Nuclear Pores Promote Lethal Prostate Cancer by Increasing POM121-Driven E2F1, MYC, and AR Nuclear Import

Abstract

Nuclear pore complexes (NPCs) regulate nuclear-cytoplasmic transport, transcription, and genome integrity in eukaryotic cells. However, their functional roles in cancer remain poorly understood. We interrogated the evolutionary transcriptomic landscape of NPC components, nucleoporins (Nups), from primary to advanced metastatic human prostate cancer (PC). Focused loss-of-function genetic screen of top-upregulated Nups in aggressive PC models identified POM121 as a key contributor to PC aggressiveness. Mechanistically, POM121 promoted PC progression by enhancing importin-dependent nuclear transport of key oncogenic (E2F1, MYC) and PC-specific (AR-GATA2) transcription factors, uncovering a pharmacologically targetable axis that, when inhibited, decreased tumor growth, restored standard therapy efficacy, and improved survival in patient-derived pre-clinical models. Our studies molecularly establish a role of NPCs in PC progression and give a rationale for NPC-regulated nuclear import targeting as a therapeutic strategy for lethal PC. These findings may have implications for understanding how NPC deregulation contributes to the pathogenesis of other tumor types.

Keywords: E2F1; GATA2; MYC; POM121; androgen receptor; importin β; nuclear import; nuclear pore; nuclear transport; prostate cancer.

Figure 1.. Advanced Lethal PC NPCs Display a Distinct Nup Composition that Contributes to PC Aggressiveness

(A) Heatmap of Nups in primary and metastatic warm-autopsy PC tumor tissues (GSE35988). Magnitude (t-statistic) and statistical significance (false discovery rate [FDR]) of differential expression between the groups are shown as bar graph for each Nup gene. Red and blue colors indicate high and low gene expression, respectively. (B–D) Representative transmission electron microscopy imaging (scale bar, 5 μm) (B) and quantification of (C) number of pores and (D) NE spacing in primary (n = 3) and advanced (n = 3) PC cells from human tumor samples. Red arrows point to NPCs. Blue dot lines point to NE spacing. Scale bar, 500 nm. (E) Quantification of soft agar colony formation assays of aggressive PC cells transfected with control siRNA or two siRNAs targeting each upregulated Nup. (F) Quantification of cell population doublings from (E). (G) Quantification of colony formation assays of cells from (E) following 72 hr treatment with docetaxel (125 nM) and mitoxantrone (500 nM). Data represent the mean ± SD of 3 experiments. *p ≤0.05. See also Figure S1 and Table S1.

Figure 2.. Nucleoporin POM121 Promotes Tumorigenesis, Proliferation, and Survival to Standard Therapies in PC Cells

(A) Immunohistochemistry and quantification of POM121-positive cells during disease progression in a series (n = 124) of human paraffin-embedded PC tissues.Scale bar, 100 μm. (B) Transmission electron microscopy images and quantification of POM121 Immunogold stained protein localized in NPCs of PC cells from primary (n = 3) and advanced (n = 3) PC tissues. 120 NPCs where analyzed for each tumor sample from a minimum of 5 images. Scale bar, 100 nm. (C) Immunoblot of POM121 in aggressive PC cells (DU145-DR and 22Rv1-DR) compared to parental cells (DU145 and 22Rv1). (D) Immunofluorescence of POM121 and Nups (NUP214, NUP98, and NUP62) detected by mAb414 antibody in aggressive PC cells transfected with control andPOM121 siRNAs. Scale bar, 5 μm. (E) Immunoblot of POM121 and NUP62 in parental cells transfected with an empty, POM121, or NUP62 vector. (F) Soft agar colony formation assays and quantifications of cells from (E). Scale bar, 100 mm. (G) Quantification of cell population doublings of cells from (E). (H) Colony formation assays and quantification of cells from (E) following 72 hr treatment with docetaxel (5 nM), cabazitaxel (1 nM), mitoxantrone (5 nM), and singledose radiation (2 Gys). Data represent the mean ± SD of 3 experiments. *p < 0.05. See also Figure S2

Figure 3.. In Vivo POM121 Depletion Decreases PC Aggressiveness

(A) Experimental design used to test the in vivo effects of targeting POM121 on tumorigenesis, tumor growth, and response to therapy. (B) Immunoblot of POM121 in DU145-DR and 22RV1-DR cells transduced with a control shRNA and two inducible shRNAs targeting POM121 after 72 hr ofdoxycycline (1 μg/mL) exposure. (C) Tumor incidence and latency of 100 cells from (B) subcutaneously injected into NSG mice. 12 injection sites for each experimental condition were monitored weekly during 6 months for tumor formation. (D) Volumes of DU145-DR and 22Rv1-DR control or POM121 doxycycline-induced shRNAs subcutaneous xenografts after 28 days of combination treatmentwith vehicle, docetaxel (10 mg/kg intraperitoneally [i.p.] weekly), or mitoxantrone (10 mg/kg i.p. weekly). Tumor volumes of 12 xenografts for each treatment condition where analyzed. *p ≤0.05 = Comparison between control and shPOM121. **p ≤ 0.05 = comparison between shControl and shPOM121 treated with docetaxel or mitoxantrone. (E) Immunohistochemistry and quantification of Ki67 expression in tumor xenografts from (C). Scale bar, 100 μm. (F) Immunohistochemistry and quantification of cleaved caspase 3 expression in DU145-DR tumor xenografts from (D). Scale bar, 100 μm. Data represent the mean ± SD. *p < 0.05. See also Figure S3

Figure 4.. POM121 Regulates PC Cell Aggressiveness through Its Interaction with the Nuclear Import Machinery

(A) GFP and POM121 immunofluorescence and quantification of cytoplasmic versus nuclear fluorescent signal in DU145-DR and 22Rv1-DR cells stablyexpressing a (GR)-GFP reporter after transfecting control siRNA and two siRNAs targeting POM121, treated with dexamethasone (100 nM) and the export XPO-1 inhibitor selinexor (1 μM) for 10 min. Data represent the mean ± SD quantification of 50 cells for each condition. Scale bar, 5 μm. (B) Immunoblots of POM121 after immunoprecipitation of importin β in parental and aggressive PC cells. (C) GFP immunofluorescence confocal images of WT POM121-GFP-siRNA-resistant and ΔNPC POM121-GFP-siRNA-resistant DU145-DR cells. NE localization and integrity is determined by TPR staining. Scale bar, 5 μm. (D) POM121 immunoblot after anti-GFP immunoprecipitation in cells from (C). (E) Soft agar colony formation assays and quantification of cells from (C) transfected with control siRNA or two siRNAs targeting POM121. Scale bar, 100 μm. (F) Quantification of cell population doublings of cells from (E). (G) Colony formation assays and quantifications of cells from (E) following 72 hr treatment with docetaxel (125 nM), cabazitaxel (25 nM), mitoxantrone (500 nM),and single dose radiation (5 Gys). (H) Immunoblots of FG-Nups (NUP153, POM121, and NUP62) following immunoprecipitation of importin β in paired parental and aggressive PC cells. (I) SDS-PAGE Coomassie brilliant blue (CBB) of GST-tagged human WT or ΔNPC POM121 tested for binding to recombinant human His-tagged importins α and β after His-tag in vitro pull-down assays. (J) Transmission electron microscopy images and quantification of number of NPCs in parental cells transduced with an empty or POM121 vector. Red arrows point to NPCs. Scale bar, 5 μm. Data represent the mean ± SD. *p < 0.05. See also Figure S4

Figure 5.

POM121 Facilitates the Selective Nuclear Import of Cell Tumorigenic-, Proliferation-, and Survival-Conferring Transcription Factors in Lethal PC (A) Expression pattern of genes modulated by siRNA-mediated POM121 knockdown in DU145-DR and 22Rv1-DR cells. Magnitude (t-statistic) and statistical significance (false discovery rate [FDR]) of differential expression between the groups are shown as bar graph for each gene. Red and blue colors indicate high and low gene expression, respectively. (B) Modulation of E2F1 and MYC target gene signatures determined by transcriptome meta-analysis by siRNA-mediated POM121 knockdown in PC cells (GeneSet Enrichment Analysis [GSEA]). (C)Modulation of POM121, MYC and E2F1 target gene signatures determined by siRNA-mediated in vitro gene knockdown or reported in literature (Table S3) in primary and warm-autopsy tumor tissues (GSE35988). Orange and green colors indicate statistical significance (FDR) of induction and suppression of the target gene signatures, respectively (modified GSEA). (D)Immunoblot of MYC and E2F1 levels in the nucleoplasm and cytoplasm of aggressive PC cells after 72 hr of being transfected with control siRNA and twosiRNAs targeting POM121. (E) Immunoblot of β-catenin and hnRNP-A1 levels in the nucleoplasm and cytoplasm of cells from (D). (F) Immunoblots of MYC, E2F1, and β-catenin levels in the nucleoplasm and cytoplasm of parental cells transduced with an empty vector (EV) or POM121 vector. (G) Immunoblot of AR levels in the nucleoplasm and cytoplasm of 22Rv1-DR cells transfected with control siRNA and two siRNAs targeting POM121 (72 hr) and further cultured in FBS free conditions and exposed to 100 nM dihydrotestosterone (DHT) during 18 hr. (H) Quantitative analysis of AR target genes KLK3 and TMPRSS2 mRNA levels of cells from (G). Data represent the mean ± SD of 3 experiments. *p < 0.05. See also Figure S5 and Tables S2–S5.

Figure 6.. A GATA2-POM121 Regulatory Feedback Loop Modulates PC Aggressiveness

(A and B) mRNA (A) and protein (B) levels of POM121 in aggressive PC cells transfected with siRNA control and two siRNAs targeting GATA2. (C) Representation of three predicted GATA2 binding elements (GBEs) in the POM121 promoter region. (D) ChIP-qPCR of GATA2 occupancy at GBE1, GBE2, GBE3, and flanking control region (NEG) in the POM121 promoter of aggressive PC cells. (E) Luminescence analysis of HEK293 cells following co-transfection with an empty or GATA2 expression vector, a POM121 promoter luciferase reporter (wildtype [WT], mutated GBE1 [mutGBE1], mutated GBE2 [mutGBE2], or mutated GBE3 [mutGBE3]), and a Renilla transfection control. *p < 0.05. **p < 0.01 relative to control vector. (F) Modulation of POM121 target gene signature in GATA2-depleted cells (accession number GSE58966). GSEA. NES, normalized enrichment score; FDR, falsediscovery rate. (G) Immunoblot of GATA2 levels in the nucleoplasm and cytoplasm of aggressive PC cells after 72 hr of being transfected with control siRNA and two siRNAstargeting POM121. (H) Representation of the GATA2-POM121-positive feedback regulation. GATA2 regulates the transcription of POM121 and POM121 regulates the nuclear localization of GATA2. (I) Correlation between POM121 and GATA2 gene expression levels in public available clinical PC databases (accession numbers GSE35988, dbGap: phs000915.v1.p1 and dbGaP: phs000447.v1.p1). Spearman correlation test rho and p value are shown. (J) Immunohistochemical expression of POM121 and GATA2 in a PC tissue samples. Statistically association between proteins was analyzed by χ² test. Data represent the mean ± SD of 3 experiments. Scale bar, 100 μm. *p < 0.05. See also Figure S6.

Figure 7.. Targeting the POM121-Importin β Axis Decreases the Growth and Survival of Lethal PC Pre-clinical Models

(A) Soft agar colony formation assays and quantification of aggressive PC cells treated during 72 hr with vehicle (Control) or Importazole 1 μM. Scale bar, 100 μm. (B) Quantification of cell population doublings of cells from (A). (C) Representative colony formation assays and quantification of cells from (A) following 72 hr treatment with docetaxel (125 nM), cabazitaxel (25 nM), mitoxantrone (500 nM), and single dose radiation (5 Gys). (D) Immunoblot of MYC, E2F1, GATA2, and AR levels in the nucleoplasm and cytoplasm of aggressive PC cells after 72 hr treatment with DMSO orImportazole 1 μM. (E) Bioluminescence analysis of mice bearing 22Rv1-DR luciferase-expressing xenografts treated with vehicle or Importazole (100 mg/kg i.p. daily 5 days a week) during 28 days. Data represent the mean ± SD of 15 mice for each treatment condition. (F) Representative image and weight quantification of tumor xenografts from (E). Data represent the mean ± SD of 15 mice for each treatment group. (G) Representative immunohistochemistry images and quantification of cleaved caspase 3 expression in 22Rv1-DR, LPC#1, and LPC#2 xenografts after 7 days of treating mice with docetaxel (10 mg/kg i.p. weekly) and mitoxantrone (10 mg/kg i.p. weekly) alone or in combination with Importazole (100 mg/kg i.p. daily).Scale bar, 50 μm. (H) Kaplan-Meier survival curves of NSG mice following intracardially injection of 22Rv1-DR, LPC#1, and LPC#2 cells treated as in (G). 15 mice for each treatment group were analyzed. Data represent the mean ± SD of 3 experiments. *p < 0.05. See also Figure S7.