Inhibition of KPNA4 attenuates prostate cancer metastasis

Abstract

Prostate cancer (PCa) is a common cancer in men. Although current treatments effectively palliate symptoms and prolong life, the metastatic PCa remains incurable. It is important to find biomarkers and targets to improve metastatic PCa diagnosis and treatment. Here we report a novel correlation between karyopherin α4 (KPNA4) and PCa pathological stages. KPNA4 mediates the cytoplasm-to-nucleus translocation of transcription factors, including nuclear factor kappa B, although its role in PCa was largely unknown. We find that knockdown of KPNA4 reduces cell migration in multiple PCa cell lines, suggesting a role of KPNA4 in PCa progression. Indeed, stable knockdown of KPNA4 significantly reduces PCa invasion and distant metastasis in mouse models. Functionally, KPNA4 alters tumor microenvironment in terms of macrophage polarization and osteoclastogenesis by modulating tumor necrosis factor (TNF)-α and -β. Further, KPNA4 is proved as a direct target of miR-708, a tumor-suppressive microRNA. We disclose the role of miR-708-KPNA4-TNF axes in PCa metastasis and KPNA4's potential as a novel biomarker for PCa metastasis.

Figure 1. KPNA4 expression is positively correlated with human prostate cancer progression

(a). Immunohistochemical staining of KPNA4 in paraffin embedded human prostate cancer tissue microarray slides. (b) Statistics analysis of the relationship between KPNA4 expression and PCa stages as indicated. Analysis of the differential KPNA4 RNA expression level of PCa patient cases of the TCGA dataset that is classified by either (c) the pathologic T or (d) the Gleason score. *p<0.05, **p<0.01, ***p<0.001 (e) Endogenous KPNA4 protein expression level in normal and malignant prostate cell lines were determined by Western blotting, β-actin was used as loading control.

Figure 2. Silencing of KPNA4 suppresses the migration ability of PCa cells

(a) Knockdown efficiency of KPNA4 siRNAs were determined by Western blotting, α-tubulin was used as loading control. (b) Wound healing analysis and (c) Transwell assay was performed to determine the cell migration and invasion of PC3 cells which are transfected with either KPNA4 siRNA or control siRNA. *p<0.05, **p<0.01, ***p<0.001 (d) F-actin staining of PC3-shKPNA4 or scramble control, lamellipodia positive cells were counted as the invasive cells. *p<0.05.

Figure 3. miR-708 targets KPNA4 and inhibits PCa migration

(a) Sequence alignment of miR-708 and KPNA4 3′UTR. KPNA4 expression in PC3 cells transfected with either (b) miR-708 mimic/negative control or (c) miR-708 inhibitor/negative control was determined by Western blotting, β-actin was used as loading control for total cell lysate. (d) HEK293T cells were co-transfected with miR-708 mimic or negative control mimic and dual-luciferase reporter plasmid inserted with KPNA4 3′UTR/control 3′UTR. Firefly luciferase was detected to determine the binding between miR-708 and KPNA4 3′UTR, renilla luciferase was used as internal control. (e) Transwell assay was performed to determine the cell invasion of PC3 cells that were transfected with either miR-708 mimic or control mimic for 48 hours.

Figure 4. KPNA4 knockdown attenuates prostate tumor invasion

(a) Luciferase-labeled PC3 sh-scramble or shKPNA4 cells (2×10⁵) were orthotopically injected into prostate of nude mice. Bioluminescence was detected biweekly for 6 weeks to determine the primary tumor growth. (b) Tumor growth curve was generated according the luciferase activity. *p<0.05. (c) H&E staining of the invasive tumor tissue in the proximal muscle. (d) Size of invasive tumor tissue is calculated by image J. **p<0.01. (e) Samples of primary tumor tissues that were harvested in 6 weeks post the intra-prostatic injection. (f) Tumor volumes are measured immediately after harvest.

Figure 5. KPNA4 knockdown blocks bone metastasis of prostate cancer

(a–b) Luciferase-labeled PC3 sh-scramble or shKPNA4 cells (1×10⁶) were intracardiacly injected into the left ventrical of nude mice, bioluminescence imaging of the bone metastatic lesion was taken after 4 weeks post injection, *p<0.05. (c) H&E staining showing the metastatic lesions in the hind limbs. CD206 positive M2 TAMs infiltration in (d) primary tumor tissue or (e) bone marrow was determined by immunofluorenscence staining. DAPI was used as an indicator of nucleus.

Figure 6. TNF-α and β mediates the KPNA4 induced prostate cancer migration

(a–b) TNF-α and TNF-β expression was evaluated in the primary tumor tissue by immunofluorenscence staining. (c) M2-phenotype associated cytokines of primary murine macrophages that were subjected to TNF-α or TNF-β stimulation (5ng/mL) was determined by real-time PCR. (d) Transwell assay of PC3-shKPNA4 or scramble control cells lines in the absence or presence of U937 cells to determine the cell invasion, PC3-shKPNA4 cells were stimulated with or without recombinant TNF-α or β cytokines (5ng/mL). (e–f) Invasive cells were quantitated by crystal violet staining assay. *p<0.05, **p<0.01.

Figure 7. Schematic model of miR-708/KPNA4/TNF-α and β signaling pathway that regulates PCa microenvironment and bone metastasis

In primary prostate tumor, miR-708 decrease in the cancer cells causes abnormal high expression of KPNA4, which will subsequently increase the TNF-α and β expression. Enrichment of TNF-α and β in tumor microenvironment can promote both of cancer cell mobility and M2 polarization of TAMs. On the other hand, increased TNF-α and β can enhance the osteoclastogenesis in bone environment, and eventually accelerate prostate cancer bone metastasis.