Metadherin (MTDH) overexpression in diverse cancer types has been linked to poor clinical outcomes, but definitive genetic proof of its contributions to cancer remains incomplete. In particular, the degree to which MTDH may contribute to malignant progression in vivo is lacking. Here, we report that MTDH is amplified frequently in human prostate cancers where its expression levels are tightly correlated with prostate cancer progression and poor disease-free survival. Furthermore, we show that genetic ablation of MTDH in the transgenic adenomcarcinoma of mouse prostate (TRAMP) transgenic mouse model of prostate cancer blocks malignant progression without causing defects in the normal development of the prostate. Germline deletion of Mtdh in TRAMP mice prolonged tumor latency, reduced tumor burden, arrested progression of prostate cancer at well-differentiated stages, and inhibited systemic metastasis to distant organs, thereby decreasing cancer-related mortality ∼10-fold. Consistent with these findings, direct silencing of Mtdh in prostate cancer cells decreased proliferation in vitro and tumor growth in vivo, supporting an epithelial cell–intrinsic role of MTDH in prostate cancer. Together, our findings establish a pivotal role for MTDH in prostate cancer progression and metastasis and define MTDH as a therapeutic target in this setting. Cancer Res; 74(18); 5336–47. ©2014 AACR.

Prostate cancer is the most common type of cancer among men in the United States. It accounts for 28% of the total cancer incidences and 10% of the total cancer deaths of American men (1). Prostate cancer–related mortality is mainly caused by advanced cancers that have disseminated and metastasized to distant organs such as the bone, lung, and liver (2). Once cancer cells establish secondary tumors in these vital organs, treatment usually becomes more complicated with worse outcome. Therefore, defining and understanding novel molecular targets that drive prostate cancer progression and metastasis is crucial for developing effective therapeutic approaches for prostate cancer.

We previously identified metadherin (MTDH; also known as AEG-1, 3D3/LYRIC) as a prometastasis gene that resides in 8q22, a frequently amplified genomic locus linked to poor relapse-free survival of breast cancer (3). In recent years, elevated levels of MTDH have been reported in more than 20 cancer types (4), suggesting a potentially crucial and broad functionality of this gene in human cancer. Depending on the cancer type tested, recent studies using mainly cell culture systems have implicated MTDH in many cancer-related processes, including cellular proliferation, stress-induced cell death, invasion, chemoresistance, and metastasis (5, 6).

These pleiotropic tumor-promoting roles of MTDH may stem from the complex nature of this protein, as revealed by its initial identification. MTDH was originally reported as an HIV-induced gene in astrocytes (7), a cell-surface molecule mediating the homing of mammary tumor cells to the lung endothelium (8), a lysine-rich CEACAM1 co-isolated (LYRIC) protein associated with tight junctions in prostate epithelial cells (9), and as a novel transmembrane protein present in the different subcellular compartments (10). At the molecular level, the human MTDH encodes a 582-amino acid protein with no recognizable domains that could indicate its biologic function, except for a putative transmembrane domain and 3 lysine-rich nuclear localization signals (11). MTDH has recently been reported to interact with multiple proteins. In the nucleus, MTDH was shown to interact with PLZF (11), BCCIPα (12), and NF-κB subunit p65 (13, 14). In cytoplasm, MTDH was reported to interact with staphylococcal nuclease domain-containing protein 1 (SND1; refs. 15–17). MTDH has also been linked to multiple classical oncogenic signaling pathways such as PI3K/AKT and Wnt signaling (5) in a cancer cell-type–dependent manner. However, whether MTDH exerts its function through interacting with its binding partners, and how MTDH modulates abovementioned oncogenic pathways remain largely unknown.

A few studies so far suggest that MTDH may play a role in prostate cancer. MTDH was shown to be overexpressed in prostate tumor tissues and tumorigenic cell lines compared with benign hyperplastic tissues and normal epithelial cells (18, 19). Knockdown of MTDH in cultured prostate cancer cells enhanced apoptosis and inhibited Matrigel invasion (18, 19). However, the expression profiles of MTDH at different stages of prostate cancer progression, the physiologic roles of MTDH in prostate tissue development, and the in vivo functional involvement of MTDH in prostate cancer development remain unaddressed. In this study, we investigate the clinical relevance and functional importance of MTDH in prostate cancer.

Detailed descriptions of the materials and the methods are presented in the Supplementary Materials.

Mice

All experimental protocols involving mice were approved by the Institutional Animal Care and Use Committee of Princeton University. Mtdh−/− mice were generated by injecting ES cell line XB780 (Bay Genomics) containing a gene-trapped allele of Mtdh into C57BL/6 blastocysts followed by confirmation of germline transmission by PCR. All Mtdh−/− mice were backcrossed to C57BL/6 background for >6 generations before breeding with C57BL/6 transgenic adenomcarcnoma of mouse prostate (TRAMP) transgenic mice (Jackson Laboratory; ref. 20). Female TRAMP mice were bred to Mtdh−/− male mice to obtain female TRAMP/Mtdh+/− and male Mtdh+/− founder mice. Subsequently, these founder mice were bred to generate TRAMP/Mtdh+/+, TRAMP/Mtdh+/−, and TRAMP/Mtdh−/− male mice. This breeding strategy ensured that all experimental mice were heterozygous to the SV40 transgene. DNA was extracted from the tail biopsy and PCR was performed for genotypic analysis. Primer sequences are listed in Supplementary Table S1.

Transplantation

Athymic nude male mice were injected subcutaneously with 5 × 105 TRAMP-C1 cells, and tumors were measured by calipers twice a week for calculation of tumor volumes (π × length × width2/6).

Cell culture

TRAMP-C1 cell line was obtained from ATCC and authenticated using short tandem repeat (STR) profiling method. The cells were maintained in Dulbecco's modified Eagle's medium supplemented with 0.005 mg/mL bovine insulin, 10 nmol/L dehydroisoandrosterone, 5% fetal bovine serum, and 5% Nu-Serum IV.

Tissue examination and tumor grading

Tissues were fixed, embedded, and sectioned. The hematoxylin and eosin (H&E)–stained prostate sections were each blindly scored on a scale of 1 to 6 (21–23) and each section was designated with a highest grade for the area of most severe pathology.

Human samples

Tumor specimens were obtained from the Comprehensive Cancer Center at the University of Michigan with informed consent from all subjects in accordance with the Institutional Review Board of the University of Michigan. Two prostate tissue microarrays composed of 62 normal prostate tissues, 10 benign prostatic hyperplasia (BPH), 10 prostate atrophy or prostate inflammatory atrophy (PIA), 10 prostatic intraepithelial neoplasia (PIN), 72 prostate tumors, and 10 distant metastases were used in our clinical study. At the time of surgery, the ages of the patients with available information were 56 to 90 years (median = 76 years, SD = 7.6 years). All the patients were not treatment with hormone or radiation therapy.

Western blot analysis

Total cell or tissue lysates were generated using RIPA buffer. Western blots of lysates resolved by SDS-PAGE were probed with the antibodies listed in the Supplementary Table S2.

qRT-PCR analyses

RNA isolation and real-time RT-PCR were performed following standard procedures. The mRNA levels were normalized based on GAPDH. Primer sequences are listed in Supplementary Table S1.

Fluorescence in situ hybridization

Tissue FISH was performed in the Dana-Farber Cancer Institute Cytogenetic Core Facility. Probes were labeled using the Nick Translation Kit, SpectrumOrange dUTP (Vysis), and SpectrumGreen (Vysis). Hybridization signals were viewed on fluorescence microscope and analyzed.

Immunohistochemistry

Immunohistochemistry staining was performed following standard protocols. Briefly, antigen heat retrieval with citric buffer was used on paraffin-embedded sections. Sections were then incubated with primary and secondary antibodies listed in the Supplementary Table S2., followed by incubation with ABC reagents and 3,3-diaminobenzidein substrate.

Statistical analysis

All results wherever necessary were subjected to statistical analysis. A log-rank test, a nonparametric Mann–Whitney test, χ2 test, and unpaired, 2-sided, independent Student t test with equal variance assumption were used for most studies as indicated in figure legends. For all statistics test, *, P < 0.05; **, P < 0.01; and ***, P < 0.001.

MTDH is associated with tumor progression and metastasis in human prostate cancer

First, we investigated the clinical relevance of MTDH in human prostate cancer samples. Two prostate tissue microarrays, consisting samples of normal prostates BPH, PIN, prostate primary tumors, and distant metastasis were analyzed by immunohistochemistry, and the MTDH protein levels were compared between samples of different clinical stages (Fig. 1A and Supplementary Table S3). MTDH protein was undetectable or expressed at very low levels in all normal or BPH prostate tissues. The protein levels of MTDH gradually increased in premalignant tissues (PIN), primary tumors, and metastasis, with the percentages of samples expressing medium or high amount of MTDH as 10%, 47.2%, and 80.0%, respectively, revealing a tight correlation of MTDH levels with clinical progression of prostate disease (Fig. 1B, left; χ2 test; P < 0.001). A difference in MTDH protein levels was already apparent when we compared BPH and PIN, two types of noncancerous tissues (Fig. 1B, right; χ2 test; P = 0.023), indicating an increase of MTDH expression during the transition from benign to premalignant stage. MTDH levels were also correlated with Gleason scores of primary tumors, as evidenced by higher average staining intensity of MTDH (Fig. 1C, blue curve) or a greater percentage of samples with at least medium levels of MTDH (Fig. 1C, red curve) in late-stage tumors. Importantly, the prostate-specific antigen (PSA)–based recurrence rate in patients with medium or high levels of MTDH was significantly higher than that in patients with low levels of MTDH in their prostate tumors (Fig. 1D). Together with the fact that MTDH protein levels in prostate cancer distant metastasis are markedly higher than those in primary tumors (Fig. 1A and 1B), these data document a strong positive correlation of MTDH levels with prostate cancer recurrence and metastasis.

Figure 1.

MTDH levels are associated with tumor progression and metastasis in human prostate cancer. Two prostate tumor tissue microarrays were stained with an anti-MTDH antibody. MTDH levels were scored as negative (0), low (1), medium (2), or high (3). A, representative images of MTDH immunostaining of prostate tissue of different stages and distant metastasis. Scale bar, 40 μm. B, left, MTDH expression levels in normal tissues (n = 62) and BPN (n = 10), PIN (n = 10), primary tumors (n = 72), and distant metastasis (n = 10). P < 0.001 by the χ2 test; right, MTDH levels in PIN (n = 10) and BPH (n = 10). P = 0.023 by the χ2 test. C, correlation of MTDH levels with prostate tumor Gleason scores. Blue curve represents average MTDH score (mean ± SEM) and red curve represents the percentage of samples with medium/high levels of MTDH within indicated groups. D, the Kaplan–Meier analysis of recurrence-free survival of patients with prostate cancer based on MTDH expression in their tumors.

Figure 1.

MTDH levels are associated with tumor progression and metastasis in human prostate cancer. Two prostate tumor tissue microarrays were stained with an anti-MTDH antibody. MTDH levels were scored as negative (0), low (1), medium (2), or high (3). A, representative images of MTDH immunostaining of prostate tissue of different stages and distant metastasis. Scale bar, 40 μm. B, left, MTDH expression levels in normal tissues (n = 62) and BPN (n = 10), PIN (n = 10), primary tumors (n = 72), and distant metastasis (n = 10). P < 0.001 by the χ2 test; right, MTDH levels in PIN (n = 10) and BPH (n = 10). P = 0.023 by the χ2 test. C, correlation of MTDH levels with prostate tumor Gleason scores. Blue curve represents average MTDH score (mean ± SEM) and red curve represents the percentage of samples with medium/high levels of MTDH within indicated groups. D, the Kaplan–Meier analysis of recurrence-free survival of patients with prostate cancer based on MTDH expression in their tumors.

Close modal

MTDH genomic gain is associated with MTDH overexpression and clinical progression in prostate cancer

We previously mapped MTDH at the center of a common 8q22 gain that is associated with poor prognosis of breast cancer (3). To test whether there is also increased genomic copy number of MTDH in prostate tumors, we analyzed a prostate tissue microarray with interphase FISH using probes against either the MTDH genomic locus or the chromosome 8 centromere region (as control). FISH analysis revealed a significant portion of prostate cancer samples harbored extra genomic copies of the MTDH gene (Fig. 2A). Importantly, MTDH DNA gain was significantly associated with higher levels of protein in the cells (Fig. 2B), indicating that DNA gain is a mechanism for MTDH overexpression in prostate cancer. Furthermore, MTDH DNA gain was tightly correlated with clinical stages and prognosis, such that of samples before or at the premalignant stages (normal, BPH, or PIN), local tumors, and distant metastasis, the portions with DNA gain were 0%, 24%, and 50%, respectively (Fig. 2C). In addition, patients with increased MTDH copies suffered from earlier recurrence than those without MTDH genomic gain (Fig. 2D). These clinical data demonstrate the prognosis values of DNA and protein status of MTDH, and suggest a potentially crucial role of MTDH in prostate cancer progression and metastasis.

Figure 2.

MTDH genomic gain is associated with MTDH protein levels and clinical progression in prostate cancer. A, a prostate tumor tissue microarray was analyzed for MTDH genomic copy number by FISH. Shown are examples of tumors without (left) or with (right) MTDH genomic gains. SpectrumGreen (green) and SpectrumOrange (pale orange) probes detect chromosome 8 centromere and the 8q22 region, respectively. Scale bar, 1 μm. B, MTDH genomic gain is correlated with MTDH protein levels. Samples with MTDH gain, n = 11; samples without MTDH gain, n = 64. P = 0.0018 by the χ2 test. C, frequency of MTDH genomic gain in prostate noncancerous tissues or tumors. Normal/benign/premalignant, n = 38; primary tumors, n = 29; distant metastasis, n = 8. P values by the χ2 test are shown. D, the Kaplan–Meier analysis of recurrence-free survival of patients with prostate cancer with and without MTDH genomic gain in their tumors. Cox proportional hazard ratio (HR) is shown.

Figure 2.

MTDH genomic gain is associated with MTDH protein levels and clinical progression in prostate cancer. A, a prostate tumor tissue microarray was analyzed for MTDH genomic copy number by FISH. Shown are examples of tumors without (left) or with (right) MTDH genomic gains. SpectrumGreen (green) and SpectrumOrange (pale orange) probes detect chromosome 8 centromere and the 8q22 region, respectively. Scale bar, 1 μm. B, MTDH genomic gain is correlated with MTDH protein levels. Samples with MTDH gain, n = 11; samples without MTDH gain, n = 64. P = 0.0018 by the χ2 test. C, frequency of MTDH genomic gain in prostate noncancerous tissues or tumors. Normal/benign/premalignant, n = 38; primary tumors, n = 29; distant metastasis, n = 8. P values by the χ2 test are shown. D, the Kaplan–Meier analysis of recurrence-free survival of patients with prostate cancer with and without MTDH genomic gain in their tumors. Cox proportional hazard ratio (HR) is shown.

Close modal

Generation and characterization of Mtdh-deleted TRAMP mice

To investigate the roles of Mtdh in normal development and cancer, we first created whole-organism Mtdh-knockout (KO) mice using ESC line XB780 from Bay Genomics gene trap database (24). The mutant allele contains a LacZ transgene insertion into the second intron of Mtdh, which results in premature termination of transcription. Mtdh-KO (Mtdh−/−) mice were viable and fertile, and no overt abnormality in any organs were seen at autopsy. Specifically, the gross morphology (Supplementary Fig. S1A) and relative weights (Supplementary Fig. S1B) of the lower genitourinary tract that includes both the prostate and seminal vesicles were comparable between wild-type (WT, Mtdh+/+), heterozygous (Mtdh+/−), and Mtdh−/− male mice. When examined at microscopic levels, there was no morphological abnormality in the prostate epithelium of Mtdh−/− mice (Supplementary Fig. S1C).

To examine the consequence of Mtdh ablation in prostate tumor formation, we used the TRAMP model, in which the expression of the SV40 T antigen is controlled by the rat Probasin promoter and induced during sex maturity of the mice, leading to prostate cancer development that resembles the clinical progression of human prostate cancer (20, 21). TRAMP/Mtdh+/+, TRAMP/Mtdh+/−, and TRAMP/Mtdh−/− mice in C57BL/6 background were generated as described in Materials and Methods (Fig. 3A) and the genotypes were confirmed by PCR (Fig. 3B). As an initial step to assess the role of Mtdh in prostate tumorigenesis, we isolated prostate tissues from normal and TRAMP mice and examined the mRNA (Fig. 3C) and protein (Fig. 3D) levels of Mtdh. As expected, no Mtdh mRNA was detected in Mtdh−/− prostate tissues (Fig. 3C), confirming that the gene-trapped method completely abolished Mtdh expression. Furthermore, we observed more abundant Mtdh mRNA in prostate tissues from TRAMP/Mtdh+/+ and TRAMP/Mtdh+/− mice compared with matched control mice that do not express SV40 T antigen (Fig. 3C). Consistently, a significant increase in protein levels of Mtdh was detected in TRAMP/Mtdh+/+ prostates compared with normal controls (Fig. 3D). These data indicate that Mtdh is upregulated during prostate tumorigenesis in mice and suggest that higher levels of Mtdh may confer a competitive advantage for tumor cells.

Figure 3.

Generation and characterization of TRAMP mice with different Mtdh status. A, cross scheme for the generation of TRAMP mice in C57BL/6 background with different Mtdh status. B, genotyping of mice generated in A. Top, the detected WT (602 bp) and gene-trapped mutant alleles (472 bp) of Mtdh. Bottom, the detected TRAMP transgene (600 bp) and internal genomic control (324 bp). C, Mtdh mRNA in prostate tissues from mice with indicated genotypes analyzed by qPCR. Mtdh mRNA was undetectable in Mtdh−/− tissues and was elevated in TRAMP-positive prostate tissues as compared with normal glands. **, P < 0.01 and ***, P < 0.001 based on the Mann–Whitney test. D, Western blot analysis of Mtdh in prostates from WT nontransgenic mice and tumors from TRAMP/Mtdh+/+ mice. Arbitrary levels of Mtdh protein after normalized to β-actin are shown at the bottom. E and F, representative immunohistochemical staining of SV40 Tag oncoprotein in prostate glands from 8-week-old (E) and 28-week-old (F) TRAMP/Mtdh+/+ and TRAMP/Mtdh−/− mice. Scale bar, 50 μm.

Figure 3.

Generation and characterization of TRAMP mice with different Mtdh status. A, cross scheme for the generation of TRAMP mice in C57BL/6 background with different Mtdh status. B, genotyping of mice generated in A. Top, the detected WT (602 bp) and gene-trapped mutant alleles (472 bp) of Mtdh. Bottom, the detected TRAMP transgene (600 bp) and internal genomic control (324 bp). C, Mtdh mRNA in prostate tissues from mice with indicated genotypes analyzed by qPCR. Mtdh mRNA was undetectable in Mtdh−/− tissues and was elevated in TRAMP-positive prostate tissues as compared with normal glands. **, P < 0.01 and ***, P < 0.001 based on the Mann–Whitney test. D, Western blot analysis of Mtdh in prostates from WT nontransgenic mice and tumors from TRAMP/Mtdh+/+ mice. Arbitrary levels of Mtdh protein after normalized to β-actin are shown at the bottom. E and F, representative immunohistochemical staining of SV40 Tag oncoprotein in prostate glands from 8-week-old (E) and 28-week-old (F) TRAMP/Mtdh+/+ and TRAMP/Mtdh−/− mice. Scale bar, 50 μm.

Close modal

To test whether Mtdh loss affects the expression of the PB-Tag transgene, we immunostained SV40 T antigen in prostates from TRAMP mice. The T antigen was detected in prostatic epithelial cells as early as 8 weeks of age (Fig. 3E) and continuously expressed in the prostatic epithelium and tumor cells at later stages (Fig. 3F). There was no noticeable difference in the T antigen immunoreactivity in individual prostatic epithelial cells between TRAMP/Mtdh+/+ and TRAMP/Mtdh−/− mice at either early or late stages of cancer progression. Furthermore, SV40 t and T mRNA expression were comparable in TRAMP mice with different Mtdh status (Supplementary Fig. S2A and S2B). As expected for the negative control, SV40 antigens were undetectable in prostates from normal mice (Supplementary Fig. S2A and S2B). These results clearly demonstrate that Mtdh deletion does not affect the PB-Tag transgene expression in prostate epithelial cells.

Loss of Mtdh suppresses prostate tumor formation and increases survival rate

To investigate the overall effect of Mtdh on oncogene-induced prostate tumorigenesis, we studied cohorts of TRAMP mice (n > 100) with different Mtdh status for their prostate tumor development and cancer-related mortality. First, as genitourinary weight is a reliable indicator of tumor burden in TRAMP mice (22), we dissected and measured the urogenital apparatus from TRAMP mice at different ages to monitor tumor formation at a gross level. As Mtdh expression was comparable between Mtdh+/+ and Mtdh+/− prostate tissues (Fig. 3C), we grouped these mice as Mtdh-positive group (Mtdh+) as opposed to Mtdh-negative group (Mtdh). The SV40 T antigens induced a drastic expansion of the urogenital mass in TRAMP/Mtdh+ mice as a function of age (Fig. 4A, red), such that many of these mice displayed severely enlarged abdomen at 36 weeks of age. In contrast, the wet weights of genitourinary in TRAMP/Mtdh−/− mice remained relatively constant or increased only at a much slower rate (Fig. 4A, blue). Similar results were obtained when the relative genitourinary weight (as % of body weight) was compared (Fig. 4B). Consistently, almost all the urogenital complex dissected from 36-week-old TRAMP/Mtdh+ mice exhibited grossly visible signs of tumor formation (Fig. 4C, top), whereas those of TRAMP/Mtdh−/− mice remained similar to normal controls (Fig. 4C, bottom).

Figure 4.

Loss of Mtdh in mice inhibits tumor formation, reduces tumor burden, and increases survival rate. A and B, wet weight (A) or relative weight as % of body weight (B) of lower urogenital tracts excised from TRAMP/Mtdh+/+ and TRAMP/Mtdh+/− mice (denoted as Mtdh+) or TRAMP/Mtdh−/− (denoted as Mtdh). Data represent mean ± SEM. **, P <0.01 based on the Mann–Whitney test. C, representative images of lower urogenital tracts (B, bladder; P, prostate; SV, seminal vesicle) excised from 36-week-old male mice with indicated genotypes. Scale bar, 1 cm. D, tumor incidence scored by examining histologic sections of prostate glands from cohorts of TRAMP mice with indicated Mtdh genotype at different ages. The numbers on top of each bar graph indicates the number of mice with prostate cancer versus total number of mice examined in a given group. *, P < 0.05 and **, P < 0.01 based on the χ2 test. E, mortality rate of TRAMP mice with indicated Mtdh genotypes by 1 year of age. ***, P < 0.001 based on the χ2 test.

Figure 4.

Loss of Mtdh in mice inhibits tumor formation, reduces tumor burden, and increases survival rate. A and B, wet weight (A) or relative weight as % of body weight (B) of lower urogenital tracts excised from TRAMP/Mtdh+/+ and TRAMP/Mtdh+/− mice (denoted as Mtdh+) or TRAMP/Mtdh−/− (denoted as Mtdh). Data represent mean ± SEM. **, P <0.01 based on the Mann–Whitney test. C, representative images of lower urogenital tracts (B, bladder; P, prostate; SV, seminal vesicle) excised from 36-week-old male mice with indicated genotypes. Scale bar, 1 cm. D, tumor incidence scored by examining histologic sections of prostate glands from cohorts of TRAMP mice with indicated Mtdh genotype at different ages. The numbers on top of each bar graph indicates the number of mice with prostate cancer versus total number of mice examined in a given group. *, P < 0.05 and **, P < 0.01 based on the χ2 test. E, mortality rate of TRAMP mice with indicated Mtdh genotypes by 1 year of age. ***, P < 0.001 based on the χ2 test.

Close modal

To further assess the occurrence of prostate cancer at histologic level, we sectioned prostate tissues from TRAMP mice and performed H&E staining. About 50% of 23- or 28-week-old TRAMP/Mtdh+ mice developed prostate cancer lesions, and by 36 weeks of age, almost all the mice in this group developed multifocal prostate tumors (Fig. 4D, red). In contrast, no prostate cancer was detected in 28-week-old TRAMP/Mtdh−/− mice, and prostate cancer lesions were detected in only 50% of 36-week-old TRAMP/Mtdh−/− mice at very limited regions of the prostate epithelium. These histopathologic results are consistent with the morphologic observations showing that most TRAMP/Mtdh−/− mice did not display visible signs for prostate tumor formations by 36 weeks of age (Fig. 4C). Of note, tumors formed in TRAMP mice exhibit diverse histologic features that are characteristics of adenocarcinomas (Supplementary Fig. S3A, top), phyllode-like tumors (Supplementary Fig. S3B, top), and neuroendocrine tumors (Supplementary Fig. S3C, top), and it has been suggested that these morphologically distinct tumors may originate from different cell lineages within the prostate or develop independently from common early progenitors (25). Interestingly, TRAMP/Mtdh−/− mice showed a delay in the occurrence of all these different tumor subtypes (Supplementary Fig. S3), suggesting a possible lineage-independent role of Mtdh in prostate cancer.

As a consequence of malignant prostate cancer development, a significant percentage (∼50%) of TRAMP/Mtdh+ died before reaching 1 year of age. In contrast, only 1 of 14 TRAMP/Mtdh−/− examined died from cancer-related disease (Fig. 4E, P < 0.001). These data together demonstrate that Mtdh deletion in mice inhibits oncogene-driven formation of prostate cancer and extends the life span of TRAMP mice.

Inactivation of Mtdh impedes prostate cancer progression

Prostate cancer progression consists of multiple stages starting from premalignant lesions such as PIN to well, moderately, and poorly differentiated carcinoma. To investigate how Mtdh loss affects the initiation and progression of prostate cancer, we performed histologic examination on dorsal and lateral lobes of each prostate gland, the most frequent and severe sites subjected to tumorigenesis in TRAMP mice (Fig. 5A; ref. 21). On the basis of previously described grading systems for TRAMP tumors (21, 22), each prostate was assigned a highest score ranging from a score of 1 for normal prostate and 6 for poorly differentiated carcinoma including neuroendocrine tumors. PIN lesions were readily detected in both TRAMP/Mtdh+ and TRAMP/Mtdh prostates from 8 to 12 weeks of age (Fig. 5B, first panel), as evidenced by epithelial hyperplasia and the presence of an intact basal cell layer (Fig. 5A, T8 and T12). This was consistent with comparable proliferation (Supplementary Fig. S4A) and apoptosis (Supplementary Fig. S4B) indices in prostate epithelium from 10-week-old TRAMP/Mtdh+ and TRAMP/Mtdh mice. A substantial fraction of TRAMP/Mtdh+ mice developed well or moderately differentiated adenocarcinoma with invasive lesions at 23 and 28 weeks, respectively, and by 36 weeks, 84% of TRAMP/Mtdh+ mice displayed moderately to poorly differentiated adenocarcinoma characteristic of anaplastic sheets of tumor cells. In comparison, the majority of TRAMP/Mtdh mice only developed PIN or early-stage, well-differentiated adenocarcinomas or phyllode-like tumors by 36 weeks. Consistently, the expression of E-cadherin was gradually reduced or even became absent as prostate tumors progressed to less-differentiated stages in TRAMP/Mtdh+ mice (Fig. 5C, top), whereas abundant E-cadherin was detected in prostate tissues in TRAMP/Mtdh−/− mice across different ages examined (Fig. 5C, bottom). In addition, malignant progression of Mtdh+ tumors was accompanied by a significantly higher percentage of proliferating cells (Fig. 5D, P < 0.01), whereas the percentage of apoptotic cells were similar in Mtdh+ and Mtdh tumors at this stage (Supplementary Fig. S4C). Taken together, these findings suggest that inactivation of Mtdh arrests prostate cancer progression at early stages of tumorigenesis and prevents premalignant lesions to expand and progress into advanced stages.

Figure 5.

Loss of Mtdh inhibits malignant progression of prostate cancer. A, H&E-stained histologic sections of prostates dissected from TRAMP/Mtdh+ and TRAMP/Mtdh mice at indicated ages. Scale bar, 200 μm. B, each prostate from mice with indicated genotypes and ages was assigned a single highest grade. + and − indicate TRAMP/Mtdh+ and TRAMP/Mtdh mice, respectively. Grade scores: 1, normal; 2, low grade PIN; 3, high-grade PIN; 4, well-differentiated adenocarcinoma and phyllode tumor; 5, moderately differentiated adenocarcinoma; 6, poorly differentiated adenocarcinoma and neuroendocrine tumors. The grading scheme followed standard protocol as previously described (23). The numbers at the bottom of each column indicate total number of prostate glands evaluated in each group. P < 0.05 by the χ2 test in T36 group. C, analysis of E-cadherin during prostate tumor progression. Sections from dorsal–lateral lobes were immunostained with an antibody against E-cadherin and images shown are representative of n > 3 animals per group. Scale bar, 25 μm. D, left, representative images of Ki67-stained sections of prostate tumors from 36-week-old TRAMP/Mtdh+ (n = 6) and TRAMP/Mtdh (n = 3) mice. Scale bars, 25 μm. Right, quantification of Ki67-positive epithelial cells. Data represent mean ± SEM. **, P < 0.01 based on the Student t test.

Figure 5.

Loss of Mtdh inhibits malignant progression of prostate cancer. A, H&E-stained histologic sections of prostates dissected from TRAMP/Mtdh+ and TRAMP/Mtdh mice at indicated ages. Scale bar, 200 μm. B, each prostate from mice with indicated genotypes and ages was assigned a single highest grade. + and − indicate TRAMP/Mtdh+ and TRAMP/Mtdh mice, respectively. Grade scores: 1, normal; 2, low grade PIN; 3, high-grade PIN; 4, well-differentiated adenocarcinoma and phyllode tumor; 5, moderately differentiated adenocarcinoma; 6, poorly differentiated adenocarcinoma and neuroendocrine tumors. The grading scheme followed standard protocol as previously described (23). The numbers at the bottom of each column indicate total number of prostate glands evaluated in each group. P < 0.05 by the χ2 test in T36 group. C, analysis of E-cadherin during prostate tumor progression. Sections from dorsal–lateral lobes were immunostained with an antibody against E-cadherin and images shown are representative of n > 3 animals per group. Scale bar, 25 μm. D, left, representative images of Ki67-stained sections of prostate tumors from 36-week-old TRAMP/Mtdh+ (n = 6) and TRAMP/Mtdh (n = 3) mice. Scale bars, 25 μm. Right, quantification of Ki67-positive epithelial cells. Data represent mean ± SEM. **, P < 0.01 based on the Student t test.

Close modal

Ablation of Mtdh reduces systemic metastasis of prostate tumor cells

We examined metastatic disease in cohorts of TRAMP mice at around 50 weeks of age, a time point when most of TRAMP/Mtdh−/− mice have developed prostate cancer lesions. We dissected livers, lungs and periaortic lymph nodes, frequent sites of metastasis in the TRAMP model (26), and examined the appearance of metastatic lesions (Fig. 6A and B) at both macroscopic (Fig. 6C) and histology levels (Fig. 6D). Multiple large metastatic nodules were observed in livers from 34% of TRAMP/Mtdh+ mice, resulting in markedly increased liver size. Conversely, none of TRAMP/Mtdh−/− mice examined developed visible metastases in the liver. Similarly, multiple metastatic deposits in the lung were found in 39% of TRAMP/Mtdh+ mice, but only 2 of 17 TRAMP/Mtdh−/− mice had 1 or 2 metastatic lesions detected in their lungs. Enlargement of lymph node were frequently identified in TRAMP/Mtdh+ mice (9/20) and this incidence decreased to 30% in TRAMP/Mtdh−/− mice. To demonstrate that the metastatic cells were of prostatic origin, immunohistochemical analysis was performed on serial sections of the liver, lung, and lymph node to detect the T-antigen oncoprotein. We observed a uniform expression of the T-antigen oncoprotein confined to the metastatic deposits but not surrounding normal cells (Fig. 6E), in agreement with the tissue-specific pattern of PB-directed transgene expression (26). These data together clearly show that Mtdh deletion in mice significantly reduced systematic metastasis in the TRAMP model. Of note, because of the fact that close to 50% of TRAMP/Mtdh+ mice had died from large primary and likely metastatic tumors before reaching one year of age (Fig. 4E), the observed difference in metastasis between Mtdh+ and Mtdh groups was very likely to be underestimated.

Figure 6.

Ablation of Mtdh reduces systemic metastasis of prostate cancer. A, incidence of lung, liver, and lymph node metastasis in cohorts of 1-year-old TRAMP mice with indicated Mtdh genotypes. P values based on the χ2 test. B, bar graph presentation of A. C and D, images (C; scale bar, 1 cm) and H&E-stained sections (D; scale bar, 200 μm) of livers and lungs from mice in A. E, left, representative immunohistochemical staining of SV40 T-antigen oncoprotein in liver, lung, and lymph node metastases; right, IgG control on a section of lymph node metastases. Scale bar, 50 μm.

Figure 6.

Ablation of Mtdh reduces systemic metastasis of prostate cancer. A, incidence of lung, liver, and lymph node metastasis in cohorts of 1-year-old TRAMP mice with indicated Mtdh genotypes. P values based on the χ2 test. B, bar graph presentation of A. C and D, images (C; scale bar, 1 cm) and H&E-stained sections (D; scale bar, 200 μm) of livers and lungs from mice in A. E, left, representative immunohistochemical staining of SV40 T-antigen oncoprotein in liver, lung, and lymph node metastases; right, IgG control on a section of lymph node metastases. Scale bar, 50 μm.

Close modal

Knockdown of Mtdh in prostate cancer cells reduces proliferation in vitro and tumor formation in vivo

As Mtdh is widely expressed in mice (data not shown), the tumorigenesis defects in whole-organism Mtdh-KO mice could result from either loss of Mtdh in prostate epithelial cells or other cell/tissue types. To distinguish between these two possibilities, we examined the necessity of Mtdh in tumorigenic ability of TRAMP-C1 cell line, which is characteristic of an advanced prostate tumor in the TRAMP model (27). We knocked down (KD) Mtdh (Fig. 7A and 7B) and found that Mtdh-KD decreased proliferation of TRAMP-C1 cells in vitro (Fig. 7C). To examine the consequence of Mtdh-KD in tumor formation in vivo, we injected control and Mtdh-KD cells into the flanks of recipient mice and monitored tumor occurrence over time. The occurrence of palpable tumors was significantly delayed in the KD groups (Fig. 7D), and tumors dissected from the KD groups were much smaller than those from the control group (Fig. 7E and 7F). Similar results were obtained in another 2 independent of experiments (Supplementary Fig. S5). Interestingly, tumors that eventually formed in the Mtdh-KD groups expressed similar levels of Mtdh as tumors in the control group (Fig. 7G), which could result from either the clonal expansion of the cells without Mtdh inhibition or from re-expression of Mtdh in Mtdh-KD cells.

Figure 7.

Silencing of Mtdh in TRAMP-C1 prostate cancer cells decreases proliferation in vitro and tumor formation in vivo. A and B, Mtdh was KD by two independent shRNA, as quantified by qPCR (A) and Western blot analysis (B). C, proliferation rate of control and Mtdh-KD TRAMP-C1 cancer cells after 48 hours. D, kinetics of prostate tumor onset in male mice subcutaneously transplanted with control (n = 16), Mtdh-KD1 (n = 8), and Mtdh-KD2 (n = 12) TRAMP-C1 cells. P value based on the log-rank test. See also repeat experiments in Supplementary Fig. S5. E, tumor volumes 5 weeks postinjection in D. ***, P < 0.001 based on the Mann–Whitney test. F, images of tumors dissected 5 weeks posttransplantation in D. G, Mtdh mRNA levels in tumors formed in control and KD groups. Note the tumors that eventually grew in the KD groups expressed similar levels of Mtdh as the controls. A, C, and G, data represent mean ± SEM and P values based on the Student t test. **, P < 0.01; ***, P < 0.001; n.s., not significant.

Figure 7.

Silencing of Mtdh in TRAMP-C1 prostate cancer cells decreases proliferation in vitro and tumor formation in vivo. A and B, Mtdh was KD by two independent shRNA, as quantified by qPCR (A) and Western blot analysis (B). C, proliferation rate of control and Mtdh-KD TRAMP-C1 cancer cells after 48 hours. D, kinetics of prostate tumor onset in male mice subcutaneously transplanted with control (n = 16), Mtdh-KD1 (n = 8), and Mtdh-KD2 (n = 12) TRAMP-C1 cells. P value based on the log-rank test. See also repeat experiments in Supplementary Fig. S5. E, tumor volumes 5 weeks postinjection in D. ***, P < 0.001 based on the Mann–Whitney test. F, images of tumors dissected 5 weeks posttransplantation in D. G, Mtdh mRNA levels in tumors formed in control and KD groups. Note the tumors that eventually grew in the KD groups expressed similar levels of Mtdh as the controls. A, C, and G, data represent mean ± SEM and P values based on the Student t test. **, P < 0.01; ***, P < 0.001; n.s., not significant.

Close modal

Human prostate cancer is characterized by rampant recurrent amplifications and deletions, suggesting that along with a few well-known genetic lesions, such as loss of tumor suppressors PTEN and p53, there remain many uncharacterized genes governing the genesis and progression of this malignancy (28). In this study, we demonstrate that MTDH is frequently overexpressed and amplified in human prostate cancers and its expression levels strongly correlate with disease progression and poor survival outcome. In addition, our genetic studies utilizing Mtdh-KO mice provide the first in vivo evidence that Mtdh plays a critical role in spontaneous prostate cancer progression and metastasis without affecting normal development.

The TRAMP model closely resembles human prostate cancer as it spontaneously develops progressive prostate cancer that can metastasize to multiple different organs. Importantly, we show Mtdh levels are elevated in both human and SV40-driven murine prostate tumors compared with normal prostate tissues. Interestingly, Mtdh was recently shown to be overexpressed through genomic amplification in a p53/Pten-deficient metastatic prostate cancer mouse model, in which telomerase activity was reactivated following telomere dysfunction (29), suggesting that Mtdh can be activated in mice through genomic amplification mechanism similar to what we found in human prostate cancer. Future studies are needed to investigate the molecular basis of Mtdh overexpression in SV40-transformed tumors.

We show that Mtdh knockout in the TRAMP mice significantly impaired tumor formation and decreased cancer-related mortality. The findings that silencing of Mtdh in TRAMP-C1 cancer cells prolonged tumor-free survival and reduced tumor growth in vivo lend further support for a prostate tumor cell-intrinsic role of Mtdh in prostate cancer. Of note, tumors eventually formed by TRAMP-C1 cancer cells that presumably carried Mtdh-targeting shRNA regained the expression of Mtdh, reinforcing the notion that Mtdh is essential for prostate cancer formation in vivo. We observed a blockage of prostate cancer progression at benign or well-differentiated stages in TRAMP/Mtdh−/− mice, a notion that is further supported by constitutively high levels of epithelial marker E-cadherin in TRAMP/Mtdh−/− prostate tissues. Consistently, although TRAMP/Mtdh+ mice displayed enlarged prostate glands and obstructed seminal vesicles, a phenomenon often resulted from invasive and malignant growth of prostate tumor cells into the urethra and seminal vesicles, urogenital track from TRAMP/Mtdh−/− mice seemed largely normal. In line with our findings from mouse models, MTDH exhibits a progression-correlated expression pattern in human prostate cancer. These results may suggest the potential use of MTDH as a biomarker for prostate cancer progression and a possible therapeutic target to halt malignant progression of human prostate cancer.

The fact that a small percentage of TRAMP/Mtdh−/− mice still developed invasive prostate cancer after long latency and died from the disease (1 of 14) suggests that some TRAMP/Mtdh−/− tumors have overcome the deficiency of Mtdh and utilized Mtdh-independent pathways to support their malignant growth and progression. Indeed, the induction of SV40 oncogene occurs uniformly in prostate epithelial cells as early as puberty; however, the development of invasive prostate cancer and metastasis occurs much later and seems to be highly heterogeneous, suggesting that individual tumor may acquire different genetic and epigenetic alterations to fully transform SV40 antigen-positive cells. Although inactivation of Mtdh inhibits Mtdh-dependent signaling and prostate cancer progression as observed in the majority of TRAMP/Mtdh−/− mice, it would not block tumor growth that is supported by Mtdh-independent mechanisms. This is also in line with the observation that MTDH is highly expressed in a large percentage but not all human prostate cancer. Nevertheless, cancer cells that have depended on high levels of MTDH for their progression to a malignant stage would be sensitive to MTDH-targeting therapeutics, as supported by the finding that inhibition of Mtdh in Mtdh-positive TRAMP-C1 cells significantly impairs the tumorigenic potential of these cells.

Metastasis to distant organs is responsible for the majority of cancer-related death in patients with solid cancer (30). We previously demonstrated that MTDH is amplified and/or overexpressed in human breast cancer with a higher risk of metastasis (3). In this study, we found that a substantial fraction of distant metastases of human prostate cancer exhibited high levels of MTDH and harbored amplification of the 8q22 genomic loci. Functionally, Mtdh ablation significantly decreased the incidence and burden of metastases in distant organs in the TRAMP model. In concordance with our findings, a recent study shows that telomerase reactivation following telomere dysfunction in a prostate cancer–prone mouse model driven by Pten and p53 loss facilitates the selection of copy number alterations at cancer-relevant loci, and Mtdh was found to be at the top of a selective list of genes whose genetic amplification is strongly associated with cancer progression and metastasis in this mouse model (29). This suggests a possibly crucial role of Mtdh in prostate cancer driven by genetic events in addition to SV40 oncogenes. Finally, it should be noted that the metastasis-promoting function of MTDH is likely to exist in other cancer types, beyond breast and prostate cancers. Several recent studies indeed showed abnormal MTDH expression was associated with lymph node metastasis and a worse prognosis in additional cancer types including laryngeal squamous cell carcinoma (31), squamous cell carcinoma of the head and neck (32), and hepatocellular carcinoma (33), and overexpression of MTDH increases experimental metastasis of hepatocellular carcinoma cells (34).

In summary, our study reports that MTDH is overexpressed in both mouse and human prostate cancer and this elevated level of MTDH is critical for spontaneous prostate cancer progression and metastasis in vivo. Given the broad overexpression of MTDH in diverse cancer types, the tumor-promoting roles of MTDH may not be restricted to the models used in the study. Indeed, ongoing studies in our laboratory have also found that deletion of Mtdh drastically inhibits spontaneous mammary tumor formation and metastasis (unpublished data). Future studies are needed to understand the underlying mechanisms and signaling pathways that responsible for the tumor-promoting function of MTDH and to facilitate the development of MTDH-targeting therapeutics to control human cancer.

No potential conflicts of interest were disclosed.

Conception and design: L. Wan, Y. Kang

Development of methodology: L. Wan, Qi. Yang

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Wan, G. Hu, M. Yuan, R.T. Bronson, Qi. Yang, J. Siddiqui, K. J. Pienta

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. Wan, G. Hu, J. Siddiqui

Writing, review, and/or revision of the manuscript: L. Wan, J. Siddiqui, K. J. Pienta, Y. Kang

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Wan, M. Yuan, Qi. Yang, J. Siddiqui

Study supervision: L. Wan, Y. Wei, Y. Kang

Other (did the pathology for this paper): R.T. Bronson

The authors thank J.T. Eggenschwiler, J. Levorse, and M. Bhaumik for helping with the generation of Mtdh-knockout mice, and M. Shen for assistance and helpful discussion.

The work was supported by a Charlotte Elizabeth Procter Fellowship (L. Wan), grants from the Brewster Foundation, and the NIH (R01CA134519; Y. Kang), and grants from the NCI (U54CA143803, CA163124, CA143055, and CA093900; K.J. Pienta). This work was also supported by the Transgenic/Knockout and Flow Cytometry Shared Resources of the Cancer Institute of New Jersey (P30CA072720).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Siegel
R
,
Naishadham
D
,
Jemal
A
. 
Cancer statistics, 2013
.
CA Cancer J Clin
2013
;
63
:
11
30
.
2.
Hess
KR
,
Varadhachary
GR
,
Taylor
SH
,
Wei
W
,
Raber
MN
,
Lenzi
R
, et al
Metastatic patterns in adenocarcinoma
.
Cancer
2006
;
106
:
1624
33
.
3.
Hu
G
,
Chong
RA
,
Yang
Q
,
Wei
Y
,
Blanco
MA
,
Li
F
, et al
MTDH activation by 8q22 genomic gain promotes chemoresistance and metastasis of poor-prognosis breast cancer
.
Cancer Cell
2009
;
15
:
9
20
.
4.
Sarkar
D
,
Fisher
PB
. 
AEG-1/MTDH/LYRIC: clinical significance
.
Adv Cancer Res
2013
;
120
:
39
74
.
5.
Emdad
L
,
Das
SK
,
Dasgupta
S
,
Hu
B
,
Sarkar
D
,
Fisher
PB
. 
AEG-1/MTDH/LYRIC: signaling pathways, downstream genes, interacting proteins, and regulation of tumor angiogenesis
.
Adv Cancer Res
2013
;
120
:
75
111
.
6.
Wan
L
,
Kang
Y
. 
Pleiotropic roles of AEG-1/MTDH/LYRIC in breast cancer
.
Adv Cancer Res
2013
;
120
:
113
34
.
7.
Su
ZZ
,
Kang
DC
,
Chen
Y
,
Pekarskaya
O
,
Chao
W
,
Volsky
DJ
, et al
Identification and cloning of human astrocyte genes displaying elevated expression after infection with HIV-1 or exposure to HIV-1 envelope glycoprotein by rapid subtraction hybridization, RaSH
.
Oncogene
2002
;
21
:
3592
602
.
8.
Brown
DM
,
Ruoslahti
E
. 
Metadherin, a cell surface protein in breast tumors that mediates lung metastasis
.
Cancer Cell
2004
;
5
:
365
74
.
9.
Britt
DE
,
Yang
DF
,
Yang
DQ
,
Flanagan
D
,
Callanan
H
,
Lim
YP
, et al
Identification of a novel protein, LYRIC, localized to tight junctions of polarized epithelial cells
.
Exp Cell Res
2004
;
300
:
134
48
.
10.
Sutherland
HG
,
Lam
YW
,
Briers
S
,
Lamond
AI
,
Bickmore
WA
. 
3D3/lyric: a novel transmembrane protein of the endoplasmic reticulum and nuclear envelope, which is also present in the nucleolus
.
Exp Cell Res
2004
;
294
:
94
105
.
11.
Thirkettle
HJ
,
Girling
J
,
Warren
AY
,
Mills
IG
,
Sahadevan
K
,
Leung
H
, et al
LYRIC/AEG-1 is targeted to different subcellular compartments by ubiquitinylation and intrinsic nuclear localization signals
.
Clin Cancer Res
2009
;
15
:
3003
13
.
12.
Ash
SC
,
Yang
DQ
,
Britt
DE
. 
LYRIC/AEG-1 overexpression modulates BCCIPα protein levels in prostate tumor cells
.
Biochem Biophys Res Commun
2008
;
371
:
333
8
.
13.
Emdad
L
,
Sarkar
D
,
Su
ZZ
,
Randolph
A
,
Boukerche
H
,
Valerie
K
, et al
Activation of the nuclear factor kappaB pathway by astrocyte elevated gene-1: implications for tumor progression and metastasis
.
Cancer Res
2006
;
66
:
1509
16
.
14.
Sarkar
D
,
Park
ES
,
Emdad
L
,
Lee
SG
,
Su
ZZ
,
Fisher
PB
. 
Molecular basis of nuclear factor-κB activation by astrocyte elevated gene-1
.
Cancer Res
2008
;
68
:
1478
84
.
15.
Blanco
MA
,
Aleckovic
M
,
Hua
Y
,
Li
T
,
Wei
Y
,
Xu
Z
, et al
Identification of staphylococcal nuclease domain-containing 1 (SND1) as a metadherin-interacting protein with metastasis-promoting functions
.
J Biol Chem
2011
;
286
:
19982
92
.
16.
Yoo
BK
,
Santhekadur
PK
,
Gredler
R
,
Chen
D
,
Emdad
L
,
Bhutia
S
, et al
Increased RNA-induced silencing complex (RISC) activity contributes to hepatocellular carcinoma
.
Hepatology
2011
;
53
:
1538
48
.
17.
Meng
X
,
Zhu
D
,
Yang
S
,
Wang
X
,
Xiong
Z
,
Zhang
Y
, et al
Cytoplasmic metadherin (MTDH) provides survival advantage under conditions of stress by acting as RNA-binding protein
.
J Biol Chem
2012
;
287
:
4485
91
.
18.
Thirkettle
HJ
,
Girling
J
,
Warren
AY
,
Mills
IG
,
Sahadevan
K
,
Leung
H
, et al
LYRIC/AEG-1 is targeted to different subcellular compartments by ubiquitinylation and intrinsic nuclear localization signals
.
Clin Cancer Res
2009
;
15
:
3003
13
.
19.
Kikuno
N
,
Shiina
H
,
Urakami
S
,
Kawamoto
K
,
Hirata
H
,
Tanaka
Y
, et al
Knockdown of astrocyte-elevated gene-1 inhibits prostate cancer progression through upregulation of FOXO3a activity
.
Oncogene
2007
;
26
:
7647
55
.
20.
Greenberg
NM
,
DeMayo
F
,
Finegold
MJ
,
Medina
D
,
Tilley
WD
,
Aspinall
JO
, et al
Prostate cancer in a transgenic mouse
.
Proc Natl Acad Sci U S A
1995
;
92
:
3439
43
.
21.
Gingrich
JR
,
Barrios
RJ
,
Foster
BA
,
Greenberg
NM
. 
Pathologic progression of autochthonous prostate cancer in the TRAMP model
.
Prostate Cancer Prostatic Dis
1999
;
2
:
70
5
.
22.
Kaplan-Lefko
PJ
,
Chen
TM
,
Ittmann
MM
,
Barrios
RJ
,
Ayala
GE
,
Huss
WJ
, et al
Pathobiology of autochthonous prostate cancer in a pre-clinical transgenic mouse model
.
Prostate
2003
;
55
:
219
37
.
23.
Hurwitz
AA
,
Foster
BA
,
Allison
JP
,
Greenberg
NM
,
Kwon
E
. 
The TRAMP mouse as a model for prostate cancer
.
Curr Protoc Immunol
2001
;
Chapter 20:Unit 20.5
.
24.
Stryke
D
,
Kawamoto
M
,
Huang
CC
,
Johns
SJ
,
King
LA
,
Harper
CA
, et al
BayGenomics: a resource of insertional mutations in mouse embryonic stem cells
.
Nucleic Acids Res
2003
;
31
:
278
81
.
25.
Chiaverotti
T
,
Couto
SS
,
Donjacour
A
,
Mao
JH
,
Nagase
H
,
Cardiff
RD
, et al
Dissociation of epithelial and neuroendocrine carcinoma lineages in the transgenic adenocarcinoma of mouse prostate model of prostate cancer
.
Am J Pathol
2008
;
172
:
236
46
.
26.
Gingrich
JR
,
Barrios
RJ
,
Morton
RA
,
Boyce
BF
,
DeMayo
FJ
,
Finegold
MJ
, et al
Metastatic prostate cancer in a transgenic mouse
.
Cancer Res
1996
;
56
:
4096
102
.
27.
Foster
BA
,
Gingrich
JR
,
Kwon
ED
,
Madias
C
,
Greenberg
NM
. 
Characterization of prostatic epithelial cell lines derived from transgenic adenocarcinoma of the mouse prostate (TRAMP) model
.
Cancer Res
1997
;
57
:
3325
30
.
28.
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
.
29.
Ding
ZH
,
Wu
CJ
,
Jaskelioff
M
,
Ivanova
E
,
Kost-Alimova
M
,
Protopopov
A
, et al
Telomerase reactivation following telomere dysfunction yields murine prostate tumors with bone metastases
.
Cell
2012
;
148
:
896
907
.
30.
Wan
L
,
Pantel
K
,
Kang
Y
. 
Tumor metastasis: moving new biological insights into the clinic
.
Nat Med
2013
;
19
:
1450
64
.
31.
Liu
Y
,
Su
ZW
,
Li
G
,
Yu
CY
,
Ren
SL
,
Huang
DH
, et al
Increased expression of metadherin protein predicts worse disease-free and overall survival in laryngeal squamous cell carcinoma
.
Int J Cancer
2013
;
133
:
671
9
.
32.
Yu
C
,
Liu
Y
,
Tan
H
,
Li
G
,
Su
Z
,
Ren
S
, et al
Metadherin regulates metastasis of squamous cell carcinoma of the head and neck via AKT signalling pathway-mediated epithelial-mesenchymal transition
.
Cancer Lett
2014
;
343
:
258
67
.
33.
Zhu
K
,
Dai
Z
,
Pan
Q
,
Wang
Z
,
Yang
GH
,
Yu
L
, et al
Metadherin promotes hepatocellular carcinoma metastasis through induction of epithelial-mesenchymal transition
.
Clin Cancer Res
2011
;
17
:
7294
302
.
34.
Yoo
BK
,
Emdad
L
,
Su
ZZ
,
Villanueva
A
,
Chiang
DY
,
Mukhopadhyay
ND
, et al
Astrocyte elevated gene-1 regulates hepatocellular carcinoma development and progression
.
J Clin Invest
2009
;
119
:
465
77
.