Coordinate loss of MAP3K7 and CHD1 promotes aggressive prostate cancer - PubMed (original) (raw)

. 2015 Mar 15;75(6):1021-34.

doi: 10.1158/0008-5472.CAN-14-1596.

Leah Rider 2, Cera Nieto 2, Lina Romero 2, Anis Karimpour-Fard 2, Massimo Loda 3, M Scott Lucia 4, Min Wu 2, Lihong Shi 5, Adela Cimic 6, S Joseph Sirintrapun 6, Rosalie Nolley 7, Colton Pac 2, Haitao Chen 8, Donna M Peehl 7, Jianfeng Xu 9, Wennuan Liu 9, James C Costello 2, Scott D Cramer 10

Affiliations

Coordinate loss of MAP3K7 and CHD1 promotes aggressive prostate cancer

Lindsey Ulkus Rodrigues et al. Cancer Res. 2015.

Abstract

Prostate cancer subtypes are poorly defined and functional validation of drivers of ETS rearrangement-negative prostate cancer has not been conducted. Here, we identified an ETS(-) subtype of aggressive prostate cancer (ERG(-)MAP3K7(del)CHD1(del)) and used a novel developmental model and a cell line xenograft model to show that cosuppression of MAP3K7 and CHD1 expression promotes aggressive disease. Analyses of publicly available prostate cancer datasets revealed that MAP3K7 and CHD1 were significantly codeleted in 10% to 20% of localized tumors and combined loss correlated with poor disease-free survival. To evaluate the functional impact of dual MAP3K7-CHD1 loss, we suppressed Map3k7 and/or Chd1 expression in mouse prostate epithelial progenitor/stem cells (PrP/SC) and performed tissue recombination experiments in vivo. Dual shMap3k7-shChd1 PrP/SC recombinants displayed massive glandular atypia with regions of prostatic intraepithelial neoplasia and carcinoma apparent. Combined Map3k7-Chd1 suppression greatly disrupted normal prostatic lineage differentiation; dual recombinants displayed significant androgen receptor loss, increased neuroendocrine differentiation, and increased neural differentiation. Clinical samples with dual MAP3K7-CHD1 loss also displayed neuroendocrine and neural characteristics. In addition, dual Map3k7-Chd1 suppression promoted E-cadherin loss and mucin production in recombinants. MAP3K7 and CHD1 protein loss also correlated with Gleason grade and E-cadherin loss in clinical samples. To further validate the phenotype observed in the PrP/SC model, we suppressed MAP3K7 and/or CHD1 expression in LNCaP prostate cancer cells. Dual shMAP3K7-shCHD1 LNCaP xenografts displayed increased tumor growth and decreased survival compared with shControl, shMAP3K7, and shCHD1 xenografts. Collectively, these data identify coordinate loss of MAP3K7 and CHD1 as a unique driver of aggressive prostate cancer development.

©2015 American Association for Cancer Research.

PubMed Disclaimer

Conflict of interest statement

The authors disclose no potential conflicts of interest.

Figures

Figure 1

Figure 1. MAP3K7 and CHD1 deletion and mRNA downregulation correlate with increased biochemical recurrence

A) Frequency of single and co-occurring MAP3K7 and CHD1 deletions in 4 CaP datasets. B) Oncoprint of MAP3K7 and CHD1 CNAs in ERG fusion-positive and negative tumors (MSKCC dataset, n=157 primary tumors). C) Kaplan-Meier analysis of recurrence-free survival in patients with MAP3K7 and/or CHD1 deletions (MSKCC, n=157) (log-rank test). D) (Left) Correlation of MAP3K7 and CHD1 mRNA expression (MSKCC, n=109)(P<0.0001). Patients were divided into upper (purple), middle (blue), and lower (black) quartiles for MAP3K7 and CHD1 mRNA expression. Samples in which the gene expression of CHD1 and MAP3K7 did not fall into the same quartiles are shown as open circles. (Middle) and (Right) Kaplan-Meier analysis of recurrence-free survival in patients with lower, middle or upper quartile expression of MAP3K7 and CHD1. E) Oncoprint of MAP3K7 and CHD1 CNAs and mRNA expression (MSKCC, n=109). Samples are ranked from lowest (blue) to highest (red) mRNA z-score. CHD1 and MAP3K7 samples were ranked independently.

Figure 2

Figure 2. Chd1 suppression decreases growth of shControl but not shMap3k7 PrP/SCs

A) Strategy for dual suppression of Map3k7 and Chd1 in PrP/SCs. B) Protein lysates from knockdown cell lines immunoblotted with indicated antibodies. C) and D) 5-day monolayer growth assay for shControl (C) or shMap3k7 (D) PrP/SCs +/− Chd1 knockdown. Data represent means ± SEM (n=3; RM-ANOVA; *=P<0.0001 for shControl-puro vs. shChd1-1/shChd1-2 in (C); **=P<0.001 for shControl-puro vs. shChd1-1/shChd1-2 in (D)). E) 10-day clonogenic growth assay stained with crystal violet. F) and G) Quantification of average well confluence (F) and average colony size (G). Data represent means ± SEM (n=3; One-way ANOVA; (F): P<0.01; (G): P<0.05; bars with different letters above are statistically significantly different).

Figure 3

Figure 3. Dual shMap3k7-shChd1 recombinants display high-grade PIN and intraductal carcinoma in vivo

A) (Left) H&E staining of representative shControl, shMap3k7, and shChd1 tissue recombinants after 12 weeks in vivo (100x, 400x, scale=50μm). (Right) IHC with AR and p63 antibodies. Arrowheads indicate p63+ cells. B) Representative images of shDouble recombinants stained with H&E, AR, and p63. Recombinants from 3 different shDouble cell lines are shown. Arrowheads indicate p63+ cells. Arrows indicate goblet or signet ring-like cells. C) AR quantification. Data represent means ± SEM (One-way ANOVA; P<0.01; bars with different letters above are statistically significantly different). D) Representative high magnification images from shControl, shMap3k7, shChd1, and shDouble recombinants (1000x). Arrowheads indicate multinucleated cells. Arrows indicate cells with prominent nucleoli. Arrows with tails indicate large cells.

Figure 4

Figure 4. Co-suppression of Map3k7 and Chd1 alters differentiation of PrP/SCs

A) Representative images of recombinants stained with CK18, CK14, and CK5 antibodies. Arrowheads indicate CK14+ or CK5+ cells. B) Images from shDouble recombinants stained sequentially for AR, p63, CK18, CK14, and CK5. Circles indicate the same region in each image. C) Quantification of different graft phenotypes (Fisher’s exact test; *=P<0.05). D) SYP staining. Arrowhead indicates SYP+ cluster. E) SYP quantification. F) Nestin staining. Arrowhead indicates non-specific staining inside lumen. G) Nestin quantification. Data represent means ± SEM (One-way ANOVA; P<0.05; NS=not significant). H), I), and J) Correlation of SYP, CHGA and NES mRNA expression with MAP3K7 and CHD1 expression (MSKCC dataset, MAP3K7-CHD1 quartiles). Data represent means ± SEM (One-way ANOVA; *=P<0.0001; ***=P<0.01; ****=P<0.05; NS=not significant).

Figure 5

Figure 5. Co-suppression of Map3k7 and Chd1 promotes E-cadherin loss

A) Representative images of recombinants co-stained with Alcian Blue and Nuclear Fast Red. B) Alcian Blue quantification (Fisher’s exact test; *=P<0.05). C) E-cadherin staining. D) E-cadherin quantification. E) Sequential staining of shDouble grafts for Alcian Blue and E-cadherin. F) Ki67 staining. G) Ki67 quantification. Data represent means ± SEM (One-way ANOVA; P<0.01; bars with different letters above are statistically significantly different).

Figure 6

Figure 6. Co-suppression of MAP3K7 and CHD1 promotes robust growth of LNCaP xenografts

A) Strategy for dual suppression of MAP3K7 and CHD1 in LNCaP cells. B) Protein lysates from knockdown cell lines immunoblotted with indicated antibodies. C) 5-day monolayer growth assay. Data represent means ± SEM (n=3; RM-ANOVA; *=P<0.05 for shCHD1 vs. shControl, shMAP3K7, and shDouble). D) Bioluminescence images of mice from each experimental group 4 weeks after xenograft injection (exposure time=1s). E) Average flux measured by bioluminescence imaging weekly for 56 days (exposure time=1s). Twenty-eight-day time point from (D) indicated by boxed area. Data represent means ± SEM for each time point (RM-ANOVA; *=P<0.01 for shDouble vs. shControl, shMAP3K7, and shCHD1 for 42, 49, and 56 days). F) Average tumor volume over 56 days. Data represent means ± SEM for each time point (RM-ANOVA; *=P<0.05 for shDouble vs. shControl, shMAP3K7, and shCHD1). G) Kaplan-Meier analysis of mouse survival over 112 days (log-rank test). P-values for shControl vs. shMAP3K7, shCHD1, or shDouble are indicated. NS=not significant.

Figure 7

Figure 7. MAP3K7, CHD1, and E-cadherin are coordinately lost in high-grade tumors

A) Fifty whole-mount prostatectomy samples were stained with MAP3K7, CHD1, and E-cadherin antibodies. Representative images from four samples are shown. For each sample, the same region was imaged for all three stained sections. Arrowheads indicate benign glands. B), C), D) Quantification of MAP3K7, CHD1, and E-cadherin protein expression. MAP3K7 and E-cadherin expression were scored on a scale of 0 (no expression) to 3 (strong expression) (One-way ANOVA; *=P<0.0005); CHD1 expression was scored on a scale of 0 to 300 (Welch’s t-test; *=P<0.0005). E) and F) Correlation of _CDH1 (_E-cadherin) mRNA expression with MAP3K7 and CHD1 expression (MSKCC, TCGA). Data represent means ± SEM (One-way ANOVA; *=P<0.0001; **=P<0.001; ***=P<0.01; ****=P<0.05; NS=not significant). G) Model of single or dual Map3k7 and Chd1 loss during PrP/SC differentiation.

References

    1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA: a cancer journal for clinicians. 2014;64:9–29. - PubMed
    1. Sun J, Liu W, Adams TS, Sun J, Li X, Turner AR, et al. DNA copy number alterations in prostate cancers: a combined analysis of published CGH studies. The Prostate. 2007;67:692–700. - PubMed
    1. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, et al. Integrative genomic profiling of human prostate cancer. Cancer cell. 2010;18:11–22. - PMC - PubMed
    1. Berger MF, Lawrence MS, Demichelis F, Drier Y, Cibulskis K, Sivachenko AY, et al. The genomic complexity of primary human prostate cancer. Nature. 2011;470:214–20. - PMC - PubMed
    1. Barbieri CE, Baca SC, Lawrence MS, Demichelis F, Blattner M, Theurillat JP, et al. Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer. Nature genetics. 2012;44:685–9. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources