Manipulation of prostate cancer metastasis by locus-specific modification of the CRMP4 promoter region using chimeric TALE DNA methyltransferase and demethylase

Abstract

Prostate cancer is the most commonly diagnosed non-cutaneous cancer and one of the leading causes of cancer death for North American men. Whereas localized prostate cancer can be cured, there is currently no cure for metastatic prostate cancer. Here we report a novel approach that utilizes designed chimeric transcription activator-like effectors (dTALEs) to control prostate cancer metastasis. Transfection of dTALEs of DNA methyltransferase or demethylase induced artificial, yet active locus-specific CpG and subsequent histone modifications. These manipulations markedly altered expression of endogenous CRMP4, a metastasis suppressor gene. Remarkably, locus-specific CpG demethylation of the CRMP4 promoter in metastatic PC3 cells abolished metastasis, whereas locus-specific CpG methylation of the promoter in non-metastatic 22Rv1 cells induced metastasis. CRMP4-mediated metastasis suppression was found to require activation of Akt/Rac1 signaling and down-regulation of MMP-9 expression. This proof-of-concept study with dTALEs for locus-specific epigenomic manipulation validates the selected CpG methylation of CRMP4 gene as an independent biomarker for diagnosis and prognosis of prostate cancer metastasis and opens up a novel avenue for mechanistic research on cancer biology.

Keywords: CRMP4; epigenetic manipulation; metastasis; prostate cancer; transcription activator-like effectors (TALEs).

Figure 1. Regulation of CRMP4 promoter activity by CpG modification

(a) Illustration of the four CRMP4 promoter-driven luciferase reporters designated as A+ (−867/+114), A− (−839/+114), B+ (−717/+114), and B− (−656/+114). (b) Luciferase activities of the four CRMP4 promoter reporters that were pre-treated with or without M.SssI. One-way ANOVA was used to analyze the difference among the four groups luciferase reporters designated, and the differences between groups determined by the Student's t-test were considered to be significant at a P value less than 0.05/3 after correction. The error bars in b are s.e.m.

Figure 2. Locus-specific modulation of CRMP4 expression by dTALEs

(a) Illustration of the dTALEs. The synthetic TALE DNA-binding domain, the 23 bp targeting sequence from CRMP4 promoter region, nuclear localization signal (NLS), the truncated N-terminal domain (N95), the catalytic domain of Tet1 (Tet1c), the catalytic domain of DNMT3A (3Ac), and the other functional domain such as GFP are shown. The CRMP4 promoter structure (middle panel of Figure 2a) is drawn on a non-proportional scale. TSS: translation start site. (b) Luciferase activities altered by dTALEs through locus-specific CpG modification. Co-transfections with dTALEs and the CpG-free CRMP4-pCpGL reporter pre-treated with and without M.SssI were performed in HEK293 and COS-1 cells. (c) Alteration of endogenous CRMP4 mRNA expression in prostate cancer cells detected using qRT-PCR. The PC3 and 22Rv1 cells were transfected with CRMP4-TAL-3Ac, CRMP4-TAL-Tet1c, and empty phCMV1 vector to induce locus-specific CpG modifications. (d) Alteration of endogenous CRMP4 protein expression in prostate cancer cells detected using Western blotting. The prostate cancer cells were treated as described for Figure 3c. (e) CpG methylation frequencies of CRMP4 promoter Region A and Region B detected in the PC3 cells using pyrosequencing. The PC3 cells were transfected with CRMP4-TAL-Tet1c or empty phCMV1 vector as control. (f) CpG methylation frequencies of CRMP4 promoter Region A and Region B detected in the 22Rv1 cells using pyrosequencing. The 22Rv1 cells were transfected with CRMP4-TAL-3Ac or empty phCMV1 vector as control. The P values in b–f were determined with the Student's t-test. The error bars in b–f are s.e.m.

Figure 3. Histone modifications in the PCa cells expressing dTALEs

Ectopic expression of CRMP4-TAL-Tet1c in PC3 cells and CRMP4-TAL-3Ac in 22Rv1 cells induced the histone modifications at H3K9me3, H3K27me3 and H3K79me3 in (a) CRMP4 promoter Region A (−867/−839), (b) CRMP4 promoter Region B (−717/−656), (c) CRMP4 alternative promoter located 56 kb to the 5′ of the start codon, (d) CRMP4 terminal exon located 60 kb to the 3′ of the start codon. Empty phCMV1 vector was transfected in PC3 and 22Rv1 cells as controls. CRMP4 gene is located in chromosome 5. The P values in a–d were determined with the Student's t-test. The error bars in a–d are s.e.m.

Figure 4. In vitro and in vivo manipulation of prostate cancer cell metastasis by dTALEs

In vitro migration and tissue invasion of the PC3 cells (a) and the 22Rv1 cells (b) transfected with specified dTALEs or empty phCMV1 vector as control. Left panel: representative images; Right panel: results in means of three independent experiments. Significance was determined with the Student's t-test. (c) Xenogen images of mice with orthotopic implantation of the dTALE-expressing PC3 and 22Rv1 cells infected with luciferase-expression lentivirus LP-RLUC-LV. (d) Tumor volume of PC3 and 22Rv1 cells with and without expression of the specified dTALEs or empty phCMV1 vector as control. The volume was calculated from the Xenogen images. (e) Dot plot depicting number of metastases per mouse in animals injected with PC3 or 22Rv1 cells expressing specified dTALEs or empty phCMV1 vector as control. (f) Organ distribution frequency of tumor metastasis. All mice were autopsied and the organs were measured by detecting luciferase activity, respectively. The P values in a, b were determined with the Student's t-test. The error bars in a, b, d are s.e.m.

Figure 5. Akt-Rac1-MMP9 signaling pathway in CRMP4-mediated suppression of metastasis

(a) Western blot detection of phosphorylation and expression of Akt and Rac1 in the PC3 cells transfected with CRMP4-TAL-Tet1c or empty vector phCMV1 as control. CRMP4 siRNA was utilized to verify the loss-of-function. (b) Akt and CRMP4 interaction detected in the PC3 cells with differential Co-IP and Western blot. (c) Rac1 and CRMP4 interaction detected in the PC3 cells with differential Co-IP and Western blot. (d) Subcellular co-localization of CRMP4 with Akt and Rac1, respectively, in the CRMP4-TAL-Tet1c-expressing PC3 cells detected using confocal images. (e) Rac1 GTPase activity detected in the CRMP4-TAL-Tet1c-expressing PC3 cells using a Pull-down assay. (f) MMP-9 activity detected in the CRMP4-TAL-Tet1c-expressing PC3 cells using a Gelatin zymography assay. (g) IHC detection of opposite expression of CRMP4 and MMP-9 in the primary tumors from mice injected with PC3 or 22Rv1 cells expressing specified dTALEs. (h) Schematic of CRMP4-mediated signaling pathway involving phosphorylation of Akt and Rac1, activity of Rac1 GTPase and MMP-9, and repressed expression of MMP-9, VEGFB and VEGFC¹⁵, collectively leading to suppression of prostate cancer metastasis (Solid-lines: findings of this manuscript. Dashed-lines: previously published data and data not shown). The P values in e and f were determined with the Student's t-test. The error bars in e and f are s.e.m.

Figure 6. Differential survival of prostate cancer patients with positive and negative CRMP4 CpG methylation

(a–d) Kaplan–Meier graphs representing the probability of cumulative (a) biochemical recurrence-free survival, (b) clinical progression-free (free of a biopsy-proven local recurrence or imaging-identified systemic metastasis lesions) survival, (c) overall survival and (d) prostate cancer-specific survival in prostate cancer patients stratified according to positive and negative CRMP4 CpG methylation status in their primary tumors. The log-rank test P value reflects the significance of the correlation between CRMP4 CpG methylation and survival outcome. MSP: methylation-specific PCR.