5 μL/mL) or enzalutamide (Enza, 5 μM) supplied with 1 nM DH for 12~48h. The proliferation of A) LNCaP and B) VCaP cells was measured by the CyQuant NF Cell Proliferation reagent. The PSA production in the culture medium of C) LNCaP and D) VCaP cells was detected by PSA (human) ELISA kit. Total protein of AR and PSA in E) LNCaP and F) VCaP cells was detected by western blot, and mRNA of AR and PSA in G) LNCaP and H) VCaP cells was quantified by qRT-PCR. The results are mean ± standard deviation of three independent experiments, p values were determined by multiple Student’s t-tests. *p<0.05, **p<0.01, ***p<0.001. ns. No significant difference.

Figure 2.. WBM extract interrupts AR expression, AR nuclear-cytoplasmic distribution, and transactivation.

Cells were treated with the mushroom extract (WBM, 1 μL/mL~5 μL/mL) or enzalutamide (Enza, 5 μM) supplied with 1 nM DHT for 48h. The AR cellular localization in A) LNCaP and B) VCaP cells was determined by immunofluorescence staining. Magnification at 20X and 100X with a scale bar of 100 μm and 20 μm, respectively. The protein amount of AR in the cytoplasmic and nuclear fractions of C) LNCaP and E) VCaP cells was determined by western blot. GAPDH and nucleolin were used as the cytoplasmic and nuclear loading controls, respectively. The AR reporter activity in D) LNCaP and F) VCaP cells was determined by PSA promoter-Luciferase assay. **p<0.01. ns. No significant difference.

Figure 3.. WBM extract suppresses the expression of AR responsive genes.

LNCaP cells were treated with 1 nM DHT, with or without 6X mushroom extract (3 μL/mL), for 48h. Total RNA was sequenced. A) Significant regulated genes between Ctrl (1 nM DHT, 3 samples) versus Treatment (1 nM DHT and WBM extract, 3 samples) were demonstrated by a heat map. There were 155 up-regulated genes (Cluster A, Fold-change ≥ 2, p<0.05) versus 185 down-regulated genes (Cluster B, Fold-change ≤ 0.5, p<0.05). B) Identification of pathways was analyzed by Gene Set Enrichment Analysis (GSEA) with hallmarks of datasets. False discovery rate (FDR) was <0.01, Number of significant genes (NES) >40. C) Three RNA-Seq datasets derived from LNCaP cells with treatments of mushroom extract (3 μL/mL WBM with 1 nM DHT, 48h), enzalutamide (10 μM Enza with 1 nM DHT, 48h,) and DHT (1 nM DHT, 48h) were applied to enrich AR responsive genes. D) Venn Diagram Analysis demonstrated the distribution of the number of genes in three datasets. There were 39 common genes among 3 datasets. Top 5 GSEA-Hallmarks of 39 genes (p<0.05, FDR<0.25). Key Genes involved in Top5 hallmarks were quantified by qRT-PCR in LNCaP E) and VCaP F) cells.

Figure 4.. CLA-9Z11E inhibits AR coactivator interaction and PSA expression.

AR Coactivator Interaction Assay in A) antagonist and B) agonist models were performed to test the ability of CLA-9Z11E to affect coactivator peptide recruitment and binding to AR. Cyproterone acetate (IC50: 672 nM) was used as a positive control for the antagonist assay and R1881, a synthetic AR agonist, was used for the agonist assay. Data are expressed as mean ± standard deviation of duplicate assays. LNCaP and VCaP cells were treated by 1 nM DHT with CLA-9Z11E (10 μM~500 μM) for 48h. AR reporter activity in C) LNCaP and D) VCaP cells was determined by PSA promoter-luciferase assay. The mRNA levels of AR and PSA in E) LNCaP and F) VCaP cells was quantified by qRT-PCR. The results are expressed as mean ± standard deviation of three independent experiments, p values were determined by multiple Student's t-tests. *p<0.05, **p<0.01, ***p<0.001. ns. No significant difference.

Figure 5.. Intake of WBM extract suppresses PDX tumor growth and PSA expression.

A) Male 6^~8-week-old NSG mice bearing PCa PDX fragments were gavaged with WBM extract (6 mg/mice/day, 5 mice in WBM group), enzalutamide (300 μg/mice/day, 2 mice in Enza group) or PBS (2 mice in Ctrl group) for 6 days. B) The tumor volume in mm³ is represented by individual values in each group by different colors. C) Tumor xenografts in each group are illustrated and tumor weights are represented by individual values in each group by different colors. D) Hematoxylin and eosin (H&E) and immunohistochemistry (IHC) staining of AR, PSA, and Ki67 on formalin-fixed tumor tissues in each group. The IHC staining was scored (upper panel) by the QuPath software (version 0.2). Immunoreactivity of AR was reported as H score (upper left panel). Ki67 and PSA were reported by their percentage of positive cells (upper right panels). Scale bar of representative images is 40X by 100 μm. E) The mRNA level of AR, PSA, and TOP2 in tumor fragments of each group was quantified by qRT-PCR. F) The PSA concentrations in serum were detected by PSA (human) ELISA kit. G) The prostate gland weight in each group is presented as a median±standard error for the mean. H) Hematoxylin and eosin (H&E) and immunohistochemistry (IHC) of AR staining on the formalin-fixed prostate gland in each group. The AR-positive complex acini in anterior prostate gland are shown. Scale bar for 10X and 40X, with scale bar of 400 μm and 100 μm, respectively. p values were determined by multiple Student's t-tests. *p<0.05, **p<0.01, ***p<0.001. ns. No significant difference.