Advertisement
Cancer Genes and Networks

Telomere DNA Damage Signaling Regulates Prostate Cancer Tumorigenesis

Jianchun Wu and David L. Crowe
DOI: 10.1158/1541-7786.MCR-19-1129 Published September 2020

Abstract

Telomere shortening has been demonstrated in benign prostatic hypertrophy (BPH), which is associated with prostate epithelial cell senescence. Telomere shortening is the most frequently observed genetic alteration in prostatic intraepithelial neoplasia, and is associated with poor clinical outcomes in prostate cancer. Gene expression database analysis revealed decreased TRF2 expression during malignant progression of the prostate gland. We reasoned that reduced TRF2 expression in prostate epithelium, by activating the telomere DNA damage response, would allow us to model both benign and malignant prostate disease. Prostate glands with reduced epithelial TRF2 expression developed age- and p53-dependent hypertrophy, senescence, ductal dilation, and smooth muscle hyperplasia similar to human BPH. Prostate tumors with reduced TRF2 expression were classified as high-grade androgen receptor–negative adenocarcinomas, which exhibited decreased latency, increased proliferation, and distant metastases. Prostate cancer stem cells with reduced TRF2 expression were highly tumorigenic and maintained telomeres both by telomerase and alternative lengthening (ALT). Telomerase inhibition in prostate glands with reduced TRF2 expression produced significant reduction in prostate tumor incidence by halting progression at intraepithelial neoplasia (PIN). These lesions were highly differentiated, exhibited low proliferation index, and high apoptotic cell fraction. Prostate tumors with reduced TRF2 expression and telomerase inhibition failed to metastasize and did not exhibit ALT.

Implications: Our results demonstrate that the telomere DNA damage response regulates BPH, PIN, and prostate cancer and may be therapeutically manipulated to prevent prostate cancer progression.

Introduction

Telomeres protect chromosomes from DNA damage (1). Telomeres form a loop mediated by invasion of the G-rich DNA strand (2). Short telomeres induce DNA damage responses at chromosomes resulting in apoptosis or cellular senescence.

The double stranded DNA binding protein telomere repeat factor 2 (TRF2) protects telomeres (3). Loss of TRF2 induces fusion of telomeres, and cells undergo apoptosis and senescence (4). ATM and p53 mediates DNA damage signaling due to loss of TRF2, resulting in apoptosis (5). These results illustrate the importance of TRF2 in regulating telomere DNA damage response.

Cells stabilize telomeres by upregulation of telomerase (6). Telomerase lengthens telomeres via its Terc RNA template (7). In Terc−/− mice, cells with critically short telomeres demonstrate DNA damage signaling, chromosome fusions, aneuploidy, and programmed cell death (8). Telomerase can inhibit telomere DNA damage and result in cellular immortalization. Telomeres also are maintained by alternative lengthening of telomeres (ALT) associated with ALT-associated promyelocytic leukemia bodies (APB; ref. 9). Circular C-rich strands are products of this pathway and are present in ALT+ cells (10). Telomerase is frequently upregulated in cancer cell lines and tumors (11).

Prostate epithelial cell senescence is a common feature of age-related conditions such as benign prostatic hypertrophy (BPH; ref. 12). The cause of senescence in prostate epithelia in BPH is unclear, but telomere shortening has been demonstrated in these lesions (13). BPH also is frequently characterized by smooth muscle hyperplasia, which may be driven by signaling from senescent epithelial cells (for review, see ref. 14).

Prostate cancer is the second most common malignant tumor in men, with over 160,000 cases and 26,000 deaths each year in the United States (15). Prostate cancer may begin as precursor lesions termed prostatic intraepithelial neoplasia (PIN) which is characterized by hyperplasia of luminal epithelial cells (for review, see ref. 16). Telomere shortening is the most frequently observed genetic alteration in PIN and is predictive of prostate cancer (17–20). Telomere DNA damage response including telomere fusions also was observed in PIN (21). The vast majority of prostate cancers are classified as androgen receptor positive luminal adenocarcinoma. Telomere shortening was associated poor clinical outcomes in prostate cancer, including early recurrence and metastasis (22–28). High levels of telomerase activity were also detected in prostate cancer (22, 29, 30). Prostate cancer cells with high telomerase activity exhibited cancer stem cell properties (31). A mouse prostate cancer model with short telomeres and telomerase activation resulted in bone metastases (32).

Stem cell populations in prostate cancer can self-renew and give rise to basal and luminal cells (33). Lineage tracing studies revealed that prostate basal or luminal cells generated unique self-sustaining populations in vivo (34). Prostate cancers containing basal and luminal cells could be initiated from basal-specific loss of the PTEN tumor suppressor. Prostate cancer could also be initiated from luminal-specific loss of PTEN, which contained cells expressing both basal and luminal markers. A separate study demonstrated rare bipotent basal progenitors in the prostate gland, and that basal cells could initiate prostate tumors (35).

Given the importance of the telomere DNA damage response in regulating prostate epithelial cell homeostasis and disease, we hypothesized that this signaling pathway regulates benign conditions such as BPH, potentially malignant PIN lesions, and prostate cancer. Gene expression database analysis revealed reduced TRF2 expression during malignant progression of the prostate gland. We reasoned that reduced TRF2 expression in prostate epithelium, by activating the telomere DNA damage response, would allow us to model both benign and malignant prostate disease. Our results demonstrate that the telomere DNA damage response regulates BPH, PIN, and prostate cancer and may be used therapeutically to prevent prostate cancer progression.

Materials and Methods

Mouse breeding and procedures

Mouse strains and experimental procedures were approved by the institutional animal care committee. The Tg(KRT14-Cre)1Amc/J, B6;129P2-Terf2tm1Tdl/J, B6.129P2-Trp53tm1Brn/J, C57Bl/6-Tg(TRAMP)8247Ng/J, and B6.Cg-TercTm1Rdp/J–mutant mouse strains were purchased from The Jackson Laboratory. Mice were crossed to create K14Cre;TRF2+/+;p53+/+; K14Cre;TRF2f/f;p53+/+, K14Cre;TRF2f/f;p53f/f, K14Cre;TRF2+/+;TAg, K14Cre;TRF2f/f;TAg, K14Cre;Terc−/−;TAg, K14Cre;TRF2f/f;Terc−/−;TAg offspring. Littermates were used to control for genetic background and assigned to experimental or control groups by genotype. The latency, number, and volume of tumors were recorded for each animal. Prostate glands and tumors were fixed in 4% buffered formaldehyde, flash frozen, and stored at −80°C, or trypsin dissociated and cryopreserved in liquid nitrogen.

qRT-PCR

RNA was extracted from prostate glands and tumors and reverse transcribed according to manufacturer's instructions (Invitrogen; ref. 36). cDNA was amplified using TRF2 5′- ACTAGCTTACGGAGTCTGC -3′ and 5′-AAGGGGGAGTTTCAGGAGAG -3′ and p53 5′- TTGATGGAGAGTATTTCACCC -3′ and 5′- TTGACCATTGTTTTTTTATGGC -3′ primers. β-Actin was amplified using primers 5′-AAAAGCCACCCCCACTCCTAAG-3′ and 5′- TCAAGTCAGTGTACAGGCCAGC-3′ at 94°C for 25 seconds, 55°C for 1 minute, and 72°C for 1 minute. Quantitative PCR was performed using the StepOnePlus system (Thermo Fisher Scientific) and normalized using the 2ΔΔCt method.

Histopathology, FISH, immunofluorescence, and IHC

Fixed mouse prostate glands and tumors was dehydrated in ethanol, cleared in xylene, and embedded in paraffin (36). Sections were deparaffinized and stained with hematoxylin and eosin. For telomeric FISH, deparaffinized tissue sections were denatured with Cy3-labeled telomeric peptide nucleic acid probe (TTAGGG)3 in 70% formamide at 80°C for 10 minutes, followed by overnight incubation at room temperature. After washing, sections were blocked with 10% normal serum and incubated with anti-53BP1 antibody overnight at room temperature. After washing, sections were incubated with anti-IgG secondary antibody conjugated to AlexaFluor 488. After washing, colocalized DNA damage and telomere signals were visualized by fluorescence microscopy (Zeiss LSM 710 META). In separate experiments, PML protein was localized at telomeres of tumor sections using the same immunofluorescence/FISH protocol. Phospho-ATM, phospho-Chk2, p53, lactate dehydrogenase C, golgin B1, Sca1, and CD49f proteins were localized in mouse prostate glands and tumors using the same immunofluorescence protocol. For IHC analysis, sections were blocked using 10% normal serum and incubated with anti-PCNA, keratin 14, keratin 18, chromogranin A, androgen receptor, pATM, pATR, pChk1, pChk2, p53, smooth muscle actin, or laminin antibodies overnight at room temperature. After washing, sections were incubated with biotinylated secondary antibody at room temperature for 10 minutes. After additional washing, sections were incubated with streptavidin-conjugated horseradish peroxidase for 10 minutes at room temperature. Antigen–antibody complexes were detected by incubation with peroxide substrate solution containing aminoethylcarbazole chromogen. Data were obtained using image analysis software and analyzed by ANOVA with Tukey post hoc test.

Cell death analysis

Prostate glands and tumor sections were incubated with terminal deoxynucleotidyl transferase and dUTP-fluorescein for 1 hour at 37 °C according to the manufacturer's recommendations (Roche Applied Sciences). After washing, apoptotic cells were visualized by fluorescence microscopy. The percent fluorescent cells were determined using image analysis software. Data were analyzed by ANOVA with Tukey post hoc test.

Telomere length analysis of stem cells

We used quantitative PCR to measure average telomere length ratios of sorted cells from mouse prostate glands and tumors (36). Telomeric primers were 5′-CGGTTTGTTTGGGTTTGGGTTTGGGTTTGGGTTTGGGTT-3′ and 5′-GGCTTGCCTTACCCTTACCCTTACCCTTACCCTTAACCCT-3′. Primers for the mouse acidic ribosomal phosphoprotein PO (36B4) internal control gene were 5′-ACTGGTCTAGGACCCGAGAAG-3′ and 5′-TCAATGGTGCCTCTGGAGATT-3′. Each reaction for the telomere assay included 12.5 μL of Sybr Green PCR Master Mix (Applied Biosystems), 300 nmol/L of each primer, and 20 ng of genomic DNA. Samples were amplified in triplicate with reaction conditions of 95°C for 10 minutes followed by 30 cycles at 95°C for 15 seconds, and 56°C for 1 minute. For the 36B4 assay, reaction conditions were 95°C for 10 minutes followed by 35 cycles at 95°C for 15 seconds, 52°C annealing for 20 seconds, and extension at 72°C for 30 seconds. The relative input amount of telomere PCR was divided by the relative input amount of the 36B4 PCR. PCR was performed three times for each sample, and the average of these ratios was reported as the average telomere length ratio. Data were analyzed by ANOVA with Tukey post hoc test.

Transmission electron microscopy

Prostate glands were fixed in phosphate buffered 2% glutaraldehyde, postfixed in phosphate buffered 1% osmium tetroxide, dehydrated in acetone series, infiltrated in propylene oxide/resin, and embedded in Epon resin. Sections (0.1 μm) were stained with uranyl acetate and lead citrate and imaged using a JEOL JEM-1220 electron microscope.

Western blot analysis

Protein was extracted from prostate glands and tumors in 1× Laemmli buffer. Seventy-five micrograms total cellular protein was separated by SDS-PAGE. Proteins were electroblotted to polyvinylidene difluoride membranes. Blots were incubated with antibodies to phospho-ATM, total ATM, pChk2, total Chk2, p53, or β-actin for 16 hours at 4°C. After washing, blots were incubated for 30 minutes at room temperature with anti-IgG secondary antibody conjugated to horseradish peroxidase. Bands were visualized by the enhanced chemiluminescence method, normalized to β-actin expression, and quantitated by image analysis software (36).

Metaphase chromosomal spreads

Primary metastatic prostate tumor cells from both genotypes were treated with 0.1 μg/mL colcemid for 3 hours at 37°C. Cell pellets were suspended in 60 mmol/L KCl and incubated at room temperature for 30 minutes. After centrifugation, cell pellets were fixed three times in 3:1 methanol:acetic acid and spotted onto microscope slides. Cy3-labeled telomeric probe was hybridized to the metaphase spreads overnight at room temperature. After washing, slides were coverslipped with antifade mounting medium containing DAPI, photographed using fluorescence microscopy, and analyzed using digital image software and t test (36).

FACS

Prostate tumors were dissociated by trypsinization, washed in PBS, and incubated with phycoerythrin conjugated Sca1 and AlexaFluor 488–conjugated CD49f antibodies. Samples were washed in PBS followed by FACS. Data were analyzed by t test.

Tumorigenicity experiments

A total of 102–105 Sca1/CD49f+ or Sca1/CD49f− cells from primary tumors were mixed with 106 urogenital sinus mesenchymal cells in Matrigel and injected subcutaneously into 2-month-old NU/J male mice. Mice were examined weekly for tumor formation for up to 6 months. Five transplanted tumors per group were analyzed. Mice underwent complete necropsy; tumors were dissected and portions of each were fixed in 4% buffered formaldehyde for microscopy or dissociated for flow cytometry.

Tumorsphere culture

Sorted Sca1/CD49f+ and Sca1/CD49f− cells were cultured in suspension in stem cell medium (3:1 DMEM/F12 medium, 1× B27 supplement, 10 ng/mL EGF, 25 ng/mL basic fibroblast growth factor, 0.2% heparin, 40 μg/mL gentamicin, 2.5 μg/mL amphotericin B for 7 days at 37° C in a humidified atmosphere of 5% CO2. Tumorspheres were photographed using an Olympus X51 inverted phase contrast microscope.

Telomeric circular DNA analysis

Genomic DNA extracted from sorted tumor cells was digested with HinfI and RsaI restriction enzymes (36). Telomeric circular DNA was amplified with ϕ29 DNA polymerase in the absence of dCTP and blotted to nylon membranes. Blots were hybridized with 32P-labeled (TTAGGG)4 probe overnight at 50°C. After washing in 4× SCC for 30 minutes at room temperature, blots were exposed to autoradiographic film for 16 hours at −80°C.

Telomeric repeat amplification protocol

Telomeric repeat amplification was performed using a commercially available kit (TeloTAGGG Telomerase PCR ELISA) according to manufacturer's protocol (Roche Applied Science).

Statistical

Telomere DNA damage in human prostate glands and tumors were analyzed by Fisher exact test. Flow cytometry of prostate cancer stem cells from TRF2+/+ and TRF2f/f tumors were analyzed by t test. Other data were analyzed by ANOVA with Tukey post hoc test. All samples were analyzed in each experiment using at least three experimental replicates. We did not exclude data from the study. Investigators were blinded with respect to genotype. All tests were two tailed.

Antibodies

The following antibodies were used: keratin 14AB7800, keratin 18 AB668, laminin AB11575, ATM AB78, pChk2 AB85743, TRF2 AB13579, golgin B1 AB24568, pChk1 AB47318 (Abcam); PCNA SC7907, pATM SC47739, PML SC5621, AR SC815, chromogranin A SC13090, pATR SC109912, ATR SC1887, Chk1 SC7898 (Santa Cruz Biotechnology); 53BP1 NB100-304 (Novus Biologicals); p53 BD5544147 (Becton Dickinson); Sca1 60032PE (Stem Cell Technologies); SMA A5228, CD49f 60037AD (Millipore-Sigma); and β-actin 4970T (Cell Signaling Technology).

Data availability

Data are available to qualified investigators on reasonable request.

Results

Telomere shortening is a frequent genetic alteration in prostate carcinogenesis, presumably resulting from the telomere DNA damage response. To characterize telomere DNA damage response in human prostate carcinogenesis, we analyzed colocalization of 53BP1+ DNA damage foci at telomeres in 80 prostate tumors and matched histopathologically normal adjacent prostate tissue. Telomere DNA damage was significantly increased in prostate tumors (70% vs. 26%; P < 0.0001; Supplementary Fig. S1A–S1C). Histopathology of prostate gland and tumor is shown by hematoxylin and eosin staining in Supplementary Fig. S1D and S1E. Telomere DNA damage was detected in tumors of all histopathologic grades, sizes, Gleason scores, and PSA levels in this patient cohort (Supplementary Table S1). To identify potential mechanisms for telomere DNA damage in prostate carcinogenesis, we searched the Gene Expression Omnibus database. Analysis of GDS1439 revealed reduced TRF2 expression in primary (P < 0.003) and metastatic (P < 0.0006) prostate cancers compared with prostate gland (Supplementary Fig. S1F). Expression of telomere DNA damage signaling proteins ATR (P < 0.0002) and Chk2 (P < 0.009) was increased in metastatic prostate tumors compared to prostate glands and tumors (Supplementary Fig. S1F). We used the K14Cre;TRF2f/f conditional mutant mouse to model reduced TRF2 expression in prostate epithelium. During mouse prostate development, Cre-mediated recombination of TRF2 exons 1 and 2 in the keratin 14+ urogenital sinus epithelium results in TRF2-null basal and luminal prostate epithelial cells (37). TRF2 expression in wild-type and null prostate epithelium is shown in Supplementary Fig. S2A. By 9 months of age, K14Cre;TRF2f/f prostate glands exhibited prominent hypertrophy with ductal dilation (4-fold increase in lumen area; P < 0.004) and smooth muscle hyperplasia (4-fold increased smooth muscle thickness; P < 0.01) compared with K14Cre;TRF2+/+ glands of the same age (n = 20 glands; Fig. 1A and B). K14Cre;TRF2f/f prostate epithelium exhibited pathologic degeneration as shown by low cuboidal morphology compared with the columnar epithelium of K14Cre;TRF2+/+ glands (Fig. 1D and E). Smooth muscle hyperplasia was confirmed using smooth muscle actin IHC (Fig. 1G and H). K14Cre;TRF2f/f;p53+/+ prostate epithelium also exhibited loss of basement membrane integrity as shown by laminin IHC (Fig. 1J and K). Strikingly, K14Cre;TRF2f/f prostate glands exhibited large numbers of sperm in ductal lumina that was confirmed using the sperm-specific marker lactate dehydrogenase C (Fig. 1M and N). To determine whether these changes in the K14Cre;TRF2f/f prostate gland were due to telomere DNA damage response, we crossed the K14Cre;TRF2f/f mouse with p53f/f animals to inhibit telomere DNA damage signaling in prostate epithelium. Nine-month-old K14Cre;TRF2f/f;p53f/f prostate glands did not exhibit hypertrophy, ductal dilation, smooth muscle hyperplasia, epithelial degeneration, loss of basement membrane integrity, or sperm influx observed in K14Cre;TRF2f/f;p53+/+ glands (Fig. 1C, F, I, L, and O). Testes (which do not express keratin 14) were histopathologically normal in all three genotypes and fertility was not impaired (Supplementary Fig. S2B–S2D). These results indicate that telomere DNA damage signaling resulting from decreased TRF2 expression produces prostate gland pathology exhibiting features of human BPH.

Figure 1.
Figure 1.

Smooth muscle hyperplasia and epithelial degeneration in K14Cre;TRF2f/f prostate glands. Histopathologic analysis of prostate glands from K14Cre;TRF2+/+;p53+/+ (A), K14Cre;TRF2f/f;p53+/+ (B), and K14Cre;TRF2f/f;p53f/f (C) mice was determined by hematoxylin and eosin staining. Scale bar, 50 μm. Histopathologic analysis of prostate glands from K14Cre;TRF2+/+;p53+/+ (D), K14Cre;TRF2f/f;p53+/+ (E), and K14Cre;TRF2f/f;p53f/f (F) mice was determined by hematoxylin and eosin staining. Scale bar, 10 μm. Smooth muscle actin localization in prostate glands from K14Cre;TRF2+/+;p53+/+ (G), K14Cre;TRF2f/f;p53+/+ (H), and K14Cre;TRF2f/f;p53f/f (I) mice was determined by IHC. Nuclei were counterstained with hematoxylin. Scale bar, 10 μm. Basement membrane localization in prostate glands from K14Cre;TRF2+/+;p53+/+ (J), K14Cre;TRF2f/f;p53+/+ (K), and K14Cre;TRF2f/f;p53f/f (L) mice was determined by laminin IHC. Nuclei were counterstained with hematoxylin. Scale bar, 10 μm. Lactate dehydrogenase C (a specific molecular marker for sperm) localization in prostate glands from K14Cre;TRF2+/+;p53+/+ (M), K14Cre;TRF2f/f;p53+/+ (N), and K14Cre;TRF2f/f;p53f/f (O) mice was determined by immunofluorescence microscopy. Nuclei were counterstained with DAPI. Scale bar, 10 μm. Representative photomicrographs are shown.

We next characterized K14Cre;TRF2f/f;p53+/+ prostate gland pathology at the cellular and molecular levels. Given that prostate epithelial senescence is a feature of human BPH, we determined expression of senescence associated heterochromatin foci (SAHF) in mouse prostate glands by histone H3K9me3 immunofluorescence microscopy. K14Cre;TRF2f/f;p53+/+ prostate glands exhibited increased SAHF+ epithelial cells (51% vs. 1%; P < 0.0005; Supplementary Figs. S3A and S3B and S4A). K14Cre;TRF2f/f;p53+/+ prostate epithelial cells exhibited increased apoptotic cell fraction (54% vs. 0.1%; P < 0.0001; Supplementary Figs. S3D and S3E and S4B) compared with K14Cre;TRF2+/+;p53+/+ epithelium. K14Cre;TRF2f/f;p53+/+ prostate epithelium exhibited decreased androgen receptor–positive cell fraction (0.1% vs. 11%; P < 0.002; Supplementary Figs. S3G and S3H and S4C) compared with K14Cre;TRF2+/+;p53+/+ glands, likely due to luminal cell apoptosis. K14Cre;TRF2f/f;p53+/+ prostate epithelium exhibited decreased proliferating cells (5% vs. 0.1%; P < 0.005; Supplementary Figs. S3J and S3K and S4D) compared with K14Cre;TRF2+/+;p53+/+ glands as determined by PCNA IHC. K14Cre;TRF2f/f;p53+/+ prostate epithelium exhibited decreased basal cells (12% vs. 0.1%; P < 0.004; Supplementary Figs. S3M and S3N and S4E) compared with K14Cre;TRF2+/+;p53+/+ glands as determined by K14 IHC, likely due to basal cell apoptosis. Remaining ductal epithelial cells were luminal phenotype in both K14Cre;TRF2f/f;p53+/+ and K14Cre;TRF2+/+;p53+/+ prostate glands as determined by K18 IHC (Supplementary Fig. S3P and S3Q). Inhibiting telomere DNA damage response in K14Cre;TRF2f/f;p53f/f prostate glands prevented epithelial cell senescence and death (Supplementary Figs. S3C, S3F, S3I, S3L, S3O, and S3R and S4A–S4E). These results indicate that TRF2-mediated telomere DNA damage inhibits proliferation of prostate epithelium and induces cell death.

Our histopathologic analysis of K14Cre;TRF2f/f;p53+/+ prostate glands revealed transition from columnar to low cuboidal epithelium with loss of apical cytoplasm (Fig. 1). Second, K14Cre;TRF2f/f;p53+/+ prostate glands exhibited large numbers of sperm in ductal lumina. This finding is anatomically possible via retrograde migration of sperm into the prostate gland. Taken together these two findings suggest a severe secretory defect in K14Cre;TRF2f/f;p53+/+ prostate epithelium. To begin our examination of the secretory apparatus of prostate gland epithelium, we first used golgin B1 expression that is a specific molecular marker of Golgi apparatus. Golgin B1 immunofluorescence microscopy revealed strong expression in the apical cytoplasm of K14Cre;TRF2+/+;p53+/+ prostate luminal epithelial cells that was not detected in K14Cre;TRF2f/f;p53+/+ prostate epithelium (Supplementary Fig. S5A and S5B), indicating reduced golgin B1 expression or Golgi apparatus in these glands. Golgi apparatus was detected in both K14Cre;TRF2+/+;p53+/+ and K14Cre;TRF2f/f;p53+/+ prostate epithelium by transmission electron microscopy (Supplementary Fig. S5C and S5D). However, secretory vesicles were markedly different in K14Cre;TRF2f/f;p53+/+ prostate epithelium. Secretory vesicles in normal K14Cre;TRF2+/+;p53+/+ prostate epithelial cells were large, elongated, and contained electron dense contents (Supplementary Fig. S5E). The ductal lumina of these glands were filled with electron dense granules released from secretory vesicles. In contrast, K14Cre;TRF2f/f;p53+/+ secretory vesicles were smaller, ovoid, and contained less electron dense contents (Supplementary Fig. S5F). The ductal lumina of K14Cre;TRF2f/f;p53+/+ glands largely lacked the electron dense secreted granules found in the normal prostate gland. These results indicate that the telomere DNA damage response induces molecular and ultrastructural changes in prostate epithelium resulting in severe secretory defects.

We next characterized the telomere DNA damage response in prostate glands from the three genotypes. K14Cre;TRF2f/f;p53+/+ prostate epithelium exhibited increased cells with 53BP1+ DNA damage foci at telomeres (28% vs. 1%; P < 0.002; Supplementary Figs. S6A and S6B, S7A), ATM+ cells (32% vs. 1%; P < 0.001; Supplementary Figs. S6D and S6E and S7B), pChk2+ cells (38% vs. 2%; P < 0.009; Supplementary Figs. S6G and S6H and S7C), and p53+ cells (25% vs. 1%; P < 0.0007; Supplementary Fig. S6J and S6K and S7D) compared with K14Cre;TRF2+/+;p53+/+ glands. K14Cre;TRF2f/f;p53f/f prostate epithelium exhibited telomere DNA damage response with the exception of p53 induction, reflecting its downstream function in the signaling pathway (Supplementary Figs. S6C, S6F, S6I, and S6L and S7A–S7D). K14Cre;TRF2f/f;p53+/+ and K14Cre;TRF2f/f;p53f/f prostate epithelial cells exhibited significant telomere shortening (1.4 vs. 1.8; P < 0.04; Supplementary Fig. S6M) compared with K14Cre;TRF2+/+;p53+/+ glands. These results indicate that reduced TRF2 expression induces telomere DNA damage signaling and telomere shortening in prostate epithelium.

Human BPH is not associated with increased risk of prostate cancer (38), and we did not detect prostate tumors in K14Cre;TRF2f/f;p53+/+ or K14Cre;TRF2f/f;p53f/f mice. However given the importance of reduced TRF2 expression and telomere shortening in the development of human prostate cancer, we characterized the effects of reduced TRF2 expression in the K14Cre;TRF2f/f;TAg model. Interestingly, there was a 25% reduction in prostate tumor formation in K14Cre;TRF2f/f;TAg mice at 1 year (P < 0.05; n = 20; Fig. 2A), indicating that telomere DNA damage is detrimental to tumor initiation in this model. However, mean prostate tumor latency was significantly shorter in K14Cre;TRF2f/f;TAg mice (152 vs. 177 days; P < 0.02; Fig. 2B) compared with K14Cre;TRF2+/+;TAg glands. K14Cre;TRF2f/f;TAg prostate tumors were histopathologically classified as poorly differentiated adenocarcinoma (85%), while K14Cre;TRF2+/+;TAg cancers were well differentiated adenocarcinoma (80%; P < 0.0001; n = 20 tumors/group; Fig. 2C, D, and K). Prostate tumors from both genotypes metastasized to abdominal lymph nodes in 60% to 70% of cases, and to lung and liver in 10% to 15% of cases (no significant difference; Fig. 2E–J). These results indicate that reduced TRF2 expression inhibits prostate tumor initiation but decreases latency and impairs differentiation.

Figure 2.
Figure 2.

Rapid onset of high-grade prostate tumors in K14Cre;TRF2f/f;TAg mice. A, Fraction of K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg mice that developed prostate tumors. B, Latency of K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors. Percent tumor-free mice is shown on the y-axis and time (days) is shown on the x-axis. Histopathology of K14Cre;TRF2+/+;TAg (C) and K14Cre;TRF2f/f;TAg (D) primary prostate tumors was determined by hematoxylin and eosin staining. Histopathology of K14Cre;TRF2+/+;TAg (E) and K14Cre;TRF2f/f;TAg (F) lymph node metastases was determined by hematoxylin and eosin staining. Histopathology of K14Cre;TRF2+/+;TAg (G) and K14Cre;TRF2f/f;TAg (H) lung metastases was determined by hematoxylin and eosin staining. Histopathology of K14Cre;TRF2+/+;TAg (I) and K14Cre;TRF2f/f;TAg (J) liver metastases was determined by hematoxylin and eosin staining. Scale bar, 10 μm. K, Quantitation of well- and poorly differentiated prostate tumors as determined by histopathology in K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg mice.

Prostate tumors of both genotypes were luminal phenotype as demonstrated by K18 IHC (Supplementary Fig. S8A and S8B). No prostate tumors from either genotype exhibited basal or neuroendocrine differentiation as demonstrated by K14 and chromogranin A IHC (Supplementary Fig. S8C–S8F). K14Cre;TRF2f/f;TAg prostate tumors exhibited decreased AR+ cell fraction (2% vs. 29%; P < 0.006; Supplementary Fig. S8G–S8I) compared with K14Cre;TRF2+/+;TAg cancers. K14Cre;TRF2f/f;TAg prostate tumors exhibited increased proliferating cells (72% vs. 21%; P < 0.001; Supplementary Fig. S8J–S8L) compared with K14Cre;TRF2+/+;TAg cancers as demonstrated by PCNA IHC. Apoptotic cell fractions in prostate tumors from both genotypes were low (<1% positive; no significant difference; Supplementary Fig. S8M and S8N). These results indicate that reduced TRF2 expression inhibits AR+ differentiation and increases cell proliferation in prostate tumors.

K14Cre;TRF2f/f;TAg prostate tumors exhibited increased 53BP1+ DNA damage foci at telomeres (31% vs. 0.8%; P < 0.0008; Fig. 3A and B) compared with K14Cre;TRF2+/+;TAg cancers. K14Cre;TRF2f/f;TAg prostate tumors exhibited 11-fold increased pATM, 4-fold increased pChk2, and 16-fold increased p53 expression by Western blot analysis (P < 0.001; Fig. 3C). Telomere FISH on metaphase spreads of K14Cre;TRF2f/f;TAg tumor cells grown as tumorspheres revealed increased signal free ends (18% vs. 2%; P < 0.01; Fig. 3D and E) and aneuploidy (modal chromosome number 45 vs. 40; P < 0.05). K14Cre;TRF2f/f;TAg prostate tumor cells exhibited multiple shortened chromosomes indicative of breakage and nonhomologous end joining during tumorigenesis (48%; P < 0.00003; Fig. 3E). These results indicate that reduced TRF2 expression activates telomere DNA damage response and induces genomic instability in prostate cancers.

Figure 3.
Figure 3.

Telomere DNA damage signaling and genomic instability in K14Cre;TRF2f/f;TAg prostate tumors. Localization of the DNA damage response protein 53BP1 at telomeres (FISH) in K14Cre;TRF2+/+;TAg (A) and K14Cre;TRF2f/f;TAg (B) prostate tumors is shown by immunofluorescence microscopy. Arrowheads indicate colocalization of 53BP1 protein at telomeres. Nuclei were counterstained with DAPI. Scale bar, 5 μm. C, TRF2 expression and activation of DNA damage signaling proteins ATM, Chk2, and p53 in K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors was determined by Western blot analysis. β-Actin expression was used to demonstrate equal protein loading in each lane. Quantitation of normalized signals by digital image analysis is shown at right. Telomere FISH of metaphase spreads from K14Cre;TRF2+/+;TAg (D), and K14Cre;TRF2f/f;TAg (E) prostate tumor cells. Metaphase chromosomes were counterstained with DAPI. Scale bar, 1 μm. Signal free ends are indicated (sf) and shortened chromosomes are shown at arrowhead.

We next examined a well characterized prostate cancer stem cell population (Sca1/CD49f+) in tumors from both genotypes (39). K14Cre;TRF2f/f;TAg prostate tumors exhibited decreased average telomere length (0.5 vs. 1.3; P < 0.004; Fig. 4A) in Sca1/CD49f+ and Sca1/CD49f− cells. K14Cre;TRF2f/f;TAg prostate cancers exhibited depletion of the Sca1/CD49f+ stem cell population (1% vs. 29.9%; P < 10−6; Fig. 4B and C) as determined by flow cytometry. Sca1/CD49f+ cells from both genotypes formed tumorspheres in suspension culture that were not observed when Sca1/CD49f− were cultured under the same conditions (Fig. 4D and E). Sca1/CD49f+ cells also were demonstrated in prostate tumor sections by immunofluorescence microscopy (Fig. 4F). As few as 103 Sca1/CD49f+ cells from both genotypes formed tumors when transplanted to immunodeficient mice (Fig. 4G and H) while 105 Sca1/CD49f− cells were required for tumor formation in this microenvironment. Transplanted K14Cre;TRF2+/+;TAg Sca1/CD49f+ prostate cancer cells reconstituted well-differentiated adenocarcinoma in immunodeficient mice (Fig. 4I). Transplanted K14Cre;TRF2f/f;TAg Sca1/CD49f+ prostate cancer cells reconstituted poorly differentiated adenocarcinoma in immunodeficient mice (Fig. 4J). Sca1/CD49f+ cells were detected in transplanted tumors from both genotypes by flow cytometry (Fig. 4K and L) and immunofluorescence microscopy (Fig. 4M). These results indicate that reduced TRF2 expression produces telomere shortening and depletion of prostate cancer stem cells. However, these cells meet the functional criteria for cancer stem cells of tumor initiation, self-renewal, and reconstitution of other tumor cell types.

Figure 4.
Figure 4.

Telomere shortening and cancer stem cell depletion in K14Cre;TRF2f/f;TAg prostate tumors. A, Average telomere length ratios in cancer stem (Sca1/CD49f+) and non-stem (Sca1/CD49f−) cells from K14Cre;TRF2+/+;TAg and K14Cre;TRFf/f;TAg prostate tumors. Error bars, SEM. Prostate cancer stem cell fractions in K14Cre;TRF2+/+;TAg (B) and K14Cre;TRFf/f;TAg (C) prostate tumors (n = 20/group). Sca1 and CD49f log scales are shown. K14Cre;TRFf/f;TAg Sca1/CD49f− (D) and Sca1/CD49f+ (E) prostate tumor cells in suspension culture for 7 days is shown by phase contrast microscopy. Scale bar, 10 μm. F, K14Cre;TRFf/f;TAg Sca1/CD49f+ prostate cancer stem cells are shown in tumor sections by immunofluorescence microscopy. Nuclei were counterstained with DAPI. Scale bar, 10 μm. G, NU/J mouse with tumor derived from transplanted K14Cre;TRFf/f;TAg Sca1/CD49f+ prostate cancer stem cells. H, Dissected tumor derived from transplanted K14Cre;TRFf/f;TAg Sca1/CD49f+ prostate cancer stem cells. Histopathology of tumors from transplanted K14Cre;TRF2+/+;TAg (I) and K14Cre;TRF2f/f;TAg (J) Sca1/CD49f+ prostate cancer stem cells was determined by hematoxylin and eosin staining. Scale bar, 10 μm. Prostate cancer stem cell fraction in tumors derived from transplanted K14Cre;TRF2+/+;TAg (K) and K14Cre;TRF2f/f;TAg (L) Sca1/CD49f+ prostate cancer stem cells. Sca1 and CD49f log scales are shown. M, K14Cre;TRFf/f;TAg Sca1/CD49f+ prostate cancer stem cells are shown in transplanted tumor sections by immunofluorescence microscopy. Nuclei were counterstained with DAPI. Scale bar, 10 μm.

Despite telomere DNA damage response and shortening, chromosomes from K14Cre;TRF2f/f;TAg prostate cancer cells exhibited telomere FISH signals (Fig. 3) indicating functional telomere maintenance mechanisms. We examined alternative lengthening of telomeres (ALT) in Sca1/CD49f+ and Sca1/CD49f− cell populations from K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors using specific and sensitive telomere circular DNA analysis. Sca1/CD49+ cells from K14Cre;TRF2f/f;TAg prostate tumors were ALT+ in this analysis (16-fold increased signal; P < 10−6; Fig. 5A), but not Sca1/CD49+ cells from K14Cre;TRF2+/+;TAg tumors. Sca1/CD49− cells from prostate tumors of both genotypes were ALT negative. We also examined K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors for ALT associated PML bodies (APB), a second marker of ALT. K14Cre;TRF2f/f;TAg but not K14Cre;TRF2+/+;TAg prostate tumors exhibited APB (P < 0.00001; Fig. 5B–D). In contrast, Sca1/CD49f+ and Sca1/CD49f− cells of both genotypes were telomerase positive (Fig. 5E). These results indicate that telomerase activity is present in prostate cancer stem and non–stem cell populations, but only K14Cre;TRF2f/f;TAg Sca1CD49f+ cells were ALT+.

Figure 5.
Figure 5.

Alternative lengthening of telomeres in cancer stem cells from K14Cre;TRF2f/f;TAg prostate cancer. A, Circular telomere DNA analysis in Sca1/CD49f+ cancer stem and luminal (Sca1/CD49f−) cells from K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors. U2OS was used as the positive control cell line, and template free reactions were used as the negative control. Hybridization signals were quantitated by digital image analysis. Experiments were performed three times with similar results. A representative blot is shown. ALT associated PML bodies (APB) were demonstrated by PML protein localization at telomeres (FISH) in K14Cre;TRF2+/+;TAg (B) and K14Cre;TRF2f/f;TAg (C and D) prostate tumors using immunofluorescence microscopy. E, Relative telomerase activity in Sca1/CD49f+ cancer stem (SC) and luminal (Sca1/CD49f−) cells from K14Cre;TRF2+/+;TAg and K14Cre;TRF2f/f;TAg prostate tumors was determined using a commercially available kit. Negative control is shown.

Reduced TRF2 expression inhibited prostate tumor initiation (Fig. 2), suggesting that the telomere DNA damage response could be therapeutically manipulated to improve clinical outcomes. Telomerase inhibition induces telomere DNA damage and apoptosis (8), has detrimental effects on human prostate cancer cell lines (40), and is being tested on other cancers in clinical trials. To obtain in vivo evidence for telomere DNA damage as a targeted therapy for aggressive prostate cancers, we created the K14Cre;TRF2f/f;Terc−/−;TAg mouse that lacks telomerase activity (Fig. 6H) in the TRF2-null prostate tumor prone background. K14Cre;TRF2f/f;Terc−/−;TAg mice exhibited a 73% reduction in prostate tumor initiation (P < 0.003; Fig. 6A). In contrast K14Cre;TRF2+/+;Terc−/−;TAg mice exhibited no reduction in prostate tumor initiation (data not shown) suggesting that despite the aggressive nature of K14Cre;TRF2f/f;TAg cancers, reduced TRF2 expression makes these tumors vulnerable to additional telomere DNA damage. K14Cre;TRF2f/f;Terc−/−;TAg mice without prostate tumors were healthy throughout the 12-month experiment (Fig. 6B) at which point prostate glands were dissected for histopathologic, cellular, and molecular analyses. Surprisingly, 18% of K14Cre;TRF2f/f;Terc−/−;TAg prostate glands were histopathologically classified as normal or low-grade PIN (Fig. 6C and D). The majority of K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (55%) were classified as high-grade PIN (Fig. 6E), and 27% of mice with this genotype developed prostate tumors that were classified as poorly differentiated adenocarcinoma (Fig. 6F). Despite this classification, all K14Cre;TRF2f/f;Terc−/−;TAg prostate tumors failed to metastasize to abdominal lymph nodes, lungs, or liver (P < 0.00002; Fig. 6G). These results indicate that telomerase inhibition dramatically reduces prostate tumor initiation and metastasis in prostate tumors with reduced TRF2 expression.

Figure 6.
Figure 6.

Telomerase inhibition blocks TRF2-null primary and metastatic prostate tumorigenesis. A, Fraction of K14Cre;TRF2+/+;TAg, K14Cre;TRF2f/f;TAg, and K14Cre;TRF2f/f;Terc−/−;TAg mice with prostate tumors. B, Latency of K14Cre;TRF2+/+;TAg, K14Cre;TRF2f/f;TAg, and K14Cre;TRF2f/f;Terc−/−;TAg prostate tumors. Percent tumor free mice is shown on the y-axis and time (days) is shown on the x-axis. Twenty mice per genotype were included in this experiment. Histopathology of prostate gland (C), early prostate intraepithelial neoplasia (D), advanced prostate intraepithelial neoplasia (E), and prostate tumors (F) from K14Cre;TRF2f/f;Terc−/−;TAg mice was determined by hematoxylin and eosin staining. G, Fraction of tumor bearing K14Cre;TRF2+/+;TAg, K14Cre;TRF2f/f;TAg, and K14Cre;TRF2f/f;Terc−/−;TAg mice with metastasis. H, Relative telomerase activity in K14Cre;TRF2+/+;TAg, K14Cre;TRF2f/f;TAg, and K14Cre;TRF2f/f;Terc−/−;TAg prostate tumors.

We characterized both high-grade PIN and prostate tumors from K14Cre;TRF2f/f;Terc−/−;TAg mice at the cellular level. K14Cre;TRF2f/f;Terc−/−;TAg high-grade PIN and prostate tumors were luminal phenotype as shown by K18 IHC (Supplementary Fig. S9A and S9B). Neither basal nor neuroendocrine differentiation was detected in high-grade PIN nor in prostate tumors as determined by K14 and chromogranin A IHC (Supplementary Fig. S9C–S9F). High-grade PIN exhibited high AR+ cell fractions (48% vs. 0.5%; P < 10−7; Supplementary Fig. S9G–S9I) compared with prostate tumors. The proliferating cell fraction was low in K14Cre;TRF2f/f;Terc−/−;TAg high-grade PIN and tumors (12% and 29% vs. 72%; P < 0.0004; Supplementary Fig. S9J–S9L) compared with K14Cre;TRF2f/f;TAg cancers. K14Cre;TRF2f/f;Terc−/−;TAg high-grade PIN exhibited increased apoptotic cell fraction (52% vs. 2%; P < 0.0007; Supplementary Fig. S9M–S9O) compared with prostate tumors. These results indicate that telomerase inhibition reduces cell proliferation in high-grade PIN and prostate tumors and induces apoptosis in PIN with reduced TRF2 expression.

Both K14Cre;TRF2f/f;Terc−/−;TAg high grade PIN and prostate tumors exhibited increased 53BP1 foci at telomeres (56% and 75% vs. 31%; P < 0.002; Fig. 7A, B, and M) compared with K14Cre;TRF2f/f;TAg tumors. K14Cre;TRF2f/f;Terc−/−;TAg high-grade PIN exhibited increased pATM+ (74%), pATR+ (88%), pChk2+ (72%), and p53+ (90%) cell fractions (P < 10−11; Fig. 7C, E, G, I, K, and M) compared with prostate tumors (Fig. 7D, F, H, J, L, and M) as determined by IHC. Neither K14Cre;TRF2f/f;Terc−/−;TAg high-grade PIN nor prostate tumors expressed APB (Fig. 7N and O). These results indicate that high levels of telomere DNA damage can block prostate tumor formation. Tumors that initiate and progress under these conditions have inactivated downstream DNA damage signaling.

Figure 7.
Figure 7.

High levels of telomere DNA damage signaling in K14Cre;TRF2f/f;Terc−/−-;TAg prostate glands and tumors. Localization of the DNA damage response protein 53BP1 at telomeres (FISH) in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (A) and tumors (B). Nuclei were counterstained with DAPI. Scale bar, 10 μm. pATM expression in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (C) and tumors (D) was determined by IHC. Nuclei were counterstained with hematoxylin. Scale bar, 10 μm. pATR expression in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (E) and tumors (F) was determined by IHC. pChk1 expression in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (G) and tumors (H) was determined by IHC. pChk2 expression in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (I) and tumors (J) was determined by IHC. p53 expression in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands K, and tumors (L) was determined by IHC. M, Quantitation of positive cells expressing DNA damage signaling proteins shown at right in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (PIN) and tumors (adenoCA). Lack of ALT-associated PML bodies in K14Cre;TRF2f/f;Terc−/−;TAg prostate glands (N) and tumors (O) was determined by colocalization of PML protein at telomeres (FISH) by immunofluorescence microscopy. Nuclei were counterstained with DAPI. Scale bar, 10 μm.

Discussion

Prostate epithelial cell senescence is a common feature of age-related conditions such as BPH (12). The cause of senescence in prostate epithelia in BPH is unclear, but telomere shortening has been demonstrated in these lesions (13). BPH is predominantly characterized by smooth muscle hyperplasia, which may be driven by signaling from senescent epithelial cells (for review, see ref. 14). In fact the mechanism of action of some drugs used to treat BPH is inhibition of prostate smooth muscle contraction. Our TRF2-mutant model exhibits prostate epithelial cell senescence and marked smooth muscle hyperplasia resembling human BPH. Although some dog breeds develop BPH with age (41), to our knowledge, this is the first mouse genetics model of the disease. Additional studies in this model will reveal paracrine mechanisms by which senescent prostate epithelium can promote smooth muscle hyperplasia leading to this condition frequently encountered in older men.

Another important finding of our study is the decompensation of prostate epithelial function resulting from telomere DNA damage response. Prostate epithelium transitioned from columnar to low cuboidal epithelium with changes in gene expression (e.g., golgin B1) affecting the cellular secretory pathway. Our ultrastructural data indicated that secretion was markedly reduced in prostate epithelium with reduced TRF2 expression. It is likely that the markedly reduced flow of prostate secretions allow retrograde migration of sperm in prostate glands with reduced TRF2 expression. This finding reflects the diminished prostate gland function in these mice rather than a correlation with human disease. Our study describes a unique model for understanding prostate epithelial biology and tumorigenesis.

Telomere DNA damage induced profound depletion of Sca1/CD49f+ cancer stem cells in our prostate cancer model. This is likely due to double strand break activation of ATM/Chk2/p53 signaling that induces stem cell apoptosis (42), reflected in the reduced onset of prostate tumors in the conditional TRF2-mutant mouse. It is likely that the small stem cell fraction survived due to telomere DNA damage–induced genomic instability induced by chromosomal fusions and translocations (43) that altered survival mechanisms. This altered gene expression also resulted in reduced prostate tumor latency, altered primary tumor differentiation, increased proliferation, and inhibition of cell death pathways. GEO data demonstrated that reduced TRF2 expression is frequently observed in prostate cancer. Our data demonstrate that reduced TRF2 expression promotes prostate tumor progression, and provides a mechanism for this outcome. Additional studies will be needed to characterize genetic rearrangements and gene expression changes leading to aggressive prostate cancer phenotypes in this model.

An important finding of our study is that the Sca1/CD49f+ prostate cancer stem cell population expresses ALT activity in the context of telomere DNA damage. ALT is DNA repair mechanism that uses the homologous recombination pathway to maintain telomeres (44). A previous study using ALT+ cells demonstrated that PML induced TRF2 depletion and telomere DNA damage response (45). Our previous studies of stem cells in stratified epithelia suggest that ALT induction is common when these populations experience telomere DNA damage (46), which is independent of the cause of the damage. Cdk-dependent TRF1 phosphorylation was required for ALT induction, which was ATM, Mre11, and BRCA1 dependent (47), again highlighting the role of recombination in this repair process. This study also demonstrated that ALT was sensitive to transcriptional inhibition and RNase H1, suggesting that ALT may be induced from processing telomere DNA-RNA hybrids. Mutations in the chromatin remodeling gene ATRX have been associated with ALT in some cancers (48), and may be due to inability to resolve sister telomere cohesion thus promoting recombination (49). ALT+ tumor cells may be sensitive to ATR inhibition, suggesting another targeted therapy for these cancers (50). Additional studies will be needed to determine why telomere DNA damage activates ALT in prostate cancer stem cells but not the Sca1/CD49f− luminal population.

Another key finding of our study is that the level of telomere DNA damage signaling regulates prostate tumor phenotype. Reduced TRF2 expression promotes tumorigenesis in surviving prostate cancer stem cells; however, combined TRF2 and Terc deficiency is extremely detrimental to both primary and metastatic tumor formation. Both increased and decreased DNA damage response has been associated with major alterations in cancer phenotype such as metastasis. Targeting of DNA damage response in cancer stem cells may improve clinical outcomes. Additional studies will be needed to determine how telomere DNA damage response could be exploited as a targeted cancer therapy.

Prostate tumors that survive combined TRF2 and Terc deficiency do not exhibit APB (a marker of ALT) nor do they exhibit telomerase activity. It is possible that lack of telomere maintenance is responsible for the dramatic reduction in prostate tumor initiation in the TRF2f/fTerc−/− genotype. However the question of telomere maintenance mechanisms in surviving tumor cells remains. Additional studies will be necessary to determine how these tumors maintained telomeres. An interesting observation in our TRF2f/fTerc−/− prostate tumors is the apparent lack of telomere DNA damage signaling. These results suggest the existence of an uncharacterized mechanism of telomere maintenance, or alternatively that telomere maintenance in certain cancer contexts is not strictly required if the DNA damage response is overcome. This latter hypothesis would suggest high levels of genomic instability in these tumors, which may be required for initiation but also are detrimental as reflected by the absence of metastatic tumor formation in the TRF2f/fTerc−/− prostate cancer genotype. Additional detailed genomic analyses of telomere length and DNA damage in these rare tumors are needed to distinguish between these mechanisms.

Disclosure of Potential Conflicts of Interest

J. Wu reports receiving grants from U.S Department of Defense during the conduct of the study. D.L. Crowe reports receiving grants from U.S. Department of Defense during the conduct of the study. No other potential conflicts of interest were disclosed.

Authors' Contributions

J. Wu: Formal analysis, investigation. D.L. Crowe: Conceptualization, formal analysis, supervision, funding acquisition, writing-original draft, project administration, writing-review and editing.

Acknowledgments

We thank Dr. Ke Ma and Wei Feng (University of Illinois Research Resources Center, Chicago, IL) for assistance with microscopy and flow cytometry. We thank Drs. HuangHui Tang and Erwin Goldberg (Northwestern University, Evanston, IL) for Ldhc antibody. This study was supported by Department of Defense grant W81XWH-10-1-0081 (to D.L. Crowe).

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.

Footnotes

  • Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/).

  • Mol Cancer Res 2020;18:1326–39

  • Received November 24, 2019.
  • Revision received March 30, 2020.
  • Accepted May 21, 2020.
  • Published first May 28, 2020.

References

  1. 1.↵
    1. Palm W,
    2. de Lange T
    . How shelterin protects mammalian telomeres. Ann Rev Genet 2008;42:301–34.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Arnoult N,
    2. Karlseder J.
    Complex interactions between the DNA damage response and mammalian telomeres. Nat Struct Mol Biol 2015;22:859–66.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Martinez P,
    2. Blasco MA.
    Role of shelterin in cancer and aging. Aging Cell 2010;9:653–66.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Celli GB,
    2. de Lange T
    . DNA processing is not required for ATM mediated telomere damage response after TRF2 deletion. Nature Cell Biol 2005;7:712–8.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Denchi EL,
    2. de Lange T
    . Protection of telomeres through independent control of ATM and ATR by TRF2 and POT1. Nature 2007;448:1068–71.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Choi J,
    2. Cheung P,
    3. Beaudry V,
    4. Moore H,
    5. Artandi SE.
    1. Rajaraman S
    . Choi J, Cheung P, Beaudry V, Moore H, Artandi SE. Telomere uncapping in progenitor cells with critical telomere shortening is coupled to S phase progression in vivo. Proc Natl Acad Sci U S A 2007;104:17747–52.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Lloyd NR,
    2. Dickey TH,
    3. Hom RA,
    4. Wuttke DS
    . Tying up the ends: plasticity in the recognition of single stranded DNA at telomeres. Biochemistry 2016;55:5326–40.
    OpenUrl
  8. 8.↵
    1. Blasco MA,
    2. Lee HW,
    3. Hande MP,
    4. Samper E,
    5. Lansdorp PM,
    6. DePinho RA,
    7. et al.
    Telomere shortening and tumor formation by mouse cells lacking telomerase RNA. Cell 1997;91:25–34.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Wu G,
    2. Jiang X,
    3. Lee WH,
    4. Chen PL
    . Assembly of functional ALT associated promyelocytic leukemia bodies requires Nijmegen breakage syndrome 1. Cancer Res 2003;63:2589–95.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Nabetani A,
    2. Ishikawa F.
    Unusual telomeric DNAs in human telomerase negative immortalized cells. Mol Cell Biol 2009;29:703–13.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Hockemeyer D,
    2. Collins K.
    Control of human telomerase action at telomeres. Nat Struct Mol Biol 2015;22:848–52.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Castro P,
    2. Giri D,
    3. Lamb D,
    4. Ittmann M
    . Cellular senescence in the pathogenesis of benign prostatic hyperplasia. Prostate 2003;55:30–8.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Heaphy CM,
    2. Fleet TM,
    3. Treat EG,
    4. Lee SJ,
    5. Smith AY,
    6. Davis MS,
    7. et al.
    Organ wide telomeric status in diseased and disease free prostatic tissues. Prostate 2010;70:1471–9.
    OpenUrlPubMed
  14. 14.↵
    1. Roehrborn CG.
    Pathology of benign prostatic hyperplasia. Int J Impot Res 2008;20:S11–8.
    OpenUrl
  15. 15.↵
    American Cancer Society. Cancer facts and figures. Atlanta, GA: American Cancer Society; 2016.
  16. 16.↵
    1. Shen MM,
    2. Abate-Shen C.
    Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev 2010;24:1967–2000.
    OpenUrlAbstract/FREE Full Text
  17. 17.↵
    1. Meeker AK,
    2. Hicks JL,
    3. Platz EA,
    4. March GE,
    5. Bennett CJ,
    6. Delannoy MJ,
    7. et al.
    Telomere shortening is an early somatic DNA alteration in human prostate tumorigenesis. Cancer Res 2002;62:6405–9.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Vukovic B,
    2. Park PC,
    3. Al-Maghrabi J,
    4. Beheshti B,
    5. Sweet J,
    6. Evans A,
    7. et al.
    Evidence of multifocality of telomere erosion in high grade prostatic intraepithelial neoplasia (HPIN) and concurrent carcinoma. Oncogene 2003;22:1978–87.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Meeker AK,
    2. Hicks JL,
    3. Iacobuzio-Donahue CA,
    4. Montgomery EA,
    5. Westra WH,
    6. Chan TY,
    7. et al.
    Telomere length abnormalities occur early in the initiation of epithelial carcinogenesis. Clin Cancer Res 2004;10:3317–26.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    1. Joshua AM,
    2. Vukovic B,
    3. Braude I,
    4. Hussein S,
    5. Zielenska M,
    6. Srigley J,
    7. et al.
    Telomere attrition in isolated high grade prostatic intraepithelial neoplasia and surrounding stroma is predictive of prostate cancer. Neoplasia 2007;9:81–9.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Tu L,
    2. Huda N,
    3. Grimes BR,
    4. Slee RB,
    5. Bates AM,
    6. Cheng L,
    7. et al.
    Widespread telomere instability in prostatic lesions. Mol Carcinog 2016;55:842–52.
    OpenUrl
  22. 22.↵
    1. Sommerfeld HJ,
    2. Meeker AK,
    3. Piatyszek MA,
    4. Bova GS,
    5. Shay JW,
    6. Coffey DS
    . Telomerase activity: a prevalent marker of malignant human prostate tissue. Cancer Res 1996;56:218–22.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Engelhardt M,
    2. Albanell J,
    3. Drullinsky P,
    4. Han W,
    5. Guillem J,
    6. Scher HI,
    7. et al.
    Relative contribution of normal and neoplastic cells determines telomerase activity and telomere length in primary cancers of the prostate, colon, and sarcoma. Clin Cancer Res 1997;3:1849–57.
    OpenUrlAbstract
  24. 24.↵
    1. Donaldson L,
    2. Fordyce C,
    3. Gilliland F,
    4. Smith A,
    5. Feddersen R,
    6. Joste N,
    7. et al.
    Association between outcome and telomere DNA content in prostate cancer. J Urol 1999;162:1788–92.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Fordyce CA,
    2. Heaphy CM,
    3. Joste NE,
    4. Smith AY,
    5. Hunt WC,
    6. Griffith JK
    . Association between cancer free survival and telomere DNA content in prostate tumors. J Urol 2005;173:610–4.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Baydar DE,
    2. Ozen H,
    3. Geyik PO,
    4. Gurel B
    . Can telomere alterations predict biochemical recurrence in prostate adenocarcinoma? Pathol Res Prac 2010;206:700–4.
    OpenUrl
  27. 27.↵
    1. Treat EG,
    2. Heaphy CM,
    3. Massie LM,
    4. Bisoffi M,
    5. Smith AY,
    6. Davis MS,
    7. et al.
    Telomere DNA content in prostate biopsies predicts early rise in prostate specific antigen after radical prostatectomy for prostate cancer. Urology 2010;75:724–9.
    OpenUrlPubMed
  28. 28.↵
    1. Heaphy CM,
    2. Yoon GS,
    3. Peskoe SB,
    4. Joshu CE,
    5. Lee TK,
    6. Giovannucci E,
    7. et al.
    Prostate cancer cell telomere length variability and stromal cell telomere length as prognostic markers for metastasis and death. Cancer Disc 2013;3:1130–41.
    OpenUrl
  29. 29.↵
    1. Botchkina GI,
    2. Kim RH,
    3. Botchkina IL,
    4. Kirshenbaum A,
    5. Frischer Z,
    6. Adler HL
    . Noninvasive detection of prostate cancer by quantitative analysis of telomerase activity. Clin Cancer Res 2005;11:3243–9.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Dalle Carbonare L,
    2. Gasparetto A,
    3. Donatelli L,
    4. Dellantonio A,
    5. Valenti MT
    . Telomerase RNA detection in serum of patients with prostate cancer. Urol Oncol 2013;31:205–10.
    OpenUrl
  31. 31.↵
    1. Zhang K,
    2. Guo Y,
    3. Wang X,
    4. Zhao H,
    5. Ji Z,
    6. Cheng C,
    7. et al.
    WNT/β-catenin directs self renewal symmetric cell division of hTERThigh prostate cancer stem cells. Cancer Res 2017;77:2534–47.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    1. Ding Z,
    2. Wu CJ,
    3. Jaskelioff M,
    4. Ivanova E,
    5. Kost-Alimova M,
    6. Protopopov A,
    7. et al.
    Telomerase reactivation following telomerase dysfunction yields murine prostate tumors with bone metastases. Cell 2012;148:1–12.
    OpenUrlCrossRef
  33. 33.↵
    1. Lawson DA,
    2. Witte ON.
    Stem cells in prostate cancer initiation and progression. J Clin Invest 2007;117:2044–50.
    OpenUrlCrossRefPubMed
  34. 34.↵
    1. Choi N,
    2. Zhang B,
    3. Zhang L,
    4. Ittmann M,
    5. Xin L
    . Adult murine prostate basal and luminal cells are self sustained lineages that can both serve as targets for prostate cancer initiation. Cancer Cell 2012;21:253–65.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Wang ZA,
    2. Mitrofanova A,
    3. Bergren SK,
    4. Abate-Shen C,
    5. Cardiff RD,
    6. Califano A,
    7. et al.
    Lineage analysis of basal epithelial cells reveals their unexpected plasticity and supports a cell of origin model for prostate cancer heterogeneity. Nature Cell Biol 2013;15:274–83.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. Wu J,
    2. Crowe DL.
    Molecular and cellular basis of mammary gland fibrosis and cancer risk. Int J Cancer 2019;144:2239–53
    OpenUrl
  37. 37.↵
    1. Wang Y,
    2. Hayward SW,
    3. Cao M,
    4. Thayer KA,
    5. Cunha GR
    . Cell differentiation lineage in the prostate. Differentiation 2001;68:270–9.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Chang RT,
    2. Kirby R,
    3. Challacombe BJ
    . Is there a link between BPH and prostate cancer? Practitioner 2012;256:13–6.
    OpenUrlPubMed
  39. 39.↵
    1. Ruscetti M,
    2. Quach B,
    3. Dadashian EL,
    4. Mulholland DJ,
    5. Wu H
    . Tracking and functional characterization of epithelial mesenchymal transition and mesenchymal tumor cells during prostate cancer metastasis. Cancer Res 2015;75:2749–59.
    OpenUrlAbstract/FREE Full Text
  40. 40.↵
    1. Marian CO,
    2. Wright WE,
    3. Shay JW
    . The effects of telomerase inhibition on prostate tumor initiating cells. Int J Cancer 2010;127:321–31.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Prins GS,
    2. Jung MH,
    3. Vellanoweth RL,
    4. Chatterjee B,
    5. Roy AK
    . Age dependent expression of the androgen receptor gene in the prostate and its implication in glandular differentiation and hyperplasia. Dev Genet 1996;18:99–106.
    OpenUrlCrossRefPubMed
  42. 42.↵
    1. Roos WP,
    2. Thomas AD,
    3. Kaina B
    . DNA damage and the balance between survival and death in cancer biology. Nature Rev Cancer 2016;16:20–33.
    OpenUrlCrossRefPubMed
  43. 43.↵
    1. Bunting SF,
    2. Nussenzweig A
    . End joining, translocations, and cancer. Nature Rev Cancer 2013;13:443–54.
    OpenUrlCrossRefPubMed
  44. 44.↵
    1. Durant ST.
    Telomerase independent paths to immortality in predictable cancer subtypes. J Cancer 2012;3:67–82.
    OpenUrlCrossRefPubMed
  45. 45.↵
    1. Osterwald S,
    2. Deeg KI,
    3. Chung I,
    4. Parisotto D,
    5. Worz S,
    6. Rohr K,
    7. et al.
    PML induces compaction, TRF2 depletion, and DNA damage signaling at telomeres and promotes their alternative lengthening. J Cell Sci 2015;128:1887–900.
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    1. Bojovic B,
    2. Booth RE,
    3. Jin Y,
    4. Zhou X,
    5. Crowe DL
    . Alternative lengthening of telomeres in cancer stem cells in vivo. Oncogene 2015;34:611–20.
    OpenUrl
  47. 47.↵
    1. Wilson FR,
    2. Ho A,
    3. Walker JR,
    4. Zhu XD
    . Cdk dependent phosphorylation regulates TRF1 recruitment to PML bodies and promotes C-circle production in ALT cells. J Cell Sci 2016;129:2559–72.
    OpenUrlAbstract/FREE Full Text
  48. 48.↵
    1. Koschmann C,
    2. Lowenstein PR,
    3. Castro MG
    . ATRX mutations and glioblastoma: impaired DNA damage repair, alternative lengthening of telomeres, and genetic instability. Mol Cell Oncol 2016;3:e1157158.
    OpenUrl
  49. 49.↵
    1. Ramamoorthy M,
    2. Smith S.
    Loss ATRX suppresses resolution of telomere cohesion to control recombination in ALT cancer cells. Cancer Cell 2015;28:357–69.
    OpenUrlCrossRefPubMed
  50. 50.↵
    1. Flynn RL,
    2. Cox KE,
    3. Jeitany M,
    4. Wakimoto H,
    5. Bryll AR,
    6. Ganem NJ,
    7. et al.
    Alternative lengthening of telomeres renders cancer cells hypersensitive to ATR inhibitors. Science 2015;347:273–7.
    OpenUrlAbstract/FREE Full Text
Back to top
View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
Telomere DNA Damage Signaling Regulates Prostate Cancer Tumorigenesis
Jianchun Wu and David L. Crowe
Mol Cancer Res September 1 2020 (18) (9) 1326-1339; DOI: 10.1158/1541-7786.MCR-19-1129
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement