Fig 2: ADT-upregulated ZBTB46 is associated with the NGF and promotes neuroendocrine differentiation of prostate cancer.a GSEA of TCGA prostate cancer dataset showed that higher ZBTB46 expression of prostate tissues was positively associated with NEPC-responsive gene signature. NES normalized enrichment score, FDR false discovery rate. b Relative expression of ZBTB46, neuroendocrine markers, and androgen-responsive genes determined by RNA-Seq analysis with LNCaP cells expressing the NGF vs. LNCaP cells expressing an empty vector (EV). The relative expression was determined by the ratio of LNCaP-NGF/LNCaP-EV. c Western blotting of ZBTB46, NGF, CHGA, ENO2, NKX3-1, and KLK3 in various prostate cancer cells. d Relative mRNA levels of ZBTB46, NGF, CHGA, CHGB, SYP, and ENO2 in PC3 cells with stable expression of the non-target control (NC) or ZBTB46 shRNA vector. e, f ZBTB46, NGF, CHGA, CHGB, SYP, and ENO2 mRNA levels in NC and ZBTB46-knockdown C4-2 (e) and LNCaP (f) cells cultured with 10% charcoal stripped serum (CSS)-containing medium for 1 week. * vs. Fetal bovine serum (FBS); # vs. the NC. g Western blotting of ZBTB46, NGF, CHGA, and ENO2 in ZBTB46-knockdown C4-2 and LNCaP cells treated with CSS-containing medium for 1 week. h, i ZBTB46, NGF, CHGA, and ENO2 mRNA (h) and protein (i) levels in C4-2 and MDV3100-resistant C4-2 (C4-2-MDVR) cells following ZBTB46-knockdown. * C4-2 vs. C4-2-MDVR; # vs. the NC. Data from the quantification of mRNA are presented as the mean ± SEM of three independent experiments; n = 3 per group. *p < 0.05, **p < 0.01, ***p < 0.001; by a two-way ANOVA. j ZBTB46, NGF, CHGA, and ENO2 protein levels in LNCaP and C4-2 cells stably expressing an EV or a ZBTB46 cDNA expression vector.

Fig 3: CHRM4 activates AKT and MYCN signaling and promotes neuroendocrine differentiation of prostate cancer.a GSEAs of TCGA prostate cancer dataset showed that higher expression levels of both the NGF and ZBTB46 in prostate cancer tissues were positively associated with a NEPC-responsive gene signature25. NES normalized enrichment score, FDR false discovery rate. b, c Relative mRNA levels of the top 10 candidate genes from Supplementary Table 6 in LNCaP cells with an empty vector (EV) or NGF cDNA vector overexpression (b) or in PC3 cells with non-target control (NC) or NGF shRNA vector expressions (c). d Western blotting of CHRM4 in various prostate cancer cell lines. e Western blotting of CHRM4, CHRM1, CHRM3, p-AKT, AKT, and MYCN in LNCaP cells treated with various concentrations of the NGF protein in charcoal-stripped serum (CSS)-containing medium for 1 week. Fetal bovine serum (FBS)-containing medium without the NGF protein served as a control. f Western blotting of CHRM4, p-AKT, AKT, and MYCN in PC3 cells stably transfected with the NC or CHRM4 shRNA vector, and cells were treated with 100 ng/ml NGF protein in CSS-containing medium for 1 week. g CHRM4, p-AKT, AKT, and MYCN protein levels in C4-2 and C4-2-MDVR cells following stable CHRM4-knockdown. h CHRM4, CHRM1, and CHRM3 protein levels in LNCaP and C4-2 cells cultured in CSS-containing medium for 1 and 2 weeks, followed by treatment with 10 nM dihydrotestosterone (DHT) for 1 day. i Protein levels of CHRM4, p-AKT, AKT, and MYCN in LNCaP and C4-2 cells following stable CHRM4-knockdown and cultured in CSS-containing medium for 1 week. j Relative mRNA levels of CHRM4, CHGA, CHGB, SYP, and ENO2 in PC3 cells following stable expression of NC or CHRM4 shRNA vector and treated cells with 100 ng/ml NGF protein in CSS-containing medium for 1 week. * vs. Phosphate buffered saline (PBS); # vs. the NC. Data from the quantification of mRNA are presented as the mean ± SEM of three independent experiments; n = 3 per group. *p < 0.05, **p < 0.01, ***p < 0.001; by a two-way ANOVA.

Fig 4: The effects of tPBM on PTZ-induced neuroinflammation and astrogliosis in the CA3 region. Double immunofluorescence staining of NSE (red) and GFAP (green) in the CA3 illustrates the effect of tPBM on neuroinflammation (indicated by NSE) and astrogliosis (GFAP) and their combination. NSE was stained in the cytosol of inflammatory neurons. Pyramidal neurons were aligned at SP. Neurons located at other layer such as at SO were GABAergic interneurons. Astrocytes were stained for GFAP, and an increase in GFAP immunoreactivity, somata size, and branching indicated astrogliosis. DAPI (blue) was used as a counterstain for the nucleus (scale bar=50 µm). The two images in the lower panel are enlargements of the merged images from the PTZ and tPBM + PTZ groups, respectively. Enlargement: a GABAergic interneuron (labeled “G”) based on the morphology with smaller nuclear–cytoplasmic ratio compared with that of principle cells; the process of an astrocyte (A) was adjacent to the GABAergic interneuron (scale bar=10 µm). Stratum lucidum, SL; stratum radiatum, SR.

Fig 5: Immunocytochemical analysis of ADSCs on d 30 after induction (a), with β-ME, NIM, and AZC showed NSE positivity and NF-L negative. The resultant cells are spheroids with minute neurite projections, resembling poorly differentiated structures (a, b). ADSCs on d 30 after induction with β-ME and NIM without AZC showed NSE and NF-L positivity. They resultant cells are not spheroids (c, d) (Magnification 20 x). Semi-quantitative gene expression profiling by RT-PCR of ADSC-derived neural-lineage cells at d 30 after induction (b). Neural – lineage cells obtained through the treatment of NIM without 5-azacytidine (NC WO AZC) and Neural – lineage cells obtained through the treatment of NIM with 5-azacytidine (NC with AZC). Untreated ADSCs used as control. The analysis shows reduced expression of neurofilaments.